User talk:Jimbo Wales/Archive 208

Latest comment: 8 years ago by MediaWiki message delivery in topic The Signpost: 15 June 2016
Archive 205Archive 206Archive 207Archive 208Archive 209Archive 210Archive 215

Correction

Hi Mr. Wales, I had made some comments in the Wikipedia talk pages. These are the talk pages I have commented in: Sasanian Empire, Parthian empire, Achaemenid empire. Now I want to correct those comments.

I had no intention to make the comments in the talk pages. I am not so good in English. I just wanted to include some maps in some of the articles. But then the other editors said I must have a consensus to include my edits. So I had to make those comments. But then I realized it was a mistake. There are many mistakes in my language. So I want to correct those comments.

Mr.Wales, I have made three accounts for this edit. This is the fourth one. It will probably not be possible for me to make a new one. The other editors are not allowing me to do this. So I have come to you Mr. Wales. You are the owner of Wikipedia. You have the power to let me edit the comments. So please let me do this a bit.

I don't have the capability to write any article in Wikipedia. I am not so good in English. I just want to correct my comments. I will only do some grammatical change. I have to correct the comments, I have to correct them at any cost. I am feeling very uncomfortable about the comments in the talk pages. This is the last thing I want to do in Wikipedia. After doing this I will quit Wikipedia. So please mr. Wales, let me do this a bit.Arman ad88 (talk) 19:47, 12 May 2016 (UTC)

Jimmy isn't the owner of Wikipedia. He is a founder and on the board of trustees of the Foundation, and an editor. Consequently he gets a certain amount of respect, and flack, and requests to intervene.
If you have grammatical errors in your comments, don't worry. As long as the meaning is clear it is fine.
As to the maps, it is a complicated situation. I have cleaned up one of the maps of that area we have on Commons, but I can't remember if this used on any of those pages.
All the best: Rich Farmbrough, 21:50, 13 May 2016 (UTC).
Wow, Arman, it seems you sure know how to get blocked. And even lose your Talk Page access too. Most people just try and stick to one account. I guess your current account will soon be blocked too. Which may be unfortunate, as you seem to want to just correct what you see as mistakes that you've made. Martinevans123 (talk) 22:05, 13 May 2016 (UTC)
I agree that would be unfortunate. I haven't reviewed his edit history but he seems to be wanting to make things right and then stop. I see no harm in letting him try to do that.--Jimbo Wales (talk) 10:32, 14 May 2016 (UTC)

Thank you very much Mr. Jimbo Wales for understanding it. I just want to correct my comments. I just want to do some grammatical change in my comments. I want to do this because this is an encyclopedia. The people of the world will watch this. If they see my incorrect comments there it will look very odd. This is why I want to correct the comments. If you want to block my account after this I am ready to accept it.Arman ad88 (talk) 12:11, 14 May 2016 (UTC)

Arman, assuming (as I do) that you are editing here in good faith, I recommend that you switch instead to improving the Wikipedia articles in the language(s) in which you are more fluent, all of which need new editorial participants. --Orange Mike | Talk 01:25, 15 May 2016 (UTC)

No no it is not possible for me write any article in Wikipedia. I don't have the skills to write any article in Wikipedia. I just came in Wikipedia as a novice, then I saw some problems in the articles and then had to make those comments. The problem here is that, I don't think my incorrect comments should to be there. These are very important articles. Many people of the world will watch them. I don't want people to see my incorrect comments. That is why I want to correct the comments. That's it, nothing more than that. So please let me do this a bit.Arman ad88 (talk) 10:31, 15 May 2016 (UTC)

If you feel that strongly, about correcting obvious mistakes, you could actually log out and correct them anonymously. I'm sure no-one would complain about that. If they are just Talk Page comments, I'm really not sure they will be read by "many people of the world". Martinevans123 (talk) 15:24, 15 May 2016 (UTC)

OK if I log out and correct my comments anonymously, will it be accepted? There is a problem. I had tried few times to correct my comments. Other editors have reverted my edits. And finally they even blocked my IP adress. I fear that this kind of thing can happen again. So please tell me what can I do about that.Arman ad88 (talk) 16:04, 15 May 2016 (UTC)

I'd suggest that the best course of action is for you to discuss the things you want corrected at the Talk Page(s) of the relevant article(s). Martinevans123 (talk) 16:08, 15 May 2016 (UTC)

If I discuss about the edits in the talk pages they will be more messy. The other editors may be more angry with me. They may even block my account. So I don't think that is a right idea.Arman ad88 (talk) 18:00, 15 May 2016 (UTC)

Arman, I have reviewed the talk pages that you mentioned, as well as some of the past history - sock puppet investigation, etc. I did not review whatever behaviors led to your first block, as you have admitted to mistakes in the past, so I see no reason to go into that. Here is my warm recommendation to you, and I hope you will take it. The standard offer strikes me as your most useful approach, and I would encourage you to follow it, and upon your return, let people know that you are sorry for what happened before. It seems to me that you do have something to offer the project if you remain calm and take suggestions from others as to how to improve and compromise on various map details.--Jimbo Wales (talk) 19:08, 16 May 2016 (UTC)

I think you don't trust me because I have made many mistakes in the earlier times. I admit that I have done many mistakes. But it is also true that the articles I was trying to improve were at very poor condition. So I wanted to improve them by any means. You can consider it as inexperience of a newcomer.

I just want to correct my comments. Mr. Wales, it is a very minor edit. It will only take few minutes. Then the problem will be solved. This is the last thing I want to do in Wikipedia. And then I will leave Wikipedia. So please let me do this a bit.Arman ad88 (talk) 04:12, 17 May 2016 (UTC)

OK I think we can solve the problem another way. I don't want to edit the comments myself. But can someone of you help me correct the comments in the talk pages? Then the problem will be solved. Can some one of you help me improve the English in my comments in the talk pages?Arman ad88 (talk) 10:59, 18 May 2016 (UTC)

Howdy Jimbo.

Howdy, I think you need to change the image policy, Im from the NIWA (Nintendo Independent Wiki Alliance) they allow non-free images, but not Wikipedia, and I heard also on the Wikias, you allowed them to use non-free images also. Knowing you are one of the founders of Wikipedia, can you allow Non-free images? I wan't a non-free image as a Personal Image. And please only allow non-free images allowed for Personal Images if they are NOT intended for a article. Thanks! DatNuttyWikipedian (talk) 23:34, 16 May 2016 (UTC)

Hello, DatNuttyWikipedian. Wikipedia allows limited use of non-free images, within articles only, and only when the image is irreplaceable. Please read WP:NFCI. In my personal opinion, the chance for a dramatic liberalization of this policy is negligible. Cullen328 Let's discuss it 00:27, 17 May 2016 (UTC)
.......even though it would make for a far better encyclopedia and would represent extremely minimal legal risk to WMF. Carrite (talk) 02:33, 17 May 2016 (UTC)
Note that legal risk has very little to do with it. Reusability - a core principle of software freedom and of Wikipedia - is the primary reason. Keeping the website safe is only a part of our responsibility - we want people to be able to take our work, follow the license, and reuse it.--Jimbo Wales (talk) 13:30, 17 May 2016 (UTC)
Sometimes "free to use" gets in the way of using things that are freely usable. For example, there are lots of CC-by-NC videos from major universities, foundations, and museums that we would like to be able to use - that would greatly improve the encyclopedia, but just don't have a chance to appear on Wikipedia. I suppose the argument is that if we "hold the line" and don't allow marginally less free material, the institutions will eventually come around to our view and start using CC-BY-SA. But practically, that's not happening. Please consider allowing these, at least on an institution-by-institution basis. Another example of where "total freedom" gets in the way of "practical freedom" is with MP4's. Putting videos on Wikipedia in a "totally free" format, limits the number of people who can view them. Is there a real hangup about using MP4s or is the lack of freedom purely theoretical? People do get carried away on this stuff. Smallbones(smalltalk) 17:37, 18 May 2016 (UTC)

April 2016 volunteer stats

Once again we find that reports of WP's demise have been greatly exaggerated... April 2016 stats show a count of 3,309 Very Active Editors (100+ edits/mo.) at English Wikipedia, an increase of 5.1% over the figure for April 2015. This number surpasses the same-month figures for 2012, 2013, 2014, and 2015. New Articles Per Day is up slightly, from 804 last year to 834 this year.

The count of Very Active Editors across all the language Wikipedias is flat at the 10,000 mark — basically 3 times more than the En-WP total. There were 15 language Wikipedias with 100 or more Very Active Editors in the month, with English, German, French, Russian, and Spanish being the top five in sequence. Some 59% of Very Active Wikipedians participate through these five projects. Carrite (talk) 19:08, 18 May 2016 (UTC)

Hey Jimbo

Will you let <your daughter> edit Wikipedia? If so, what would you say about child protection? Because harassment is rampant here. — Preceding unsigned comment added by Peter Eron Quinn (talkcontribs) 08:37, 15 May 2016 (UTC)

Who is <removed>? There isn't a <name removed> registered user with this name. Some possible explanations here.--♦IanMacM♦ (talk to me) 11:07, 15 May 2016 (UTC)
I am guessing that <name removed> might be the name of one of Jimbo's daughters, and if I am right, he would possibly not appreciate questions about his parenting on WP. EdChem (talk) 12:40, 15 May 2016 (UTC)
This article might give us a few clues, I guess. Martinevans123 (talk) 13:07, 15 May 2016 (UTC)
I strongly support our policy on Wikipedia:Child_protection and indeed would not mind seeing it strengthened.--Jimbo Wales (talk) 12:31, 16 May 2016 (UTC)
What specific changes would you see as strengthening the child protection policy? I think it's adequate and reasonable as is, but would be interested to know what you have in mind. HolidayInGibraltar (talk) 02:10, 17 May 2016 (UTC)
I don't have a specific proposal in mind. My point was: the implicit hostility in the question (bringing my own children into the discussion for no reason) appeared to me to be an attempt to suggest that I'm somehow reluctant on this issue, when in fact I'm quite hawkish on it. I think that harassment of all kinds, and particularly that involving sexual harassment of children, is to be fought vigorously.--Jimbo Wales (talk) 10:18, 17 May 2016 (UTC)
The problem some of us who are 'hawkish' about these things face is that the administrative corps (as a group) and Arbcom are incredibly bad at dealing with child protection issues. In some cases this is a justified 'We dont get paid or trained for this' response, the worse cases however are the 'shoot the messenger' responses that are all too commmon. The most notable one being an editor who repeatedly brought up (with evidence from editing history) where another editor was advising a minor to circumvent his parents prohibition on contact with him. The reporting editor was blocked, and it was a significant time before the problematic editor engaging in inappropriate contact was globally banned by the WMF. A significant time I might add in which repeated complaints were made to the WMF and the local police authorities about the editor. Did the original complainent ever get an apology or even a thank you? I think you can guess the answer. This clearly has a chilling effect on editors in general, as it encourages turning a blind eye in case you get sanctioned. Why report something you know is wrong when you will end up blocked for it. Only in death does duty end (talk) 11:19, 17 May 2016 (UTC)
I'd love an email with the details of this case, and especially timings. My understanding is that the situation has significantly improved fairly recently, due to the Foundation investing more in this area. If there is still a problem, I will advocate for spending more. Responses to concerns like this should be prompt.--Jimbo Wales (talk) 13:32, 17 May 2016 (UTC)
I would suggest asking Kiefer Wolfowitz for a timeline but he is still banned. It was on your page in August 2013 and Demiurge1000 was banned in December 2014. You might want to ask the foundation directly regarding the details. Only in death does duty end (talk) 17:20, 17 May 2016 (UTC)
Children may face potentially very serious problems (e.g. sexual abuse), but it is easy to go overboard with protective measures. Well meant measures can lead to disruption, interfering with normal everyday activities. Cases like this are deeply troubling. When I was 5 years old, I used to walk a lot farther to Kindergarten alone. Today, this is taboo (at least in the US), it's not even good enough for a young child to be accompanied by a ten year old. You have to wonder how children raised this way will turn out as adults. Count Iblis (talk) 20:58, 17 May 2016 (UTC)
One doesn't have to wonder. Just look at college campuses today with their safe spaces, trigger warnings, prohibitions against microagressions, and other efforts to protect students from any kind of intellectual or emotional discomfort. Deli nk (talk) 18:14, 18 May 2016 (UTC)
What is your familiarity with this phenomenon? Do you work or teach at a college? Or do you have a child who attends college? Or is just an notion you got from reading blogs? There are thousands of institutes of higher education in the U.S. alone and it's a mistake to think an outlier or two is representative of the majority of colleges and universities. Liz Read! Talk! 22:12, 18 May 2016 (UTC)
I'm quite familiar with this phenomenon. It's my own experience as a student. If your college experience didn't involve these things, consider yourself fortunate. Deli nk (talk) 01:41, 19 May 2016 (UTC)
  • I wondered about the same user page to which Sir Joseph refers. In that case, the child identified his age on ANI in a comment to an Arbitrator, and shortly afterwards the user page content and ANI comment were gone, presumably oversighted. The problem was addressed quickly and quietly, an example of the system working, IMO. EdChem (talk) 15:09, 18 May 2016 (UTC)

Have you uploaded any images?

Hey Jimbo, I suppose I could just check through the proper links but, have you ever uploaded any images and if so, do you still take pics and upload? If not....can you be talked into it. LOL! ;)--Mark Miller (talk) 04:58, 19 May 2016 (UTC)

Jimbo's upload log is easy to find, for goodness' sake. He hasn't uploaded very many images and they date mostly (apart from the deleted ones) from the period 2004-6.-♦IanMacM♦ (talk to me) 05:48, 19 May 2016 (UTC)
I should upload more. The truth is, I don't take a lot of pictures anyway, other than family snapshots.--Jimbo Wales (talk) 16:41, 19 May 2016 (UTC)
Please take a picture of my future corner office, with view. I aim to move up the corporate ladder. Drmies (talk) 16:42, 19 May 2016 (UTC)
I love to take pictures and contribute them as much as researched and referenced, written content. I always wondered if video is not taken advantage of here as much as it could? Perhaps I just need to understand how to upload video better but, it does seem a bit complicated. Anyway....make sure Drmies office has a nice view. ;)--Mark Miller (talk) 06:41, 20 May 2016 (UTC)

Update re: Attawapiskat Suicide Crisis among Canadian/Native youth

Great news re: encouraging the youth there to edit Wikipedia, join our community, and converse with Wikipedians world-wide. No luck yet via the teachers and a high school administrator whom I contacted by email; however, I added to my efforts by going to facebook to contact people who live in Attawapiskat, and that approach seems to be really taking off ! I have so far gained 2 facebook "friends" who live there as well as I joined a facebook group that is specific towards their suicide crisis. You can see what's happening and will happen there at Attawapiskat Suicide Awareness Nocturnalnow (talk) 03:35, 20 May 2016 (UTC)

This looks very worthwhile. Can you let editors know (here or maybe in the Signpost) what we can do to help. Maybe a few mentors would help? Smallbones(smalltalk) 16:31, 20 May 2016 (UTC)
Beyond the Attawapiskat region, there are articles for see-also in the pages "Suicide" and "Suicide prevention" to inform readers about the issues, such as "List of suicide crisis lines" or "World Suicide Prevention Day". -Wikid77 (talk) 17:04, 21 May 2016 (UTC)
  • Improve crisis awareness: I also think WP could do more to help improve other crisis awareness among our readers. For example, hurricane victims need to learn about the new U.S. Government plans to post an "inundation map" (as a predicted flood map) plus "storm surge warning" messages, to better alert inhabitants about the dangers of flooding in their specific areas. Previously, the complex concept of "storm surge" (as the extra level of water in the storm tide, excluding waves) has not been well understood where people were shocked how a hurricane could have such extremely high waters with waves. For example, Hurricane Katrina in Alabama had waves over 22 feet (6.7 m) high, although the flood was officially only 14 feet (4.3 m) above the high tide on the morning of 29 August 2005, and when homes were destroyed, the refrigerators or other floating appliances were pushed along the ditches of highways, while some fishing boats landed in the limbs of large trees. -Wikid77 (talk) 15:02, 20 May 2016 (UTC)

Why not measure effect of providing desktop interface to mobile users?

Jimbo, when you opposed measuring whether mobile users offered the ordinary desktop interface edit or donate more, you did not seem to offer an actual reason against making the measurement although I think it's fair to say that you implied someone would have to be neo-luddite to think of doing away with the Foundation's custom mobile app content viewers and reassigning their engineering headcount to getting a unified mobile and desktop web experience. While I am vaguely inclined to agree that the Foundation should support diversity in content viewing applications, I just want to be sure that we are not doing so at the expense of editors and donations. And I think it's clear that the existing diversity of web browsers on the different desktop and mobile platforms causes custom content viewer applications to add needless and expensive complexity to an already very complicated software support task, for which there is no clear cost-benefit justification. I doubt dropping mobile apps, if doing so can increase editing contributions and financial donations, will cause anything like the state of mobile app development to return to horse-and-buggy days. Are there any reasons to oppose the measurement? EllenCT (talk) 20:28, 21 May 2016 (UTC)

GMO's and concerns about corporate influence

Hi Jimmy,

Previously I expressed concerns about how the GMO articles appear to have the hand of the GM industry in them. There is discussion about the rules of a soon to be launched high stakes RfC regarding language of GMO safety that is in the lede of many GM articles.

FYI, I made the following suggestion here to address concerns of possible influence from industry:

Choice of Closing Admins
COI concerns: It is hard to ascertain whether anyone involved in GMO articles edits with a financial motivation. As we saw with WifiOne ([1][2]), anonymous paid editors can be very crafty in avoiding scrutiny and gaining positions of power at Wikipedia.
With billions [3] at stake, the GM industry spent ~$100 Million fighting U.S. GMO labeling laws ([4][5][6]).
The industry could easily afford to hire a team of full-time anonymous Pro-GM editors like WiFiOne to make sure the articles reflect their views in each article lede, in the same way that BP was able to write 44% of its article, including the Deepwater horizon oil spill [7].
Wikipedia's policies of anonymity combined with WP:AGF and ArbCom's GMO ruling protect editors from criticisms for similar behavior.
In this high stakes RfC, I suggest that the three closing admins make a declaration that they have no direct or indirect financial or fiduciary conflict of interest (COI) by participating in this RfC. Such COI might include, for example, employment or contracting for a GM company, holding a GM patent, doing GM research, or working at a PR firm.

--David Tornheim (talk) 03:34, 18 May 2016 (UTC)

  • Jimmy, I second David's concerns. It appears to me that the proposed Rfc on GMO's is largely an attempt to silence those with legitimate concerns and questions regarding those editors who edit relentlessly in favor of corporate interests. Under the rubric of "casting aspersions" which can be used as a chilling effect to threaten editors like David, the impression I get is that some editors seek to inoculate themselves from any scrutiny whatsoever. In my view this is a recipe for turning Wikipedia into a PR machine that serves industry interests. I have suggested the proposed Rfc be scrapped. Jusdafax 13:02, 18 May 2016 (UTC)
I generally hate using quippy internet slang but LOL. Most of the editors that favor the wording you oppose have been here for years and have tens of thousands of edits across a spectrum of subjects but it's your contention that they're paid sleeper-shills embedded by food companies years ago just for this moment? If there's anything WP has to worry about it's the newer anti-GMO SPAs that cyclically crop up much the anti-vaxers do. Capeo (talk) 16:08, 18 May 2016 (UTC)
The irony is that we see these aspersions all the time from those trying to dispute the scientific consensus referencing "corporate interest", but they never seem to be engaging in this inappropriate behavior against corporate on the other side of the issue. That includes the organic, "natural" foods, etc. lobbies that have been actively campaigning and pushing often shoddy research to claim the food is unsafe to create marketing to sell their own products. There's a lot of money on that side too (not exactly David and Goliath if you pit the two groups of companies against each other) That's why at least editors keeping clear of behavior issues are sticking to scientifically reliable sources that reflect the consensus of academics and focus on content instead of resorting to shill gambits, erroneously claiming the research was just bought off on one side, or raising these boogeyman specters to try to sway content.
Editors really shouldn't be complaining that they can't interject this stuff into the RfC anymore. It's like a kid on the playground complaining because they got caught putting ice chunks in their snowballs when no one else was even throwing even regular snowballs. The situation would be worlds worse if editors like myself started slinging aspersions that editors disputing the consensus are favoring the organic industry's interests. I've seen no evidence of anyone on any side having that kind of COI or advocacy connection, and I'm not aware of anyone like me who's been characterized as the "pro-industry" side (that irony astounds me as an equal opportunity scrutinizer of industry) going anywhere near claiming that. It's interesting what "side" the chaffing at our ArbCom ruling on aspersions solely comes from in this dispute with that in mind. I'm hoping the ArbCom supervised RfC helps excise that distraction so we can solely focus on content, and the current admins seem to be doing a good job of weeding out distractions. Kingofaces43 (talk) 19:30, 18 May 2016 (UTC)
Luckily it appears the admins watching the RFC draft page have already dealt with this appropriately. Further shill gambits will be removed and editors repeating them will be removed as well. Capeo (talk) 22:13, 18 May 2016 (UTC)
The problems here should not be construed in terms of COI. What does happen from time to time is that a COI editor edits in some dark corner of Wikipedia with little scrutiny leading to a POV article that can exist for many years. Then this is discovered by the wider Wikipedia community, people with the opposite agenda jump on that fact to try to get their way by invoking COI as a systemic problem even when there is a lot of scrutiny. But that's a false assertion. It's impossible to get away with one biased, one sided POV edits in a controversial topic with half of the community looking over your shoulders. The real source of the dispute lies elsewhere. The core of the dispute is that the anti-GMO activists evaluate things by assuming a priori that every aspect of GMO technology is extremely dangerous and then update this prior assumption using scientific evidence. Mainstream science, in contrast, will base their prior assumption on the totality of all of the prior existing evidence, here some safety margins will be build in, so there is then still a bias toward assuming potential danger, but it's not as extreme what the anti-GMO movement is doing. Count Iblis (talk) 17:10, 18 May 2016 (UTC)
  • It would fun to see a fork of some GMO articles written by David alone on his userpage. A rigorous peer review of such articles by uninvolved editors and David's reply to that should keep the whole lot busy for quite a while. But this would at least focus all the discussions on the actual topic. Count Iblis (talk) 01:21, 19 May 2016 (UTC)
@Count Iblis: I have done this at Talk:Genetically modified crops#Fourth proposal. I welcome anyone to comment there on what I prepared. I hope the experience is as fun as the Count predicts. --David Tornheim (talk) 05:15, 20 May 2016 (UTC)
Wow, very revealing. TimidGuy (talk) 11:16, 19 May 2016 (UTC)
@TimidGuy: Indeed. Here is another good example of similar editing, where language pointing out that the U.S. and other countries have no special regulations for GMO food was removed, and replaced with language claiming that it is tested on a case-by-case basis (like the E.U.). There is ample RS showing the huge difference between U.S. and E.U. approaches to GMO's. Any attempt to put that many countries mandate labeling and/or ban GMO's in a conspicuous place is summarily deleted from the GMO food and crops articles and kept buried. Why? The answer is here: [11] [12]. These countries supposedly hold "fringe" views. Is it true the leaders of the E.U. are no better than flat earthers? --David Tornheim (talk) 04:02, 20 May 2016 (UTC)
Indeed, and rather ironic too considering Petrarchan linked to one of their poor personal interpretations of sources in question rather than something that should make people shudder about me. At the end of the day, Petrarchan's interpretation of the source didn't gain traction in consensus discussion. The short of it was that the paper I was citing discussed major analysis errors in fringe sources the source directly analyzed and also mentioned it was a problem overall for people trying to dispute the consensus. When claims for anything in science are found to be primarily based on flawed research (e.g. vaccines and autism), that isn't a left-field criticism. That's especially when other fringe sources supportive of the no consensus idea uncritically accept that research without checking basic things like that. Anyways, I don't really plan to hang around this page too much more, but I figured I'd fill people in a bit on what we've been dealing with (and address personal attacks) as we're always looking for more eyes in the topic. Kingofaces43 (talk) 16:06, 19 May 2016 (UTC)
@Petrarchan47: I don't see that edit by Kingofaces43 as any sort of smoking gun. It actually seems pretty sensible. When a paper talks about "statistically unaccounted multiple comparisons", it is implicitly talking about statistical significance. Multiple comparisons are a problem because they can lead one to report a statistically significant effect where none actually exists. Unaccounted multiple comparisons are, in fact, a flaw in statistical methodology, as Kingofaces43 summarized. This isn't a "misrepresentation" by Kingofaces43; it's an indication that he has a decent level of statistical sophistication. (We have an article at multiple comparisons problem with more detail; it is marginally less impenetrable than our average statistics article). In this topic area, there seems to be a very powerful, and counterproductive, tendency to immediately assume that any edit that one disagrees with (or simply doesn't comprehend) is motivated by malice or undisclosed conflicts of interest. MastCell Talk 16:22, 19 May 2016 (UTC)
Here is the edit in question. I'm not concerned with the statistical significance part, sorry to waste your time refuting that. To be specific, what I consider a highly egregious misuse of this site and misrepresentation of science is:
  • Some studies have claimed harm from GMOs...but review of these studies show the statistical methodologies were flawed and do not show evidence of harm."
Take a look at the studies that found harm, and the piece used to refute all of them, with MEDRS in mind. This is indefensible; King should not be anywhere near the GMO suite after such an edit. petrarchan47คุ 03:07, 20 May 2016 (UTC)
Above, KingofAces43 identifies the sources of Domingo[1] and Krimsky[2] as "Fringe". Both are REVIEW articles published in peer reviewed journals, the highest standard required of WP:MEDRS. In addition, Domingo's expertise is in the right field to address food safety--toxicology [13]. This expert in toxicology sounds very different than the many published scientists like Pamela Ronald who make their livings inventing and promoting new GMO products, rather than testing them for safety.[3] To equate Domingo with what our WP:UNDUE policy describes as akin to a "flat earth" theory is hard to fathom. The Modern flat Earth societies do not have a single peer reviewed article supporting their views. What source(s) is King relying on to claim that Domingo and Krimsky are fringe?
--David Tornheim (talk) 05:08, 20 May 2016 (UTC)
  1. ^ Domingo, José L.; Bordonaba, Jordi Giné (2011). "A literature review on the safety assessment of genetically modified plants" (PDF). Environment International. 37: 734–742. doi:10.1016/j.envint.2011.01.003.
  2. ^ Krimsky, Sheldon (2015). "An Illusory Consensus behind GMO Health Assessment" (PDF). Science, Technology, & Human Values: 1–32. doi:10.1177/0162243915598381. [E]ight review articles were mixed in their assessment of the health effects of GMOs.
  3. ^ Ronald, Pamela (May 5, 2011). "Plant Genetics, Sustainable Agriculture and Global Food Security". Genetics. 188: 11–20. doi:10.1534/genetics.111.128553. {{cite journal}}: Cite has empty unknown parameter: |1= (help)
I think this concern should be taken seriously, though not perhaps directly as suggested. From conversations I've seen here recently, and my attempts to make small, common sense changes on a few articles, it seems like Wikipedia's "skeptics" have some very sharp elbows. Part of this is that they seem to identify with a community rather than simply holding a belief. They have favorite Wikis and favorite sources. There are people who have articles on Wikipedia primarily for their skeptical activism, and even primarily for their coordination of skeptical edits on Wikipedia! I doubt any large number of them are literally employed by corporations to promote their particular style of rationalism, and probably not many even receive grants toward this end. But there is a feeling, reading some of these sources and articles, that being a Noted Internet Skeptic is something that people can proudly put on a resume, and backing down from a fight is not something they can proudly put on a resume. And this casts a broader shadow of cultural fallout.
In the specific case of GMOs, what we need to bear in mind is that it is not really very difficult to make GMOs that are immensely powerful weapons, capable of bringing down nations. For example, there's no fundamental reason (though I think there are specific technical hindrances) against putting abrin into a common bean, and a few bushels of seeds and satchets of pollen in the hands of a terrorist network could end production of the crop for decades. A basket of such GMOs against its major crops might leave a nation dependent on food imports and international goodwill. So when people add statements that genetically engineered crops are recognized as safe, what they mean is that rationally designed, and honestly approved and tested genetically engineered crops are recognized as safe. A corrupt approval, of course, is no safeguard; what they mean is an honest, thorough approval by a civilized country with strong regulatory resources and effective law enforcement. How honest, how thorough, how strong, how effective? Why, enough to be safe, of course! It's a tautology masquerading as incontrovertible fact. Wnt (talk) 10:36, 20 May 2016 (UTC)
This is stupid. By this reasoning, pillows aren't safe, since you can suffocate someone with one. Water isn't safe; in fact almost nothing is safe.
"Safe" means "it doesn't accidentally hurt people". The fact that something can be used to deliberately hurt people doesn't mean it isn't safe. Ken Arromdee (talk)
The issue might be "unsafe at any speed" such as a technology which allows "just anyone to edit" where it would be very risky, such as in Lua script modules used on hundreds of pages. Plus we have gluten-free diets, after the short Franken-wheat replaced the prior vast "amber waves of grain" where tall wheat had been planted decades ago. How much has gluten increased in modern wheat?-Wikid77 (talk) 17:20, 21 May 2016 (UTC)
@Wikid77: I looked up "Frankenwheat" and was directed to einkorn as an alternative. Some articles on the Web like [14] bragged that because einkorn has a different number of chromosomes it has a different gluten structure (!), and cited a manufacturer of einkorn that was studying whether people with celiac disease could eat einkorn. Following that link, I reached the manufacturer's sales page, with a terse statement that einkorn has gluten and if you have celiac disease don't eat it. [15] I do not claim any understanding or take any side regarding gluten, save to say I never worried about eating it. Wnt (talk) 18:31, 21 May 2016 (UTC)
@Ken Arromdee: People keep telling me never to attribute to malice what can be attributed to stupidity. I am skeptical of that particular dictum, but to such extent as it is true, the malicious possibility implies we should watch out for accidents of comparable magnitude. Not every harmful protein has already been labelled as such in GenBank.
Now as for pillows -- pillows, or at least some pillows, aren't safe. Manufacturers put warnings right on the packaging,[16] after coroners deal with the casualties.[17] Major websites advise against them.[18] Does that mean we should all throw out our pillows? Of course not. But we should avoid blanket statements that they're safe, and recognize that there can be (though not necessarily are) valid reasons to require labelling. The 'skeptical' attitude around here is to pooh-pooh the very idea that a pillow, or a GMO, could be unsafe, dismiss any further thought about it, and treat any consideration of the opposing possibility as a fringe idea. I disapprove of this because this is the well-worn road of nuclear power, for example, where people insisted there was no possible risk right up until the point we just stopped using it because we couldn't trust the industry. I'd rather acknowledge that there's potential risk, acknowledge that "safe GMOs are safe" isn't really much of a promise, but it's a starting point, encourage public discussion of exactly what the GMOs are and what they do, and hope we move onto a mature consensus. On Wikipedia, that can be done by keeping an open-minded article, not buying into every fringe anti-GMO statement but not dismissing them out of hand either. Wnt (talk) 18:22, 21 May 2016 (UTC)
One should apply the same standards to all related topics. You can debate on where to draw the line in general, but this issue is far less of a problem if the same standards are applied uniformly. The problem is that in the real world this isn't always done and Wikipedia then has to deal with the problems this poses. E.g. if we take a very skeptical attitude w.r.t. the safety of cooking oils, refuse to compare olive oil to saturated fats, instead consider if using oil itself is a problem (so we have to compare using oil to using no oil at all), then you get results that point to oil not being safe at all. There is a lot of evidence that a plant based diet where no cooking oil is used reduces the incidence of heart disease to almost zero. This is a far more spectacular result than the "not proven to be 100% safe" objection the anti-GM campaigners have against GMO foods. Yet, olive oil bottles are not going to be labeled any time soon with statements that it can cause heart disease. Count Iblis (talk) 20:15, 21 May 2016 (UTC)
You just proved my point. Industry would strongly oppose such labeling, if the research you cite is accurate. And why is it that consumers get so much misleading information about what is and is not healthy? The FDA entrusted with regulating food and drugs is too cozy with industry.[19], [20], [21] --David Tornheim (talk) 21:19, 21 May 2016 (UTC)
I think human intervention plays an important role. You should be able to sell an apple without having to summarize every scientific paper about apples. But if you sell a cake with an artificial flavored frosting, people have a right to know how much arsenic was used to used to make it. It's the unknown quality of an item that has been altered that brings labelling into play. Now, labelling of cakes with chemicals has political limits - "artificial flavors and colors" - to what degree it should is questionable. There should be certain kinds of GMOs, simple gene deletions, controlled processes of hybridization, that I can definitely picture exempting in this way. But then there are other ideas, that can range anywhere from the equivalent of partially hydrogenated fat to (hypothetically) well beyond arsenic. There are GMOs like StarLink that have been treated as banned; there are others that are controversial. Rather than having somebody else make a yes-or-no decision for us, we could leave it to labelling. Now my purpose here isn't really to argue for labelling; my point is only that we can see this as a political issue with multiple points of view, not just a "rational" point of view and a "fringe" point of view. Wnt (talk) 21:53, 21 May 2016 (UTC)
^I agree. I really appreciate this discussion. I also agree with your assessment of the Sketpic movement's advocacy here on Wikipedia. Indeed what has happened in the GMO articles is that any concerns raised by scientists and doctor's associations get scrubbed or hidden in the article leads with language that misrepresents the disparate views on the subject. The justifications on the talk page for adding concerns in RS are always similar: (1) "fringe" (2) undo (3) advocacy. Any concerns about GMO are attributed instead to a paranoid misinformed public and radical activists groups like Greenpeace.
Industry has recruited scientists and other academics to their PR campaign--because the public trusts scientists and doctors. See [22]. I would hope that our articles would be more NPOV, but if you try to change the lede to be more NPOV and put any mention of dissension among scientists or academics on GMO's, it will usually be reverted within minutes and threats will be issued along with other drama if you don't immediately back down.
Constant bullying and accusations of "fringe" beliefs, anti-GMO activism, being akin to a "climate change denier" and anti-vaxxer are the norm.
If you challenge the name-calling ([23], [24]), not only will you lose, but the act of raising the issue will be used against you as evidence to get you topic-banned [25]. After being topic-banned, that editor asked to be blocked for 6 months and the user's talk page is Retired [26]. It's no wonder that non-involved editors are so afraid touch the GMO pages and disputes. I believe this kind of bullying is why we have fewer editors than in the past. --David Tornheim (talk) 01:41, 22 May 2016 (UTC)
@David Tornheim: Well, looking at that case I see more there than what you say. And the anti-GMO side is certainly capable of all the same clannishness and even careerism as the pro-GMO side. From what I see I wouldn't pick that ground to make a fight - I want to side with people when they are trying to add a wider variety of sources, not take them out. Wnt (talk) 15:17, 22 May 2016 (UTC)
  • The dangers of too much free information: I thank God that many terrorists tend to have poor education (and Plato reasoned that educated people would act more civilized). The 2001 September 11 terrorist attacks required the terrorist pilots to attend flight school to navigate aircraft, and someone had to pay for it (and some U.S. Government officials had to ignore warnings about it). Consider the danger if more intelligent, educated people wanted to cause harm, on a massive scale, for decades or longer. Is there a link between intelligent people receiving visions, or the "gift of prophecy" (book A Gift of Prophecy re Jeane Dixon), compared to hateful people living in darkness about divine powers? One of the recent PowerBall lottery jackpot winners said they received the winning numbers in a dream, or at least that is what was reported. -Wikid77 (talk) 15:30, 20 May 2016 (UTC)

