Wikipedia talk:Verifiability: Difference between revisions

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
Line 1,150: Line 1,150:
::The key here is whether the information is '''harmful''' to the article (as outlined in WP:CITE)... since listing someone as a Rumanian Actor is hardly harmful (to the article or the person), the correct approach is to add a citation request (a <nowiki>{{fact}}</nowiki> tag) and leave it in the article. If, after a reasonable period (I would suggest at least a month), a citation has not been provided, ''then'' you can delete it or move it to talk. [[User:Blueboar|Blueboar]] 11:33, 8 July 2007 (UTC)
::The key here is whether the information is '''harmful''' to the article (as outlined in WP:CITE)... since listing someone as a Rumanian Actor is hardly harmful (to the article or the person), the correct approach is to add a citation request (a <nowiki>{{fact}}</nowiki> tag) and leave it in the article. If, after a reasonable period (I would suggest at least a month), a citation has not been provided, ''then'' you can delete it or move it to talk. [[User:Blueboar|Blueboar]] 11:33, 8 July 2007 (UTC)
:::If I look up Romanian actors, and add some to my report that are not romanain actors becuase WP put uncited information back in, that is harmful. Unreliable information is a detriment to an article. [[User:Until(1 == 2)|Until(1 == 2)]] 13:48, 8 July 2007 (UTC)
:::If I look up Romanian actors, and add some to my report that are not Romanian actors because WP put uncited information back in(after another editor challenged it veracity), that is harmful. Unreliable information is a detriment to an article. This is basic academic responsibility, show your work. [[User:Until(1 == 2)|Until(1 == 2)]] 13:48, 8 July 2007 (UTC)
===Unsourced data===
===Unsourced data===

Revision as of 13:49, 8 July 2007

The project page associated with this discussion page is an official policy on Wikipedia. It has wide acceptance among editors and is considered a standard that all users should follow. Before you update the page, make sure that changes you make to this policy really do reflect consensus.


Self-published sources: suggestion

I suggest the following phrasing for this section:

Anyone can create a website or pay to have a book published, then claim to be an expert in a certain field. For that reason, self-published books, personal websites, and blogs are not usually acceptable as sources.
Self-published material may, in some circumstances, be acceptable when produced by a well-known expert on the topic of the article whose work has previously been published by reliable third-party publications. However, caution should be exercised when using such sources: information in self-published sources has not been subject to independent fact checking, and may be less reliable than professional-published work by the same author. For this reason, it is usually necessary to provide a prose attribution for material sourced to self-published works. Self-published works that make controversial claims are not usually considered reliable sources.
Self-published sources should never be used as third-party sources about living persons, even if the author is a well-known professional researcher or writer; see WP:BLP.

This seems to me to be a reasonable start. It relaxes the requirements on the writer of the source to be a professional researcher, now only requiring them to be a "well-known expert" (i.e., it must be documented somewhere reliable that they are an expert on the subject), but requiring them to have been published on the same topic by a reliable source (rather than merely having been published as is in the current text). It also requires a prose attribution and suggests that controversial claims should not be sourced from self published works, both of which should help reduce the problems of undue weight being placed on contentious sources.

Any comments on this suggestion? JulesH 14:18, 27 June 2007 (UTC)

Sounds reasonable. The word "expert" is good, especially "expert" and "same topic" and "reliable source" all mentioned together. --Aude (talk) 14:34, 27 June 2007 (UTC)
"whose work in the relevant field [or in the field in question] has previously been published..." I'd also suggest dropping "well-known" for "established". Marskell 14:42, 27 June 2007 (UTC)
I'd also leave in "if the information in question is really worth reporting, someone else is likely to have done so." Another formulation I've thought of previously "editors should ask themselves if the person needs to self-publish because reliable publishers would reject the material" or some such thing. Marskell 14:46, 27 June 2007 (UTC)

"Scholarly or non-scholarly" - I see no reason to say this. It is likely to encourage yet more use of dubious sources, which we have a problem with already. If it's either/or, then let's not mention it at all. Tom Harrison Talk 19:37, 27 June 2007 (UTC)


