Wikipedia talk:Attribution/Archive 6

Archive 1

←

Archive 4

Archive 5

Archive 6

Archive 7

Archive 8

→

Archive 10

Questionable or self-published sources in articles about their authors.

The opening to that subsection in the current proposal, reads:

Material from questionable or self-published sources may be used as primary source material in articles about that source's author only rarely, under exceptional circumstances, when other reliable referencesfor non-disputed events cannot be found elsewhere. The material: must be relevant to the person's or organization's notability;; should not be contentious or unduly self-serving; must not be used to support claims about topics not directly related to the source or about third parties.

Current wording in WP:RS:

Self-published sources, whether published online or as a book orpamphlet, may be used as sources of information about in articles aboutthe writers/publishers of those sources, so long as there is no reasonable doubt who wrote them, and where the material is: relevant to the self-publisher's notability; not contentious; not unduly self-serving or self-aggrandizing; about the subject only and not about third parties or events not directly related to the subject;

As you can see the new wording is a big departure from previous understanding, and it is extreme. With this new wording, for example, we will not be able to quote from company's websites, official press releases, and other literature published by corporations such as annual reports, for example. The wording needs to be tone down to make it closer to the origianl formulation. ≈ jossi ≈ t • @ 03:48, 11 November 2006 (UTC)

• Actually, I don't see that at all. The only change is that we cannot quote from a self-published source if there is a better (i.e. independent) source for the same information. If an independent source exists, why not use it instead? If none exists, how do you figure we can't quote from an official press release, annual report, etc.? Robert A.West (Talk) 04:17, 11 November 2006 (UTC)

• I think the difference is relatively small also. It leaves the exception open, as it was before, but uses language to guide this use to only if other sources are not available. An unreliable or self-published source should be the last resort. Atom 05:07, 11 November 2006 (UTC)

Sometimes it makes sense to quote from a primary source, even if the same information is repeated in a secondary source. If that secondary source added no analysis or other value, why not use a primary source for basic uncontested information? I would only require a secondary source in the event of controversy arising over the material. JulesH 09:26, 11 November 2006 (UTC)

The primary source is unfiltered information, and you act in a primary gathering role (original research). The quality and reliability of the information is suspect, and we should either not use that information, or only in the rarest of circumstances. In other types of articles, we quote reliable and reputable sources, and the reliability of those sources is judged by the ability of that journal, newspaper or publisher to do good research and apply/interpret/express it in an applicable and ethical way. We (wikipedia) take on that role, and thereby jeopardize our reliability and credibility in trying to report and publish original research. The reasons we prohibit and express that we are not publishers of original research are sound. This exception should either not be allowed, or only allowed under rare circumstances where the information is necessary to express the notability of the person involved, and is not controversial or disputed. Atom 11:44, 11 November 2006 (UTC)

As JulesH emphasizes (and others earlier on this Talk page), a primary source is usually a more reliable source about itself than a secondary source. It happens to be for the reason of reliability and verifiability that reliable articles always quote original sources. Wikipedia would jeopardise its reliability and credibility if it neglects this universal rule.

Thus I agree on this point with Jossi. IMO the current proposal is not only not a neccesary consequence of any existing policy, but even in conflict with WP:NPOV.

And please note that here Wiki-lawyering is happening that has become rather unnecessary: it's rather overdone now that we have an improved introduction which emphazises that a source must be reliable for the claim being sourced. From that automatically a rule can be deduced like the one in the current policy. Harald88 12:18, 11 November 2006 (UTC)

I agree - remove "only rarely, under exceptional circumstances." Furthermore, the bit after that is confusing and unclear (It seems to be saying that self-published sources about the author can't be used at all to present the author's POV of a disputed event, which is a catastrophe for BLP). Remember - this isn't just the rule that applies to corporate press releases. It's the rule that applies to an author's blog where he talks about his latest book, or where J. Michael Straczynski talks about Babylon 5, or where the White House releases the text of one of George Bush's speeches. We can't strengthen this rule to ward off corporate vanity at the expense of legitimate contributions, and we shouldn't attempt to state the case more strongly than we mean to. Lots and lots of topics rely heavily on this kind of source - it's not a case of "rare use" at all, and to call it rare is just misleading. Phil Sandifer 16:21, 11 November 2006 (UTC)

Agree. That was exactly the concern that made me raise this issue. ≈ jossi ≈ t • @ 22:43, 11 November 2006 (UTC)

You've convinced me. Would it be overemphasizing things to make two points?

• Editors should avoid using self-published or questionable sources when reliable sources would serve equally well or better.
• If a questionable source is cited by a reliable source, it may be quoted or summarized as any other primary source.

The first reduces the command to a caution. The second covers those few autodidacts whose self-published work is cited by later academic research. Robert A.West (Talk) 16:41, 11 November 2006 (UTC)

Both look very commonsense to me. Phil Sandifer 17:23, 11 November 2006 (UTC)

I think that "Only rarely, under exceptional circumstances" is precise and describes the exception perfectly. "Material from questionable or self-published sources may be used as primary source material in articles about that source's author only rarely, under exceptional circumstances, when other reliable references for non-disputed events cannot be found elsewhere."

By wording it without that, "Material from questionable or self-published sources may be used as primary source material in articles about that source's author, when other reliable references for non-disputed events cannot be found elsewhere." Almost reverses the meaning.

The first one says that it should hardly ever be done. (what we want)

The other almost gives permission to use the exception. Many people (too lazy to find a good reference) will merely quote the blog, or personal web site of the living person that the article is written about and say that they could find no reliable reference elsewhere. (not what we want).

What the editor should think from the policy is: "I can' find a reliable reference, should I use this quote from the person themself? (original research) Hmmm. It isn't really important to the notability, I shoud leave it out. Maybe I should look for a reliable reference to support the event.". Or, "This is important to the notability, and I think this may be the exceptional case for putting the claim since I am pretty sure it is true, and I have looked and can find no reliable source." Or maybe "The subject claims this event happened and was covered by the media, I wonder why I can't find any record of that ever happening. Maybe I should not include it in the article."

Atom 22:35, 11 November 2006 (UTC)

The exceptions are needed, Atom. Imagine a biography of a living person, in which a source makes a highly critical statement. Imagine that then, the person publishes on his website a rebuttal of that criticism, shall not that be included as well for NPOV? In many articles, we have to rely on information provided by the source, if properly attributed (e.g. "He claims", "The company declares ina press release that", etc.) ≈ jossi ≈ t • @ 22:48, 11 November 2006 (UTC)

What is needed is to decouple "questionable sources" from "self-published sources", as not all self-published sources are questionable, and in many situations a self-published source is needed to provide the viewpoint of the author, organization, etc. ≈ jossi ≈ t • @ 22:57, 11 November 2006 (UTC)

Phil Sandifer mentioned that this section applies to a Bush speech released by the white house. It does not. The US government is a reliable publisher, and material from the US governement is not self-published. The same might be said of many corporate sources, so long as the material is released thorough a reliable publication process. --Gerry Ashton 01:09, 12 November 2006 (UTC)

I would disagree that the US Government is not self-publishing when it publishes things like the Congressional Record, Presidential speeches, etc. Regardless, the point stands - the policy that protects against what this is trying to block is the vanity policy. This adds no extra protection, and causes headaches elsewhere. It's exactly the sort of stern, redundant overreach that caused WP:RS to become such a disaster. Phil Sandifer 01:14, 12 November 2006 (UTC)

Title

Attribution sounds like it could be as manual as Wikipedia:Citing sources, or confined only to Wikipedia:Reliable sources. Perhaps "attributability"? What are some other options? —Centrx→talk • 21:16, 11 November 2006 (UTC)

"Attribution" is good because it implies that edits must be attributed to sources rather than being theoretically "attributable". Beit Or 11:17, 12 November 2006 (UTC)

That's not policy. Unsourced contributions are welcome and always have been, and unsourced statements are not and should not be uniformly removed. Information on Wikipedia need only be verifiable, except in the special case of controversial material about living persons. That the information be in fact cited is optimal, but the mere absence of such citation is not a reason to delete half the articles on Wikipedia. Even "attributable" is problematic because it implies that obvious deductions of the kind under section What is not original research are not acceptable. —Centrx→talk • 20:25, 12 November 2006 (UTC)

• Perhaps this page should take the place of either page it supersedes. (Radiant) 11:57, 13 November 2006 (UTC)

Attributability? There's no such word. SlimVirgin ^(talk) 10:32, 14 November 2006 (UTC)

