Wikipedia talk:Attribution/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

→

Archive 10

---

Popular Culture (yet again)

Attempting to compartmentalize:

These sources are acceptable:

• In articles about themselves (e.g., Usenet posts may be used as examples on the Usenet article)
• For a plot synopsis.

Are not acceptable:

• For "hard" data, including but not limited to, sales figures and direct quotes.
• Comparisons and judgments on quality (e.g., "X is the greatest game of the 1990s").
• To describe development history.
• When the source is anonymous.

If reliable sources cannot be found for these latter points, information should not be added. Whenever used, dubious sources must be attributed in the text (e.g., "Website X has said..."). Marskell 13:47, 14 October 2006 (UTC)

I feel like you're trying to address a case other than the one we have. I'm really talking more about something like Outpost Gallifrey (Which, until it stopped doing it, was the absolute bees' knees of Doctor Who news), or the Doctor Who Reference Guide (Still the best place for summaries, including summaries of some seriously arcane shit in Doctor Who). These are sites that nobody with any knowledge of the fandom disputes the reliability of it. If you have any knowledge of the subject, you recognize that the sites are every bit as good as a book on the subject would be. (And often better) And, as I've said many times, I have 0 doubt that there are other cases outside of popular culture that require the same sort of exception. The problem is that I just don't know whether the authoritative source on contemporary model trains is a website somewhere. Or whether somewhere out there is a massive and authoritative index of quilting patterns that every quilter worth her salt uses. What I'm looking for is some mechanism whereby we can identify those sources and let them in. My instinct remains to defer to ad hoc consensus of editors, as we should for any editorial decision that is tricky and subject-specific. I'm open to other ideas, but I think it's vital to this page that we have some mechanism to do this.

You are absolutely correct that we need to not use such sources for comparisons and judgments ever, though. Phil Sandifer 14:11, 14 October 2006 (UTC)

I'm glad we agree on the judgments, at least. I'm reminded of the joke that two out of three dentists recommend every brand of toothpaste available when I look at some of our actors and actresses—an awful lot of them are the greatest of their generation, most unappreciated, etc.

I'm not sure about your main point though, because you're approaching this subject-by-subject, which is far outside of the scope of what we can achieve on this page. The above list was meant to suggest content issues that apply across subjects to break the impasse over I want it in/I want it out above. Marskell 14:52, 14 October 2006 (UTC)

I'm trying to do the exact opposite - to acknowledge that there are weird exceptions on the subject level and to let them get hashed out there, just so we don't have rules lawyers engaging in terminal density. Phil Sandifer 15:21, 14 October 2006 (UTC)

Someone mentioned Buffalo nickels. There was a situation with Jefferson nickels; see Wikipedia_talk:Verifiability/archive13#How to handle this case?. Ken Arromdee 15:01, 14 October 2006 (UTC)

That case would not be a problem for this wording of policy because of the part about experts. WAS 4.250 15:38, 14 October 2006 (UTC)

I got confused over Buffalo/Jefferson Nickels, but that was the disscussion I was thinking of. I do think the allowance for experts, whether professional or amateur, will solve this issue. --Birgitte§β ʈ Talk 16:03, 14 October 2006 (UTC)

Perhaps the relevant criteria for the popular culture thing

Perhaps the relevant criteria for the popular culture section is not popular culture, but rather "areas there is little controversy about." All the examples I've thought of that might require these exceptions are deeply placid subjects. Quilting. Model trains. Doctor Who. Fans and afficianados may be passionate, and there may be subjective debates within the area ("No, the sixth Doctor is best!") but there's no real factual disputes. ("Mariner's Compass was Hitler's favorite pattern") And certainly no real controversy with the outside world. ("Model Trains: Harmless Hobby, or Gateway to Hell?") The important thing, to me, is to let such topics about which there is not a bunch of raging controversies just waiting to happen play it a bit looser. Think of it as the opposite of BLP - "Nothing ever goes seriously and dangerously wrong with fly-fishing, so editors in that area should go ahead and use common sense." Phil Sandifer 14:18, 14 October 2006 (UTC)

Phil, what is in the current version that you need an exception to? I know of nothing in it that would be a problem for you. WAS 4.250 14:30, 14 October 2006 (UTC)

I'm not entirely convinced the "recognized expert" quite covers it. Though I admit I can't quite fully articulate my problem, so give me a bit to think about that. Phil Sandifer 14:32, 14 October 2006 (UTC)

I think the policy does give leeway for a group of Dr Who experts or afficionados to accept a website as a source when they agree on its accuracy and quality. I see your concerns as being met by the clause which addresses such circumstances, by the undeprecation of reliable primary sources, by the scope the policy gives to commonsense, expertise, and judgement as final arbiters in assessing sources, and by, in a tight corner, the magic pebble of "ignore all rules". This policy really has become admirably tolerant, in my opinion, which I'm sure is the way to go. qp10qp 18:54, 14 October 2006 (UTC)

Man o man are we on the same page! WAS 4.250 22:57, 14 October 2006 (UTC)

Would the proposed new pop culture clause allow this kind of article, which is based on Usenet posts? If it would, is that a good thing or a bad thing? SlimVirgin ^(talk) 00:52, 15 October 2006 (UTC)

It's a tough one, and one I'm not sure I feel qualified to judge. Within mathematics as a whole, I would not consider such self-published material OK, because there are a large number of cranks and lunatics. On the other hand, if we narrow to fractal mathematics, and particularly the Mandelbrot set, it may well be an area of hobbyists where such concerns don't apply. Also, I'm not sure the new PC clause is the most relevant one - is the creator of Buddhabrot a recognized expert? Phil Sandifer 14:41, 15 October 2006 (UTC)

Replace "Model Trains" with "RPGs/D&D" and you have something controversial. ColourBurst 02:50, 15 October 2006 (UTC)

No you don't. The sources you can use for an anti-D&D article do not offer a substantial or important POV at this point. The 80s controversy should be reported, but it's a historical phenomenon. Phil Sandifer 14:40, 15 October 2006 (UTC)

Expert self-published

"When the credibility of a source is unquestionably established by experts (in their area of expertise) then other things that sometimes indicate a lack of credibility (such as being self published) may not apply." I didn't understand what this was suggesting at all--self-published sources that other experts have looked at? Marskell 16:18, 14 October 2006 (UTC)

I've since fixed it to clarify that we are talking about materials vouched for by the community of experts in a field and not just anybody claiming to be an expert. Hope that helps. Also note the "unquestioned" part. Provide meaningful doubt and it is no longer unquestioned. Creationists doubt evolution. Most peope doubt Scientologists. Detailing criteria so the clueless can tell one from the other is worse than pointless for reasons expressed elsewhere on this page. WAS 4.250 16:47, 14 October 2006 (UTC)

But in what sense do a community of experts vouch for a self-published work? Comments on the dust jacket? Marskell 17:00, 14 October 2006 (UTC)

Independently published reviews or recommendations? By making frequent citations? JulesH 17:52, 14 October 2006 (UTC)

If someone has independently published reviews or citations in secondary sources, then the self-published sourced wouldn't need to be used, yes? Perhaps it's the slightly tortured syntax of the sentence, but I don't see how this was an improvement on what was there. Marskell 17:57, 14 October 2006 (UTC)

(<---) It is a matter of common sense by editors making these choices and mutual material use within an expert community and acceptance of that community by societ in general. For example, society accepts some organizations expertise in maters of coin knowledge and many of those use some website for up to date data on coin value. Society vouches for that organization. That organization by its use vouches for that website for its price data. The flow of credibility is like that. The exact method of citing it is whatever makes the most sense in a given situation. Police use a chain of custody concept. Think of this as a chain of credibility concept where society as a whole is the ultimate fount of assigning credibility and each assignee can in turn pass it on to a degree. If the physics community agrees Dr. Smith is an expert on black holes and he has a blog that says something about black hole then we can say "Dr. Smith in his blog says ... about black holes." WAS 4.250 18:21, 14 October 2006 (UTC)

The definiton of reliable source currently reads "Reliable sources are credible published materials" (my emphaisis). The "Credibility established by the expert community" paragraph currently begins "When the credibility of a source...." So as it currently reads, the credibility of a particular publication must be established by the expert community, even if it is written by a recognized expert. So for example, if I want to quote from the Film Developing Cookbook by Anchell and Troup, published bo Focal Press, I can't use the corrections to formulas posted on Troup's website, because it's self-published and I can't find a formal review of the website.

An important use of this paragraph is to allow the reputation of an expert to carry over from the expert's publications by reputable publishing houses to self-published works that may not be important enough to be commented on by other experts in places where we can find the commentary. The present wording defeats that purpose. --Gerry Ashton 19:44, 14 October 2006 (UTC)

If you read my explanation above about "chain of credibility" you will see that was not my purpose. None the less as it apparently is too easy to misinterpret, obviously other wording is needed. WAS 4.250 22:55, 14 October 2006 (UTC)

The way the policy is worded now, a source is only a publication, not a person or organization. Since the wording that I changed referred to a source having credibility, not a person, that exact source would have to be recognized by the community of experts. That is too narrow which is why I changed it. --Gerry Ashton 23:26, 14 October 2006 (UTC)

Bulaluh. I need to dunk my head somewhere before commenting at too great length. I see above no argument as to why suggested was better than previous. Jossi reverted, so the dog can sleep. Marskell 21:40, 14 October 2006 (UTC)

The best sources

From "Key principles"

In general, the most reliable sources are books and journals published by universities; mainstream newspapers; and magazines and journals published by known publishing houses.

