Wikipedia talk:Verifiability/Jguk's version

From Wikipedia, the free encyclopedia

Contents

[edit] To start with

I'll want some time to digest this, and thoroughly compare it to the existing policy. One quick note, in 'Citing sources', you have, Any edit lacking a source may be removed. If an article topic has no third-party sources, Wikipedia should not have an article on that topic. If you doubt the accuracy or origin of an unsourced statement that has been in an article for a long time, delete it or move it to the talk page. My feeling from other discussions is that we are not ready to endorse the first sentence. And I suspect that you intended to say in the last sentence, If you doubt the accuracy or origin of an unsourced statement that has not been in an article for a long time, delete it or move it to the talk page. -- Dalbury(Talk) 19:55, 29 December 2005 (UTC)

The first sentence has been endorsed already, though it was recently questioned on the talk page. But it has had widespread support for as long as I've been editing. And for your second point, it doesn't matter how long a claim has been in an article: if you doubt the accuracy or origin of it, you may remove it. SlimVirgin (talk) 20:02, 29 December 2005 (UTC)
I would love to just rip out unsourced material, but I think doing that wholesale would be disruptive. I do try to source some articles I find, but that can be time consuming, and leads me to related articles that need work, as well. I am experimenting with marking articles [[unreferenced]] and linking to them from a 'suspense' page, and that has provoked an attempt to supply references in a couple of cases, but I'm still trying to figure out how to escalate without disrupting. -- Dalbury(Talk) 21:15, 29 December 2005 (UTC)

The first sentence has never been so directly stated in policy before. SlimVirgin is right, however, in saying that it is accepted. Good editors already appreciate the need for sources, and I think this has improved a hundredfold in the last year. If you revert something with the edit summary "Revert - removing unreferenced information. Please only re-add if you can cite a source, see Wikipedia:Verifiability", you will generally find a positive response. (By the way, I wasn't sure of this until I experimented for an hour myself doing just that. I found that, except on an article where an edit war was in progress, info either stayed out or was re-added but with references.

On the second point, I didn't omit "not". Whilst you'll find good editors accept that new edits need sources, there is much unreferenced older material on Wikipedia. People do have problems if this is removed wholesale, so greater care is needed. Even if there is something that you think is wrong or dubious it may be best to raise the issue on the talk page first before deleting everything. In essence, there is a transitional "rule" for old material. All I'm trying to do is make reference to it, jguk 20:34, 29 December 2005 (UTC)

(after edit conflict) The strongly phrased version was only recently added by SlimVirgin [1]. Prior to this, and IMO, what is still generally accepted practice is that you should only summarily remove unreferenced edits that you have reason to believe are false or inaccurate. I very much do not like giving license to remove *any* edit simply because it is not sourced. I have not encountered editors summarily removing unreferenced material simply because it was unreferenced--nearly always the material was rather obviously dubious and could have been removed on any number of grounds. There are differences between handling edtis by anons and registered users as well. For an anon with only a few questionable edits, few would consider it worth the effort to try and engage in dialogue. But for a registered user, I would hope we would assume good faith and try to engage the user rather than simply summarily remove the edits. older≠wiser 20:47, 29 December 2005 (UTC)
WP:V and WP:NOR are inextricably linked, because the only way to show you're not doing the latter is to comply with the former. NOR does not allowed unsourced material either, so any wording that suggests unsourced material can stay in an article for reasons of politeness runs counter to the entire philosophy of Wikipedia. Realistically, have you ever encountered anyone systematically going through articles and removing every unsourced edit in order to be disruptive? SlimVirgin (talk) 21:24, 29 December 2005 (UTC)
Realistically, have you ever encountered anyone systematically going through articles and removing every unsourced edit in order to be disruptive? No, of course not. But what is the point of such a strongly phrased policy when it does NOT correspond to actual practice? Since so many unsourced edits are accepted without question, it makes this policy appear ineffectual. older≠wiser 21:43, 29 December 2005 (UTC)

Bkronrad, it does correspond to actual practice. New information can and is regularly removed when it isn't sourced, and old information, when questioned, can be removed.

On SlimVirgin's point, I beg to differ in approach (though not in consequence). To my mind "no original research" follows immediately from the three Verifiability criteria I've listed. WP:NOR doesn't need to be (and therefore shouldn't be) policy. If something is original research it will not be possible to come up with a reputable source for it - either no source will exist, or the source will not be reputable. Although I'm certainly not going to argue we dispense with the guidance on WP:NOR, I would suggest that, if the wording I suggest here on Verifiability is adopted, WP:NOR need not be policy (though I'd suggest that later rather than now), jguk 17:32, 30 December 2005 (UTC)

Editors need no special authorization to remove or edit content. This strong phrasing gives license to remove ANY edit that is not sourced. While it has not been a problem yet, with this being accepted as policy, on what basis would you stop someone who did go about removing ANY edits that did not cite a source? I'm sure that the community would intervene quickly enough, but simply on the basis of how this policy is phrased, such editors would have a strong basis for claiming such removals are authorized.
Also, you didn't address my second point. Yes, while some unsourced edits are removed, many, many more are not. This discrepency between actual practice and the strongly phrased policy makes it appear ineffectual. This is sort of a "Do as I say, not as I do" situation, where the authority of the policy is undermined because there is in fact so much unsourced material that is accepted. Dubious unsourced material should be removed and it usually is. But so many edits are made in good faith and are quite factual and are also unsourced. I don't think this policy should imply that the proper conduct is to summarily remove such edits. older≠wiser 23:14, 30 December 2005 (UTC)
On the first point, I would encourage people to go round new edits reverting those that insert information that isn't sourced. Do so politely, of course - "Revert edit as it lacks a source, please only re-add the information if you have one" will go down much better than "Revert unreferenced POV crap". I have, by the way of an experiment, tried this for an hour or so by monitoring new edits over a five minute period one hour after they were created (after all, existing policy does not ban doing this). I found there were surprisingly few that were unreferenced and that material if it was re-added was re-added together with a reference! I genuinely don't think any good editor would mind being asked to reference his or her work - most good editors are already doing this.
On your second point - wikipedia is quickly getting better at references. Ideally every snippet of information should be sourced. In the past, whole articles were being written without references, and no doubt wholesale deletion of those would not go down well. But we need to promote references. A strongly worded policy gives good editors the ammo they need to shoot down dodgy unreference material. Let's not beat about the bush - every bit of info ought to be referenced. At present, that isn't generally true, but we're getting better all the time, jguk 11:41, 31 December 2005 (UTC)
Perhaps you're using a different selection criteria. I see many, many unsourced edits that go completely unquestioned. Many of these are trivial additions. Or are of the nature of "plain as the nose on your face" type of edits, but no source is provided and I don't see editors requesting sources (and I really don't think sources are needed for ALL edits, as some seem to). What I object to is that this strong phrasing, Any edit lacking a source may be removed, is rather indiscriminate. With that blunt phrasing, it explicitly authorizes persons to remove any edits for which no source is provided. And, while I have often removed dubious materials with only an edit comment, for unsourced material added by registered users, it is rather rude to simply remove good-faith edits with only an edit comment--by far more common in this case (at least in my experience), is either politely asking the editor for a source on the user talk page or else raising a question on the article's talk page (and sometimes moving the content to the talk page). The current phrasing appears only intended to provide blunt force instruments to remedy the worst aspects of unsourced contributions. older≠wiser 12:39, 31 December 2005 (UTC)
It's meant to be a blunt tool - those ones are the hardest to find loopholes to! Also as noted above, there's nothing stopping anyone removing unsourced new information at present, this is just a direct restatement of what is current acceptable practice. Let me ask you a question, are there any edits where you are adding information where you'd consider it unreasonable to ask for a reference? Is it really rude to ask for references by way of an edit summary? Anyone seeing the comment on their watchlist knows immediately what the issue is, and how to rectify it. Do we really want to risk a weeklong discussion on the talk page and then debate whether or not at the end of it, that material should be removed, when all that's really needed is a good, honest reference.
As restated, the WP:Verifiability policy would be both a blunt tool and a straightforward one. One that firmly places the onus on editors, rather than requiring any work other than a simple revert from editors seeking to prevent unreferenced material being added and encouraging use of good sources, jguk 14:12, 31 December 2005 (UTC)
Is it really rude to ask for references by way of an edit summary? For good faith additions by registered users (with more than a few edits), yes, I would think it rather rude to simply remove the addition with merely an edit comment. Individual edit comments can easily get missed -- not everyone monitors every edit to every article on their Watchlist -- only the most recent edit summary appears on Watchlists. [A]re there any edits where you are adding information where you'd consider it unreasonable to ask for a reference? This is an entirely different question -- of course references should be provided if requested. There are countless edits I make that I do not explicitly source because they are relative commonplaces (like so and so place is in such and such township). I don't think requiring explicit references for such uncontroversial information is particularly useful in every case -- but if someone questioned whether I might have made a mistake about something, then sure, I would dig around a bit to find a reference to substantiate the edit. But I'd probably be a little POed if someone came along and simply removed the detail without asking me about it, either on my talk page or the article talk page. I think what I'm getting at is that the bluntly worded, strong policy needs to be tempered with some acknowledgement of WP:CIVIL, WP:EQ, and WP:ASG, as well common sense. What I find missing from this version is any sense of discrimination regarding the quality or nature of the edits. older≠wiser 15:00, 31 December 2005 (UTC)