Many of the people involved in this discussion have done excellent work writing articles, with many featured articles, though covering different topic areas with different issues. I haven't followed SlimVirgin's articles closely, so not familiar with what problems she has had with WP:RS and guidelines/policies on self-published sources. I'm more familiar with what Tim Vickers has worked on, the climate/global warming articles that Raymond has worked on, and some of the work done by Marskell. Reliable sources definitely follows a continuum, from scholarly peer reviewed works (should generally be expected on science articles), to reputable journalists and news media outlets, to some other works (e.g. blogs or other material published by known experts in a field; for biographies and history articles, books and material written by people involved in the event likely would be good sources, etc.) But then we have the arbcom case on pseudosciences that should also guide us. What other examples and situations do people have? --Aude (talk) 14:30, 27 June 2007 (UTC)

  • For social sciences, I still think scholarly peer reviewed material should be preferred. For topics like criminology, there is scientific rigor in studying topics like gun violence that help elevate the article above a political debate - for political aspects of the topic we have the article on gun politics. The topic also is looked at from a public health/epidemiology standpoint, with scientific rigor. There may be room for some exceptions to scholarly sources, but it is wise to stay above politics with such articles. --Aude (talk) 14:32, 27 June 2007 (UTC)
    • On medical topics, obviously, peer-reviewed journal articles are generally the best source of information. Review articles or textbooks are best, because they're secondary sources, generally written by recognized experts, and draw conclusions (thus avoiding problems with synthesizing primary sources ourselves). Not all peer-reviewed material is created equal - after all, both the New England Journal of Medicine and the Journal of Scientific Exploration claim to be peer-reviewed - but it's a start. There are far too many instances of self-published health-related websites being cited as equivalently reliable sources of medical information, being given equal weight, and being used to challenge or "rebut" accepted medical consensus. WP:WEIGHT and WP:FRINGE are already nearly completely disregarded on many medical topics. Giving "non-scholarly" sources essentially equivalent weight, in this sub-area, will worsen the problem. MastCell Talk 15:46, 27 June 2007 (UTC)
      • Perhaps there should be a requirement not to use self-published sources where they are contradicted by professionally published sources? Would this help with this issue? JulesH 15:50, 27 June 2007 (UTC)
  • For science articles, current scientific data, hypotheses and theories should be taken from peer-reviewed publications such as scholarly reviews and research papers. General information on the background of a topic can come from textbooks, but these should not be used to support specific arguments if these points of view conflict with the peer-reviewed literature. For example, the standard undergraduate biochemistry textbook written by Stryer is wrong when it classifies ribozymes as enzymes, a mistake which has caused much discussion on at Talk:Enzyme#Ribozymes. The history of scientific ideas is usually covered in books by historians. The only place where a significant amount of non-academic sources is appropriate would be in discussing people's opinions about science, for example in the Green Movement. Tim Vickers 15:52, 27 June 2007 (UTC)
  • In regards to the other side of the spectrum where there are sources we may be sometimes in danger of excluding, are the societies that are organized around different authors or types of literature which publish, sometimes self-publish, quality material. The material can range from 8-volume books, to monthly or quarterly journals, to websites. Although I have not used these sources to write a Wikipedia article, I have relied research from such sites for misc. stuff at Wikisource. One extreme example the Kipling Society which in the process of developing a web-only "New Readers' Guide" to update and replace the self-published "Readers' Guide" from the 60's of which only few of 100 copies are publicly available. An example of the detailed information that is unlikely to found outside of these sources are these notes on a poem.--BirgitteSB 17:27, 27 June 2007 (UTC)
  • Sources/publishers such as the Kipling Society have been mentioned many times by reputable news media such as the New York Times and the BBC. [1] This BBC news article, for example, mentions the society in a way that indicates it is indeed a reliable source and authority on the topic. No reason that we can't include such societies in the guideline. --Aude (talk) 17:33, 27 June 2007 (UTC)
OK, this might be covered by an exception saying that if multiple reliable sources cite a self-published source as an authority, this source may be used in a similar manner on Wikipedia. Tim Vickers 17:46, 27 June 2007 (UTC)
  • The Old Readers Guide was supported by the society, but it was definitely self-published. (I am lucky enough to live near a University that has a copy and have personally seen it; every page is printed with something along the lines of "For private use only"). The New Readers Guide gets a lot of information from this Old Readers Guide, but it is far from complete currently, so in some case you will have to refer to the Old Readers' Guide. This is the most extreme example of a source I can think of which, I feel people use and regard as reliable, but has few of the benchmarks that are used here to judge reliability.--BirgitteSB 17:57, 27 June 2007 (UTC)

