Wikipedia talk:Canvassing/Archive 6

Latest comment: 5 years ago by Gwillhickers in topic Recent changes
Archive 1Archive 4Archive 5Archive 6Archive 7

Productiveness of the guideline

I have been warned about "canvassing" over and over. Is the guideline against WP:NOTBURO? If not, how are we supposed to attract any discussion? Very little number of editors participated and are aware of discussions. WP:NOTAVOTE, Stanton, is now an essay. Why not the same for WP:CANVASS? George Ho (talk) 19:06, 3 December 2016 (UTC)
Oops. I accidentally pinged you because I originally intended to post this at WP:VPM, which has the "canvassing vs notice" talk taking place. George Ho (talk) 19:07, 3 December 2016 (UTC)

George Ho, some people misuse the guideline. In what ways were you accused of canvassing? If you contacting editors complies with the "Appropriate notification" section of the guideline, you are fine. Flyer22 Reborn (talk) 22:39, 3 December 2016 (UTC)
Won't matter whether I followed the rules or not, Flyer22 (reincarnated actually?), unless I'm wrong. I found past discussions about the page's guideline status: /Archive 4#Who ever decided that this should be a guideline?, Wikipedia talk:Spam/Archive 2#Factor canvassing, Wikipedia talk:Spam/Archive 2#WP:CANVASS. One of the discussions was about splitting the canvassing portion from WP:Spam. Looking at the merger discussion, there wasn't an established majority, a vote, or something. Perhaps HighInBC can explain all of this. To me, the guideline has not produced any good results. Many editors or whoever were warned and/or punished per this rule. Also, this rule can be abused in many ways, whether it was violated or not. --George Ho (talk) 23:19, 3 December 2016 (UTC)
The guideline is needed because the problematic aspects it mentions have proven to be, well...problems. Flyer22 Reborn (talk) 00:56, 4 December 2016 (UTC)
Here's my interpretation of the rule's spirit: editors must may notify a just small number of people, write neutral messages, and select a good number of people of at least two sides. Isn't that right? If that's it, I wonder whether this method has produced any good results. I'm thinking scenarios when small number of people are notified:
  1. Not one person participated in the discussion. A closer may interpret this as no opposition to the proposal. Something changes per closure.
  2. Those people participated, but the participation is very small. Results vary, depending on the closer's rationale.
  3. The amount of participants is very large. Results vary.
Shall we find examples? The result of a discussion can be good or bad, depending on canvassing. I want to prove that canvassing is a good thing, but how do I prove it? Or maybe I was discussing "mass posting", which looks vague to me. --George Ho (talk) 03:09, 4 December 2016 (UTC)
That's not really what it says though. It's possible to notify a very large number of people, e.g. by posting to WP:VPPOL, either as the forum in which to host the discussion, or just a pointer to an already existing discussion. The issues with canvassing are primarily of two sorts: 1) notifying specific individuals you know are likely to vote your way; 2) notifying specific groups of editors (e.g. a particular wikiproject that is not really closely tied to the subject or nature of the discussion; or, among many actually relevant projects, such as biology ones, notifying only one or a few specifically chosen ones) whose general position on the matter you think will support your version among the potential outcomes. The second of these comes up frequently in WP:CONLEVEL disputes. If a particular wikiproject has a history of defiance against site-wide rules in favor of its own WP:LOCALCONSENSUS, and few editors outside that little wannabe-fiefdom care or have historically cared all that much, then it's fairly easy to get 5 or 10 or even 20 people from that project to come bloc-vote against anything that is a move toward increased site-wide consistency and less "we make our own rules at our project" territoriality, and this is often enough to drown out more diverse voices, from all over Wikipedia, in favor of the shift toward having a consistent approach, at least early on. These little "rebellions" tend not to last indefinitely, but some have managed to persist for several years to the detriment of the encyclopedia and its readers. The general nature of this second, more clever form of canvassing was recently discussed elsewhere, e.g. at Wikipedia:Village pump (miscellaneous)/Archive 54#Need opinions on canvassing vs. notices, at two now-archived WT:MOS threads about repeated canvassing of anti-MoS editors from WT:FAC, as just one multi-page example. The former kind of canvassing – just rounding up one's buddies who all think alike or who like to back each other up in a WP:TAGTEAM – is more obvious and less systemically problematic, though often higher profile because of its blatantness.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  12:50, 4 December 2016 (UTC)
George, I added a note at Wikipedia:Redirects for discussion/Log/2016 November 28#Wikipedia:JUSTAVOTE that I didn't think you were canvassing. But clearly some people don't really think they were pertinent to invite to the discussion directly.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  13:03, 4 December 2016 (UTC)

Is the consensus still against everything that the page says, i.e. canvassing? Can we be certain that the rule produce more good than bad? Can we lose a number of good editors due to this rule? George Ho (talk) 20:29, 4 December 2016 (UTC)

Not sure why I was pinged, can you refresh my memory as to how I am involved? HighInBC Need help? {{ping|HighInBC}} 23:13, 4 December 2016 (UTC)
You were "Chillum", HighInBC. I saw your former name in one of archived threads at "Archive 4". You said this as response to someone: "The prohibition on canvassing is not a matter of opinion, but something that is regularly enforced through warnings and blocking. Being a guideline is far more representative of its status than calling it an essay." George Ho (talk) 08:59, 5 December 2016 (UTC)
I stand by that position. HighInBC Need help? {{ping|HighInBC}} 03:10, 6 December 2016 (UTC)
HighInBC, Stanton, and Flyer22 (reborn or whatever), if no one notifies others about the discussion, and the discussion lacks participation, and the administrator decides to be bold as "no opposition" toward some proposal, would that be gaming the system or something? This is George Ho actually (Talk) 01:25, 9 December 2016 (UTC)
I think this is relevant: Wikipedia:Silence and consensus. I am not familiar with the circumstances of the specific issue you are concerned about but I think the essay gives good advise for implementing ideas that lack community attention. HighInBC Need help? {{ping|HighInBC}} 18:23, 9 December 2016 (UTC)
Some of the common things to do if one thinks that participation is insufficient in numbers or two narrow in breadth of interest/opinion:
  1. "Advertise" the discussion at the appropriate Village Pump page (if it's not a trivial matter). If it's a really big deal that could affect many pages, also add it to WP:CENT.
  2. Notify the talk pages of policies, guidelines, or other "big deal" pages that are populated by editors who focus on matters that are central to the discussion. E.g., if it's an RfC about whether a particular use of a tertiary source at an article is appropriate, WT:RS, WT:V and WP:NOR might be good places to post a pointer to the discussion, especially if it's an interpretational problem that could indicate one or more of those pages needs to be clarified in some way.
  3. Notify all of the participants in previous rounds of an extended dicussion, by pinging them all in the new discussion (there are limits to how many can be pinged in a single post; see WP:PING).
  4. Notify all of the relevant wikiprojects if it's something that directly (not vaguely) relates to them; or if this would be excessive (e.g. all the biology projects) notify the main/central/meta project(s) that is/are their "parent(s)" (e.g. WP:WikiProject Tree of Life, WP:WikiProject Animals, WP:WikiProject Botany for a general biology RfC).
  5. Convert a non-"process" discussion into a process one, e.g. start and actual WP:RM as a subthread of a move discussion that isn't resolving, or start an actual RfC as a subthread of some other dispute resolution that isn't coming to a consensus; this will auto-expand the level of input since WP:RM and WP:FRS are site-wide "services" that get the attention of many editors who are not connected to the issue.
  6. Use a topical WP:NOTICEBOARD, or WP:3O, to get additional input from people who aren't already in the dispute.
Some other people can probably think of others. As I said above, the main canvassing problems are inviting specific involved editors but not others, and drawing in specific groups (wikiprojects or whatever) of editors with a "stake" but not others, both of which have a vote-stacking effect.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  19:23, 9 December 2016 (UTC)

"...and do not send messages to users who have asked not to receive them."

Please add a minimal amount of info on how you do that (even a link to another help page would suffice).

How do you "ask" to not recieve canvas messages? Is there a template you put on your user talk, or what? CapnZapp (talk) 23:11, 25 January 2017 (UTC)

Inappropriate Actions / Secret

After becoming aware of an issue within deletion discussions, I wonder if the above can be expanded to include not notifying interested parties in an effort to achieve a consensus based upon only the nominators rationale and discussion among those who habitually participate in that space. While notifying those in the editing history of items up for deletion is not mandatory, the fact that it is recorded within the guidelines suggests that it is optimal and Wikipedia:Consensus indicates that it should be expected that there might be opposing views that need to be weighed for a sound decision to be made. Since deletion discussions are time limited, I would suggest that not allowing all potentially interested parties the opportunity to contribute makes any decision made from reviewing only the arguments of those notified through a more limited notification more likely to be challenged after a decision of delete - with the subsequent expense of time and energy. Comprehensive notification is a more efficient method of allowing consensus to form.

If it felt that the above sub-section is not the appropriate one to include deliberate non notification, perhaps another might be suggested? LessHeard vanU (talk) 21:02, 17 February 2017 (UTC)

Note that this proposal derives from a discussion at WP:ANI (permalink) and at at User talk:VegaDark(permalink). For the sake of transparency, it would have been better for LHvU to disclose that this all relates to the highly unusual situation of a CFD debate on a category relating to LHvU. Not compulsory of course, but it is surely optimal to disclose a direct personal interest like this so that editors can make their own assessment of how much is concern for healthy process and how much is personal interest in the fate of a category about LHVU, populated by LHVU's wiki-friends.
At the second venue, I offered some comments[1] on LHvU's ideas. In summary, LHvU seems to be seeking a degree of notification which I have rarely seen at AFD, and AFAICR have never seen at CFD. It would impose a high burden on editors who nominated categories for renaming for deletion or merger, and in most cases I think that burden would be massively disproportionate to any potential benefit. In particular, CFD and TFD do not deal with substantive content. CFD deals solely with navigational devices (categories), and TFD deals primarily with a different form of navigational device (navboxes). The nature of the decisions thee is very different to AFD, which is all about whether to remove substantive content.
Both CFD and TFD are an ongoing maintenance process, adjusting navigational aids as the encyclopedia grows and develops. It would be wrong to impose on them the sort of burdens imposed at AFD, because the stakes are so much lower at CFD and TFD, and the effect of those burdens would be so much higher. In particular, LHvU proposed to VegaDark that it was wrong to launch a CFD without advising those who have edited or populated categories, and since LHvU has not clarified the scope oh their proposal here, I presume that they also intend it to apply to TFD.
The problem is that this would require require a revision history search of each page in a category, which could easily take many hours for a larger categories. In 11 years at CFD I do not recall ever seeing that done, just as I don't recall from sporadic visits to TFD any sign of nominations accompanied by systematic notifications of those who applied a template to pages. 8 days ago I opened a CFD discussion on 596 categories, at WP:CFD 2017 February 10#Subcategories_of_London_Boroughs. Those categories between them contains many thousands of articles, and there is just no human way that it is feasible to identify and notify all the contributors. That CFD was unusually large, but in the last 8 weeks I alone have opened at least half a dozen CFDs which involved more than 100 categories each. The sort of notification which LHVU seeks would make that sort of maintenance impossible, and it directly contradicts the explicit guidance in the nutshell at the top of the guideline When notifying other editors of discussions, keep the number of notifications small.
So I think that a one-size-fits-all approach to desirable XFD notification is unfeasible. If LHVU wants to take specific proposal to individual XFD pages, they would be better considered against the needs of that process.
In any case, the clear focus of this guideline guideline is about how to avoid votestacking. It is largely a collection of negatives, and LHVU's desire for positive burdens would fit poorly into the overall purpose of this guideline. -BrownHairedGirl (talk) • (contribs) 11:22, 18 February 2017 (UTC)
The genesis of this discussion was User:BrownHairedGirl's comments at User talk:VegaDark#Ultra-Speedy deletions, which I read after going to the talk page of User:Vega Dark. It seems to me that there might be an issue where an understanding of policy disallows discussions to occur before action is taken, which I feel is contrary to the concept of consensus i.e. as a category is unpopulated for 7 days it might be deleted even if notice of the intent to delete is only provided late on the 6th qualifying day. Where depopulation was the result of error or inattention the opportunity to Wikipedia:FIXIT is potentially lost, unless there is the even more burdensome business of undeletion requests and discussion - not all parties are admins with the ability to reverse an admin action, and most will correctly not without attempting consensus with the original admin. There are good reasons why it is noted on deletion guidelines that interested parties should be notified, in case the issue can be resolved before going to a final determination of consensus on the original request.
Yes, I went to that page after the matter referred to above as I had been worried about a comment made my User:VegaDark in response to a query why people with a category in userspace were not advised that it had been nominated for deletion - the response was that such notice might cause a delay in the prompt deletion of the category in question as notified parties might argue for keep under Wikipedia:ILIKEIT. My understanding of Wikipedia:CONSENSUS is the very light weight attached to such arguments should not be a reason for not providing the opportunity for people to wield it. In the event, those who were alerted to the matter provided policy and guideline based reasons why deletion was not in the best interests of the project - and the result was keep. I am concerned that where notification is suggested within the guideline and that they are either not given or only belatedly, in order to ensure the prompt consideration of consensus only under the "strict" understanding of policy as provided by the proposer without the benefit of other views - particularly of those who have contributed to the subject - that there might be the potential of the charge of Wikipedia:VOTESTACKING by omission.
Even when I was active as a contributor and admin, I was very rarely found in the Deletion discussions area of WP except with regard to articles. I am and was aware of the sheer amount of drudgery involved in dealing with the wiki-gnoming aspects, and how there was always more work available than there were editors to deal with it. However, Wikipedia is NOTFINISHED and arguments about the increased load upon those involved in the less acclaimed back rooms should be viewed in the light that time is not a limiting factor. Yes, editor and admin burn out is - so it needs to be made clear that the most efficient way of handling tiresome issues is clarification on what needs to be in place to ensure consensus is properly arrived at. It is under that reasoning and that reasoning only that I have sought to extend the meaning of "silence" in a type of inappropriate conduct in deletion discussions to include the non notification of some parties by choice or habit.
I take User:BrownHairedGirl's point about transparency, in how I became aware of a potential issue, but am unsure if it is germane to the broader point. Perhaps, as an example of giving proper and full notification of all matters potentially effecting a discussion, then I should have. It is, after all what I hope to achieve by opening this discussion here. LessHeard vanU (talk) 15:17, 18 February 2017 (UTC)
LHVU, I am sorry to say that your reply comes across as thoroughly disingenuous. As a former admin, you will undoubtedly be aware that speedy deletions are based on a long-standing policy at WP:CSD that a consensus-forming discussion is not required before the speedy deletion of a page which meets one or more of certain specified criteria. In fact, I just checked your own deletion log[2], where I see close to 1000 deletions over a period of about 4 years, nearly all of them speedies. So it's very clear that you are well-aware of the distinction, and well-aware that speedy deletions exist precisely to allow minimal process for uncontroversial actions.
Your attempt to conflate the question of canvassing with the question of under-notification is tenuous enough, but this latest attempt to tie speedy deletions into concerns about consensus-forming is utterly mischievious -- because the whole point of speedy deletion is that per the policy's nutshell under certain limited conditions, a page may be deleted by an administrator without waiting for any discussion. WP:CANVASS is entirely about discussions, and the WP:CSD is about actions without discussion, so trying to tie the two together is plain silly.
You didn't mention speedy deletions when you posted on VegaDark's talk page[3], and you didn't mention speedy deletions when you first posted on this page[4]. You raised it now only because after those two posts because it was after them that I resumed[5] a discussion with VegaDark about some nuances of the speedy process. You would of course have been welcome to join in that discussion, but your attempt to conflate that with CFD notifications has all the hallmarks of post-facto muddying-of-waters.
So, leaving aside the red herring of speedies, we return to your original point here about CFD notifications. I hope that you can agree that when considering notifications, there is a balance to be drawn between a whole range of factors. The desirability of bringing more people into a discussion is balanced against the intrusiveness of excessive notification, the risks of vote-stacking, and the burden on editors launching a discussion. You are briefly dismissive of the burden on editors proposing actions, saying that time is not a limiting factor. Actually, no. There is no deadline, but time most certainly is a limiting factor, and if the sort of burdens you seek were ever imposed at CFD, most of its works would simply stop. It wouldn't be delayed or postponed, because there is no reason to expect that any editor would devote the utterly huge amount of time required for the sort of notification you seek.
Now, if there was some sort of systemic problem here, then you would have a point in suggesting that the balance should be adjusted. But the crucial fact of all this is that (apart from the speedy deletion red herring) your grounds for concern are based solely on the thoroughly exceptional case of a category relating to you personally, a conflict of interest which you failed to disclose. You have offered no evidence that there is any wider problem of under-notification which would justify considering some sort of change. If you do have some such evidence, then please present it ... but I will be very surprised if you can find anything other than very rare cases. I closed dozens of CFDs each year, sometimes hundreds, and it is exceptionally rare to find any sort of post-closure complaint based on lack of notification. You can look at the archives of WT:CFD and find next to nothing on the issue, and similarly with ANI.
Ironically, the only ANI discussion I can recall on this was last month, when another admin tried ramming through some speedy renamings of untaggedd categories, and ignored objections to the lack of notification. When I reverted, he got exceptionally irate and started a multi-venue shitstorm, but the discussion ended up back at back at CFD. Despite the very high profile of that brouhaha, nobody has sought any notification of that discussion other than the tagging of the categories.
So all of this is a non-issue, an extrapolation of the exceptional circumstances of LHVU's personal category into an counter-factual claim of a wider problem. Naughty. -BrownHairedGirl (talk) • (contribs) 21:46, 18 February 2017 (UTC)
  • I mostly agree with everything User:BrownHairedGirl had to say here. I think if this type of non-automated notification were expected prior to a nomination, virtually all XfDs would grind to a halt. Nobody would bother working on the category system when it could potentially take up hours of volunteer time just to make sure all those notifications were made. With the focus on having a "balanced namepace" in your edit history as of late with WP:RFA, certainly no administrator hopefuls would ever venture into XfD else they would need 100 user talk edits for every semi populated category nomination. I can't imagine a single CfD regular bothering making nominations on anything but incredibly new and easy to notify categories - no changes to existing structure would ever be done. Also I can't imagine the nightmare of notifications any semi regular category gnomes would get if every category they edited triggered a CfD notification (although if nominations grinded to a halt I guess this wouldn't be an issue). To implement this sort of notification requirement would be so overly burdensome that literally nobody would deal with category work anymore. The same argument goes for user categories in particular - notifying all members of a category would be so overly burdensome you would effectively stop all user category nominations. VegaDark (talk) 04:51, 23 February 2017 (UTC)
  • The idea that user categories (which is what are pertinent here) can be deleted without even notifying the editors in those category is unreasonable in the extreme. The idea that tagging the category pages is sufficient notification is absurd. VegaDark asserts that the notification burden is an unreasonable burden might be arguable if the only means of notification was manual editing of every editor's user talk page, but it isn't. There are means of having a bot notify every editor on a list, and a bot to automatically notify all editors in a specified category would not be difficult to create. The reason to avoid notifying editors in a category you want to delete is because you know it will be contentious, and choosing not to notify for this reason is both a deliberate attempt to skew the debate and profoundly disrespectful to the editors you seek to avoid including. Anyone working in user category maintenance should have the decency to engage with the editors effected by the actions you propose. LessHeard vanU is correct, in my view, what has been occurring is unacceptable attempts to avoid engagement with key stakeholders in general in user category discussion, and the fact that the specific example cited is a category related to him is irrelevant. The editors in user categories are people and deserve to be treated as such. Some categories will need deleting, some even speedily, and policy must guide the process, but the fact that maintenance is easier if the people effected don't know until after decisions are taken does not justify their exclusion. "Consensus" by omission of interested parties lacks legitimacy and integrity. It should not be necessary to have a rule to ensure ethical / principled operations of user CfDs, but if it is, the so be it. EdChem (talk) 13:29, 25 February 2017 (UTC)
  • @EdChem: To be clear, you are suggesting that our failure to go above and beyond what policy requires (and on top of that, in my view an act that would be detrimental to the interests of the encyclopedia), by failing to use a hypothetical bot that could be created to notify individual users, is "Unreasonable in the extreme?" Sorry, but notifying individual users in user categories, even if it were as easy as an automated bot (which it currently is not) is an activity that would make the encyclopedia worse, because notifying such users is not reflective of consensus. People have a tendency to want to keep things they have self-selected to be placed on their userpage, wouldn't you agree? How is reaching out to a subgroup of people more inclined to support keeping a better measure of consensus over random, disinterested (not to mention generally more policy-aware) people who happen to come across the CfD?. As I've previously stated, notifying members of a category is like notifying NRA members that a gun restriction is about to be passed. This is based off of over a decade of dealing with user categories and personal experience - often such notification results in ILIKEIT votes not based in policy regardless of the reasons for deletion. Unlike AfD, CfD has lower participation. You can't rely on many neutral people coming in and giving their .02, particularly when it comes to user categories which many users expressly avoid caring about. If you are worried about notification, I liked SmokeyJoe's proposed solution which would assign every user category to a Wikiproject, and notification could go to that Wikiproject's talk page. This would, in my view, accomplish concerns regarding lack of notification while simultaneously not reaching out to specific users likely to try to sway discussion a particular way. VegaDark (talk) 01:26, 26 February 2017 (UTC)
  • @VegaDark: Perhaps you would like to go and review your own comments from debates in 2007/8 about LGBT categories, and reflect on your own comments, then re-read what I have said. In the meantime, I did not suggest just a hypothetical bot, I stated that "[t]here are means of having a bot notify every editor on a list." A system like is used by ArbCom for each case, which the Signpost uses for deliveries, or like the admin newsletter uses. This is not something novel on WP, nor is it difficult to implement. That policy does not require notification of users in user categories you seek to delete is a flaw in policy, one likely resulting from the group of editors working on category maintenance is small, like-minded, and rarely confronted by outside editors on matters of principle. So, yes, not notifying is unreasonable in the extreme. Choosing to not notify the group most directly by the proposed decision regarding a user category to which they have taken a deliberate decision to join is indefensible. Your view that these editors might be biased is akin to saying you don't like what they might say. I wonder if you've ever considered that, since user categories are supposed to have a collaborative purpose, the editors in the category are the ones most likely to have direct knowledge of its collaborative value? Your view comes across as that your opinion is correct and those who value something and disagree with you must be wrong or biased. You state that neutral people don't come to CfDs, but presume that you and the regulars at CfD are neutral and unbiased, and the "not to mention more generally more policy-aware" comment underlines the "we're right, outsiders are wrong" attitude. I don't doubt that there are a lot of categories that need removal, and that category maintenance is necessary, and even that it is thankless and under-recognised work. But when it comes to user categories, groups of editors who are also people, there have been serious mistakes and harm done, and the idea that the problem can be resolved by trying to remove categories without telling the editors in them is offensive. And, I believe that a properly constructed RfC on this point, held at a widely read page and well publicised, would show a community consensus to that effect. EdChem (talk) 12:53, 27 February 2017 (UTC)