It is a combination of "attributable" and the suffix "-ility", with a well-defined meaning. You can find its widespread use at [1], including in the Stanford Encyclopedia of Philosophy, the IEEE, and the UK Ministry of Defence. —Centrx→talk • 16:58, 14 November 2006 (UTC)

Changes

I don't know what to do about the repeated changes to this proposal, which are either pointless, or which make it inaccurate or unclear. For example,

1. To "Material from questionable or self-published sources may be used as primary source material in articles about that source's author," someone added "only rarely, under exceptional circumstances, when other reliable references for non-disputed events cannot be found elsewhere." (a) This is completely false. If George Bush's blog gives info about George Bush, or John Doe's about John Doe, we can use it so long as the person is notable enough for an article in the first place. (b) What is the difference between "only rarely" and "under exceptional circumstances"? (c) What is the point of saying "other" reliable references "elsewhere"? (d) It's not even clear what it means: it seems to be saying we may only use the self-published source for non-disputed material that no one else has written about, which implies we may NOT use it if there's a dispute. So if the NYT says "Johnny is a killer," and Johnny's blog says: "The NYT is wrong. I'm not a killer," we're not allowed to quote Johnny? Who decided that?
2. What is the point of changing "any edit that is challenged or likely to be challenged" must be sourced, which is already policy and is widely used, to "any edit that is contested or likely to be contested"?
3. We say of OR: "it is legitimate to make simple mathematical calculations or straightforward logical deductions based on properly attributed, relevant data that neither change the significance of that data nor require additional assumptions ..." Then someone added "A purely logical entailment in the mathematical sense is valid." (a) What is the difference between the first sentence and the second? (b) What exactly is a "purely logical entailment," as opposed to any other kind? (c) What is a logical entailment "in the mathematical sense", and are only deductions "in the mathematical sense" allowed? And (d) what exactly is meant by "is valid"?

There is no point in producing a proposal that is unclearly written or internally inconsistent, but what's particularly worrying is that edits are being made to change existing policy (such as point 1(d) above) and it's not clear the writer knows that he's actually changing policy. At some point, people are going to have stop making tweaks for the hell of it. SlimVirgin ^(talk) 04:38, 14 November 2006 (UTC)

There are undoubtedly changes and parts of the article that need improvement, but that does not warrant doing a mass revert on edits which are mostly minor tweaks; and doing so was not going to convince other editors—oh, I see you reverted again, with the insulting edit summary of "not carefully considered". To continue, reverting varied changes wholesale is pointless, is not going to convince anyone that you are right, and is not going to lead to the adoption of this policy; you do not have a monopoly on this page. If you wish to correct the errors you think are in these changes—and you only mention the above three—then you should not revert the many productive changes along with them. To address the specific objections you raise above:

1. Yes, that's correct; this is erroneous and should be removed.
2. Challenge has several different shades of meaning many of which imply hostility. Contested is a much more specific word, but disputed may be superior to both.
3. This may not be optimal, but the first sentence does not properly encompass the full meaning of allowable deductions. A purely logical entailment is one in which there are well-formed premises and well-defined rules for manipulating those premises and well-formed statements derived from those premises. Arithemetic calculations, for example, are allowable because there is a well-defined system of arithmetic which may be independently verified by anyone. This change was partially prompted because of past concerns regarding the inclusion of examples of mathematical concepts—which cannot be legally reproduced verbatim from textbooks and which cannot be altered without following a system of deduction that are not as "straightforward" as arithmetic calculations. This also has applications in philosophy and the natural sciences, and theoretically includes the variations in wording that are common to all articles. Please help come up with a better solution.

—Centrx→talk • 05:10, 14 November 2006 (UTC)

On further consideration of number 1, we should not blanketly allow self-published sources on articles about the author, as this will simply allow an article that presents the author's views. The author's own book is a reliable source with regard to what his ideas are, but it is not necessarily a reliable source with regard to his biographical information, for example. Can you give an example of what remains after this is excluded, other than rarely? —Centrx→talk • 05:24, 14 November 2006 (UTC)

1. If it's erroneous and needs to be removed, why did you revert to it, or add it in the first place?
2. The phrase the policies use is "challenged or likely to be challenged." There is no reason to change it. Disputed and contested mean the same thing as challenged.
3. What does the second sentence say that the first sentence does not? No handwaving, please; be precise.

Centrix, please read the current policies. SlimVirgin ^(talk) 05:32, 14 November 2006 (UTC)

Read the policies for when self-published sources are allowed. This page isn't going to change it.

I don't know what you mean by "The author's own book is a reliable source with regard to what his ideas are, but it is not necessarily a reliable source with regard to his biographical information, for example. Can you give an example of what remains after this is excluded, other than rarely?" SlimVirgin ^(talk) 05:33, 14 November 2006 (UTC)

These were by no means "minor tweaks", Centrx. Established policy concerning self-published sources was changed quite substantially, for example. ≈ jossi ≈ t • @ 06:12, 14 November 2006 (UTC)

I did not make that change, and that doesn't explain the numerous other changes which SlimVirgin simply reverted back to his previous edit. —Centrx→talk • 06:25, 14 November 2006 (UTC)

As for your last comment, Centrx, note that a published autobiography, can be cited for biographical information, if properly attributed. ≈ jossi ≈ t • @ 06:15, 14 November 2006 (UTC)

Its reliability depends on the person writing it; evaluations of that reliability are made by historians corroborated in other sources, which results in having a secondary source that should be used instead of or alongside the autobiography. —Centrx→talk • 06:25, 14 November 2006 (UTC)

And this kind of edit: Discussing reliable sources, we say "their authors are generally regarded as trustworthy ..." This was changed to RS are "authored by persons who are generally regarded as trustworthy ..." Why authored and not written? And if not written by persons, then by what?

These changes are not improvements. SlimVirgin ^(talk) 06:20, 14 November 2006 (UTC)

So change it to "written". That doesn't call for reverting every change made over 4 days. This is not your page. —Centrx→talk • 06:25, 14 November 2006 (UTC)

1. I didn't add it. It was most recently added by User:Atomaton on [2]. I specifically did not revert its removal after seeing it by the above comment.
2. These words do have different meanings; "challenged" is not the appropriate word. You added "challenged or likely to be challenged" 3 weeks ago in this edit. You probably used the word because it was used previously "may be challenged and removed", but these are different contexts. An edit is challenged to demand that it be substantiated; a statement (or other material) is contested or disputed or likely to be contested or disputed even if no one who happens upon Wikipedia ever actually challenges it.
3. "Straightforward", coupled with an example of simply calculating a percentage, does not allow for more complex mathematical examples necessary to demonstrating a topic. The recent addition that "any reader" must be able to understand it is even more restrictive. We cannot simply copy these examples out of a textbook.

Regarding self-published sources, for example, a person's memoirs are likely not a reliable source on the person's life. The statements may not be contentious or self-serving, but they are nevertheless based on that single person's recollections and would need to be corroborated by secondary sources, such that the secondary sources would be required.

Also, please explain what was wrong with all of the other changes made between your edit on November 10 and your edit on November 14.[3]. I think there are some ownership problems when you are just reverting back to your last edit and other people are editing with sheepish edit summaries like "SlimVirgin: is this clearer? You should be discussing this on the talk page anyway as you promised." —Centrx→talk • 06:25, 14 November 2006 (UTC)

There is no problem in citing a memoir, if properly attributed. You are misunderstanding existing policies. If Joe Blow says in his memoir that he can read minds, the article Joe Blow can contain a statement such as" Joe Blow, in his Memoirs of Joe, described his abilities to read minds." That statement is verifiable to that source, and the fact that it may be most probably dubious, not relevant. If Joe Blow is notable, that claim made on his own autobiography, is by all means, encyclopedic. ≈ jossi ≈ t • @ 06:46, 14 November 2006 (UTC)

I've explained above why I reverted, Centrix. As for the self-published thing, the proposal and the existing policies say "The material must be relevant to the person's or organization's notability; should not be contentious or unduly self-serving; and must not be used to support claims about topics not directly related to the source or about third parties." Which of your concerns do these points fail to cover? SlimVirgin ^(talk) 06:52, 14 November 2006 (UTC)

No, you have not explained it. What, for example, is wrong with this edit, or this edit; you have also not addressed the specific reasons given in the edit summaries of [4] and [5], or how [6] is not an accurate representation of policy. You need to stop thinking of this as your proposal; essays dominated by one person do not become policies. —Centrx→talk • 07:01, 14 November 2006 (UTC)

I'll have to answer these individually. The first one [7] misses the point of the edit. We don't necessarily cast doubt on X by attributing inline, but we do cast doubt when it is someone discussing HER OWN birthday. The section cautions people about being careful not to use inline attribution to cast doubt, not refraining from using it altogether. SlimVirgin ^(talk) 07:58, 14 November 2006 (UTC)