Aude raises an interesting point above. Where there are reliable sources that give a self-published source a good review, it should usually be considered reliable, IMO. Should this rule-of-thumb be added to the policy? It would certainly help with cases like the one Birgitte mentions. JulesH 18:58, 27 June 2007 (UTC)

I don't think there's a way of formulating language to cover that issue. Many blogs get some kind of favorable nod from columnists writing in reliable sources. We don't have a system where a reliable source can make another source reliable by giving it a good review. The trouble with self-published sources is that they may not be consistent from one period of time to another. We already exempt self-published people who are acknowledged experts writing within their field of expertise. Why wouldn't that cover the Old Readers Guide? ·:· Will Beback ·:· 19:59, 27 June 2007 (UTC)
Do we really "exempt self-published people who are acknowledged experts writing within their field of expertise"? I don't see anything in WP:V, WP:NOR, or WP:RS that says so. Raymond Arritt 20:05, 27 June 2007 (UTC)
(edit conflict)The man behind the Old Readers Guide is described here when he became president of the Society:He was a Banker all his working life, and retired in 1946 after twenty years as Manager of West End Branches. From 1911 he was an officer in a Territorial Battalion of The Buffs, and served for over five years during and after World War One, mostly in the East and much of the time as a Brigade Major. Then it describes his work done for the Society or towards the Society's goals (amoung others things For many years he has given invaluable help to the past two Hon. Librarians, and his knowledge of our Library and indeed of all Kipling's writings is encyclopaedic.). I guess the real question is where does someone cross the line from "enthusiast" to "expert". I honestly don't know how we judging if someone qualifies as an expert right now.--BirgitteSB 20:16, 27 June 2007 (UTC)

Proposed new draft

Incorporating the suggestions generated in the discussion above. Please edit and comment. Tim Vickers 19:02, 27 June 2007 (UTC)


See also: Wikipedia:Reliable sources and Wikipedia:Biographies of living persons

Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy. Sources should be appropriate to the claims made, with exceptional claims requiring multiple, independent reliable sources. The most reliable sources for subjects such as medicine and the natural sciences are generally published by scientists, scholars, and researchers, particularly peer-reviewed publications in academic journals, textbooks, and books published by academic presses. However, alternative reliable sources can be used if scholarly publications do not cover a topic - which may sometimes be the case in the arts, humanities or current events.

Sources of questionable reliability

In general, sources of questionable reliability are sources with a poor reputation for fact-checking or with no fact-checking procedures or editorial oversight. Sources of questionable reliability should only be used in articles about themselves. (See below.) Articles about such sources should not repeat any potentially libelous claims the source has made about third parties, unless those claims have also been published by reliable sources.

Self-published sources (online and paper)

Anyone can create a website or pay to have a book published, then claim to be an expert in a certain field. For that reason, self-published books, personal websites, and blogs are usually unacceptable as sources.

Self-published material may be acceptable when produced by an experienced professional researcher, well-known journalist, or other expert in a relevant field. Such authors should have previously published work in that specific field, in publications that are clearly reliable sources. In this case, citation of these self-published sources may be acceptable, particularly if these sources have previously been cited by other reliable third-party publications. However, exercise caution: if the information in such a self-published source is really worth reporting, someone else is likely to have done so in a reliable source.

Self-published sources should never be used as third-party sources about living persons, even if the author is a well-known professional researcher or writer; see WP:BLP. Such sources are also unacceptable as support for exceptional claims, as these require multiple, independent reliable sources.