"Canvassing" re Donald Trump

Hi, could I ask about how WP:CANVAS applies to this thread? Is it bad to link to the discussion at Donald Trump? Should I only discuss this general issue at a specific example of that issue Siuenti (talk) 19:30, 2 April 2017 (UTC)

Notice of Discussion

Hopefully I am not violating this policy by this notice, please revert if you feel like I am. This policy and the specifics herein are being discussed at Wikipedia:Village pump (policy)#WP:CANVASS and input there from the editors that have watchlisted this policy could be helpful. Endercase (talk) 17:20, 19 April 2017 (UTC)

Canvassing?

Is this a sign of canvassing?

The out-of-the-blue editor admits that the other editor, who he is supporting, contacted him soliciting his input in the discussion. By his talk page, this seems to be a common practice of him inviting him, and the other editor then supporting him. (See the "Hey Friend" message regarding Fairview Alpha, Louisiana here, and other similar ones here).

And the two editors who are agreeing with each other never mentioned this in the talk page discussion, till the issue was raised by the editor with whom they disagree.

But he says he "I don't see how this is inappropriate." Because he says he is one of the most active editors on "political geography articles in Michigan" (though not on the subject of the discussion -- inclusion of notable people in lists of people).

If this is canvassing, what does one do, where in addition to the two disagreeing original editors we only have the editor that joined out of the blue, supporting his wp "Friend," who contacted him, leading him to join the discussion? --2604:2000:E016:A700:79D7:9430:713A:70E2 (talk) 01:22, 27 April 2017 (UTC)

When you discuss another editor's behavior on a "WP" page you are required to notify then in writing at their talk page. I explained my actions at Talk:Plymouth, Michigan and also told you what you could do to garner wider participation in the discussion. And I am one of the most active editors on political geography articles in Michigan. Political geography (also known as social geography), broadly defined, is articles about things man has imposed on the landscape, such as (but not limited to) roads, schools, settlements, other political divisions (such as counties, townships and states) and buildings. Except roads and buildings, all articles on these subjects have either notable people sections or associated freestanding list articles for them. So yes, I am quite familiar with them. Your attitude shows a remarkable lack of good faith, bordering on WP:OWN. If you wish to make a complaint about editor behavior, the proper place is WP:ANI. I would not suggest going there tho. Just lose the chip on your shoulder and work the conflict through. Believe it or not, the editorial control on Wikipedia is driven on conflict by design. John from Idegon (talk) 03:26, 27 April 2017 (UTC)
I don't think the question is if it is Canvassing or not. I think it should be "Was it inappropriate canvassing?" Endercase (talk) 19:42, 28 April 2017 (UTC)

Method of delivery

Regarding Do not use a bot to send messages to multiple pages, is the "how" of the delivery something that still matters? Note, I am not in any way talking about the content of the delivery - but assuming the content is acceptable and the audience is acceptable - using the system bot via Special:MassMessage should be preferable to individually sending talk messages to an list if the acceptable audience is in anyway sizable. Thoughts? — xaosflux Talk 03:36, 23 December 2016 (UTC)

That specific guidance needs to go. It might make sense to dedicate a paragraph or so to "notification by (semi-)automated methods" in its place. --Izno (talk) 10:04, 23 December 2016 (UTC)
Unless I'm missing something, that sentence only means sending a one-sided message that way is inappropriate. In other words, don't mass message something like "please vote support for xyz". There's certainly no problem with using a bot for neutral messaging, and the sentence probably should be clarified in that regard. Maybe: Do not use a bot to send non-neutral messages to multiple pages. --Tryptofish (talk) 00:52, 24 December 2016 (UTC)
The guideline already has "appropriate" and "inappropriate" sections speaking to the content and the audience ("non-neutral messages to multiple pages" are inappropriate regardless of how they are delivered), I have no desire to change that - just that the "how" of the delivery doesn't belong (and if a sizeable audience is appropriate - then using the Mass Message function is likely the best way to deliver the message from a technical point of view). — xaosflux Talk 02:25, 24 December 2016 (UTC)
The only way I can read the current text is "Do not use a bot to send messages to multiple pages" like, ever, for any reason. Super clear: if you want to send the kind of messages we're talking about here, you can't use a bot to do it. If the intent is to say "do not use bots to send inappropriate messages" or if the intent is "using bots to send appropriate messages is fine", we have a massive failure of communication. Since, you know, the text says the complete opposite of that, in every way I can interpret it. CapnZapp (talk) 23:17, 25 January 2017 (UTC)
Basically, if it is appropriate to copy-n-paste the same message en masse (say with dozens or hundreds of browser tabs) then using MassMessage should be just as appropriate. — xaosflux Talk 02:41, 2 May 2017 (UTC)

Issue with lead wording, and possible improvements

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.


Many editors (recent example here) use the term "canvassing" in a way that clearly leaves the implication that canvassing is wrong. They often use the term "canvass" (a broad term coving both acceptable and unacceptable actions) as a negative, or they use "WP:CANVASS" as itf is only covers the unacceptable actions.

That is not what this guideline says. Canvassing is an action that can be done properly or can be done improperly. This guideline specifically notes that contact of other editors may be perfectly acceptable and then goes on to identify cases where it is not acceptable.

Is my opinion that the wording should be tightened up in this guideline, as it is perfectly possible for someone to read this casually and come away with the impression that this is a guideline about canvassing which is always inappropriate.

I believe the problem arises from this key sentence:

However, canvassing which is done with the intention of influencing the outcome of a discussion in a particular way is considered inappropriate.

The construction of the sentence sounds like it could be defining canvassing as something that is done with the intention of influencing the outcome of discussion in a particular way. But that's not what it's trying to say. It is trying to distinguish acceptable canvassing from unacceptable canvassing.

The subsequent sections are fine. One section identifies ways of informing that are considered appropriate. Some of these actions could be considered canvassing but would be considered acceptable canvassing. A subsequent section talks about inappropriate notification. Those examples are typically canvassing and would be considered unacceptable canvassing. Fixing the wording should be relatively simple, subject to the usual discussions of exactly what wording is best. I also think we should consider whether this guideline should be retitled as "inappropriate canvassing". I fear that the ubiquitous history of the use of wp:Canvassing and WP:CANVASS may make it difficult to rename this guideline.

Proposed wording

Replace:

With

Title of guideline

I hope someone with more knowledge of the technical implications of change in the name will weigh in regarding whether this is feasible or not worth the effort.--S Philbrick(Talk) 19:06, 7 May 2017 (UTC)

Discussion

I feel that in the case you mentioned providing examples of comments on an ongoing discussion is bad canvassing. As long as the message is neutral then it should remain okay. I cant speak about project notifications though. Would it be acceptable for example to inform the Liberal projects about an ongoing discussion related to Donald Trump? Would it be okay to notify WikiProject Palestine about an ongoing discussion about Israel? The problem is that even if an editor knows that a project might be interested it can still be seen as biased. - Knowledgekid87 (talk) 16:14, 8 May 2017 (UTC)

I think it would be better to adjust and combine the first paragraph with your proposed changes, to avoid repetition. Perhaps something like this:

Notifying other editors of ongoing discussions, also known as canvassing, is acceptable when done with the intent to improve the quality of a discussion by broadening participation to more fully achieve consensus. It is inappropriate to selectively notify editors with the intention of influencing the outcome towards a specific bias, subverting the normal consensus process.

To be honest, though, I think there generally is a promotional connotation to the word canvass, and I expect people will continue to use it in a negative sense, regardless of any changes made. isaacl (talk) 20:48, 8 May 2017 (UTC)

  • Oppose - There is appropriate notification and inappropriate notification. Appropriate notification is not canvassing, while inappropriate notification is canvassing. That's how I've always viewed it at least, but I could be incorrect on the semantics, so I'll be interested to read others opinions on the matter. — Godsy (TALKCONT) 00:07, 9 May 2017 (UTC)
  • Strongest possible oppose - "Canvassing" IRL does not have a negative implication, but in the Wikipedia sense, it is quite simply the disruptive behavior of soliciting input in order to artificially create a consensus, or lack thereof, that would not otherwise exist. The whole point of WP:CANVASS is to identify this behavior as being distinct from simply notifying people. It is extremely important that the distinction is made, as we are almost wholly governed by consensus, and consensus building is impossible if canvassing is going on. Changing the term "canvassing" from a specific, negative behavior to a broad general term that may be considered good or bad would create unnecessary confusion, muddle the distinction between what is acceptable behavior and what is not, deprive the community of the clear, longstanding label for what is a severely destructive behavior, and provide a free strawman defense for those accused of canvassing (I can already see canvassers citing this proposed wording as a defense). Canvassing has a specific meaning. De-fanging the use of the word in order to remove the negative connotation would not provide any benefit to the project. We do not need users accused of canvassing making the argument that there's "nothing wrong with canvassing". That's not the point of this policy. The point is to identify a specific, inherently destructive behavior which we call "canvassing". Strongly recommend we retain the longstanding position of this policy and do not soften our stance on what "canvassing" is. Swarm 19:56, 10 May 2017 (UTC)
  • Support Isaacl's version Because "canvassing" does have these different senses – soliciting opinion, selectively soliciting opinion, trying to influence opinion – it's better not to rely on this word to convey what we mean. The summary simply needs to get across that there are appropriate and inappropriate ways of going about notification. The rest of the page sets out in detail what these ways are: Noyster (talk), 21:14, 11 May 2017 (UTC)
  • You're implying that the word "canvassing" doesn't refer to something specific on Wikipedia, when in reality, it does and has. If you want to be unnecessarily pedantic, you can claim some sort of false confusion as to how the word "canvassing" can be defined, but the current definition as a negative behavior is a sufficient answer to any such confusion (which doesn't actually seem to be a problem). I can't even comprehend how changing the policy to say that canvassing can be perfectly fine or absolutely horrible would be easier to understand for anybody. We would literally be removing a longstanding term for a very serious disruptive behavior and that would have the effect of deemphasizing just how serious "bad faith notification" is. Christ, maybe it's just because I'm an administrator but I can already envision someone accused of canvassing referencing that same, pedantic, empty argument that "there's nothing wrong with canvassing". Swarm 19:31, 12 May 2017 (UTC)
  • Oppose 'Notifying' when done according to neutral criteria, is quite distinct from 'canvassing' and the proposed change simply 'muddies the water'. Even in the real world, a common meaning of 'canvassing' is contacting with the intent to persuade. I support all the arguments made by Swarm above. Pincrete (talk) 15:46, 21 May 2017 (UTC)
  • Support some change, but not the one proposed. In the real world, canvassing can just mean finding out how people feel about an issue, so it would be helpful to clarify this. [6] I wouldn't oppose adding something like "While notifying editors in accordance with this guideline may meet some dictionary definitions of canvassing, on Wikipedia cavassing generally refers to notifying editors with the intention of...". This clarifies the use of the term "canvassing" on Wikipedia and avoids calling appropriately notifying editors canvassing, which is a generally negative phrase here. --- Patar knight - chat/contributions 20:18, 26 May 2017 (UTC)
  • oppose Swarm nailed it. Providing notifications in a neutral and non-selective way is fine; "canvassing" is a technical term in WP that means providing notifications in a non-neutral and selective way. Jytdog (talk) 05:17, 30 May 2017 (UTC)
  • Comment - Agree that the Wikipedia sense of "canvassing" is and should remain negative. But the existing sentence muddies the message by suggesting that some canvassing is not "done with the intention of influencing the outcome of a discussion in a particular way". It's like saying "However, murder which is done with the intent of killing the victim is against the law." The word "canvassing" should be replaced by "notification" in that sentence.
    Re Patar knight's comments, there is no need to explicitly state that Wikipedia's definition is different from that of the dictionary. Every editor eventually learns that Wikipedia has its own definitions for many such terms, notably WP:Notability, and they learn that they need to read past the title of a policy or guideline. The title alone is fairly meaningless and we often get into trouble when we try to apply the dictionary definition. WP:Canvassing might as well be WP:Guideline 217. ―Mandruss  22:36, 12 June 2017 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Followup