I think this sentence needs improving. Specifically:

1. Most university presses are no longer affiliated with the university for which they are named. I would assume that books and journals published by Princeton University Press are considered A-list sources?
2. What about academic presses other than university presses?
3. Don't we need to distinguish between
4. What about books published by known publishing houses other than university presses? Are these truly lower quality than magazines published by the same house or a house with the same reputation?

Accordingly, I have replaced it with the following:

In general, the most reliable sources are peer-revised journals and books published by specialized publishers, such as university presses, for an audience of professionals or academics. Next are books, mainstream newspapers and periodicals published by other recognized publishing houses.

It is half again as long, but I think that clarity is important, and that we have stated elsewhere that peer-reviewed publications are generally to be taken in preference to newspaper articles, particularly when specialized topics are considered. Robert A.West (Talk) 17:45, 14 October 2006 (UTC)

Looks good to me. JulesH 17:56, 14 October 2006 (UTC)

I restored the previous version, as I am unhappy about the suggested wording. In particular the qualifier " for an audience of professionals or academics". That is a too narrow formulation. ≈ jossi ≈ t • @

But, the current wording is overly narrow, because it excludes professional-level textbooks and peer-reviewed journals that are not published by universities. Robert A.West (Talk) 22:02, 14 October 2006 (UTC)

How about just removing the "for an audience..." qualifier? Andrew Levine 01:52, 15 October 2006 (UTC)

Then you get "journals and books published by specialized publishers, such as university presses." But a vanity press is also a specialized publisher, and we don't want to include them. So, how about

In general, the most reliable sources are peer-reviewed journals and books published by academic publishers. Next are books, mainstream newspapers and periodicals published by other recognized publishing houses.

I also replaced "peer-revised" (a term I never heard) by "peer-reviewed". -- Jitse Niesen (talk) 03:14, 15 October 2006 (UTC)

Thanks for catching my typo. Robert A.West (Talk) 12:56, 15 October 2006 (UTC)

I don't think we can simply dismiss this issue with a [1] best [2] next-best [3] there is no more. Reliable Sources are more important to us, I think, than they are to any previously published Encyclopedia because articles are constantly evolving and because there is no other foundation to build a NPOV on. Reliable Sources are frequently discussed and are a cornerstone of our work. Editors accept (and I'm sure the public appreciates) stable references. I believe we need to view "reliable sources" as a spectrum, a continuous spectrum of increasing reliability with a bottom (note stapled to a telephone pole) and a top (consitution of the USA, perhaps). We need to specify the parameters which actually make a source of information reliable so that editors understand it is not simply "this is reliable, that is not", but understand the qualities which make a source reliable. It might be more than can be done in a policy because a policy is a statement of intent. But wherever we do it, we should do it throughly. Terryeo 17:34, 27 October 2006 (UTC)

"Material challenged or likely to be challenged"?

SV, why the change to "Material challenged or likely to be challenged" as the primary criterion for using an inline citation? Is that supposed to sum up "substantial, surprising, or controversial"; if so, I don't think it does. "Challenged" implies that someone will think the information incorrect, and will be vocal enough to say something about it, implying that we need citations only when someone disagrees with what is written. There are many other reasons to have citations than the possibility of somebody thinking something is incorrect (research, ease of verification, sourcing an essay, etc.), and this wording should reflect that. At the very least, I would expect something like "At a minimum, material that may be or has been challenged should be accompanied by an inline citation." Better would be the previous wording (something like "At a minimum, any substantial, surprising, or controversial content should be accompanied by an inline citation.") --Spangineer^es (háblame) 18:03, 14 October 2006 (UTC)

For my part, I think that inline citation is not always appropriate, especially in shorter articles that have around three decent sources, all of which amount to a few pages, and all of which support one another. Accordingly, I object to the word "substantial" in the old version. I think that the current wording describes both actual and reasonble practice. Robert A.West (Talk) 18:24, 14 October 2006 (UTC)

The way it currently reads "Material challenged or likely to be challenged, and quotations, should be accompanied by an inline citation" is better than adding "substantial". A page whose content can be found in every text book on the subject should not have inline links to each substantial claim. WAS 4.250 18:29, 14 October 2006 (UTC)

I also found the words "substantial, surprising, or controversial" were a bit misleading as to best practise. We often need to supply references for points that are none of the above, and often don't need to supply refs for substantial material, unless it's challenged or likely to be. SlimVirgin ^(talk) 01:15, 15 October 2006 (UTC)

Would it be a good idea to state more about appropriate challenging methods? See Hippie and Talk:Hippie for a recent RfC where this became an issue. Durova 03:03, 15 October 2006 (UTC)

No. We don't need to discuss how to communicate with other editors in a policy on content. WAS 4.250 03:59, 15 October 2006 (UTC)

Foreign language titles

Angus, is it not standard to supply the English translation of foreign-language title? [1] SlimVirgin ^(talk) 01:15, 15 October 2006 (UTC)

Slim, is that for titles of sources or titles of foreign books and films? I think we're very inconsistent in the latter (almost all Hong Kong film titles are translated into English while almost all French film is not) and probably just as inconsistent with the former. ColourBurst 02:07, 15 October 2006 (UTC)

The edit in question is talking about titles of sources exclusively. And in the academic and non-fiction-literature worlds, foreign-language sources are almost always cited without translating their titles. So I agree with Angus here. Andrew Levine 02:15, 15 October 2006 (UTC)

I think it was titles of sources that was being discussed. Okay, I'm fine with it if others are. SlimVirgin ^(talk) 02:17, 15 October 2006 (UTC)

It's not uncommon or unreasonable to translate titles which are in non-Latin scripts, but leave languages like French alone. But still, I wouldn't make any requirements here. Perhaps a recommendation, particularly if an authoritative translation is available (as opposed to a multi-lingual Wikipedian coming up with one themselves). --EngineerScotty 02:56, 15 October 2006 (UTC)

Except we don't. Some French film titles are translated, some not; some Greek (as another example) film titles are translated, some not (only romanized). Some of it has to do with whether or not they're exported to English film markets, so I tend to leave the issue alone; but still, we're not consistent on this, but we're also not terribly consistent with other foreign language conventions either, so this isn't really a big issue. ColourBurst 03:24, 15 October 2006 (UTC)

The issue at hand has nothing to do with film titles. The current proposal is about citing sources, and that's what Angus' edit was about. Andrew Levine 03:42, 15 October 2006 (UTC)

Sorry. About reciting sources in a foreign language, what little I've come across suggests that the titles are cited in the original language, even if non-Latin. ColourBurst 05:13, 15 October 2006 (UTC)

I sometimes translated book titles e.g. here[2] Andries 11:57, 15 October 2006 (UTC)

Bear in mind that titles written in non-European scripts might not display on all computers. I would suggest that a transliteration into the Roman alphabet at the very least would be advisable, otherwise readers might not be able to get enough information to even identify the source. JulesH 09:21, 16 October 2006 (UTC)

Film titles should be given in English where they have been translated and in the original language where they have not. This is so obvious as barely to need discussing, surely? This is not a Cantonese-language encyclopaedia and it should not be expected to become one. We absolutely should not offer novel translations though, although transliterations should be okay. -- GN.

Reliable sources and authors

"Reliable sources are credible published materials whose authors are regarded as trustworthy or authoritative in relation to the subject at hand" is incorrect because there are reliable sources with authors that don't fit that description. The King James Bible for example. WAS 4.250 04:22, 15 October 2006 (UTC)

The authors of that, whoever they were, are regarded as authoritative concerning what they said, but not about anything else that I can think of. SlimVirgin ^(talk) 04:26, 15 October 2006 (UTC)

That the KJV is reliable is certainly debatable, as many people have undertaken new translations because they disagree with your statement.--Birgitte§β ʈ Talk 04:28, 15 October 2006 (UTC)

Taking the question seriously, I would think the authors would be considered the group of people who assembled and translated the KVG. Similar cases would be volumes which have one or more editors instead of an overall author, or publications that identify only a publisher, in which case the publisher is also the author. --Gerry Ashton 04:34, 15 October 2006 (UTC)

The King James Bible is a reliable source only for the King James Bible. There are many translations of the bible, and there is apocrypha. None of these texts can be considered reliable sources for anything else than these sources themselves. ≈ jossi ≈ t • @ 04:35, 15 October 2006 (UTC)

I think the reliability of the King James Bible is beside the point. Those of us who have read a few style manuals and written a few kilograms of university papers probably understand that author is often used as shorthand for individual human author, corporate author, or editor. The question is will a person reading this policy see the word author in the reliable sources section and think it only means individual human author?