Comments on the new draft

Please comment on the new draft here:

Tim, while your suggestions are an improvement over the existing policy, the problem that we continue to run into is with regard to judging the reliability of the fact checking sources. I have seen greatly extended arguments dealing with who or what is a reliable fact checking source. And the arguments go nowhere. If we require that the WP editor has to source the specific fact that they are alleging in the WP article, then we bypass a great deal of debate. For example, imagine a book appearing saying that the earth revolves around the sun in a book publishing environment where the common conception is that the sun revolves around the earth. We know that this one book is right but by WP standards, this book is not a reliable source. Unless I'm mistaken in my understanding of the policy, this simply is an afront to new and provable research. Again, my I suggest this article as a source about the state of reliable sources and fact checking: "The Factchecking Facts"? It's quite an eye-opener and a serious wake up call to the Verifiability vs. Truth discussion. Thank you for your time and effort. Jtpaladin 19:42, 27 June 2007 (UTC)

If there is only the one book which makes the claim, then, yes, it couldn't be used as a source. But it could be used as an article on its own, so long as it was published by a reputable publisher. But if it's the embodiment of a movement which says that the earth revolves around the sun, then the book could be used in an article about counterexamples, i.e., 'In his book The Solar Sytem, Nicholas Copernicus asserts that, in fact, the Earth revolves around the Sun. The theory has been supported by Galileo Galilei, Fred Hoyle and ..." (then provide references). Do you see where I'm going with this? Corvus cornix 20:12, 27 June 2007 (UTC)
Tim, you've never edited any of the content policies before, and it's showing in some of the edits you're making, which are causing subtle contradictions in some areas, both internally and across policies. SlimVirgin (talk) 20:16, 27 June 2007 (UTC)
Thanks for your feedback, could you be more specific? Tim Vickers 20:44, 27 June 2007 (UTC)
Yes, I certainly respect User:SlimVirgin's experience in crafting policies, but it would be more useful to identify specific failings of the draft rather than simply reasserting her greater experience. MastCell Talk 21:06, 27 June 2007 (UTC)

I don't think that the preference for academic sources should be restricted to natural science and medicine, as it is in TimVickers' current draft. IMHO, academic sources, if they are present at all, will tend to be at least preferable in almost any field. "Popular views" and crank theories are extremely dangerous in the social sciences and in history, and I can imagine them causing problems in the humanities as well. As for SV's Holocaust issue (which seems to be the main example of an alternative non-scholarly view that the policies must endorse): I find it difficult to believe that no scholar at all, of any conviction or affiliation, has ever defended what she describes as "the non-scholarly view". But if that is indeed the case, then I just don't see how that view could claim extensive coverage in Wikipedia (it would inevitably get some coverage, since even the fact that it is dismissed by scholars implies that it has to be described). --Anonymous44 02:56, 28 June 2007 (UTC)