Civility is covered by a separate policy. But I take your point that it is especially important to consider it when removing material. I have added a sentence to the proposed revision to refer to it, jguk 22:28, 2 January 2006 (UTC)

It's better now, thanks. I'm still a little unhappy with the "big stick" position taken here, but I guess it may be a necessary evil. I do my best to avoid vexacious topics, so I'm really not in good position to say what sort of weapons are needed. But as in RW politics, if you surrender too many liberties in the name of defending those liberties, you may have ended up with a bad bargain. older≠wiser 03:25, 3 January 2006 (UTC)

[edit] SlimVirgin's last amendments

I have reverted SlimVirgin's second amendments [2] as they reproduce much material in Wikipedia:Citing sources, and I think we should make reference to that page, but not get bogged down in how to cite sources. I'm open to re-wording my phrasing here, but I'm reluctant to do much other than direct the reader to Wikipedia:Citing sources. (Mind you, the citing sources page is too complicated for me to understand at the moment! Maybe I'll have to tackle that one next!) jguk 20:34, 29 December 2005 (UTC)

The problem with the way you have it, Jguk, is it's incomplete and incorrect. "If the reference is a book, cite the title, author, ISBN and page number. If it is a journal, cite the title, author of the article, volume and issue number and date and page number." We don't do that in the text, and in the references section, we don't include page numbers and we do include the name of the journal/newspaper. Also, a policy page shouldn't refer people to a guideline without itself giving any guidance on the issue, if it's one that matters and this done. No point in telling people to cite sources without giving them at least one example of how to do so. SlimVirgin (talk) 20:38, 29 December 2005 (UTC)

Maybe this page should say nothing on how to cite sources, but instead just refer to the Wikipedia:Citing sources page, jguk 17:34, 30 December 2005 (UTC)

[edit] unimpeachable sources

I am repeating part of a comment I made on the actul policy talk page. I do have one specific comment about the policy — this is in Jguk's proposed version, but he adopted it from this version. I am referring to the statement that sources should be unimpeachable. I agree that the policy should convey to editors the importance of highly reputable sources. However, when it comes to contentious topics, some sources will not be "unimpeachable," indeed, the sources themselves may be objects of contention. I do not think such sources should be eliminated. Rather, in compliance with NPOV, we should make it clear what the source of contention is, and provide some context about the parties who contend over the un/impeachability of the source. The source's verifiability of course must be unimpeachable by which I mean that the provenance must be clear and accurate (i.e. verifiability means, if we say "this is the source," i.e. provide information to help people find the source (so they can read it for themselves), this information must be incontrovertably accurate. Slrubenstein | Talk 21:05, 29 December 2005 (UTC)

I agree. I was uncomfortable transferring the word "unimpeachable", but unsure what to do with it. I think the whole clause can safely be removed [3], jguk 17:37, 30 December 2005 (UTC)

[edit] Verifiability, not truth

I have removed the following addition to the first part of the restated policy. It was added by SlimVirgin: " The criterion for entry into Wikipedia is verifiability, not truth". The reason is that I think the policy is clear that the entry is verifiability. Any statement as to what it is not is superfluous - it is not anything other than verifiability. I think it's important to try, as far as possible, to stress the positives (although I'm aware I'm just about to add a section on no original research:) ), jguk 22:33, 2 January 2006 (UTC)

I always like the "Verifiability, not Truth" meme very much. It is very effective for ending talk page discussions that go nowhere because an editor tries to convince the others what is really true. I would like it back. Arbor 17:44, 14 January 2006 (UTC)
Yes, it's essential we make clear that truth, by itself, is entirely inadequate. Many editors fail to appreciate the distinction betwen "truth" and "verifiable", and take a suggestion that something is not verifiable, as an personal attack on their honesty. When challenging unverified claims, it has to be possible to say "I'm sure what you say is true, but it must still be verified to be used". --Rob 17:59, 14 January 2006 (UTC)

I think it's important to describe something in terms of what it is, rather than what it is not. Whilst I don't disagree with "verifiability, not truth", the real point is that it is "verifiability, not anything else that you may care to mention that is not verifiability". Rob's last sentence strikes me as a good way to put the point by employing positive terms. Maybe adding a sentence along the lines of "When challenging unverified claims, you may consider taking the approach of writing "I'm sure what you say is true, but it must still be verified to be used", jguk 21:38, 14 January 2006 (UTC)

Njaa... the problem is more the other way around. Every editor who spends some time on a really touchy subject, like those on the borderline between science and politics will have met this. Take this mock article:

Origin of Foo

About the origin of foo, some say X and some say Y.

Those who say X hold that... (long, brilliant presentation of X follows).

Those who say Y hold that ... (long, brilliant presentation of Y follows.

Now, assume that User:Bob hates Y. He will use a lot of screen time to convince the other editors that Y is in fact a lie (producing statistics, and papers, and really bullet-proof arguments and whatnot). And lies have no place in an encyclopedia. After all, the very inclusion of the arguments pro Y in these pages give a false impression of their merits. User:Bob wants to remove the entire paragraph.

Now, the point of Wikipedia is that User:Bob cannot do that. Even if all editors of Origins of Foo agreed, the paragraph explaining Y has to stay. But User:Bob will go with very disruptive edits, including "but this is disputed, see [5]" after each and every sentence in the paragraph about Y. Because he cares about truth. Who knows, I might even agree with him! But the result is (1) lots of time wasted on the talk page arguing about the relative merits of X and Y, a discussion that belongs to the pages of some scientific journal (2) A really weak paragraph about Y. Especially, the page degenerates to a discussion forum about what is the truth about the origins of Foo.