Snuge purveyor, regarding this, we should perhaps change "How to respond to inappropriate canvassing" to "How to respond to canvassing" since the former title suggests that there are appropriate forms of canvassing. For the record, like others, I always considered the guideline as stating that canvassing may be appropriate or inappropriate. As for your edit, I guess it solves the above dispute, although I do wonder about "notification" being interpreted as applying to WP:Pings. Whether or not WP:Pinging should be considered inappropriate canvassing has been discussed before. It has not yet been added because it can be complicated. I see such pinging all the time at WP:ANI, for example, and the pings are usually used by editors (administrators and non-administrators) for support on whatever they are arguing. Flyer22 Reborn (talk) 07:39, 6 July 2017 (UTC)

Flyer22 Reborn, thank you for the followup. I changed the section heading, which I should have caught earlier. I too had previously considered the term canvassing to cover both appropriate and inappropriate actions, but my reading of consensus shows the above discussion reaching a different conclusion.
It's my understanding that this discussion grew from the "are pings canvassing?" question which came up at Dane's RfA, but the RfC was not phrased to address that topic and none of the participating editors mentioned pings. I agree that most of the time when I see editors in a dispute pinging other editors, it usually seems to constitute canvassing, and would support adding a bit about inappropriate use of pings to the guideline. Maybe we could change the "Vote-stacking" bullet point from "Posting messages to users selected based on their known opinions…" to "Pinging or posting messages to users…". We could also expand the "Votestacking" subsection to include something about appropriate use of pings, perhaps in the final sentence about "appropriate notification". Snuge purveyor (talk) 09:27, 6 July 2017 (UTC)
Snuge purveyor, no, I wasn't aware of that RfA. When it comes to whether or not pinging is canvassing, I mean Wikipedia talk:Canvassing/Archive 5#Would this be canvassing? and Wikipedia talk:Canvassing/Archive 5#Pinging?. Personally, because of those discussions and the grey area mentioned regarding pinging, I would be against adding pinging to the guideline. Before such an addition, I would prefer a big discussion on the matter since adding pinging will affect Wikipedians on a mass scale. Flyer22 Reborn (talk) 09:43, 6 July 2017 (UTC)
Reading Wikipedia talk:Requests for adminship/Dane, it's even clearer to me that adding pinging would be contentious. The addition would definitely need substantial discussion. Flyer22 Reborn (talk) 10:00, 6 July 2017 (UTC)
Flyer22 Reborn, thanks for pointing out those discussions. That's important background for this question. I believe there are cases where pinging constitutes canvassing (like pinging a half-dozen known-friendly admins to review your unblock request), but it's usually a greyer area and definitely seems contentious. Agree we should tread lightly and gain consensus before addressing pings in this guideline. Snuge purveyor (talk) 17:44, 6 July 2017 (UTC)

Campaigning

Hi all,

Can someone identify when "campaigning" was added to this page and where the discussion was that indicated it was consensus to have this short section included? I think it is a rather peculiar point to have on this page. People should try to sway people with wording, argumentation, and opinions.

jps (talk) 21:37, 6 August 2017 (UTC)

[7], 11 years ago. So it is long standing. By the way, people should try to use wording, argumentation, and opinions in debates. That is not the same as posting a message designed to drum up support by using language deigned to encourage a certain type of person to edit or join a discussion.Slatersteven (talk) 22:39, 6 August 2017 (UTC)
The old wording seemed to apply specifically to user talk pages. Present wording is not so clear. Can you figure out when that change occurred?


jps (talk) 00:53, 7 August 2017 (UTC)

Question regarding IRC

Would using an IRC channel (like #wikipedia-en) to encourage participation in something, like an AfD, be considered canvassing? Note that I am not talking about privately messaging specific users or pinging specific users, but just a general notice. RileyBugz会話投稿記録 19:26, 23 September 2017 (UTC)

Appropriate and inappropriate canvassing

Jc37, regarding this edit you made, there was a discussion about that wording. Sphilbrick, Knowledgekid87, Isaacl, Godsy, Swarm, Pincrete, Patar knight, Jytdog, and Mandruss weighed in. Snuge purveyor closed the thread. I commented in a "Followup" subsection after that. Considering that you reverted Snuge purveyor, you must have seen Snuge purveyor edit's summary about consensus on the talk page. Flyer22 Reborn (talk) 04:35, 28 October 2017 (UTC)

I agree with Flyer's analysis that the conclusion of that RfC was that such changes of the text were counter-productive. 'Canvassing' is distinct from apt 'notifying' in WP-speak (I personally think they are different in the real-world too) . Pincrete (talk) 08:52, 28 October 2017 (UTC)

Abuse of language

IMO it's just wrong to use canvass - without qualification - to mean something negative. Just today I wrote on a talk page that I want to "canvass opinion" before completing a significant change that I'm proposing. Had I known of the discussion earlier this year I would certainly have weighed in!

Why should there be any WP-speak, as Pincrete calls it, for any but the most technical of Wikipedia concepts? We owe it to each other to speak as clearly as possible, and to the purpose. Abusing the English language in this way just adds to the confusion surrounding the arcane mysteries of editing Wikipedia. We're not a "walled garden" and shouldn't be creating a priesthood of cognoscenti – let's leave that to the academicsDisclaimer: I was one … and other politicians ;-)

Canvassing opinion is not wrong. Vote stacking is. yoyo (talk) 01:34, 20 November 2017 (UTC)

Nobody objects to you using a word in a way that makes sense to you on talk pages. In a policy it makes sense to minimise ambiguity, if I canvass on behalf of a politician in the RW, am I not hoping to persuade, as well as to inform and enquire? Words don't have 'special' esoteric meanings on WP, they do (necessarily IMO) sometimes shed the ambiguities that exist in the RW. "Canvassing opinion", "selectively notifying" ONLY those who are likely to agree, "making enquiries" to a selective few are precisely the means by which "vote stacking" is achieved. Why introduce ambiguity here? Pincrete (talk) 10:53, 20 November 2017 (UTC)
Canvass, sense 3: "to go through (a district) or go to (persons) in order to solicit orders or political support or to determine opinions or sentiments" (my emphasis). So our usage is correct per one sense (or sub-sense) of the word, according to Merriam and Webster, and this is hardly a case of Wikipedia language distortion. ―Mandruss  12:17, 20 November 2017 (UTC)
The way we use words on Wikipedia talk pages matters because, for example, one editor might think that an essay is a policy or a guideline, or that someone has violated some policy or guideline. And, of course, there is the WP:Civil aspect of things. I used to dislike when editors would state "canvassed" to accuse people of wrongdoing when what the people actually did was appropriate notification. In those cases, I would state that the guideline was not violated and/or that it was appropriate canvassing. And as we see now, the guideline has made it clear that "canvassing" specifically refers to inappropriate notification. Flyer22 Reborn (talk) 20:54, 20 November 2017 (UTC)

Groups

Can I organize a special group of Wikipedians that would like to help as a team? GermanGamer77 21:19, 18 December 2017 (UTC)

GermanGamer77, I don't see why not. Flyer22 Reborn (talk) 21:41, 18 December 2017 (UTC)

Wikiprojects

Can the list of appropriate notifications be expanded to include WikiProjects?

(I first became aware of the Canvassing policy when I posted neutral non-partisan notices to a WikiProject saying a discussion was underway at AfD, and that, I was told, was a violation of the Canvassing policy. My accuser then engaged in extensive partisan canvassing, and one of his recruits even added to the AfD discussion a dishonest statement to the effect that I had contacted people who I knew would agree with me on the AfD discussion. How there could be any such people, I still don't know.) Michael Hardy (talk) 15:14, 28 December 2017 (UTC)

@Michael Hardy: ... It's the first bullet? --Izno (talk) 15:21, 28 December 2017 (UTC)
@Izno: oh. That explains why I didn't notice it. Michael Hardy (talk) 16:59, 28 December 2017 (UTC)

Notification of RFC on proposed amendment to Article Rescue Squadron guidelines

There is currently a discussion to amend the usage guidelines for Wikipedia:Article Rescue Squadron – Rescue list located here. Hijiri 88 (やや) 00:12, 10 February 2018 (UTC)

Policy

Would it make sense to promote this guideline page to be a policy? (This is a question, not a formal proposal.) I'm surprised that it isn't a policy already, since it is widely treated as though it were. --Tryptofish (talk) 00:45, 13 March 2018 (UTC)

Not sure. Flyer22 Reborn (talk) 00:42, 14 March 2018 (UTC)
As someone who understands and sympathizes with the purpose behind the guideline, but also believes it to be woefully misguided and counter-productive, I would certainly not endorse elevating it further. Unschool 07:28, 14 March 2018 (UTC)
No, not until we find a more acceptable name for it: two discussions within the past year show substantial disagreement over whether the term "canvassing" in itself denotes a sin: Noyster (talk), 10:26, 14 March 2018 (UTC)
OK, thanks everyone. That's why I asked. --Tryptofish (talk) 18:21, 14 March 2018 (UTC)
Yeah, I thought about those discussions before replying to you above. People can be sanctioned for canvassing whether it's a guideline or policy, though. Flyer22 Reborn (talk) 20:27, 14 March 2018 (UTC)
Thanks! --Tryptofish (talk) 23:18, 14 March 2018 (UTC)

Canvassees' responsibilities

I wonder if the guideline needs a couple of sentences to guide editors in how best to proceed, if they feel they have been the target of inappropriate notification about a discussion (either expressly, or through the ignorance of a new editor).

I notice that the guideline doesn't currently mention anything about the responsibilities of someone who is the target of possible canvassing. My guess is that this is by design, but I just wanted to throw it out there. Clearly, there's no sin in being the target of it, because if there's a sin involved, you're the "victim" so to speak, and not the agent in any inappropriate notification.

However, I wondered if someone canvassed me about something, whether or not I would feel duty-bound *not* to go !vote on some subject under discussion. My first reaction, was that I should stay away in an abundance of caution, to avoid any possible misinterpretation of my own action, or to avoid tainting the voting.

On reflection, however, I think an editor should feel free to !vote even if they suspect they are canvassed, because the contrary leads to an absurdity: if canvassed votes were prohibited a priori, that could lead to "reverse votestacking" by a partisan individually notifying opponents in order to disqualify them, while notifying projects neutrally so their supporters see it, thus tilting the balance their way. That would be an absurd side-effect of prohibiting canvassed votes; that may be one reason why there isn't a statement in the guideline about this now.

But, not saying anything about it at all, might not be ideal for NPOV, either. Because I would still feel funny about !voteing, if I felt canvassed, and might well avoid it anyway. So how to handle this? I think a "best practices" move here, would be that an editor who feels canvassed should vote, if they would have voted had they been notified appropriately, and that they should self-tag their vote with {{canvassed}}. This alerts the closer to the situation, who can then decide what to do with such votes. As this is a bit subtle, in order to make this clear to editors who want to do the right thing, we might need to say something explicit about this in the guideline. So, my attempt:

If you feel you may have been inappropriately canvassed to participate in a discussion, you are neither obliged to vote, nor to avoid voting because of it. Consider whether you would have voted, had you received the notification in an appropriate way. It is the responsibility of the closer to evaluate the consensus of a discussion, and you can help by including the {{canvassed}} template with your vote, and then leave it up to them.

Any thoughts on this? Mathglot (talk) 00:33, 5 June 2018 (UTC)

Consultation request

Perhaps this belongs at Wikipedia:Requests for comment, but since it concerns alleged canvassing on my part, I will first post here. The exchange at Wikipedia talk:WikiProject Ukraine/Archive 5#New Kiev/Kyiv naming discussion, which was further developed near the bottom of the discussion at Talk:Kiev/naming, suggested that my posting at Talk:WikiProject Ukraine the sentence, "Editors who may wish to participate, can access the discussion at the page specially designed for that topic, Talk:Kiev/naming" constituted canvassing.

Are we to assume that only like-minded editors keep Wikipedia talk:WikiProject Ukraine on their watchlists, while everyone else ignores that talk page. It was certainly noticed immediately by an opponent of the proposal and it was possible or even likely that the announcement I posted would have attracted more votes against my position than in favor of it. The venue itself, however, is the precise spot where it would seem such an announcement should be posted in the same manner that Wikipedia talk:WikiProject Film is the venue to post announcements about ongoing film discussions.

A complaint might have been legitimately lodged if I picked through Category:Ukrainian Wikipedians and posted on the talk pages of individual members but, as I indicated at the time, this was my only announcement regarding the discussion in question. Does anyone else feel that such an announcement constitutes canvassing?    Roman Spinner (talkcontribs) 13:27, 5 July 2018 (UTC)

Intent

GreenMeansGo, I can see why you made this edit, but notification can influence a discussion in a particular way even when the notification is neutrally worded. Canvassing is more about the intent. I'm trying to think of unintentional canvassing. Examples? Flyer22 Reborn (talk) 17:05, 10 July 2018 (UTC)