New Idea

Mr. Wales, I think I have a new idea to solve the problem. I don't want to correct the comments myself. But can some one of you help me correct the comments in the talk pages. Then the problem will be solved. Can some one of you please help me improve the comments in the talk pages?Arman ad88 (talk) 21:12, 19 May 2016 (UTC)

In general, users should avoid correcting the talk-page comments of other editors, except perhaps by adding rare editorial phrases in square brackets ("[...]") or citation-needed "{{cn}}" tags, but instead, a reply can be added after an incorrect message of the OP (original post) to refute the text by the opinion of the next editor. Also see: wp:Refactor for guidelines about redacting libelous statements or restructuring of talk-pages. -Wikid77 (talk) 14:40, 20 May 2016 (UTC)

OK I think I have found a new article here. The article is Wikipedia: Talk page guidelines. It is mentioned in the section Editing own comments, point 4, that I can correct the comments by inserting a message in square brackets after the end of the comment. I think it is a legal way to correct the comments. For example if I try to correct a particular comment it will turn to be like this:

Wikipedia is an Internet encyclopedia supported and hosted by the non-profit Wikimedia Foundation. It is a free-of-cost encyclopedia with its articles being free-content; those who use Wikipedia can edit almost any article accessible. Wikipedia is ranked among the ten most popular websites and constitutes the Internet's largest and most popular general reference work. [Corrected]

So can I correct the comments this way?Arman ad88 (talk) 13:17, 21 May 2016 (UTC)

Wait, the plan to alter another editor's post, and merely append "[Corrected]" afterward, is likely to cause severe arguments. Instead, try to add replies which refute a post where you disagree, in the talk-page. The use of bracket-text "[...]" is mainly for editorial notes, such as noting a key misspelled word, as "Beetooven" by putting "Beetooven [Beethoven]" with the bracket-text inserted immediately after the word, but not rewrite a prior editor's post to reword (nor copy-edit) their text. Some users have even complained when the spacing in their messages was changed over 2 years later to better fit mobile-phone views. Avoid changing any other user comments. -Wikid77 (talk) 14:07, 21 May 2016 (UTC)
It is actually stated at WP:Talk "Never edit or move someone's comment to change its meaning, even on your own talk page. If something needs correcting, I usually leave a message "I think you meant...". Most editors than change their own post, or make a fresh posting. DrChrissy (talk) 15:26, 21 May 2016 (UTC)

Well I will tell you to look at this section Editing comments in the talk page article. I think there are some conditions given there for editting the comments. Look at the 18th point. It is mentioned there that "If you have his permission". It means "you can correct the comments if you have his permission". I think this is the main point that allows you to correct the comments. I think this is the proof that you can correct the comments.

I am giving you the permission to correct the comments. I have no objection if someone of you can correct the comments. You don't have to correct the other editors' comments. There are no problems in that. Just my ones needed to be done. I want help from some one here to correct my comments in the talk pages.Arman ad88 (talk) 19:24, 21 May 2016 (UTC)

Please give us an example. I am (moderately) sure that you will get help. All the best: Rich Farmbrough, 23:45, 21 May 2016 (UTC).

OK these are the talk pages I want to correct: Sasanian Empire, Parthian empire and Achaemenid empire.Arman ad88 (talk) 05:13, 22 May 2016 (UTC)

@Arman ad88: I looked at Talk:Sasanian_Empire and didn't anything that needs correction. What, specifically, concerns you?--S Philbrick(Talk) 15:34, 22 May 2016 (UTC)

Mr. Philbrick, I am not so good in English. I cannot express myself clearly in English. I think the message should be like this:

I think the map given in the infobox in the article is wrong. If you look at the map, it looks like Turkey and Egypt were parts of the Sasanian empire, which were not. It may confuse the readers. I wanted to place my map in the infobox. And place the current map in the later section Second golden era.

The next message should be like this:

I don't see any problem with the military routs of Ardashir and Shapur. Those two kings founded the Sasanian empire. So they are very much relevant with the map. The problem with the map is that, it gives the impression that, Turkey and Egypt were parts of the Sasanian empire, which were not. This is actually an exagarration of the original map. I think this is not the right place for the map to be placed, it should be placed in the later section Second golden era. Remember emperor Khosrau II coquered those territories but could not hold it on for so long.

This is how I want to correct the messages.Arman ad88 (talk) 18:49, 22 May 2016 (UTC)

Search engine hiccup

First, a quick word thanking whomever at WMF is responsible for the newish change to the internal WP search engine that pops up a small list of pictures together with names and a short descriptions in a list when a person is trying to locate a page on WP. It is very helpful. However, a small weirdness popped up when I ran a quick search to get to the page I'm working on, Vera Figner. The search engine gives a portrait, a name, and the description "Russian writer." Now, the narodnovoletsa Vera Figner certainly is the author of a famous memoir, but calling her a "writer" as a fundamental description would be like calling Donald Trump a "writer" because he wrote The Art of the Deal. She was, first and foremost, a political activist. Second descriptive would be political prisoner. Third descriptive would be memoirist. But not a writer.

My question is this, and hopefully somebody can answer it: how do you change the quick description given by the search engine? Where does that information come from? The word "writer" is not used in the lead of the Figner article (although she is justifiably included in a few writer-related categories). It has to come from somewhere... Where? Carrite (talk) 18:00, 22 May 2016 (UTC)

Wikidata, I presume. See d:Q266783. BethNaught (talk) 18:20, 22 May 2016 (UTC)
You presume correctly, BethNaught. :) Carrite, thank you for the feedback. I'm glad search is getting better for you. I'll make sure to pass it along to the search team. CKoerner (WMF) (talk) 03:09, 23 May 2016 (UTC)
@Carrite: I see you fixed the "occupation" field, but I believe the field that was causing the problem was the description itself, so I've gone and changed that to "Russian political activist." If that's lacking, obviously you can go ahead and tweak it. — PinkAmpers&(Je vous invite à me parler) 04:01, 23 May 2016 (UTC)

Correction

Hi, I will tell you to look at this section Editing comments in the talk page article. I think there are some conditions given there for editting the comments. Look at the 18th point. It is mentioned there that "If you have his permission". It means "you can correct the comments if you have his permission". I think this is the main point that allows you to correct the comments. I think this is the proof that you can actualy correct the comments.

OK, then, I am giving you the permission to correct the comments. I have no objection if someone of you can correct the comments. You don't have to correct the other editors' comments. There are all right. Just mine. I want help from some one here to correct my comments in the talk pages.Arman ad88 (talk) 19:24, 21 May 2016 (UTC)

Please give us an example. I am (moderately) sure that you will get help. All the best: Rich Farmbrough, 23:45, 21 May 2016 (UTC).

OK, I am mentioning here. These are the talk pages I want to correct: Sasanian Empire, Parthian empire and Achaemenid empire.

Hi anyone there? Can anyone help me with the correction of the comments. You have to just improve the Enlish in my comments. You have to just correct them in a way that, they don't look odd in the eyes of the other people. That's it. Just that will be enough.Arman ad88 (talk) 05:19, 24 May 2016 (UTC)

Would it be satisfactory to you if someone simply deleted your comments? It is a pretty uncomfortable thing to edit someone else's remarks even if just to correct the English. Also, just FYI, in my experience native English speakers are very very tolerant of unusual or mistaken variants of English, because it is such a popular second language. Particularly at Wikipedia and particularly on talk pages, no one should think poorly of someone for being a non-native English speaker. So another alternative for you is to just stop worrying about it.--Jimbo Wales (talk) 07:13, 24 May 2016 (UTC)
Jimbo is correct, this was not worth worrying about. However I have tweaked your comments on those pages. All the best: Rich Farmbrough, 10:16, 24 May 2016 (UTC).

Mr. Wales, I want you to delete my comments. I don't want anyone to see my comments. I think this is the best way to solve the problem. So please help me deleting the comments.Arman ad88 (talk) 14:46, 24 May 2016 (UTC)

OK Mr. Wales, thank you very much. Thank you very much for correcting the comments. I think the problem has been solved.Arman ad88 (talk) 20:52, 24 May 2016 (UTC)

One last thing

Mr. Wales, one last thing. I want you to please remove all the comments from my talk page. From my original talk page ( This talk page ). There are too many negative comments. It will be really very humiliating for me, if someone see them. So I want you to please help me remove the comments. Then all the problems will be solved.Arman ad88 (talk) 20:52, 24 May 2016 (UTC)

I rarely understand your requests, Arman ad88. Why don't you delete the comments yourself? You can take care of that immediately. Liz Read! Talk! 21:19, 24 May 2016 (UTC)

No, there is a problem. My original account is actually blocked. If I try to remove my comments from my talk page, there is a great possibility that the other editors may block my account. I just want to correct my comments and then I will leave Wikipedia.Arman ad88 (talk) 22:10, 24 May 2016 (UTC)

OK thank you very much. Thank you very much all for helping me out with the corrections. All my problems have been solved. I am going to leave Wikipedia now. You can block my account now.Arman ad88 (talk) 22:43, 24 May 2016 (UTC)

Fine, the original page, User_talk:Arman_ad60, has been blanked earlier, and could be restored later if wanted. -Wikid77 (talk) 22:46, 24 May 2016 (UTC)
Wait, you say your original account is blocked. Are you evading your block using a new account? --Kinu t/c 23:06, 25 May 2016 (UTC)
Actually, to correct and expand on my previous statement, it's pretty obvious that this new account was to get the original account's edits refactored (as discussed in the previous section) and to get the original account's talk page courtesy blanked. For example, see also this from a previous sock. It's happened before, but now that it's done, perhaps the socking will stop. I don't see a problem with either of those actions, but I don't see any reason to allow this block evasion to continue without substantive discussion of the original block (i.e., via UTRS), so I've blocked indefinitely. --Kinu t/c 23:23, 25 May 2016 (UTC)

Hi Jimbo, do you or anyone else in WMF have any contacts within the United Nations? Per the discussion at Commons:Commons:Deletion_requests/Template:PD-UN-map we need to update a decade-old OTRS authorization if we are not to delete close to 1,000 high quality maps from across the encyclopaedia. Oncenawhile (talk) 07:57, 25 May 2016 (UTC)

I think some people there are making absurd claims that you can't CC-license things they admit that you could copyright and sell. But the high-level WMF response should go in a different direction entirely. The real bottom line here is that Wikipedia maps suck, and they would suck even if there weren't a crowd of people looking to ban what little we have. They are picked out of random PD publications and cover whatever they happen to cover. What WMF should do is create and anti-patent an AI map pirater!
The way the Pirater should work is this: the user uploads a copyrighted map he has no rights to as input to the program; this upload, viewed by no human being, is a transformative use. The program has access to a huge library of OpenStreetMap type elements. By visual recognition, perhaps aided by some annotation keyword matches, it automatically identifies the rivers, mountains, roads, national borders, canals, cities, shaded national territories, parks, and myriad other elements on the map. It then returns those items to the user as a list of items in text format - text that says you have corners at the given coordinates and which roads, rivers, territories etc. are on the map. It also returns as a transformative work, for academic use, to the user who uploaded the copyrighted map a "diff" that shows what elements on that map remain after all the items listed are removed from it. It is up to the user to create new free-licensed definition files that describe those features, or else they simply get omitted from the final output map. In some cases of course the maps will simply disagree; in this case the open definition should either prevail or be disputed with an alternative, which the user is also responsible for specifying, if the user really believes the river or city in that time period was in a different place or whatever.
It should however sometimes be possible to automatically specify the element. For this the user might upload several independent copyrighted sources to the program, after first highlighting the feature, and then the program develops from these a consensus profile for where that river, ancient national border, etc. is/was, and output that as a text file with coordinates for that curve which is ready to be automatically submitted to the open database.
If WMF hammered on this with some good programmers, it would be possible for Wikipedia editors to essentially recreate, by a demonstrably unoriginal method from public source data, maps corresponding to whatever maps we happen across, and modify them readily to cover a larger or smaller area, use consistent color schemes and all those lovely little details. It's not easy but if your computer people want a way to earn their chops, I think this would be a good one. Wnt (talk) 18:41, 25 May 2016 (UTC)
You could do the same thing for graphs, potentially, removing all the creative works of authorship and leaving only unadorned minimalist representations of the data, preferably in textual SVG format for internationalization and localization. EllenCT (talk) 19:18, 25 May 2016 (UTC)
See http://www.publicdomainsherpa.com/public-domain-maps.html and Category:Public domain.
Wavelength (talk) 19:34, 25 May 2016 (UTC) and 22:36, 25 May 2016 (UTC)

The maps were Kept. It really did strike me as a bizarre deletion request, but my opinion of Commons administration is now improving. Smallbones(smalltalk) 13:53, 26 May 2016 (UTC)

Surprising that you'd be able to muster an improving perspective on Commons, after that monumental beat-down the community gave you over there, over your inferior photo of the Parkway House. - Clearly clear (talk) 02:52, 27 May 2016 (UTC)
@Clearly clear: Hey now, no personal attacks. Comment on content and not on the editor. Amccann421 (talk) 03:18, 27 May 2016 (UTC)

The new bot-generated backlog list WP:POPULARLOWQUALITY is working. It shows the top 1000 articles by pageviews with Stub-, Start-, or C-class ORES article quality predictions. ORES is extremely cool; it's so much better than inscrutable neural networks because it can show the article revision's features on which it bases its prediction of an article's revision. I have no idea how that was set up, but it clearly didn't discover the features automatically because they have meaningful names (which neural networks never produce.) It uses Python's SciKit-Learn Random Forests classifier. EllenCT (talk) 05:57, 23 May 2016 (UTC)

I assume it's an artifact of how we do times. A plane crash occurred at 5:07 UTC+03 - and the readers will click because WTF is UTC, I heard of GMT, but whazzis, the Ukrainian Territorial Clock? and what is that in Eastern Daylight Time? And so you check to see, and it shows up in the stats. I'm sure I've done it myself ... multiple times. Wnt (talk) 11:53, 23 May 2016 (UTC)
In this case it's apparently transient spikes in article popularity probably caused by someone's automated article downloads, possibly used for testing or demonstration purposes, and similar to the way Google Doodles often make obscure historical figures tops in weekly popularity, except in these cases it's doubtful anyone is actually reading the articles. There is some R code on pp. 191-2 of Xu et al. (2014) which is good at measuring the modality of time series. That could be used to filter out articles which are only popular because of such transient spikes. I will have to figure out who on the Pageviews API team may be able to help with that. I'm not entirely sure we want to discard Google Doodle-caused popularity to get rid of all the transient popularity spikes, but for these purposes there are so many more of the latter that it would be best to favor the more organic article popularity of e.g. Sangeeta Bijlani instead of articles with more transient popularity. EllenCT (talk) 14:11, 23 May 2016 (UTC)
Asked at Wikipedia talk:Web statistics tool#Filtering transient spike anomalies. Those interested in economics may appreciate that central banks face an analogous problem when trying to derive core inflation from volatile consumer prices.[27][28] EllenCT (talk) 17:16, 23 May 2016 (UTC)
Re. "ORES is extremely cool" – disagree. It appears to be expert in singling out articles that consist mainly of templated content. E.g. No. 54 has over 10 sections that each fill more than a screen on my computer, it has over 200 references, and covers its subject area more than satisfactorily. Was anybody interested in "popular pages with a lot of content in templates"? "POPULARLOWQUALITY" is definitely a wrong name for the machine-generated list and should be changed ASAP, as it is a misleading name... at the very least until ORES knows how to filter out pages that derive the bulk of their content from templates. --Francis Schonken (talk) 12:49, 23 May 2016 (UTC)
Any machine learning-based assessment of natural language is going to have errors. I agree that the revision to which you refer is certainly incorrectly predicted as Start-class, and we can see the features used for the incorrect prediction, and report it as a bug to the ORES team. @EpochFail: any ideas why this happens? Why does the feature list for that one say it has only 10 "ref-tags" instead of 217? It is not the only example on the list, e.g., see also this 71 reference List of Game of Thrones episodes revision in which only 10 ref-tags are detected. EllenCT (talk) 14:29, 23 May 2016 (UTC)
Re. "Why does the feature list for that one say it has only 10 "ref-tags"?" – simple, as implied by my earlier comment: THE REF TAGS ARE IN THE TEMPLATES, and ORES doesn't detect these. For now the whole thing should be taken off-line ASAP (including a myriad of links to the "popularlowquality" page inviting to edit the articles in that list). This is extremely counterproductive: pages that need stability for their high visiting rate, and get that stability from templated content should not be pointed to in a sense of hey, go edit that page, you'll be improving the encyclopedia! --Francis Schonken (talk) 14:32, 23 May 2016 (UTC)
You are suggesting that editors who take the time to voluntarily look through a backlog list are going to have a substantial amount of time wasted by obvious classification errors? Won't the errors be immediately obvious to the editors as they look through the list? Isn't that analogous to deleting articles with errors instead of leaving them for people to fix? If these uses make such ORES errors more obvious, then they will be corrected sooner. I would like to give the ORES team a chance to see if they can improve. DataflowBot produces the report daily from fresh calls to the ORES and Pageview API, so let's wait a while and see if it gets better. EllenCT (talk) 14:39, 23 May 2016 (UTC)
My point is: no way this should be presented to volunteers, until the false positives (and that includes nearly all items on the list) are substantially reduced. --Francis Schonken (talk) 14:46, 23 May 2016 (UTC)
I will add a warning message to the top of the list. EllenCT (talk) 14:49, 23 May 2016 (UTC)
Temporary measure – please try to get this sorted before the next automated update. --Francis Schonken (talk) 14:41, 23 May 2016 (UTC)
Sheesh! How do you expect to find other bugs if you delete the entire list? There are false positives in our categories for articles needing citations, too, because e.g. people often miss citations for a statement covering multiple statements at the end of a paragraph. Should we delete those categories until that gets sorted out? EllenCT (talk) 14:46, 23 May 2016 (UTC)
My problem is the page is linked from a template that is on hundreds of pages, and the page name doesn't cover its content. The message is: this is low quality stuff, please go hack away on these pages, which are in fact pages that need a high stability. this is the permalink for the developers. --Francis Schonken (talk) 14:49, 23 May 2016 (UTC)
Unless you think the warning message added is insufficient, I feel like [29] and [30] are letting the perfect be the enemy of the good, which is not exactly Wikipedia's underlying philosophy. EllenCT (talk) 15:08, 23 May 2016 (UTC)
I rarely if ever agree with Francis Schonken, but concur entirely with his actions here. This isn't "letting the perfect be the enemy of the good", since that presumes "good"; this list has no apparent relation to reality. (Serendipodous, who compiles the weekly traffic reports for the Signpost, can testify just how little relationship a bot-generated traffic report has to what people are actually reading.) View the raw traffic report through the prism of a wonky article-assessment bot which can't parse wikicode transclusions and your results are going to be meaningless for all practical purposes. ‑ Iridescent 15:20, 23 May 2016 (UTC)
Hey folks! Thanks for the ping and for reporting issues in classification. It's great that you have flagged some obvious bugs with ORES' feature extraction process for article quality. It's important to note that the machine classifier is *highly accurate* in general. See the statistics here: m:ORES/wp10#enwiki Still, it's clear from these examples that there are some special cases where ORES fails to make a good prediction. Your reporting of these issues is awesome because *we can fix it*! I encourage you to see this as a bug to be fixed rather than a fundamental flaw with ORES.
There are a few issues that I should clarify. 1. ORES does not predict "list quality" -- just regular "article quality", so you'll get weird predictions for list articles. This is by design. We'd want a separate model for predicting list quality. 2. ORES' article quality model *currently* operates on the raw wikitext of an article as real content rarely comes via transclusion. Still, this is totally fixable. We'll likely have performance issues working with transclusions (so does MediaWiki), but we can run some experiments and see what happens.
If you can help me by producing a list of articles that are clearly misclassified with any notes about trends that you see, we can fix the models. This is a good and necessary pattern for making sure that the prediction models work in practice. Given what you have already showed me here, I have filed a couple of tasks for us to dig deeper. See Phab:T136005 and Phab:T136006. --EpochFail (talkcontribs) 15:44, 23 May 2016 (UTC)
@EpochFail: thank you; I think ORES is awesome and will be even more awesome if the community has a chance to help improve it. I will be happy to create such a list, but would you please count the ref-tags feature after template text substitution first? I'm pretty sure that will solve the lists problem. In the mean time, do you know who on the Pageviews API to ask about suppressing transience? Wikipedia:Web statistics tool links to mw:Analytics/PageviewAPI which is blank. I think they can apply an algorithm substantially less complex than the general purpose one in the R code on pp. 191-2 of Xu et al. (2014) to get 90% of those eliminated, and I know from personal experience doing the Signpost's traffic report, they waste a lot of time. EllenCT (talk) 16:03, 23 May 2016 (UTC)

(ec) I've been a bit fast in my above communications on this page, that leads to stern language:

  1. What I've seen appearing on this page on the topic of quality improvement led me to think that if we want to maximise benefit per edit ratio then naturally our top priority should be to keep high traffic pages outstanding. So I'm partial in favour of providing lists of high traffic pages that need attention. @EllenCT: kudos for your efforts, I really support what you're trying to do here.
  2. I've done roll-outs of software gizmos: if the gizmo isn't in good working order at the time of rolling out on a more or less broad scale, people will look at it, find it not doing what it should do, and you've lost those who saw it in that so-and-so state for a very long time: most of them will not return ever to the gizmo. It is very difficult and time-consuming to modify a first impression (whether that first impression is good or bad), that's the general fact behind it. The page is linked now from a handful of individual pages (no longer from the templates that are on hundreds of pages): let's consider those who find it now as the test public (maybe provide a link to this talk page section here for early adopter comments). That's why I was rash in my actions, leading to fast comments. --Francis Schonken (talk) 15:55, 23 May 2016 (UTC)
No problem! Most of the underground lair minions have advanced training in conflict resolution skills, and often remember to apply them. EllenCT (talk) 16:08, 23 May 2016 (UTC)

@EpochFail: if you have time, could you please help us understand how ORES scores this revision of "Standard deviation" with a Good Article probability of 4%? EllenCT (talk) 14:01, 24 May 2016 (UTC)

I'd love to hear the answer - sounds a bit high to me.--S Philbrick(Talk) 15:25, 24 May 2016 (UTC)
It's only predicted as B-class at 33%, when that's what its talk page says it is. Ah, but looking closer that's the WikiProject Math assessment, while the WikiProject Statistics assessment is correctly predicted as C-class; nice! I would agree it has far too much use of the <math> tag. EllenCT (talk) 16:07, 24 May 2016 (UTC)

Third update

  FYI
 – For the more technical aspects of this bot implementation see Wikipedia:Bot requests#Plot pageviews against ORES wp10 revision quality scores for popular articles?

Commenting on the POPULARLOWQUALITY's third bot update:

  • Nos. 2 and 7 are disambiguation pages, shouldn't be listed
  • No. 42 is about a future event, decently written, decently referenced (some 50 references). Can't see what is wrong with it. Indicating it as "low quality" seems like an invitation to add more speculation: a future event is more or less all speculation until concrete preparation of the actual event has begun.
  • No. 4 looks like a fairly standard article on a recent sports event: mostly listing the tournament's finalists and winners. There's only one reference, so there's certainly room for improvement, but "low quality" no way.
  • No. 65 is a bio of someone known for only two things (being an O. J. Simpson lawyer, being an ancestor to the later TV-personality Kardashians): with only one of those two the bio maybe wouldn't even have passed WP:ONEEVENT – notwithstanding that sparsity on what to base the bio on it is decently elaborated, with some 25 references. Certainly not "low quality".

@EllenCT and EpochFail: I'm beginning to doubt whether the approach is useful. If the bot can't be taught to distinguish between average quality and low quality this is an approach with little net benefit afaics. Even "mediocre" quality or "questionable" quality wouldn't apply to most of the articles listed by the bot, and these are certainly not identical to "low quality". The name of the page should not accuse them all of being low quality (unless that is what the bot can deliver): "low quality" as it stands is insulting to the Wikipedians who put their efforts in it. As it is now the page should be renamed ASAP to "popular and potentially with quality issues".