That's what I don't like. And that's the discussion that I can effectively end by pointing to "verifiability, not truth". Wikipedia is simply not a very good medium for arbitrating truth, not even for weighing the evidence. Because truth is far from agreed on. (Read Hypnosis. Or Creationism. Or Race and intelligence. Good luck arguing about Truth on those pages.) That doesn't mean I don't care about truth. But as a reader, I want the best possible presentation of viewpoints X and Y, and their counterarguments, and a description of their reception ("In a 1975 study, 86% of all biologists queried agreed or agreed strongly with X, and 96% rejected Y." That's useful.) Then I arbitrate the truth. The editors shouldn't be bothered with deciding or agreeing on that Y is in fact just a load of bullocks. Arbor 12:03, 10 March 2006 (UTC)

[edit] Reputable sources

I think a line might be added which indicates that "reputability" is a factor of the claim being justified from it. That is, a major Creationist website is not a reputable source for claims about biological evolution, but it is a reputable source for information about the Creationist POV. Similarly a major crank website about the Apollo Moon hoax is not a reputable source for whether or not people landed on the moon, but is a reputable source for information about the hoax theory itself. --Fastfission 21:40, 4 January 2006 (UTC)

I agree with what you are getting at, though I think such a statement is more appropriate on Wikipedia:Reliable sources rather than Wikipedia:Verifiability. I have removed most of the digression on reputable sources, leaving the reader to go to Wikipedia:Reliable sources, where the matter can be more fully explored, jguk 11:26, 8 January 2006 (UTC)

[edit] 'Rare occasions'

We do have one general rule though: personal websites, blogs, and other self-published material, are not acceptable as secondary sources (that is, sources about third parties or subjects other than the author), except on the rare occasion that a well-known person, or a known professional journalist or researcher in a relevant field, has set up such a website.

  • I edit in an area where these rare occasions are not very rare, and the reputed sources are not always reliable. So I would like to know whether personal websites, blogs, and other self-published material can also include usenet and message board posts - also assuming that the person posting it is sufficiently important and reliable, and that the editor is good enough to make a judgement about it. Tintin Talk 22:51, 4 January 2006 (UTC)

This is really an issue about what constitutes a reputable source, and should be discussed in Wikipedia:Reliable sources rather than on the Verifiability page. I've therefore removed that guidance from this page - but left a clear link to the Wikipedia:Reliable sources page, jguk 11:24, 8 January 2006 (UTC)

[edit] Seriously, though

Editors adding new information into an article should cite a reputable source for that information, otherwise it may be summarily removed by any editor. Do we really want to summarily remove 95% of Wikipedia? This isn't sensible. Surely, surely, SURELY this should be reduced to something like "Suspect information without a reputable source may be summarily removed to the talk page by any editor." I challenge you: Visit Random Page 10 times. I'll be generous, I'll say each paragraph is "one piece of information". Now tell me how many paragraphs you have that contain at least one cite, and how many contain no cites. Then how many articles had no paragraphs without cites. This just isn't feasible. Stevage 23:39, 4 January 2006 (UTC)

I believe that the discussion has been that this will be easier to aply to new edits, at least for the interim. I've starting doing it when I see new edits that put in unsourced material. I just revert or move the new material to the talk page with a note that sources are required. -- Dalbury(Talk) 23:50, 4 January 2006 (UTC)
In my experience, this is a shift to Wikipedia practice, and I'm not sure it's desireable. Unless the addition is controversial or very suspect, a better way is to leave the information on the page AND request a source on the Talk page. If no source is forthcoming after a period of time, and you can't easily verify the info yourself via Google, only then should it be removed. We should, of course, encourage all edits to be sourced, but to make sources mandatory significantly increases the overhead of editing Wikipedia, and we want to allow people to participate as easily as possible. Wikipedia is a work in progress, and as such, it's quite acceptable to add the info, and then source it later. — Matt Crypto 10:59, 9 January 2006 (UTC)

Matt, one key point is that when you add info to WP, you know where you got it from. If you don't tell us that we don't know whether you have made an honest mistake, misread the source, or whether, at a later date, it is out of date. Nor can we go to the source ourselves to read it in context. In short, it is impossible to assess the information you add. It really doesn't take long to say on an article page, talk page or an edit summary where you have got the info from, and this information is invaluable later, jguk 21:44, 14 January 2006 (UTC)

[edit] This is a policy change

I agree with the general thrust of this (not all details), but this isn't a redraft for existing policy, its a substantially new policy. It's also badly stated in terms of what's non-negotiable (verifiability is definately non-negotiable, but some of what said here is certainly up for debate). A large, well publicized poll (after substantial discussion), reaching a strong consensus, would be required to endorse this. --Rob 21:05, 5 January 2006 (UTC)

The intention is to restate policy, but in a much clearer way than at present. You don't say where you think that objective isn't being met - it would be useful if you could explain, as either something's missing from the redraft or there's a misunderstanding about what current policy is - whichever way it is, I'd like to see it resolved.
I've changed the term "non-negotiable" to say instead that the policy has "mandatory application". That's definitely true, whilst also allowing for some discussion over the exact wording of what is in the policy box (although the general principle that ideally everything on WP should be sourced certainly is non-negotiable).
Finally, WP works by consensus, and this is intended to be a restatement of policy, not something fundamentally different (as noted above, if you think this objective isn't being met, please say why). Once there has been time to take on board views and comments offered, and given that this page has been publicised on Wikipedia talk:Verifiability, IRC, the mailing list and the Village Pump, if it has consensus it will be moved to the main page - remember, polls are evil: However, there is, of course, no rush - all constructive comments are most welcome, jguk 22:17, 6 January 2006 (UTC)

[edit] Government sources are not 3rd party, but are sometimes appropriate

Regarding: "If an article topic has no third-party sources, Wikipedia should not have an article on that topic."

Ideally, of course, third party sources should always be used. But in some cases, a government is the only source of *certain* information about some part of itself. One has has to use common sense sometimes, if one wishes to beleive what's said. Some government statistic/census beareaus are highly reliable, while others are not, and we can't make the blanket staements. Also, in many case 3rd party sources are readily available, but seem rather redundant. 38th Canadian parliament could easily cite 3rd party (non-government) sources, but doesn't, as to do so would be utterly redundant. The single link from parliament's web site, is sufficient to verify the information, and is probably a better source then a private newspaper article. --Rob 21:22, 5 January 2006 (UTC)

A fair point. The important thing is that the source is reputable - ie it can be relied on to be accurate about the bit of information it is referencing. The comments on "third-parties" is a slight over-exaggeration, jguk 22:19, 6 January 2006 (UTC)
This also concerns me, as it has the potential to take this to an unhelpful extreme. There are plenty of sources, particularly online, where getting the information from the horses mouth is likely to be the best source - and may be the only source. As an example, if we write an article on a school, and we cite the a school history page on their website for a history section, is this really likely to be considered an unreliable source because it's not third-party? Such an attitude would be, IMHO, most unhelpful, and I'd like to either see some sort of safeguard against this or for this to go to a vote so it can be struck down. Ambi 12:05, 8 January 2006 (UTC)

[edit] Missing details

Some details in the original meaning have vanished.

(SEWilco 03:27, 6 January 2006 (UTC))

Those are good points, but they really belong in Wikipedia:Citing sources not on Wikipedia:Verifiability. I haven't checked WP:CITE, but if they are not there, I'd certainly support adding them, jguk 22:10, 6 January 2006 (UTC)

[edit] Any further comments before this rewrite goes live?