Well there's certainly a robust discussion going on here currently. I was actually mostly working on improving c:COM:CANVASSING, the absence of which was highlighted in January at c:Commons:Administrators/Requests/Yug. In both cases, the intent of the user and whether that matters if it has an effect regardless seems to be a topic of debate. For my own part, even if a user has every intent of leaving a neutral message, but how or where the message is left has the effect of inappropriately introducing bias through non-neutral recruitment, is still probably relevant to our guidance, and that's probably especially relevant for newer users and for off wiki discussion. It's not terribly uncommon for someone to simply complain off-wiki with no direct appeal for someone to involve themselves, but we get an AfD riddled with SPA tags. GMGtalk 17:17, 10 July 2018 (UTC)  
I know that people can have different views on what canvassing is, but canvassing is usually interpreted as being intentional. Even if a newbie does not know of our canvassing guideline, if they leave a message or a series of messages to bias an argument in their favor, that is trying to influence the matter in their favor. As for the SPAs, in my opinion, they should be tagged. Complaining about something that happened on Wikipedia in a forum off Wikipedia is likely to bring in one or more editors from that forum to assist the one who complained; that's common sense. The complaint might not have been to get WP:Meatpuppets, but the newbies are SPAs and meatpuppets nonetheless.
Anyway, what do others think of GreenMeansGo's wording change? This is the latest one. Swarm and Nyttend, any thoughts? Flyer22 Reborn (talk) 18:10, 10 July 2018 (UTC)
Definitely not saying that the SPAs shouldn't be tagged or anything like that. Just saying that I didn't canvass. I just complained about Wikipedia deleting my article to my 2 million twitter followers. isn't really a defense. Still pretty much going to have the same effect as saying Dear twitter, go vote keep. So it's still canvassing.
Or something like this ANI, which tl;dr, was pretty much I didn't canvass, I was just seeking outside opinion. Yeah, but if you leave pretty non-neutral notifications, it's not gonna be out of line to call it canvassing. In neither case would people really be willing to accept lack of explicit intent as a defense.
Anyway, open to other opinions. GMGtalk 18:26, 10 July 2018 (UTC)
The wording seems a bit redundant; if it's done fairly to everyone (including opposers, and including people who don't want to be notified trivially), it won't be a inappropriate. It doesn't change the meaning in a problematic manner, so I wouldn't mount some strong opposition, but I oppose it because it's not necessary. Nyttend (talk) 20:11, 10 July 2018 (UTC)
This thread caught my eye because of what is going on at Commons, and I see that it was indeed suggested by that. I think it would be a mistake to change guidelines here unless there is actually a problem here, because the culture at Commons is so different. --Tryptofish (talk) 20:19, 10 July 2018 (UTC)
Fair enough. Reverted. GMGtalk 20:20, 10 July 2018 (UTC)
The "Dear twitter, go vote keep." aspect is a good point, and funny. Flyer22 Reborn (talk) 05:15, 11 July 2018 (UTC)
And it reminds me of this recent move discussion. I and others argued that the Twitter posting was canvassing. Flyer22 Reborn (talk) 05:19, 11 July 2018 (UTC)
  • GreenMeansGo The term is often misunderstood and often misused, but there is no "debate" or lack of clarity here. Canvassing is simply the practice of attempting to artificially fabricate a consensus (or lack thereof) by introducing bias, using deceptive notification tactics of any kind. It does not matter if it's a neutral notification to a biased user, or a biased notification to a neutral user, or a completely esoteric way of notifying someone; canvassing is no specific "event" or "situation", it's the attempt at introducing biased participation. It's an intentionally malicious and abusive act. Even the first rendition of this guideline from 2006 defines it as an "attempt to influence the outcome". There's no such thing as "unintentional canvassing" because intent is the whole point. Canvassing prohibits the specific act of attempting to game the consensus-building process, by which we are governed. If there is a "debate" at Commons, that means people are trying to redefine the term for their project's purposes. That's that community's prerogative, but it's pretty simple on this side. Swarm 02:27, 13 July 2018 (UTC)
Umm...I don't draw any real conceptual borders between projects. Most people are collaborative, even when you can barely communicate. Most of the differences are fairly far on the periphery and primarily content related. The initial reference to Commons was more of why I happened to be thinking about this at the moment and looking at this policy as a reference, especially as the definitive version of guidance on Commons has to start in English. (For that matter, on some projects fully half the policies and guidelines are just links to en.wiki.)
As to whether unintentional canvassing is a thing, I simply disagree per above, with probably most of the instances being generic off-wiki complaining, and on-wiki genuine but failed attempts at neutral notification. I suppose if you define all swans as white and prefer the definition over the circumstances, then you are correct, but probably not in a way that is deeply meaningful. The only harm I expect the current framing does is setting users up for assumptions of bad faith: you canvassed, therefore you must have done so intentionally. I don't think that's particularly helpful, but if there is general opposition to any change in wording, then the wording won't change, and that's that. GMGtalk 15:31, 15 July 2018 (UTC)
@GreenMeansGo: Semantical musings aside, I'm simply explaining to you what the term "canvassing" means so that you can understand why your edit was not the superficial correction you intended it to be. The assumption of bad faith you give as an example doesn't make sense: canvassing is specifically a malicious, bad faith act. An accusation of canvassing is, and always has been, an inherent accusation of bad faith. Yes, "intent" is an intangible concept and there is the theoretical possibility of a grey area, but I doubt it's the problem you're making it out to be. Perhaps you could give me a hypothetical example about an "unintentional canvassing" scenario that wouldn't be covered by the present definition? Swarm 07:16, 20 July 2018 (UTC)
I mean, I understand the difference between the two definitions: intent vs. effect. What I've been arguing is that the community applies the second definition when there is effect without intent. There have already been given a few examples above, both real and hypothetical. But it doesn't lend any clarity to the discussion to reify canvassing as if it is a thing that exists apart from its definition. It exists however it is defined. The question is whether there is a meaningful difference between the definition de jure and the definition de facto. GMGtalk 10:50, 20 July 2018 (UTC)
If we want a counter-intuitive thought experiment, consider the opposite scenario. An established user has a dispute with a new user on Prayer. They leave a neutral notification on FTN, as a central noticeboard, but do so knowing they frequent FTN themselves, and with the expectation that they will primarily attract users they've worked with extensively in the past, as opposed to also notifying the equally appropriate Wikipedia:WikiProject Religion, with which they are not personally active. Consider that their argument on Prayer is actually quite weak, and participants from FTN generally agree with the new user. We would probably not ever characterize that as a form of canvassing, even though by their intent, the subtle difference between who they notified and why was done with the expectation they would attract others to the discussion with whom they would agree. The primary consideration would be the bare facts, without recourse to what their intention may have been, even if we could somehow read their mind, or suppose they later posted on their talk page to the effect of That discussion went off the rails. I did not expect users from FTN to side with a religious nut over someone they know. GMGtalk 12:24, 20 July 2018 (UTC)
Some editors have a, shall we say, surprising lack of awareness of the impact of their edits. I do think editors can post a message without an intent to bias a decision, and yet do so anyway. Going back to this discussion on using canvassing as a neutral term, if that had been accepted, then I think it is less clear that such a thing can happen without intent. But going with the English Wikipedia connotations of "canvassing", I believe there are cases where editors honestly think they are posting a neutral message or providing neutral notifications, but are not. We could try not to label situations that lack intent as canvassing, but I suspect it'll be difficult to get wide adoption, as it requires too much divining. isaacl (talk) 19:45, 15 July 2018 (UTC)

Proposal: social media

I propose we add a "social media" section (level3 header under "Inappropriate notification"), which would read something along these lines:

Posting links to discussions or links to articles under discussion, on social media such as Twitter and Facebook or on internet forums or mailing lists is generally considered to be canvassing. Even if the medium is public, the audience listening to one's posts in such media is often highly partisan.

Thoughts ? Icewhiz (talk) 13:45, 11 February 2019 (UTC)

It is hard to accept that being a Wikipedian means you can never again post a link to a Wikipedia discussion on Twitter or Facebook. That would be draconian and it's really not a realistic interpretation of canvassing.
Best practice help is what is needed. If you have 10,000 followers on Twitter, you might affect a vote, best practice may include being aware of that impact and if it is likely that several of your followers will go and vote, you probably need to flag that in the discussion, or raise it in advance for comments about what you plan to post on a noticeboard before posting. If you make a mistake, and realize that your "ranty" post is distorting a particular vote, again we need positive ways of allowing the poster to flag it themselves, and ask for advice on what action to take to correct the situation, without feeling that this will land them in an Arbcom case or a perma-ban.
If I use @wikilgbt with its 20,000 followers to promote a MOS related policy RFC on Wikipedia (something I may well do), if I try to make my post fair and neutral, then attracting a few more LGBT+ interested Wikipedians to participate should be considered a good thing to address our problem of having a wide base of participation, not just the policy wonks and regular long termers. -- (talk) 15:41, 11 February 2019 (UTC)
Editors may use social media to draw more readers and editors to Wikipedia. To avoid accusations of canvassing and to avoid unfairly dominating discussions, editors should follow these best practices when making social media posts about active discussions on Wikipedia:
  • Disclose your social media post by placing a link to it in the Wikipedia discussion.
  • Word your post neutrally. In other words, don't say something like, "Come to this discussion and vote 'keep'."
  • Encourage new users to follow Wikipedia rules and customs.
  • Aim to improve the quality of Wikipedia articles.
  • Avoid summoning partisans to a battleground. Sometimes an invitation just isn't helpful. Jehochman Talk 16:07, 11 February 2019 (UTC)
Looks closer to something, agree with the 'advisory' list, though the last point gives me pause. Having been involved in a rather serious discussion about a different language Wikipedia where sysop powers were being used actively to discriminate against anyone perceived to be LGBT+, we can easily run into conflict over what words like partisan, bias, and even illegal, might mean in practice. For example, a Wikipedian Twitter poster who is known for, say, right wing views and wants to draw attention to a left wing bias dominating an article rename vote for what they see as "politically correct" reasons, is unlikely to be followed by anyone other than fellow right-wingers. Now, should we be telling them that even a neutral post on their Twitter stream which links to that discussion can never be made without being called canvassing? I feel like we should be responding differently, like encouraging them to flag it, and then balancing that by encouraging other Twitter users to comment about that potential bias, and make their own Tweets to gain a wider demographically balanced set of fresh voters. The end objective of gaining more viewpoints is perfectly healthy, responding by doing precisely that seems a better answer than trying to suppress free speech. -- (talk) 16:28, 11 February 2019 (UTC)
Apart from the last bullet point, these generally look like good practices. Some people have the remarkable effect of making anything they say "controversial", "partisan" or "biased" by the mere fact of who is stating it. We don't need a guideline that invites meta-arguments. XOR'easter (talk) 19:23, 11 February 2019 (UTC)
  • It is simple: don't post off-wiki about articles that are subject to a deletion proposal. We have processes that deal with notifications for them via deletion sorting etc. - Sitush (talk) 16:34, 11 February 2019 (UTC)
If there is a dangerous cliff one builds a fence several feet back from the edge, not right up against the edge. Jehochman Talk 16:58, 11 February 2019 (UTC)
(ec, same thing, more words) That could be a guideline, but we should also cover all other types of vote or discussion where participation numbers might game the system. Further, there is a pretty obvious primary response, those with anything critical to say or promote will use anonymous accounts, and it is not unreasonable to expect that many Wikipedians already do use social media accounts that are never publicly connected to their Wikimedia account. If a policy is only effective for those that are open and transparent, then it seems only to punish the "good" users who are open, with zero effect on those most likely to cause disruption or game the system. For these guidelines to be effective they have to positively encourage best practice, not just establish a naughty list. -- (talk) 16:59, 11 February 2019 (UTC)
And the more words we use, the more people will game the system through lawyering. Accept it: we're never going to stop this entirely from happening anyway, so let's just K.I.S.S. - Sitush (talk) 17:15, 11 February 2019 (UTC)

This is not a good proposal. What people post on their social media accounts is their own business. Attempting to regulate what someone says about a deletion discussion off-site is an ill-considered idea doomed to failure and (rightful) mockery from the rest of the world. "I can't tweet about Wikipedia? The hell I can't!" There are templates anyone can use to alert an AfD closer that a deletion discussion has attracted outside notice; much better to simply use those existing tools than to play social media nanny or codify prior restraint into policy. 28bytes (talk) 17:55, 11 February 2019 (UTC)

I think for most people, their followers are family and friends, and people are essentially tweeting as part of a large group conversation. While I appreciate there is a potential problem with biasing Wikipedia editing, it seems unduly restrictive to try to dictate that one can't talk about their Wikipedia activities as part of an off-wiki conversation. I don't think calls to action are the only issue, given the targeted audience. Mentions of articles of interest, newly created articles, and so forth can attract attention. But I don't know what can reasonably be done about it. The best way is to stop making decisions by straw poll, and learn the lessons of Clay Shirky and give more weight to editors who have established their reputation as being knowledgeable in the appropriate matters under consideration. But it's unlikely to happen any time soon. isaacl (talk) 18:28, 11 February 2019 (UTC)

Appeal to authority is a logical fallacy. The solution is to ignore votes that don’t have a valid reasoning based on verifiable facts. Anybody can post whatever they want but if they are encouraging wiki warfare, they can be sanctioned for it. I agree: no prior restraint. I don’t agree: no accountability for what you say and do. Jehochman Talk 18:42, 11 February 2019 (UTC)
Who has valid reasoning is in the eye of the beholder. If many people agree that Newyorkbrad has a history of giving well-reasoned analyses of stated arguments, then they will weigh his analysis more highly. Interpretation of policy is typically not an issue of verifiable facts, but an interpretation of community priorities and desires. isaacl (talk) 18:49, 11 February 2019 (UTC)
I think I'd prefer not to implement any change such as this at this time. We already have WP:STEALTH, which covers some aspects of off-wiki contact. Also, it looks to me like this proposal comes out of some disputes that are currently happening elsewhere on-site, and so it might be best to let a little time pass in order to have more perspective. --Tryptofish (talk) 23:16, 11 February 2019 (UTC)
I agree with Sitush that it's simple: "don't post off-wiki about articles that are subject to a deletion proposal." And not just about deletion discussions, but any matter that is trying to form consensus. The last thing we want are a bunch of forum users who don't edit Wikipedia and/or are unfamiliar with Wikipedia's rules to come and vote or otherwise weigh in on whatever, which has happened times before when people have tried to sway a matter in their favor on this site. Whether we call them meatpuppets or not, it's not something we should allow. Flyer22 Reborn (talk) 06:07, 12 February 2019 (UTC)
How would you feel about standard format notices, i.e. a standard recommended wording with some rules for sticking to hash tags that match the categories of the vote or AfD?
It could also be that twitter/facebook/telegram accounts and channels related to Wikimedia Associations and WikiProjects have some sort of behavioural agreement and then are effectively community approved to share vote related notices. Keep in mind that this already happens, and nobody thinks that a self selected group of Wikipedia users has any authority to stop or control how the wikimedia-l email list works, or how @WomenInRed composes tweets, though both those examples may be seen as having already adopting best practices for social media. -- (talk) 10:54, 12 February 2019 (UTC)
If it is a (fairly) standardized message AND there is notification on the discussion itself on Wikipedia (so other discussion participants aren't in the dark on where it was advertised - allowing them to counter-publish in other venues if they feel the advertised audience is biased) - THEN I personally think that's OK. Icewhiz (talk) 11:33, 12 February 2019 (UTC)
Fæ, "twitter/facebook/telegram accounts and channels related to Wikimedia Associations and WikiProjects hav[ing] some sort of behavioural agreement [making them] effectively community approved to share vote related notices" is better. But I'm iffy on it. More from me below. Flyer22 Reborn (talk) 15:26, 13 February 2019 (UTC)

As I don't have time or impetus for contributing these days, and thus am mostly a bystander, I don't know that people will bother to read much further. But this is a manifestly poor rule to enact, because it's simply going to turn into one of those backside-of-the-card rules from The Pilgrim's Regress. Obviously the solution for anyone who wants to talk about what goes on here is to make a pseudonymous comment. On-Wiki canvassing is manifestly much more of a problem (since that is one of the purposes of the projects), and this is not a way to fulfill "everyone can edit" in its cliquishness. Mangoe (talk) 12:05, 12 February 2019 (UTC)