I've extended my window in giving this a chance: as explained above I wouldn't have returned to it if it hadn't been so that theoretically I thought this may have been a promising approach (and if I wouldn't have started commenting here). For me, and in all clarity I only speak for myself, the window for returning to this maybe one or two times before my "first opinion" is formed is closing soon. --Francis Schonken (talk) 09:01, 25 May 2016 (UTC)

How much time do you think is reasonable to allow improvements? We've already eliminated the entirety of the hundreds of articles with transient popularity spikes, the template-based lists, and addressed other issues, but that a handful of problems out of several dozen listings remain isn't improving fast enough for you? I agree your remaining concerns are valid, and can be easily addressed by showing only articles ORES predicts as Stub- and Start-class from, say, the top 100,000 articles instead of the predicted C-class articles from the top 1,000. However, that is going to take some time to accomplish because the Pageviews API doesn't currently produce more than top-1,000 lists at present. May we please have the benefit of your patience while we try to figure out how to work around that limitation in an efficient manner? Is there language we can add to the top of the page that would assuage your concerns about representing generally acceptable quality C-class articles as low quality until we can show just Stub- and Start-class predictions from the top 100,000? EllenCT (talk) 12:07, 25 May 2016 (UTC)
Re. "How much time do you think is reasonable to allow improvements?" – forever, that's not the problem. Neither is going live with an experimental tool. State that it is experimental, and give a link to the workshop for improving the tool. Above I suggested "... provide a link to this talk page section here for early adopter comments ...".
I see no such link (not to here, nor to anywhere else). Here's a proposal for replacement of the boilerplate note:
(replace the link in the last word to whatever seems most useful). Try to take the role of a volunteer arriving at that page, who doesn't know, nor is interested in the mechanics of how the list is generated, just wants to help. Currently there's only "I might be more helpful elsewhere", and click to somewhere else without the developers knowing why the visitor didn't follow through and actually started improving the articles. I checked a few edit histories of listed pages, and didn't see much change to them in the last few days, so maybe also that is an indication that volunteers don't follow through with actual improvements of the pages.
Simply with the word "experimental", the suggestion that any visitor can be helpful, and a clickable link to where to be helpful, you'd draw at least my attention for many more weeks. --Francis Schonken (talk) 12:36, 25 May 2016 (UTC)
  • Struck the first suggestion in my list above, which has been addressed in the latest update. Nos. 42, 4 and 65 of the list above have become Nos. 50, 3 and 69 respectively in that latest update. I understand that further issues are in the process of being addressed, hence placement of a link to the BOTREQ page under the subsection header. --Francis Schonken (talk) 09:55, 26 May 2016 (UTC)

I would point out that the old pageviews API had a raw top-5000 and the new one only offers the top-1000, which is the epitome of reduced functionality. There are a lot of applications of a more substantial top list, so I certainly hope the top Foundation leadership is on board with not backsliding in functionality. Those interested in economics will appreciate that the public sector faces a similar problem when deciding on the cut-off points for transfer payment benefits. EllenCT (talk) 13:14, 26 May 2016 (UTC)

I am happy to see that there are raw pageview data that include hourly 220+ MB (uncompressed) data sets with human-only filtered pageviews for all articles. Many thanks to the Analytics team and User:Milimetric_(WMF). EllenCT (talk) 16:29, 27 May 2016 (UTC)

The Wikipedia age: Peter Thonemann on The all-conquering Wikipedia?

Maybe, we all love/hate it. At any rate, will this "monopoly", finally force experts to get on board. [31] Peter Thonemann gives an interesting take including:

An army of anonymous, tech-savvy people – mostly young, mostly men – have effortlessly assembled and organized a body of knowledge unparalleled in human history. “Effortlessly” in the literal sense of without significant effort: when you have 27,842,261 registered editors (not all of them active, it is true), plus an unknown number of anonymous contributors, the odd half-hour here and there soon adds up to a pretty big encyclopedia. . . Given the manner of its compilation, the accursed thing really is a whole lot more reliable than it has any right to be...No doubt this particular case does not matter all that much. But it does illustrate some of the benefits, and many of the perils, of the brave new world of crowd-sourced online reference. Wikipedia did, eventually, get to the right answer, but it took its time about it. One of the main worries about Wikipedia is not that its content does not improve over time (it clearly does), but that it gets ­better so much more slowly than anyone would have predicted back in 2006 or 2007. . . Instead of grumbling, perhaps we ought to spend a bit more time editing Wikipedia ourselves."

(Thonemann, Peter, THE TIMES LITERARY SUPPLEMENT, MAY 25 2016)

Hopefully, not "all" experts or else, who but the non-experts will write about Wikipedia from the outside? Alanscottwalker (talk) 13:49, 27 May 2016 (UTC)

' “Effortlessly” in the literal sense of without significant effort' - I personally disagree with that. As far "One of the main worries about Wikipedia is not that its content does not improve over time (it clearly does), but that it gets ­better so much more slowly than anyone would have predicted," I wish I'd said that myself. The odd thing about it is that it seems that new articles, in their very first year, are on average better than the average old article (I should check this in more detail). Smallbones(smalltalk) 15:20, 27 May 2016 (UTC)
Oh, I think he is right in the aggregate (and really, you can improve Wikipedia in just a few minutes). -- Alanscottwalker (talk) 15:36, 27 May 2016 (UTC)
The more power Wikipedia has, the more vulnerable it becomes. We already have people talking about the "toxic community" in the discussion below and paid editing in the discussion above, and the more authoritative Wikipedia becomes, the worse these problems get. But even beyond the relevantly benevolent agenda of spam and bias that we face there, we have the deeper existential crisis of the U.S. sliding into increasing economic inequality under a command economy based on intellectual property.
The situation with Peter Thiel funding a third-party lawsuit against Gawker may well represent the shattering of an innocence - one in which we hoped that the very wealthy would, for whatever reasons, be more or less quiescent working only via paid intermediaries for small amounts of money. Under that innocence, Wikipedia imagined it could follow U.S. law and be safe; but this is no longer a country of laws, but a country of oligarchs.
Of course, Wikipedia can put up a substantial defense against attack - the question is whether its leaders take more after Poland or the advisors of Czechoslovakia. Do you roll over quietly when an oligarch tells you he wants this and that taken out of his article? Do you do it again when he comes back and complains his company needs better treatment, and you should stop featuring his opponents, and crack down on gay propaganda?
The only way to avoid apocalypse is to be ready, at a moment's notice, to go to Defcon 1, turn every key, launch every nuke - and this requires also a plan to survive after the war is over. The Internet was such a plan, because it was to be decentralized; but for that to work Wikipedia must actually be decentralized, with the mirrors already up and under independent authority, so that even if a truly terrible court order is filed today there is someone able to continue operations tomorrow. And part of that involves being ready to give up the administrative monopoly and the power it holds, particularly the power to conceal "deleted" contributions. Wnt (talk) 11:45, 28 May 2016 (UTC)
Of course, as Wikipedia gets more successful, the work gets harder and the opponents of a decentralized NPOV have more to aim at. Decentralizing administration is as easy as creating a fork. Some of that is out there already, but how much would a full scale fork in Canada or Sweden cost? Of course much of that could be raised thru online donations, but that starts the very beginnings of a potential conflict with the WMF. Different administration also means that there would be different policies, or at least different people administrating them - so it wouldn't be the "same Wikipedia."
Smallbones(smalltalk) 12:06, 28 May 2016 (UTC)

Focus on pages with technical errors or disputes

As we have discussed for years, the Wikipedia community tends to be "self-correcting" to quickly fix most problems in popular pages, except for technical issues (such as template errors) or content disputes (almost nitpicking or pettifogging over technicalities of the wording). Meanwhile, as seen by the pages in the list wp:POPULARLOWQUALITY (as pages with actually fairly high-quality content), each page when viewed many times, will attract enough editors to soon update the problems as quickly as general readers increase the pageviews. The real problem, where editors should focus extra attention, is to fix obscure template errors, technical details or calculations, or help resolve content disputes (often a case of nitpicking what to say in a page). For example, those 200,000 glaring red-error cite messages need to be fixed in popular pages (to remove "Check date values in: access-date") for wp:CS1 cite date-format errors, in the top pages among searching 28,000 various pages. The main goal is to find major technical problems in popular pages, rather than worry about general quality levels, which tend to be self-correcting by pageview level, as long as any technical or nitpicking content-dispute problems are fixed by special attention. I guess the general rule is: Nitpicking about issues, format or template data is the greatest source of glaring errors in articles. Long term, we need to reduce the nitpicking, to allow more viewpoints, and autofix template data, but the nitpickers tend to argue about the need to nitpick the issues. -Wikid77 (talk) 00:06, 25 May 2016 (UTC)

Perhaps we need to give editors a ration of some number of edits they can make to an article per month. This would then push editors toward making larger edits instead of making many small edits. Count Iblis (talk) 00:26, 25 May 2016 (UTC)
A ration? Really? Personally I tend to do both. I create pages and do minor corrections (listed here for instance). Bots can do a lot of this work, but in some cases it needs human intervention. Karst (talk) 10:57, 25 May 2016 (UTC)
Indeed a single correction could fix a glaring error, such as duplicate wp:reftag name "autogenerated1" could be fixed as "autogenerated2" to remove a hideous reftag error message from a major page. -Wikid77 (talk) 20:58, 26 May 2016 (UTC)


I do not consider the rate at which systemic biases are addressed to be particularly quick. There are several WP:BACKLOG categories full of errors far more glaring and higher priority than technical template issues. EllenCT (talk) 22:26, 25 May 2016 (UTC)
Well, the problem with wp:BACKLOG is the numerous pages already fixed where removal of their backlog-tags is the real "backlog" due to the tedious complexity of judging when a page is good-enough to clear a backlog entry. Imagine the frustration for an editor who edits several backlog pages and cannot find any problems now. -Wikid77 (talk) 20:58, 26 May 2016 (UTC)
I don't have to imagine, as I remember it, but I think most backlog characterization templates are valid. Do you want to try a small sample tally to estimate the percentages of issues corrected without accuracy tag removals? EllenCT (talk) 12:51, 27 May 2016 (UTC)

The over-reach of tagging among the wp:BACKLOG categories varies by type, where the copy-edit category has typically been correct over 98% of the time (perhaps because so many details only get fixed when untagging), while other categories still list many pages already fixed, which should have been untagged. -Wikid77 (talk) 15:17, 28 May 2016 (UTC)

A new study was just released showing a link between cell phones and cancer Wall Street Journal 5/27/2016. It will be interesting to see if this is covered as NPOV. I have a feeling a bunch of "neutral" "pro-science" editors will make sure it gets buried, say it is too new, WP:OR, WP:Fringe, a "primary study", WP:UNDO and every excuse possible to keep the news from messing with cell phones sales. And any editor trying to insert it will be labelled an anti-cell phone activist, WP:SPA, disruptive, lacking competence or works for some industry that competes with cell phones... Well, let's go see... --David Tornheim (talk) 22:18, 27 May 2016 (UTC)

Well, it is a primary study, and partial results, and not peer reviewed yet. Also 6 Watts per kilogram far exceeds the FCC's limit of 1.6 W/kg. PMID 23320534 is the most recent WP:MEDRS-grade peer reviewed literature review on the subject, which you can see has plenty of very clear warnings fully congruent with your apparent opinion on the topic. We follow WP:PSTS generally, or I wish we did, anyway. EllenCT (talk) 22:29, 27 May 2016 (UTC)
The draft report itself says that the partial results themselves have been peer reviewed. See [32][33] and for balance [34] for secondary sources on this report. I agree that due to the high levels of exposure used, not to mention some of the apparently arbitrary findings on longevity, this study can never do more than highlight the need for a better one - and that has been highlighted many times already. — Cheers, Steelpillow (Talk) 07:09, 28 May 2016 (UTC)
PMID 9085432 has a clue, but it's getting a little stale for MEDRS. EllenCT (talk) 11:59, 28 May 2016 (UTC)
As presented, this is unconvincing. The main problem, as identified by the first reviewer in the document linked, is that multiple comparisons were made, increasing the risk of a chance result. Remember, in biology the culture is to publish data that is 5% likely to happen by chance, which is definitely not five or six sigmas like in particle physics, and so when you compare multiple test conditions to multiple controls and work statistics on them, it can get very risky that you come up with something meaningless. My own peeve (Maxwell Lee goes into this one) is that there is just one control group, no larger than the various test conditions, and so there's a lot riding on that happening to come out zero. The others certainly vary greatly (and it is not really dose dependent in most of the tables). I understand that the control tests in part how good their particular tech is at spotting lesions, but I feel like the rate still shouldn't be that much of a mystery; aren't there any other experiments, any other drugs or tests, done in that building or university by people using comparable methods and food that could have been arranged to serve as something of a verification of the control? (Lee points out the usual rate is around 1.5%)
I should add that I actually think it's likely there is a risk, but I don't like how these studies are designed. They should not be trying to duplicate a phone and look for rare cancers; they should have strong microwave/terahertz sources that are tunable and monochromatic and scan through the frequencies carefully. If there is an effect, presumably it works by displacing proteins from DNA or from binding one another, so there might be one particular frequency your phone sometimes uses that has a noticeable health effect while others don't. That effect wouldn't have to be cancer; it could be anything that can be caused by misregulating a protein. I would bet money various militaries are working on this hard - they already are interested in using terahertz, for example, to tell if bacteria in the air are anthrax based on the sequence specificity of absorption. The question is, if a piece of DNA specifically interacts, can it be targeted and what can you do with that? They might use this, say, to cause depression in a particular group of political opponents targeted from drones. But if that is possible, this kind of study will never figure it out. Wnt (talk) 15:18, 28 May 2016 (UTC)
The link between cell phones and cancer is a hypothesis that for many years has been in a desperate search for the data to support it. Count Iblis (talk) 19:13, 28 May 2016 (UTC)

The Signpost: 28 May 2016

Motherboard article about suicidal Wikipedia editor

Wikipedia Editor Says Site’s Toxic Community Has Him Contemplating Suicide Thought this might be interesting, nay, disturbing, to regular readers of this page. The editor in question claims he became suicidal because "Nobody on Wikipedia seems to be kind," and "You [i.e. Wikipedia editors] are all so busy power tripping that you forget there is a real, live person on the other side." This he says resulted from "...an ongoing disagreement with other editors on the “talk” pages of an article about a local politician. The debate devolved into name-calling, the editor wrote, and eventually he was completely banned from editing the site he had devoted so much time to." This suggests that WP:BITE may be an even more serious problem than commonly believed. Everymorning (talk) 23:14, 27 May 2016 (UTC)

Issues with situations such as these are often complex. Not thrilled about this one being aired more. Cas Liber (talk · contribs) 23:19, 27 May 2016 (UTC)
Civility is part of Wikipedia's code of conduct and one of its five pillars, and should be enforced aggressively, especially upon admins. and especially against passive aggressive , sarcastic and other nasty edits. That will quickly solve this ridiculously persistent problem, I think. Excuses about intensity, passion, zealousness etc., e.g. "The community is intense and passionate, which means its editors are often zealous in their policing" are total B.S. and lead to acceptance of the status quo which is not acceptable. We must keep moving forward and that means a zero tolerance for incivility of any kind going forward from this point in time. Nocturnalnow (talk) 21:30, 28 May 2016 (UTC)
Sorry, but I'm gonna push back a little. Not a lot, just a little. Count me as a big fan of civility; I try to live it but, being human, occasionally backslide. My pushback isn't because I think incivility is fine, or even that I worry that I might get caught in a bad moment. My pushback is motivated more by the fact that absolutist policies such as "zero tolerance for X" almost invariably turn into Kafkaesque idiocy. I was tempted to recite some examples of zero-tolerance policies gone bad but I don't want this to devolve into a political discussion, so I will simply assert that for just about any zero-tolerance policy it is easy to find evidence of absurdity arising from them.
Anyone who is ever tried to write rules whether Wikipedia or elsewhere quickly learns the near impossibility of writing absolutist rules. Wikipedia is taking, in my opinion, a generally reasonable approach, with guidelines that are short of absolute, and IAR to help cover the true exceptions. Any attempt to create a zero-tolerance policy will be the beginning of the end of this place.--S Philbrick(Talk) 11:55, 29 May 2016 (UTC)

Wikipedia is how Satan sends jerks to Hell

I would not be surprised if Satan danced with joy to see admins and zealous editors ruthlessly condemning other editors in vicious tag-teams of hateful, self-righteous whatevers. In that Motherboard article, "Wikipedia Editor Says Site’s Toxic Community Has Him Contemplating Suicide" the phrase "power trip" seems to relate to powerful editors, such as admins, being blamed for the toxic Wikipedia environment. The Wisdom of the Ages has warned, "Absolute power corrupts absolutely" and I still think WP needs to implement term-limits for admins so if they anger enough people, then they will not get re-approved, and if they cannot live with that limitation, then they should not be admins on Wikipedia. There is no intelligence test for admins (zero, nada, nichts, none), there is no systematic review of their judgmental actions, and the longer admins remain in power, the stronger the tendency toward corruption in the use of their power. Forgetting people becomes very, very easy in such a system of endless power. It is a sad situation, paved with the suffering of many trampled editors. -Wikid77 (talk) 03:06, 28 May 2016 (UTC)

Please see Tree of the knowledge of good and evil#Religious views, [35], and PMID 25328772. EllenCT (talk) 12:11, 28 May 2016 (UTC)
Just want to note that the subsection title implies that the editor who was suicidal (living in a form of hell) is a jerk. I suggest you change it. Jytdog (talk) 11:06, 29 May 2016 (UTC)
My reading was that "jerks" referred to admins, but maybe I misread. In any event, count me in support of some form of term limits (my proposal would be that all admins have to spend some period of time as a non-admin on occasion - might be insightful.)--S Philbrick(Talk) 11:58, 29 May 2016 (UTC)
We need reliable sources for what Satan does, the post above is just OR. -Roxy the dog™ woof 20:09, 29 May 2016 (UTC)
Different religions, and sects of similar religions, ascribe different modus operandi to their antagnoistic characters. The allegories of C.S. Lewis and J.R.R. Tolkien are instructive in that they both attempted to re-construct the Judaeo-Christian mythos and metaphysics by replacing actors with actions and placing the two trees in a swamp instead of a garden; and using elements of Germanic and Norse mythology and sticking the two trees on a different continent instead of a garden, respectively. We can also look towards Southern Asian and aboriginal religions for alternative creation myths with alternative antagonists.
However, getting our economics articles harmonized with Ostry, Jonathan D.; Loungani, Prakash; Furceri, Davide (June 2016). "Neoliberalism: Oversold?" (PDF). Finance & Development: 38-41. should be a far higher priority than bookkeeping all the creation myths and their antagonists.
Also I think creating a bot capable of using the Socratic method to produce epiphanies in people should be a higher priority than improving religion articles. If I was going to work on metaphysics articles, I think I would wait until new contemporary creation myths appear in the peer reviewed academic journal literature reviews. EllenCT (talk) 15:22, 30 May 2016 (UTC)
@S Philbrick. I agree with you. Furthermore, I have suggested elsewhere that those applying to be an admin should spend some time in a Topic Ban of their favoured subject, or even a Site Ban, to understand the implications and consequences of them imposing these on editors. DrChrissy (talk) 20:29, 29 May 2016 (UTC)

Is this an improvement to [36]: "...the extent to which cultural diversity should be permitted to influence bioethical judgments in Africa, which at present is burdened with many diseases, should be of concern to researchers, ethicists and medical experts taking into considerations the constantly transforming global society"? EllenCT (talk) 19:07, 30 May 2016 (UTC)

Wikipedia:Wikivillages of Cameroon

Need help. User:David.moreno72 is interfering with new users creating articles for Wikipedia:Wikivillages of Cameroon. — Preceding unsigned comment added by De Geescht vum Herer (talkcontribs) 22:22, 29 May 2016 (UTC)

See also User talk:Charlipro for how rude and disrespectful this new editor was treated by others. They were accused of vandalism and treated like a criminal for participating in a sanctioned Wikipedia contest and creating a new article! — Preceding unsigned comment added by De Geescht vum Herer (talkcontribs) 22:35, 29 May 2016 (UTC)
Where to begin. First off, sign your posts when you are talking on talk pages. This can be done with four tildas ~~~~. Second, Jimbo does not deal with conduct disputes. Third, have you read the response that David gave you on their talk page? Just because something is part of a competition does not equate to immediate acceptance of the material. It still must follow our policies and guidelines. Wikipedia is a collaborative project. You need to continue discussing this with David before you do anything else. And certainly before complaining to Jimbo who does not deal with this stuff anyways. --Majora (talk) 22:43, 29 May 2016 (UTC)
Where to begin indeed! Just where do you get off talking to and treating people like this? There is collaborative editing on this topic from three different editors whom David Moreno ignored! — Preceding unsigned comment added by De Geescht vum Herer (talkcontribs) 22:50, 29 May 2016 (UTC)
I think that Wikipedians from areas with poor internet service are being treated unfairly. When Wikipedia first began operations, editors did not insist on references for every article. It was sufficient that a place really existed to have an article on it. I imagine that there are not as many newspapers online in Cameroon as in Britain, making it harder to link to a source. That said, I imagine people in Cameroon can cite paper documents the same as anywhere else, and it would make it easier. There is also considerable confusion because Jakiri makes it sound like this village is just one of many neighborhoods inside the city. I don't understand whether, if you go to https://www.google.com/maps/@6.0808654,10.6395755,3856m/data=!3m1!1e3 you would see this village as something inside Jakiri, or if you would need to move the map far away from it. Despite all, the village does have quite a bit of routine representation on the web: [37] [38] [39] If you can use this site to put coordinates on your villages, that's a good thing, though I don't promise it will help in this dispute. But I hope that the deletionists will back off and recognize that they're being unfair - we have articles on every single school district in the U.S., which are certainly no more notable than these. Wnt (talk) 21:28, 30 May 2016 (UTC)
Here are three Cameroonian newspapers.
Wavelength (talk) 22:00, 30 May 2016 (UTC)
See also http://www.onlinenewspapers.com/cameroon.htm
and http://www.abyznewslinks.com/camer.htm
and http://www.4imn.com/cm/.
Wavelength (talk) 22:07, 30 May 2016 (UTC)
See https://botw.org/top/Regional/Africa/Cameroon/ (by Best of the Web Directory)
and http://www.dmoz.org/Regional/Africa/Cameroon/ (by DMOZ)
Wavelength (talk) 22:28, 30 May 2016 (UTC)
See http://www.searchenginecolossus.com/Cameroon.html
and http://radiostationworld.com/Locations/Cameroon/
and Cameroon Radio Television (http://crtv.cm)
Wavelength (talk) 22:57, 30 May 2016 (UTC)

I suppose I should frame this in the form of a question (paid editing)

There is a very good blog post at Wikipediocracy by Mike Wood, who lost his job due to WMF or ArbCom retaliation over paid editing. Hopefully this will be cause for introspection about Check User abuse, outing, and the mentality that leads to this sort of disgusting behavior against one's wiki-opponents. Carrite (talk) 15:49, 23 May 2016 (UTC)

Some users have been fired when using workplace computers to edit Wikipedia while becoming notorious for arguing with admins or other users, perhaps on company time, and then the employers fire them. -Wikid77 (talk) 13:18, 24 May 2016 (UTC)
Do you think it's fair to say that the root of the problem was the attempt to leverage Wikipedia for personal gain by moonlighting at a paid position for large corporations who usually wish to provide only one side of their stories, against NPOV, WP:BALANCE in particular, after repeated warnings, and then bragging about it in the commercial press? EllenCT (talk) 16:25, 23 May 2016 (UTC)
Yeah, agree with Ellen here. As you (Carrite) presumably know, I was one of those who consistently supported allowing and regulating paid editing rather than trying to suppress it (if not, I dare say a 2601: IP can pop up shortly to remind you), but someone who makes a public statement like As for me, I have made money from Wikipedia and will continue to do so as long as the community continues to create the environment for it. With each article that is deleted by the community, there is another client looking to pay me to get the article back up. For every article that is tagged, there is a marketing company contacting me to re-write the article and remove the tag. For every person who tries to find my account and block it, there are ten emails that I am responding to, giving quotes to write their article. No one can make money from Wikipedia? can hardly be surprised when the WMF starts sending out the cease-and-desists. ‑ Iridescent 17:24, 23 May 2016 (UTC)
I find him refreshingly honest, open and believable. I think he's learned a lot since then. Running this business on his lunch hour wasn't the best move (c'mon now, was it all during lunch?); he should have limited his moonlighting to after work hours, from home. I'm not sure that Elen or the WMF could have known precisely who was editing from the casino, couldn't it have been customer(s), assuming they have free Wi-Fi? The casino's tech guys probably traced it to a specific computer. He seems to treat his current customers well, and knows how to deliver value in terms of what edits will be accepted. Three Areas of Wikipedia Marketing Wikipedia Marketing – Understanding the Deletion Process (podcasts). I don't mind if someone gets paid a bit for producing quality articles of this sort. Among my least favorite editing activities is cleaning up ("Wikifying") new articles for style, left by single-purpose editors who don't take the time to learn. If his editors produce quality work that I don't have to clean up for free, I don't mind this much. And talk me about "personal gain" after you read the WMF's recently released tax return. You know they wouldn't volunteer this information if the government didn't require them to. – wbm1058 (talk) 18:18, 23 May 2016 (UTC)
I think Wikipedia should not encourage admins contacting employers. Admins are community leaders, not official WMF spokesmen. When they go after someone's employer, basically, it's cyberbullying. There's no formal line of distinction between our admins doing this and if Russian admins decided to notify employers of people who write about gay nightlife - only community policy. Now in this case, tracking back to Wikipedia:Long-term_abuse/Morning277, substantial abuse of multiple accounts was alleged. Maybe WMF could have issued a cease-and-desist, then complained to the company's internet about TOS violations. But why? Isn't it enough to hard block the account? Of course, there's always a chance that someone at the site goes to edit Wikipedia and starts asking questions, but it wouldn't then be Wikipedia's initiative. You can say that this was so important that it had to be stopped... but what is stopped? The way he tells it he's gone from editing in his lunch hour to making a business out of it. Some coup for Wikipedia, eh?
Best way to fight this should be familiar to the drug legalization crowd. You simply allow articles about companies with relatively mild notability requirements. Most of what they care about is simply having an article with their name and a link to the company website. If you let Wikipedians easily cover that basic iota of information, what's left for them to pay to get done? Thus increasing the importance of volunteers while decreasing importance of paid editors. Wnt (talk) 19:00, 23 May 2016 (UTC)
Technically this must have been a checkuser at the time (or someone who previously had the rights and access). Which is a small subset amongst admins. Only in death does duty end (talk) 19:53, 23 May 2016 (UTC)
The IP who posted the "Catch Me If You Can" taunt here is in the UK, so a checkuser on that wouldn't have located Mr. Wood. wbm1058 (talk) 19:58, 23 May 2016 (UTC)
So I'm wondering if there was a valid subpoena or other compulsory request from law enforcement that required the WMF (or a volunteer "checkuser") to contact the casino. wbm1058 (talk) 20:13, 23 May 2016 (UTC)
I imagine they had a suspicion and checked the user most likely to be the paid editor - which threw up the casino IP. Meta privacy policy for the time allows for contacting third parties with private data for the purposes of preventing distruptive editing. As the editor was editing from a work network, on his work computer - whoever did contact his employer was (while being a complete idiot) within the policy at the time. They didnt even notify a third party, they notified the owner of the IP doing the editing. Only in death does duty end (talk) 20:19, 23 May 2016 (UTC)
So if the culture here is "preventative, not punitive", and the issue needing prevention was "disruptive" editing, wouldn't the block, done from afar, be sufficient to prevent the disruption, and any contact of the casino, if they indeed knew that Mr. Wood was employed there (I guess they likely put two and two together), wouldn't that have simply been punitive? wbm1058 (talk) 20:29, 23 May 2016 (UTC)
It would depend on the circumstances. Personally in the case of paid editing - absent obvious nono's like whitewashing a murderer or corporate wrongdoings - I think the reaction was a little harsh. He was creating puffy articles on non-notable companies for pay and adding neutral/postive information about them to already existing articles. Not the end of the world. The counter would be he had a LTA with socks going back to 2010. 3 years of the above would certainly qualify as 'disruptive' in the WMF's and most wikipedians (who had to deal with the cleanup of the editing) books. And generally with long term abuse, you start to look for more serious solutions. The WMF has not really had much success in this area. I dont recall Grawp being handled particularly well despite the WMF knowing exactly who they are. Only in death does duty end (talk) 20:45, 23 May 2016 (UTC)

Wood is absolutely disgusting. He knows what he is doing is against the rules (See e.g. WP:NOADS "content hosted in Wikipedia is not for .... Advertising, marketing or public relations," and his book "Wikipedia as a Marketing Tool"). What he is doing is selling products (or company reputations) under the guise of an objective report. That's fraud in my book. He is also extracting money from a non-profit with an educational mission by degrading the educational content. That's theft in my book. His only justification is that he can't be caught - and then when he does get caught he moans that somebody else must have broken the rules. Disgusting. Smallbones(smalltalk) 02:09, 24 May 2016 (UTC)