Are there any further comments before this rewrite goes live? I don't want to rush things, I just want to be clear as to whether there are any outstanding points, and if so, what they are, jguk 11:29, 8 January 2006 (UTC)

See my concern above. Unless that is allayed, I would have great concerns about this going anywhere without a full vote. Ambi 12:05, 8 January 2006 (UTC)

Thank you for your comment. Hasn't the text you were concerned about above now been removed, or am I missing something? Also, do we really need a full vote on a rewrite if there is no real dissent (which is my intention before allowing this to go live)? If people have an issue with things, let them say now and we'll discuss it, rather than stay silent and vote a straight "oppose" with no commentary later, jguk 12:10, 8 January 2006 (UTC)

On second thoughts, it might have been - if so, I apologise. The one lingering thing that concerns me about this is that it leaves no room for common sense - removing something when there's no reason to suggest that it's incorrect, simply because no one has even tried to do a cursory Google search, is not helpful. Ambi 12:18, 8 January 2006 (UTC)

All Wikipedia policies have to be applied using common sense:) Also, this policy doesn't stand in isolation, the policies on behaviour are still there to deal with problem cases. In the case you say, common sense would suggest (assuming the information is not potentially damaging or libellous) that a "dubious" tag be added, or for the info to be moved to the talk page, to allow people to try to look for information. Mind you, if the information is newly added, I think, if it is unsourced, it is perfectly acceptable to remove it immediately together with a polite request not to re-insert it without also providing a reputable source, jguk 12:23, 8 January 2006 (UTC)

I like this version definitely better than the present wikipedia:verifiability, primarily because it's shorter, clearer, less depending on complicated hermeneutics. Nonetheless, might I suggest the following:
Why not provide:
  • {{details|wikipedia:citing sources}} immediately under the "Cite sources" section title, and,
  • {{details|wikipedia:reliable sources}} immediately under the "Reputable sources" section title?
IMHO this would show better how these guideline and policy pages are connected. --Francis Schonken 12:30, 8 January 2006 (UTC)
Thanks, I have amended the project page to take account of your suggestion, jguk 12:57, 8 January 2006 (UTC)

Maybe too picky, but could the word onus be replaced by burden/task/work or something like that. I speak english, even legal english and I had to think a few seconds on the meaning. I can imagine that some non native speakers wouldn't understand it. Garion96 (talk) 13:11, 8 January 2006 (UTC)

Obligation, good change. Garion96 (talk) 21:36, 8 January 2006 (UTC)
One comment I have is about the links to guidelines on biographical articles. The guideline on biographies of living persons is still a proposal. There are also guidelines on fiction, music and numbers, and proposed guidelines for corporations and websites. If the policy is going to link to the guidelines on biographies (criteria for inclusion of biographies and auto-biography), I think it should also link to other current guidelines. If this policy links to proposed guidelines, it should note that they are proposed guidelines, and while they may be of interest to individuals making decisions about notability, they currently have no force in wikipedia. -- Dalbury(Talk) 15:07, 8 January 2006 (UTC)
Thank you for your comment. Aren't those notability discussions though, rather than Verifiability discussions? The comments on Biographies of living persons, if it ever goes live, will stress that everything should be verified to death - it's not about whether we should have a biography in the first place - that's why it's on here and the others aren't. Please correct me if I've gotten something wrong here, jguk 17:19, 8 January 2006 (UTC)
First, this is not a show stopper for me. I'm content to see the proposal go live as is. However, it currently points to [[Wikipedia:Notability (people)]] (redirected from [[Wikipedia:Criteria for inclusion of biographies]]), but not to any other notability guidelines. I suggest it would be better to point to [[Wikipedia:Notability]] for questions of notability, and drop [[Wikipedia:Criteria for inclusion of biographies]]. -- Dalbury(Talk) 19:02, 8 January 2006 (UTC)

I oppose this going live without a full vote. It's a change in core policy, with signficant impact, as it declares a substantial number of articles to be unacceptable topics, due ot a lack of 3rd party sources. I'm not saying this is necessarily a bad change in policy, but it is a change in policy. Lessor changes in policy have goten a full formal vote, and so should this. --Rob 16:43, 8 January 2006 (UTC)

The "change" you state is already part of WP:V, although I accept some choose to ignore it (and often end up on AfD for their troubles:) ).
If anyone has objections to the proposed rewrite or suggestions for improvements, they really should say so here and now. I'm very much in favour of looking for consensus. The problems with votes are manifold: (1) people might object because they don't like one clause or they'd really like to see one new sentence added - that's silly - we can work round those points in the same way as we've been doing; (2) people might object because they don't like WP's Verifiability policy, full stop; (3) it gives equal (and once we include sockpuppets, unequal) weight to WPians who have never made any substantial edits or contributions to WP and are unfamiliar with our processes; (4) m:Votes are evil;). The proposed rewrite has been publicised twice now on Wikipedia talk:Verifiability, the mailing list and the village pump. It has been added to RfC and it has been publicised on IRC for many days now too. Anyone who wishes to express a view and suggested improvements is encouraged to do so, and the proposal has changed substantially [4] as a result of those views. If you think it needs fuller publicity - then please publicise it further, though I'm not aware of having missed any further way to publicise it. Also, there's no hurry to adopt it - constructive discussion is good. But I do believe that it is fair to ask those who disagree with aspects of it to say so and allow the draft to change, if necessary, to accommodate those views rather than come in at the thirteenth hour and vote a simple "oppose" - indeed that approach would be very disrespectful to those seeking to improve our statement of our Wikipedia policy, jguk 17:16, 8 January 2006 (UTC)

Actually, one version has it, another doesn't, or worded in meaningful different manner. People are constantly changing WP:V and other policy documents, and I've lost track of what's old and what's new. As said previously, official sources from government entitities are often accepted as authoritative sources, and accepted, though not 3rd party. I find it absurd we're constantly changing core policy. It seems policy is now directed at whoever edits the policy document, not by broad consensus. --Rob 17:45, 8 January 2006 (UTC)

The bit about 3rd parties has been removed, hasn't it? (Please correct me if I'm wrong.) As the discussion on what is or isn't a reliable source should be on Wikipedia:Reliable sources not Wikipedia:Verifiability, I removed that bit and instead directed the reader to Wikipedia:Reliable sources. On your final point, that is why this rewrite is progressing slowly and with wide publicity, so that it can proceed with consensus, jguk 17:59, 8 January 2006 (UTC)
In the current temp version there is the text "If an article topic has no third-party sources, Wikipedia should not have an article on that topic". Now, we have lots of government entitties where the only sources cited is the government itself. It's true this version of the main page says essentialy the same thing. When I claimed this temp version made a change, I was going on memory, comparing it to the older main version. There have been one hundred revisions of WP:V since beginning of December. I can't possibly carefully read them all. I doubt anybody can. This temp version is also changing to much *if* you want to copy it over now. I submit, we should do the following: freeze the main version, freeze the temp version for at least 5-days before its copied over; and vote on whether to overwrite the main with the temp. Either way, freeze the main version after. Wikipedia should have about two or three core policy pages which rarely changes (by rarely I mean less then once a month). Anyways, I'm quite sorry, if I've mixed up my facts above, or missed something. I think people who follow this all closely, don't realize the confusion that exists. --Rob 18:21, 8 January 2006 (UTC)
The only adjective now applied to "sources" in the project page is now "reputable". If I've missed any instances by mistake, change them, jguk 21:59, 14 January 2006 (UTC)
Before new policy goes live, a straw poll may be helpful to determine consensus. xaosflux Talk/CVU 03:37, 11 January 2006 (UTC)

Some form of ratification process will be needed. However, I do think it's reasonable to ask anyone who opposes either the whole re-write or aspects of it to say why that is the case before that process starts. It may very well be possible to come to an accommodation - after all, the best wording will have the input of many people who have different outlooks in life, jguk 21:59, 14 January 2006 (UTC)

[edit] I don't get it

What happens when this doctrine goes live? Looks the same to me. Lotsofissues 23:20, 8 January 2006 (UTC)

It's meant to be just a re-write, not a change in policy, so I'm glad it looks the same to you - that's the intention. What happens when it goes live is that it will be easier for those unfamiliar with the policy to understand what it is and where to go for further guidance, jguk 22:00, 14 January 2006 (UTC)

[edit] Good idea, but...