  • I don't agree with Sitush's suggestion - it throws out the good and the neutral along with the bad. I do agree though that simple is good so I would suggest something like "If you post links to a Wikipedia discussion on external forums or social media, best practice is to word posts neutrally and, where possible¸ post links to your post in the discussion." In the past I have used social media to attract the attention of people I know who have relevant subject knowledge (this is little different to posting on a wikiproject talk page), and on one occasion used facebook to get an uninvolved editor to close a discussion I was involved in when all on-wiki channels had failed to get someone to do the job (iirc it was listed at ANRFC for about 2 weeks with no takers). Thryduulf (talk) 14:23, 12 February 2019 (UTC)
Posting to social media "to attract the attention of people who [...] have relevant subject knowledge" is not egregious as long as the posting is not to influence a deletion discussion or some other WP:Consensus matter. Even then, though, it can have the effect of a mishandled WP:Class assignment. Often, we get an influx of students who mean well coming to edit Wikipedia, but end up making a mess of articles. And that is even when they are somewhat trained. Think of how much worse it is to send a bunch of random people here from Facebook or Twitter. And as for neutral posts? Off-wiki canvassing was a big issue during the Gamergate controversy. It's still an issue. It does not matter how neutrally worded a post is if one is posting to a men's rights group or similar on Reddit about coming over to Wikipedia to weigh in on Gamergate issues. So the choice of a site/forum itself can be non-neutral. Those non-Wikipedians do not belong here. If one is contacting an uninvolved Wikipedian on Facebook to close a discussion, that's not the same thing. But people will find the matter questionable if that uninvolved Wikipedian is the editor's on-wiki friend. Flyer22 Reborn (talk) 15:26, 13 February 2019 (UTC)
On the surface these words seem like it could be a sensible precaution, but look at the severe lack of good faith that resulted from forcing me to Arbcom. I have been happily, openly, Tweeting about Wikipedia projects for 8 years, encouraging the public to support our projects. If I ever Tweet about Wikipedia again, it will not be from any account that you know about, because I do not want another experience of being hounded and pilloried. Driving Wikipedians active on social media underground as a necessary act of self-preservation, by using fear of witch hunts to punish their enthusiasm for our project and attempt to control their free speech, is the only consequence you will see from a strategy that is not supportive and positive to adopt. -- (talk) 15:44, 13 February 2019 (UTC)
I'm lost on what your point is, Fae, but I stand by what I stated above. I've never seen anything positive result from canvassing non-Wikipedians to come vote or otherwise weigh in on matters here. And if we don't want to call it canvassing, it's still a problem and has been a problem countless times. I'm not talking about "driving Wikipedians active on social media underground." I'm talking about bringing a bunch of newbies here who don't know what the heck they are doing. In my opinion, no newbie should be weighing in on a AfD...at least not without a few months experience with how WP:Notability and related matters on Wikipedia work. But when it comes to trying to contact Wikipedians on Twitter or Facebook, etc., the notification process can be sketchy in terms of selection bias. Facebook may have a number of pages devoted to whatever topic, and each page has a select group who may think somewhat differently from the other group devoted to the same topic. Unless an editor is going to seek to contact all associated Wikipedia Facebook pages devoted to the topic, there is some selection bias there. Meanwhile, Wikipedia has one WikiProject for the topic and it comes with transparency. Flyer22 Reborn (talk) 16:29, 13 February 2019 (UTC)
Regarding Those non-Wikipedians do not belong here: That's not how it works. Anyone can edit. Even newbies and/or clueless people. If their arguments are poor or unsupported by policy or common sense, the closer will take that into consideration. If their arguments make sense, we should listen to them, even if they were "canvassed" via a tweet. 28bytes (talk) 18:24, 13 February 2019 (UTC)
I knew someone might bring up that comment, but I stand by it. Non-Wikipedians canvassed here don't belong here. WP:Meatpuppet is pretty clear about that. I clearly stated, "It does not matter how neutrally worded a post is if one is posting to a men's rights group or similar on Reddit about coming over to Wikipedia to weigh in on Gamergate issues. So the choice of a site/forum itself can be non-neutral. Those non-Wikipedians do not belong here." I was referring to those types being brought here by an editor specifically to influence a matter on Wikipedia. Just no. Flyer22 Reborn (talk) 02:03, 15 February 2019 (UTC)
Hold on, I was accused of meatpuppetry and canvassing, this ended up with an Arbcom case request. The meatpuppetry policy states that you should not go around accusing people of meatpuppetry, which is a reminder to stick to the facts. The fact is, that someone tweeting a link to an AfD, is not by itself a request to your friends to set up SPAs and manipulate a vote. Even if someone were stupid enough to openly tweet from a non-anonymous account, asking their followers to create an account just to bias a vote, that tactic will always fail, as it is blooming obvious if a vote has an influx of SPAs or newbie accounts.
Setting unrealistic "rules" for Twitter, Facebook, and everywhere else off-wiki that nobody can ever link to a Wikipedia vote, is equivalent to shouting out of your bedroom window, about a problem that in practice is easy to identify and correct when it happens. The easiest solution to discouraging real meatpuppeteers is to handle it in an undramatic and civil way, in the very, very, rare occasion when it happens and might have a tangible affect, we already have sufficient policies and guidelines to do exactly that, without chasing unicorns by policing everybody's Twitter account and making everyone paranoid or driving good faith editors away with unpleasantly bad faith witch hunts. -- (talk) 11:13, 15 February 2019 (UTC)
Like I stated, we shouldn't be pulling in editors and non-editors in the way that I mentioned with regard to my Reddit example. The WP:Meatpuppetry policy is clear on that. The WP:Meatpuppetry policy does not state "that you should not go around accusing people of meatpuppetry." What it states is the following: "The term 'meatpuppet' may be seen by some as derogatory and should be used with care, in keeping with Wikipedia's civility policy. Because of the processes above, it may be counterproductive to directly accuse someone of being a 'meatpuppet', and doing so will often only inflame the dispute." And? When obvious meatpuppetry is going on, we note that. It's not like we are never to call someone a meatpuppet or point to the WP:Meatpuppetry policy. It's not like accusations of meatpuppetry are usually without common sense and therefore care. Flyer22 Reborn (talk) 03:38, 17 February 2019 (UTC)
This seems a tangent. Posting a link on Twitter, Facebook or elsewhere off-wiki, asking for more people to give their opinions, is not meatpuppetry. Neither would an effective ban from posting links off-wiki do anything to stop true meatpuppetry should anyone want to try it. -- (talk) 09:36, 17 February 2019 (UTC)
Posting a link on Twitter, Facebook or elsewhere off-wiki, asking for more people to give their opinions, is meatpuppetry when done within the realm of what the WP:Meatpuppetry policy covers. And ether way, it's problematic for reasons that I and others have noted in this discussion. It's canvassing that Wikipedia does not tolerate, as seen time and time again. An editor can claim that they were being neutral all they want to by posting a link to a Twitter or Facebook page that conforms to their view, but it will not work. And like Isaacl stated below, "It also provides incentive for pile-on comments: if the supporters for one viewpoint canvasses and supporters of other viewpoints do not, there is a disproportionate weight given towards one viewpoint." This is similar to what I stated above; I stated, "But when it comes to trying to contact Wikipedians on Twitter or Facebook, etc., the notification process can be sketchy in terms of selection bias. Facebook may have a number of pages devoted to whatever topic, and each page has a select group who may think somewhat differently from the other group devoted to the same topic. Unless an editor is going to seek to contact all associated Wikipedia Facebook pages devoted to the topic, there is some selection bias there. Meanwhile, Wikipedia has one WikiProject for the topic and it comes with transparency." Flyer22 Reborn (talk) 11:42, 17 February 2019 (UTC)
It's a tangent because posting a link is not asking your friends and relatives to create accounts to manipulate a vote. In the example Arbcom case, there were zero SPAs, yet I was still accused of meatpuppetry and canvassing.
The are no good outcomes here. Do we start banning well established contributors because they openly posted a link on their Twitter accounts, and make enough fuss so that everyone with anything to say about Wikipedia in the real world, will now ensure they do so with no link to their Wikipedia accounts? This is paranoid rubbish. As Nemo has pointed out, all you do with daft unenforceable faux policies like this will be to put more power in the hands of malicious actors who would like nothing better than create plenty of drama that gets long term Wikipedians banned. -- (talk) 12:02, 17 February 2019 (UTC)
We aren't going to agree on this. Like I stated, posting a link on Twitter, Facebook or elsewhere off-wiki, asking for more people to give their opinions, is problematic and is not something that Wikipedia tolerates. At this point, I'm just repeating myself. If we held a big RfC on this, I very much doubt that most Wikipedians are going to agree that posting a link on Twitter, Facebook or elsewhere off-wiki, asking for more people to give their opinions, is just fine. Flyer22 Reborn (talk) 14:00, 17 February 2019 (UTC)
It's not clear to me that the community would agree with you. Consider GorillaWarfare's statement that she has expressed opinions in other venues about Wikipedia activities. I think trying to restrict how much people can talk about their Wikipedia contributions in other places is difficult. isaacl (talk) 16:08, 17 February 2019 (UTC)
Considering how the Wikipedia community has repeatedly treated this matter (including at WP:ANI), I think it is clear that they would agree with me. Regardless of the comment by that editor, Wikipedia has been like this for the longest, which is why Arbitration was even considered in that case. Plus, ArbCom's opinions have commonly been out of step with the wider Wikipedia community's opinions. I'm stating that if an AfD is going on, and an editor advertises that AfD on social media and their followers and/or friends, who are not Wikipedia editors, come pouring in, it's been called canvassing or otherwise inappropriate many times and the votes and other arguments by those one-time posters are discounted. One doesn't even have to know who brought in those influx of newbies; it's obvious that someone did unless the article is getting a lot of traction already, like the kind of traction that the Chelsea Manning article was getting at one point regarding the article title and what pronouns to use in the article; that article was also in news. And discounting the comments of obviously canvassed individuals is the same regardless. Even if all or some of the editors are Wikipedia editors, it's seen as problematic because it's highly likely that those editors share that editor's point of view, while editors contacted at different related WikiProjects are likelier to have more diverse opinions on the matter. The same thing has applied to move discussions. If it's a Wikipedia dispute of any kind, and influx comes in, we know that canvassing (whether or not one wants to call it canvassing) has occurred (again, unless the article's traction indicates that the influx may not be due to canvassing) and an admin may be called in to handle it. It might be taken to WP:ANI. This has certainly been the case with articles under WP:Discretionary sanctions. The vast majority of editors do not want to allow canvassing, whether it's canvassing Wikipedia editors or the general public, and it's for reasons I stated and what you stated below. Flyer22 Reborn (talk) 23:45, 18 February 2019 (UTC) Flyer22 Reborn (talk) 23:55, 18 February 2019 (UTC)
The question is what's special about an AfD? If I post a tweet saying I'm working on improving coverage of person X's political views, then it draws the attention of all my followers to the person X article, and they could show up on the talk page to ostensibly help me. A prohibition on using social media would have to cover any mention of Wikipedia-related matters to be effective. It's unclear to me that a consensus can be obtained to agree to this. isaacl (talk) 03:29, 19 February 2019 (UTC)
I didn't just focus on AfD. And AfD is obviously special in terms of consensus; it's about whether or not Wikipedia should have a Wikipedia article on the topic. It's something that is very often subject to a significant dispute, and WP:Single purpose accounts showing up, and WP:Gaming the system issues. A dispute over what image to use does not have the same level of conflict, in part because it's not advertised with a tag like AfD is. As for "a prohibition," we already have the WP:Canvassing guideline, the WP:Meatpuppetry policy and the WP:Common sense supplement page. If you contacting your followers or friends about an AfD or whatever dispute results in the matter leaning in your favor, common sense dictates that you alerting your followers was inappropriate. And you can bet that the AfD or whatever will be considered tainted and that comments from the people you pulled in will be discarded. Again, this has happened numerous times on Wikipedia. I am speaking from experience. I also watch and edit enough controversial topics to know what I'm talking about. But, hey, we can always take the matter to WP:Village pump (policy) and see just how many are going to agree that a person contacting their followers or friends about an AfD or whatever dispute, resulting in those people coming to Wikipedia and weighing in on the matter so that the dispute (artificially) leans in that editor's favor, is a fine and dandy. No one wants to be in a dispute with someone and to then have that person go whine on Twitter or whatever and bring their followers into the dispute. Flyer22 Reborn (talk) 06:01, 19 February 2019 (UTC)
Yes, I am also speaking from what I've seen and experienced. "Please vote for X" is frowned upon, and new accounts coming to Wikipedia will be found and dealt with. "I'm writing articles about group X" doesn't cause anyone to bat an eye, and it's really hard to tell how an established editor decided to weigh in a discussion. A prohibition on overt canvassing will just push people to less obvious ways of attracting like-minded people. No one likes outsiders who don't understand the context of the current discussion to come in and shoot off an uninformed opinion, no matter if they are new Wikipedia editors or old hands. But English Wikipedia's decision-making tradition gives everyone equal weight, which provides an incentive to canvass. To really put an end to it, we need a decision-making process that makes it less effective. There are different, non-mutually exclusive options, but all of them will to some degree move away from treating every individual editor's viewpoint as equal, and the English Wikipedia community does not seem amenable to following this direction at present. isaacl (talk) 07:02, 19 February 2019 (UTC)
Again, when it comes to Wikipedia tolerating the type of canvassing or "pulling in people" this discussion has concerned, Wikipedia in general does not tolerate it. And for good reason. That has been my point. You stated, "A prohibition on overt canvassing will just push people to less obvious ways of attracting like-minded people." Eh? Overt canvassing is already covered by the guideline. Contacting off Wikipedia? The "Stealth canvassing" section states, "Because it is less transparent than on-wiki notifications, the use of email or other off-wiki communication to notify editors is discouraged unless there is a significant reason for not using talk page notifications. Depending on the specific circumstances, sending a notification to a group of editors by email may be looked at more negatively than sending the same message to the same group of people on their talk pages." From what I've seen, editors do not restrict the "to notify editors" part to editors only; they apply it non-Wikipedians who may become editors at any moment as well. And editors already do "less obvious ways of attracting like-minded people." I don't agree that explicitly stating "don't post about disagreeing with an AfD nomination on social media while the AfD is ongoing" will increase canvassing or make things worse. That's like stating that we shouldn't be strict at all when it comes to our rules because it just makes breaking the rules that more alluring or makes people that more determined to break the rules, like telling a child not to do something. Either way, I'm not arguing for anything to be added to the WP:Canvassing guideline. When the "pulling in people" thing this discussion has talked about comes up, editors already point to WP:Canvassing or WP:Meatpuppetry; so the WP:Canvassing guideline and WP:Meatpuppetry policy are working fine in that regard. Editors already use common sense on the matter. Flyer22 Reborn (talk) 01:29, 21 February 2019 (UTC)
My apologies for eliding the full statement; of course I meant if the prohibition in this proposal is implemented, then people will just find other ways not covered by this proposal to attract like-minded people. But if you're not arguing in favour of this proposal, then we don't really have to continue this thread. isaacl (talk) 02:30, 21 February 2019 (UTC)
The problem with canvassing is that English Wikipedia discussions are typically closed by straw poll rather than weighing the arguments directly. The number of supporters for a position is used as a proxy to indicate the strength of the sum total of the arguments for the position. This fails to isolate the strength of each individual argument, which is necessary to weigh them against each other properly. It also provides incentive for pile-on comments: if the supporters for one viewpoint canvasses and supporters of other viewpoints do not, there is a disproportionate weight given towards one viewpoint. As I said in my comments in the case, moving away from straw polls to make decisions and focusing on weighing arguments for and against is the best way to reduce the incentive for canvassing. isaacl (talk) 16:40, 15 February 2019 (UTC)
I have seen more good closers than I have seen bad closers. A good closer should know not to close an AfD based solely on how many people voted "Keep" or "Delete." WP:Consensus is clear that consensus is not supposed to be about votes, except for those discussions that are specifically designed for it...such as an RfA. If it's an image discussion, it's also common for it to be about the number of votes in favor vs. against. But if I see a matter closed solely or mostly based on votes when it should not have been, I am likely to address the closer about it. Flyer22 Reborn (talk) 00:09, 19 February 2019 (UTC)
Other than cases of obvious bad rationales, or a spate of newly registered accounts, I've rarely seen closures go against the straw poll results, and those that do get raised at deletion review, where the number of supporters is used as a proxy for strength of argument. Canvassing of persons who joined Wikipedia just to chime in on the issue in question is the easy case to deal with. But if already-registered editors are drawn to a conversation, it's hard to discount their views, particularly when it's not a black-and-white problem but one of interpreting community priorities. isaacl (talk) 03:29, 19 February 2019 (UTC)
Unless a discussion specifically has a straw poll section as a last resort, I often seen closers weighing the strength of the arguments, as they should. It's often that that the majority of the votes just so happen have the stronger argument(s). If it's often that the majority of the votes have poor arguments and the closer closes in favor of those poor arguments, that is a problem and it's something that needs to be addressed elsewhere for broader discussion. Flyer22 Reborn (talk) 06:01, 19 February 2019 (UTC) Flyer22 Reborn (talk) 06:09, 19 February 2019 (UTC)
Almost all discussions end up having support and oppose statements, as it seems people really like to write them (and put words in bold)... As I wrote above, if closers really closed on the basis of strength of argument, they'd line up the arguments, evaluate them, evaluate each option on the basis of the arguments, and see which option yields the best net positives. Instead, they all make statements about how many people supported each option, and how much they discounted individual opinions based on the arguments. The focus is on the number of people supporting each option, instead of weighing the options based on their pros and cons. isaacl (talk) 07:02, 19 February 2019 (UTC)
The support and oppose statements usually have rationales. If those rationales are not based on policies or guidelines, they are often ignored. If they are not ignored, they usually should be. Sometimes editors will specifically address the editor with a poor rationale; for example, if the editor has made a poor WP:OTHERSTUFFEXISTS argument. You stated, "As I wrote above, if closers really closed on the basis of strength of argument [...]." Are you really arguing that editors never or rarely close on the strength of the arguments? If so, I can point to numerous closes where the editor clearly closed on the strength of the arguments. In this case, we can see that Galobtter discounted "the other arguments against [inclusion that] were personal opinions on the importance of [the material]." Galobtter also stated that "the not yet arguments are somewhat less strong 15 days later." In this case, we can see that Swarm thoroughly weighed the strength of the arguments, and the pros and the cons, and that when I addressed him about the way he closed...he clarified. There is no need to "line up the arguments" (especially the poor ones), but Swarm did line up matters in a way that helped clarify how he closed. In this case, we can see that Wugapodes weighed the strength of the arguments (although I and others disagree that it's a WP:NOTCENSORED matter). Yes, Wugapodes stated that "the numbers were evenly split between the two positions," but Wugapodes immediately followed that up with "both of which have reasonable policy based rationales for their side." As I mentioned, there are a lot of other examples. So I don't agree that "the focus is [always or solely] on the number of people supporting each option, instead of weighing the options based on their pros and cons." Yes, the number of people supporting each option is a part of closing, but good closers also weigh the strength of the arguments. So to repeat: If we have more poor closers than good closers, that is an issue to address elsewhere. Flyer22 Reborn (talk) 01:29, 21 February 2019 (UTC) Flyer22 Reborn (talk) 01:43, 21 February 2019 (UTC)
Your examples show closures that basically follow what I said most closers do: they looked at the straw poll results and then looked for reasons to discount the views for each option based on strength of argument, and they all followed the straw poll results. Swarm explicitly counted up the number of people supporting each option. Wugapodes found there was no consensus after saying the numbers were split. These cases don't show the closer making a decision based on strength of argument, since the closer did not provide a description of each option's strengths and weaknesses on that basis. In the real world, if a decision is being made by consensus and not a vote or benevolent dictator, the typical approach is to tally up the relative pros and cons of each option and weigh them on the basis of the strength of arguments. If you have examples of any closures that took an argument-first approach, breaking down and analyzing the options on the basis of their pros and cons, I'd be glad to see them. isaacl (talk) 03:03, 21 February 2019 (UTC)
At the risk of taking this tangent further off topic, I don't believe your characterization of my close is quite accurate enough. I did see consensus, there was near unanimity to keep the image but there was no agreement on what to do with it so it defaulted to the status quo. In fact, that close was partly diplomatic. I think a reasonable editor could have interpreted that discussion as having consensus against obfuscation, and I personally didn't find the arguments for obfuscating it strong. However the result between consensus against and no consensus would be the same (it stays in the lead unobfuscated), so given the wide range of disagreement I opted to take the more conservative approach and leave open the possibility of compromise. That range of disagreement was exemplified by the tallies, but it doesn't have to be. Though there was near unanimity in wanting the image somewhere, there were 7 different proposals on what to do with it: uncollapsed in lede, collapsed in lead, move to mid article and don't collapse, move to mid article as part of a gallery, keep in lede but use a progressively sharpening gif, keep but unspecified on what to do with it, and the 2 uncompromising remove opinions. By far the most common opinion was keep in the lede uncollapsed, and the opinions are only "split" if you do what I did and summarize the other proposals under some common theme: they all didn't want the image to stay as it was. So even within the "not status quo" camp there was no consensus on what to do. Though I grouped these into "obfuscate" and "don't obfuscate" for ease of exposition, I could have done that close without any reference to the number of opinions in each of those camps. My job as a closer is fundamentally not to weigh arguments, except where the rationale violates our fundamental principles, it is to neutrally summarize the opinions of the community. If an argument has merit it should have convinced the community and would be obvious from the discussion through widespread agreement. Our goal in discussions should not be to convince some random person who stumbled across WP:ANRFC, but to convince and compromise with others in the discussion. This is why, in that close, I link to WP:Not counting heads which says "If the discussion shows that some people think one policy is controlling, and some another, the closer is expected to close by judging which view has the predominant number of responsible Wikipedians supporting it, not personally select which is the better policy." So Flyer is right, there is disagreement in whether NOTCENSORED applies, but that disagreement was brought up and not resolved in the 2013 discussion, and brought up and not resolved in the 2015 discussion. I did not close as I did because I thought they had equal merit (I don't, I think NOTCENSORED applies), but because there is and has been disagreement about whether it does and so me deciding which one wins isn't an adequate representation of the lack of agreement in the community for the past 5 years. Wugapodes [thɑk] [ˈkan.ˌʧɹɪbz] 07:57, 21 February 2019 (UTC)
Isaacl, I'm not going to keep debating you on this. You argued that closers don't really close on the basis of the strength of the arguments. You stated, "Instead, they all make statements about how many people supported each option, and how much they discounted individual opinions based on the arguments. The focus is on the number of people supporting each option, instead of weighing the options based on their pros and cons." You are wrong, per what I stated above (including the examples I pointed to). Flyer22 Reborn (talk) 10:32, 21 February 2019 (UTC)
Wugapodes, sorry if you felt pulled into this. I figured that If I was going to point to examples, it was fair that I ping the closers I've mentioned. And your response to Isaacl above shows that I was right to ping you. I want to point out, though, that the gallery discussion wasn't technically a part of the RfC. Had that option been presented in the RfC -- the suggestion that we use a less trypophobic-inducing image as the lead image and move the lotus image to a gallery -- I think a number editors would have voted for that option. If I do another RfC on the matter, I will present that option in the RfC. Flyer22 Reborn (talk) 10:47, 21 February 2019 (UTC)
The sentence The numbers were evenly split between the two positions, both of which have reasonable policy based rationales for their side and so I'm loathe to discount them. is the problematic one. A evaluation of the community's viewpoints on the strongest arguments has no need to discount the numbers. But I will agree that my viewpoint is mainly a semantic difference for this case, since the discussion is weighing the strength of opposing priorities for image placement, and there's no evident way to weigh that without counting heads. I appreciate the greater explanation provided by Wugapodes here. I disagree that neutrally summarizing the opinions of the community precludes determining its consensus view on the pros and cons of the various options. I think finding an approach closest to true consensus, where the greatest net gain is achieved across the community, is what is sought. The community includes everyone, not just those participating in the conversation. I disagree that convincing others in the discussion is necessary for one argument to be viewed as stronger than the others, as those participating are just a small sample and self-selected. I agree though that there are multiple ways to achieve this, but some are more vulnerable to canvassing than others. English Wikipedia's current decision-making tradition is vulnerable to self-selection, which is why we try to stop canvassing. But this exacerbates the small sample size problem, as there are only specific acceptable ways to try to increase the sampling of people weighing in. isaacl (talk) 18:32, 21 February 2019 (UTC)
To repeat: Your argument was that closers don't weigh the strength of the arguments. Per my above examples, I view you as wrong on that. And that is the case regardless of if you are speaking of "generally don't" or "never don't." Flyer22 Reborn (talk) 03:41, 23 February 2019 (UTC)
  • This kind of rule on off-wiki behaviour can only serve to discourage transparency and empower abuse. We're never going to find out about private chat rooms or Facebook groups created with malicious intents, while you can be certain that good faith activity done in the open (and probably innocuous) would get targeted by malicious actors. Nemo 14:56, 12 February 2019 (UTC)
  • Oppose AfD is moribund and so discussions often have to be relisted again and again to attract any input. This is a symptom of a general decline in Wikipedia participation. It is therefore sensible to publicise discussions. We put a big banner tag on any article up for deletion to attract attention but Wikipedia's mobile app doesn't show this – not the slightest clue. Other means are therefore needed. And you simply can't stop people talking to each other. Andrew D. (talk) 09:05, 3 May 2019 (UTC)
  • Weak support A while ago I would have said no, but after what happened over Mark Dice a couple of months ago I now lean yes. It caused far to much hassle to try and deal with the wave of meat puppets who fetched up.Slatersteven (talk) 14:51, 3 May 2019 (UTC)