Good point. What about?: "Reliable sources are credible published materials whose authors are regarded as trustworthy or authoritative persons, in relation to the subject at hand" ≈ jossi ≈ t • @ 04:48, 15 October 2006 (UTC)

The KJV does illustrate a point though. If we were writing the 1615 Wikipedia, then the KJV would be an authoritative source on the proper rendering of the scriptures into English. Now it is either a source requiring expert interpration (primary source in the historiographer's sense) or an obsolete source, depending on the use one makes of it. Robert A.West (Talk) 13:22, 15 October 2006 (UTC)

I am with WAS's original point.

The bible needs to be routinely used as a reference in Wikipedia. You do it all the time in history articles to source what historical figures are saying. For example, if John Knox calls Mary I a "Jezebel", you can give the bible reference; and the bible is a perfectly reliable source for what Knox said (while a particular history book or document would be the reference for the fact that he said it). There shouldn't be a problem with quoting sources that aren't in themselves reliable for the truth, so long as they are reliably published (usually there will be no shortage of sources to contradict them, if required). So I don't agree with the formula that authors of sources have to be reliable; it is the publication process that has to be reliable. qp10qp 13:47, 15 October 2006 (UTC)

Quotations from the KJV are fine as they would be from any other primary source. What you are concerned with is using a reliable edition of the source rather than the reliability of the source itself. In the case of the KJV, nearly any edition is reliable, so we tend not to think that way, but that is what we are doing, when we are doing it right. Robert A.West (Talk) 14:24, 15 October 2006 (UTC)

WP:BLP Controversial (positive or negative)

Please don't add positive or negative. It's not in the main BLP article for a reason. It is not all inclusive and will only be used to limit BLP actions, not enhance it. BLP used to be only "negative" and it was changed to "controversial" because people will claim that "true" is neither positive or negative. By adding these extra qualifiers, it logically limits the definition of "controversial" to only "positive" or "negative". Rather than debate people on the point of "positive" or "negative", leave it as broad here as it is in the main article of BLP and we can simply delete the unsourced controversial material regardless of whether someone views it as positive or negative. --Tbeatty 07:44, 15 October 2006 (UTC)

Yes, fair point. SlimVirgin ^(talk) 08:01, 15 October 2006 (UTC)

Legalistic synonym bloat

"It includes unpublished facts, arguments, concepts, ideas, statements, theories, and neologisms."

Unless you think that people are likely to complain "sure, this is an idea, but it's not a concept, so you should allow it", one of the two words "concepts" and "ideas" should be removed.

Okay. SlimVirgin ^(talk) 09:28, 15 October 2006 (UTC)

"Reliable sources are credible published materials whose authors are regarded as trustworthy or authoritative in relation to the subject at hand."

Are there likely to be materials that are regarded as authoritative but are not trustworthy?

Yes, an extremist website is authoritative as a primary source about itself, but is not in any sense trustworthy. SlimVirgin ^(talk) 09:28, 15 October 2006 (UTC)

"The analysis or synthesis of published materials to produce novel interpretations, arguments, or to advance a position are excluded."

I find it difficult to believe anyone's going to say "sure, I'm synthesizing published materials to advance a position, but I'm not making an argument, so that's okay". The need for the "advance a position" part is dubious.

I don't understand your first point. Re your second point: that phrase is essential. All WP articles are syntheses of publihed material; it is when material is put together in a way that serves to advance a position that we have a problem. SlimVirgin ^(talk) 09:28, 15 October 2006 (UTC)

The point is that saying that you aren't allowed to produce arguments and you aren't allowed to advance a position is redundant. Advancing a position requires an argument (unless you're making a completely unsupported statement, in which case the problem is a lack of sources, not OR). There is no need to include the "advance a position" language since "argument" already covers all the cases we're worried about. Ken Arromdee 18:04, 15 October 2006 (UTC)

See also Wikipedia talk:Notability (companies and corporations), where I doubt that we will have to apply the rule to many corporations that are not companies. Ken Arromdee 08:20, 15 October 2006 (UTC)

I don't follow this point. SlimVirgin ^(talk) 09:28, 15 October 2006 (UTC)

I think that the point is that "company" is a superset of "corporation," so we can dispense with the latter word. Since the term, "company" does not bring to mind "non-profit corporation", I think that Ken Arromdee is wrong on this point. Robert A.West (Talk) 13:00, 15 October 2006 (UTC)

In that case, there are four categories:

1) Corporations, profit

2) Non-corporations, profit

3) Corporations, non-profit

4) Non-corporations, non-profit.

"Company" includes 1 and 2; "corporation" includes 1 and 3. This still doesn't do what you want since you missed 4.

Wouldn't it be better to just say "(companies)" and put a note in the text saying exactly what's included? Ken Arromdee 18:04, 15 October 2006 (UTC)

SlimVirgin, it was bad form here to split up Ken's post for a point-by-point reply; others can't participate without making it hard to see who the original poster was.

Regarding Ken's question about the difference between trustworthy and authoritative, I'd prefer to avoid using those terms altogether. I guess I'm the only one around here who feels this way, but I'd much rather defer to WP:RS for the definition of reliable sources, and just say here "Reliable sources are sources that meet criteria for reliability, as defined in the WP:RS guidelines." Otherwise, I feel we are going to get into debates over having to define what "credible", "published", "regarded", "trustworthy", and "authoritative" mean for various subjects, as well as arbitrarily elevate guidelines to policy level. People will challenge every one of those words: What counts as "published"? Who does the "regarding"? Where can I find some guidelines to weigh how "credible" "trustworthy" and "authoritative" a source is? Doesn't it depend on the subject area? (We've already posited that it does; e.g., pop culture). Providing those guidelines is what WP:RS should be for, IMHO. I don't see any reason they need to be defined here. Beef up WP:RS if need be.

Regarding the definition of original research and "I'm not making an argument", I would just add that it's not necessary for the argument to be explicitly stated, and that it's the editor's intent that matters. I propose stating it this way:

• "Original research is any analysis, synthesis, or extrapolation of material from any source (reliable or not), excluding simple calculations and logical deductions, with the intent to produce a novel interpretation or advance a position, even if that interpretation or position is not explicitly stated."

I also propose defining original thought:

• "Original thought is any argument, concept, data, statement, theory, or neologism that is not attributable to a reliable source."

Both of these definitions deliberately avoid the term 'published', which I feel is part of the RS criteria. —mjb 17:07, 15 October 2006 (UTC)

I believe one of the intents of this proposed policy is to eliminate the need for WP:RS, so references to it would not be a good idea. I also feel that any definition for purposes of this policy that differ from the general definiton should be flagged as such: "For purposes of this policy, original thought is any argument, concept, data, statement, theory, or neologism that is not attributable to a reliable source." After all, people have original thoughts all the time, and only publish a tiny fraction of them in reliable sources. --Gerry Ashton 17:24, 15 October 2006 (UTC)

It seems to me that all theories involve concepts, so the use of the word "theory" is redundant. Ken Arromdee 01:47, 16 October 2006 (UTC)

Gerry- I am pretty sure SlimVirgin backed off of the proposal to change WP:RS, after it was pointed out that WP:RS's status as a guide to finding and ascertaining the reliability of sources is pretty well entrenched. This proposed policy (WP:ATT) is replacing WP:V and WP:NOR only.

I agree re: "for purposes", although that phrasing has always seemed a little awkward. "In Wikipedia articles" would work as well. —mjb 10:21, 16 October 2006 (UTC)

Using Stormfront as an example

Currently Attribution is worded to say: (the website of Stormfront would be an example of an extremist site) This is probably counter-productive from what is intended, as it inadvertently limits what counts as an extremist site. By listing a well known hate group/site, you inadvertently raise the standard against which all other extremist groups/sites are to be judged. This works to the advantage of most extremist groups, as it gives them a foil. Editors who support including material that originated from extremist groups simply say “the policy is designed to keep out the Ku Klux Klan and the neo-Nazis, neither of which describe group XYZ” even when group XYZ is clearly white nationalist in nature. For example: The main journal of White Nationalist thought in the US is the American Renaissance (magazine) or Amren. Even thought Jared Taylor the founder/publisher of Amren is a close personal friend of Don Black, the founder/ publisher of Stormfront, and much of the material in Amren can be found on Stomfront, Amren’s supporters will always claim the high road and scream loudly if you call Amren an extremist site. (I have already had this discussion more or less on the White nationalism page, apparently it is a Paleoconservative publication, according to some)

In short it would be better to remove the example of Stormfront and leave things more open. By using Stormfront as an example, you have set the benchmark so high that most extremist groups can not reach it.

An alternative would be use one or more professionally maintained lists of extremist groups as a general point of reference. Such lists are not foolproof, nor all inclusive, and not every group listed would necessarily meet everyone’s definition of extremist, however, the only “harm” caused by being list inadvertently is that your group/website can not be cited directly. It simply forces editors to find an acceptable alternative source when citing facts. In my opinion, the “good” (having a vented source of extremist who where off-limits, thus reducing edit wars) of using such a list would outweigh the “harm” (legitimate statements would have to be cited using alternative sources).