It is too damn hard to cite resources... add a way in Mediawiki to cite sources MUCH easier and then, and only then, will I support this.. Hell, a text box right below "edit summary" that asks you where you got the information... perhaps a dropdown box with a bit of AJAX/DHTML to let you easier cite things... Also, there needs to be an implimentation guideline written up before this goes live... because I don't know if I should be ripping up every article on the place or not... --Nick Catalano (Talk) 18:39, 9 January 2006 (UTC)

Why is it so hard to add sources? You shouldn't be adding information into Wikipedia unless you are looking at a good source. You are not supposed to be relying on your memory. -- Dalbury(Talk) 19:22, 9 January 2006 (UTC)

The only "hard" part of adding sources is the trivial matter of formatting so it looks "proper". Simple solution: don't worry about formatting. Add all the relevant details of the source, in whatever format is convenient/practical for you. Let those obsessed with formatting worry about formatting. I agree that the software could be designed much better, but it's not. What counts is whether people can find where you got your stuff. If you know your sources, citing them should be easy. If you don't know your sources, you're not ready to make an edit. Also, I find half the time one adds stuff, it's based on references already cited, and I don't think it's necessary (as some do), to put a redundant footnote next to every sentence, as long as the references are cited in the article. --Rob 20:01, 9 January 2006 (UTC)

True enough. But if there are no existing references listed for the article, then you really do need to provide sources for what you add. I would also note that if are adding references reguarlarly, you quickly learn how to enter references in pretty good form. And there is now a book reference citation generator available which produces plain Wiki text that you can paste into an aticle. See [5]. -- Dalbury(Talk) 20:21, 9 January 2006 (UTC)
We have a little different experience then. I do cite references regularly, and what I've learned is there's a wild west, and lack of standards. And each person "knows" their approach is best, and "the" approach. That's why, I've adopted the approach of citing as best I can, and not yet worrying about form. p.s. Is there one of those generators for a magazine or newspaper article? --Rob 21:15, 9 January 2006 (UTC)
Sure there is a generator for Newspapers, etc: Easybib.com --Nick Catalano (Talk) 21:40, 9 January 2006 (UTC)

OK, lets do an example of how many cool ways an anonymous editor who has no experience with Wikipedia but has something useful could cite:

[edit] Example Sources

  • james book ISBN 19239102 Author: Cool Guy 102390
  • Author: Goodguy Book: Biography of Cool Guy
  • Interview with Cool Guy, New York Times, January 12, 2005
  • interview with Cool Guy, NY Times, 2005/1/12
  • The source of this entry is http://www.coolguywebsite.com/biographyexample.htm

Sorry... no... I shouldn't have to look up a style guide every time I want to make a change... I should be able to simply add the URL or Citation information below my entry and it should show up under the sources... Or on a seperate page that gives the sources... it isn't that difficult to impliment and makes it acceptable... Also, I still don't have anyone telling me how this is going to be implimented with current articles... If it means I can blank every article I find on Wikipedia well hot damn lets do it! In all seriousness, lets get our heads out of our asses and think over some of the difficulties of implimenting a major change that will overhaul the entire editing experience here at Wikipedia --Nick Catalano (Talk) 21:40, 9 January 2006 (UTC)

The current WP:Verifiability shows that citations with more detail are preferred, so you should provide as much as you know. The material is more likely to be challenged if there is not enough detail for finding the source (including if the URL no longer works and nobody can find what you were trying to link to). As with everything in Wikipedia, it is expected that others will improve the citations (except for me because the ArbComm doesn't follow policies such as this). (SEWilco 04:41, 11 January 2006 (UTC))

Actually a new way to automatically build reference lists from inline citations was just added to MediaWiki a couple weeks ago. Take a look at http://meta.wikimedia.org/wiki/Cite/Cite.php. Kaldari 22:52, 10 January 2006 (UTC)

[edit] This policy change could lead to massive disruption of Wikipedia

Specifically point 3:
3. The obligation to provide a reputable source is on editors wishing to include information, not on those seeking to remove it.
I see this point of policy being used to delete any information that POV-pushers don't agree with, even if it's been in an article for years. If we must have this sentence (which seems mostly redundant with the previous one), it should clarify that this policy is not retroactive, i.e. it only applies to new material. Otherwise someone could stroll over to an article like Persecution of Jehovah's Witnesses and just blank the entire article and the blanked version would have to stand (since there are no references). Wikipedia is full of good articles with no references. This is due to the fact that Wikipedia did not originally emphasize the creation of references, nor provide an easy way to create them until quite recently. Do we really want all those articles blanked? If you consider individual article sections, virtually every article in Wikipedia is vulnerable. I certainly applaud encouraging editors to add references to articles, especially when adding new material, but can't we do it in a way that will be less disruptive? Kaldari 23:08, 10 January 2006 (UTC)

Actually I lied, Persecution of Jehovah's Witnesses has 2 inline citations. It was just a random example though. I'm sure there are plenty of controversial articles which have few or no references. How about Foreign relations of Israel (1 citation) or Vietnam War (only 4 citations in an 85K article)? None of those articles have a references section. Even many older featured articles would not survive this policy change. Kaldari 23:38, 10 January 2006 (UTC)
I don't think the policy (old or proposed) means every sentence needs it's own inline citation. It means collectively the cited sources must back up all that's in the article. If none of the sources mentioned in the article, back up a claim, and I challenge it, I may remove it. That's good. If I link to imdb in an actor's bio under "External links", I needn't cite every acting job he's had with an inline citation. The articles you mention are good examples of controversial topics, where removal of material quickly results in it's re-addition with better sources. Frequently articles go unattended/unfixed until somebody "shakes things up" with a large textual removal. Tempers may flare, but the problems are generally fixed. I do think we must put the onus on the adder of the info, as it's often impossible for the "challenger" to ever "prove" something is false. For instance, how do you "prove" somebody wasn't suspected in the Kennedy assasination? You can't. However, I do agree with you, that we should be softer on older established material. Keeping a year-old paragraph one more day, will do little extra harm, but keeping a new unsourced claim just one day, could do great harm, as we spread new rumors/hoaxes that way. --Rob 23:57, 10 January 2006 (UTC)
The way this proposal is worded makes me fear neo-nazis blazing through and systematically deleting every unsourced paragraph about the holocaust. Or a Christian fundamentalist deleting every unsourced paragraph about evolution and then the evolutionists fighting back by deleting every unsourced paragraph about Jesus (and there are certainly a few). Of course the end result for Wikipedia may be positive, but in the short run it will be armegeddon. There needs to be a certain allowance or grace period for older material IMO. Kaldari 04:37, 11 January 2006 (UTC)
I assert that's already been attempted continuously (almost) since our founding. That's why every important word on the holocaust is well sourced (prove me wrong). Same deal with other extremely controversial topics. It's more likely the neo-Nazis, and other hate groups, are some of the ones who already spread maliscous rumors, by making allegations against people they hate, without proper sources. If we make those wishing to remove the statements prove them false, the statements will be left in for long enough to acheive their ends. In your Jesus example, I disagree. We don't (shouldn't) have unsourced claims of Jesus. We do state what *others* claim to be true about his life, and hopefully show alternate views as well. If we're currently stating as fact (not as an attributed opinion) something about the life of Jesus, then I do hope its removed promptly, and I don't much care who does it. Basically, I'm saying people "ready for a fight" have always done what you fear, which is mass removal of unsourced stuff they don't agree with. But, many non-controversial articles are left unsourced/unreliable, because somebody sees unsourced material, doubts it, but isn't "ready for a fight". We want to tell that person, acting in good faith, that they should go ahead, and remove the content. The person acting in bad faith, has already acted, and won't be reading this policy document. --Rob 07:35, 11 January 2006 (UTC)