I agree completely with your sentiments Smallbones. Although I agree that checkuser should not be abused, I am troubled that there was not near unanimous condemnation of Mr. Wood's article bragging that Wikipedia is a great place to make money with paid COI editing, with statements like these:
For every article that is tagged, there is a marketing company contacting me to re-write the article and remove the tag.
Having authored a new book, and collaborating on a regular basis with a team of editors, I am able to assist clients in every aspect of Wikipedia.
How can we maintain a reputation for neutrality when we allow paid editing to determine the content of our articles? Wasn't it a problem when 44% of the article about BP was written by BP [40], or is that the new model of Wikipedia we are shooting for? --David Tornheim (talk) 03:21, 24 May 2016 (UTC)
Oh come on. If neutrality is paramount, then we should move towards the Sanger model. The encyclopedia anyone can edit becomes the encyclopedia only neutral editors can edit. Prohibit anonymous editing – all editors must register and identify themselves before they are allowed to edit. Employee of BP? Sorry, you're not allowed to edit that article, but an article on oil exploration technology is fine. I wouldn't mind a move in this direction, but it probably means the Foundation will have to start paying for content. There aren't enough people without agendas who are willing to work for free. What's the difference between an article by an editor paid by a record label to promote a new singer, and an article by a fan who loves their voice and wants to promote the singer. One is paid and the other's not, but neither is likely to include less than favorable criticism of the singer's talent. And besides, all this promotional and fan cruft stuff about obscure pop-culture is hurting the educational mission by degrading the educational content. So if Wikipedia isn't going to make any changes to cope with the inevitable editing that comes with being a top-ten website that Google loves, then y'all should stop whining about it. wbm1058 (talk) 04:45, 24 May 2016 (UTC)
You do notice, I chose to use my real name for transparency. I don't understand why anonymity is so important to editors. And I do agree it creates the issue that it is very hard to prove COI if the only evidence you have is the edits.
As for the fan vs. the big record label, I trust the fans more, since they are motivated to share the positive experience they had of the band and will likely include RS that reflects that view, just as those who hated the band might be motivated to include RS with critical reviews. As long as the lovers and haters of the band can work together and fairly represent the RS that is out there, I don't see a problem. I think most fans understand that other people don't always like the music they like. And, I have met fans who are near experts on every little detail they could find about the band, including negative things like drug addiction, drama, history, etc. I imagine such fans might add some interesting things to the article, if they can find the RS. But, it is true, some fans and haters can be a over the top and refuse to accept any view counter to their own, even if it is clearly in the RS.
The big record labels are not like this. Their goals are not to educate, inform or provide healthy or meaningful or deeply moving experiences from music. Their primary agenda is simple: to make money for their owners. They will push music that is formulaic and proven to appeal to a wide audience by bean counters [41] rather than appeal to artists, music critics and musicians. It doesn't matter if the music is actually of good quality, unique, expressive or not: As long as they can sell it, then it is "good". I am not interested in reading (or helping to provide) an article written by paid editors to publicize bands that are mediocre and trite, which make them sound better than they are. Let them pay for advertising and not use our encyclopedia for promotional purposes.
I don't see why "obscure pop-culture" is "degrading" to our educational mission. Are we only permitted to have articles on "high" art? Despite being an encyclopedia that anyone can edit, is Wikipedia actually classist and would eschew talking about the Harlem Renaissance as it was happening? Would it have been a mistake for Wikipedia to cover grunge before the record industry found a way to make tons of money from popularizing it and turning it into a formula?
As for the difference between paid editors who are here simply to promote the interests of the person who pays their salary, and those who come here because they are interested in improving the content, I see a huge difference, which I hope I illustrated by the example you chose. --David Tornheim (talk) 06:33, 24 May 2016 (UTC)
Arturo from BP (or whatever his name was) helped enormously with the coverage of BP. And he didn't even cover some of the terribly inaccurate material we had relating to Deepwater Horizon.
Moreover he followed best practice throughout. The knee-jerk reaction by some of the community was not only an unnecessary insult to Arturo, but to the editors who considered and moderated his work.
All the best: Rich Farmbrough, 10:23, 24 May 2016 (UTC).
Rich, that is a ridiculous and inaccurate statement. While I disagree that Arturo serves as a good example of why we should ban paid editing (being that he outed himself, he was less able to be overtly manipulative), it is not true according to many editors who worked with him and on the BP article, that his presense was more helpful than harmful. I think people like Sarah SV and Coretheapple would agree that Arturo's work caused quite a lot of unnecessary drama. The BP oil spill article went through a giant overhaul and almost nothing was found to be without support, it remained entirely intact save for some minor details, changes and creation of daughter articles. To my knowledge the only involvement you had on these articles was to remove the entire "environmental damage" section one night, creating a daughter article, without leaving a summary, or consulting other editors. So I am unsure why you are making this comment. (If I have you mistaken with someone else, my sincere apologies, bad memory these day.) petrarchan47คุ 20:41, 25 May 2016 (UTC)
@Petrarchan47: That's a fair statement. Arturo's job is to spin the article to make it more favorable to BP, which include not just what he does but what he does not do, which is to not correct egregious errors when those errors make BP look good. As he did (at least) once. He is WP:NOTHERE to build an encyclopedia but to repair a reputation. Coretheapple (talk) 20:15, 28 May 2016 (UTC)
@Rich Farmbrough: I have no issues with Arturo. I don't consider him a paid editor, but more like a paid lobbyist, as an official voice of BP at the article. I approve of having such a mechanism to give non-COI editors a way to hear from the subject of the article, allowing us to become aware of common misconceptions, false rumors, RS we might not have been aware of, and helping us to avoid potential libel. As a reader, I prefer to be able to hear the voice of the entity the article is about, not just those who would like the entity run out of town on a rail (consider the Séralini affair, where Seralini's views are almost impossible to find). This is what I look for from journalists: NPR on Fairness and the Fairness Doctrine. I would like to see the same at Wikipedia. Instead, I am seeing increased use of WP:undo and WP:Fringe to create one-sided content in disregard of the fairness doctrine, even when opposing views are available in RS.
So yes, I support BP having a voice at the article that is about it, and I commend BP for its transparency. I see no problem with the person who works for BP being paid to express BP's official view, just so long as they have no decision-making power over the content.
From what I have heard, Arturo was easy to work with and followed the rules. So I support allowing lobbyists at the articles, as long as they don't edit the article, and as long as they do not misuse their position to disrupt the talk page, disrupt consensus of non-COI editors, vote on rules at Wikipedia, and also long as they stick solely to their mission of representing their client at the article (and related articles) in full disclosure.
That said, BP no doubt pays PR firms to maintain its image. If one of those PR firms in addition to maintaining advertising has a team of paid Wikipedia editors, that is potentially a big problem. Editors from such a PR firm could argue for or even insert Arturo's draft directly into the article or have other undue influence--not acceptable. And since it is the PR firm and not BP that would be breaking the rules, BP appears to be clean, even if the PR firm is caught.
When Arturo's version became the Wikipedia version, that shows to me that something is awry. I have trouble believing BP's version is NPOV. Just as I would have a problem if a disclosed lobbyist from Greenpeace showed up and their draft was added verbatim. It shows a failure on our part that this happened. And I have seen no major changes to address these problems. --David Tornheim (talk) 20:21, 24 May 2016 (UTC)


@Smallbones: I don't dispute that Wood's service was against the rules. The problem is that we enforce these rules kind of like the way the government enforces laws against hacking, where the nerd kid from next door gets hauled off in cuffs for screwing around, while the syndicate in China breaks into whatever it likes with impunity. The mere fact that Wikipedia could crack down on this guy in this way is proof that he was never really a threat. We have obvious, obvious signs of crookedness around here, like the way that Square Enix gets a main page article about their products every six months (well, now their fictional characters - I think they've run out of products to feature) as if they were 1/180 of the entire world. No, I don't know who is to blame there. It feels more and more dangerous even to complain about it. So I take no pleasure in seeing some guy fired because in his lunch hour he gave some companies nobody ever heard of the thrill of having their very own Wikipedia page for a while. It looks like enforcement, but it's more like the exception that proves the rule. Wnt (talk) 09:47, 24 May 2016 (UTC)
Wnt I agree with almost everything you say here, including the part about the danger of even talking about the problem. I received threats from 3-4 editors, 2 of which are from admins for posting this: User_talk:Jimbo_Wales/Archive_208#GMO.27s_and_concerns_about_corporate_influence. I agree with you that the focus of COI investigations are on small entities, while the big fish get away. I do not trust the objectivity of at least one of the major players at COI. It feels to me like the fox in the hen house.
I honestly wouldn't care if Wikipedia were to support a bunch of short yelp-like entries on non-notable businesses, giving basic information, as long as everything is based on RS. The same for films. I am not sure why we even have a notability requirement. It's not like this would eat up a huge amount of disk space, but maybe it is a burden on our bots to have "too many" articles?
As for Woods, I am less sympathetic. We just have his side. We don't know that he worked exclusively for small entities. With his article in Business Insider, big entities might have contacted him looking for advice. As for his claim that his employer violated his "privacy", it sounds like he may not know the law. See for example: [42]. (FYI. I am not an attorney). But worse, he admits to deliberately breaking the rules. If it were a minor violation--I do not consider his misuse of Wikipedia minor--or if the interpretation of the rules was subjective, calling the ISP might be over the top. But in this case, if other methods had not worked, it does sound justified. I do not see Wikipedia to blame for his lost job. We do not know all the reasons for his termination. What we do know is his own actions at Wikipedia were unethical, he got caught, actions were taken against him, his employer found out and there were consequences. If he had not wanted his employer to find out about the unethical things he was doing at Wikipedia, he should have used a different ISP. --David Tornheim (talk) 21:50, 24 May 2016 (UTC)
Gee who would that be that you don't trust David? You are a pixel waster. Jytdog (talk) 06:34, 25 May 2016 (UTC)
That's a personal attack, Jytdog, and in my view worthy of a preventative block per WP:NPA. Jusdafax 05:22, 26 May 2016 (UTC)
So barky. Jytdog (talk) 05:09, 27 May 2016 (UTC)
  • Per WP:SLEUTH, "Outing is not a suitable strategy to control COI editing." Martha Stewart comes to mind. We shouldn't have selective enforcement of guidelines for those who aren't significant donors to the Foundation. – wbm1058 (talk) 14:05, 24 May 2016 (UTC)
  • I'd like to state that I favor strong enforcement of the TOS. I'm pleased that Mike Wood got fired as a consequence of abusing the system, and I wish there were further consequences strong enough to motivate him to stop editing in a way that violates the TOS. Looie496 (talk) 15:04, 24 May 2016 (UTC)
  • The arguments seem to shake out this way:
    • Inserting paid ads into Wikipedia is not against the rules.
    • They can't catch me
      • They did - too bad, stop your cry-baby act
    • They had to break the rules to catch me.
      • I haven't seen any evidence of that, If you have any evidence take it to ArbCom or the WMF
    • "They" outed me.
      • You outed yourself with your advertising/editorializing, not to mention your current webpage.
    • The rules are applied inconsistently
      • That sounds like an excuse that could stop the enforcement of any rules. A real cry-baby act IMHO
    • Other people do it.
      • That sounds like an excuse that could stop the enforcement of any rules.
    • We've got a userfied essay that says something about Martha Stewart.
      • We've got real rules that deal with the problem. See the TOU, WP:NOADS, WP:Sock, etc. etc.
    • The WMF tolerates donors breaking the rules.
OK. Does any paid edit to any article, by definition, make that article a paid ad? wbm1058 (talk) 16:19, 24 May 2016 (UTC)
Why do you care? It's plain you don't like the rules, here, which is fine, it does not change anything about them. Alanscottwalker (talk) 16:22, 24 May 2016 (UTC)
Does Wikipedia:Reward board violate any rules? I see there that someone is offering AUD $20 for an image of both members of the band M2M together. Would a paid image of the band on their article constitute paid advertising for the band, and thus be a rules violation? Someone is offering £10 GBP for getting Buxton to GA status. Doesn't that amount to paid promotion or advertising for that spa town? I've yet to see any money offered for making spelling corrections. But, theoretically, would an offer of money for that be a rules violation? wbm1058 (talk) 17:52, 24 May 2016 (UTC)
Really? Your asking about something that is explicitly governed by WP:COI? Read it. To answer you, it depends on how it's done. Alanscottwalker (talk) 21:48, 24 May 2016 (UTC)
Reward board
Another benign example of paid editing is the reward board, where editors can post incentives, usually to raise articles to featured-article or good-article status. If you participate in this kind of paid editing, transparency and neutrality are key.
So it is theoretically possible to promote that spa town in a neutral way, with no conflict of interest, even though you're getting paid to do that, but only when you transparently advertise in an approved venue that money is being exchanged?
However, identical text written by Mr. Wood, for an identical amount of money, is illegal if there is no transparent reporting in an approved venue of the amount of money exchanged? wbm1058 (talk) 23:08, 24 May 2016 (UTC)
No. You do have a COI, which is why it's covered in COI and transparency is one of the keys for dealing with that (disclosure is key). Alanscottwalker (talk) 23:17, 24 May 2016 (UTC)
I'm sorry, I have trouble parsing your answers. Are you saying that I personally have a COI (with regard to what?), or that offering £10 GBP for getting Buxton to GA status is illegal because the person making that offer has a COI? I'm not aware of any disclosed interests of the person offering the reward. wbm1058 (talk) 23:53, 24 May 2016 (UTC)
No, not "you" in the sense of talking about your personal situation (which as far as I am aware is not being discussed), "you" in the sense of you asking is the Reward Board under COI. It is, explicitly so. And transparency is one of the keys for dealing with that COI. Alanscottwalker (talk) 00:01, 25 May 2016 (UTC)
Right, so let me put it a different way. Assume that:
  1. An editor is paid £10 after getting Buxton to GA status – assume that it passes neutrality concerns because it is a Good Article. I also think it will "promote" tourism in Buxton, which is the point of the £10 offer
  2. If the £10 transaction is transparently documented at an approved venue, such as Reward board, no rules were violated and everyone is happy, although this was technically conflict-of-interest editing
  3. If the £10 transaction was not transparently documented, then that is a violation of the rules
Is that a correct assessment? wbm1058 (talk) 00:31, 25 May 2016 (UTC)
The Buxton article is interesting, as it is a feature determined by a hot mineral spring; the question is, why is it hot in a sedimentary area? This provides an explanation, which also explains the magnesium content, while emphasizing that our unsourced list of geothermal springs in the United Kingdom is woefully incomplete. (I mean, the link mentions "almost 100" in Harrogate) A little digging by independent editors may divert away the benefit paid editing might be intended to have, and as such should be considered as a real and more pleasurable method of 'enforcement'. Wnt (talk) 13:00, 27 May 2016 (UTC)
Also consider the externality penalties. Gibraltar lost a whole lot of their front banks used as offshore tax havens after they started paying for tourism PR. Talk about a win-win for everyone except tax evaders. EllenCT (talk) 20:24, 24 May 2016 (UTC)

Reward board

Continuing the discussion from above. I see that the reward board has survived two deletion discussions, the most recent was a bit over two years ago. There have been less than 200 offers made over the history of the board, they can be perused HERE. Most of the offers made have been of insignificant monetary value (barnstars and such), however the offer that triggered the last deletion discussion was for $150-$300. I'm not sure I see much point or value in this board if it can't offer reasonably attractive rewards. The way to keep underground crime syndicates from forming is to offer people a legitimate means for making some money. – wbm1058 (talk) 14:39, 25 May 2016 (UTC)

I was not familiar with the "reward" board. I certainly see your arguments above that it could cause similar problems to paid editing. Nonetheless, from looking at it, when the rewards are low-stakes, such as barnstars or brownies or feelings of appreciation, it doesn't really raise any red flags for me, unless the requested help is related to content or decisions. And it does have the advantage of transparency. I would be okay with having people paid to help with bots or some kind of programming, but would prefer that be done through the WMF with paid staff rather than "paid" volunteer editors. So I could see an argument for getting rid of the reward board, or making sure it can't be used to cause bias in articles and decisions. --David Tornheim (talk) 02:16, 26 May 2016 (UTC)

Terms of use: then and now

Mr. Wood has raised an interesting point. The terms of use in effect on January 9, 2013 did not say anything about paid editing. I see no basis there for contacting the casino on January 10, 2013. While § 4. Refraining from Certain Activities says WMF can exercise their enforcement discretion with respect to the above terms, I'm at a loss as to which specific term he violated. Only on 16 June 2014 was Paid contributions without disclosure added to the TOU. Terms of use/Paid contributions amendment was created on 11 February 2014‎, and Terms of use/FAQ on paid contributions without disclosure was created on 6 May 2014‎. So it seems the only basis for contacting the casino was (perhaps) something in the local community policies or guidelines?

Now, with regard to his current business operations post June 2014...  wbm1058 (talk) 21:00, 25 May 2016 (UTC)

The WMF made that change to get itself a stronger basis for action. The ToU has always said that editors have to follow community policies and guidelines. With regard to the WP community, WP:PROMO has been policy for a long time and this is what the COI Guideline looked like on Jan 6 2013. Jytdog (talk) 02:22, 26 May 2016 (UTC)
The summary box at the top of the ToU says You adhere to the below Terms of Use and to the applicable community policies when you visit our sites or participate in our communities.
However, while the current ToU has a clause "Applicable law, or community and Foundation policies and guidelines", the January 2013 ToU did not use the word "guideline".
The WP:COI guideline "is a generally accepted standard that editors should attempt to follow, though it is best treated with common sense, and occasional exceptions may apply." (Jan. 2013)
Where does the WP:Checkuser policy authorize anyone to directly contact the owner of an IP address? wbm1058 (talk) 04:34, 26 May 2016 (UTC)
Morning277 wasn't just engaged in paid editing, he was also engaged in sockpuppetry leading to his block in 2012. I'm not arguing for contacting his employer - personally I'm completely against that - but in regard to the Terms of Use he was in violation because of his long term and large scale use of sock accounts. - Bilby (talk) 06:05, 26 May 2016 (UTC)
Even the current ToU doesn't use the words "sock" or "puppet". "Multiple" is only used in the sense of "access blocked on multiple Project editions", not multiple accounts on the same Project edition. "(If) any individual has had his or her account or access blocked under this provision [§10. Management of Websites], he or she is prohibited from creating or using another account on or seeking access to the same Project, unless we provide explicit permission." I see nothing prohibiting users who have not been blocked under §10 from creating multiple accounts. So, again, this is just under community policies. WP:CheckUser is a policy though, while WP:COI is just a guideline. wbm1058 (talk) 12:12, 26 May 2016 (UTC)
As I recall, the argument was that using socks to evade scrutiny when editing, or to create a false consensus, comes under "Attempting to impersonate another user or individual, misrepresenting your affiliation with any individual or entity, or using the username of another user with the intent to deceive". Certainly, the intent is to deceive people into believing that you are someone else, so they don't detect the paid editing. - Bilby (talk) 12:37, 26 May 2016 (UTC)
Just to clarify, I don't understand why the focus here is on COI editing. We know he was a paid editor, but he was blocked for abusing multiple accounts, and later banned when it was found that he had been running one of the biggest sockfarms in WP's history. He was doing this in order to deceptively engage in large scale paid editing, but it wasn't the paid editing for which he was blocked per se, but the sockpuppetry. - Bilby (talk) 13:01, 26 May 2016 (UTC)
SOCKing has been against policy for ever. See ToU on following policy Jytdog (talk) 05:07, 27 May 2016 (UTC)
  • If I understand this chain of events correctly, Woods was fired for running a paid editing business out of his company computers. If so, and that is the most unfavorable interpretation of what happened, then the WMF instigated a user's firing. I'm not comfortable with that, sorry anti-paid-editing peeps. Particularly since the TOU back then did not explicitly ban paid editing. Yes, the TOU does have general statements concerning proper behavior but I find this case disturbing, much as I despise the business that Woods is in. Coretheapple (talk) 20:29, 28 May 2016 (UTC)
I also disagree with contacting a person's employer to complain about their editing practices. However, to be more accurate, Mike Wood was not banned and blocked for paid editing - he was initially blocked, and subsequently banned, for engaging in large scale sockpuppetry. That was against the Terms of Use, as it was an attempt to deceive people into thinking that the account belonged to someone else. After being blocked he continued to edit Wikipedia as a paid editor, leading to his ban, and has since that date engaged on ongoing paid editing through new accounts.
There is a story floating around that Mike Wood was banned for paid editing - it is important to be clear that this is not the case. His ban was for deceptive practices which he undertook in order to run his paid editing business. - Bilby (talk) 00:24, 29 May 2016 (UTC)
OK, thanks for clarifying that. I'm still not crazy about WMF contacting people's employers except in rare and extreme circumstances. Coretheapple (talk) 00:29, 29 May 2016 (UTC)
Nor am I. I do not, and will not, support contacting an employer simply over editing practices. It would take a lot more than that before I'd see it as an option. It shouldn't have happened. - Bilby (talk) 00:41, 29 May 2016 (UTC)
I have not seen any description other than Wood's for the fact that the contact was made. (it probably was) We have no idea what was said in the call nor who made it from "our side" nor why. What we do know is that his employer acted on whatever information they were given and that the employer decided to fire him. Saying that WP/WMF "got Woods fired" is Woods' spin on the story, and that is as far as that goes as far as I am concerned. His employer fired him.
I've never looked into the on-Wiki history of this. So I did. I was wondering what account it was - above someone said it was Morning277 and based on this that appears to be true. They were blocked for socking in August 2012 (the call to his employer was apparently made in Jan 2013). I hope anybody reflecting on this takes time to read the vast SPI archive to see what all unfolded there before the call was made. Wood had already sucked up a lot of community resources - time people could have spent doing more constructive things. Finally, I checked the archives of this page and there was no discussion here around that time. Here is the post that was made on this page when the Business Insider article appeared in Jan 2013; the next dif after the one I just gave is the one where it was removed. Quite a taunt; it is unsurprising that people responded.
Here is the AN where Wood was community banned in July 14 2013. This WT:ARB discussion from Oct 2013 is related to Wiki-PR but discusses Wood/Morning277 and Dennis Brown; it is also relevant and makes it clear that a lot of the discussions about what to do about Wood/Morning277 were held offline. Jytdog (talk) 01:54, 29 May 2016 (UTC) (note - i just edited this to make it more chronological and added dates and other clarifications Jytdog (talk) 18:31, 30 May 2016 (UTC); added dif of the edit Jytdog (talk) 18:33, 30 May 2016 (UTC))
I would be curious to hear what the WMF has to say about this serious allegation. Coretheapple (talk) 14:55, 29 May 2016 (UTC)
What serious allegation are you referring to? Jytdog (talk) 01:48, 30 May 2016 (UTC)
Keep in mind, running a private business using your employers hardware/network will get you fired in pretty much every country. Blaming the WMF/Wikipedian that contacted them for him getting fired is a bit pointless. You might as well blame the restaurant customer who spots the waiter stealing from the till and tells the manager. Only in death does duty end (talk) 15:36, 30 May 2016 (UTC)
I don't think so. AFAICT bullying works on employers 100.00% of the time. You start a poison pen campaign against some employer, launch a Twitter feed, get it on the news, get a cop to ask them a couple questions, whatever route you go by, that calls negative attention to an employee, they get fired, regardless. And a lot of the employers don't wait for it to be highly publicized before they do so. Bullying is not just a bad idea, it's the law. Wnt (talk) 21:41, 30 May 2016 (UTC)
I think that the claim made that the WMF "got him fired" is serious, notwithstanding the questionable venue in which it was made. I'd like to see what the WMF has to say about it, rather than just assume that it is a false charge. I'm not sympathetic with any paid editor but I get queasy when I hear things like this. Coretheapple (talk) 02:33, 31 May 2016 (UTC)
fwiw, his writing at the end "I guess I should thank everyone involved for pushing me into a more profitable career." pretty much blew a huge hole in any effort he might make to collect damages. Jytdog (talk) 13:09, 31 May 2016 (UTC)
He may be completely full of crap for all I know, but these are serious allegations nonetheless. WMF strikes me as conservative and cautious, and it just doesn't make sense they'd go around "getting someone fired." It may be that they reported a particular source of abuse and that the company tracked it down to this person. That is not getting him fired. I'd like to know the other side. Coretheapple (talk) 15:17, 31 May 2016 (UTC)
I think the WMF is conservative and cautious, and they are not known for even paying much attention (it isn't really their job) to detailed problems with abusive sockpuppeteers. In my own opinion, investing more staff resources in reporting ongoing longterm abuse from corporate networks to the network administrators would be a good investment. But as things currently stand, and I haven't looked into this at all, it seems highly unlikely to me that the WMF did anything at all here.--Jimbo Wales (talk) 21:14, 3 June 2016 (UTC)

False information and assumptions

Removed edit from the usual banned editor. Please use the unprotected page if you'd like. Smallbones(smalltalk) 13:26, 27 May 2016 (UTC)

  • Quote from the usual banned editor removed - Smallbones(smalltalk) 02:13, 28 May 2016 (UTC) - I'm willing to restore this legitimate IP query under my own handle and at my own risk in accordance with my belief that free speech and a variety of perspectives is a good thing and that censorship is not. It doesn't sound like GK's prose, for what it's worth. Carrite (talk) 21:12, 27 May 2016 (UTC)

Carrite, See WP:UP#OWN "Traditionally Wikipedia offers wide latitude to users to manage their user space as they see fit." ArbCom has made it clear several times that this applies to Jimbo's talk page.

Jimbo says at User:Jimbo Wales "Over 3,000 Wikipedians monitor my user and user talk page via a watchlist, and I trust them to edit and remove errors or attacks." i.e. that others should moderate his talk page.

I've moderated this page about 4-5 times and been dragged to ArbCom on it about 3 times. Each time they've upheld the right of Jimbo to have others moderate his talk page. And Jimbo has said there that he sees nothing wrong with my moderation. He has specifically said that others need to moderate because some folks are not there to engage in a discussion and that there are trolls who should be removed.

He has mentioned the usual banned editor specifically several times. Forcing that person's text onto this page via a quote just looks like a cheap trick trying to get around the rules.

Since Jimbo is involved in most discussions here, I think you can see why it is difficult for him to moderate this page directly himself, and why it is better to have others doing the moderation. For example you can probably imagine some trolls would love to say "Jimbo Wales kicked me off his talk page, rather than answer my questions." As a matter of fact, I've seen where the usual banned editor says that in a fourth rate online "newspaper." I don't take orders from Jimbo or anybody else on this - it is my judgement only. However, if Jimbo were to say - by any means available, public or private, that he thinks my judgement is wrong, either in general or in a specific case, then I would respect his views.

The usual banned editor can edit at User talk:Jimbo Wales/Unprotected if he just wants to communicate with Jimbo. But if he is just trying to gain attention by posting his opinions (which always degrade into trolling) on this page, rather than on his own fifth-rate website that nobody takes seriously, then just forget it. Consider it automatic that he is reverted on this page, whether a direct edit, a quote, or in Morse code sent by the Dali Lama. Of course if Jimbo wanted to quote him, that's another matter.

So feel free to ask Jimbo if you can insert the banned editor's comments - but after 10+ years of being trolled and harassed by that editor, I doubt Jimbo will agree.

Smallbones(smalltalk) 02:13, 28 May 2016 (UTC)

Yes, except when JImbo has not backed you up. Stop puffing yourself up this way Smallbones. I do agree with your suggestion they just ask Jimbo. Jytdog (talk) 05:25, 28 May 2016 (UTC)
  • Let's try that again. You, Smallbones, have no evidence that this IP is a "banned editor." Stop casting aspersions. We won't even get into your bullying behavior. Restoring again the IP's deleted legitimate query over my signature: removed again Smallbones(smalltalk) 22:35, 28 May 2016 (UTC) Fin. Carrite (talk) 16:50, 28 May 2016 (UTC)
    • You haven't got a leg to stand on Carrite. Just ask Jimbo if my removal of this quote is consistent with his moderation policy, or whether he would like to make an exception for this particular case. Otherwise the usual banned editor can edit on the unprotected page. But it is automatic that the usual banned editor will be reverted on this page. End. Smallbones(smalltalk) 22:35, 28 May 2016 (UTC)
@Smallbones: Your position is not reasonable. You are saying that because editors foolishly agreed to the notion of creating a separate page that is left unprotected so that IPs can post to it, that authorizes you to remove a quotation by an editor in good standing because it came from a source who was able to post here. There is zero logic, not even DEA-grade logic, to connect the one with the other. Some IP believes that some people here have "perpetuated" misinformation about Mike Wood's engagement on Wikipedia and is curious whether the Wikipedia community is "solid enough" to allow him to offer corrections to that point of view, and suggests that this conflicts with claims of openness or transparency. It's not a bad argument, really. If Wikipedia is going to generate stories that reflect negatively on someone's character, they really *do* deserve opportunities for fair response, and I thought that the "open door policy" that had been advertised for this page was one such opportunity.
I should add although I do not support enactment of EU-style laws on databases and privacy, many of them are good practices, and one of these practices (codified e.g. at item 36 here) is to allow people to correct databases that reference them. Some might say that letting them mail Arbcom is a way to do that, but it's not so. To illustrate: I were to post something libelous about Mr. Wood or anyone else, ArbCom would not actually take responsibility for it; they might take it down if they don't like it, but if they decided it was OK, even if a court disagreed, they would not be held liable for that either. And - thank God - Wood would not have any easy mechanism to send me cease and desist letters in the mail, either. The downside of that though is that here I could be, typing away data about Wood to go on some Wikipedia long term abuse page, and it could be complete balderdash, and short of getting a subpoena and going to WMF and then my internet provider, he doesn't have a route to correct it. I would be the one maintaining that database, not ArbCom, not Wikipedia, at least on that day. Unless, that is, we let him post somewhere on Wikipedia and simply tell me "@Wnt:, that story is wrong, and here's why." He could ping me, or he could post here or on his talk page and someone else could tell me, and that way I might be talked out of believing something bad about him before it even rises to the point of libel. This should pretty clearly be to our interest, and his, and there is no great downside to it, because we don't have to let him post everywhere, just somewhere. Oh sure, someone could use the mechanism to harass, but there are a lot more straightforward ways to harass! Wnt (talk) 11:57, 29 May 2016 (UTC)
The place for banned editors to post is at User talk:Jimbo Wales/Unprotected. What's the problem? Smallbones(smalltalk) 04:26, 30 May 2016 (UTC)
As I recall the drama that ensued in the past on this, Jimbo has explicitly delegated to other editors the monitoring of his talk page for removal of banned editors he doesn't want on this page. I personally wouldn't engage in such activity as I think it is wearisome, but if he wants to he has Jimbo's backing. That's been explicitly stated. Coretheapple (talk) 14:49, 29 May 2016 (UTC)
Well, if you truly know this is a particular banned editor Jimbo said he doesn't want, I guess I can't really tell you you're wrong to delete him here; yet I think it's important that any editor, even a banned one, has an open door policy somewhere to help head off trouble. Wnt (talk) 22:47, 29 May 2016 (UTC)
The place for banned editors to post is at User talk:Jimbo Wales/Unprotected. What's the problem? Smallbones(smalltalk) 04:26, 30 May 2016 (UTC)
Well, you said "banned editors he doesn't want" before, now you said all banned editors, which seems contrary to the open door policy as I understood it; honestly, I don't know what he wants. I can't deny it's his page to decide on the rules, and the same is true of the "unprotected" version but if such rules have been decided on I wish I knew where they were. I thought I'd have noticed if he actually said all that; last I remember people were complaining because he didn't object to banned editors. Wnt (talk) 11:03, 30 May 2016 (UTC)
  • Anyone smart enough to edit WP can easily find the IP's question in the history. And the sanctimonious, thuggish, axe-grinding, edit-warrior behavior of some overwrought people is also readily evident to anyone who can read. Carrite (talk) 17:51, 29 May 2016 (UTC)
That sure has "scripted reply" written all over it. Wild guess: are you in telemarketing? Coretheapple (talk) 02:36, 31 May 2016 (UTC)
No. My turn: are you a professional badger trainer? Carrite (talk) 16:22, 31 May 2016 (UTC)
Weak. Coretheapple (talk) 17:15, 31 May 2016 (UTC)

Case study: Bullies claiming to be bullied

Carrite above seems to be referring to me when he writes "And the sanctimonious, thuggish, axe-grinding, edit-warrior behavior of some overwrought people is also readily evident to anyone who can read." Really? He's also writing in that section about the person I call "the usual banned editor", about Wikipediocracy in general, and about a banned editor who readily admits that he is a serial violator of our paid editing rules. The usual banned editor has been banned for over 10 years now, but has seldom gone more than a couple of weeks without harassing Jimbo on this page during that time. He claims to operate a paid editing service in violation of our rules. He has personally harassed me. Wikipediocracy - which he appears to own - is well known for harassing and outing editors. Jimbo has several times clearly and directly told the usual banned editor to stay off this page. He has approved moderation by myself and other editors that removes the usual banned editor's "contributions" to this page. In short, the usual banned editor is widely recognized as a serial troll, and both Jimmy and ArbCom have approved my removals of his edits. Carrite and the ube can't be bothered to take the simple step of either posting on the User talk:Jimbo Wales/Unprotected page (instead of this page) or asking Jimbo whether he can now post on this page.