Probably the most widely cited list at the moment, IMO only, is the SPLC’s list of hate groups. It covers perhaps a couple hundred groups divided into seven categories, making it a good starting point. If a statement can be traced back to a group listed here, it’s a pretty safe bet that the source should never be cited in Wikipedia. (I have no connection what so ever to the SPLC) Thanks Brimba 09:34, 15 October 2006 (UTC)

Brimba, the recommendation of the SPLC list could perhaps be placed on the FAQ page; the problem with it is that we don't control its contents; not everyone agrees with it; and it covers the U.S. only. What we need on this page is a clear example of a group "widely acknowledged as extremist," and Stormfront is a good example of that. Whatever Amren says about itself is irrelevant; what matters is what other reasonable people think of it. Also, we're not talking only about hate groups, but about extremist groups or people of any kind. SlimVirgin ^(talk) 11:37, 15 October 2006 (UTC)

Due to the context, I read it as saying that some sites can not be trusted to be used a source for anything other than themselves such as gossip columns, promotional sites and extremely biased sites. Which makes sense. Every site has biases that need to be taken into account but only extremely biased sites need to be basicly ruled out. "Extremeism for accuracy is no vice" (to paraphrase Goldwater)? WAS 4.250 14:24, 15 October 2006 (UTC)

I would welcome placing the SPLC list on the FAQ page. For our purposes it is far from a conclusive list, but I never assumed that it would be conclusive. I think using a list of extremists is better than making an example of any one website. I spent a lot of time yesterday thinking of ways to solve a particular problem that I have seen more then once (call it a shortage of reasonable people when you need them), and this was the best solution I could come up with. Thanks again for you time, Brimba 14:56, 15 October 2006 (UTC)

I agree. Let's leave the wording "extremist" without a specific example. These can go in the FAQ, together with the SPLC and other lists. ≈ jossi ≈ t • @ 15:06, 15 October 2006 (UTC)

direct quotation of sources

Forgive me if I'm missing something basic, but I'm surprised to find no mention in the guidelines of when quotation marks should be used when using material word-for-word from another source. Isn't this a major part of attribution? My suspicion is that WP contains a lot of directly quoted material that is properly referenced but that wrongly (even dishonestly) lacks quotation marks. Tony 13:04, 15 October 2006 (UTC)

This proposed policy is about the need to attribute (currently verifiability) and what can be done with attributed material (currently NOR), not about attribution style. For that, see WP:CITE#When you quote someone. Robert A.West (Talk) 13:18, 15 October 2006 (UTC)

I see your distinction, but the link concerns quoting what other people have said, which is a little different in scope from coping slabs of material without attribution. This issue needs to be explicitly dealt with here and at CITE, in my view. There are warnings at the bottom of every edit box, but little advice as to when quote marks are required to avoid charges of plagiarism. Tony 16:03, 15 October 2006 (UTC)

A copyvio, with a reliable source cited, does fulfill the attribution requirement. I think your concern would be addressed if we clarify that this policy must not be interpreted in isolation from not just WP:NPOV, but also WP:C. —mjb 17:16, 15 October 2006 (UTC)

Well, I have seen the following runaround on at least two articles, with editor A claiming academic credentials.

1. Editor A copies a paragraph from source X, which he cites, but he does not make it explicit that he is quoting.
2. Editor B objects on the grounds of copyvio.
3. Editor A defends on the grounds of copyvio that the text lifted is less than 500 words (or some other magical number gotten from an academic handbook).
4. Editor B objects on the grounds of plagiarism.
5. Editor A defends on the grounds of plagiarism that the material is sourced, and that since Wikipedia is a group project disclaims original thought, it logically cannot be guilty of plagiarism.
6. Editor B objects that this is not how Wikipedia does or should do things, regardless of Wikilawyering.
7. Editor A calmly claims that Editor B is obviously biassed and is violating WP:POINT.
8. AGF, NPA are quoted and an edit war continues.

That said, I don't think this proposal is the place to address it. Perhaps it could be made clearer in WP:CITE or the MOS? Robert A.West (Talk) 17:31, 15 October 2006 (UTC)

It sounds like we are in agreement that it shouldn't be directly addressed here. Here in WP:ATT we're putting forth various shoulds and musts to encourage the use of citable material, and to actually cite it. We already point to WP:CITE (how to cite), which ideally should tell people how to go about citing lifted text. If we also point to WP:C, which tells people what text is OK to lift, then I think we'd have all bases covered, no? —Editor B, mjb 10:09, 16 October 2006 (UTC)

"Word - of - mouth" material is likely to become more and more used as broadband increases and subsequently, video and audio become more commonly used on the internet. It needs to be addressed. But Policy is probably not be the best place to address the rules which need to be deliniated about that. Policy is a statement of intent. It is beyond the scope of policy to spell out such specific rules, that belongs in guidelines, I believe. Not to step on your intent, Robert and some mention of that specific kind of reference might find a link in policy. Terryeo 17:43, 27 October 2006 (UTC)

What is not original research

The phrasing, "Examples would be" is clumsy, so I tried rephrasing. I believe this version expresses the same intent, but more smoothly.

I note that Jossi reverted my note about being careful when doing calculations. I think some sort of caution is needed, because it is really easy to introduce new assumptions without realizing it.

A question: should we specially note the exception for original photographs and graphics? As I understand, NOR is interpreted loosely in such cases because of the difficulty of finding appropriate public-domain or GDFL-licensable illustrations? I have seen editors create a graph because of the complexity of fair use with respect to using the copyrighted version, and then worry about NOR. While I commend their concern, I think there is no objection to a graphic that is really just a good illustration of what the source is saying, and I think it improves Wikipedia to have it. Robert A.West (Talk) 13:46, 15 October 2006 (UTC)

Proposed third paragraph in the section:

Unpublished photographs may be included as images in articles provided that there are no published photographs of similar quality that can be used under the GFDL and that the accuracy of the photograph can be readily verified, as by comparison to copyrighted, published sources. Similarly, original graphs of numerical data may be produced, provided that the underlying data is from a published source, the source is cited, and the graphic introduces no interpretation not already contained in that source.

While I can't imagine this being controversial, I have recently been reverted for another edit I thought non-controversial, so I will look for consensus. Robert A.West (Talk) 14:04, 15 October 2006 (UTC)

There is a wide variety of opinion on Wikipedia regarding NOR as regards images. For example, a picture of a thing on a page used to represent that thing (blue jay, centipede, RAM chip). What is acceptable in determining if the picture represents that thing or not? We have people going to botanical gardens and photographing plants and bird watchers photoing birds in the wild and tech buffs photoing computer parts, OR? I don't think the needed conversations have even begun in earnest on the subject much less reached a consensus. I suggest we don't mention images at all in this policy, rather than prematurely make policy on something without any real concensus by wikipedians as a whole. WAS 4.250 14:40, 15 October 2006 (UTC)

I am confused. Is not WP:NOR#Original images policy now? My proposal says essentially the same thing, albeit in fewer words. Robert A.West (Talk) 15:16, 15 October 2006 (UTC)

WP:NOR#Original images is more essay than policy. It is exactly what we want to leave behind when we replace the ramblings we have now with something both nondogmatic and more to the point. WAS 4.250 17:21, 15 October 2006 (UTC)

So that part of the policy page is not policy? It has been there for a while, and looks like policy to me. I have cited it when someone produced an original graph to stand in for the actual graph that could be copyvio if included. I have cited it when considering whether to delete images from a private collection were being used in place of similar copyrighted images. Given that the current policy page addresses it, and given that GA and FA both encourage images, I don't think we can or should duck it. Fear of instruction creep aside, is there any objection? Robert A.West (Talk) 19:54, 15 October 2006 (UTC)

An "original" graph created to stand-in for a copyrighted graph is probaly a copyright violation regardless. As it is a derivative of the coyrighted graph. In such a case the source graph is either not a creative effort and therefore uncopyrightable, or if does pass the threshold for being copyrighted any derivative work would be a copyright violation. --Birgitte§β ʈ Talk 00:11, 16 October 2006 (UTC)

Copyright for an image is in the form of presentation, not the facts presented. The creative effort exists in the choice of form. To be original, the form of the new graph would need to be materially different from the graph in the journal, but a different graph produced from the same underlying published tabular data need not be a derivative work. At least, that is the advice my counsel gave me in a similar instance. Robert A.West (Talk) 00:24, 16 October 2006 (UTC)

So the question is if a graph produced from the same underlying published tabular data as a published graph is original research. I would think not. As long as it encompasses the complete data that was published it could be sourced to the published version. It would be inappropriate to make a graph with a selection of the data from the published graph however. --Birgitte§β ʈ Talk 00:56, 16 October 2006 (UTC)

Agreed. I can think of some exceptions, but they would be rare, and done for clarity of presentation, not to prove a point. IAR may cover those cases. The question is, should we include a paragraph on this point in the proposal? Robert A.West (Talk) 01:13, 16 October 2006 (UTC)

Orignal images of well known objects is "original research"? Enact such a policy, and not only will your source of images dry up, but you'll need to delete most of the current collection. Images are but highly detailed descriptions of things, indistinguishable from a bunch of text (information content wise). "A picture is worth a thousand words", etc. Looking over this current proposal though, I get the impression that like the current policy, the "image exemption" will flow naturally out of it. It it can be shown not to be the case, then there is a bug that needs to be fixed, or this policy needs to die in committee. mdf 20:15, 17 October 2006 (UTC)

What about material that is verifiable but not attributable?