I share this concern. Also, I would like to see a clear summary of what this changes about policy. I don't want to have to do my own point-by-point comparison. Offhand, this seems to go farther in a direction about which I have already raised concerns. -- Jmabel | Talk 08:34, 11 January 2006 (UTC)

Good idea. It would be nice to have an item-by-item explanation of the changes rather than just "Do you like this document?". (SEWilco 04:21, 12 January 2006 (UTC))

At the very least can we change the following statement: "If the article has many unsourced statements that have been there a long time, you may request sources on the talk page before removing them" to: "If the article has many unsourced statements that have been there a long time, you should request sources on the talk page before removing them" as this has always been standard practice and seems necessary to comply with Civility and Assume Good Faith. Kaldari 14:26, 12 January 2006 (UTC)

I think it is more likely that the new page would be used as a justification to go on a slash-and-burn campaign against genre-fiction-related entries. Consider (to pick a random example) the Chewbacca article. No references whatsoever. Most entries regarding fan culture, be it anime or Rugrats, are in a similar state. As current discussion over at WP:VPP suggests, there are a few folks who would love to delete all such entries. The explicit wording of item 3 would give them the justification (and possibly the motivation) to do so. Ξxtreme Unction|yakkity yak 16:05, 12 January 2006 (UTC)

Those aren't totally unsourced, because the widely known published works of fiction are primary sources, which may be used (but shouldn't be used alone, and explicit citations aught to occur). However, if somebody removes material, and its added back with better sources, more explicit sources, that would be good. I dislike reading such fiction articles, and not being able to whether something is nonsense vandalism, or nonsense fiction. Cleanup would be good, even if it's induced by temporary content removal. --Rob 17:23, 12 January 2006 (UTC)
And sometimes the only way to get the attention of someone who can supply sources is to remove the material. Most requests to cite sources are simply ignored, or reverted without bothering to try to supply sources. -- Dalbury(Talk) 17:58, 12 January 2006 (UTC)

[edit] Dubious sources

Dubious sources are sources that are reliable persuant to our policy, but which must be named when used as a sole-source of fact (compare -> "Something is true (source: New York Times)" vs. "The New York Gossip Magazine said something is true. (source: New York Gossip Magazine)." This section of the policy is removed from the current draft. I must request it be returned. Hipocrite - «Talk» 22:04, 11 January 2006 (UTC)

The reputability (is that a word?) of sources is discussed elsewhere - namely in Wikipedia:Reliable sources, which is why it isn't on this project page. I agree with your point that we should have guidance on that point - it should, however, be in Wikipedia:Reliable sources, which is link to very clearly from this page, jguk 22:11, 14 January 2006 (UTC)

[edit] Inappropriate citations

It is misleading to assert that reputable sources are a sure sign of good editorial scholarship. Bad citations can point to reputable sources and while remaining inappropriate in a number of ways. These include:

  1. Misrepresenting the facts of a source.
  2. Misrepresenting the opinions of an author.
  3. Citing a tangentially related fact and claiming it has direct relevance to the article.
  4. Burdening a primary (factual) source with interpretive statements.
  5. Naming one source in the text while citing a different source in the associated footnote.

Durova 21:40, 14 January 2006 (UTC)

I don't think we are asserting that reputale sources are a sure sign of good editorial scholarship, are we? I think it's implicit that the statement a source is being used to reference must be entirely consistent with what the reference says, otherwise it's just not a reputable source for that information. As this discussion is more to do with the reliability of sources, perhaps Wikipedia talk:Reliable sources would be a good place to have the discussion, jguk 22:16, 14 January 2006 (UTC)

[edit] Vandalism?

I strongly objected to this edit which said, in part "Indeed, removing large chunks of information without asking for sources first may be deemed vandalism.". That's rubbish. Good faith edits are not vandalism. In the case of very recent additions of highly dubious claims, its best to promptly remove something. In cases where there is potentional defamation, I'll remove unsourced accusations promptly. We forget, that Wikipedia can be, and is, used to intentionally spread rumors, and some things simply shouldn't be allowed to sit around. Key point: Adding misinformation to Wikipedia is vandalism, removing what appears to be misinformation is not. --Rob 20:35, 17 January 2006 (UTC)

What I'm trying to do is take account of people's concerns that those acting in bad faith will use this policy to remove large chunks of info on WP. After all, large chunks of WP are unreferenced - now these all should be referenced, but it isn't going to happen overnight. I agree good faith editors will remove dubious information on occasion - but the emphasis will be on trying to encourage people to supply evidence to support assertions that have been added in good faith rather than see their outright deletion. It's a judgment call as to whether to remove the info immediately or open a discussion on the talk page. The real message is "use common sense", which is always hard to phrase. Please feel free to propose wording that would help address the concerns raised above, but that better hit the mark, jguk 20:52, 17 January 2006 (UTC)
The problem is that those "bad faith" users have been doing what your worried about for a long time. They're not the people reading this page, and won't care what it says. This page is read by good-faith contributors. We to-often, frighten good-faith editors (like RC patrollers), into simply leaving dubious unsourced material. People trying to remove bogus info often feel like they're walking on egg shells, because they can be accused of "vandalism" for removing something somebody thinks is essential. Its often tempting to simply leave dubious unsourced stuff alone. This is why so much unsourced stuff is left around. It's far to easy for sneaky vandals to plant "plausible" stuff, that's mixed in. We must encourage the removal of unsourced material if doubted. No good editor should ever feel worried that they will be called a "vandal". Bad faith editors couldn't care less what you call them. I find that word, "vandalism" extremely offensive in the context it was used. I saw nothing of value in the text I removed. Also, quite frankly, many editors who regularly add huge amounts of seemingly unverifiable information, won't stop doing so, until they find some information removed. --Rob 21:01, 17 January 2006 (UTC)
Good statement of the problem. -- Dalbury(Talk) 22:40, 17 January 2006 (UTC)
Bad faith users can and will twist policy to suit their agenda if it exists. This allows the removal of perfectly good - and I stress not dubious - information simply because it isn't sourced, which provides a handy tool in the bow of anyone willing to be a particular pain in the ass. Ambi 23:16, 17 January 2006 (UTC)
And "bad faith" editors can argue that they thought the material they removed was "dubious". The solution, of course, is to provide reliable sources. Just leave it that unsourced material can be removed. If something that is verifiable is removed, it can be put back in with sources cited. Keep it simple, assume good faith, and then deal with anyone who acts in a disruptive manner. -- Dalbury(Talk) 23:43, 17 January 2006 (UTC)
How do you propose to deal with people who are causing disruption (for example, by blanking all articles without sources) but are fully in line with policy? Kaldari 00:11, 18 January 2006 (UTC)
I think anyone who makes a habit of blanking articles simply because they have no sources cited will quickly run afoul of Wikipedia:Etiquette, Wikipedia:Civil and other guidelines. I just don't want to worry too much about protecting articles that violate policy by not citing sources. I think a little edge of jeopardy for such articles will be a goad to get sources cited in them. -- Dalbury(Talk) 00:57, 18 January 2006 (UTC)