Okay. —Centrx→talk • 17:19, 14 November 2006 (UTC)

Second one: [8] it's equivalent in terms of ... not equivalents with similar quality. Different meaning and arguably it wouldn't be an equivalent if it weren't of a similar quality. SlimVirgin ^(talk) 08:00, 14 November 2006 (UTC)

It is an equivalent in attesting to the same information, but it need not be equal in quality and relevance. If a foreign-language source is only slightly better than the English-language source in terms of quality and relevance, should it be used instead of the English-language source? —Centrx→talk • 17:19, 14 November 2006 (UTC)

Third: [9] You changed authors to "authored by persons". Do you feel that "authored by persons" is good writing? And better than "authors"? SlimVirgin ^(talk) 08:01, 14 November 2006 (UTC)

The purpose of the change was to more explicitly link the reliability of authors to the reliability of sources. The exact wording is open to change. —Centrx→talk • 17:19, 14 November 2006 (UTC)

Fourth: [10] Already discussed. You're making changes for the sake of change and ignoring that the current policies use the word "challenged." SlimVirgin ^(talk) 08:03, 14 November 2006 (UTC)

You just added the "challenged" wording 3 weeks ago, read above. If I go correct the wording on WP:V, are you going to change this page or are you going to revert it? Also, this diff includes more changes which you have not addressed. —Centrx→talk • 17:19, 14 November 2006 (UTC)

Fifth: [11] Change for the sake of change again, and your version is repetitive, and doesn't need most of the last sentence. SlimVirgin ^(talk) 08:05, 14 November 2006 (UTC)

No, material about living persons must follow this policy regardless of whether it is in the main biography about the living person. Also, what do you mean when you keep saying "change for the sake of change"? These changes were made for specific reasons to improve the page, not simply to make a change for no other reason. —Centrx→talk • 17:19, 14 November 2006 (UTC)

How about the other changes and the lines of argument above that you seem to have simply dropped? —Centrx→talk • 17:19, 14 November 2006 (UTC)

How about [12] and [13]? I really don't understand why you reverted these. JulesH 10:19, 14 November 2006 (UTC)

I didn't notice them. Thanks for changing them back. SlimVirgin ^(talk) 10:37, 14 November 2006 (UTC)

I would like to see little tinkering from now on; reduction, yes.

And I would say this: there are relatively few good writers on Wikipedia—plenty of good scholars, analysts, researchers, toilers, yes; but not many good writers, as is inevitable—and Slim Virgin is one of them. In my opinion, we should give her her head on the writing; without doubt we'd then end up with a sharp document. Do the arguing here on this page, and allow the best scribes at the wording.

I said from the start that I thought this page should set itself a 1000-word limit. Anything longer invites subclausing and elegant (or inelegant) variation. One idea to a sentence is the best principle. And as few subordinate clauses as possible. qp10qp 14:41, 14 November 2006 (UTC)

For information, we're currently somewhere between 1700 and 1800 words, excluding bits like the contents table, notes & references section, further reading, etc. I suspect getting it all down to 1500 would improve the clarity and make more people likely to read it, but at 1000 words it may well be too dense. JulesH 15:39, 14 November 2006 (UTC)

Possibly. But the way to keep a document concise is to know what it aims to do: if it aims to set out principles, stick to that aim. Examples, exceptions, ifs, buts, and bunions can go in the FAQ. People elaborate the core text because they think the idea is to address as many individual issues as possible. Well-written principles and clear terms dissuade elaboration. qp10qp 16:04, 14 November 2006 (UTC)

Here's an example of the opposite:

Building an encyclopedia requires the use of good editorial judgment and common sense.

Formal rules are not a substitute for good judgment and intellectual honesty. Formal rules cannot always determine whether material is reliable, notable, or relevant to a particular topic. In rare cases, the best source for a particular article may be one that this policy would describe as "questionable." (See below.) Editors are most likely to encounter these exceptions in the areas of popular culture and fiction, where professional sources offer shallow coverage, or none, than for established subjects where professional sources dominate. In these cases, and where there is consensus that the sources in question are accurate, trustworthy, and reasonably free of bias, such a source may be used. If reasonable objections are raised about the use of a questionable source, it should not be used. Questionable sources must never be used to support biographical claims about living persons, or to support fringe theories or marginal positions in the areas of history, politics, current affairs, science, religion, and other academic disciplines.

Horrible, eh? I mean, even the heading is too fat:

Building an encyclopedia requires (patronising, portentous, redundant)...the use of (redundant)... good (redundant, obvious)... editorial judgement and common sense (they amount to the same thing in this context. Since common sense is required in all walks of life, and since the advice that follows is about editing decisions, I'd set that aside); result:

Editorial Judgement

I could go through the above paragraph in the same way, cut it considerably without losing a jot of meaning, and make it clearer in the process. I'd like to see the page more settled before I attempt this type of editing, though. For the time being, it would probably be wasted effort.

qp10qp 16:42, 14 November 2006 (UTC)

I think you're taking it a little too far. The heading should be a statement of a principle, which "Editorial judgement" isn't. I'd agree that the current one is a little too wordy, and perhaps doesn't address the point quite as directly as is desirable. Maybe "Selecting sources requires common sense" would be better. I also think the exceptions and explanations do need to be here, at least the most important ones. JulesH 19:27, 14 November 2006 (UTC)

OK. But "Selecting sources requires common sense" is not only trite but not really a heading that fits this paragraph. The content of the paragraph calls for a title along the lines of "Using questionable sources". Since there is a sectrion titled more or less that further down the page, this material would be better fitted there rather than sit in a part of the page where its heading needs to be a "statement of principle". It's not a statement of principle to describe when exceptions to principle may be allowed; just the opposite.

But my point was not so much to edit this section now as to demonstrate how the length of the policy could be cut by rinsing it of word and sentence pileups of the sort this paragraph is stuffed with. And length, of itself, isn't the issue for me, but the fact that stodgy language obfuscates the core principles. qp10qp 21:08, 14 November 2006 (UTC)

Getting back to the point, "only rarely, under exceptional circumstances, when other reliable references for non-disputed events cannot be found elsewhere. is unclear, redundant, poor writing, and most importantly, contradicts current policy. Jayjg ^(talk) 19:56, 14 November 2006 (UTC)

I agree. And we have rather too many sentences with this comma'd hedging style now. They are tricky to absorb and remember; and I'm afraid they give the impression of pusillanimity. qp10qp 21:23, 14 November 2006 (UTC)

My absentee ballot: Non!

Since User:SlimVirgin isn't accepting any changes to his proposal other than those changes which result in an identical proposal, I think I can confidently cast my ballot based on the original proposal:

Reject.

Sorry. Bi 09:25, 14 November 2006 (UTC)

To be fair, looking through the changes reverted, I actually agree on most points. The return of the linking of Wikipedia:Attribution back to itself strikes me as illogical -- why would you want a page to link to itself? -- and there's a typo that I removed (an extra full stop) that has been reintroduced, but other than that, I think every reversion made is an improvement. The resulting policy is clearer, less wordy, and adds less additional restrictions that aren't present in current policy. JulesH 10:18, 14 November 2006 (UTC)

I agree with Bi that the proposal is currently flawed. I have noted before what makes a good policy - and few of those points have been adopted, I'm afraid. I should have added that ensuring all expressed views are properly considered when drafting the policy is another point, and SlimVirgin in particular has been failing to do that, preferring to railroad her views through. I'm sure that this has meant that many potential contributors to this page have moved away. It also means that, regardless of any claims to the contrary on this page, there is likely to be a lack of buy-in amongst Wikipedians more generally, jguk 12:33, 14 November 2006 (UTC)

• Out of interest - you say that this proposal fails your notes of what makes a good policy (and can I have a link to those notes please?). Do you believe that WP:V, WP:RS and WP:NOR are, by their present wording, good policies? (Radiant) 12:38, 14 November 2006 (UTC)

It's at [14]. I think WP:V comes closest to it (though it is not perfect). WP:NOR fails, as it doesn't give the reader enough to answer "what is encouraged/what should I do?" after reading it, just "what should I not do?", jguk 12:52, 14 November 2006 (UTC)

• Okay. To summarize (and please correct me if I get you wrong), your issues with this page are the following: (1) the name is unclear, which could be solved by having it replace WP:V; (2) it is too verbose, which is a good point; and (3) there is an emphasis on prohibition, e.g. of OR, which I don't really see a solution to. So we could start by trimming this page somewhat to make it more concise and clear, would that help? (Radiant) 13:29, 14 November 2006 (UTC)

Your summary's pretty much correct. I do, however, think that the emphasis on prohibition can be rectified. By all means outline that the effect of the policy is to prohibit certain things (though carefully so as not to confuse), just stress clearly first what editors should be doing.