I like that people are putting a lot of energy into making content policies clearer, as this is an area that newbies struggle with a lot. I like the desire to aid both the sincere newbies and established editors resisting POV edits. But one thing troubles me: what about material that is verifiable but not attributable?

It seems to me that a major difference between Wikipedia and traditional academic efforts is that at least for popular articles, our reader provide a vast fact-checking appratus of their own. For example when I look at San Francisco or Hearts (game) I see a lot of useful, unattributed information that I can personally verify to some extent.

Reading the current version, I gather content like this would be allowed to remain as long as it weren't challenged, but I'm a little worried that some well-meaning but rules-focused editor would go on a tear and prune back the Hearts article to the three sentences he could verify in his Pocket Guide to Card Games. Although I deeply appreciate the iron-sharpening-iron benefits of these policies on points where balance and factual accuracy are crucial, I worry that they might force us into a disservice to readers elsewhere. William Pietri 17:17, 15 October 2006 (UTC)

William Pietre wrote "I see a lot of useful, unattributed information that I can personally verify to some extent." I wonder what he means by "personally verify"? Does he mean rely on his own experience playing hearts? Unfortunately there are topics other than hearts were people might exaggerate or even misrepresent their personal experience.

One area I could see attribution through something other than a traditional publication is commonly available objects. If I want to write that a North American 120 V 15A grounded power plug has 3 prongs, I don't see any problem. These objects are far more available than any official specification would be. --Gerry Ashton 17:33, 15 October 2006 (UTC)

Wrong. I am in Europe. I cannot get a 3 prongs grounded power plug, all those i see around me have 2 prongs. Please see Worldwide_view. On the other hand source giving a specification on a department of energy or ISO website would make this claim verifiable for me.-- ExpImp^talkcon 14:44, 30 October 2006 (UTC)

It is not a requirement that every Wikipedia reader have equally convenient access to references. Such a requirement would destroy Wikipedia, because many readers are children who have convenient access only to an Internet connection and a small school library. Furthermore, I'll bet it would be cheaper for you to buy a plug from a North American online electrical supplier and pay the international postage than to by a standard from ISO; only a few of their standards are free. --Gerry Ashton 18:18, 30 October 2006 (UTC)

All claims in Wikipedia articles must answer the credibility issue ("Who says so and why should I believe them") in a way understandable to the average English language speaker. For example "Everything on this page can be found in any physics testbook on the subject" at the bottom of a short article on a physics particle. Or "This article is from the 1911 Encyclopedia Britannica" at the bottom of such an article. The use of a game or DVD as a primary source in wikipedia is under debate at the moment and the solution is to tell the reader whatever you can concerning the credibility of the article and let's let best practices evolve without trying to prescibe what is best here and now. WAS 4.250 17:39, 15 October 2006 (UTC)

We do well to remember that sometimes to IAR is best. There was a case of a Pennsylvanis covered bridge that was descibed by a written source as unused except for storage, yet can be seen from a nearby highway (it is now private) as having cars path through it. Sourcing objects "anyone" can see with their own eyes is something difficult to put in policy except to descibe it as original research and rule against it. Yet original research was properly used to delete a properly sourced claim about a bridge. WAS 4.250 17:53, 15 October 2006 (UTC)

Yes, that's exactly the kind of case I'm thinking about. Or looking at San Francisco I note that it doesn't cite definitions for the neighborhoods, or give proof that Lombard Street is "famously crooked". It's possible that those claims aren't sourceable, but any San Francisco resident can verify them. If I have to haul out IAR to get away with that, I happily will, but as long as we're looking at replacing WP:V, I was hoping that we could come up with an R that needed less I. William Pietri 17:59, 15 October 2006 (UTC)

Yes. We are doing that. The solution is not a bunch of ad-hoc exceptions, though. Being nondogmatic, careful to not assume too much, and short and to the point is what we want and pretty much what we have. It looks to be very much a policy proposal that is "an R that needs less I". WAS 4.250 18:18, 15 October 2006 (UTC)

I don't agree with WAS 4.250 that relying on objects anyone can see with their own eyes is necessarily original research. At least in the case of manufactured goods that are sold widely, they have similar characteristics to books: they are all the same, and they are at least as easy to buy as books. They are marked with a brand, so their origin can be traced. Really, they have most of the characteristics of publications. --Gerry Ashton 18:27, 15 October 2006 (UTC)

If all we have to say about an object is what can be gleaned by looking at one, then I would say that our article isn't very useful. An article on a three-pronged plug should discuss the evolution and usage of the device, which involves an understanding of electrical power and grounding techniques. While hundreds of Wikipedians could competently describe that in their sleep, we still should be able to find a source if demanded. For one thing, we all carry a lot of misinformation, even about our own fields. As for the appearance, what is the specification? What range of shapes and sizes are in use? That information is a pain to dig up, but that is the task we set ourselves when we joined this nuthouse. Robert A.West (Talk) 18:47, 15 October 2006 (UTC)

I'd agree with all of that, but I don't think it quite addresses the point that concerns me. A relatively small portion of the world population can verify that "The Avenues" refers to a particular part of San Francisco. That fact may not be in any reliable source. But enough of the people who do know edit Wikipedia that I'm not worried about that being wrong. However I am worried, and possibly needlessly, that a shift in focus from verifiability to attributability would make certain articles worse even though I could well believe it will make a lot of articles better. William Pietri 19:09, 15 October 2006 (UTC)

The policy should be framed not just to identify sources that are to underpin large portions of a new article, but also to identify sources that are suitable for writing one sentence in a large article, or to correct a minor error in an existing article. West argues "If all we have to say about an object is what can be gleaned by looking at one, then I would say that our article isn't very useful." That may be so, but the article may not be about the object; the object may only be mentioned in passing. --Gerry Ashton 20:33, 15 October 2006 (UTC)

As for the covered bridge point, I have deleted things I know to be mistakes, such as saying that a building is in the wrong municipality. If compelled, I could dig up a source, but it usually suffices to say, "I live near there; that's wrong." One editor calls this "zero order verification," and it is a common practice. In the case of the covered bridge, odds are that it is on the local register of historic places, which would have records of its current usage. Shortcuts are just that, and should not be enshrined in policy. Robert A.West (Talk) 18:57, 15 October 2006 (UTC)

Ok. It looks like people feel that my concern is either not a big deal or covered under IAR, so I'll let the hypothetical problem go until it becomes an actual one. Thanks for the feedback. William Pietri 21:56, 15 October 2006 (UTC)

No - you're absolutely correct. Especially since this is one of the major reasons we're redoing the policy. That said, is the existing line "This policy should never be used to cause disruption by prematurely removing material for which reliable sources could reasonably be found." insufficient for your concerns? Phil Sandifer 21:58, 15 October 2006 (UTC)

Unfortunately, yes. I'm saying that there is a class of material that may be unsourceable but that is verifiable by a significant number of editors. Maybe a good term for what I'm thinking of is "folk knowledge". As I mention above, common local knowledge and popular card games are two examples where I think our articles go well beyond the sources I've seen, and beyond what I expect we could get from any source. I think it's related to the pop culture problem above, but different enough that I thought I'd mention it as long as we were on the topic. William Pietri 22:10, 15 October 2006 (UTC)

What if we changed the line to also include a warning against tagging information that does not appear questionable? Phil Sandifer 22:22, 15 October 2006 (UTC)

That would obliterate the distinction between {{fact}} and {{dubious}}. It is legitimate to request sources for significant assertions that you believe. Robert A.West (Talk) 22:31, 15 October 2006 (UTC)

But we ought not consider this grounds for removal or a "challenge" so much as a note that this would be a helpful way to improve the article. Phil Sandifer 23:01, 15 October 2006 (UTC)

Take a real case. An editor posted something that I 99% believed, but sourced it to confidential documents. I strongly felt that the statements should be sourced to publicly-available documents. The reaction was, "I'm not here to do research for you," until I used WP:V as a club. I do not believe that I was being obnoxious, but I strongly felt that the statement should be verifiable. Maybe the 1% doubt just bothered me too much. How much doubt should an editor have (or claim) to justify insisting on a source? If we don't put the burden squarely on the shoulders of the editor adding material, we may never get sources. Robert A.West (Talk) 00:05, 16 October 2006 (UTC)

Given that confidential documents were involved, it sounds like a different case than the one I'm describing, which is more about knowledge common among a sizable group. Perhaps the difference lies in an editor knowledgeable about the subject expressing doubt versus demanding attribution for its own sake? William Pietri 02:58, 16 October 2006 (UTC)

There are several aspects to a Reliable Source of information. Verifiability is one of them and Attributabiliy another one of them. Verifiability, not truth places that quality highest but attributability is important, too. For example, several anonomous websites can be easily verified for their information, but don't have any attributability at all. So neither quality by itself is sufficient. Terryeo 17:49, 27 October 2006 (UTC)

Fact-checking processes with inherent POV?