There are academics who support it, but Holocaust historians do not, because they say it relies on a simplistic understanding of the material. Regardless of their views, the debate is an important one, and it would be obtuse of us to leave it out. One of the editors supporting the change to this page said today that alternative views have been deliberately kept out of Global warming, even though there exist good non-scholarly sources for those views. That's the kind of thing I find very objectionable. We are here as librarians only, as Crum375 said. We're not scientists or researchers who are trying to promote certain views and suppress others. We're here only to direct people to good sources of all kinds and persuasions, and allow readers to read about the debates and judge for themselves. That doesn't mean we let in nonsense, but we're not allowed to keep something out just because it didn't emanate from an ivory tower. SlimVirgin (talk) 03:25, 28 June 2007 (UTC)
The question is precisely how a source, especially in a scientific filed such as global warming, can be "good" without being scholarly. I don't see that. As for letting wikipedians "judge for themselves", your version is the one that gives more of a researcher's "freedom" and "autonomy" in that respect. The editor is left almost completely free to decide arbitrarily what constitutes a "reliable" or "good" source and is likely to be influenced by his own agenda. We always have to decide which sources/views deserve more trust/attention, how "significant/prominent" they are per WP:Undue Weight etc. In Vickers' proposal, there is some guidance for that. To me, reliability is associated with academic, peer-reviewed stuff, and with reputed media. If we stop preferring this "ivory tower", there are just too few objective barriers against the tsunamis of "alternative" nonsense coming from outside. --Anonymous44 12:09, 28 June 2007 (UTC)
Fundamentally, the question is whether "good non-scholarly sources" are equal to scholarly sources. Some of us have a hard time accepting that an interview with Dr. Krakpott in USA Today has to be considered on the same level as a report from the National Academy of Sciences. It will be interesting to see how this ends up. Raymond Arritt 03:46, 28 June 2007 (UTC)
Nobody says that an interview with a single eccentric scientist on USA Today carries the same weight as a report published in a reputable peer reviewed journal. The point is that we need to assign to each source a weight, compare them all, try to gauge the prevailing views based on these sources, and then present a neutral and balanced picture of the state of the published literature. A tiny minority, such as Dr. Krakpott, could well be excluded, per WP:UNDUE, despite being published in a newspaper. But the point is that all significant views should be represented, per WP:NPOV. It is up to us as editors to assign these weights and decide how to meet our NPOV requirement. Crum375 04:41, 28 June 2007 (UTC)
(ec)My experience editing highly disputatious articles such as global warming and medical scams tells me that your view of how the process works is touchingly naive. If primacy isn't given to peer-reviewed science, we'll have even more of a struggle against tenacious POV-pushers. I can bet you as surely as the sun rises in the east that they'll holler "You can't leave out Dr. Krakpott! USA Today is a perfectly good newspaper, and according to WP:V you can't say that the Journal of Climate carries any more weight just because it's peer reviewed!" (Actually many of their favorite sources are worse than USA Today, but they still meet the very weak requirements established for WP:V.) Raymond Arritt 04:59, 28 June 2007 (UTC)
Absolutely. I don't want to sound apocalyptic, but the truth is that there is always literally an ocean of nonsense and agenda-driven stuff out there, threatening to engulf us and turn us into just another place that makes the Internet suck (paraphrasing Jimbo). --Anonymous44 12:09, 28 June 2007 (UTC)
I agree, the problem is that the current policy gives no guidance on how to assess sources for relative weight - a critical omission. It does a reasonably good job of explaining what cannot be included, but does not help a reader who is wondering how to apply the statement that the "relative weight" of sources must be considered. I have tried to fix this omission in this new wording. However, nothing in the proposed wording on which sources are usually most reliable makes fixed rules, instead it offers general and useful guidance. Tim Vickers 04:51, 28 June 2007 (UTC)
My own opinion is that we should not constrain editors from using their common sense, as each context may be different. But in general, mainstream reputable sources with more vetting layers carry more weight than others. Crum375 04:57, 28 June 2007 (UTC)
Yes, very true. This is why the propose wording allows editors to use their common sense and include alternative sources if they are needed to cover aspects that are not dealt with in the higher-quality sources. However, I'm sure you see my point in that giving no guidance at all on an relative weight - an important point made explicitly in WP:NPOV - is a major omission. However, as you say, this can't be any more than guidance and needs to allow editors to use their own judgement. I get the feeling we may be coming close to agreement. Tim Vickers 05:04, 28 June 2007 (UTC)

An alternative is to omit any mention of subject areas:

The most reliable sources for subjects are generally published by scientists, scholars, and researchers, particularly peer-reviewed publications in academic journals, textbooks, and books published by academic presses. However, alternative reliable sources such as newspaper or magazine articles, books from non-academic publishers and government reports can be used if scholarly publications do not cover all aspects of a topic.