I'd add that I really see WP:NOR as following on pretty much automatically from WP:V. I've asked before, but haven't had any response apart for from SV, as to what violates WP:NOR but is acceptable to WP:V. It seems simpler all round, just add to the pre-existing WP:V the answer to that (if there is one), and perhaps have a short resumé that WP:V leads to NOR in WP:V, rather than to throw WP:V away entirely for a verbose alternative. If anyone can answer the question, what would violate WP:NOR but not WP:V, please tell me. I think it would be useful in framing the policy, jguk 13:43, 14 November 2006 (UTC)

I've given you examples many times of edits that violate NOR but not V. Here's another: this edit violated NOR but did not V. Yale did a study on antisemitism. They concluded that people most likely to hold certain beliefs about Israel were most likely to hold certain beliefs about Jews (too much money, too many strange habits). I referenced the correlation to a secondary source per WP:NOR. Another editor came along and changed the correlation to the edit I linked to above; that is, he used the study as a primary source, deciding for himself which statements to highlight, and in so doing violated NOR, even though he was able to source what he wrote. This is an example of why it's important to retain the primary/secondary distinction. SlimVirgin ^(talk) 14:09, 14 November 2006 (UTC)

• So that makes NOR a corollary of V, that deals with the synthesis of otherwise verifiable information, into a conclusion that may be considered obvious or common sense by some, but is not in fact documented anyplace else. Is that broadly correct? Another rationale for NOR that I could think of is that if something is obviously OR, there's no point in seeking out sources for it; kind of a shortcut for editors. I'm not saying that it's proper to do it that way, but it seems to me that people do so anyway. (Radiant) 14:25, 14 November 2006 (UTC)

I may be missing something here because I am unable to follow the above. My understanding i that this proposal is not changing policy, rather, merging existing policies of V, NOR an the RS guideline into one simple-to-follow policy. If the attempt as expressed by some editors above is to change policy, this may be not the best place forum to discuss such change. That would be better discussed at the Village Pump. ≈ jossi ≈ t • @ 17:03, 14 November 2006 (UTC)

• I believe this was mentioned on the village pump several weeks ago. At any rate, the goal isn't so much to change things, as it is to reword policy to match what we actually do. (Radiant) 17:10, 14 November 2006 (UTC)

The aim of this proposal - achieved or failing? retrievable?

Having just re-read the text, and considered what is already in WP:V and WP:NOR (as well as the now-sidelined WP:RS), I must say I'm not sure what is actually being achieved by this proposal.

I don't mean this in a bad way - many people have spent many minutes or hours trying to craft it, but it seems more like a discussion page rather than one that is actually suited to being a policy. I say this as, whilst it explains and elaborates on points already contained within WP:V and WP:NOR (which is a useful discussion to have available to people), it is not clear and does not immediately outline what the key points are. Or put it another way, as an exposition of policy, it fails.

Not only that, in terms of concise policy, WP:V does the trick already. The policy is outlined briefly in a box at the top of the page, and the rest of the text is short, but offers a little extra guidance on how the policy should be/is implemented. It also largely covers the essence of what is in WP:RS and WP:NOR. OK, some may argue that it does not cover everything, but WP:V could easily be expanded by a little bit so that it does. So if we are to have a merged WP:V, WP:NOR and WP:RS policy, it's best to start with WP:V as it now stands, and look at what tweaks are necessary to stress the WP:NOR parts.

No doubt some will want more guidance than most - at which point a discussion page along the lines of WP:ATT is helpful. Well-constructed, it would have persuasive advice - but by not being policy, we can deviate from it when it is appropriate to do so. Whatever is written anywhere will never convince those not employing common sense to employ it - but good advice for well-intentioned editors who want it is welcomed, jguk 12:46, 14 November 2006 (UTC)

That's what this is. Using V as a base and adding what was neeed from NOR. What you want would leave us with no content policies at all. SlimVirgin ^(talk) 12:49, 14 November 2006 (UTC)

I really can't see that. It looks completely different, it reads completely different, hence my conclusion that it is different. The wording of WP:V as a policy is still 100 times better than WP:ATT. WP:ATT is much more discursive, and a discursive exposition of the issue may well have much value, but as a discursive exposition, not as the policy itself, jguk 12:55, 14 November 2006 (UTC)

Except for the ones we already have... :-) -- Visviva 12:57, 14 November 2006 (UTC)

Well no, he's just tried to remove key sections of WP:V. Were it up to Jguk, we'd be left with some vague principles and absolutely no way to sort out disputes between editors. SlimVirgin ^(talk) 13:02, 14 November 2006 (UTC)

(edit conflict) I'm not sure what Visviva means, perhaps she could elucidate.

And SV, stop misrepresenting me! I have not tried to remove any concepts from WP:V at all - indeed, I'm suggesting adding one, namely a definition of "reliable sources", which we do not have. WP:V is already based on well-worded, precise principles. And having a principle-based approach to deciding what is and is not ok is necessary on content issues, unless you want to write an extremely long policy on the matter which no-one will read. (I don't recommend this.)

I'm not misrepresenting you. Here is the diff so that people can judge for themselves. You removed four sections. SlimVirgin ^(talk) 14:11, 14 November 2006 (UTC)

I think my disagreement with SlimVirgin boils down to this: I believe that we should have a small number of well-worded principles that tell the reader concisely what they should do. SlimVirgin prefers to have a list of prohibitions setting out things she does not like. My approach to resolving content disputes would be to invite the editor to consider whether his edit really does fit within our inclusion criteria - highlighting, where appropraite, where I feel it fails to. That editor can then address what he has to do to make edits stick in Wikipedia. SlimVirgin prefers a more confrontational approach along the lines of "you can't do that because it is written that you can't". Only that fails to encourage positive behaviours. Especially as the next stage might be for the editor to copy SlimVirgin's approach and get a further rebuke because he has done something further wrong, because he has not been told how to do things right. It's the simplest thing in the world to say, "these are our inclusion criteria, I don't believe your edit meets those criteria because of X, could you address the point". That offers a solution. The "don't do this because it's not allowed" offers no solution.

The other side effect of SlimVirgin's approach is to have much, much longer policies. This isn't a good idea. Most Wikipedians never read beyond the policy title anyway. (What does "attribution" mean as a policy anyway?) Very few will ever read more than a few paragraphs. The solution to me is to have a content policy that has a title that is self-explanatory, and only a few paragraphs of content. We just have then to make sure that those few paragraphs are really, really good - but that's ok. jguk 13:15, 14 November 2006 (UTC)

I believe the aim of merging V, NOR and RS is a good one, especially given the recent controversy around RS. (Radiant) 13:08, 14 November 2006 (UTC)

I agree.

This should be done by taking the best of those pages, WP:V, and tweaking it to merge the other ideas into it. Plus by making sure we keep things simple and easy to follow (WP:V largely does this, WP:NOR much less so, WP:RS, well, I think you said it yourself, Radiant:) ).

I have proposed what I think is a much neater, shorter, merging of the underlying concept of WP:RS with WP:V. It is on Wikipedia:Verifiability/temp. Your thoughts on the approach would be most welcome, jguk 13:18, 14 November 2006 (UTC)

This was Jguk's last attempt to rewrite policy: Wikipedia:Verifiability/Jguk's version. SlimVirgin ^(talk) 13:22, 14 November 2006 (UTC)

Jguk, I have respect for your abilities. But please use them here to help make the Attribution page more precise (the paragraphs are plainly too long in the top half). Help trim with a penknife, though; leave the machete aside for the moment.