I think this could be clearer about press releases. Under WP:RS I had always thought of them as self-published sources, but this page pretty clearly excludes them. Could we make it clearer that they are questionable sources, perhaps by naming them? If we don't name them, then perhaps inserting "independent" before "fact-checking process" would do the trick. As an example of why this matters, consider PA_Consulting_Group, an article written by in-house PR people sourced almost entirely from the company's own press releases. On the talk page, involved parties claim that the company has a great fact-checking process before publishing, and so are just fine as sources. However, as somebody who has seen the the press release process from both sides, this strikes me as true but missing the point: even if a press release is full of verified facts (which is certainly not a given), they are ones carefully chosen to put the subjects in the best possible light. William Pietri 17:36, 15 October 2006 (UTC)

This is where the primary/secondary source distinction would be useful. A press release is a primary source, and articles shouldn't be based entirely on primary sources. SlimVirgin ^(talk) 01:08, 16 October 2006 (UTC)

I'm in favor of adding "or conflict of interest" somewhere appropriate. WAS 4.250 17:43, 15 October 2006 (UTC)

Care is needed in placing a phrase like "conflict of interest." A great deal of useful technical information is provided by product manufacturers, who have an inherent confict of interest; they want people to buy lots of the kind of product they make, and they want people to buy their brand rather than competing brands. But if we disallow this type of source, some essential information would be unavailable. --Gerry Ashton 18:16, 15 October 2006 (UTC)

I remember having to rewrite assembly code because the product specs for D/A chips were incorrect - a little too optimistic you might say. WAS 4.250 18:26, 15 October 2006 (UTC)

Was that an error of erroneous documentation, or intentionally misleading documentation (boasting of specs that the manufacturer knew the part could not meet)? No amount of policy on our part, or fact-checking on the part of a publisher, will completely omit error. - unsigned

I'm sure it was a deliberate lie to promote sales. It was off enough to cause problems but not enough to go to a competitor after we had already run all our tests. By advertising a sightly better spec on their product they got our company to use their product in our initial mockups. The chip companies involved were all top notch by the way (Motorola, Intel, you get the idea). But the competition was too fierce for the marketing departments to let the engineers tell the truth. WAS 4.250 22:18, 15 October 2006 (UTC)

Unnecessary. The problem with this article is not its sourcing - it's that it's a vanity hatchet job. We ban those already - no need to adjust this policy to remedy against something we already have means of dealing with. Phil Sandifer 18:27, 15 October 2006 (UTC)

I agree it's not a perfect example, but I still think making it clearer that biased fact-checking does not make something a reliable source would improve this proposed policy. As a practical matter, I passed over pruning that article because I didn't have the energy to argue them back to first principles, but if there had been somethig clear in WP:RS it wouldn't have been as much trouble. I don't know whether or not the help in resisting corporate puffery is worth the clearly significant cost of a few more words here, but it would have made a difference to me. William Pietri 18:52, 15 October 2006 (UTC)

Aren't biassed sources and their use properly covered at WP:NPOV? Robert A.West (Talk) 18:59, 15 October 2006 (UTC)

Yes, they are. I would argue that trying to design policy by exception is a mistake. The current formulation WP:NPOV and WP:ATT gives us a good platform from which to explore scuch exceptions. ≈ jossi ≈ t • @ 19:26, 15 October 2006 (UTC)

Yes, you are both right. I had forgotten all about NPOV when I was talking about "conflict of interest". WAS 4.250 20:27, 15 October 2006 (UTC)

Hmmm... If that's the case, shouldn't we drop the hints and specific examples of biased sources under "questionable sources"? Or could we replace them with a more generic reference to NPOV? I'd rather see press releases treated explicitly since they're so common, but I'm glad to pull people over to NPOV as long as this article doesn't give the impression that fact-checked press releases are ok. William Pietri 20:33, 15 October 2006 (UTC)

The only warning I can find concerns "extremely biased" sources, which are beyond the pale insofar as reliability is concerned. I think it is legitimate to distinguish those from run-of-the-mill bias. Robert A.West (Talk) 21:10, 15 October 2006 (UTC)

I think this is a VERY poor idea. The point of this page should be to sketch out a pretty good idea of what a good source is and how/when to use it. Specific issues like "This article is a POV hackjob of advertising" are NOT something that we can meaningfully guard against on the policy level, and doing so wouldn't help anyway. You're dealing with the malicious and the clueless. We can't legislate them out of existence. Go shoot them yourself. Phil Sandifer 21:37, 15 October 2006 (UTC)

We have a policy for that. It's WP:AFD, and in some cases, WP:CSD. --EngineerScotty 21:41, 15 October 2006 (UTC)

Exactly. Phil Sandifer 21:45, 15 October 2006 (UTC)

Hmmm... I feel like I'm not getting my point across, but I don't think saying things again will improve the situation. Sorry to have troubled you folks. William Pietri 21:54, 15 October 2006 (UTC)

(<----) I added ":See Wikipedia:Neutral point of view for bias issues." to the "Problematic sources" section. I don't know if WP:NPOV currently adequately deals with these concerns, but that is the place for it. WAS 4.250 22:32, 15 October 2006 (UTC)

(edit conflict)

Whoa! AFD and CSD are not content policies, they are procedures for enforcing policy in extreme cases where a Wikipedia article cannot be made policy-conforming. If an article uses an extremely biased source, but other sources are available, the solution is to remove the problematic material, not delete the whole article. Robert A.West (Talk) 22:07, 15 October 2006 (UTC)

In the case of an article that has always been corporate vanity, CSD is the correct place for it. In the case of an editor that works purely to promote themselves, WP:BP is also relevant. Phil Sandifer 22:09, 15 October 2006 (UTC)

The article seems to be used as an example of why we should clarify our policy on self-promoting sources. One reason to clarify is to inform AFD debates. Another is to give ammunition to an editor who wants to make a good-faith effort to fix such an article. Whether this particular article should be deleted is beside the point of this page. Robert A.West (Talk) 22:23, 15 October 2006 (UTC)

But the problem with the article isn't its sourcing. Phil Sandifer 22:30, 15 October 2006 (UTC)

Then why is it here? Robert A.West (Talk) 22:37, 15 October 2006 (UTC)

(<---) Sounds like a game of "Who is on first." WAS 4.250 22:43, 15 October 2006 (UTC)

Wouldn't this whole problem fit under a questionable source as source with "no independent editorial oversight". Yes there are editor and fact chaecking but they are not independent. --Birgitte§β ʈ Talk 01:16, 16 October 2006 (UTC)

Yes, but I think the sentence could be read different ways. That's why I opened by suggesting that we put "independent" before "fact-checking process" as well just to make clear that an internal fact-checking process isn't a substitute for independent editorial oversight. A good press release would have fact-checking and come from a company with a reputation for such, and would lack the other signs mentioned later in that paragraph. But it still would be a questionable source for anything other than the company's views. William Pietri 01:50, 16 October 2006 (UTC)

Primary/secondary

I'm minded to put the primary/secondary source distinction back in. If you look at the discussion above that was needed to decide on a press release, all that needed to be said is that it's a primary source and articles shouldn't rely entirely on primary sources (notwithstanding exceptions that may be made for pop culture).

A revert war was recently stopped by pointing out to someone that he was using a primary source to make an interpretive claim, and that was enough of an explanation. Without that distinction, I suspect we'd be discussing it still. SlimVirgin ^(talk) 01:28, 16 October 2006 (UTC)

I understand why you would want to do this but I believe everyone would be better served if you simply described what you really mean to say (i.e. a source created by primary players as a opposed to a source created by independent parties). I believe the primary/secondary terminology is flawed as all newspapers when commenting on current events can be used as primary sources, and a presidential speech were it talks about historical precedents can be used as a secondary source.--Birgitte§β ʈ Talk 01:41, 16 October 2006 (UTC)

But primary players doesn't necessarily cover it. The Bible isn't a "primary player." I'm unclear of the benefits of avoiding a term that is otherwise widely used, and which is largely understood (although I know our use of it is simplistic). SlimVirgin ^(talk) 01:50, 16 October 2006 (UTC)

The Bible was certainly created by primary players. Unless you are discussing the later translators, but they are still translating a document that was created by primary players. I question that this simplistic definition of the term is widely understood by people not already familiar with the precedents for its use at WP. Especially if you plan to link to primary source, which clearly talks about things that you do not mean to include in this simplistic use. That said I am certainly open-minded to reading some proposed wording using these terms. --Birgitte§β ʈ Talk 01:57, 16 October 2006 (UTC)

But who those "primary players" who created the Bible is much debated, and basically not clearly known. The Bible as a primary source, however, is a clearly understood fact - the Bible says a bunch of things, and various secondary (and further removed) sources interpret it to mean various things as a result. "Players" is a terminology that will mean little to the reader, "sources" is a terminology that is well-established both on and off Wikipedia. Jayjg ^(talk) 03:38, 16 October 2006 (UTC)

I am not trying to promote my own wording of "created by primary players". If someone can come up to with a clearer way to explain this I would be very happy. That said I cannot think of a single theory of authorship that does not include a person I would consider a "primary players", in either the actual events described or the promotion of the narrative as the "truth". If someone can come up with a better wording to describe this please suggest it. The problem with the term primary source is this. If you want to say "primary source" actually describes a type of document, and I dispute that (see previous disscussion above), the term encompasses any document that is written about contemporary events. Most all newspaper articles would be included in this. What you are really wanting to describe is not contempary works but works whose authors are too close the the subject to be trully reliable. I think the problem with these documents really stems from the authorship. And "primary source" includes closeness in time period as well as closeness of authorship--Birgitte§β ʈ Talk 03:55, 16 October 2006 (UTC)

What counts as a primary source depends on context and on the relationship between the researcher and the source. Most articles in the New York Times today constitute secondary source material, except perhaps for some op-eds and letters to the editor which may be written by people directly involved in an event. But in 100 years time, those articles will be regarded as primary source material about the state of the world in October 2006. Similarly, if I'm writing about the Arab-Israeli conflict and I use Benny Morris's books as a source, they're secondary sources; but if I'm writing about Benny Morris, the same material is a primary source.