Tim Vickers 04:59, 28 June 2007 (UTC)

I don't think that I like that wording, because my own preference is to give more weight to mainstream, well vetted, reputable publications. So you could have an obscure scientific journal, with 2 peers who are in collusion with the author, on one side, and reports in the Washington Post and the New York Times on the other. I would not give the obscure journal any more weight than the papers, yet your proposed wording relegates the papers to a lower tier, while promoting the obscure journal. In my opinion, we need to decide which sources are significant and discard those that aren't, and then we need to prioritize the survivors, and present them based on their prevalence and quality. To blindly declare that a scientific journal trumps other publications is wrong, and would violate NPOV. The bottom line is that we need common sense, not an exact prescription. Crum375 05:24, 28 June 2007 (UTC)
The problem is that common sense is so uncommon. If guidelines can be disqualified by the presence of absurd counterexamples (such as your two-author "journal"), why have guidelines at all? Why not just replace WP:V with a one-sentence statement "Use whatever sources you think best"? Raymond Arritt 05:31, 28 June 2007 (UTC)
The problem with just saying "mainstream" and "reputable" is that these are very woolly concepts and impossible in practice to define or apply. But the point you raise is a good one - there may be the occasional exception to this general statement, but do you agree that this is true in most cases? I've made the change to "generally published by", which will allow editors to use this guidance (which is not worded to be at all prescriptive) flexibly to accommodate the rare exceptions, such as the one you raise. Tim Vickers 05:36, 28 June 2007 (UTC)
For the record, what is excluded on Global warming is non-science sources to introduce scientific skepticism. The principal alternative to human-induced warming, solar variation, has a nice, tidy section, properly sourced to scholarly material, which introduces a note of skepticism without relying on conservative newspapers. The section on global political issues is also of appropriate scope for an introductory page, and here newspapers are used, because newspapers are appropriate for describing politics.
I did not want to introduce the example only to have it twisted. This is top of google on easily one of the top-ten issues of our time and, having gone through it extensively on FAR, I think it's a fine example of our best work. If it's not clear, prioritizing textbooks and peer-reviewed journals is a no-brainer for scientific theory itself. A science page in its entirety may, of course, have other material where we have no need to demand peer-reviewed work. Marskell 13:31, 28 June 2007 (UTC)
Regarding Crum375's comment, I don't think anyone would interpret any of these drafts to exclude the New York Times or Washington Post as a reliable source on pretty much anything. We're not talking about the Times, though; we're talking about much more questionable "non-scholarly" news sources. For many topics where a clear scientific consensus exists despite a vocal sociopolitical minoritarian dissent (e.g. vaccination, evolution, AIDS reappraisal), you have the consensus of the scientific community on one side and a number of blogs, websites, newsletters, organizations, and alternative media criticizing the science on the other. Inevitably, dedicated single-purpose editors show up to demand that these alternative sources be given equal weight in terms of how we present the science. WP:WEIGHT is vital here, as is the fact that scholarly sources are typically given precedence over non-scholarly sources when describing the scientific understanding of a topic. If we alter WP:RS or WP:V to put "scholarly and non-scholarly" sources on an equal footing, the integrity of the encyclopedia (as far as scientific articles, at least) is going to suffer. Note that I'm not talking about excluding dissenting views; I'm talking about properly characterizing them. I have a lot of respect for editors who've worked on policy extensively, and I recognize my inexperience here. I'm just asking that the views of editors who try to maintain medical/scientific articles be appreciated as well, as I think the implications of this change on such articles has not been fully taken into account. MastCell Talk 18:43, 28 June 2007 (UTC)
No matter how much some scientists detest it, alternative views about vaccination must be published if there are reliable sources, and there are. That doesn't mean blogs and newsletters. It means books and papers by reliable sources; mainstream newspaper articles, and so on. You write as if scientists were always right, but you must know that's very far from the truth. Take some recent news about asthma. The Buteyko method of controlling attacks was widely ridiculed, with all the same arguments being trotted out (not peer-reviewed, no clinical trials, not scientific!!), even though thousands of asthma sufferers were able to stop their medication because of it. The difficulty was that no one wanted to fund a clinical trial, because no drug company would benefit from it; quite the reverse, in fact. Then a hospital took it up, developed it a little, and conducted a trial, and recently the method has been confimed to work very well. Yet for over a decade, the story was that it was baseless and dangerous. This is exactly why we have our NPOV policy. We must never fall into the trap of thinking that just because a powerful special interest group says X, X is the only story in town. SlimVirgin (talk) 05:47, 29 June 2007 (UTC)
It's a matter of choosing the lesser evil and being cautious. This Buteyko could have turned out to be a charlatan as well, and for every example of good "alternative" stuff, there will be thousands of bad ones; if we have to choose between missing some useful info and allowing some BS in, we should choose the former (that's one point of deleting unsourced stuff, articles etc). You talk about "books and papers by reliable (non-scholarly) sources", but during this whole discussion, I haven't heard how one can establish that these are "reliable", with no criterion left but the fact that they are "kinda popular", that one likes their style, the design of their cover, or their agenda. Apart from the "not self-published" requirement, everything else is relative, and there are tons of agenda-motivated pseudoscience being published by third parties. --Anonymous44 11:47, 29 June 2007 (UTC)