By the way, I agree with you that Verifiablity is a well-written page (despite its unhelpful title); it's NOR and Reliable Sources that are rotten. A page combining all three is therefore a valid proposal. Please work with SV rather than against her, because we need both your efforts on the same page.qp10qp 15:10, 14 November 2006 (UTC)

I don't see how Jguk's minimalistic approach could work. Leaving the "dirty work" to WP:RS is not a solution... given its poor state. —Preceding unsigned comment added by Jossi (talk • contribs) [15] 16:57, 14 November 2006 (UTC)

I'm not suggesting having WP:RS. Just a principle-based definition and explanation. Ultimately you can't legislate for common sense, which is really what it boils down to, jguk 17:19, 14 November 2006 (UTC)

Which is why one of the key principles tells editors that if there's general agreement that common sense disagrees with the rules here, they should go ahead and use a source anyway. JulesH 19:29, 14 November 2006 (UTC)

Oh yes. I hope it was clear from my original posting to this thread that my personal answer is that the proposal is currently failing but is also retrievable. If there is goodwill on all sides, and sensible discussion of all issues raised in good faith, I'd very much hope that the outcome of this proposal is better worded content policy for WP, jguk 20:02, 14 November 2006 (UTC)

Jules

Hi Jules, questionable sources must never be used to support any biographical claims, not just controversial ones. SlimVirgin ^(talk) 16:28, 14 November 2006 (UTC)

Nice to see that someone else agrees with me on that point. That's why I suggested a CSD criterion allowing unsourced biographies of living persons to be deleted on sight. We simply have no way of knowing what mischief might be afoot. Robert A.West (Talk) 17:31, 14 November 2006 (UTC)

But self-published sources are questionable sources, and we allow those for non-controversial claims. The phrase as it stands at the moment is contradicted by later explanations, which I think should be avoided. JulesH 19:20, 14 November 2006 (UTC)

We do not allow self-published sources to be used as third-party sources, as the current policies and the proposal both say. SlimVirgin ^(talk) 20:03, 14 November 2006 (UTC)

I think this illustrates a fault in the logic of the proposal here. It would be better not to introduce the concept of a "questionable source", then "self-published source" and then confuse them by amalgamating the ideas in the way the proposal does.

The real question is whether the source is reliable (or reputable). Where sources are self-published that certainly raises our scepticism about them - probably to the level that it is a rebuttable assumption that they should not be used. But it really depends on the claim. If I write that my fiancée is 25 years old in my self-published work, why shouldn't you use that information in the article about her (if she had one)? jguk 19:33, 14 November 2006 (UTC)

Another question, if I may

I'd like to open a thread that I hope my clear up misunderstandings and pave a way forward. It's a question:

Who are we writing for?

The reason I think we should address this is that the approach this page should take really depends on the answer to that question. If we were explaining over V/NOR/RS policies to a journalist, we'd use different language than we would to a seasoned Wikipedian, to your parents, or to an academic studying how WP resolves disputes. That's because we'd tailor our answer to address their particular needs.

Are we writing for users who wish to know the basics - what should they be doing, what content does Wikipedia find acceptable?

Are we writing for experienced users to codify the status quo in terms that experienced users (but not necessarily inexperienced users) may understand?

Are we compiling a list of prohibited material that we can put in front of people to get them to stop what they are doing wrong?

Are we compiling a list of acceptable material that we can put in front of people to get them to start doing things right?

My preferred approach is to address the first of those four audiences, but in a way that also allows us to address the fourth. My belief is that SlimVirgin (and some others) have written a policy to address the third of those audiences, but not the other three. It is there that our differences lie. Both SlimVirgin and I are arguing that we should re-present policy, not change it. Both SlimVirgin and I are arguing that the concepts of V, NOR and RS should be merged onto one page. But we have completely different ways of approaching it, jguk 19:45, 14 November 2006 (UTC)

I think the problem is that we have to do all of these things. The page must be accessible to new users, yet it must address all the points that experienced user expect it to. There are some cases when we need to prohibit material that might otherwise be used and some cases where we need to specifically allow material that a narrow reading of the rules would disallow.

And in many ways it does do all of them, despite the fundamental difficulty of writing all of these things down. The "key principles" section is accessible to new users, and introduces the basic idea reasonably well (although it could do with some editing before it is finalised IMO). The remainder addresses the concerns of advanced users, by giving specific interpretation of those principles in particular areas. There are a number of prohibitions of things that there is wide agreement should be prohibited (e.g. self-published sources) and also exceptions for those prohibitions in cases where they clearly aren't sensible (e.g. when a source was written by the subject of the article and its content is credible).

The result is a policy that is slightly wordier than I would like, but hopefully that isn't an irretrievable situation. JulesH 19:58, 14 November 2006 (UTC)

For "users who wish to know the basics" we have WP:PILLARS. This page (when upgraded to policy and replace V, NOR and RS) is the policy as stated by the community in all the detail needed. The companion FAQ at WP:ATT/FAQ, providing examples and answers to commonly asked questions.≈ jossi ≈ t • @ 20:05, 14 November 2006 (UTC)

Thank you for your comment, which I for one find useful. If I may be pedantic though, may I ask who do you believe this policy should be written for? If I may take your phrase, "The policy as stated by the community" to whom? Who are the audience? jguk 20:17, 14 November 2006 (UTC)

People with a sincere interest in learning about how this project works. I doubt that occasional users will read this page. For them and for newbies we could just have WP:PILLARS or a similarly constructed page featuring just the polciy "nutshells". ≈ jossi ≈ t • @ 20:27, 14 November 2006 (UTC)

The "use common sense" principle

I'm trying to correct a grammatical error in this principle. One sentence currently reads:

Editors are most likely to encounter these exceptions in the areas of popular culture and fiction, where professional sources offer shallow coverage, or none, than for established subjects where professional sources dominate.

Note that the conjunction of "most [...] than" doesn't make any sense. The sentence should read "more [...] than" as it did when it was originally written. I've made this change twice and have been reverted on both occasions.

I also suggest removing "or none" as it is redundant and makes the structure of the sentence rather confusing. This change has also been reverted twice. JulesH 19:51, 14 November 2006 (UTC)

I restored your edit. ≈ jossi ≈ t • @ 20:10, 14 November 2006 (UTC)

I was just about to ask "more likely in these areas than where"? Editors are most likely to encounter it in these areas, not more likely, because we're not comparing it to anything. SlimVirgin ^(talk) 20:27, 14 November 2006 (UTC)

"[...] than for established subjects where professional sources dominate". It's the "or none" that confuses the sentence to the point where it's hard to read, I think, that's the problem here; with it in, you can't easily see that the two clauses are connected like that. OTOH, switching to "most likey" and dropping everything from "than" onwards makes the sentence even clearer, and I see no downside to it. JulesH 20:32, 14 November 2006 (UTC)

The "or none" isn't important so by all means remove it, but "most likely" is correct. SlimVirgin ^(talk) 20:36, 14 November 2006 (UTC)

That section refers to " professional sources", but I am not sure we have defined that distinction anywhere. ≈ jossi ≈ t • @ 20:41, 14 November 2006 (UTC)

The grammar is much better now; the trouble is that the following sentence is untrue:

Editors are most likely to encounter these exceptions in the areas of popular culture and fiction, where professional sources offer shallow coverage.

If ever there was an area of knowledge where "professional sources" offer thorough coverage, it is in the area of fiction; apart from anything else, the books, films, TV programmes etc. are professionally produced through a collaborative effort involving many people; then we have all the newspapers, magazines, and critical works that report on them. It is possible that fiction is not meant here, but only popular fiction (in which case, bad English); but popular fiction, too, is often reported and the writers interviewed. I've made this point twice over the past month and challenged editors to give me examples of fiction covered so shallowly that questionable sources may be used, without takers.

However, always the optimist around here, I have a proposal that would keep what seems to be the favoured wording but change the punctuation in a way that eases my objection:

Editors are most likely to encounter these exceptions in the areas of popular culture and fiction where professional sources offer shallow coverage.

That removes the impression that fiction is an area with shallow sources but leaves room for the possibility that some works of fiction are shallowly covered. And I think it keeps to the spirit of the idea that such examples will be exceptions, rather than that the whole areas of fiction and popular culture are shallowly covered by secondary sources.qp10qp 22:12, 14 November 2006 (UTC)

Then, my next question would be, what is this doing under "Key Principles" rather than under "Using questionable or self-published sources"? qp10qp 22:24, 14 November 2006 (UTC)

And again

Atomaton, can you explain why you re-added: "Material from questionable or self-published sources may be used as primary source material in articles about that source's author only rarely, under exceptional circumstances, when other reliable references for non-disputed events cannot be found elsewhere.

I explained above that this is wrong: not what the current policies say or what current best practise is. It would mean we couldn't use a self-published source in an article about himself regarding an issue that was disputed. Also, could you answer my question above: what is the difference in that edit between "only rarely" and "under exceptional circumstances"? SlimVirgin ^(talk) 20:01, 14 November 2006 (UTC)

I'm getting the impression that the people currently editing this page don't know what the current content policies say, which is why they're having difficulty editing the proposal without adding internal inconsistencies or unworkable new ideas. Am I right about that? If I am, it's a problem, because you need to very thoroughly understand the current policies in order either to streamline them or change them coherently. SlimVirgin ^(talk) 20:06, 14 November 2006 (UTC)

My take is that WP:RS is in such a mess, that is creating false impressions of current policy. The sooner we promote this proposal, the better. ≈ jossi ≈ t • @ 20:11, 14 November 2006 (UTC)

My take is that it was this kind of editing that created RS in the first place. I'm sorry to write this way, but I don't know what else to say. People are making changes that are pointless at best, senseless at worst. If you feel you can really improve the writing, by all means edit it, but causing the writing to become even looser is taking us in the wrong direction. We need to tighten without altering meaning. SlimVirgin ^(talk) 20:26, 14 November 2006 (UTC)

I'm sorry, I missed the commentary. I had thought my wording was changed without explanation again, and I see that there is some discussion.