Yet @Carrite: is accusing me of being a bully for removing the usual banned editor's edits. Correct me if I'm wrong on that, Carrite. I will not take this to ANI - I don't want to play his game. But if any admin reads this, I wouldn't mind them blocking Carrite for a day, or at least giving him a warning for the most bizarre personal attack of all time.

A pattern of bullies claiming to be bullied is widely reported. This appears to me to be a classic case. Just taking a few minutes to check the facts should make it clear who is the real bully.

My advice for dealing with Wikibullies:

1. Know the rules, follow them precisely and then stand up to the bullies.

2. Shaming the bullies when they make outrageous statements.

That is what I am doing now.

Smallbones(smalltalk) 17:19, 31 May 2016 (UTC)

No, what you are doing now is acting as a self-appointed censor, making bad faith assumptions about an IP, engaging in edit-warrior-type behavior to defend your unilateral decisions (which others have warned you against above) and taking up a lot of time and space with some rather pointless and unfocused prattle. You are the "bully claiming to be bullied," not I. Carrite (talk) 06:13, 1 June 2016 (UTC)
It's good to see that we can agree on something - that one of us is a bully claiming to be bullied. So that future readers can see what I consider to be a clear case of this type of behavior, I'll likely copy this section to User:Smallbones/Case study - Bullies claiming to be bullied. I'll summarize briefly here and if @Carrite: can summarize in a similar amount of space, I'll include his summary too.
  • The "usual banned editor" is a serial troll, who has harassed Jimmy on this page many times in the 10 years that he has been banned
  • Jimmy has requested that other editors moderate this page and has specifically told the ube to stay off this page.
  • I have removed the ube's edits on this page several times, including edits quoting the ube's comments inserted by others.
  • I've been taken to ArbCom several times on this, and ArbCom confirmed the Jimmy has a right to let others moderate the page for him, and Jimmy confirmed that my moderation was within his personal talk page moderation policy.
  • Carrite reinserted the ube's comments on this page after I removed them and refused to use the alternative of editing at User talk:Jimbo Wales/Unprotected, or simply asking Jimmy to change his policy of not letting ube edit on this page.
Carrite can you summarize your position in about 5 points of about the same length? Smallbones(smalltalk) 13:36, 1 June 2016 (UTC)
According to Wikipedia:Arbitration/Requests/Case/Banning_Policy#Tarc's_conduct, the ArbCom has confirmed that when Tarc (or Carrite, or anyone else) reinserted the comments of a banned editor, they take responsibility for those edits. They did so again at #Banned editors on the same page. You are well within policy to initially remove the comments of a "banned editor", but continuing to edit war to remove the comments after they have been reinserted is directly against what the ArbCom ruling states. I thus invite Carrite to restore the comments of any "banned editor" he deems necessary, but also know that he is personally responsible for anything he reinserts. Libelous and harassing material, of course, will be removed. Pinguinn 🐧 14:44, 1 June 2016 (UTC)
Which same case recorded only slightly further down that Smallbones [yes, specifically Smallbones] is warned to refrain from edit warring and needlessly inflammatory rhetoric in the future. Further instances of similar misconduct may result in serious sanctions. If anyone chooses to reinsert the comments from the editor that Smallbones is assuming is Thekohser, then Smallbones should leave them there.  — Scott talk 17:01, 1 June 2016 (UTC)
The overriding decision in that case is Wikipedia:Arbitration/Requests/Case/Banning_Policy#User_talk_pages, that Jimmy gets to decide how his talk page will be moderated. I interpret his policy to include removing any comments from the usual banned editor in whatever shape or form they appear. The cheap trick of quoting ube does not pass muster, in my book. The civilized way to dispute this is simply to ask Jimmy. But everybody has a pretty good idea of the likely answer, don't they? Smallbones(smalltalk) 22:32, 1 June 2016 (UTC)
LOL. You were specifically censured for misconduct. No amount of absurd wikilawyering will change that.  — Scott talk 23:40, 1 June 2016 (UTC)
A vs B:
A = Would-be paid editor.
B = Anti-paid editing.
Muffled Pocketed 06:27, 1 June 2016 (UTC)
@Smallbones and Carrite: Guys, please, chill pills! As I complained above, we don't actually have a clear written guideline, so everyone's making best efforts. I'd like to have a full "open door policy", and I don't like it when a good argument gets deleted, but even I will get fed up sometimes if they mess with my edit button [43], and anyway it's up to Jimmy Wales. This is a fair difference of opinion between well meaning editors. Wnt (talk) 14:39, 1 June 2016 (UTC)
I agree. To all: more light and less heat, please. --Guy Macon (talk) 16:53, 1 June 2016 (UTC)
Following recommendations

<snip> Taking ownership of the above comments, as I'm quite allowed to.  — Scott talk 23:36, 1 June 2016 (UTC)

I will grudgingly admit that they have not done quite such terrible things recently, but I was there (and here, in the archives of this page) when a lynch mob of editors based at Wikipediocracy went after User:Fæ, one of Wikipedia/Commons' most productive and well-respected editors, out of some kind of salacious condemnation of a couple of innocent photos. Wikipedia never lets bygones be bygones, especially when it is wrong, and as long as Fae is under any kind of restriction whatsoever, there will be good reason for people to bear grudge against the lies and technical tricks and the tidal wave of utter balderdash levied by the cyberbullies from Wikipediocracy in that situation. As for the article, it has enough connected contributors making large reverts in its history that I wouldn't trust it to be fully comprehensive or neutral. As for the ban, well, [44] says that TheKohser was blocked 2006-08-21 to 2006-09-12, 2006-09-18 to 2007-07-20, 2007-07-21 to 2009-06-23, 2009-08-05 to 2009-08-07, 2009-09-27 to (2016-06-01 and counting). That leaves 7 days + 1 day + 44 days + 51 days that he was unblocked in 10 years less 82 days, i.e. he was actually blocked 185 days less than 10 years; but he was also blocked 2 days before then as MyWikiBiz. So fine, 9.5 years, not 10, you have a point there. Wnt (talk) 00:11, 2 June 2016 (UTC)
I am not sure what is to be gained - aside from deliberate antagonism - by proxying for a banned editor and/or representative of a troll website over what is little more than a nitpick. Other than a brief (and miserably failed) attempt at allowing an unban in the summer of 2009, Kohs has been banned almost continuously since late 2006. But technically, the anon is correct. Kohs has been banned for a little under 10 years, not a little over. Resolute 00:18, 2 June 2016 (UTC)
@Scott: seems to misunderstand - I did *not* revert the above edit by the banned editor. I hatted it, as is quite common on talk pages. I want people to be able to see that edit (after a warning that it's his usual stuff), because it clearly shows the troll trolling. And it shows that he thinks this is all about him, that other people must follow the most picayune rules that he wants to impose, but that he thinks he doesn't have to follow any rules.
The problem with taking full responsibility for the edits of a troll is that you may not understand what the troll is actually saying. It would be much better if you put the points you think he is making into your own words. Just leave the troll out of it. But if you want to push the troll onto a page and make the point that he can edit wherever he wants, then you are taking full responsibility for whatever he said.
So for example, it looks like ube is saying that I'm lying by saying that he has been banned for over 10 years. Sorry, it turns out to be only 9.5 years. That's minor imprecision, not lying. Are you really willing to take full responsibility for this?
He is also saying that Wikipediocracy does not have a reputation for harassing other editors. Do you really want to say that? It's just laughable.
More seriously, he is saying that he has not harassed me. Make that you are taking responsibility for saying that ube has not harassed me. How do you know that? I'm willing to present the evidence to ArbCom by private e-mail, and I believe it is very convincing. So you've made yourself responsible for his trolling. He trolls, you pay the penalty.
I've noticed from your user page that you are an administrator both on Wikipedia and on Wikipediocracy. Do you think that an admin on Wikipediocracy should be trolling for the ube and still be an admin on Wikipedia?
I think you owe me an apology. Please re-hat ube's edit and clearly state that you are not taking responsibility for that edit. Smallbones(smalltalk) 01:23, 2 June 2016 (UTC)
TL;DR. I think we're done here.  — Scott talk 01:29, 2 June 2016 (UTC)
Cut out the ad hominem attacks, Smallbones. Whatever Scott does on Wikipediocracy has absolutely nothing to do with his comments here, smells like the same ad hominem attacks you've made about me in the past, and reak of the same attacks that got you a slap on the wrist at Arbcom. Knock it out. Ad hominems are not a mature and valid form of argument. It screams "I have nothing better".--v/r - TP 01:57, 2 June 2016 (UTC)
  • I've got nothing further to say, the IP's legitimate question is easy to find in the history if anybody missed it. I will chime in that Wikipediocracy is no monolith, it is more or less a message board with a blog attached to the front end. There are a wide range of perspectives exchanged there and nobody marches in lockstep. For example, WPO Administrator Hex and I are not exchanging Christmas cards, which doesn't mean I don't read his stuff and respect his point of view. WPO is not in existence for the purpose of outing, nor does it engage in such conduct frequently or without reason. —Tim /// Randy from Boise on WPO (but not on WP) /// Carrite (talk) 02:24, 2 June 2016 (UTC)
    • "WPO is not in existence for the purpose of outing, nor does it engage in such conduct frequently or without reason." - I'm glad they are so reasonable.
    • The bottom line is that the usual banned editor may edit at User talk:Jimbo Wales/Unprotected, but not on this page. That will be enforced automatically every time I see him here, until Jimbo says not to do it, or ube is given back his editing privileges.
    • Everybody seems to be in agreement that there is nothing more to say on this matter. Smallbones(smalltalk) 03:12, 2 June 2016 (UTC)
No, he won't. Let me clarify this for everyone. Any post from Mr. Kohs under any of his sockpuppets should be removed on sight, and anyone reposting to "take responsibility for the comments" is invited under the strongest possible terms to not do that, and if you do that, then again, the comments should be removed on sight. If you happen to see a question or argument from someone anywhere on the Internet and wish to ask a question about the subject matter politely in your own words, that would be different. But becoming a proxy for harassment by others is not acceptable behavior.--Jimbo Wales (talk) 11:53, 2 June 2016 (UTC)
What's the matter, Jimbo, do you have a problem addressing me by name? Since you feel that it's a problem for Kohs to point out that Smallbones is a hyperbolic liar, I'll just do it in my own words.  — Scott talk 14:13, 2 June 2016 (UTC)
No, Scott, I don't have a problem addressing you by name. It is just that I was not speaking to just you, I was speaking to everyone. I would like everyone who frequents this page to help me keep precisely this kind of argument off of it. I don't want you calling anyone a liar, and I don't want anyone telling lies about anyone. I don't want to talk about banned users, and I don't want to spend endless hours with yet the latest round of a (nearly, as we have painstakingly established for no reason whatsoever) 10 year campaign of harassment including but not limited to a facebook post by the banned user holding a gun and fantasizing about a gun battle with me, posts obsessing over my children and family life, etc. If you think it's a good idea to be a proxy for this person, then please I beg of you to step back and think about human feelings and decency and stop doing it.--Jimbo Wales (talk) 20:03, 2 June 2016 (UTC)
As an admin, you know "hyperbolic liar" is violating WP:NPA here. Kohs didn't even call Smallbones a liar; he said some things he said were "falsehoods", which allows diplomatically for a possibility of innocent error. Luckily, unlike Mark Fuhrman, Smallbones didn't make the comment about "10 years" under oath. I have seen several new editors with very reasonable frustration at WP guidelines get long term blocks for nothing worse than this, and I would suggest you retract it. This section looks like it just made a sharp dive in the direction of ArbCom and I'd guess you don't have long to pull out of it. Wnt (talk) 14:39, 2 June 2016 (UTC)
Oh please, Smallbones has thrown NPAs out all over this page. Those aren't even veiled attacks, they're overt. But I guess you and Jimbo only care about NPAs from people from bad sites. Smallbones sits on this page and blast away "bad people" and then attack them and Jimbo has just sanctified it. Unbelievable. Guess we have a new Lord Commander of the Jimbo Watch.--v/r - TP 16:20, 2 June 2016 (UTC)
Actually, you are mistaken about my position here. I have not sanctified it, I have not even commented on it. What I have said is that the right solution to the problem is not to let a banned user post here either directly or by proxy. Take it all somewhere else, I don't want to hear about the banned user, I don't want to hear from the banned user, I don't want anyone to attack him here or anywhere else, I want him to simply go away and leave me alone.--Jimbo Wales (talk) 20:03, 2 June 2016 (UTC)
@Jimbo Wales: Okay, that seems fair enough, then. So, can we take this as an exemption for this one particular person, and not the rule, to your invitations for banned editors to use you as a last line of appeals?--v/r - TP 22:16, 2 June 2016 (UTC)
I would suggest if Jimbo and I are in the same boat, it's a really big boat. I made some of the arguments that Smallbones disagreed with in the second edit you cite, and my feeling was that he was disputing my arguments, not making a personal attack on me. There's some argument to be made he was making a personal attack on Wood in the first edit, perhaps even the second; but he focuses primarily on Wood being "disgusting" for specific activities, listed one by one. For Scott to post the "liar" bit on its own, without even saying which point(s) he's saying it about - I honestly don't know which - gives the feeling it is just a general comment on the person rather than a dispute on the facts. Which is frustrating, because it makes it hard to refute, among other things. Anyway, I'm not taking anybody to ArbCom, but I've seen it happen from threads like these and it's a shame. Wnt (talk) 16:32, 2 June 2016 (UTC)
Snipped, and subsection title altered per talk page guidelines. --Francis Schonken (talk) 12:10, 2 June 2016 (UTC)
This is a useful clarification. Personally, I don't think we do Kohs any favors by allowing him to confront us, but it's your page and your call. But to avoid future dispute... I take this instruction to be specific for Kohs, not for any other banned editors who might want to appeal their cases or discuss Wikipedia social issues? Wnt (talk) 12:40, 2 June 2016 (UTC)
@JW. And how do you propose that we are to identify Greg Kohs "sockpuppets," Jimmy? Are we to guess or to follow established site procedures? Carrite (talk) 15:17, 2 June 2016 (UTC)
In this case, I'd just guess according to your best ability and err on the side of deletion. We have a perfectly effective safety valve of the unprotected page.--Jimbo Wales (talk) 19:54, 2 June 2016 (UTC)
Maybe we ought to consult Smallbones, as he seems to magically know every time. Pinguinn 🐧 18:18, 2 June 2016 (UTC)
Use judgement. Also see Some folks are easily identifiable. Smallbones(smalltalk) 18:42, 2 June 2016 (UTC)
My unsolicited and I am sure unwelcome advice to Smallbones is that if a cadre of jugheads, don't-attack-me-while-I-attack-you-ers and administrators-who-ought-to-not-be-administrators continue to bust his chops every time he complies with Jimbo's wishes on the oh so hard to identify banned user, he should cease doing so and let Jimbo take care of it himself, or better still take action against the aforementioned cadre. "What, me worry?" is an excellent philosophy in dealing with knuckleheads. Coretheapple (talk) 19:35, 2 June 2016 (UTC)
Don't forget to include the veiled-attacks-arnt-attacks-if-there-are-no-bad-words cadre.--v/r - TP 22:17, 2 June 2016 (UTC)
Or the grudge-holders-who-personally-attack-people-on-their-user-pages-and-aren't-held-accountable-because-they're-administrators. Coretheapple (talk) 22:43, 2 June 2016 (UTC)
Yeah, I'm sure every one of us falls into one of the aforementioned groups.--v/r - TP 23:00, 2 June 2016 (UTC)
Ya think? Coretheapple (talk) 23:09, 2 June 2016 (UTC)

WMF Administrative Cost

Here is a question for WMF Board member Jimmy Wales... We see from THIS META DOCUMENT that spending for the ED of WMF and her "advisor" went from $201,685 plus an amount too small to be reported in Fiscal Year 2013-14 to $609,490 in Fiscal Year 2014-15. What is the cause of this tripling in spending? Why was Sue Gardner given what effectively looks like a 50% raise to move from ED to "Advisor"? Carrite (talk) 15:09, 2 June 2016 (UTC)

I have pinged the WMF for details.--Jimbo Wales (talk) 19:51, 2 June 2016 (UTC)
Here's the answer: "The Board felt that Sue’s knowledge and experience was essential for the organization as it went through a major transition. Sue served for 6+ years as WMF’s Executive Director and led it through a tremendous period of growth. As a special advisor, she designed and executed an in-depth handover process and counseled the new ED, the Board, and other staff. In this role she also advocated publicly for the Wikimedia movement, projects, and Foundation through public speaking and media. Patricio shared a more detailed breakdown with the Signpost." Sue was not given a simple 50% raise, it is more complicated than that, and the increase in spending was temporary during the handover/transition.--Jimbo Wales (talk) 15:01, 3 June 2016 (UTC)
Thanks for that answer. Follow up question: Since WMF now finds itself in the midst of yet another "major transition," are there plans afoot for additional six-figure consulting fees for Sue Gardner or any other "advisor"? Carrite (talk) 16:10, 3 June 2016 (UTC)
Do you guys have any idea how much office space you just lost Drmies in the corner there??? Muffled Pocketed 22:34, 2 June 2016 (UTC)
Also solid, sheetrock walls are cheaper for air conditioning, so those corner windows must go. -Wikid77 (talk) 22:51, 2 June 2016 (UTC)
It would be nice if we had a salary list for all employees, I believe over $26m (right?) was paid out in salary between August 2014-August 2015 or so..I find it odd that the WMF ED gets paid a lot than my country's PM (No Form 990 yet for 2015-16)..--Stemoc 23:29, 2 June 2016 (UTC)
I hate to admit it but this is a good question. Good work, Carrite. A stopped watch is right twice a day, and this must be that time. It does appear that WMF administrative expenses have doubled at the very highest echelon. Personally I could care less, but some do. Coretheapple (talk) 23:33, 2 June 2016 (UTC)
Well, they were higher for that period of transition, but no, they haven't permanently doubled.--Jimbo Wales (talk) 15:01, 3 June 2016 (UTC)
I don't wear a watch and I'm right more than twice a day, thank you very much. Carrite (talk) 01:49, 3 June 2016 (UTC)
Good! I'm rooting for you. Bring it up to three times and I will send you a barnstar and a surefire method (administrator tested!) to make money from Wikipedia. Coretheapple (talk) 12:27, 3 June 2016 (UTC)
Coretheapple — Get back to me on that with a little less attitude when YOUR CONTRIBUTION CHART looks like MY CONTRIBUTION CHART, okay??? —Tim /// Carrite (talk) 16:38, 3 June 2016 (UTC)
I'll get back to you with a little less attitude when you stop being wrong all the time. Coretheapple (talk) 16:48, 3 June 2016 (UTC)
Pizza money. Trust me. HalfShadow 00:36, 3 June 2016 (UTC)
  • Cheese and rice. Six figures for all of them? You're welcome, all y'all, on behalf of all the editors here, not to mention the arbs and the admins and the crats. The highest one makes six times what I make, and my PhD isn't honorary at all. I wonder what it's like to walk through a grocery store and buy steak without a second thought. Drmies (talk) 01:19, 3 June 2016 (UTC)
I'm not really clear on what the mention of an honorary PhD means in this context. I don't know of any staff members with an honorary doctorate. I've got a few, but I don't even accept the normal airfare expenses of a board member to attend board meetings, much less a salary.--Jimbo Wales (talk) 15:01, 3 June 2016 (UTC)
Well this is on the Form 990 I presume. So people can go over to a repository for said forms and take a look. It seems like a lot of bucks. I am sure there is a good explanation, for otherwise there would be 600K spent on a couple of top execs. Which is a lot. not that I care, as I said It's really a donor thing, at the end of the day. I am not a donor so... what, me worry? Coretheapple (talk) 01:29, 3 June 2016 (UTC)
The Form 990 is here[45]. Yes, it would appear that $600K appears to be spent on two high-ranking positions. However, I note that there has been a substantial increase in revenues as well. If the persons at the top were largely response for that increase, this is not necessarily such a terrible thing. I know that this will surprise many people, but fundraising is one of the top priorities of a nonprofit organization. You heard it here first. Coretheapple (talk) 01:42, 3 June 2016 (UTC)
Well, that's nice, and I saw that Sue Gardner took a bunch of credit for good things. But you know, it's the chancellor of the university who takes the credit for whatever goes well, when everybody knows that it's the teachers and the janitors and the administrative staff (not the administrative hot shots) who do the real work. So it goes here. You can't raise funds of your product is a pile of crap. If we're doing good, it's cause we deliver a good product. I think one of those products is on the front page tomorrow, and it's not Sue or Erik or Jimbo or the legal team who wrote that up--it was me and he who should not be named here. Drmies (talk) 01:47, 3 June 2016 (UTC)
Hold on a second! Total revenue went up to $78 million (Form 990, part VIII, line 12) from $53 million when the Fundraising Department donations went down to $53 million from $55 million in 2014. While the calendar year is not the fiscal year, if, as I suspect, Sue was responsible for raising the approximately $25 million difference, perhaps those who think her salary is too high might consider that 2.4% is substantially less than the typical 5% to 10% paid as finders' fees by organizations compensating development officers in that manner. EllenCT (talk) 04:23, 3 June 2016 (UTC)
She did this singlehandedly? And do the editors who produce the content that apparently warrants that generosity get a cut? I know, that's a stupid question. Drmies (talk) 04:36, 3 June 2016 (UTC)
This is Wikipedia's business model. Coretheapple (talk) 12:47, 3 June 2016 (UTC)
[46] EllenCT (talk) 13:09, 3 June 2016 (UTC)
It most certainly is not a stupid question! Please see Cameron, J. and Pierce, W.D. (1994) "Reinforcement, Reward, and Intrinsic Motivation: A Meta-Analysis" Review of Educational Research 64(3):363–423. EllenCT (talk) 04:48, 3 June 2016 (UTC)
@EllenCT: I wouldn't put too much stock in that meta analysis, it has been seriously criticised

doi:10.3102/00346543066001005: "Cameron and Pierce’s overly simplistic conclusion has little theoretical or practical value and is instead the direct consequence of their systematic and consistent misuse of meta-analytic procedure"

doi:10.3102/00346543066001033: "Cameron and Pierce’s (1994) conclusion ... is a misrepresentation of the literature based on a flawed meta-analysis. Their call to abandon cognitive evaluation theory is more an attempt to defend their behaviorist theoretical turf than a meaningful consideration of the relevant data and issues."

doi:10.3102/00346543066001001: "Cameron and Pierce’s meta-analysis ... is unpersuasive by virtue of its methodology, its tendency to ignore important distinctions, and its failure to include certain evidence"

Cameron and Pierce defended their work - doi:10.3102/00346543066001039 - concluding that "the results and conclusions of our meta-analysis are not altered by our critics’ protests and accusations"

doi:10.3102/00346543071001001: "(Cameron & Pierce, 1994) concluded that the undermining effect was minimal and largely inconsequential for educational policy. However, a more recent meta-analysis (Deci, Koestner, & Ryan, 1999) showed that the Cameron and Pierce meta-analysis was seriously flawed and that its conclusions were incorrect."

And, the 1996 Cameron response was also poorly received, doi:10.3102/00346543071001043: After having "pointed out flaws in Cameron and Pierce’s (1994) meta-analysis. Cameron’s (2001) commentary did not reveal any problems with our meta-analysis, nor did it defend the validity of Cameron and Pierce’s. Instead, Cameron referred to a fourth meta-analysis by her group; little detail was presented about the new meta-analysis, but it appears to have the same types of errors as the first three. Cameron also presented a new theoretical account of reward effects—the fourth by her group, which sequentially abandoned the previous ones as they were found wanting. Cameron concluded again that there is no reason to avoid using performance-contingent rewards in educational settings, yet her application of the research results to education lacks ecological validity."