Do you have suggested wording of your own, Ambi? jguk 23:34, 17 January 2006 (UTC)

I'd just like to see a little more focus on removing information if it is dubious, rather than giving anyone who sees an unsourced article carte blanche to trash it until the author provides sources. If there's any reason to doubt it, then fire away, but at least a cursory Google search should be made before firing. Ambi 01:27, 18 January 2006 (UTC)
Just because you can find it in a Google search doesn't necessarily mean it isn't dubious. To me it keeps coming back to credibility. Is an article credible if it isn't backed by credible sources? I don't think so. How do you prove that an item isn't dubious unless you supply a credible source supporting it? I'm afraid of protecting unsourced information because a majority of editors "think" it's correct. I see a lot on things in Wikipedia articles that I have doubts about. If those items are unsourced, it is not my responsibility to prove those items are wrong; it is the responsibility of whoever added the item to the article, or wants to keep the item in the article, to provide credible sources for it. "Dubious" is a judgment call, and arguing about whether or not an iotem is dubious detracts from the need to cite sources. -- Dalbury(Talk) 02:26, 18 January 2006 (UTC)
We could redefine things along the lines of:
  • Removing material without an explanation (edit summary or talk page) is vandalism.
  • Removing material as "unsourced" without an explanation (something longer than "unsourced") on the talk page is vandalism.
  • Repeated removal as "unsourced" (and no other valid reason) of more than one paragraph of material that has been present for more than a few days without prior notice or discussion could be interpreted as "bad faith".
  • Restoring material removed as "unsourced" without providing sources will require an explanation on the talk page of why it is vital to do so even though sources aren't being provided (although I'm having problems thinking what a valid reason would be).
-- Dalbury(Talk) 12:23, 18 January 2006 (UTC)

If you are removing material because it is unsourced, what more is there to say than that it is "unsourced"? There's nothing more to say on the matter. OK, you could say "unsourced, please see WP:V" or "unsourced, please do not re-add without providing a reference, see WP:V", but I can't think of much more that can be said, jguk 18:15, 18 January 2006 (UTC)

OK, good point. I'm trying to work through to a formulation that will allow material to be removed simply because it is unsourced without being seen as creating an open season for vandalism. I don't like any formulation that says only "dubious" material can be removed. That seems to be protecting unsourced material from ever being removed, and I think that is bad for Wikipedia. I think we need a formula that allows known bad material to be removed immediately, and other unsourced material to be removed after adequate notice. I see this as one step in changing the default Wikipedia culture of shovelling all kinds of stuff into articles without providing sources. That makes Wikpedia more like a giant communal blog than an encyclopedia. -- Dalbury(Talk) 18:36, 18 January 2006 (UTC)

[edit] Reputable sources and reliable sources

This article uses both expressions more or less as synonyms. This can be confusing to some people. As there is an official Wikipedia policy Wikipedia:Reliable sources, I changed the "reputables" to "reliables", as reliable is the broader application (any reputable source is reliable, but not every reliable source is reputable). Reputable is a factor in scientific publications, but there are many subjects in Wikipedia which are not scientific, and there reliable source fits better. Irmgard 10:12, 20 January 2006 (UTC)

I prefer "reputable". Nature is reputable, but not always reliable (they have made numerous retractions over the years). Non-scientific areas also have reputable sources - for example, in cricket, Wisden, Cricinfo and CricketArchive would all be considered reputable, jguk 07:42, 26 January 2006 (UTC)
My point is not which I prefer, but that the existing policy is named "reliable sources". Using different terms for the same thing in a set of articles is confusing to the reader (that's one of the basics of technical writing). From this point of view, you can either change the "reliable sources" article or use the same term as they use - but if you talk of Z, don't call it X in one place and Y in another place.
Moreover, there is a good reason, to talk of reliable and not of reputable sources: Reputable sources are in that policy only discussed in regarding academic publications - where reputable is the better word - but there are many more kinds of sources for an encyclopedia. Take the local history of some place in the outbacks compiled from local archives, graveyards and family documents by two diligent but not otherwise distinguished local teachers. Solid work, probably more reliable regarding local data than anything written by a historian of renown. But which reputable source would publish it? And the reason why no reputable journal does is not, that it is bad quality but that it is of too limited interest for their readers. Ok, the teachers get the local newspaper to publish their research in a booklet which is printed in 1000 exemplaries and finds its way also into some libraries. A reputable source - no way. A reliable source - yes. Useful information - yes. I own a "History of the Canton of Zurich" - folio format, 3 volumes each above 500 pages, written by a team of academics of several disciplines, in itself sure a reputable source. And if I look at the list of sources they give, its four pages of original sources, and ten pages with some reputable authors and sources and a majority of those very very specific secondary sources which have no chance to get published in a reputable journal: "Social history of doctors in the Zurich countryside 1700-1850" "Students from Zurich in Halle in the 19th century", "History of the community of Zell" "History of the paper mill of Zurich" "Tiled stoves from Winterthur used in town halls" . You can find the same kind of sources in Wikipedia as well, e.g. "First Minnesota Volunteer Infantry",
As the majority of Wikipedia articles is not about science or academic subjects, I strongly plead to use "reliable" - it fits the reality of the average Wikipedians better. --Irmgard 00:05, 27 January 2006 (UTC)

[edit] Information does not always need to be supported by citations to be verifiable

I think the wording as it was placed too much emphasis on citing a source for every fact. It is very much Wikipedia policy that information is verifiable, but this does not mean that every single little statement needs a citation. Many kinds of fact that are in an encyclopedia don't need a specific source to be verifiable - they are common knowledge of experts in the topic. The vast majority of basic facts in a scientific field, for example, are easy for an expert in the topic to look up in an appropriate reference. If an editor doesn't know where to verify a fact like "the atomic number of Einsteinium is 99", they frankly don't have the expertise to reliably edit or check that article in the first place, and a citation would add very little of use. Citing sources is very important for controversial, surprising, or little known information. But for many basic facts, from science to geography to sports results, most statements are readily verifiable without citations.

Ultimately, having a citation for every fact on Wikipedia is neither practical nor desirable, and attention should be focussed on providing citations where they are genuinely needed - i.e. to assist editors in verifying information in more difficult situations.