I think my point is that the way that it is worded it explicitly gives people permission to add materiable from questionable or self-published sources. My edit does not prohibit it, it merely says that that it should only be done rarely, and under exceptional circumstances.

Can you say what you mean by "rarely," on the one hand, and "under exceptional circumstances," on the other? SlimVirgin ^(talk) 20:43, 14 November 2006 (UTC)

You make it sound as though it had been prohibited. The way it is written now, many people will quote this when it is poolicy and suggest that they do not need other sources, that sourcing the subject of the articles blog, or personal web page is sufficient, even if other sources are available.

No, if you read what it says, it tells you when you can use self-published sources. The limits are explained. Did you read what I wrote about this above? SlimVirgin ^(talk) 20:43, 14 November 2006 (UTC)

I'm aware that there are at least a dozen people here who have substanitally more experience and understanding of the large view of Wikipedia policies than I do. In fact I continue to be humbled by the careful thought and experience that many of the people here have and express. I stay limited to this part of the discussion because it is one that I have had trouble with myself. From my reading, this proposed policy seems to be an attempt at changing policy as well as better clarifying it. So, if my wording does not exactly match current policy, that would be intentional. Obviously I think clarification of that part of the policy is important.

No, this is not an attempt to change policy, and the change you made would be completely unworkable, and indeed absurd, because it would mean a self-published source, if accused of something by a reliable source, would not be allowed to defend himself. SlimVirgin ^(talk) 20:43, 14 November 2006 (UTC)

I can't agree with you. Wikipedia is not a forum for resolving dispute (what a topic that is...) Indeed, if someone is accused of something by a reliable source, Wikipedia is not the forum for resolving that. The appropriate manner would be that a newspaper or magazine would interview the person regarding the dispute, the person would express their view, and then a wikipedia editor would quote the article and cite it appropriately. Atom 21:26, 14 November 2006 (UTC)

Unfortunately that means that unless you are a "public figure," you can't defend yourself from libelous claims since you will not have access to the media. I'm ok with that as long as we can delete all living bios that do not meet the standard of being a "public figure." --Trödel 22:15, 14 November 2006 (UTC)

I see your point about access to the media. However, we have a very tight policy regarding Biographies of Living Persons, and slander or unsourced negative comments are not allowed in them. If a person can make what could be viewed as negative descriptions in a BLP, and find proper and reliable sources for it, then those belong in the article. If there other views, even if it helps balance NPOV, but there is no reliable reference for the view, then that is unfortunate, but does not belong in the article. Although we should be intimately concerned about acting ethically, it is not our job to resolve conflict for controversies described in wikipedia, nor to do reporting on events. As an encylopedia we are here to report facts as cited by verifiable sources, not try to act as a judge, mediator, forum for fairness or collection of all opinions and viewpoints about a person or topic. We can have compassion for people who are treated unfairly by culture, society, or the government, but it is not our job to offer a place where every notable person is represented to their liking. Living BIO's should have content with facts as supported by other peoples research, not our own, or our or the subjects research or opinions. Atom 22:28, 14 November 2006 (UTC)

That is incredibly irresponsible and poor scholarship too. Sorry you can't rebut this lie because it isn't in a RS - that is a ridiculous standard. --Trödel 04:38, 15 November 2006 (UTC)

You say If there other views, even if it helps balance NPOV, but there is no reliable reference for the view, then that is unfortunate. That is totally wrong understanding of current policy! What we are discussing is that in BLPs and in articles about organizations, the POV of that person or organization can be sourced to their self-published material, be that a press relelase, a website, an annual report, etc. That has been the policy so far, as it is consistent with NPOV that states that all significant viewpoints need to be presented. ≈ jossi ≈ t • @ 01:22, 15 November 2006 (UTC)

You say "it would mean we couldn't use a self-published source in an article about himself regarding an issue that was disputed." I am not sure I understand your point. If the source is self-published, then it is original research.

No, you've misunderstood what original research is. SlimVirgin ^(talk) 20:43, 14 November 2006 (UTC)

I'll go review Original Research again, but IMO a wikipedia editor gathering data from primary sources is reporting, or original research. Citing a Time Magazine article where a reporter gathered the data from primary sources would not be reporting, or original research. Doesn't the NOR policy say "any facts, opinions, interpretations, definitions, and arguments published by Wikipedia must already have been published by a reliable publication in relation to the topic of the article" and "The prohibition against original research limits the possibility of an editor presenting his or her own point of view in an article. By reinforcing the importance of including verifiable research produced by others, this policy promotes the inclusion of multiple points of view in an article" When I interview Barack Obama and ask his opinions about the failings of the current administration, and then use that in the Wikipedia biography of Barack Obama, regardless of the content, isn't that original research? Has this already been printed by a reliable publication before? If it has, is it okay for me to refere to Mr. Obama's web page, rather than the Time article? How is what I put in the Wikipedia article now considered to be "verifiable research?" Atom 21:26, 14 November 2006 (UTC)

Good idea, go and read WP:NOR, as you are indeed confusing things a bit. ≈ jossi ≈ t • @ 21:29, 14 November 2006 (UTC)

If you are not a journalist in a reputable publication, your interview of Mr. Obama cannot be included in his article as that would be OR. But we certainly can cite his website (if he as one) to describe his views of the current administration. That is what is being discussed. ≈ jossi ≈ t • @ 21:32, 14 November 2006 (UTC)

Interviewing the person, or taking it from his blog or his web site is all the same thing. Why would speaking with him in person be original research, but quoting what he put on his web site NOT be original research? All three are one persons opinion expressed three different ways. Us quoting any of the three is OR. Atom 21:59, 14 November 2006 (UTC)

Your question, points to the fact that you may not have not understood the principle of NOR. NOR is relatred to 'editors making that research, not persons about which the article is about. If that person would want to edit Wikipedia, that would be a different story and a;ready covered in WP:AUTO and WP:COI ≈ jossi ≈ t • @ 00:38, 15 November 2006 (UTC)

One particular issue that may be the source of your confusion, is that Wikipedia content policies work together. You may be interpreting one, without the benefit of another. NPOV cannot be bypassed, so we can and should cite self-published sources in articles about their author if only to provide the viewpoint of the author. ≈ jossi ≈ t • @ 21:35, 14 November 2006 (UTC) NPOV allows us to publish POV, but says that we should be balanced and offer all (citeable, referenced) views. Gathering his views by interviewing him (or from his web site) may be in alignment with NPOV, but not NOR. We can site the Time article where he expresses his view, but interviewing him, or quoting his web site, or Blog is "reporting" and OR. "any facts, opinions, interpretations, definitions, and arguments published by Wikipedia must already have been published by a reliable publication in relation to the topic of the article" Atom 21:59, 14 November 2006 (UTC)

Atomaton, you're mixing up a bunch of different issues. The sentence you changed was talking about self-published sourcces in articles about themselves, and that is, of course, allowed within the limits laid down by the policies, which are retained here. Your interviewing someone and using it would be OR, but that's not what is being discussed. SlimVirgin ^(talk) 21:36, 14 November 2006 (UTC)

Our specific discussion went to discussion or NOR. I was talking about that generally, and not specific to about themselves I should have made the point that the general view of NOR should extend to articles about the person themselves. Atom 21:59, 14 November 2006 (UTC)

Getting information (regardless of the content) directly from the person, or from their blog, or the web page that they design and maintain as an extension of themsleves and interests is nothing more than original research. Not using such material might lead to inconvenience, but it is better than an entire biography where the subject of the biography is the primary source for 90% of the facts stated in the article. The question is then, "When is using unreliable or self-published sources appropriate?" I would say "Only rarely, under exceptional circumstances". Otherwise the policy that you propose says "Material from questionable or self-published sources may be used as primary source material in articles about that source's author." My wording says it should almost never be done, that one should resort to digging to find reliable sources. This is a pain, I know, however, when Wikipedia, or an editor from it attempts to do "reporting" by collecting and using primary sources rather than reliable and citable sources, that level of work and detail should be taken (if you feel that reporting is sometimes appropriate, which I could argue agains). To allow reporting, and to give "permission" via the policy with the wording you suggests, introduces a hole in which unreliable and uncertain information enters into wikipedia, and then is quoted by other wikipedia articles, and by reliable sources such as newspapers and magazine. It is only a matter of time before this creates a problem where the reliability and fact checking of articles by "reporters" at wikipedia comes into question, and the reliability of the information in Wikipedia as a whole is subject to scrutiny. Although it would be my personal preference to never allow unreliable and self-published sources, I accedde to those with more experience than myself, such as you and others here who say that there are times when it is sometimes necessary. Hence my wording "Only rarely, under exceptional circumstances".