This debate has gone back and forth for ages, but it is my understanding that Cameron's group's view is considered to be in the minority. EdChem (talk) 14:14, 3 June 2016 (UTC)

On the contrary, my understanding is that Cameron & Pierce prevailed in the face of fierce critiques.[47][48] I didn't click on all your links. Are any of them more current than those two? EllenCT (talk) 14:34, 3 June 2016 (UTC)
For non-profits and charities this will often be down to one person yes. No your (and other editors) contributions to wikipedia are not a big factor in large donors contributions when compared against the schmoozing ability of someone who can get them to open their wallets. If this was actually what Sue Gardner was doing, comparatively she is getting equal or less than other people doing similar fundraising jobs at similar size organisations. -ninja edit- I dont particularly like this is the way the world works, but it is how it works in fundraising huge amounts (million+). Only in death does duty end (talk) 07:22, 3 June 2016 (UTC)
Is there a list somewhere of vague references to real "bad" people so I know who we're talking about. He who should not be named and the certain banned user should both be included. That way I can keep up with the conversation.--v/r - TP 02:56, 3 June 2016 (UTC)
TParis, I cleverly hid the name in a blue link. Don't tell anyone. Drmies (talk) 04:14, 3 June 2016 (UTC)
@Drmies: Thank you for being considerate of the tech-illiterate like myself, but I may still need a list for next time.--v/r - TP 05:00, 3 June 2016 (UTC)
I agree that the volunteer model is unsatisfying. After a sustainable endowment perpetuity is established, or even while it is being established, the Foundation should definitely experiment with ways to compensate editors. Please see my suggestions at meta:Talk:2016 Strategy/Draft WMF Strategy#Goal proposal for strategic priorities 2 and 3: rewarding contributors.
Which reminds me, Jimbo, do you know when the Strategy Critical Question Synthesis is due out? It has been "(coming soon)" since the Strategy consultation closed in mid-February. EllenCT (talk) 04:44, 3 June 2016 (UTC)
No, I don't know.--Jimbo Wales (talk) 15:01, 3 June 2016 (UTC)
It was only a matter of time before Too Much Money for WMF led to proposals of this nature... I'd like to see The Usual Suspects get as worked up about this as they do about paid editing freelancers, but I doubt they will. Carrite (talk) 10:50, 3 June 2016 (UTC)
The problem is that discussions like this are almost invariably commenced either by a banned paid editor or by his acolytes and pals, such as you. That doesn't help. So what you really should do is have someone with more credibility post messages like this on this talk page. (There is also a paid-editing method, administrator-tested, involving much the same technique. PM me for details!) Coretheapple (talk) 13:29, 3 June 2016 (UTC)
Are you saying you are in favor of paid editing funded from corporate PR departments who want to whitewash their articles, but not from donors? EllenCT (talk) 11:15, 3 June 2016 (UTC)
Conspiracy theories involving the WMF are favorites of the paid editing community and their apologists, because they are a "look! look! there's worse over there!" form of Chewbacca defense to the impact of the gravy train on Wikipedia's credibility. That's why you see a stream of WMF criticism out of that quarter (a businessman and his unpaid acolytes). On occasion a valid question is asked, such as this one. Coretheapple (talk) 12:42, 3 June 2016 (UTC)
Indeed.--Jimbo Wales (talk) 15:01, 3 June 2016 (UTC)
Well, that's an interesting reduction of Wikipediocracy — the "paid editing community and their apologists"/"a businessman and his unpaid acolytes." I'd say paid editing is probably issue no. 12 on the hit parade over there. Or lower... Things would start with programming incompetence, move to the chronic wasting of donor funds by the bureaucracy, moving to administrative abuse of "regular editors," moving to the deformity of the governance structure, moving to improper handling child protection matters if I were to rank a Top 5. Of course there is an anti-JW drumbeat as well in the background, it should be said, since this project came from somewhere and there has been one unchanging feature about the WMF Board of directors. I'm sure I could come up with 6 or 7 bigger issues about WP at WPO very easily before the paid editing question ever reared its head. So, this is completely a strawman argument here, a diversion. See: Goldstein, Emmanuel. (For what it's worth, GK is to paid editing on WP what Just Add Water Records is to the music industry... And he edits within our NPOV and sourcing standards, which is why his pages stay up and he never gets caught — not that he could be banned anyway until we institute One Account and Sign In To Edit practices...) Carrite (talk) 16:21, 3 June 2016 (UTC)
The stopped watch speaketh, but this is not one of the two times a day it is right. It's not a "reduction of Wikipediocracy." It's a description of you. I made no mention of that website so there is no reason for you to wrap yourself in its banner. How do you manage to be wrong so often? The fact is that if this WMF stuff was not advanced by trolls and people with agendas it might get a lot more mileage than it does. The spectre of a paid editor lecturing people on the morality of the WMF is an exercise in hypocrisy that is visible to most people. As for the stuff you added: thanks for the stirring defense of "NPOV paid editing" but I hope you're aware that your invoking of Orwell is Orwellian. Coretheapple (talk) 16:45, 3 June 2016 (UTC)
Break down the above paragraph and one is left looking at a steaming plate of ad hominem argument with a side order of incivility. Carrite (talk) 16:52, 3 June 2016 (UTC)
Not incivil at all. Wrong again. You're going for a record. Coretheapple (talk) 16:56, 3 June 2016 (UTC)
@Jimbo Wales: Please take a moment to yell at me, lest I be accused of brownnosing, a practice that one of your administrators seems to feel is rampant in this very discussion[49]. Coretheapple (talk) 16:03, 3 June 2016 (UTC)
It is worth saying right here, right now, that all paid editing is a slippery slope that leads to articles filled with public relations and perception management, not a neutral encyclopedia. Paid editing must be ended without delay. And with all due respect to the WMF, money corrupts. It's time to put a cap on the WMF budget and stop the endless growth. This process will likely have to come from the community of editors, if it is possible. Jusdafax 12:56, 3 June 2016 (UTC)
Pulling the plug on Wikipedia would not stop paid editing, it would simply result in the use of AdWords to set Google Answer Box content. Always go with the lesser of two dystopias. EllenCT (talk) 13:14, 3 June 2016 (UTC)
Meh. The WMF is not generally about content, explicitly legally so (so sorry, you volunteered to write that article, but don't complain you volunteered, that is just very odd). The WMF is about infrastructure for multiple projects and whatever that kind of weird thing called the "movement" is ('something something . . . free knowledge . . let's have an event and an affiliation and a committee, . . . is this a cult, etc, etc.'). As for the editors/entities who want to create content to make money off the content they create, that's a different kettle of fish, and yes it's odd when all these fish and fowl tangle, you get the 'professional WMF', 'professional WMF critic', 'professional manipulator of the WMF project's content', etc, and then they sometimes wear double or triple hats. Alanscottwalker (talk) 13:19, 3 June 2016 (UTC)
The WMF should be more focused on shielding its volunteers from legal liability. That is my only bone to pick with them, that and being downright cowardly in dealing with paid editors. Coretheapple (talk) 13:35, 3 June 2016 (UTC)
I am sympathetic to the second part - I think the WMF should be much more proactive in leading the fight against paid advocacy editors, although admittedly it is a difficult issue for them to deal with - there are no magic bullets. But on the first part, that the WMF should be more focused on shielding its volunteers from legal liability, I'm absolutely in agreement with the sentiment, but I think the Foundation does a great job on this right now, and I'm always open to hearing about case where there was a failing, or additional principles that could be instituted to improve matters.--Jimbo Wales (talk) 15:01, 3 June 2016 (UTC)
Well for starters the WMF could release every subpoena for information on Wiki editors and every lawsuit against a Wiki editor, with summary information on the Foundation's action, if any, in response to each. It's hard to make suggestions without information. Coretheapple (talk) 18:22, 3 June 2016 (UTC)
Ok, let's chew on this further. According to the WMF transparency report there were only 25 requests for user data, and only 4% of them resulted in any information at all being produced. (4% of 25 is 1 case, to be clear.) Assuming that we are collectively happy with the 24 cases where nothing was produced, our only actual concern is the 1 case (in 6 months) where information was produced, right? So it doesn't strike me as particularly beneficial to know all the details of the 24. And of course there's a significant privacy downside to even disclosing that a (possibly unjust) legal request has been made - I can imagine trolls going after a user with repeated endless claims for years that he or she is either doing something wrong or putting the project at risk or whatever. On the 1 case, I don't know what it was personally, so I don't have an opinion about whether the details should be released. But I'm not inclined to think that the full details should be published. I am inclined to think that I'd be interested in a high level description of it, with particulars omitted.
Anyway, I think the evidence backs up my overall point - the WMF does a great job of fighting for users.--Jimbo Wales (talk) 20:05, 3 June 2016 (UTC)
OK, thanks. No lawsuits during this period against editors, so far as you know, or the WMF stemming from content? Coretheapple (talk) 21:01, 3 June 2016 (UTC)

A different form of paid editing

I agree with most of the anti-paid editing comments above, and the "let's have the WMF pay its current editors" seems quite naive. I think that might turn "The encyclopedia that anybody can edit," into "The encyclopedia that will pay anybody to edit" which would be an obvious black hole that would suck up a whole universe of money.

Folks may however be surprised that I am willing to support paid editing - just not commercial paid editing on Wikipedia. For example I would love to see all the PR folks get together, write all the CC-by-3.0 text that they'd like and put it all on their own pedia, call it Adopedia for now. I'm sure it would be very fun reading and several years after the fact, when it becomes totally ridiculous, we could show folks why we don't allow undeclared paid editing on WP.

More seriously, I think "we" could bring in some academics to fill a few holes around here. Starting with - "what do we have enough of already?" Biographies, Popular Culture, and Geography (73.6% of all articles) see File:Size of English Wikipedia (1000 vol).svg. These 3 subject areas are not really academic areas (at least in the way we present them). That is not to say that we should remove them, they are perfectly good encyclopedic articles, but the hole is our coverage is pretty much everything else. Such as science, technology, philosophy - really anything technically difficult.

How to do that? I'll suggest a different foundation, formally independent but on speaking terms with the WMF. Let's say foundation 2 (F2) has $1 million to spend. They could request applications for 6 month or 1 year positions - something for folks on sabbaticals or a just retired prof who always wished he/she had the time to work on Wikipedia. Maybe pay the equivalent of $100,000 per annum. Not too many academic folks in SF, NY, or London would probably want to do this, but there are likely very, very many qualified academics who'd enjoy this for one year. I'd require something like the following to be done on Wikipedia: 1. light edits (e.g. inserting good sources and pix, copyediting) of at least 10 articles each week 2. substantial edits (re-ordering sections, adding sections, removing outdated material, integrating topics that aren't well linked) of 1 article per week. 3. Complete re-write of a major article (e.g. Physics or Biology) every 6 months

Of course as paid editors they must declare, work with regular editors, and can't claim any special privileges. On Wikipedia they'd just be another editor.

Substantial edits and complete re-writes should be archived on F2's website, cc-by-3.0 of course.

$1 million could get Wikipedia a lot of material where we need it.

$10 million per year for 10 years would be a revolution.

F2 would be formally independent of WMF, probably run by internet savvy academics.

Formally, I don't think it would even be necessary to ask the WMF's permission to do this. But of course it would be nice to let everybody know what's happening, and a chance to object. If anybody has a few spare $millions, please do let me know. Smallbones(smalltalk) 16:20, 3 June 2016 (UTC)

@Smallbones: how does your local chapter feel about doing that? EllenCT (talk) 18:18, 3 June 2016 (UTC)
@EllenCT: I've got 2 local chapters, DC and NY (DC caught me for my $12 fee). Short answer - I haven't asked them. I'm guessing your point is "Wouldn't they think you are turning your back on them?" I'll say no, what this proposal would do is get in experts in areas where we don't do a good job - the big, hard-to-do, technical articles - most Wikipedians don't even try in these areas, and that's the problem. Current editors who do edit in these areas would be welcome to apply, and experience editing would be a definite plus, but obviously academic qualifications would also be a plus.
Another possible objection might be "didn't Jimbo try this and fail with NuPedia?" Things change. This might be a higher-paid gig. A "nu" project is difficult to start, but now that Wikipedia is going, folks can see that it wouldn't be lost effort, the articles are bound to stay around for awhile. It might even be the type of thing that in the future would count towards tenure. Input from academic Wikipedians would be appreciated. Smallbones(smalltalk) 19:35, 3 June 2016 (UTC)
I'm mainly interested in this idea for smaller language Wikipedias in the developing world. $1 million a year to fund graduate school scholarships at universities in Africa in exchange for 10 hour a week writing fellowships could do a lot of good. I have no idea of the credibility of this particular list, so I don't endorse it, I just point to it as an example of the kinds of schools I have in mind: 78 Master's Degree programs in Africa. --Jimbo Wales (talk) 20:16, 3 June 2016 (UTC)
Of course, it might be possible for undergraduate students there to get some "pay" if they might find some public domain documents which would be useful to the project and scan them onto commons. I am thinking particularly of any older biographical dictionaries, some of which I think exist, older texts regarding native mythology and religion, etc. They might not be in English, but maybe the Education groups to translate them, into or out of those languages. John Carter (talk) 21:16, 3 June 2016 (UTC)
In the developed world, this (John Carter's comment) is the type of thing handled by GLAMS, Wikipedians-in-residence, Wikipedia Visiting Scholars, and interns at university libraries. I'm not against further developing these programs, but Jimmy's suggestion to concentrate on lesser-developed countries would make a lot of sense here. Smallbones(smalltalk) 22:35, 3 June 2016 (UTC)

For once I fully agree with you Jimbo, invest money into universities and colleges in the developing world in reutrn for students improving content in the developing world. Something where students are going to earn something and gain academic esteem and to really benefit them, and improve wikipedia at the same time. I was thinking about running a contest for Africa sometime, but wasn't sure if WM would support it or the idea of getting universities and schools involved. The prize for improving the most content in a month to whatever could be a grant to study or something. I do think it's something WM should really start to discuss and seriously consider, as our developing world content of course is generally very poor. Brazil given its size in particular the lack of good content on important subjects is severe. For Africa it could also involve Swahili, Afrikaans, Arabic etc wikis and the smaller language ones which have an extreme lack of basic content. Something like this I think would be a good incentive to start to get the smaller language wikis improved, or something which supports translation from English. And then on the charity side, something like $1 million funding annually into African education is going to be immensely productive and help a lot of people. If you're genuine about this Jimbo I'm pleasantly surprised as I didn't think it was something WM would ever be interested in.♦ Dr. Blofeld 20:42, 3 June 2016 (UTC)

Agree, more below. Smallbones(smalltalk) 22:35, 3 June 2016 (UTC)
Why does it have to be in the developing world? Why not subjects that are neglected involving communities that aren't top-heavy with agenda-pushing COI editors? For instance, I sometimes dabble in labor issues relating to Mexican-Americans. The Bracero program, for instance, is an example of an article that could use attention from experts. We have a disconnect here. Persons with COIs and paid editors swamp articles on subjects of minor and secondary interest, making odd areas (minor companies, certain BLPs, certain historical figures) subject to lavish coverage while important topics of no commercial interest, not attracting COI editors, get short shrift. This is a serious source of systemic bias. I'd have no problem with scholars of Mexican-American history being paid to beef up the articles on subjects that currently get little attention in that realm. Just as one example. However, it would be essential that such articles be policed so that scholars don't use such articles to promote their work. That's a pitfall. But it's a far cry from Editor X adding crap about daddy and Editor Y being paid to santitize the article about Jughead Industries. Coretheapple (talk) 21:22, 3 June 2016 (UTC)
Agreed, to the extent that minority history and culture are pretty poorly covered on Wikipedia. Certainly in the US, but I expect that it is true in the UK and the rest of Europe (can I still include the UK in Europe?) as well. What Jimmy is talking about could be run much more simply than what I proposed above. The WMF could probably handle it, though it might be spun off like the WikiEd program after a few years. Projects that include the writing of major or contentious articles might be best handled by an outside foundation - so that the WMF is not responsible for writing content that might be seriously challenged. I expect that minority material (in any country) would raise more than hackles. So I'll repeat my original question. "Anybody know where to find a couple of $million?" I'd actually guess that it is out there somewhere (Carlos Slim??) and suggest the academic/ outside foundation model I suggested above. Even though I was thinking about science articles, same arguments apply. Smallbones(smalltalk) 22:35, 3 June 2016 (UTC)
There are some real pitfalls here. Even in the kind of articles that I'm suggesting there are potential POV issues. Academics often are axe-grinders particularly in contentious areas. So while I think the idea has some merit I think that it could be problematic if the WMF finds itself subsidizing editing that is criticized as POV. Coretheapple (talk) 23:56, 3 June 2016 (UTC)
That's why it should be an outside independent foundation. There are a series of imperfect safeguards. The academics who select the writers should (try to) be impartial (at least to offset any bias by the funders); the writers (paid by the independent foundation) should be impartial; since they disclose their paid status, Wikipedians will review the articles, hopefully removing any bias. The WMF wouldn't actually be involved in the writing or funding at all.
But with Jimbo's proposal, I don't think it would be controversial to fund graduate students in Africa, and their choice of what to write or translate should be pretty much left up to them. A non-WMF committee (say composed of Community members and academics) could select the candidates. Smallbones(smalltalk) 02:43, 4 June 2016 (UTC)
That would be tricky to execute. See my further thoughts below. Coretheapple (talk) 14:53, 4 June 2016 (UTC)
I think that handing out money to a limited group of people will breed resentment, but I think it's possible to make a big difference with some very limited support with few strings attached. There is actually a broad swath of the planet earning no more than a dollar a day, though this is in a way an exaggeration since much of the economy is not cash. If Wikipedia offered dollar-a-day, $30 a month, $360 a year support to people who make strong, sustained contributions, without telling them what to write about or limiting their behavior in any way beyond the logical don't get blocked to the point where you can't contribute, well, for 10,000 active editors that would be $3.6 million a year, i.e. readily affordable on the current budget. (If it weren't, we might have to let go a few of our elite overlords to afford it ... my, that would be a pity!) Wnt (talk) 13:17, 4 June 2016 (UTC)
Rather than pay some people a modest stipend to make strong, sustained contributions, without telling them what to write about, I'd be tempted to go in almost the opposite direction. Identify some needed tasks that just aren't getting done, and pay people to take on these tasks. In many cases, it may be very uninteresting work but it is worked needs to get done and it isn't getting done. That would avoid any potential resentment — if any existing editor expresses concern that someone is getting paid to do these tasks tell them they are eligible as well.--S Philbrick(Talk) 13:39, 4 June 2016 (UTC)
I think that either approach may have problems, because then we would be creating a class of compensated editors. There are plenty of good editors out there who labor quietly in fields of consequence who might resent it. So while the idea has appeal I can see pitfalls. It happens any time money is introduced into the equation. For instance, using a foundation is not a panacea as foundations often have political agendas, or are perceived that way. Even history can be politicized (look at Israel-Palestine). A tricky area, but interesting. Fact is, there is a serious systemic imbalance (in terms of size of articles and overall emphasis) caused by 1) POV editors 2) Editors with COI of various kinds and degrees and 3) Paid editors. None of the three factors may be resolvable.
Wikipedia stuns me sometimes by the disparate way it treats subjects. I was watching an old VHS tape the other day on a World War II heroine. I then looked her up here. To my amazement the article on her was small and tacky despite ample sources. It just hadn't received any sustained attention and she had no relatives to push her article or the article on her colleagues, now dead, all true heroes. It pained me to see this, but it is more common than not. In this and similar situations I feel like a caretaker of a neglected cemetery. Coretheapple (talk) 14:51, 4 June 2016 (UTC)
Indeed, the feeling of "caretaker of a neglected cemetery" reminds me of trying to check the "Slavery Amendment" and why the American Civil War was called the "War of Northern Agression" and was Abraham Lincoln the Northern tyrant (re "Sic semper tyrannis")? -Wikid77 (talk) 18:40, 4 June 2016 (UTC)

This conversation brought to mind the topic of digital labor and the essay "Thoughts on Wikipedia Editing and Digital Labor", which, as The Signpost reported, received some media attention. I have several work backlogs that I would like to hire someone to work on. One potential pitfall to avoid is having a system where vandals break things intentionally, to fill work queues so that they can get paid for fixing things they broke themselves. Perhaps @EllenCT: and @Hexatekin: can work together to give this topic more exposure. Regarding GLAMS, Wikipedians-in-residence, Wikipedia Visiting Scholars, and interns at university libraries, I'm not sure we're getting our money's worth. I haven't noticed much impact from these on content improvement, though maybe I haven't looked in the right places. What I have noticed of some of the initiatives that bring in new editors, is that generally they are more likely to increase my workload, rather than reduce it. wbm1058 (talk) 18:44, 4 June 2016 (UTC)

I'm not aware that my status as a Wikipedia Visiting Scholar is costing the WMF anything. The contrary, I suspect, because better quality boosts contributions, or so I hope. Its not unreasonable to ask that funds be re-invested in the editorship. And it certainly has improved content, being able, to give one example, to read through the legislative history from the comfort of my laptop improves my coin articles no end, and the ability to access scholarly articles doesn't hurt my political ones, to say nothing of the ability to borrow books (as I happen to be local to my sponsor, many are not).
I don't think it unduly arduous or expensive to reimburse (after application and vetting) expenses such as books, or allow a small stipend, again on application and with all public disclosure, which would obviate concerns about getting paid under the table by Big Biz. In other words, not pay for editing but a stipend to free up time (and focus attention of the editor, who probably wants it renewed), and the editor has no incentive to slant (the opposite, in fact) Much of the conversation above seems to revolve around doing that for outside editors brought in for the purpose, but the learning curve to be an effective Wikipedia editor is a steep one and it would probably be better to use those who have already picked up the necessary skills, and, more than that, have proved to be effective editors. And WMF or an allied organization paying small sums lessens the incentive for people to go to the "dark side". User:Darth Vader stays User:Anakin because people, as has been pointed out, are likely to be loyal to the funding source.--Wehwalt (talk) 08:18, 5 June 2016 (UTC)
Book stipends would be very valuable. I suppose there are already such grant programs out there if one knows where to look. The current situation, which we don't talk about, is that people are using the fraction of pages readily viewable on Google Books partial uploads, skipping the content they can't see on the major or minor fraction of pages that aren't viewable. That's not a healthy way to use a source. Now, would it be cost-effective to spring a considerable sum of money to buy that book? It depends, I suppose, it's a cost-benefit question... Carrite (talk) 13:14, 5 June 2016 (UTC)
@Wbm1058: This comment surprised me. While I am not that close to GLAM initiatives, my impression is that they been enormously successful. The WIR and Visiting Scholar positions don't qualify as "our money" unless I'm missing something. I do agree that some of the initiatives to bring in new editors have generated significant workloads for existing editors.--S Philbrick(Talk) 12:28, 5 June 2016 (UTC)
Sure, I'm not in a position to really judge the success of GLAM initiatives; I'm just not aware of content additions or improvements that can be attributed to GLAM – that doesn't mean they aren't there. Maybe they could do more to promote their results; maybe they have and I just missed it or forgot about it. Confess I have trouble remembering what the acronym stands for – I know it's libraries and museums and other things like that. wbm1058 (talk) 12:41, 5 June 2016 (UTC)
  • I suppose it is healthy to do some collective thinking outside the box, but the bottom line reality is that the current WP system of (mostly) volunteer-generated content works pretty darned well and there isn't anything that time and perhaps new efforts to direct volunteers to deficient areas can't fix. It wouldn't take all that many years of university classroom projects, by not all that many university classes, to make a real impact. Instead of pointless term papers, if instructors turned to the writing of a comprehensive WP article as an assignment, the content would flow. This model does produce significant usable and lasting content.
Additionally, Professors will retire and more of them will come to see WP's vital educational purpose and massive penetration in the internet age and will themselves be motivated to share specialist knowledge if we (meaning now the WMF) make efforts to gather and direct them. Carrite (talk) 13:08, 5 June 2016 (UTC)

I'm glad to see this thread has generated some interesting thoughts and discussion. I agree that

  • payment for books and similar to current editors can be (and is being) done effectively, but
    • any sizable stipends to editors by the WMF should be introduced only very gradually because they potentially might cause backlash among other editors, and might involve a contentious political process to determine the awardees.

Many of our current programs have a limited, but meaningful effect, e.g.

  • GLAM programs mostly affect art articles (often pre-1923), but if you are interested in those 0.5% (guess) of our articles, their affect has been revolutionary.
  • Similarly, any paid/professional/subsidized WMF program will only have a marginal effect on the number of articles, but could have a revolutionary effect on the quality of these articles.

These are the main reasons I suggest the large, technical, super-important science articles, e.g. physics and biology. I don't think most of our current part-time editors have the skills, knowledge, time, and patience to edit these type of articles. Current editors could apply of course, and editing skills should certainly be considered. Focusing on 2,000 such articles (about 0.04%) over 10 years probably wouldn't drive away many editors, but could have a revolutionary effect on the whole encyclopedia.

I do recommend that an outside foundation, not the WMF, do this. So if any potential donors (Bill, George, Warren - this means you!) are reading this and agree, please let me know! I do think this could be done by an outside foundation successfully given our current rules. But perhaps it is all just a pipe dream.

Jimbo's suggestion of graduate fellowships for Africans wouldn't require an outside foundation, shouldn't be controversial, and would have a significant effect on some of the small language editions.

Smallbones(smalltalk) 14:36, 5 June 2016 (UTC)

Very sound, and I agree with you on the political question re stipends, but I think the conversation should be had at some point, because there are key articles that should be improved for Wikipedia to be more credible, and fifteen years of waiting to see who will improve them hasn't led to improvement. I'd also be glad for a reference to a program that actually reimburses for books, not "only in this highly specialized field" or "just submit your grant proposal along with the guy who wants to transmit Wikipedia to the Klingons (in Klingonese) and who will probably get the funding because he knows all the people in the WikiProject".--Wehwalt (talk) 15:17, 5 June 2016 (UTC)
Try the new Rapid grants. Individual engagement grants are also linked there. I think several chapters have this type of small grants program as well, in fact I'd thought DC was one of them. I wouldn't worry too much about Klingons (but they seem to welcome all types of folks there). And if Wehwalt doesn't know the right folks who make these grants, I'd guess they know about Wehwalt very well (the positive side that is :-) ) Smallbones(smalltalk) 16:44, 5 June 2016 (UTC)

The Signpost: 05 June 2016

Epiphany-producing socratobot?

Jimbo, should the Foundation fund development of a chatbot designed to use the Socratic method to cause epiphanies?[50][51] EllenCT (talk) 14:48, 3 June 2016 (UTC)

I'm not sure about that, but I certainly find the concept discussed in the first paper interesting. Neither paper appears to mention "Socratic method" or "epiphanies" by the way, and I think we are very very far away from machine intelligence (that is affordable at our budget levels anyway) that could usefully do that.
In terms of what the first paper says - and keep in mind that I am not an expert in this area and that I have only read the abstract and introduction so far - I think that this is the sort of thing that we should be researching (and are researching) as part of the "Discovery" project. Suppose someone comes to the web with a question like "What are some good horror movies for Halloween?" They likely type that into Google. The 9th result is List of films set around Halloween, which is a roughly decent answer. But how could a basic machine intelligence approach improve that answer easily? Ok, our article is a list of films that are set around Halloween, that's a good start. But the question asked about "horror movies" - so it seems straightforward to exclude the ones which aren't in horror movie categories. And then what about "good horror movies"? Hard for a machine to decide the answer, but using the kinds of techniques this paper talks about, it seems quite possible to look for supporting evidence. So we could easily enough (not trivially, but easily enough) return a results page in our own internal search engine that extracts the list of horror movies and rather than having a simple list, also have a quote from each article indicating the quality.
That's not even an ideal example - it's one I got from the paper itself. But I think it's interesting enough and easy enough to think about how a more generalized system might work that it is worth further investment.--Jimbo Wales (talk) 15:18, 3 June 2016 (UTC)
So, "Tell me more about your mother..." has been THE question in AI systems, but I think we should encourage a group of "wikinannies" to patiently talk with users, to help them better consider their attitudes and actions. Meanwhile, the wp:Reference_desk has typically attracted many interesting-to-converse people, although some subjects might require a few days for fuller responses. -Wikid77 (talk) 15:37, 3 June 2016 (UTC)
You are thinking of ELIZA's person-centered, Rogerian and Freudian approaches. But I agree with both of you. [52] would have been a better review to cite, and [53] explains the Socratic method connection. You can't make a good question answering system that doesn't ask its own questions, which is one of the things our disambiguation pages do. If the system's goal is to cause epiphanies then I believe the outcomes will be more successful, and failures more helpful, than if the goal is merely to answer questions instead of both answering and asking them. EllenCT (talk) 16:06, 3 June 2016 (UTC)
You could also consider training a large neural network using a database that contains questions and what we would prefer as the best answers. Count Iblis (talk) 21:20, 3 June 2016 (UTC)
Questions are ontolo-grammatical graphs (like parse trees with more and different kinds of arcs), but neural networks are trained on vectors. The ORES team selected random forest techniques. Those are based on symbolic features so conceivably they are subject to human review and possibly editing. There would certainly need to be a way to model the beliefs of multiple agents, and I would recommend a goal-directed problem solving system based on the symbolic and probabilistic logic of those agents. Here are some more sources on those things: [54][55][56][57][58]. EllenCT (talk) 00:40, 4 June 2016 (UTC)

The random forest approach seems better. BTW: I see WP (up to May 2016) has had almost nothing about Standford PhD and AI researcher Chuck Rieger, University of Maryland. -Wikid77 (talk) 17:55, 4 June 2016 (UTC)

@Wikid77: which of Rieger's works do you find most pertinent? Of the problems with Wikipedia I wish I could solve the most, the questions involved would be something like, "How do I optimize outcomes?" but the assumption that outcomes can be optimized is not in evidence, so the system would need to try to get the questioner to cooperate. Therefore, answering a question with a question can be helpful, which validates the Socratic method in general. However, questions are not always preferable or even desirable, for example as in the case of yes-or-no questions about arithmetic. Chatbot engineers get that wrong, because setting up the system to respond to answers to questions is a huge multi-agent sensory-and-belief modeling task, which is why I think the probabilistic methods compatible with general symbol systems of logic (Bayesian methods) are likely to be the most helpful for that task. The control flow loop would need to be generated by a goal-directed problem solving-oriented planning system, with initial goals to "correctly understand statements and answer questions" which might have hundreds of subgoals about language and I/O. So, let's say the system is asked about decision making authority in the hierarchy. Let's take a practical concern.
Question q1: Is authority the consent of the governed or the threat of force?
In order to respond to that question, the system would need to understand the motivations of the society of the questioner which involves things like their respect for strict hierarchy. A good example answer might be,
Answer q1a1: Forces of nature have authority apart from forces of laws.
But describing how to get from Q to A is not easy, because it depends on the details of society, let alone biology, physics, and math.
If this was going to be an ongoing research project that the Foundation wanted to support independently, it would dovetail with content improvement. In order to be ambitious enough to have a greater-than-even chance of achieving success, it would need levels of content improvement which I doubt the Foundation could fund for less than USD 150,000,000 per year. That would work out to an endowment goal of around $10 billion under present low-growth assumptions, but I expect the improvements would likely increase the rate of growth, so maybe it's only $2 billion. EllenCT (talk) 16:40, 5 June 2016 (UTC)
On second thought, q1a1 is an unsatisfying answer, and a question like,
Answer q1a2-q2: Why not both?
would be better. I am aware there is a meme for that, the subject of which probably reads our articles by now. EllenCT (talk) 17:05, 5 June 2016 (UTC)
People are still using troff to write papers? Gasp! :) --David Tornheim (talk) 05:16, 6 June 2016 (UTC)

Objectivist tradition

[59] says, "... objectivist tradition ('Objective stimuli are important')...." Which is the least dystopian of the objectivist traditions? EllenCT (talk) 04:52, 6 June 2016 (UTC)

Processing false dichotomies

Frequently conflicting authorities include those involving the age of the universe. [60] says it is, "... helpful to bridge the false dichotomy ... between atheistic evolution versus religious creationism." How would a question-answering system best respond to situations where our NPOV policy would likely result in presentation of mainstream and fringe points of view? EllenCT (talk) 04:59, 6 June 2016 (UTC)

Journalists to benefit advertisers

Returned from the archives for a follow-up question. EllenCT (talk) 17:34, 5 June 2016 (UTC)

Time Inc. evaluates the work of its journalists on the basis of its benefit to advertisers.