I've adjusted the wording to emphasise the need for verifiability rather than citation as the policy; comments welcome. Enchanter 03:16, 26 January 2006 (UTC)

What may be 'common knowledge' to you may not be 'common knowledge' to others. I'm not arguing that every single statement needs it's own citation, but the article needs citations for all of its substantive content. It should be the burden of the editor introducing a statement to provide a source. I don't have time to go chasing down a source for every edit I see added to articles, especially as so much of my time is spent providing sources for my edits. I resent the idea that someone else is free to add whatever junk they want to an article, and it's somehow my responsibility to prove its not verifiable. If we protect irresponsible edits by making it too hard to clean out unverified information, only Wikipedia will suffer. -- Dalbury(Talk) 03:47, 26 January 2006 (UTC)
My edit stated very clearly that it should be the responsibility of the person adding information to demonstrate that it is verifiable. My point is simply that adding citations and references is in many cases neither sufficient nor necessary to make a fact verifiable. Facts that are widely available in reference sources can be easily verified without providing a citation. In many cases, providing a citation is of no practical help in verifying an article - it is often easier for another editor to verify an article with whatever reference materials they themselves have to hand than to go searching for the particular publication given in a citation.
What is common knowledge does indeed vary from person to person. However, it is reasonable to expect that someone checking an article has a reasonable expert knowledge of the topic. If someone is not an expert on a difficult, technical topic, they won't be able to verify it no matter how many sources they are given. Giving sources for facts that are widely known by experts in these fields does nothing to assist verifiability, and can easily just build up into a useless clutter of poorly chosen references at the bottom of an article.
Enchanter 02:46, 27 January 2006 (UTC)
I just want to say that I very much agree with what Enchanter says here. It very much echoes some of the concerns I have mentioned earlier on this page and elsewhere. Verifiability is not the same as providing a citation. I think we are doing editors and readers alike a disservice by 1) conflating the two related but distinct concepts and 2) by implicitly (or in some cases explicitly) endorsing a culture of "delete first, ask questions later". older ≠ wiser 03:16, 27 January 2006 (UTC)
I think we should bear in mind that there's such a thing as over-citation. Adding a tag such as citation needed can be used as a way of promoting a POV by calling the the facts most often quoted by the other side into question, and in any case the tag is just plain ugly. Adding a reference to a footnote to every factoid makes the article read like a research paper (or sometimes like a college assignment) not an encyclopedia article, and again subtly undermines the credibility of the material. An excessively long list of references doesn't aid verification, just the opposite, it conceals the useful ones among the junk.
Every external link is a maintenance overhead. Links to highly POV sites, even if clearly labelled as such, are a particular problem because these sites tend to change their directory structures frequently, which breaks the links here.
Excessive citation of offline resources has its own problems. We do have some experts here who will spot misquotations and authoritative-sounding references to vanity press (or even nonexistent) works, but they can't be everywhere. A check to see whether the title exists in any major University library online catalogue is good, but time-consuming, and doesn't do anything to confirm that the citation is accurate. Andrewa 17:47, 13 February 2006 (UTC)

[edit] Enchanter's edit

I fully support Enchanter's edit to this policy. If we leave it at that, I will withdraw my previous objections. Kaldari 03:04, 26 January 2006 (UTC)

I don't. It puts too much of the burden for supplying sources on the rest of us. If editors are too lazy to provide sources, their edits shouldn't be protected. -- Dalbury(Talk) 03:14, 26 January 2006 (UTC)
I'm with Dalbury, jguk 07:41, 26 January 2006 (UTC)

[edit] Comments

I oppose this policy rewrite. My thought process is something like this:

  1. Ultimately, unverifable material cannot remain in an article. We want sources to cover all the facts, more or less, in the article. I think we all agree on this. The question is what process do we go through to challenge and find sources for unsourced assertions.
  2. In the majority of cases, a Wikipedia article is better off containing as-yet unsourced information than without that information at all. Evidence: imagine Wikipedia today with all the unsourced material removed. We'd more than decimate the encyclopedia.
  3. Therefore, unsourced additions will, in a majority of cases, improve the article.
  4. Improvements to an article should not be undone without good reason.
  5. Accordingly, I oppose the proposed policy that it should always be a favoured procedure for new additions to be "summarily" removed by any editor.
  6. Instead, I would advocate a range of approaches, dependent on the exact situation. There are many different cases, and we need to be flexible. Yes, sometimes it is appropriate to summarily remove a suspect addition. Sometimes it's better to move it to the Talk page and request a source. Or to add a "{{fact}}" template next to it, or post on the editor's talk page requesting a source. Sometimes it's even appropriate to spend 30 seconds Googling.
  7. We're balancing two things here: the immediate pragmatic use of an article (that is, more information the better) with the need to cite our sources. One requirement should not dominate. We should take a balanced approach that encourages -- even demands, occasionally -- sources without it being too disruptive, or without making it burdensome for people to participate in Wikipedia. — Matt Crypto 14:01, 26 January 2006 (UTC)
You seem to be opposing the rewrite because you oppose current policy. If you wish to change policy, by all means propose a new one (though I would note that I for one am likely to disagree with it), but the question here is whether the rewritten page better conveys current policy than the current page, not "what do you think of current policy?". (For the avoidance of doubt, note that the current policy uses the following wording: "The burden of evidence lies with the editor who has made the edit. Editors should therefore provide references. Any edit lacking a source may be removed.") Incidentally, I do agree with you that it is better to provide or obtain references for unsourced materials so that they can remain than to removed unsourced materials, but I believe current policy is completely correct in placing the obligation to provide references on those wishing for material to remain, jguk 17:49, 26 January 2006 (UTC)
If this is just a rewrite, it should not say "summarily remove". This implies that editors are free to remove uncited information without any discussion whatsoever no matter what the circumstances. Clearly in certain circumstances, it is advisable to attempt some form of discussion before gutting an article. This is common sense and civil behavior. That's why the current policy says simply "remove". It then goes on to say that in certain circumstances you should move it to the talk page and in other circumstances you should request a source. It says nothing about "summarily removing" information. Your characterization of this proposal as only "a rewrite" seems a bit disingenuous to me. If you changed it to just say "remove" in the main policy summary, and then explained that dubious or controversial additions can be "summarily removed" I think that would be more in line with the intent of the original policy. Kaldari 18:48, 26 January 2006 (UTC)
And who decides whether an item is sufficiently 'dubious' or 'controversial' to summarily remove? I can see revert wars over whether something is 'dubious/controversial' and requires a source, or isn't 'dubious/controversial' and can be re-inserted without bothering to supply a source. If any item can be removed because it isn't sourced, it can be re-inserted as soon as a source is supplied. -- Dalbury(Talk) 20:19, 26 January 2006 (UTC)
Whoever is removing the information decides if it is dubious or controversial. If someone disagrees with the deletor's assessment, then yes, discussion should happen before the item is removed. Personally I don't think that all information in Wikipedia has to be sourced. I think that all information in Wikipedia has to either be sourced or easily verifiable, which is how I interpret the current policy. I think that is the essential difference between the supporters of this "rewrite" and the detractors. Kaldari 21:19, 26 January 2006 (UTC)
Say I decide that something in an article is dubious and remove it, and someone disagrees with me. How do we discuss it before it is removed if I've already removed it as 'dubious'?? If every item must be discussed before it can be removed, we will be bogged down in endless discussions. The best way to prevent something being removed is to cite good sources. and if something is easily verifiable, why not just cite the sources instead of fighting to keep it in without sources? -- Dalbury(Talk) 00:13, 27 January 2006 (UTC)

[edit] Big Bang

I appreciate there are concerns about disruptive editors using this policy to remove large sections of currently unsourced information with no attempt to get references for the information, but the general thrust appears to be that this rewrite is better than the current wording of the policy (plus the objections are equally valid for the current wording as this is just a rewrite). I'm therefore being BOLD and, after tweaking the wording of this proposal to state explicitly that there is no carte blanche for mass deletions, making this version live. Once-live further tweaks will and should happen, but I think it is now time to allow this version to live, jguk 20:44, 31 January 2006 (UTC)