There's no need for us to go into that degree of detail here. We can make use of the distinction and explain the bare bones of it, and go into more detail on the FAQ page if we want to. SlimVirgin ^(talk) 10:23, 16 October 2006 (UTC)

<<< Let's not re-invent the wheel, shall we? There are established definitions of what is a primary sources and what is a secondary source. One example (from bgsu.edu) [3]

• Primary sources are the "materials on a topic upon which subsequent interpretations or studies are based, anything from firsthand documents such as poems, diaries, court records, and interviews to research results generated by experiments, surveys, ethnographies, and so on." Primary sources are records of events as they are first described, without any interpretation or commentary. They are also sets of data, such as census statistics, which have been tabulated, but not interpreted.
• Secondary sources, on the other hand, offer an analysis or a restatement of primary sources. They often attempt to describe or explain primary sources. Some secondary sources not only analyze primary sources, but use them to argue a contention or to persuade the reader to hold a certain opinion.Examples of secondary sources include: dictionaries, encyclopedias, textbooks, and books and articles that interpret or review research works.

Another example (this one from Berkeley library guide):[4]

• What are Primary Sources? Primary sources enable the researcher to get as close as possible to what actually happened during an historical event or time period. Primary sources were either created during the time period being studied, or were created at a later date by a participant in the events being studied (as in the case of memoirs) and they reflect the individual viewpoint of a participant or observer.
• What Are Secondary Sources? A secondary source is a work that interprets or analyzes an historical event or phenomenon. It is generally at least one step removed from the event. Examples include scholarly or popular books and articles, reference books, and textbooks.

There are also good examples of primary, secondary and tertiary sources in this page from JCU [5] ≈ jossi ≈ t • @ 04:38, 16 October 2006 (UTC)

The JCU article is pretty good, and makes excellent points about the different applications according to the domain of knowledge being discussed. A must read, IMO [6]. ≈ jossi ≈ t • @ 04:42, 16 October 2006 (UTC)

I agree; it's very clear. SlimVirgin ^(talk) 10:27, 16 October 2006 (UTC)

It is not reinventing the wheel to say newspaper articles on current events are primary sources. Really it quite common for historians to use contempary newspapers articles as primary sources. And this is not even dealing with the completely seperate definition of primary source that is used in science. The terms are problamatic, I think we can be more clear--Birgitte§β ʈ Talk 05:02, 16 October 2006 (UTC)

Maybe we are misunderstanding each other. Do you intend for a New York Times article annoucing the results of an election to treated as a "primary source" like the Bible? --Birgitte§β ʈ Talk 05:13, 16 October 2006 (UTC)

No, the election returns themselves would be. HTH. -- GN.

It is clear that whatever the "real" or "correct" meaning of "primary source" is; we do not agree amomg ourselves what that meaning is or what its significance is. Any such term can be used by people with authority to shut others up and that is not a good reason to reintroduce such a term. Please, please do not reintroduce "primary source" or "secondary source". Talk about bias or conflict of interest or lack of perspective or whatever it is that constitutes a direct objection without going through a term we can't even agree on the definition of. WAS 4.250 05:19, 16 October 2006 (UTC)

I think we do agree on the definitions; where we differ, I think, is in the degree of complexity that needs to be introduced. SlimVirgin ^(talk) 10:25, 16 October 2006 (UTC)

I know beyond doubt that there is disagreement on the definitions; but just don't use the terms in policy and then it doesn't matter. WAS 4.250 14:57, 16 October 2006 (UTC)

If you believe you can encapsulate definitions of those three categories in a few words, then go for it; I couldn't begin to define those terms myself because they mean different things to me in different contexts. The thing I'd least like to see would be any restoration of the blanket preference of secondary sources over primary ones.

I see little in common between the historian's notion of a primary source and the idea of a first-person account on a blog or a website, often also on Wikipedia called a primary source. The first sort comes to us via an edited publication whereas the second comes to us raw and unpublished (which is why I prefer the criterion of a distinction between published and unpublished sources rather than between primary and secondary sources).

As for tertiary sources, they are a conceptual will o'the wisp to me: encyclopedias may base their texts on secondary sources but so do many books (popular history books, for a start), and yet we call the latter secondary sources. And some entries in encyclopedias are written by named authors who may have used primary sources in compiling their articles. So in my opinion we don't need the term "tertiary" at all. For all Slim Virgin's promethean work on these policies, I fear the eternal pecking of her liver by the quibble eagles if she attempts this particular semantic feat. qp10qp 15:38, 16 October 2006 (UTC)

I think we are closer than we think. The issue is not so much about stating a preference for secondary sources over primary sources, but about the use of primary sources to deduce and advance a position that is not reported in a secondary source. We can keep the wording simple by explaining what a primary soure and a secondary source is, and how to best use these, as per the JCU article at http://www.library.jcu.edu.au/LibraryGuides/primsrcs.shtml ≈ jossi ≈ t • @ 17:25, 16 October 2006 (UTC)

Popular Culture section removed

So are we saying that popular culture is no longer an exception, or is it in revision? I'm confused because the talk page leads me to think that some people want this in some form, but that doesn't match the page. ColourBurst 04:05, 16 October 2006 (UTC)

I'm confused about that too. SlimVirgin ^(talk) 10:02, 16 October 2006 (UTC)

We are expecting popular culture editors to rely on the writings of experts in the subject area per:

• • Expert self-published sources

When a recognized expert writing in his or her area of expertise, or a well-known professional journalist or commentator, produces self-published material, we can rely on it so long as material produced by the writer would normally be regarded as a reliable source. However, exercise caution: if the material is really worth reporting, someone else is likely to have done so. If there is reasonable doubt about the reliability of the source or the relevance of the material to the subject matter, err on the side of caution and don't use the self-published material.

--Birgitte§β ʈ Talk 04:09, 16 October 2006 (UTC)

Someone removed "professional" from the above, so I've restored it. If we allow anyone recognized by some people as an "expert," we open the floodgates to nonsense. Jack Sarfatti, for example, is a physicist who is regarded by some serious people as a very eminent one, a genius even. But we would not allow his self-published material, which is idiosyncratic, to be used in physics articles. The way we prevent sources like that from being used is with the caveat "professional." Someone has to be paying that expert for his expertise or we don't recognize it. SlimVirgin ^(talk) 10:01, 16 October 2006 (UTC)

Yes. Also:

Reliable sources are credible published materials with a reliable publication process; generally meaning that their authors are regarded as trustworthy or authoritative in relation to the subject at hand. How reliable a source is depends on context. In general, the most reliable sources are books and journals published by universities; mainstream newspapers; and magazines and journals published by known publishing houses. What these have in common is process and approval between document creation and publication. As a rule of thumb, the more people engaged in checking facts, analysing legal issues, and scrutinizing the writing, the more reliable the publication. Material that is self-published is generally not regarded as reliable, but see below for exceptions.

Relying on experts or relying on a reliable publishing processes for credibility while at the same time not being prohibited from use of primary sources for verifyability should be sufficient for anything Wikipedia should actually be standing behind. (Except for the odd item that needs the IAR clause.) WAS 4.250 05:33, 16 October 2006 (UTC)

Unfortunately the only current exception for primary sources is this one:

Material from a problematic source, including a self-published source, may be used as primary source material in articles about that source's author.

If we want to allow primary sources in non-reliable publications (which I think we should, with safeguards) we need to have something additional about it. JulesH 09:34, 16 October 2006 (UTC)

Can you provide an actual example of a real case so we can see what you are talking about? WAS 4.250 14:52, 16 October 2006 (UTC)

I think we have to face the fact that a policy can't cover every angle: articles like Spoo and that one about the funny little maths creature are acceptable under "ignore all rules" and within the scope given for commonsense and the judgement of editors.