Wavelength (talk) 14:43, 29 May 2016 (UTC)

"The highest and primary obligation of ethical journalism is to serve the public. Journalists should... Deny favored treatment to advertisers, donors or any other special interests, and resist internal and external pressure to influence coverage; Distinguish news from advertising and shun hybrids that blur the lines between the two. Prominently label sponsored content." -- Society of Professional Journalists' Code of Ethics. EllenCT (talk) 15:31, 29 May 2016 (UTC)
Thank you. See Code of ethics in media#Society of Professional Journalists: Code of Ethics.
Wavelength (talk) 17:38, 29 May 2016 (UTC)
Is there an impartial entity evaluating news organizations on the basis of their adherence to ethics?
Wavelength (talk) 19:42, 29 May 2016 (UTC)
Category:Media analysis organizations and websites, but it is missing newcomers like PolitiFact.com and things like Snopes.com. It's also missing the Columbia Journalism Review which might be the oldest such organization in the US, and Project Censored which attempts to rank problems with the specific issue you point out by their impact on society, using mean opinion scores of a panel of academic journalism studies judges.
Also, this is one of those very frustrating areas where actual impartiality does not yield what the average, unsophisticated person considers political neutrality. If you type "facts have a" into a Google search form, it will auto complete into multiple forms of "well known liberal bias." I've never heard anyone claim that facts ever seem biased towards conservative viewpoints, because, as far as I can tell, the problems you pointed out with Time have already permeated almost all of the mainstream corporate for-profit media in the developed and developing world, which is one of the reasons that our policies hold peer reviewed secondary sources as more reliable than even well-respected commercial news sources.
Rupert Murdoch's recent purchase of the National Geographic Society is something to keep an eye on, because it's clear what happened to the Wall Street Journal after he bought it. EllenCT (talk) 22:59, 29 May 2016 (UTC)
Facts have no bias, but different politics may be biased against certain facts. Some may even be hostile to the very idea of establishing the facts - climate change (on the right) or sexual violence (on the left). Rhoark (talk) 12:52, 6 June 2016 (UTC)
@Rhoark: to what "sexual violence (on the left)" do you refer? EllenCT (talk) 15:21, 6 June 2016 (UTC)
"listen and believe": examining the facts is sexist (unless the accused is Muslim in which case women can go hang themselves and examining the facts is racist) Rhoark (talk) 15:47, 6 June 2016 (UTC)
Is any entity evaluating peer-reviewed secondary sources on the basis of their reliability?
Wavelength (talk) 23:15, 29 May 2016 (UTC)
Yes, when literature reviews conflict, meta-analyses are usually published to quantify their respective claims. EllenCT (talk) 23:27, 29 May 2016 (UTC)
P.S. http://guides.library.yale.edu/c.php?g=295905&p=1975646 EllenCT (talk) 18:47, 30 May 2016 (UTC)
Thank you.—Wavelength (talk) 20:05, 30 May 2016 (UTC)
@Wavelength: do you agree with the assessment at [61] that, "most suppression of free speech ... comes from conservative and corporate forces, not overly aggressive leftists"? EllenCT (talk) 17:34, 5 June 2016 (UTC)
@EllenCT: In my judgement, this is too close to being a political question, and I respectfully decline to take a position on it. Anyway, I do not know what sources and measurement criteria the author used.
Wavelength (talk) 23:25, 5 June 2016 (UTC)
It is certainly political, as Censorship and Censorship of Wikipedia indicate. I suggest a tally of the instances in those articles may be a helpful way to arrive at a reasonably objective answer. EllenCT (talk) 23:41, 5 June 2016 (UTC)
EllenCT, which news sources are rated as being the most beneficial to their readers?
Wavelength (talk) 23:30, 5 June 2016 (UTC)
Because sources are usually corporations subject to change of hands, there aren't stationary answers to that question. Category:Journalism awards may be helpful, as may [62] and [63] for example. EllenCT (talk) 23:41, 5 June 2016 (UTC)
Apparently, most articles in that category are about journalists, although I found the subcategory Category:Pulitzer Prize-winning newspapers, with 91 articles. I also found Pulitzer Prize#Criticism and studies (version of 14:27, 31 May 2016).
Wavelength (talk) 01:14, 6 June 2016 (UTC)
  • I suppose it is healthy to do some collective thinking outside the box, but the bottom line reality is that the current WP system of (mostly) volunteer-generated content works pretty darned well and there isn't anything that time and perhaps new efforts to direct volunteers to deficient areas can't fix. It wouldn't take all that many years of university classroom projects, by not all that many university classes, to make a real impact. Instead of pointless term papers, if instructors turned to the writing of a comprehensive WP article as an assignment, the content would flow. This model does produce significant usable and lasting content.
Additionally, Professors will retire and more of them will come to see WP's vital educational purpose and massive penetration in the internet age and will themselves be motivated to share specialist knowledge if we (meaning now the WMF) make efforts to gather and direct them. Carrite (talk) 13:08, 5 June 2016 (UTC)
@Carrite: Would you be willing to direct some volunteers to help out with CCI? There are over 155 open cases, many of which have several hundred articles, each of which have several edits. Probably over a quarter million edits. I've worked on a few, it is mind-numbing work. There are cases open since January of 2010, and the list is growing. How do we persuade thousands of volunteers to spend considerable time on this task?--S Philbrick(Talk) 13:43, 5 June 2016 (UTC)
Well, that's a new thread, but what needs to be done with CCI is that it should be blown up and started over. Its methodology is fundamentally flawed, it does not scale. And so the backlog grows and grows and grows... What should happen: 1. copyright bots identify violations. 2. volunteers (the current CCI crew) jump in to offer advice to the violator to make sure it doesn't happen again. 3. persistent problem editors get tossed. All this fast and in the present, within hours or days of the violation(s). Only then, if it is a long-term editor being hauled in, should there be sampling of the contribution history to identify problems for cleanup.
As it stands now, three minor-ish violations out of an entire contribution history can be used to start a case, which triggers a complete audit of every edit ever made by the subject. It's clearly an unsustainable system, I've got no clue why it has lasted so long. The way to fix it is to blow it up and start fresh, it's beyond reform. Carrite (talk) 13:57, 5 June 2016 (UTC)
Well, I think the reason why people don't want to do that stuff is because it seems like something that could readily be automated. Maybe instead of hiring editors WP should consider putting up a bounty for a piece of free software to do the job; the winner of course would also get a great resume bullet-point. Wnt (talk) 13:51, 5 June 2016 (UTC)
See User:EarwigBot/Copyvios/Exclusions to see why relying on purely automated systems has problems in the copyright area. And automated systems are very likely to miss "close paraphrasing" cases, which also violate copyright. Also note that "minor-ish" is generally not "minor-ish" if you look at the cases at CCI. Collect (talk) 14:16, 5 June 2016 (UTC)
True, but realistically CCI isn't gonna catch much close paraphrase either unless they are working really, really slowly. Direct cut-and-paste can be caught by automation and should result in a fast warning followed by a fast ban-hammer. Close paraphrase, well that's a subjective thing anyway, and something that really has to be taught and learned. That sort of copyvio has been with Wikipedia since 2001 and will be with it until the end of the project — we are a work in progress and that applies to the people writing content as much as the content being written... All we can do is work closely with people of good intentions as such violations are found and rapidly toss those who do not have good intentions or a willingness to learn the proper way to use sources to write content. CCI is too obsessed with the process of purification of the entire contribution histories of malefactors; it should be concerned with fast identification of malefactors and their ruthless removal. If we treated copyvio like we treat No Legal Threats — "fix it fast or you are out of here" — the problem would be greatly reduced and a sustainable system of copyright monitoring would be possible. Carrite (talk) 14:44, 5 June 2016 (UTC)
  • The classic example of how CCI should not work is the case of Richard Arthur Norton (1958- ). Enormous contribution history dating back to the wild west era of WP, 2004-05, in which there were problems with a certain percentage of edits. Turned in to the copyright cops, it became an intractable case — tens of thousands of edits, too big to fix even though the problem was past history not current behavior. Scores or hundreds of hours wasted and no way in hell that his case will ever be resolved by CCI, with an emphasis on no way in hell...ever. Sanctions against him which impede him from creating content, encyclopedia the worse for it, no way to clear his name unless he self-polices (!!!) his ancient editing, which would probably take about a full year of his volunteer time, just guessing. Not gonna happen. So it is lose-lose for WP and for him. Meanwhile, current serious copyright violators escape sanction because CCI is so bogged with intractable cases.
CCI seriously needs to be completely shut down, its current cases dismissed without action, and a new software-based speedy system of quick identification and fast blocks of copyright violators established. Carrite (talk) 14:13, 5 June 2016 (UTC)
Existing software does a decent job of starting the process, but it isn't close to sufficient, especially for older cases. I don't think people eschew it because it seems like a job for a computer, I think it is avoided because it is mind-numbingly boring. But not easily automated. If something thinks it can be automated, I'll support WMF funding for a system. But we don't yet have a system, and I cannot possibly support dismissing current cases without action. In OTRS work, one of my common phrases is "Wikipedia takes copyright seriously." I don't want to modify that to ""Wikipedia takes copyright seriously, except for edits before 2016, in which case we don't care". We have many copyright violations in Wikipedia. One reason this isn't a more serious problem is that some outsiders accept that we have a process, even if backlogged, to track them down. If we throw up our hands, and declare all old transgressions get a pass, we might find ourselves in legal trouble.
You might be right that there are better ways to address the problem going forward, but even if true, that doesn't make the old problems go away. I'd support payment, either for software that can do the job, or cadres of individuals to do it manually. Someone expressed the concern that paying someone for edits could cause resentment.I don't think see that as an issue with CCI edits. If it means paying existing volunteers a buck a day, in addition to the recruited specialists, it would be small beer, and worth it.--S Philbrick(Talk) 15:06, 5 June 2016 (UTC)
Well, some people (not me) seem to be motivated by contests to do things (just takes someone or two to organize, make rules, and "prizes") . Alanscottwalker (talk) 15:18, 5 June 2016 (UTC)
I am a fan of contest or contest-like initiatives, with the obvious caveat that contest can lead to unintended consequences. I think contests work best when it is relatively easy to monitor that the goals were accomplished. For example, imagine a contest in which participants are given a list of articles without references or without enough references and asked to add at least one reference. It isn't difficult, in the case of online references, for someone to review and ensure that, yes, a reference was added and the reference supports the claim. No doubt, there can be edge cases but it at least sounds doable.
Compare that to a typical CCI check. Participant A does it right, maybe run the article through some software as a start, choose some particular phrases to search, be on the lookout for certain types of phraseology which experienced checkers will recognize, and after a quarter hour of checking you conclude that the article is fine and you accept it. In contrast, participant B glances at it, decides it's fine and checks it off as Ok. A reviewer will see both participants marking the edit is fine but there is no trivial way to determine that participant A is doing it correctly and participant B is going through the motions. A sloppy confirmation of no problem is arguably worse than doing nothing at all.
By "contest like" I mean something along the idea of a contest but fewer incentives to take shortcuts. For example, editors might commit to checking 10 items a week, and someone creates a table with involved editors, and you simply add a check box if you meet your weekly quota. Creates a little peer pressure to keep up with others, and unfortunately, has the potential that you might go through the motions but somewhat less likely to encourage sloppiness as there is no big prize for working faster than you can reasonably review them.--S Philbrick(Talk) 15:50, 5 June 2016 (UTC)
So, you would probably want to make it a short contest (5Day), with a step checkoff box, and an estimated time of average review, where the 'winner' is likely to to do no more than say a dozen, and those can be spot checked by the judge(s) (and by fellow contestants) for the prize (and if they don't get it goes to the next in line, the person with say 8 that check or 5, etc), and the failed CCI is returned to the que, or sometimes is fixed by the sofixit type editor). Alanscottwalker (talk)
That might well work for some tasks, but my very crude estimate of a quarter million edits to check is almost certainly low. A hundred participants doing a dozen over a 5-day period rounds to 0% of the task. We need 100 participants agreeing to do a dozen a week for four years.--S Philbrick(Talk) 16:57, 5 June 2016 (UTC)
Well, this CCI topic was started with the obvious acknowledgement that you are not getting that and you won't be getting that. So, either shut it down and give up, or like actually experiment with getting people involved, some may like it, who knows. (A dozen edits? Well, ok. I was actually thinking of whole dozen articles being cleared but if you want to divide it by edits just tranch the edits, but the simplest tranch would be a whole article where a potential CCI problem has been flagged - and giving the article a clean-bill) Alanscottwalker (talk) 18:19, 5 June 2016 (UTC)
Here's the thing though: checking every edit is ridiculous and that's the CCI way. Their SOP is akin to proving that somebody has violated traffic laws three times, which triggers an impound of a car and manual review of every minute of driving history shot on a dashcam going all the way back to drivers' education class... The backlog that CCI finds itself with is absolutely irresolvable; continuing the same bad system forward isn't going to positively resolve any old cases and save the day for god, country, and copyright law, it is merely going to divert attention from new malefactors who need to be stopped post haste. I'll tell you right now that 250,000 edits to be reviewed is a pipe dream, it is far higher than that, Norton got hauled to the proctologist for 70,000 or something edits himself — which was akin to trying to fix gophers in your yard with half a dozen bombers on an air strike. The system is broken. Carrite (talk) 03:44, 6 June 2016 (UTC)
You keep bringing up Norton as if his case somehow supports yours. He was sanctioned and is currently restricted because not only were problems identified, once they were pointed out to him, he refused (repeatedly) to agree they were in fact problems and refused to help identify where he had gone wrong (because he didnt think he had). At that point there was no choice but to restrict him. Under your 'best process' example above, his entire history would still end up looked at, because his contributions were so prolific. He actually was given the 'fix it or get out of here' option. He chose not to fix it, so his editing was restricted. His userpage has what, 500? 600 subpage biographies, a large portion that contain links to findagrave (a user-generated problematic site in itself) pages written by himself containing content that in a number of cases violates our close paraphrasing and extended quote usage here. Norton actually tried to change the External Links perennial websites in order to allow linking to material that Wikipedia would find problematic. Lets not mention his issues with extensive quote usage on-wiki that was found to be excessive. No one wants to deal with RAN's work because it is time consuming and just results in arguments with RAN. I dont see you clearing that backlog of subpages he is basically using as a self-promoting webhost. Literally every single one that has a find-a-grave link has to also have the find-a-grave page checked in order to see that it also complies with wikipedia's policies. The thing is, there are *really* easy fixes to most of RAN's problems. He just refuses to deal with them. Only in death does duty end (talk) 10:06, 6 June 2016 (UTC)
I think the philosophy is off here. We normally say focus on the edit, not the editor. When a CCI investigation is merited, maybe all you need to know is - is an editor particularly prone to copyvios? Then you run each of the articles he edited through a general plagiarism finder, and highlight all the text that comes up with a match. Ordinarily there is a high risk such matches go from Wikipedia to other sites rather than from there to here; but when there is reason for doubt, it seems fair (once sites with massive Wikipedia content are excluded) just to flag all the questionable text for a rewrite. Few of our articles are so perfect that they can't be rewritten on a source by source basis. Certifying some old edit is or isn't a copyvio is really uninspiring. It would be a lot more inspiring to get editors to work through individual articles, producing drafts known to be free of copyright and plagiarism issues in which the sources have been rechecked and expanded and rewritten from, until all articles with questionable parentage are accounted for.
That said, in the RAN case as I recall the situation is aggravated by people coming up with special standards just for him. Wikipedia has a general philosophy that if it thinks someone is having a hard time following the rules, it makes him follow more and more rules until they are impossible for anyone to follow. That can be all well and good for bullying people off the project, but if you want volunteers to go back and edit our actual content to follow special rules only for one person, it isn't so much fun. A lot of those findagrave quotations are very clearly not copyright violations, because they're much too short to be, and they are properly attributed so they are not plagiarism. Our adaptation from findagrave is a transformative use anyway. So no, people don't want to run around taking a sentence out of an article here and there because some other editor was supposed to be under a special restriction against it. Wnt (talk) 10:31, 6 June 2016 (UTC)
You have it backwards. Special standards were not invented just for him. His inability to follow the normal standards resulted in harsher and harsher sanctions. The problem with the findagrave ones is not that all the quotations he uses on-wiki are problematic, its that some were and in combination with the fact our and findagrave standards are not the same AND that as someone who wrote a lot of findagrave biographies, all of them require checking for compliance. ELNEVER requires we do not link to sites that have copyright issues, findagrave may or may not, when you factor in a known editor with copyright issues who is active on both wikipedia and findagrave, it requires any linked to findagrave site by them is checked *as well as* the wikipedia article. Or remove any references/links to findagrave. Which RAN resists. Its a massive time-sink, and since RAN doesnt want to do it, no one else does either. Only in death does duty end (talk) 10:40, 6 June 2016 (UTC)
Wait a minute - you want Wikipedia volunteers to run around checking whether entries on some other site are copyright violations so that you know, case by case, whether it is worthy to link to? Or threaten just to scrap all the citations to a widely used resource on the web? Wnt (talk) 10:54, 6 June 2016 (UTC)
Find-a-grave is ineligible to be used as reliable source, and generally should not be used as an external link. This has been the community consensus (see the archives at RSN etc for the history) for a long time. If it was almost any other editor, then no checking would be necessary because it would be accepted on good faith that the editors use of findagrave is compliant. Because it is RAN, and he often links to pages he actually wrote on findagrave, more diligence is required - the other option being just to remove the link which would make no checking necessary, not to mention EL's to FAG add zero extra info to an article anyway. Keep in mind this is just *one* of the issues with actually having to check RAN's work. There are others. Which is why no one is making a concerted effort at it. Only in death does duty end (talk) 11:01, 6 June 2016 (UTC)
@Only in death: Now I'm genuinely confused. Stipulating that find-a-grave really isn't a reliable source and can't be linked, why is there any problem at all? (I don't actually want to evaluate that point here, because this discussion is really about CCI in general, i.e. why it exists) Couldn't you nuke every reference to them all at once by bot, whether it was RAN or someone else who added it? I don't get how someone else's use can be assumed to be compliant if it's not usable as a source. Wnt (talk) 17:57, 6 June 2016 (UTC)

* Well, fortunately, as the one person who painstakingly worked through the sub-pages intended to be the seeds of proxy-starts listed on RAN's user page and who meticulously eliminated all the extended quotations from the citation templates that he was using so that the wolves wouldn't have anything to howl about, I can state with authority that Only in death is constructing a diversionary strawman with all this hoo-haw about Find-a-Grave. Norton's stubby pseudo-starts derive almost exclusively from the New York Times (as copyright compliant rewrites) and generally have little, if anything, to do with Find-a-Grave. ArbCom, in all their short-sighted genius, has subsequently shut down RAN's ability to even do these start-seeds. He has been a bullying victim since day one. He's far from the greatest content writer, but he's a dedicated creator of notable starting stubs and start-quality pieces, or at least he was before the attack dogs tore one of his arms off... Carrite (talk) 16:41, 6 June 2016 (UTC)

    • Its not a strawman. Its one aspect of why RAN (specifically) has to have his work closely vetted, and its one example of why no one is doing it. As an editor in good standing, if you were to promote any of the articles in his subpages to mainspace, no one would bat an eyelid. But you havnt. If you have checked them all, move them to mainspace yourself and let them stand on their merits. RAN's restrictions do not prevent him from working on articles, only creating new ones due to his problems. And dont lay this at Arbcom's feet, the community was tired with RAN before Arbcom got involved and after. Hell, if you say you have checked them all for copyright, close paraphrasing and the extended quotation issues he was sanctioned for, I will move them all myself if you dont want to do it. Some of his subpages have not been touched in years and userspace is not for pseudo-articles. Only in death does duty end (talk) 16:55, 6 June 2016 (UTC)
If you move them, you move them on your own authority, not mine. But so far as I am aware, they are all clean. Carrite (talk) 17:11, 6 June 2016 (UTC)

Attawapiskat Suicide Crisis update.

Jimbo, there seems to be something much broader and not indigenous specific going on, in Canada at least, so I will close this discussion here because I do not see how you can help with this. Regarding the Attawapiskat situation, our federal and provincial government are pouring 500 million dollars into 1st. nations initiatives...especially mental health..so maybe that will help. I established contact with 2 teachers in Attawapiskat as well as about 7 "friends" on Facebook including a specific "Attawapiskat Suicide Awareness" group but did not find any interest at all in editing Wikipedia, even though their own articles.... e.g.Cree...have lots of room for expansion. As you can see, there are lots of resources being applied to the issue...its downright horrible that many young people are so messed up. I've been to Woodstock, Ontario and its a gorgeous smaller town...all I can say is I have no idea wtf is going on. Nocturnalnow (talk) 21:44, 6 June 2016 (UTC)

Why Johnny can't wikisearch

 
Wikimedia Commons Kitten after being told En-Wikipedia has The Lord of the Rings: The Battle for Middle-earth II as the 3rd result on a sitewide search for 'Earth'.

For further discussion about improving search-in-Wikipedia to pinpoint answers, I have checked the problems caused by unusual page names, or when omitting synonym phrases inside a page. In particular, the renaming of page "Area of a circle" as pagename "Area of a disk" has limited the search listing for "Area of a circle" (as noted in May 2016 RfC: "Talk:Area of a disk#RfC article title: "Area of a circle" or "Area of a disk""). The circle-area problem is a complex case, where extreme use of the term "disk" seems to deter the common search for "area...circle" and hence Google Search lowers WP to pagerank #15 (2nd page of results "area of a circle"), although Bing.com lists Wikipedia on the first page of results. By comparison, a search for "area of a disk" shows WP as result#1, but also lists "Area of an annulus - Math Open Reference" where people think disk area can exclude the hub of the disk (see: Annulus (mathematics)).

Similarly, to broaden a search for related words, then WP pages must contain synonym phrases, either blended into the text or within footnotes, because the current wp:wikisearch does not match as many various synonyms as either Google Search or Bing.com have indexed. For example, if a song contains a famous phrase, then the WP page about the song must be modified to discuss the words mentioned in the phrase; otherwise, wikisearch cannot find the song unless those specific words are noted within the WP song page. WP is severely hindered when people demand a "circle has area zero" because the interior of a circle is a so-called "disk" while ignoring the common-sense tradition of "area of a circle" just like "area of a square" is the area inside the square. The confusion stems from the mathematical view, because a disk (mathematics) is the area inside a circle and the "area" is the "size of the disk".

As discussed for years, some people's unusual warping of terminology really hinders access to information, but with wikisearch, the peculiar phrasing almost totally derails the search function. So we find "one-bad-apple wording" can totally throw the whole bunch of wikisearch into the far reaches of the twilight zone. Hence, bizarre article names (and unusual contents) can almost completely destroy the reader's access in searching. Instead, WP needs more articles targeted to specific common topics (such as "area of a circle" or similar topics), where the internal wording includes popular related phrases, and the page title should not be a purist's demand for unusual wording. Fortunately, most page titles tend to be move-renamed into the wp:COMMONNAME phrases, and the search can be rescued. -Wikid77 (talk) 16:24, 6 June 2016 (UTC)

The search system exists in some sort of bizarro world of design with underlying good intentions. Suppose I am on the page Earth, and wish to know more about the Earth and what pages talk about it. I type in Earth into the search bar and press enter, only to be redirected to Earth again. Truly bizarre, but even more so when you consider that searching for "Earth" works only when in the search page already and not typing from the bar. So either you have to type obvious gibberish into the bar in order to get to the search page (where it will dutifully tell you there is nothing by that name, and you can then search Earth), or type nothing into the search bar and press enter (where it will bring you to a generic search page), a wholly unintuitive maneuver. So not only is the system wholly broken, but the more frustrating thing is that there is a working shell of a system buried deep within, begging to get out.
We haven't even touched on what the search system displays. The first result for Earth, is obviously Earth, and the second is Geocentric orbit. Seems pretty good so far. The third is The Lord of the Rings: The Battle for Middle-earth II, a real time strategy video game. Why is this here? It's a FA and a good long article, but is wholly irrelevant to Earth, the planet. The only common feature is that it has "Middle-earth" repeated a lot of times throughout it. The rest of the search results carry the same pattern, interspersing relevant pages with irrelevant ones like Live Earth and Geastrum triplex (???). Pinguinn 🐧 18:19, 6 June 2016 (UTC)
See Help:Searching.Wavelength (talk) 18:45, 6 June 2016 (UTC)
Here's a riddle for you: search Allegheny County,PA and you get lots of places in Allegheny County, Pennsylvania, but not that article itself. To find that you need to search Allegheny County, PA. Wnt (talk) 22:56, 6 June 2016 (UTC)

FWIW, if one wishes to find other articles than just Earth, one need not go to the route of using gibberish to get to the general search page when all one needs to do is add "a" (or "q" or whatever) to the regular search. And, yes, "middle earth" will always show up. Really. Collect (talk) 23:48, 6 June 2016 (UTC)

Why not measure effect of providing desktop interface to mobile users?

Jimbo, when you opposed measuring whether mobile users offered the ordinary desktop interface edit or donate more, you did not seem to offer an actual reason against making the measurement although I think it's fair to say that you implied someone would have to be neo-luddite to think of doing away with the Foundation's custom mobile app content viewers and reassigning their engineering headcount to getting a unified mobile and desktop web experience. While I am vaguely inclined to agree that the Foundation should support diversity in content viewing applications, I just want to be sure that we are not doing so at the expense of editors and donations. And I think it's clear that the existing diversity of web browsers on the different desktop and mobile platforms causes custom content viewer applications to add needless and expensive complexity to an already very complicated software support task, for which there is no clear cost-benefit justification. I doubt dropping mobile apps, if doing so can increase editing contributions and financial donations, will cause anything like the state of mobile app development to return to horse-and-buggy days. Are there any reasons to oppose the measurement? Shameless Hoompa (talk) 17:53, 5 June 2016 (UTC)

@Shameless Hoompa: You placed this discussion at Talk: Lucas Johnson. I have moved it here as I presume you meant to edit Jimbo's talk page and not that of a fictional character :)--5 albert square (talk) 18:40, 5 June 2016 (UTC)

I asked that question May 21st. The subsequent discussion is at meta:Talk:Fundraising#Please measure effect of providing desktop interface to mobile users. EllenCT (talk) 19:37, 5 June 2016 (UTC)

Can someone please point me to any statement resembling me opposing measuring whether mobile users offered the ordinary desktop interface edit or donate more? In the links provided, I said nothing of the sort.--Jimbo Wales (talk) 16:32, 6 June 2016 (UTC)
Jimbo, I can only think that this has been taken from what you said here.5 albert square (talk) 00:33, 7 June 2016 (UTC)

Mobile carrier

Jimmy, I am originally from New Jersey, but now for some years in England. For business, I frequently travel to Nigeria and to Kenya, and the mobile phone situation is maddening to me. I have an Android on Airtel for my Nigeria trips, and also a different SIM chip that I plug in to get Econet in Kenya. When in the UK, I just use my Windows Phone on Lebara's network. It's crazy that I have to have three different mobile numbers for people to reach me, though I'm considering using Google Voice to try to solve for that problem. I am wondering if you could give some advice, as a world traveler, how do you handle your cell phone situation, and what carrier(s) do you use? Don't call me shorely (talk) 12:29, 8 June 2016 (UTC)

I agree with you that it's a bit of a nightmare. I use TPO, but I don't know that our plans would be the best for you - I have no idea what sort of rates we have in Kenya and Nigeria, but my guess is that local carriers are the way to go there. I basically have several numbers - my TPO ones but I also have a Tmobile number in the US because they have a fantastic global roaming plan (free data worldwide in most countries, albeit at 2G speeds). I'm testing the YouMail app on Android to pick up my voicemails on my old US number and transcribe them so that I can listen to them and/or read them wherever I happen to be.--Jimbo Wales (talk) 15:36, 8 June 2016 (UTC)

Research-based RFA reform proposal from a 2013 ACM conference

Please see WT:RFA#Research-based RFA reform proposal from a 2013 ACM conference. EllenCT (talk) 14:24, 11 June 2016 (UTC)

Using a weighting for voting which places more weight on voters who vote often is intriguing. As noted in the paper, the goal was to identify editors who might be "sleeper" POV editors with intentions to affect controversial areas. (As an aside, I'd like to see how a weighting based upon the log of the editors edit count performed.)
However, while the measure apparently succeeds in identifying some areas of concern, I suspect it would have some consequences which are troubling. We already have a concern that admins are part of a cabal. One negative attribute of a cabal is that they become self-perpetuating by only allowing new members who support the cabal - the so-called "old-boys-network" which is rightly excoriated for its lack of diversity. Actually implementing this scoring system would help institutionalize an anti-diversity approach, which, in light of the upcoming diversity conference, would be a step in the wrong direction.--S Philbrick(Talk) 21:36, 11 June 2016 (UTC)
Maybe we should make a list of N administrators who wouldn't have made it by the widest margins if people who hadn't already voted at least M times weren't counted, in an NxM matrix, so those interested could see what they have been up to? Then people could vote on the median best M with M=0 being the status quo. EllenCT (talk) 23:53, 11 June 2016 (UTC)

Jimbo, please imagine that you wanted to hire professionals to do exactly half as much work of generally the same quality and quantity as volunteers do, but at a competitive wage. How much do you estimate that would cost? EllenCT (talk) 14:31, 8 June 2016 (UTC)