What About Off-Site Organizing & Education?

As a relatively New Editor, I've been confronted with several challenges, and wished that there were some kind of off-site resource I could go in order to have open and wide-ranging conversations (meaning outside Wikipedia Rules) about "How to Edit Wikipedia", to the extent that I sometimes toy with the idea of forming such a site on my own. Last night I happened upon this Article, read it, and wondered if there was anything outside of this Article that would prohibit, or regulate, such an activity. I've heard rumors that sometimes college students get together and make a project out of editing Wikipedia and always assumed this was considered okay, but when I contemplate doing, or facilitating this kind of thing myself, I wonder if the idea be doomed from the start for some unknown reason.Tym Whittier (talk) 17:30, 13 May 2019 (UTC)

A rule clarification that is (apparently) needed

Hi all. :) I'm proposing adding the following sentence to the "Vote stacking" section. (My addition is in bold):

Votestacking is an attempt to sway consensus by selectively notifying editors who have or are thought to have a predetermined point of view or opinion (which may be determined, among other ways, from a userpage notice, such as a userbox, or from user categorization), and thus encouraging them to participate in the discussion. (Note that "notifying" an editor includes, but is not limited to, leaving a message on their talk page, pinging them, or contacting them outside of Wikipedia.)

This is in response to a particular user (user:Flyer22 Reborn) arguing, in an ongoing move discussion on the Genderqueer talk page, that pinging someone by mentioning them on a talk page does not count as "notifying" them. While Flyer22's stance obviously goes against the plain language of the guideline*, if they are confused about what "notifying" someone means perhaps other people are too. More clarification wouldn't hurt.

I BOLDly already made the change. Predictably, Flyer22 reverted it (in an edit incorrectly marked minor). In their edit summary they said Editor trying to use this page to influence a discussion. This isn't true, I haven't cited my guideline change in the ongoing dispute and had no intention of doing so. Instead, I'm trying to prevent people from misunderstanding (or trying to Wikilawyer their way around) the guideline in the future.

Note that this rule change would not completely prohibit talk page pings during contentious discussions. The intention here is just clarify that a talk page ping is essentially the same thing as a user page ping which essentially is the same thing as an email, etc. Sometimes, as the guideline notes, it is acceptable to notify certain users during a discussion and this clarification would not change that.

*See: the Wiktionary definition of notify: To give (someone) notice of (something). See also Wikipedia:Notifications, which does not distinguish between user page messages and talk page pings, and lists them both as a form of notification/pinging.

WanderingWanda (they/them) (t/c) 01:49, 3 May 2019 (UTC)

Regarding this, as seen here (followup note here), I reverted the following addition by WanderingWanda: "Note that 'notifying' an editor includes, but is not limited to, leaving a message on their talk page, pinging them, or contacting them outside of Wikipedia."
As I've made clear to WanderingWanda, what counts as canvassing when pinging and whether or not pinging should count as canvassing is debated among editors. This is seen, for example, with the Wikipedia talk:Requests for adminship/Dane discussion. It's been discussed at this talk page before, and in other venues, like WP:ANI. The community has not been able to find a consensus on this. This is because of the intent aspect that comes along with canvassing and because of complicated cases like the Dane one. Like I mentioned before, "I see such pinging all the time at WP:ANI, for example, and the pings are usually used by editors (administrators and non-administrators) for support on whatever they are arguing." And regarding notifying editors off Wikipedia, I noted to WanderingWanda that, as seen in the #Proposal: social media discussion above, even what constitutes canvassing when alerting a Wikipedia editor off Wikipedia has recently been debated. And yet WanderingWanda made the "contacting them outside of Wikipedia" addition.
I was clear to WanderingWanda that pinging editors who were involved in a previous discussion, who watch the talk page where that previous discussion was had, and who asked to be kept informed is not canvassing. Notice the use of "and" instead of "or" in that previous sentence. Not only was that the case, I was mentioning two editors who researched a matter. I was not going to mention them and not ping them. Even after I pointed WanderingWanda to Wikipedia:Canvassing#Appropriate notification, which notes the exemptions to the canvassing guideline, including "Editors who have made substantial edits to the topic or article," "Editors who have participated in previous discussions on the same topic (or closely related topics)," "Editors known for expertise in the field," and "Editors who have asked to be kept informed," WanderingWanda still chose to view the matter as a canvassing violation (as is clear by their "Flyer is confused" nonsense above). Like I told WanderingWanda, "If one or two editors asked to be kept informed, I do not need to then notify every editor who participated in the discussion but argued in the opposite direction." Predictably, WanderingWanda came to this guideline to add "pinging" to it, despite the Dane example I pointed to. Despite all of the past discussions on pinging and editors being cautious about adding anything on it to this guideline for a number of reasons, including those gone over by Collect, WanderingWanda, a newbie who isn't really a newbie, somehow knows better than the significantly experienced Wikipedians. I'm apparently confused despite pinging editors the way I did being quite clearly allowed under the "Appropriate notification" section. Apparently, we need clarification on the a rule that WanderingWanda understands better than us. WanderingWanda speaks of wanting to stop editors from gaming the system... Well, adding "pinging them" without context is ripe for Wikilawyering. And, indeed, "notification" being interpreted as including pinging was discussed after it was added to the introuction. This type of interpretation by WanderingWanda is what I was concerned about. As for WanderingWanda not trying to influence the Genderqueer discussion, I disagree. There was no reason to even mention that article here in this section, for example. Flyer22 Reborn (talk) 02:27, 3 May 2019 (UTC)
Whether you were justified in pinging the specific editors you did in that specific discussion is beside the point. It seemed like canvassing to me, but, whatever, that's not what I'm here to talk about.
We're both agreed that there are exceptions where it is appropriate to notify editors of a discussion. Here's where we seem to disagree: you've said (in that other ongoing discussion) that WP:Canvassing [does] not address WP:Pinging and (just now) whether or not pinging should count as canvassing is debated among editors. To me, that position doesn't even make sense. It's literally incomprehensible to me. How is notifying someone on a talk page different from notifying someone on a user page? How is one a notification and one not a notification? They're obviously both notifications!
I fully admit I'm not a very experienced editor (although, hey, I did just get my apprentice badge. :) so maybe there's something obvious here I'm missing, but I'm completely baffled by your position. WanderingWanda (they/them) (t/c) 02:54, 3 May 2019 (UTC)
How are you "completely baffled by [my] position" when looking at cases like the Wikipedia talk:Requests for adminship/Dane case? Why do you think, in that discussion, Sir Joseph stated that "pinging is not canvassing."? Why do you think editors were arguing over pinging in relation to canvassing in that discussion? How many more similar discussions should I point you to for you to get the point that it's the case that WP:Canvassing clearly does not specifically address WP:Pinging and that whether or not pinging should count as canvassing is debated among editors? Should I ping everyone involved in that Dane discussion? I think I should. We obviously need a big discussion on this. And by "a big discussion," I mean a discussion at WP:Village pump (policy) about it. It's long overdue. I personally can see pinging as canvassing in a particular case, but adding "pinging" to the guideline needs much discussion per past arguments on the matter. You interpreting my having pinged those editors as canvassing is a prime example of that. Flyer22 Reborn (talk) 03:16, 3 May 2019 (UTC) Flyer22 Reborn (talk) 03:19, 3 May 2019 (UTC)
I have no idea what you guys are talking about, but in general, for a discussion, as you can see at WP:AN, you can't ping an editor, you need to notify them on the talk page. People don't necessarily read pings. In order to be properly notified, it must be on the user's talk page. That is my opinion. Sir Joseph (talk) 04:04, 3 May 2019 (UTC)
A formal notification requires a user talk page message, yes. But canvassing can be done with a ping, email, or other off-site channels. Whether or not a specific message is canvassing is a different question. isaacl (talk) 04:12, 3 May 2019 (UTC)
--
So let's back up and ask a few basic questions.
1. What is a ping? The page WP:Notifications seems to indicate that a ping is anything you do on Wikipedia that causes another user to receive a notification pop-up, including sending them a Wiki email or leaving a message on their user talk page. However, in this discussion I think we're using the term more narrowly to mean: when you link to a user's page on a talk page (which automatically sends a notification to that user). For clarity, from now on I'm going to call that specific form of ping a talk page ping.
2. Does the guideline Wikipedia:Canvassing cover talk page pings? It does not mention talk page pings specifically, but what it does cover is notifying in general.
3. Does a talk page ping count as notifying? Well, it sends a notification to a user, so yes. How could it not? By any reasonable definition of the word notify, to ping someone on a talk page is to notify another user. Note that WP:Notifications treats pinging and notifying as synonymous: Notifications ...commonly called pinging. Therefore, any time the canvassing guideline talks bout notifying other users, the word notifying encompasses talk page pings.
4. So what, in the end, is the current guideline for talk page pings? Simple, any time you see a word like notify replace it with something like talk page ping, and you'll have your answer. Let's try it:

When talk-page-pinging other editors of discussions, keep the number of notifications small, keep the message text neutral, and don't preselect recipients according to their established opinions. Be open!

...In general, it is perfectly acceptable to talk-page-ping other editors of ongoing discussions, provided that it be done with the intent to improve the quality of the discussion by broadening participation to more fully achieve consensus.

Canvassing is talk-page-pinging done with the intention of influencing the outcome of a discussion in a particular way, and is considered inappropriate. This is because it compromises the normal consensus decision-making process, and therefore is generally considered disruptive behavior...

Votestacking is an attempt to sway consensus by selectively talk-page-pinging editors who have or are thought to have a predetermined point of view or opinion (which may be determined, among other ways, from a userpage notice, such as a userbox, or from user categorization), and thus encouraging them to participate in the discussion.

In conclusion, I disagree with the notion that the guideline doesn't currently cover talk page pings, it does cover them, just not explicitly. In the discussion you linked, the person who seemed to have the most reasonable point of view was Swarm:

It's a matter of common sense as to whether pinging constitutes canvassing—it depends on the intent, obviously, and can only be judged on a case by case basis. If one opposes an RfA, and unnecessarily pings another editor who was in a dispute with the nominee and will likely oppose as well, I don't understand how you can say it's objectively not canvassing. Per AGF, the default assumption should be that canvassing was not the intent, but surely you could see why one might view such a move as suspect.

In the case of that other contentious discussion we're both engaged in currently... if you pinged those editors because your intent was to get additional voters that would vote in line with your own POV, it was canvassing, if you had another intent, it wasn't. WanderingWanda (they/them) (t/c) 04:09, 3 May 2019 (UTC)
The reason that the guideline does not yet explicitly cover pinging is because of the issues I mentioned. And regardless of whether or not an editor is going to agree with the person doing the pinging, and the one doing the pinging knows that the editor will agree, it's still the case that pinging an editor who asked to be kept informed is an exemption. Nowhere does the guideline state that an editor should only keep another editor informed if the editor who asked to be kept informed will disagree with them. Nowhere does the guideline state that an editor should only keep another editor informed if, after pinging the editor who asked to be kept informed, that editor pings all others who were involved in the previous discussion. Nowhere does the guideline state that pinging an editor who is already watching the talk page is canvassing.
Speaking of pinging, I'm pinging the following editors who took part in that Dane pinging discussion: TParis, Beeblebrox, Ivanvector, Primefac, RileyBugz, Juliancolton, Lepricavark, Swarm, Ritchie333, clpo13, and Yintan. Also pinging SMcCandlish because he's contributed to many of our policies and guidelines and is good at proposing matters at WP:Village pump (policy) if they need to be presented there. He often has good ideas about crafting policies and guidelines. I didn't ping Coffee because he's currently indefinitely blocked.
If we are going to add pinging to the guideline, let's get it right. Flyer22 Reborn (talk) 05:07, 3 May 2019 (UTC)
  • There seems some disagreement on the necessity of a) whether individuals in prior discussions had to have stated their desire to be informed to be pinged in future similar discussions and b) whether a minority of individuals desiring to be informed, could be pinged without needing to ping all individuals. I would say a is clearly "if they sounded interested, you can ping them, if you're even-handed" and b is "yes, you can first ping the "enthusiastic" group, without needing to ping all. So long as this is not part of a shared tactic to enable canvassing". Nosebagbear (talk) 14:47, 3 May 2019 (UTC)