Thanks for your patience with listening to my opinion. Atom 20:37, 14 November 2006 (UTC)

I see your point, but note that you are taking it to an unnecessary extreme. An article on Acme, Inc. can (and should) include material about the way that Acme, Inc describes itself in their company literature. If there are competing views about Acme, Inc, we list them. All to abide by NPOV. In the case of a person, sames applies. If the person declares in his website or perosnal home page that he considers himself an African-American, we can state that even if other reliable sources may provide the viewpoint that he is not from African ancestry, etc. ≈ jossi ≈ t • @ 20:53, 14 November 2006 (UTC)

Aside

As an aside, I'd like to welcome SlimVirgin's approach here of questioning the amendment on the talk page rather than reverting. This allows for propoer discussion of Atomaton's suggestion - and after the discussions today in particular, I hope this presages a return to constructive collaboration from all interested parties, jguk 20:05, 14 November 2006 (UTC)

I would gather that's only because another revert would violate 3RR. —Centrx→talk • 20:30, 14 November 2006 (UTC)

I'm not going to continue responding to Jguk, but for him to turn up here talking about collaborative editing after the revert war he engaged in with Radiant over the three content policies today, against multiple editors, is stunning. That's my last comment to you, Jguk, because you're doing this deliberately; you know you are; and everyone who knows your editing knows you are. SlimVirgin ^(talk) 20:39, 14 November 2006 (UTC)

Regardless, I welcome the move towards active discussion of differences, a position I have advocated consistently since my recent return to WP, jguk 20:58, 14 November 2006 (UTC)

Primary sources, again.

I realise we're trying to slim this down, which is certainly a good idea, but in the current draft it appears that there's neither discussion of how primary sources can be used, nor the previous "sources requiring in interpretation" text. That primary sources can be reliable sources was stated very clearly in the polcies we're attempting to replace, but the text describing reliable sources here, under key principles, seems to be written with secondary sources in mind; the reliability of primary sources doesn't necessarily depend on "process and approval between document creation and publication", "people engaged in checking facts, analysing legal issues, and scrutinizing the writing", etc. Is it possible this text could be added to, or slightly tweaked, to acknowledge that primary sources can be reliable sources? Is not, can it be addressed elsewhere on this page? -- Bailey^(talk) 14:27, 15 November 2006 (UTC)

We have addressed that in WP:ATT/FAQ. ≈ jossi ≈ t • @ 15:12, 15 November 2006 (UTC)

I see that, and I'm fine with the bulk of the discussion of how primary sources can be used remaining on the FAQ page. But the idea that primary sources can be reliable sources seems like an important enough idea to be mentioned here directly. That concept doesn't follow logically from the present text describing reliable sources, and it's an important yet widely misunderstood concept in Wikipedia. -- Bailey^(talk) 16:42, 15 November 2006 (UTC)

I think there needs to be a section about primary/secondary. I've wavered back and forth on this point, but I've been keeping track for the last couple of weeks of how often the distinction's come in handy and I see that it's used a lot. SlimVirgin ^(talk) 23:34, 15 November 2006 (UTC)

Maybe it's time to set up a vote (I mean opinion poll) about keeping that unnecessary issue or not --IMO it is both time-wasting and artificial, for a reminder see Wikipedia_talk:Attribution/archive5#Delete_all_talk_about_primary.2C_secondary_etc._sources.3F. ::: It may seem a bit drastic, but by now I see no other effective way to stop never-ending talk and expansion of invented issues. Of course it could still be mentioned in WP:ATT/FAQ.

However, the proposal should be brought to the attention of a large amount of Wikipedia editors. How can that be done, as apparently most editors never look (or don't know where to look)?

Harald88 21:34, 15 November 2006 (UTC)

I'm not 100% certain we're talking about the same thing here. I'm presently neutral on the question of whether referring specifically to "primary sources", "secondary sources" etc. is especially confusing; however the basic concept -- that certain kinds of published sources are always reliable sources for purely descriptive claims about their own contents -- needs to be addressed directly, regardless of what language we use to describe that concept. IMO, a lot of the never-ending talk on this subject comes from confusion about when it's okay to use a source reflexively vs. when that's original research -- this is a distinct issue. I feel we need to be clear about this in the main policy itself, rather than just in the FAQ. -- Bailey^(talk) 21:46, 15 November 2006 (UTC)

"the basic concept -- that certain kinds of published sources are always reliable sources for purely descriptive claims about their own contents" This statement about reliabilty is true of all published sources not just "certain kinds". The issue is that while all claims of this kind are reliable, it is diffcult to write purely descriptive claims and maintain them in a collaborative enviroment and they may not be appropriate for WP for other reasons (i.e. NPOV).--Birgitte§β ʈ Talk 21:57, 15 November 2006 (UTC)

Via edit conflict:You're correct; I was just using cautious language for the sake of being cautious. The only exceptions I can think of off the top of my head would depend on the specific definition of "published" -- there are some things that one could concievably argue were published once that are no longer widely available -- for example a 19th century newspaper could be considered published, but would not make a good primary source if it wasn't widely available currently in some form.

Regardless, NOR does presently allow for the use of primary sources, and whether or not they're difficult to maintain in a collaborative environment, it would represent a pretty major change to suddenly discourage that. I'd be happy with even a single sentence acknowledging that published sources are reliable sources for descriptive claims about their contents, with whatever caveats are necessary. -- Bailey^(talk) 22:20, 15 November 2006 (UTC)

I just reinserted the nutshell phrase that had gone lost from the article, and with which we had come up in the abovementioned discussion:

a source must be reliable for the claim being sourced

That summarises all such rules and exceptions. Harald88 22:03, 15 November 2006 (UTC)

Not so sure about that summary. The preceding paragraphs explains it quite clearly. Regarding your idea posed above for a poll, I would say, God forbid! This proposal is developing nicely and many editors have contributed of the last month or so. The existence of this proposal as been posted on the Village pump and discussed on the Wiki-En mailing list. Those editors interested in policy know about it, and that is the way it stands now. ≈ jossi ≈ t • @ 22:21, 15 November 2006 (UTC)

I disagree with the addition of that sentence. Being reliable for the claim being sourced is not enough in many cases for a source to be reliable. Removed again. ≈ jossi ≈ t • @ 22:23, 15 November 2006 (UTC)

As policy affacts all editors, I disagree with your opinion that only editors who are interested in policy should be consulted for this move to simplify the policy pages.

And I strongly disagree with the removal of that phrase by you [16]. If the earlier sentences would explain it as nicely, there would be no need to mention exceptions that are to be explained later. I am sure about that exception-free phrasing: it was the consensus that resulted from an extensive discussion, see [17],and [18], and which was next transfered into the article by WAS, [19]. Your claim that "Being reliable for the claim being sourced is not enough in many cases for a source to be reliable" puts things on their head - what matters for Wikipedia is that Wikipedia is reliable.

It's that kind of misrepresentation that causes much of the Wikilawyering that next needs to be countered with policies such as WP:IAR. Revert. Harald88 23:09, 15 November 2006 (UTC)

I would appreciate if you lower your tone and spare us the rhetoric. The discussions you refer to are of no consequence as these were made at a different time, and since then a lot has changed in the current formulation. As for your "I strongly disagree with your removal", I counter that with "I strongly disagree with your addition. ≈ jossi ≈ t • @ 23:16, 15 November 2006 (UTC)

The addition wasn't mine while the removal was yours; thus your "counter" is misleading. No rhetoric is needed, and the "different time" happens to be the time period to which you just referred. Thus I would equally appreciate it if you lower your tone as well. Harald88 23:42, 15 November 2006 (UTC)

Bailey, I don't think you need worry that the use of primary sources is being discouraged here. Can you really see any sign of that? What has happened is that consensus has emerged that all sources are subject to the same need for attribution and that we mustn't add our own original thought to any type of source. Overall, I believe this page is more encouraging to the use of primary sources than previous ones because it doesn't get into calling self-published websites and blogs primary sources, thereby tarring all primary sources with the same brush amd causing unnecessary anxiety to editors—of history articles, for example—who use primary sources in the accepted way (in other words, by referring to edited and published versions). qp10qp 22:38, 15 November 2006 (UTC)

I do agree that the distinction you refer to is important. I'm just iffy about emphasis that's being placed on fact-checking as a determiner of reliability without some explict exception being made for primary sources. I may be needlessly worrying, but on the other hand, there are plenty of editors who are quick to label perfectly legitimate uses of primary sources "original research".