I don't support the popular culture exception at all because I believe that very few popular-culture articles would be inhibited by the present text. I don't believe for one moment that the worlds of clock-making or quilting lack published sources such as books and magazines. Most popular-culture activities and artefacts do spawn secondary sources; and those few that don't can be judged on their merits at the level of the article without needing a sanction here that blemishes the simplicity of the proposed policy. qp10qp 16:07, 16 October 2006 (UTC)

I feel that it would be subject to bias (in particular, bias towards forms of popular culture of the United States, where there are more people that care to apply IAR and less people willing to vigorously argue against it.) ColourBurst 16:54, 16 October 2006 (UTC)

Original research def

We currently have:

• Original research refers to material for which no reliable published source can be found

I previously changed this to:

• Original research refers to material for which no reliable published source may be found

but it was changed back. There's a problem with the phrasing "can": it suggests that if an editor searches for a source and can't find one, that constitutes original research, when a source may well exist that they couldn't find. JulesH 09:41, 16 October 2006 (UTC)

Can you say how you're using the word "may" here? "Can" is being used to indicate that no reliable source can be found by anyone who looks, not just by one editor. SlimVirgin ^(talk) 09:55, 16 October 2006 (UTC)

Perhaps "may" isn't a good choice of word. I wasn't happy with it at the time and I'm still not. Basically I'm trying to get the idea across that what we mean is that it isn't possible to find sources, not that a search didn't turn any sources up, both of which are valid interpretations of what we have at the moment. JulesH 17:30, 16 October 2006 (UTC)

I thought that if you synthesized material from primary sources only, it's still original research. Has that changed or should we say no reliable secondary published source? ColourBurst 20:55, 16 October 2006 (UTC)

If you synthesize material from any reliable sources to form a new idea, it will not be possible to find the new idea in any reliable source and it will be original research. I hope Colourburst isn't suggesting that primary sources be prohibited altogether. --Gerry Ashton 21:05, 16 October 2006 (UTC)

Oh no, using primary sources for an article is perfectly fine, though solely using primary sources or if the majority of the information in an article uses a primary source, that's a problem (because it violates the idea that Wikipedia is a tertiary source). ColourBurst 22:02, 16 October 2006 (UTC)

Wikipedia IS a tertiary source. It therefore follows that primary and secondary and tertiary sources fill its basket, so to speak. Terryeo 17:55, 27 October 2006 (UTC)

Yes. "Tertiary source" doesn't mean that the source is necessarily the third to report. It merely means it is neither a primary source, which is able to verify facts directly, nor a secondary source, which is necessarily in contact with the primary ones, but rather one only reporting others and not necessarily having a fact-checking ability, relying on other sources instead. CP/M ^comm |Wikipedia Neutrality Project| 18:31, 27 October 2006 (UTC)

Professional expert

(copied from SV's talk page) It was me that pulled professional out of Wikipedia:Attribution. There was discussion at Wikipedia talk:Attribution#Old popular culture section. I've pulled it out again, because I think it was a good change, but I'm commenting here too to let you know and generate discussion, although it's probably best to have the discussion there rather than here. If you want to put it in again, fair enough, as long as we do discuss it. My feelings are that professional is a hard one to quantify, and I think recognised works. The idea is that where fan research is recognised within the field, I don't see that it should be excluded. The internet has challenged many ideas, and the self published idea is one of them. Fanzines and fan published magazines have long been used as sources, The Comics Journal, for example, is one such reliable source which is also, basically, a fan published magazine. The internet has moved such publications, and potential contributors online, for example the newsarama website, would, years ago, have seen publication. I think we have to work out some issue here, and professional is a hard to define term. I think I said on the talk page, what constitutes professional? Does the blogger with adsense on his blog constitute professional? Anyway, there's a starter. Hiding Talk 10:26, 16 October 2006 (UTC)

If we want to make an exception for pop culture, that's fine, but we can't dilute the definition of a reliable source for other areas. As I wrote above, if we allow anyone recognized by some people as an "expert," we open the floodgates to nonsense. Jack Sarfatti, for example, is a physicist who is regarded by some serious people as a very eminent one, a genius even. But we would not allow his self-published material, which is idiosyncratic, to be used in physics articles. The way we prevent sources like that from being used is with the caveat "professional."

Someone has to be paying that expert for his expertise (and that's not what ads do) or we don't recognize it. SlimVirgin ^(talk) 10:32, 16 October 2006 (UTC)

Why don't we simply include the pop culture exception paragraph (and add some of the examples that Phil gave, such as quilting, clock-making, where the key issue is that professional sources don't exist). SlimVirgin ^(talk) 10:36, 16 October 2006 (UTC)

• I don't see the problem that you're describing, that we weaken the barriers. A recognised expert is something that would have to be consensually agreed upon, and contentious instances wouldn't get through. That's the basis of how we work, that we build by consensus, and I think the guidance we have works to support the idea that you can't call anyone an expert as you fear. I think the addition of the term professional creates an extra barrier that isn't needed, and isn't always applicable, and is capable of misuse.

• Sarfatti's research isn't used in certain articles, not because of his credentials but because of our founding principle, that we write from a neutral point of view. I think the insistence on professional experts may actually violate our founding principle of writing from a neutral point of view in certain areas. For example, Eddie Campbell's opinions are extremely useful to the comics field, and he is a recognised expert on the field, despite the fact that he self-published the main body of his work. The same is true of Dave Sim. Self-publishing should not be the bar to people's opinions, rather the credence to which their views are reported should be the issue.

• The way we prevent people presenting information which unbalances an article by giving undue weight to an opinion is through our founding principle of writing from a neutral point of view. It's not through our reliance on professional experts. It can't be, since Wikipedia is compiled by people who very clearly aren't professional experts. The goal, as I understand it, is to that the "wiki process" is the final authority on content. I think we should be allowed to utilise expert opinion where people are recognised experts in their field.

• The newsarama website is supported by advertising, I fail to see how one differentiates between that and a magazine which is supported by advertising. The internet has changed the forums in which people can publish their views, and I think that we need to recognise that fact. Experts are people whose views are given credence. We shouldn't allow the bias of which views commerce wants to support challenge the idea that we write from a neutral point of view. If people are afforded recognition, then we should represent that recognition. Andy Bleck's work on early comics is widely regarded in comics scholarship. He has given talks at conventions. That he has never received any money for his work seems an arbitrary reason to refuse using his work as a source. Hiding Talk 11:19, 16 October 2006 (UTC)

The merging and streamlining of the policies is not supposed to be a complete rewrite of them. It's one thing to introduce an exception for pop culture, but if you want the criteria to be changed for all subject matter, then it's a major change, and I doubt it would get community support.

It's precisely because we're not written by professional experts, for the most part, that we have to rely on other professionals as sources. If any group within any field can decide that someone is an "expert" within the terms of our policy, even though they've never been paid for their work as experts, we would be radically changing what kinds of sources could be used. SlimVirgin ^(talk) 11:27, 16 October 2006 (UTC)

No. That's what we have the neutral point of view for. Our attribution policy allows people to verify we have presented information correctly. Whether content belongs in the encyclopedia is dictated by the neutral point of view policy, which this should support, but not trump. Writing from a neutral point of view is a foundation issue, and should not be superceded by this policy. Verifiability and no original research were built to support the neutral point of view policy, not to supplant it. If any group within a field can determine who an expert is, we cannot, without breaching point of view, refuse to reflect that. We should, however, present that information, where it is within context, with due weight.

And I don't understand this sentence: "If any group within any field can decide that someone is an "expert" within the terms of our policy, even though they've never been paid for their work as experts, we would be radically changing what kinds of sources could be used." Do we not have featured articles which refute the point being made? Hiding Talk 12:32, 16 October 2006 (UTC)

I don't understand this sentence: "The newsarama website is supported by advertising, I fail to see how one differentiates between that and a magazine which is supported by advertising."

If the newsarama website generates enough advertising to sustain a professional staff of writers and researchers, copy editors, fact-checkers, lawyers, etc, then it probably counts as a reliable source. SlimVirgin ^(talk) 11:32, 16 October 2006 (UTC)

Yes, you do misunderstand it. My point is that self published websites don't meet the reliable sources guidance. There's no exception which says that if someone fact checks their material, he becomes a reliable source. Your point about needing a lawyer seems redundant in areas where people publish on areas other than living people. Hiding Talk 12:32, 16 October 2006 (UTC)

The removal of professional was WAS 4.25's compromise to the roadblock caused by the removal of the popular culture paragragh. I do not believe this policy can show consensus without some kind of compromise on this matter. As the policy stands now, Featured Articles such as Spoo are not acceptable. If "professional expert" is to be insisted on, then you must offer an alternate compromise on this issue.--Birgitte§β ʈ Talk 12:58, 16 October 2006 (UTC)

• Actually, it was me that removed it, BirgitteSB, in this edit. WAS 4.25 edited after me. You do however, have my motives right. Hiding Talk 13:45, 16 October 2006 (UTC)

Sorry I should have read the above more carefully.--Birgitte§β ʈ Talk 14:52, 16 October 2006 (UTC)

Anyone willing to tell lies for financial gain can be a professional. Many "professionals" assured us tobacco smoking was not harmful and "orofessionals" today tell us there is no global warming. Being bought with a salery and being an unquestionably recognized expert are two different things. Soviet Russia was famous for having recognized experts in prison. Issac Newton was not a professional. Einstein wrote his initial papers while not a physics professional. Recognized expert. Being paid is not a good criteria for who the expert is. Its only a criteria for who has a boss that can tell him what to say. WAS 4.250 15:10, 16 October 2006 (UTC)

I changed you last edit to "indisputably". I don't think this changes the meaning but I think "indisputably" is a more familair word--Birgitte§β ʈ Talk 17:01, 16 October 2006 (UTC)

A more up-to-date example of experts who (in some sports) are amateurs are Olympic atheletes. Also, if one finds a self-published website that seems useful, it may be easier to establish that the author has publised in reputable journals than to discover the author's employment status. --Gerry Ashton 19:06, 16 October 2006 (UTC)