Poll

  • I support the proposed addition as a useful clarification and an anti-WP:GAMING effort, even if I deplore the tone of this proposal's presentation and the major response to it. Just the amount of WP:WIKILAWYERING going on in the discussion following the proposal is strong evidence why this is needed. I would say that I've noticed that attempts to skirt the edge of WP:CANVASS have gotten much worse and more frequent over the last couple of years, and that the community enforcement against it has weakened, in part due to lack of specificity in the guideline wording. Some meta-commentary on this thread: Hint 1) If your only rationale for doing something is that the formally written policies and guidelines don't quite prohibit it in exact wording, you have no argument; WP does not work that way, and you know it. Hint 2) if you're proposing a change, stick to the proposal and don't inject extraneous arguments that go way beyond it and start treading on practices that most editors firmly believe are not a problem.) In closing, please consider that the neutrality, concision, and other advice at WP:RFC is applicable to anything serving the function of an RfC, including proposals that don't have an {{RfC}} tag on them (they pre-date the existence of that template, for one thing; and see also GAMING, WIKILAWYER, WP:POLICY, WP:CONSENSUS, WP:COMMONSENSE, etc., about trying to escape a general principle on a technicality, a tactic exhibited by both of the more vocal sides in this debate). Please also consider that editor A unnecessarily personalizing a matter to seem be all about editor B doesn't require editor B to do the same, tit-for-tat; it's more constructive to rise above this (even just declaring that the personalization is distasteful and will not be addressed in any detail is often sufficient to deflate it).  — SMcCandlish ¢ 😼  05:40, 3 May 2019 (UTC)
Thanks for weighing in,  SMcCandlish. You support this exact wording? Or would you propose different wording? And what of the gray issues/context aspect that both Collect and Swarm have mentioned? And what of the "editors who have asked to be informed and/or are already watching the talk page" aspect that I mentioned above? There are many cases where a dispute is going on at the talk page of an article and editors ping other watchers of that talk page. I'll go ahead and notify WP:Village pump (policy) to this discussion. Masem and BullRangifer, who are also good at crafting guidelines and policies, might also propose something. Flyer22 Reborn (talk) 05:56, 3 May 2019 (UTC)
I don't have any issues with the original wording change proposed; it's still all subordinate to the main point, "selectively notifying editors who have or are thought to have a predetermined point of view or opinion" (i.e., vote-stacking). As I say in many matters like this, if there's some minor issue with the wording it can be tweaked, and most WP:P&G language gets adjusted over time; this is not carved in stone. Thus, I'm also not opposed to any particular tweaks to the wording if they make sense; nor to reforumlating this as a proper RfC. I'm supporting the general principle being included. I don't think any grey areas can or should be addressed in this; it's the very nature of grey areas, and WP is full of grey areas. WP:P&G and WP:CONSENSUS and related pages already take this into account, and WP:IAR exists for a reason. Next, when editors directly ask to be notified, community consensus already entirely accepts pinging them; there is no issue to resolve there. Same goes for notifying en masse all editors from a prior round of discussion about a new round. None of this relates to vote-stacking or is questionable. The proposed addition doesn't undermine these in any way; it does nothing but clarify that "selectively notifying editors [with] a predetermined point of view" isn't magically permissible if you do it via method B or method C instead of method A. It's simple, clear, and is actually already the operational community consensus that we simply haven't bothered writing down yet. It's just unfortunate that the proposal wasn't left at that, without the paragraphs of venting.  — SMcCandlish ¢ 😼  06:20, 3 May 2019 (UTC)
Just wanted to say I agree I haven't handled this proposal, or my disagreements with Flyer22, as well as I could have. Will try and be more mindful in future. WanderingWanda (they/them) (t/c) 17:10, 3 May 2019 (UTC)
  • Oppose - I'm not okay with the premise of this proposal to begin with. This is clearly inappropriate spillover from a content dispute, and it feels like a dirty move. One thing we don't do is change policy wording just so that we can write our content dispute opponents outside of policy compliance, and that appears to be exactly what is going on here. The proposed wording change itself is just a statement of common sense. It does not need to be spelled out that "pinging" is a method of "notification". It's asinine to suggest that Flyer is incapable of comprehending this, and to suggest a wording change of the policy itself in response to such a ridiculous premise is entirely inappropriate. There's nothing inherently wrong with notifying. There's nothing inherently wrong with pinging. That's not what this policy is about, with or without the proposed change. There's obviously this blatant underlying implication that Flyer engaged in canvassing via the ping function, which is sending selective notifications, with the intent to fabricate a biased result. When one suspects this is happening, one is supposed to substantiate their accusation with evidence and report the issue to administrators at WP:AN/I. Not jump straight into the policy page and try to reword it so that it seems like your opponent is more in the wrong. If Flyer canvassed, then that complaint should stand on its own merits. We should not be here, discussing changes in policy wording. Looking at the canvassing accusation itself, Flyer immediately refuted this complaint, saying "[the] editors I pinged, although they agree with me, were either involved in the previous discussions and/or asked to be kept informed. I'm not going to note who asked to be kept informed. And two of the editors -- Legitimus and Mathglot -- were central to research on the matter". In other words, these are editors who are already involved, which is a reasonable justification to notify them. That's not canvassing. If there is some evidence missing that would prove that the notifications were biased, surely Wanda would have seized on it. Yet, Wanda instead degenerated into this bizarre argument that Flyer was arguing that pinging could not be canvassing because it's not mentioned by this page. That's clearly a disingenuous straw man argument, and to take it so far as to attempt to change the policy itself is ridiculous. I refuse to take this seriously. ~Swarm~ {sting} 06:22, 3 May 2019 (UTC)
  • Oppose Pointy. Policies are not part of an extended war game. Any words along the lines of "contacting them outside of Wikipedia" encourages self appointed social media police, using fear of being outed as a weapon, damaging and potentially distressing Wikipedians interested in working collegiately and openly. This is becoming a spooks charter, making it possible to hound open contributors by asking who their friends are, which facebook account they use, which twitter account, which email groups they are part of. This already happens and those doing it should be hammered for attempted outing and harassment, rather than making policies that will be used to protect them from sanctions because they vaguely suspected someone of off-wiki canvassing. -- (talk) 08:37, 3 May 2019 (UTC)
  • Without expressing an opinion here on whether the instance that prompted this was canvassing, it seems obvious that if you take any situation that we do agree meets the threshold of canvassing, via talk pages or e-mail, etc, and replace the method with notifications sent via WP:Notifications, it's still (erm) notification, and canvassing. An addition like the one proposed would be a helpful anti-GAMING clarification, as SMcCandlish says, so I support it.
    Pace Swarm, while I know it's rough for a user to see a change be proposed in response to something they did, I think it's reasonable to expect proposals to arise in response to situations where guidelines are wikilawyered over: when users propose changes to merely 'head off' problems that haven't yet arisen, other users (including me!) routinely question whether change is needed / whether the problem actually happens, and shoot them down—having evidence that a proposal is in response to an actually-occurring issue makes it more worthy of consideration, IMO. -sche (talk) 08:30, 3 May 2019 (UTC)
  • "Pace"? Like, as in the Latin sense? Really? That is almost as insulting as this proposal itself. The key words here are "actually-occurring issue". There is no "actual issue". The purpoted "issue" is that Flyer pinged some users who ended up supporting them in a content dispute. Wanda is attempting to portray that as "canvassing", in an apparent attempt to game the content dispute. When Flyer pointed out that they actually had a perfectly legitimate reason to notify the users in question, Wanda misrepresented their argument as "pinging can't be canvassing", which is not actually Flyer's argument. That is the basis of this proposal, in lieu of actual canvassing prompted by subpar wording. This is a joke. By supporting a policy wording change on this basis, you're legitimizing a fabricated argument to win a content dispute. ~Swarm~ {sting} 08:48, 3 May 2019 (UTC)
Wanda misrepresented their argument as "pinging can't be canvassing", which is not actually Flyer's argument. It wasn't their only argument, but the idea that talk page pinging in general might not be covered under WP:CANVASSING was one of Flyer's arguments, unless I am drastically misunderstanding what they were saying. WanderingWanda (they/them) (t/c) 17:01, 3 May 2019 (UTC)
You seem to be held up on the fact that they said WP:Canvassing [does] not address WP:Pinging, at the expense of every other statement they made explaining that their notifications were appropriate and justifiable. You're apparently interpreting that as an argument that "pinging can't be canvassing". But based on Flyer's subsequent explanations, their argument is more along the lines of "pinging can't be automatically labelled as canvassing, because it's a nuanced issue, just like any other form of notification". Perhaps they didn't initially word it the best possible way, but they've certainly clarified their position by now, and you don't seem to be looking at anything other than cherrypicked statements that look bad. Maybe you're not intending to, but that's just how it looks to me. ~Swarm~ {sting} 19:42, 3 May 2019 (UTC)
Thank you, Swarm. Flyer22 Reborn (talk) 10:49, 4 May 2019 (UTC)
  • Oppose as not solving a well-defined problem. Truth be told, IRC, email and "other channels" would really have to be covered for the CANVASS rules to have any teeth at all. Collect (talk) 12:56, 3 May 2019 (UTC)
  • Oppose – partly, per my colleagues' comments above, because it's always a bad idea for editors to try and change PAGs while they're in the middle of disputes. Moreso, however, because if we were to make a change to this guideline, I would !vote that it be deleted altogether. On a collaborative project such as this, editors should be encouraged, not discouraged, from reaching out to each other and soliciting input. Concerns about votestacking, cabaling, and other such inappropriate canvassing ignore several key points, such as 1) anyone who wants to cabal only needs to follow their friends' contribs or watchlist their friends' talk pages, no actual communication is required to collude on Wikipedia, and 2) off-wiki communication make colluding trivially easy for anyone who wants to do it. As such, prohibiting on-wiki communications will only drive the "bad actors" to use off-wiki means, even assuming they need to communicate at all. TLDR: WP:CANVASS is pointless and counterproductive. Levivich 16:48, 3 May 2019 (UTC)
    I agree with that. The guideline is, effectively, "don't canvass openly where everybody can see it." We can't legislate morality and cheaters are gonna find a way to cheat. But I don't think we're going to reach consensus to delete the guideline in this discussion, which is not about that. ―Mandruss  17:21, 3 May 2019 (UTC)
    No, I don't think so either, but until then, I'd oppose any expansion of it. Levivich 17:24, 3 May 2019 (UTC)
    Even though we can't stop canvassing completely, the guideline has been helpful in a number of cases, including contentious topics such as the Gamergate controversy. It's never a good thing for an editor to go to Reddit, for example, and bring in a bunch of WP:Meatpuppets. We shouldn't think of the "we can't stop canvassing completely" matter as "Oh, well, since we can't stop it completely, we should just allow it or ignore it." I would never support abolishing this guideline. Flyer22 Reborn (talk) 10:49, 4 May 2019 (UTC)
    Oh I don't suggest just allowing it or ignoring it, but we could replace many very complicated PAGs with a single, very simple one: Don't be sneaky. That would be equally effective, or equally ineffective. We belabor precise definitions to the nth degree, as if we're programmable robots, and then criticize people for being tangled up in them, which we call wikilawyering. Every clarification creates another ambiguity which requires clarification, without end until we just give up in frustration and say we can't do any better. The entire approach is wrong, and dysfunctional. ―Mandruss  17:23, 4 May 2019 (UTC)
  • Support per SMcCandlish. The proposed change is logical and clear, regardless of the dispute that generated it. —⁠烏⁠Γ (kaw)  20:57, 04 May 2019 (UTC)
  • Support - the nutshell definition is When notifying other editors of discussions, keep the number of notifications small, keep the message text neutral, and don't preselect recipients according to their established opinions. "Notifying" is any act designed to attract someone's attention to the discussion; pinging is clearly such an action. עוד מישהו Od Mishehu 01:42, 5 May 2019 (UTC)
  • Oppose While using pings to deliberately canvas is of course wrong, if there is any grey area about the notifications an editor intends to make, the method that should be encouraged is pings on the page containing the discussion. The visibility of the ping promotes transparency, and allows any questions of canvasing to be addressed at the time, rather than when someone notices a message located elsewhere later in the discussion. Specifically calling out the use of pings I think would discourage their use in this type of grey area situation. Since the language already covers bad faith canvasing of all types, I think a specific call-out of pings would do more harm than good. Monty845 02:55, 5 May 2019 (UTC)
  • Oppose The statements by Monty, Swarm, Collect, and Levivich already cover the reasons I have for opposing, so no need to repeat them. Schazjmd (talk) 20:10, 5 May 2019 (UTC)
  • Support. I can't see how anyone can argue pinging isn't a way of notifying an editor of an ongoing discussion. Calidum 04:25, 6 May 2019 (UTC)
  • Support. The only reason I might hesitate is because it should be blindingly obvious that it doesn't matter how someone is communicating these message(s). The issue is intent or knowing disregard for disruptive results. Stacking !votes with sockspuppets, or meatpuppets, or biased participation, or a mob of clueless-junk-votes, or turning a discussion into a clusterfk, or anything else that threatens an improper outcome or threatened a confusion-based-no-consensus or threatens to impair our other work is disruptive. There are times and ways that it is appropriate to notify an individual or advertize for general participation, and that is generally addressed by Appropriate Notification. However inappropriate notification is inappropriate no matter how it's done. As an example, if there is an AFD and an editor posts about it off-wiki while the AFD is in progress and knowing that it is likely attract biased or disruptive activity in that AFD, then they are at risk of receiving a warning or sanction for the on-wiki disruption they caused. Unless canvasing is exceptionally egregious, a first instance generally results in a warning and education about the issue. But I definitely support preventative-blocks against repeat offenders, needing a clear statement that they understand the problem and will not repeat it before being unblocked. I note that these instances seem to most often occur in topics covered by discretionary sanctions. Any individual engaging in initial canvassing behavior in such a topic area should be given the relevant discretionary sanction notice, and repeated instances should receive discretionary sanction blocks. Alsee (talk) 23:13, 12 May 2019 (UTC)
  • Oppose. This confuses canvassing with letting interested, or previously involved, parties know about something. The guideline ought not to be interpreted to mean that people who care about something aren't allowed to be told about it. SarahSV (talk) 23:31, 12 May 2019 (UTC)
  • Support but I would not include pinging any editor currently active in a discussion, or all editors who were involved in a previous discussion on the same or similar issue -----Snowded TALK 04:59, 19 May 2019 (UTC)
  • Oppose per WP:COMMONSENSE, WP:CREEP, and Swarm. I agree with the supporters that this is a clarification of existing policy, but Swarm is right to point out that the proposed language is painfully obvious. The addition doesn't seem like it will make a meaningful difference except in the content dispute which motivated it. I'd still likely be opposed if this were a more carefully proposed change not tied to an ongoing content dispute, but I'd at least be more receptive to the support arguments. If such a proposal does come about, please ping me. Wugapodes [thɑk] [ˈkan.ˌʧɹɪbz] 06:01, 19 May 2019 (UTC)

Recent changes

I reverted these recent changes by Gwillhickers. We went over the meaning of canvassing last year: Wikipedia talk:Canvassing/Archive 6#Issue with lead wording, and possible improvements. Flyer22 Reborn (talk) 23:49, 21 July 2019 (UTC)


The prior discussion linked to above evidently did not resolve much of anything, given the contrary and confusing language in the opening statements that still exists on the Canvasing page.
re: This statement: Canvassing is notification done with the intention of influencing the outcome of a discussion in a particular way, and is considered inappropriate.
The statement clearly indicates that canvassing is done for only one reason, and that there is only one general opinion out there regarding it. This statement is also contrary to statements in the Appropriate notification section, a section that exists under the general page title of Canvasing. The statement is also contrary to the

The language needs to be made clear and balanced, and not completely one–sided. The language should reflect why other possible reasons canvasing may be conducted. There is also no clear distinction between notifying several users and canvasing. At what point does notification become canvasing?

Existing language :

In general, it is perfectly acceptable to notify other editors of ongoing discussions, provided that it be done with the intent to improve the quality of the discussion by broadening participation to more fully achieve consensus.

Canvassing is notification done with the intention of influencing the outcome of a discussion in a particular way, and is considered inappropriate. This is because it compromises the normal consensus decision-making process, and therefore is generally considered disruptive behavior.

Proposed changes:

In general, it is perfectly acceptable to notify other editors of ongoing discussions, provided that it be done with the intent to improve the quality of the discussion by broadening participation to more fully achieve consensus.

Canvassing is notification done with the intention of bringing other opinion(s) into the discussion, but it can be considered inappropriate if not conducted properly, as it may compromise the normal consensus decision-making process. -- Gwillhickers (talk) 00:26, 22 July 2019 (UTC)

The previous discussion closed with, "Consensus is that canvassing has a special, solely negative meaning here." In the guideline, appropriate notification is distinguished from canvassing. Flyer22 Reborn (talk) 00:29, 22 July 2019 (UTC)
Should not the page here make the distinction between canvassing and notification in the opening statements? I have seen numerous cases where when editors are notified with a neutral statement, it is referred to as "canvassing". As it is, the page starts off with the title Canvasing, and then goes into what notification is, following with a very narrow opinion about what Canvasing is. Does not the process of canvassing involve notification? How does one Canvas without notifying people? There is no clear opening statement that says canvasing involves inappropriate notification. I found the language in the opening statement contrary and unclear on that distinction, certainly others will/have also. Would someone please make the language more clear in the opening statement? -- Gwillhickers (talk) 00:46, 22 July 2019 (UTC)
You stated, "I have seen numerous cases where when editors are notified with a neutral statement, it is referred to as 'canvassing'." You and me both. Pinging the editors seen in the previous discussion I pointed to: Snuge purveyor, Sphilbrick, Knowledgekid87, isaacl, Godsy, Swarm, Noyster (now editing as Bhunacat10), Pincrete, Patar knight, and Mandruss. I didn't ping Jytdog because he currently no longer edits. I don't know if Sphilbrick was querying something different. And as for pinging, see the most recent discussion on that above on the talk page. Flyer22 Reborn (talk) 01:02, 22 July 2019 (UTC)
  • @Gwillhickers: "Canvassing" is not "notification", but "bad faith notification with the intent to fabricate a consensus". That's what this page is talking about. If you find the wording confusing, we can work on that, but that means better articulating this concept, not redefining the notion fundamentally. ~Swarm~ {sting} 01:12, 22 July 2019 (UTC)
Thanks for your prompt reply. Yes, the page is not exactly clear on the distinction in the opening language. No one is trying to 'redefine' matters, but only to bring clarity in the opening language. As I said, the statement about canvasing is a bit narrow. Who is to say that canvassing is not done with the intention of getting as many neutral opinions as is possible/practical? If anything, the statement on Canvasing should also say, Canvasing involves inappropriate notification. As it is, this sentence makes one general claim that Canvasing involves only one single objective. -- Gwillhickers (talk) 01:27, 22 July 2019 (UTC)

Unlike some issues where there are fundamental disagreements, my impression is that all of the participants in this discussion generally agree on the substance but disagree that the wording properly reflects that substance.