Alternatively, could an exception be added to to the "questionable and unreliable sources" section, parallel to the "Questionable or self-published sources in articles about their authors" exception? Something on the order of "Questionable or self-published sources in articles about the source itself"? If I'm writing, for example, about a notable propaganda film, that film might have had terrible fact-checking, but I should still be able to use it as a source for purely descriptive claims about the film itself. I can't possibly see the harm in adding that other than increasing the word count slightly. -- Bailey^(talk) 22:48, 15 November 2006 (UTC)

Okay, poor example, now that I think about it. The text about questionable sources already states that "questionable sources should not be used, except in articles about themselves". This leaves me 85% less concerned about this particular issue. Still, shouldn't that same exception apply to self-published sources notable enough to warrant their own articles? -- Bailey^(talk) 23:06, 15 November 2006 (UTC)

It is generally required to cite sources that are reliable for the claims that are made. Verifiability makes no exceptions but is demanded for all notable published information. And it is policy that articles must be properly sourced with fair citation of all notable points of view, in accordance with the key policy WP:NPOV. Harald88 23:28, 15 November 2006 (UTC)

Bear in mind that we're not here to change the policies, but to streamline then. What NOR says now about primary/secondary is what's policy, and this proposal must either say or reflect it. SlimVirgin ^(talk) 23:38, 15 November 2006 (UTC)

The sentence you are trying to add is at best confusing, and at worse opens up the policy to almost any source. Confusing and possibly dangerous qualifiers should not be added to the policy; let's keep it as concise as possible. Jayjg ^(talk) 00:00, 16 November 2006 (UTC)

The policy is not to be protected but instead Wikipdia must be protected against misuse of sources. And the threshold is notability, see WP:V and WP:NPOV. Harald88 00:16, 16 November 2006 (UTC)

Ok. After doing some re-reading, I do see how this properly reflects what's already in NOR and V. My mistake. Thanks for the patience. -- Bailey^(talk) 23:59, 15 November 2006 (UTC)

That's exactly the point: to reflect the policy better, in a more concise way, based on logical instructions. Consensus was that such an important improvement was obtained by replacing the phrase that "a source must be reliable" by the phrase that "a source must be reliable for the claim being sourced", as it neatly summarizes the discussion about primary/secondary, the exceptions on that as well as "self-published". If you disagree, do you see an occasion in which it could result in Wikipedia being less well sourced? Harald88 00:01, 16 November 2006 (UTC)

The problem is that your proposed text is both longer and less clear; thus it is neither "concise" nor "logical". Using your qualifier I could easily imagine someone arguing "Sure, this Holocaust denial website is untrustworthy in general, but regarding the claim that Hitler had an IQ of 177 it is, in fact, a reliable source". Unreliable sources are unreliable. Jayjg ^(talk) 00:06, 16 November 2006 (UTC)

The point is that it gives the possibility to delete half of the rest of the policy page, as it's already summarized by it. That opens up the space to add another one or two lines that spell it out a little more. Your comment also shows that you didn't read the discussion: "Unreliable sources are unreliable" is in conflict with WP:NOR that explains that "unreliable sources" are reliable about themselves. But do you really think that such inconsistent explanations with imprecise rules that have an incomplete list of exceptions are best for the editors and Wikipedia? Harald88 00:29, 16 November 2006 (UTC)

Harald, it's meaningless. All it does is raise questions. SlimVirgin ^(talk) 00:06, 16 November 2006 (UTC)

Slimvirgin, never mind: From this result after consensus was reached about it following a one week discussion in which all those questions were answered, I'm sure that streamlining policy is doomed to fail. As the current policy only raises questions and confuses people and the new one won't be much better, I decided to abandon it altogether. Instead I now support (and help improving) WP:IAR. Good luck! Harald88 00:29, 16 November 2006 (UTC)

No, please stay with us and help. It's not easy to get one's way here, and we have to live with that. I've argued against the popular-culture exception over reams of discussions, often with a majority agreeing, till I was blue in the face, but it's still there. Such is life (sings chorus from Chumbawumba). I believe that discussions here, though they might not always lead to a change in the text, seed group thought in ways that will eventually bear fruit. Although I agreed with the removal of the phrase in question, I continue to ponder how the matter of citations to reliable sources that are unreliable for the particular reference might be covered in this text. This all takes time and requires much patience and the ability to cope with appearing to be ignored or overruled. qp10qp 00:59, 16 November 2006 (UTC)

Harald, you've said a few times that something has consensus here just because a few editors have agreed. The problem with that position is you're assuming people don't disagree unless they say so, but that's not how consensus works. Two things are clear to me: first, that there is strong consensus for NOR and V across Wikipedia. Although a number of editors complain about them regularly, that doesn't change the fact that there is strong support for them; previous attempts to change them have attracted little or no support. Secondly, I think people will welcome a genuine tightening and streamlining of V and NOR, but only so long as there's no fundamental change, and so long as the streamlining really is an improvement in terms of clarity of thought. If we have sentences or sections where, as someone else said, you have to do a double take when reading it, it won't get support. We could have a dozen editors on this page agreeing on every single point; it'll make no difference unless the community gets behind it. SlimVirgin ^(talk) 01:09, 16 November 2006 (UTC)

SlimVirgin, the change that I had in mind would considerably unbloat the policy page (and additional explanation pages!), but I must admit that I did not take the time (indeed didn;t have the time) to work it out sufficiently.

I now do see your point that that single sentence may be less clear to some than the half a page with which everyone is familiar and with which you compare it, it's an unfair comparison but I did not deliver a full replacement when I should have. By chance I just typed it on another page, I copy here what I had in mind (it's still rough, that's the problem when starting from scratch by thinking "out-of-the box", but it may better explain what I had in mind):

All information must be notable, and all information sources must have been published (or made public) and be reliable for the purpose for which they are cited. That implies:

1. Article topics must be based on information in trusted sources; topics that require interpretation must be based on sources that provide such interpretation.
2. In particular, this policy has been sharpened as follows: topics about scientific theories must be based on peer-reviewed publications and historical topics must be based on history articles.
3. For every claim the most reliable sources must be cited. Usually that will be the original source for the claim (if available). Interpretations must be fairly cited in accordance with WP:NPOV.

Something like that would shrink things dramatically but it would imply throwing away a lot of text that has been carefully negotiated over the last years, including most (or all) talk about "primary sources". As that may never happen, and I am here not for law negotiations but for article writing, I now "unwatch" this page. Regards, Harald88 23:38, 16 November 2006 (UTC)

Time limit

In the section How to cite and request a source, how about adding in a line of this nature:

If you want to request a source for an unsourced statement, consider moving it to the talk page; or tag the sentence by adding the {{fact}} template; or tag the whole article by adding the {{not verified}} template. Material so tagged should be removed after a reasonable amount of time, if no reliable source is forthcoming. Absurd claims and original research should be removed rather than tagged; contentious claims about living persons must be removed immediately.

Of course the question will be, what is a “reasonable” amount of time. As that will change from article to article based upon the subject mater of the article and the nature of the tagged claim, it might be better not to list a specified amount of time. Brimba 14:37, 15 November 2006 (UTC)

Agreed. That's why we use the word "reasonable." SlimVirgin ^(talk) 21:45, 15 November 2006 (UTC)

Indeed that is current policy and it's rather effective -WP:CITE. But is this article meant to include WP:CITE as well? Harald88 21:32, 15 November 2006 (UTC)

Harald, I'm not sure what you mean about this proposal including CITE. SlimVirgin ^(talk) 23:52, 15 November 2006 (UTC)

Sounds like instruction creep - although you're probably on safe ground if you do it anyway, jguk 21:35, 15 November 2006 (UTC)

SlimVirgin, that's indeed what I meant: as it's already in WP:CITE#Tagging_unsourced_material, it's again more instruction creep - the very opposite of what the purpose is. Except if you want to eliminate WP:CITE.

Anyway, I now decided to give up on policy, it's hopeless, an uncurable cancer; IMO the only remedy for all this ever increasing instruction creep is WP:IAR, and I'm glad to see that Jimbo supports it. Harald88 00:10, 16 November 2006 (UTC)