A lot, but this isn't how Wikipedia works.--♦IanMacM♦ (talk to me) 14:34, 8 June 2016 (UTC)
It would be helpful to have a rational estimate. Mine is $75,000,000 per year for the English Wikipedia, which I based on the likely false assumption that it would be possible to hire half of the current number of volunteers in proportion to the time they work and who would work at about the same quality about the same fraction of the time. However, I am suspicious that there is no way to estimate volunteer time between editing sessions, so I think that could be off by a substantial margin, more than double. EllenCT (talk) 14:40, 8 June 2016 (UTC)
It certainly isn't the way Wikipedia works, and it is uncertain is such as system as you propose could work, e.g. how to do the hiring, quality control, etc. Also, who would take legal responsibility for the edits? However, before reading your last post I made several "modest" assumptions and came up with a number just for the basic labor. Assume
  • 3,000 editors needed
    • at $30,000 per year (full-time rate) = $600 per week
    • but only working quarter-time (10 hours per week)
  • equals $22.5 million per year (but change any assumptions as you like)
But I think the "production and quality control systems" would take a huge amount of labor and management just to get things working, and that we have no idea that it would work. Best to let a (mostly) volunteer system work as it amazingly has (pretty well). Sure, it's possible to bring in specialists for those really hard, technical articles; or for purely admin work, or (fill in your suggestion here). But I say let a system designed for volunteers work as it has (pretty successfully). Those who don't want to volunteer shouldn't do the work because they won't (in general) be paid. Nobody has promised them they will be paid or that they have any such prospects.
If you'd like to get a paid editor system going, conceptually it is quite simple: 1) fork Wikipedia (get the free mediawiki software and copy all the articles); 2) find the paid editors you want; 3) organize them to do the work you want; and 4) pay them. Of course, you'd need to find the money before starting all of this. Where do you think you'll find, say, $20-100 million each year? I wouldn't expect the WMF to put up any of this - nobody donated money to them based on any such a system. Smallbones(smalltalk) 15:06, 8 June 2016 (UTC)
From Jimmy Wales, who has offered to pay the server costs of any such fork so long as it rids him of the editors he doesn't like. Of course his promises are usually without merit, and soon forgotten. Eric Corbett 15:51, 8 June 2016 (UTC)
I think it is very hard to even think about the topic, because it's pretty alien to the way I think about this in the first place. It's like asking "How much would a bowling alley have to pay people to get an equivalent amount and quality of bowling done as they do now?" Among our editors, many of them have highly compensated other potential uses of their time, and many of them do not. Some are experienced professionals, some are students. Some live in expensive countries with relatively high wages and some live in poor countries with relatively low wages.
Where I'm most interested in this sort of thing is in languages that are evidently not working very well - languages of the developing world which have stagnated for years. I would be very excited and supportive of a pilot project to give part-time fellowships to 100 or 1000 graduate students fluent in a single African language to work in that language wiki say 10 hours or 20 hours a week. (I wrote about this recently on this page.) I'd recommend that such a program be set up and designed to fade out over a set time period such as 5 years... I can't really think of any valid reason not to at least try it. And I don't know specifically what salary would be necessary to attract those candidates (and staff to supervise/manage the process) - I'm just thinking that it would be affordable and super interesting. Research would be necessary to determine what is the most attractive language for something like this, but for chewing on this, I've been thinking about Bambara language.--Jimbo Wales (talk) 15:45, 8 June 2016 (UTC)
How do you feel about starting with the acknowledgment that relatively small amounts of paid editing have always been happening, usually centered around personal and corporate hagiography and scandal whitewashing, and therefore relatively small amounts of organized effort, whether paid, unpaid, or partly paid, are likely to be able to address it? I doubt you will find that controversial at all. What I would really like to engage you in particular and the much wider community on is systemic bias on both gender and supply side economics issues. You've been very supportive on the former, but I wonder if the sheer magnitude of the social impact of the latter might tend to cause a certain amount of unfounded denial. EllenCT (talk) 14:37, 10 June 2016 (UTC)
Agreed, it might work, but this will lead to an adult-dominated Wikipedia, as many minors will not be able to be paid, which will decrease the amount of potential non-paid editors. Seems a bit biased to anyone younger than 16. ThePlatypusofDoom (Talk) 16:58, 8 June 2016 (UTC)
Children would still be able to volunteer in any case, but I'm pretty sure part-time paid editing by kids wouldn't run afoul of contract issues or child labor laws. EllenCT (talk) 14:37, 10 June 2016 (UTC)
It might be cheaper to order a Watson-like AI system from IBM specifically designed for editing Wikipedia. Count Iblis (talk) 20:32, 8 June 2016 (UTC)
On the contrary, I think that would cost about fifteen times as much as human volunteer-to-hires and professionals at present, because Watson can not ask greater than 0th-order questions. EllenCT (talk) 21:02, 8 June 2016 (UTC)

As a computer scientist, I must emphasize that the most-effective AI systems would be hybrid cooperations, of humans working with computerized AI assistants, rather than computers working alone. In fact, consider how the Watson computer won Jeopardy!, by having humans provide the category data in text form (Watson was not "hearing" the spoken words but rather processing the written text) plus a human gave Watson the "beep" to indicate when the spoken words were finished, as when to respond each time after the host spoke. The human contestants noted that Watson's game advantage came mainly by responding first at the buzzer, to lock out the other contestants from responding to earn the points, which compensated for the wrong (or totally bizarre) answers which Watson gave at times. Imagine if human contestants could each run an AI quick Google search when responding. -Wikid77 (talk) 22:18, 8 June 2016 (UTC)

This is absolutely true. Even if we build the best AI assistant system based on Wikipedia, it will only be as good as the quality of the underlying content. The major pitfall preventing achievement of such goals over the past decade may have been the fallacy that it is somehow necessary to make Wikipedia less human-readable (e.g., with Wikidata code numbers) in order to improve it for machine use. There are abundant counter-examples. We absolutely need to streamline the WP:BACKLOGs by integration of article popularity, social impact, and severity of error. If paid editors know they are actually addressing the most important problems, I think it will go much better than it did with earlier, less carefully organized efforts, like Google's paid medical editing efforts. Those did good work, but not very much of it compared to what most people expected. EllenCT (talk) 14:48, 10 June 2016 (UTC)

It's not necessary to fork Wikipedia to get a paid editor system going, just the fundraising system. Wikipedia editors could establish a Wikipedia Editor's Union (WEU) with an elected Editorial Board of Directors (elected in similar fashion as the Arbitration Committee). The elected board would be responsible for organizing and supervising compensation systems. Editors not on the elected board could be appointed to help with this. Requests for Comment would likely be used to determine tasks which would be compensated, and the details of how editors would be compensated. WEU would hold annual fundraisers each June, as a bookend to the WMF's December fundraisers. Is this feasible? We would just be asking for donors to give the hardcore editors some cups of coffee and enough money to buy "Lloyd's of London" liability insurance to cover lawsuits over content. Would the WMF use "superprotect" to prevent such June fundraising banners from going online? Comparing this place to a bowling alley is pretty lame. We're not looking for money to pay the bowlers – the vandals, self-promoters and POV-pushers are quite happy to work for free (or get compensated under-the-table from elsewhere). We're looking for money to pay the volunteers who are working the shoe-rental and refreshment stands, and to pay people to clean the grafitti off the bathroom walls. The bowlers are quite disgusted that they have to take a shit in bathrooms that haven't been cleaned in months, for lack of volunteers willing to clean them. Don't tell us to go to the other bowling alley across town; nobody goes there, and who wants to bowl in a place where there are no spectators to appreciate your work? wbm1058 (talk) 19:13, 9 June 2016 (UTC)

I've actually wondered whether Wikipedia could issue a cryptocurrency to people who achieve certain article milestones, with the effort of making an article counting as "proof of work". True, it doesn't seem like it would become worth anything -- thing is, I thought the same about Bitcoins. I do not yet understand the magic by which wealth is defined out of nothing, but the whole world is governed by it. Wnt (talk) 19:27, 9 June 2016 (UTC)
I can explain bitcoin to ya: it's 1/2 secular religion and 1/2 Ponzi scheme. As long as people believe, these theoretical credits have value. I'm an atheist myself... Carrite (talk) 01:37, 10 June 2016 (UTC)
It is about as religious as any other bit of math. Considering it has no top, I don't see how it can be a ponzi scheme. I agree it is hard to understand though. HighInBC Need help? {{ping|HighInBC}} 15:57, 10 June 2016 (UTC)
All true, but the same is true of diamonds, designer dresses, or the Federal Reserve. Wnt (talk) 10:15, 10 June 2016 (UTC)
Touché! Carrite (talk) 11:54, 10 June 2016 (UTC)
Do you feel similarly about the new generation of blockchain cryptocurrencies with embedded contracts such as Ethereum? EllenCT (talk) 14:48, 10 June 2016 (UTC)
Oh, yeah... From the WP link: "The value token of the Ethereum blockchain is called ether. It is traded on cryptocurrency exchanges like any other cryptocurrency; Buterin himself sold 25% of his ETH holding in April 2016, describing this diversification as "sound financial planning." — AKA, "cashing in." Carrite (talk) 15:54, 10 June 2016 (UTC)
In the five-dimensional post-Minkowski spacetime, I'm guessing you are not tulips' favorite person. EllenCT (talk) 03:16, 11 June 2016 (UTC)
One might be tempted to say that that the tulip story presented in Mackay'sExtraordinary Popular Delusions and the Madness of Crowds is itself an extraordinary popular delusion resulting in some madness in the crowds. Mightn't one? Smallbones(smalltalk) 14:55, 11 June 2016 (UTC)\
Yes. EllenCT (talk) 15:05, 11 June 2016 (UTC)
From The Hitchhiker's Guide to the Galaxy: "Since we decided a few weeks ago to adopt the leaf as legal tender, we have, of course, all become immensely rich. But we have also," continued the management consultant, "run into a small inflation problem on account of the high level of leaf availability, which means that, I gather, the current going rate has something like three deciduous forests buying one ship's peanut. So in order to obviate this problem," he continued, "and effectively revalue the leaf, we are about to embark on a massive defoliation campaign, and...er, burn down all the forests. I think you'll all agree that's a sensible move under the circumstances."--♦IanMacM♦ (talk to me) 09:55, 11 June 2016 (UTC)
"To give real service you must add something which cannot be bought or measured with money, and that is sincerity and integrity." -- Douglas Adams. EllenCT (talk) 14:40, 12 June 2016 (UTC)

Articles about Wikipedia's Bureaucracy

I don't see that any of these have been mentioned here yet (sorry if I missed any such discussion):

One of their most striking findings is that, even on Wikipedia, the so-called “Iron Law of Oligarchy”—a.k.a. rule by an elite few—holds sway. German sociologist Robert Michels coined the phrase in 1911, while studying Italian political parties, and it led him to conclude that democracy was doomed. “He ended up working for Mussolini,” said DeDeo, who naturally learned about Michels via Wikipedia.
“You start with a decentralized democratic system, but over time you get the emergence of a leadership class with privileged access to information and social networks,” DeDeo explained. “Their interests begin to diverge from the rest of the group. They no longer have the same needs and goals. So not only do they come to gain the most power within the system, but they may use it in ways that conflict with the needs of everybody else.”
* * *
[Yun's study (below)] found that a fairly small number of Wikipedia editors exert a major influence on the site. And just as DeDeo and Heaberlin’s analysis predicts, that editing inequality is increasing over time.

--David Tornheim (talk) 02:31, 12 June 2016 (UTC)

The Gizmodo headline and the DeDeo quotes are absurd attempts at clickbait and attracting attention, respectively. The study found crystallization around content creation, moderation and social goals, not people and fiefdoms:
"Table 2. Top nine Louvain clusters, by number of nodes. Communities fall into three classifications (user-user, user-content, user-administration), based on the interactions they govern; we determine these labels by inspecting the top ten nodes by centrality within each cluster.
Rank Fraction of System Classification Topic
1 24.8% User-Content Article Quality
2 22.9% User-User Collaboration
3 17.1% User-Administration Administrators
4 14.7% User-Content Formatting Articles
5 10.5% User-Content Content Policies
6 5.4% User-User Wiki-LARPing
7 2.0% User-Content Page Templates
8 1.3% User-User/User-Content Experts and Credentials
9 1.0% User-User Humor
" EllenCT (talk) 15:21, 12 June 2016 (UTC)

More of DeDeo's research on Wikipedia found here:

--David Tornheim (talk) 02:46, 12 June 2016 (UTC)

I actually posted about one of these studies on this page in April. [64] Everymorning (talk) 20:43, 12 June 2016 (UTC)
The worst 'Fiefdoms' on wikipedia are around content creation. So many walled gardens... Only in death does duty end (talk) 12:26, 13 June 2016 (UTC)
The worst walled gardens have been around the relatively new religions. If we had automated support for article importance rankings, we could classify them between fiction and nonfiction using objective measures. EllenCT (talk) 17:24, 13 June 2016 (UTC)

"Article Quality" is the most undescriptive and nebulous category I've ever seen. It contains both gnome edits and major content-work, and as such is not representative of anything. Pinguinn 🐧 22:25, 14 June 2016 (UTC)

‎Amendment request on arbitration decision against Rodhullandemu

You are involved in a recently-filed request for clarification or amendment from the Arbitration Committee. Please review the request at Wikipedia:Arbitration/Requests/Clarification and Amendment#Amendment request: Rodhullandemu and, if you wish to do so, enter your statement and any other material you wish to submit to the Arbitration Committee. Additionally, the Wikipedia:Arbitration guide may be of use.

Thanks, --George Ho (talk) 06:07, 14 June 2016 (UTC)

FYI, it has already been closed.--S Philbrick(Talk) 00:14, 15 June 2016 (UTC)

You’re invited to teahouse!

 
Hello! Jimbo Wales, you are invited to join other new editors and friendly hosts in the Teahouse. The Teahouse is an awesome place to meet people, ask questions and learn more about Wikipedia. Please join us!

–– مجتبیٰ (Talk!) 04:20, 13 June 2016 (UTC)

So Jimbo is a "new editor" now? Well, I suppose that I'll have to welcome him to Wikipedia! ThePlatypusofDoom (Talk) 13:14, 14 June 2016 (UTC)

You've really got to be kidding us don't you? Hamham31Heke!KushKush! 04:25, 15 June 2016 (UTC)

Massive expansion of National Security Letters

While National Security Letters have been cast far and wide as an intelligence dragnet, it appears that the legal authority to demand many types of information, such as browsing history, did not actually exist when the FBI made the demands. Because of gag orders the victims have been greatly hindered from explaining this fact, but some of them were turning over only names and length of service. [65][66][67] The censorship itself, however, was actually upheld -- [68] Now that the limitation on NSLs is finally known the public, of course, the response has been to try to actually give the FBI the power to demand records of browsing history with two separate bills - a provision in S-3017 [69] and (AFAICT) a planned amendment to the "Email Privacy Act" by Senator John Cornyn (OLL16601) on which markup has been delayed [70].

Before we get into what extreme measures history usually has required to roll back such measures, there is no reason for Wikipedia not to take basic measures to head off NSLs before they happen, if it is indeed not too late. The cult of A/B testing or minor interface features are not sufficient excuses to generate records of which IP read what article when that data invites seizure. The data should simply not be generated; if generated anyway it must be immediately disposed; and a firm guarantee needs to be made of this in the privacy policy. As the world saw in the Apple case, there is still some judicial resistance to making programmers add spying features to a system, that does not exist when the information is already being collected. So don't have these features in the system! And fire any moles who try to tell you there's no possible way to do without them. Wnt (talk) 11:33, 8 June 2016 (UTC)

The Foundation already provides complete browsing histories of identifiable people to Stanford University. Please see meta:Talk:2016 Strategy/Draft WMF Strategy#Goal proposals for strategic priority 2: stop storing browsing histories and relying on compromised hardware. Would refusing to save the information in question violate any other agreements into which the Foundation has entered? EllenCT (talk) 12:23, 8 June 2016 (UTC)
I'll let someone from tech weigh in here, but I see no inherent tension between doing solid A/B testing and taking pretty strong measures to not have data that would be subject to such demands. My understanding - but I don't work day to day at the Foundation - is that the staff are quite vigilant and passionate about these issues and do a great job. The most effective techniques are to burn all the data as quickly as possible, to tokenize/anonymize (but note that for ip addresses this is of limited value since the space is relatively small), and to be thoughtful about choosing the smallest reasonable sample size for research. What I don't think is necessary is a blanket internal ban on A/B testing.
A side note that may be of some interest - storing a unique cookie as the identifier is a much stronger technique to protect privacy under these conditions than storing an ip address. You may find that counter-intuitive, so I'm happy to explain what I mean if you'd like.--Jimbo Wales (talk) 15:24, 8 June 2016 (UTC)
@Jimbo Wales: Ellen's link provided some helpful - but disturbing - background. According to User:LZia (WMF), "The browsing history data generated for this research can be built by anyone who gains access to webrequest logs and knows how to write code and run it, i.e., the fact that we are building the traces does not put the user in a riskier situation than if we would not build the traces. The raw data and the technology for building traces is there. If someone gets access to the raw data, they can create those traces themselves." And "webrequest logs are kept for 60 days. The privacy policy allows us to keep them for up to 90 days." But what is the reason for keeping these logs that long, or even having them at all? (see also Wikipedia:Village pump (technical)/Archive 144) Wnt (talk) 22:16, 8 June 2016 (UTC)
I don't know but that doesn't seem particularly problematic to me. To be clear - if we were receiving requests and responding to them on a regular basis then getting rid of the logs faster would be great. But this is really only a hypothetical problem. Review the transparency report - it's not clear to me that this is a live issue right now. I'd personally like to see the data burned much faster than 60 days, but I'm not super agitated about it because it isn't really a problem right now unless I've missed something.--Jimbo Wales (talk) 09:57, 9 June 2016 (UTC)
Well, imagine someone gets stopped at the airport for having the wrong name or making a phone call in Arabic, and rather than just admit that they were harassing the wrong person as usual, the investigators decide they want to accuse him of a terrorist conspiracy. They come up with a phone call he received last month from an alleged business client who's on their list. Now what? Well, if they can demand his last 60 days of Wikipedia browsing, maybe they can show he looked up the White House, and suggest he was casing the place with a plan for an attack. Maybe he looked up something about terrorism in his country, or how to make a knife. God help him if he actually looked up chemicals of some sort. Having access to a thorough history allows a reasonably creative investigator many opportunities to cobble together a plot. "If you give me six lines written by the hand of the most honest of men, I will find something in them which will hang him." People use Wikipedia a lot, and it might be used to tell a story. Let's make it a short story. Wnt (talk) 11:25, 9 June 2016 (UTC)
I agree with you, with the caveat that since such a thing has never happened and seems highly unlikely to happen, this should be a relatively small factor in our overall optimization.--Jimbo Wales (talk) 12:03, 9 June 2016 (UTC)
Are we allowed to see any summaries of identifiable reader information requests made by law enforcement? How about the responses? If a valid subpoena was served for the downstream Stanford data, would anyone in the Foundation ever know? EllenCT (talk) 15:25, 10 June 2016 (UTC)
@EllenCT: That's the whole point of "National Security Letters". No, you're not allowed to see the summaries. Jimbo is occasionally asked if he has received one and has said no (I asked him pretty recently myself so I won't pester him now), a tactic termed a "warrant canary", but there is really no telling when the government will just write out a paper ordering people to lie under pain of prosecution, so it's your call what you make of it. Wnt (talk) 17:14, 10 June 2016 (UTC)
Do you think such prior restraint on speech is constitutional? If not, would collective civil disobedience of the statute in hopes of provoking a constitutional review in court be preferable to unconstitutional obedience of the proposed statute? EllenCT (talk) 17:17, 10 June 2016 (UTC)
Emphatically no, but the problem is as I cited above, the court did not find that way. Collective civil disobedience may sound great, but in most implementations that amounts to some individual person likely doing hard time over what at least for now is probably going to be a short list of people to betray who might even be as unsavory as the government suggests. There is a way to do civil disobedience though - that would be to not make the logs of the information that you don't want to hand over in the first place. Wnt (talk) 23:32, 10 June 2016 (UTC)
That would be reneging on a research agreement with Stanford, but not actually disobeying the government. What is the potential penalty for cutting off Stanford's access to everyone's browsing history? Does John Hennisey make a snide comment at a faculty senate meeting?
So, let's see, on one hand you can save everyone in Russia who looked up how to make a bong from Putin's polonium-210 assassination squad, but on the other hand Bob West already defended his thesis, Jure Leskovec is probably going to make tenure if he hasn't already, but Foundation staffer User:LZia (WMF) and her fellow graduate student Ashwin Paranjape suddenly lose access to the reader interest panopticon, and who knows what that could do to their prospects? Probably not a whole lot compared to polonium-210. And is the output really worth it?
I'm guessing that is not worth risking polonium-210 for the bong browsing readers. See Table 2 on page 10 of 10 here if there is any question. We are literally exposing all the projects' readers' browsing history to anyone able to convince Stanford staff that they have a valid law enforcement request in order to brush up the "See also" sections of popular culture articles. But hey why worry, there's no reason to believe that such a thing has ever happened.[71][72] EllenCT (talk) 07:38, 11 June 2016 (UTC)
@EllenCT: Very interesting and well researched response! I feel like I cleared a cargo cult airstrip and you just landed something on it. Nonetheless, I am not so clear on what access they have. The first link you list says, "although the Wikipedia log data used in this paper is not publicly available, a powerful dataset derived from it, the matrix of pairwise transition counts between all English Wikipedia articles, was recently published by the Wikimedia Foundation" - which I might take to be reassuring. But the authors of the second paper in their discussion seemed very hot to get ahold of exactly this data - they hadn't at the time, and I definitely would not be happy to see them do so. You can't reliably keep such data secret when you're transmitting it over any kind of network connection - Snowden showed us that much - and Stanford is not a place I would trust anyway.
But there's even a broader issue here: what is "mining and modeling large social and information networks, their evolution, and diffusion of information and influence over them", and what will it accomplish to be "developing techniques in data science, data mining, network analysis, machine learning, and natural language processing", as those two people proudly advertise? Well, the answer of course comes from the source of all computer technology and law, China, and their imitators in Britain. Data mining is about making the Massa the most important thing in your life, from the moment you get up to the moment you go to bed, the only thing you are thinking about when you say something to a friend or choose a socially favorable activity to participate in. And the AIs are the Massa. We see a world of people who write every sentence for keywords as if they were hopelessly trying to get their job application read by a human or their news article to appear on Google; people for whom the content doesn't exist, only the cause of service, because the alternative is not having a place to life or a way to buy anything they need. So we need to look upon this computer science not as an enlightenment but as an enemy, and we should try our best to hinder it, not help it. Wnt (talk) 14:27, 11 June 2016 (UTC)
Are you sure you want to censor an algorithm or data set that might expose the authorship of a subversive samizdat to the Stasi, when it might also tell an epidemiologist an easy way to add a few months to everyone's productive life? Is the social context that matters. If there is no motivation for the Stasi to compile dossiers then there is no reason to be afraid they will abuse what they or anyone else can observe. But what gives the police the motivation to abuse their power? Economic conditions! Which, of course, is why I maintain that the broadly general and widely read economics articles are far more important to remove systemic bias from than, well, essentially any of the substantial remaining gender-oriented systemic biases, except those which intersect with the influential general economics topics. There has been no apparent progress at meta:Research talk:Measuring article importance for almost a year, and they never even got to the basics of simple and obvious things like fiction vs. non-fiction, historical vs. contemporary topics, pop culture vs. science, applied vs. theoretical science, or quantification of the number of people or productive hours of life involved with the topic. @EpochFail: what is the status of meta:Research:Measuring article importance and do you have any plans to measure such factors? EllenCT (talk) 14:59, 11 June 2016 (UTC)
I wasn't suggesting banning their research entirely, nor do I have the power; but I don't see the benefit to collecting data from Wikipedia readers to assist with their research, and collectively we might have the power to just keep things simple. That article link... I don't know what to make of it. It's like the Yi Jing, only less comprehensible. Wnt (talk) 15:38, 11 June 2016 (UTC)
Does [73] fail with a "network error" before closing the connection when you try to download too? I'm sorry. I'll try to find an alternative. The related work may help to explain, but it came up earlier here in the context of what a question-answering-and-asking system would need to model to understand natural language. EllenCT (talk) 16:20, 11 June 2016 (UTC)
@Wnt: what do you make of the "DIAMONDS bass-ackwards analysis" on pages 39-49 of the slides at [74] in terms of their usefulness for (a) measuring article importance and (b) supporting multi-agent modeling in a question answering-and-asking AI assistant system? EllenCT (talk) 19:26, 11 June 2016 (UTC)
I can access this stuff, but I don't know what to do with it. I'm not sure what correlations were measured in what data about it. It seems like an incomplete version of Yi Jing - the respective hexagrams being 2, 18, 29, 31, 11, 12, 7?, 16, perhaps, though as both aspire to cover all situations that could come up within a regular framework, other correspondences should exist. Wnt (talk) 19:48, 11 June 2016 (UTC)
Both are ontological frameworks designed to categorize a wider set of concepts, but DIAMONDS are designed to analyze people's motivations in a wide variety of social and otherwise contextual situations, and the similarity with folk ontology pertaining to predicting the future is very striking, now that you point out the details.
In any case, do you agree that "understanding our readers to better serve their needs" should be Communications team projects, as already reflected in the Annual Plan, and not a Research team project requiring storing browsing histories and shipping them off to third parties about whom nobody at the Foundation is allowed to know if they are subpoenaed, all for no apparent purpose other than lengthening "See also" sections on popular culture articles, especially if we want to address the risk that "a competitor provides a better reader experience for Wikimedia content, diminishing our ability to turn readers into editors and donors"? EllenCT (talk) 20:25, 11 June 2016 (UTC)
Well, understanding the readers to better serve their needs is a great idea, but for the antediluvians among us that seems best done by asking them for opinions and suggestions. Among the avant-garde of libertarian wishful thinking, it is even said that NSA-scale data collection only leads to "intelligence failure" due to the inability to process it all. I am not so optimistic as all that where the NSA is concerned, but WMF employs fewer people and there might be something to it here. Whether there is or not, I'd be content to settle for opinions and suggestions for no reason other than that is how things ought to be done. Wnt (talk) 22:23, 11 June 2016 (UTC)
And if the suggestion is to have the Research team personnel working on understanding readers choose between helping Communications with learner-centered advances and helping Research create models ranking article importance matching the median opinion scores of panels of human judges instead of storing reader-identifiable logs, and then enhancing them with specific features such as fiction vs. non-fiction, historical vs. contemporary topics, pop culture vs. science, applied vs. theoretical science, or quantification of the number of people or productive hours of life involved? EllenCT (talk) 23:43, 11 June 2016 (UTC)
Honestly, I don't fully understand what you mean with either option, but I think the first is closer to what I think. My feeling is that the encyclopedia is what it is and if editors invited more readers to the Refdesks and put more effort into filling article gaps they suggest, and if developers invited more editors to propose big/bot projects as needed and more readers to comment on what problems they have and what ideas they suggest, and if we all think about the big issues everybody knows about like the bureaucracy and nastiness and gender harassment and have pow-wows and come up with ideas, while cherishing fundamental principles, then we'll make reasonable choices. I just don't see experimentation watching reader habits as being a better way of getting to their desires. And do bear in mind there is a fundamental distinction in goals between making Wikipedia the best resource available and making WMF as large and well-funded as possible. The former includes being willing, for example, to offend people in pursuit of the truth, without consideration of how it affects their donations. Yet I feel that it is actually a better way of achieving the other goal than pursuing that goal directly, because (for example) any short-sighted compromises on something like that create a politically unstable situation that must eventually unravel. Wnt (talk) 10:21, 12 June 2016 (UTC)
"Don't put your trust in princes,
each a son of man
in whom there is no help.
His spirit departs,
and he returns to the earth.
In that very day, his thoughts perish."
(Psalm 146: 3, 4, World English Bible)
Wavelength (talk) 19:17, 9 June 2016 (UTC) and 21:11, 9 June 2016 (UTC)
"By their fruit you shall know them." -- Matthew 7:16. EllenCT (talk) 07:04, 11 June 2016 (UTC)
Ecclesiastes 9:2, 5Wavelength (talk) 16:41, 13 June 2016 (UTC)
Daniel 12:2-3; Xkcd 1450. EllenCT (talk) 17:59, 13 June 2016 (UTC)

Specific proposals

Jimbo, (1) should the Foundation adopt an explicit policy of not sending reader-identifying logs to any destination where the Foundation is not certain to be informed of law enforcement requests for them? (2) Even if that means not storing them at all? (3) Should the reader-identifying log research program used to enhance the see also sections of articles continue? (4) Should Research team members and affiliates who have been working on the reader-identifying logs be offered a choice between working on Communications' reader-understanding projects by enhancing them with recent advances in learner-centered pedagogical techniques and/or working on the dormant program to classify article importance, using measures derived from opinion scores of judges and topic attribute features? EllenCT (talk) 14:02, 12 June 2016 (UTC)

@EpochFail: and @Jbarbara (WMF): I am also very interested in your opinions on question (4). EllenCT (talk) 17:35, 13 June 2016 (UTC)

Re. (4), we already do make choices about which projects we pursue. --EpochFail (talkcontribs) 00:47, 14 June 2016 (UTC)
@EpochFail: well while I wouldn't ask you to confirm or deny it, I think your recent increase in volunteer requests strongly suggests that your team isn't getting reasonable headcount. How many FTE engineers would you need to get an automated article importance classification system with real topic attributes on the scale of ORES? How many do you think would be appropriate if Communications approved experimenting with learner-centered pedagogical advances in their reader understanding efforts? How many would you get back if we stopped storing reader-identifying logs and the personnel and affiliates involved with those had to stop using them? EllenCT (talk) 16:08, 14 June 2016 (UTC)
@EllenCT: If I had one engineer for ~3 months, I think that we could get a reliable article importance prediction system in place. It would likely use ORES' infrastructure to provide a public end-point, but that would be an engineering decision that we'd need to make based on how the system shapes up. Short of a full time engineer, I'd be happy to mentor a senior CS student (or equivalently skilled individual) on an IEG (6 months/part time) or something like that. I've had success with that pattern in the past, but it takes more time (and more of my time).
Re. "learner-centered pedagogical advances", I'm not familiar with that jargon. Further, it would be unusual, I think, to have a communications department leading a basic research effort. Generally, I think this kind of exploratory work is best lead by professional researchers and, if possible, in collaboration with academics. We do that a lot on the Research Team at the WMF since it's mutually beneficial. Are there any researchers who are actively pursuing this methodology/framework/lens who you think might be interested in looking at Wikipedia? I'd be happy to cold-email them to ask about their interest. --EpochFail (talkcontribs)
I, for one, will do what I can to support staffing an article importance prediction system, because it is sorely needed. My learner-centered suggestion is not well-formed at present and I will provide more details about the specifics on your talk page when I can compose something much more coherent and concrete than I have so far. EllenCT (talk) 12:46, 15 June 2016 (UTC)
@EpochFail: can Random Forest features be seeded with topic feature suggestions, or are they derived automatically and labeled after selection for predictive worth? Regarding reader-centeredness, how would you feel about using browser-reported geolocation and language preferences to hint language selection? Or Simple English? To what extent does that already happen? What do you think can be done with reader-identifying logs that would improve the encyclopedia the most? EllenCT (talk) 18:34, 15 June 2016 (UTC)
Hi EllenCT. The best way to get this prioritized is to write about it. If you can give me a write-up of potential uses of such a prediction model, I'll try my best to get someone (or me) on it. I think a good place to put such a proposal is the m:Grants:IdeaLab. See one of my recent proposals as an example: m:Grants:IdeaLab/Fast and slow new article review. In the short term, we could try to turn the work into an IEG or a project for a grad student. --EpochFail (talkcontribs) 19:26, 15 June 2016 (UTC)

The Signpost: 15 June 2016