There's a nice graphic (in the Inappropriate notificationon section) that illustrates four key attributes of the notification:

  • scale
  • message
  • audience
  • transparency

The graphic summarizes how they should be handled in an appropriate notification and suggest ways in which they can be abused.

I don't see any disagreements about the substance of that graphic (although perhaps I shouldn't bring it up as I'll invite challenges :).

The problem arises when we try to bring in the word "canvass".

At one extreme, we could use the term to be a synonym for notification and go on to say that some forms of canvassing are acceptable and some forms are not acceptable. At the other extreme, we could say the canvassing is a subset of notifications and exactly correspondence to those instances that are inappropriate. To put it in set theoretic terms, at one extreme set of notifications matches the set of canvassing, and as the other extreme, canvassing is a proper subset of notifications covering only those in exactly those that are inappropriate.

I don't find either extreme acceptable.

According to the graphic, a single message to a single recipient done in a transparent way but worded in a biased manner is inappropriate. I agree but I don't think people typically use the term "canvass" to refer to such a one-off situation. I think most people associate the term with multiple contacts. Nobody says "let's canvass our readers" and means a single question to a single reader. At the other extreme, I don't think we should restrict the word to only meaning inappropriate contact. Picking up on the last example, if someone enters a discussion and says "hey let's canvass our readers", we don't want the response to be that we can't do that because the canvassing is by definition bad.

Having said all this, my concern is that the current language:

Canvassing is notification done with the intention of influencing the outcome of a discussion in a particular way, and is considered inappropriate.

is, at best, awkward English.

Does the phrase "in a particular way" modify the word "influencing", "intention", or "notification done"? Does the closing phrase "is considered inappropriate" mean that all canvassing is inappropriate or only a subset of canvassing that meets certain criteria?

IMO @Gwillhickers: was attempting to clean up the language, but I also agree with @Flyer22 Reborn: that this deserves more discussion. S Philbrick(Talk) 13:07, 22 July 2019 (UTC)

My biggest problem with the language in that sentence has always been the intent requirement. If you intend to provide neutral notification, but fail, you are still canvassing. GMGtalk 13:20, 22 July 2019 (UTC)
  • I think that part of the issue here is that this page is about the concept of "canvassing" in the Wikipedia sense of the word, as opposed to the dictionary definition of the word. This is very much like the way that Wikipedia uses the word "notability" to mean something that is specific for our purposes. And I think that there is nothing wrong with that. But I think that improving/clarifying the wording would be fine. --Tryptofish (talk) 14:20, 22 July 2019 (UTC)
That's where the confusion seems to lie, i.e.that Wikipedia has assigned its own specific idea to the term. Dictionary definitions, however, are more inclusive and are not limited to a single opinion. e.g.Webster's dictionary defines the verb thusly:  1. "detailed examination or discussion";  2. "a scrutiny, esp of votes";  3. "a survey to ascertain a probable vote".  Webster doesn't even hint at the idea that canvassing is done to manipulate an outcome. Our definition of canvasing should be neutral and general and mention that it can be done appropriately or inappropriately. The idea of canvassing is currently presented here at WP in the opening statement as an endeavor that's automatically wrong. Below is a proposal that's neutral and general, where a chart on appropriate and inappropriate notification would immediately follow. -- Gwillhickers (talk) 18:39, 24 July 2019 (UTC)

Proposal

Canvassing is notification conducted with the intention of bringing other opinions into a discussion, but it can be considered inappropriate if not done properly, as it may compromise the normal consensus decision-making process..
Absolutely not, this is yet another perennial attempt to fundamentally redefine the word "canvassing" so that it matches a neutral dictionary definition, which defeats the purpose of the whole damn page. On Wikipedia, "canvassing" is the word we use to a specific, bad-faith behavior. It does not mean "notification". It is an endeavor that's automatically wrong, it's literally an extremely disruptive, bad faith action. If you're confused because you don't understand the context of it's unique meaning on Wikipedia, reading this page should clear it up for you. If you think it's too ambiguous, specify what text could be written to be more clear. But you're attempting to completely eliminate the Wikipedia definition simply because the dictionary definition does not have the same meaning and you don't like that because it confuses you. That's fair, and the article should make the difference clear. But at the same time, users are supposed to have the competence to be able to read this policy and understand the simple concept it's describing. Demanding we redefine the entire guideline just because you think it's confusing and can't figure it out is not reasonable. ~Swarm~ {sting} 03:38, 25 July 2019 (UTC)
Insert : If it's "yet another perennial attempt" it's only because other editors have made the same criticism, that the term canvassing as its used is completely one sided. -- Gwillhickers (talk) 17:04, 27 July 2019 (UTC)
The essential point is that those editors have been consistently outnumbered by editors who disagree. ―Mandruss  17:13, 27 July 2019 (UTC)
"Inappropriate canvassing"? - Ryk72 talk 14:36, 25 July 2019 (UTC)
Not sure what you're getting at? "Inappropriate canvassing"=any canvassing. "Appropriate canvassing"=no such thing. "Inappropriate notification"=canvassing. "Appropriate notification"=notification. Questions? ~Swarm~ {sting} 06:22, 26 July 2019 (UTC)
I should have been clearer that my previous comment was directly in response to specify what text could be written to be more clear. I mean "canvass" in the plain English sense, not the Wikipedia "term of art" sense. That is: "to solicit votes, opinions, or the like". Don't use a word to mean other than its normal sense, use a compound noun with a plain English meaning that matches the Wikipedia meaning. Alternatively, "inappropriate notification". - Ryk72 talk 07:56, 26 July 2019 (UTC)

Recent changes, continued

Swarm, you are clearly over-reacting. No on is attempting to "redefine the entire guideline". Did someone propose that we remove the chart on inappropriate notification?? The proposal above clearly qualifies the the idea of canvassing with "...but it can be considered inappropriate if not done properly". As it is, certain individuals here at WP have redefined the word canvassing. It instead should be neutral, with the chart on appropriate and inappropriate notification following it. WP should not be reinventing the English language. Too many times I've seen various editors sling the term "canvassing" around in cases of appropriate notification. If the statement on canvassing was general and neutral, as it is in the real world, not narrow and opinionated, this wouldn't happen near as much. WP articles are supposed to be neutral. It would seem that our polices and guidelines should be written as such also, esp in the opening language. That's all. Please take it down a notch and lighten up. -- Gwillhickers (talk) 21:58, 26 July 2019 (UTC)

As I said earlier, we do the same with "notable". I see no problem with doing it with "canvassing", although I'm quite receptive to clarification. --Tryptofish (talk) 22:20, 26 July 2019 (UTC)
Just because we can do a thing, doesn't mean we should... "Notability" is a shortcut for something like "meeting Wikipedia's standards of having an article, without considering if the subject is deemed suitable as encyclopedic content for Wikipedia", so there is a clear advantage in having a wikipedia-specific term. The only advantage to using "canvassing" solely in a negative sense is to save using the adjective "inappropriate", which isn't a big savings, so it doesn't seem worthwhile breaking away from the main meaning (as per its Wikipedia article). However, as I said last time this topic was discussed, I suspect changing the language here won't make much difference to how the term is used generally by editors in discussions. isaacl (talk) 04:18, 27 July 2019 (UTC)
We already have WP specific terms: i.e.Appropriate notification  and  inappropriate notification. Yet we have another term that's only been assigned to the latter -- Canvassing. Again, Wikipedia should not be trying to reinvent the wheel where common terms are concerned, and should employ the English language as its used in the real world. When the occasion arises, a given editor should not be accused of "canvassing", but inappropriate notification. Accusing someone of "canvassing" is like accusing someone of running. Running can be appropriate, or inappropriate, depending on where and how it's practiced. As it is, we have two ways of expressing the idea of inappropriate notification: i.e.inappropriate notification and canvassing, yet there is only one way to express the idea of appropriate notification. The opening language needs to be neutral and more clear. The above proposal is clear and neutral, and does not lend itself to a one–sided idea, i.e.inappropriate notification. -- Gwillhickers (talk) 15:08, 27 July 2019 (UTC)
This is getting unhinged. "Appropriate notification" and "inappropriate notification" are not "WP specific terms", that's just plain English. No one actually goes around using those terms. The only "WP specific term" involved here is "canvassing", which is defined as "inappropriate notification", which is a specific banned practice on Wikipedia, the banning of which is literally the entire point of this guideline. This argument that Wikipedia adopting its own terminology for certain things is ridiculous, we have our own terminology for everything, we have our own culture, and part of WP:CIR is being able to participate and comprehend all of this. This concept has been stable the entire time you and I have been editing here, it's never actually posed any sort of a problem, so it's a bit ridiculous that you want to die on this hill. ~Swarm~ {sting} 04:14, 28 July 2019 (UTC)

You're belaboring the idea of what is and isn't WP specific terms. "Appropriate notification" and "inappropriate notification" are used only on Wikipedia. Have you ever been taken to task about 'inappropriate notification' in the real world? Otoh, WP is not a 'cult' – the "WP culture" has articles about the real world, written by real people, with real backgrounds, using the same language and terminology used in the real world. Once again, many editors mis-use the term when referring to cases of appropriate notification. Your claim that "the entire point of this guideline" is banning inappropriate notification, is only half right. The page also explains appropriate notification, and has an entire chart outlining the examples. This all comes under the general title of the page, Canvassing. WP:CIR?? No solid reason has been offered to support the idea that the opening language shouldn't be neutral. WP is constantly changing for the better. All you've really given us is this assertion about WP culture, WP specific terms and a reluctance to make improvements in the opening language. There's no viable reason why we shouldn't define the general statement of Canvassing in neutral terms in the opening language. The above proposal is perfectly neutral and is clear on all points, with charts on Appropriate Notification and Inappropriate Notification that would follow. Why isn't there a "WP specific" term for Appropriate notification? The term Canvassing needs to be a neutral idea, as it involves notification, good and bad, and esp since it's the title of the page in question. -- Gwillhickers (talk) 16:11, 28 July 2019 (UTC)

Sorry, but you’re being boneheaded and pedantic. “Appropriate/inappropriate notification” is not wikispeak, nor some esoteric concept. It’s plain English. “Canvassing” is not used in the plain English sense, but is esoteric wikispeak which is amply explained by this page. I’m not giving you my subjective interpretation, I’m literally explaining the objective meanings of these terms, as has already been affirmed by consensus. If you’re still not understanding, it’s a CIR issue, if you’re understanding but wish to change this interpretation, then you are indeed engaging in a perennial proposal that is not supported by the community. ~Swarm~ {sting} 15:20, 31 July 2019 (UTC)
That you have to resort to personal attacks only emphasis the lack of gravity in your argument. Neutrality is the central theme throughout Wikipedia. The way our policies are written should also be effected this way. Easy math. You have yet to explain why using neutral and clear language, as opposed to your "esoteric wikispeak", is not desirable. Just recently a few editors expressed the idea that this "deserves more discussion"  and  "that improving/clarifying the wording would be fine"  and  "Don't use a word to mean other than its normal sense, use a compound noun with a plain English meaning". Obviously this issue needs to be surveyed again. Wikipedia is always changing for the better. Please lighten up and don't make this a personal issue. -- Gwillhickers (talk) 18:00, 31 July 2019 (UTC)
I'm not "resorting to personal attacks", I'm observing that you seem impossible to reason with.

The s~Swarm~ {sting} 01:21, 4 August 2019 (UTC)

"boneheaded and pedantic" are not a personal attack? Right... I've listened to everything you've said, have even quoted you at times, and have responded, and quite reasonably. Please don't attempt to blur the distinction between disagreement and not listening, and carrying on as if others can't remember past yesterday. Thank you. Refusing to get it, could more easily be ascribed to yourself, someone who scoffs at simple neutrality in the opening language – and for what reasons? e.g.Preference to "esoteric wikispeak"? As of yet, no one has offered any rational reason why the term Canvassing should not used as a neutral term, as it is in real life. With charts on inappropriate and appropriate notification, there is simply no reason why the term Canvassing, the title of the page, should remain one sided, and often misused by many editors when referring to cases of appropriate notification. Thanks for listening so closely. -- Gwillhickers (talk) 00:39, 7 August 2019 (UTC)
Your argument is that we should eliminate the term "canvassing" as it is known on Wikipedia, and that we should redefine it in accordance with the literal dictionary definition of the word. Yet you are simultaneously claiming that "No on is attempting to 'redefine the entire guideline'". You can't have it both ways. You want to eliminate the existing definition. However, this has already been argued, and the consensus has rejected that view. You're not going to erase previous discussions by playing word games, claiming that you're not trying to redefine "canvassing" while directly arguing that we should do away with the WP definition of "canvassing" and replace it with the dictionary definition. You need to be direct and honest about what you're trying to do, for starters, before you can even put forth a credible proposal, and you're really far away from that at the moment. Beyond that, you'll need to present a credible rationale, for doing so, and being accurate to the dictionary definition has already been considered and rejected as a rationale in previous discussions. You're trying to change a longstanding consensus. You need to come up with a good reason for doing so. And "the dictionary" is not an acceptable rationale, because it has already been directly rejected. ~Swarm~ {sting} 06:46, 8 August 2019 (UTC)
We've heard both of your points of view several times now; perhaps some others can weigh in? isaacl (talk) 18:43, 8 August 2019 (UTC)
I'm so glad you said that, thanks. I've said it earlier, but I see nothing wrong with using a Wikipedia-specific meaning of "canvassing" that differs from the dictionary meaning, but I also have no objection to clarifying anything that is unclear. --Tryptofish (talk) 20:11, 8 August 2019 (UTC)
Ummm.... I suggest all of us try not to simply repeat our viewpoints. isaacl (talk) 22:18, 8 August 2019 (UTC)
On that note of invitation, I wanted to say a few things a bit ago: You can change the words on the policy page but you can't change trivially how people refer to it nor can you particularly change the title, which influences strongly the former. So from a practical standpoint, the change suggestion doesn't work. The second point is one mentioned earlier: Some words are just terms-of-art in general, and Wikipedia is not an exception ("notability" and "neutral point of view" for a pair among our more important content PAGs). I think you probably just need to get over that. Trying to change how we use such terms is not a good investment of your time. If you personally want to say to people "that was inappropriate notification because X", I doubt anyone will stop you, but at this point, it's either "move on" or "get out the RFC" (and prepare to be disappointed), given how much argument has gone into the discussion above. --Izno (talk) 02:56, 9 August 2019 (UTC)
Change usually comes slowly. However, It'll never happen if there is no effort to make improvements. A neutral definition and the practice of using the term correctly is long over due it seems. -- Gwillhickers (talk) 22:32, 19 August 2019 (UTC)

Back to the proposal

Given the concerns about neutrality, and given the concerns of several editors who are open to the idea of clarification in the opening language, it would seem that neutrality in our titles and policies is indeed called for. What's a bit troubling is the comment that making changes to Wikipedia is something that just can't happen. Wikipedia is constantly changing for the better. As indicated before, the term Canvassing is often slung around and (ab)used in cases of appropriate notification, just as a couple of other editors have experienced also. If Canvassing was defined in neutral terms in the opening language, all one would have to do is simply refer a given finger-pointer to the general page, where charts explaining appropriate and inappropriate notification would follow the opening language. This would require any given finger-pointer to  explain  his/her accusations, rather than just chanting the word Canvassing. When it comes to how policy and guidelines are defined, the idea of Neutrality should be adhered to in the strictest sense of the idea. Several editors have already expressed the idea that this "deserves more discussion" and "that improving/clarifying the wording would be fine" and "Don't use a word to mean other than its normal sense, use a compound noun with a plain English meaning". For those who may have missed it, and for the sake of any new-comers to the Talk page here, this is the proposal that was offered. -- Gwillhickers (talk) 20:40, 19 August 2019 (UTC)

Canvassing is notification conducted with the intention of bringing other opinions into a discussion, but it can be considered inappropriate if not done properly, as it may compromise the normal consensus decision-making process.