(Response to SlimVirgin) I'm afraid I draw a different conclusion from your example (and I won't argue about Buteyko, other than to say that the story is a little more complex than the summary above). If the Buteyko method was considered bogus for years, then Wikipedia should have reflected, for those years, that it was considered bogus by the medical community, though anecdotal reports of its usefulness existed. When research was published indicating it might work, then Wikipedia should be updated to reflect that understanding. Scientists are not always right; in fact, given a long enough time frame, they are almost always proven wrong. Nonetheless, until they are proven wrong, Wikipedia should accurately reflect the current understanding. You seem to be implying that Wikipedia should have given more weight to the anecdotal claims of Buteyko's effectiveness, even before the relevant studies were published. That's convenient, with 20/20 hindsight, but as noted above it cherry-picks Buteyko as an example and ignores the fact that many proprietary health products supported only by anecdotal evidence turn out to be useless or worse.

You misunderstand if you think I'm advocating censoring anti-vaccinationist views or promoting the views of "powerful special interest groups" as the only game in town. I'm just saying that if Wikipedia's goal is truly to accurately represent the current state of human knowledge, then it has a long way to go, at least in terms of the topics I spend the most time on. From my perspective, the proposed change is a step in the wrong direction. It may lead to significant improvements in the Holocaust-related articles; I wouldn't know. It will, however, make it more difficult to accurately and neutrally present scientific topics. MastCell Talk 00:06, 30 June 2007 (UTC)

If I may butt in, I agree with much of what you say, but disagree with some. I believe that scientists should not be given an automatic 'bye' in the NPOV balance. A scientific publication should be judged on its merits, like any publication. The merits from the WP perspective are the number of vetting layers that exist and the reputability, so peer review would count, as would editorial oversight. Larger more reputable publishing houses or media would also count, as they are assumed to have more vetters and fact checkers. Given these criteria, it is likely that in most cases the scientific publication would be far 'better' wiki-wise, but not always so. We could have a case of an obscure science paper published by an obscure publisher, vs. a report in the NYT and Wash Post, for example. So we should not pre-judge sources based on arbitratry criteria - we should let our editors judge each source on its merits. Crum375 00:28, 30 June 2007 (UTC)
That makes sense. The problem is the broad-brush "scholarly and non-scholarly" wording. Yes, there are a few obscure journals with dubious editorial standards. But for scientific issues -- as opposed to things like policy implications of science -- a journal that is good enough to be indexed in the ISI Web of Science is going to trump USA Today almost every time. The "scholarly and nonscholarly" wording glides right over the nuances you've described, and makes it sound like USA Today is just as good a source on radiative transfer and its implications for global warming as is the Proceedings of the National Academy of Sciences. Raymond Arritt 00:42, 30 June 2007 (UTC)
(reply to crum -wading back in, if only a little) Crum375's comment is one of the reasons I think we should work with the RS guideline to make it more robust and comprehensive (as opposed to official policies, which should be more general in nature). It seems to me that in the event there is likely to be the occasional exception to a general rule, then we should encode it in such a way that allows for the exception without having to invoke something like IAR. Because when "rules" have to be ignored, something about the policy is probably wrong (or at least not comprehensive in scope). R. Baley 00:43, 30 June 2007 (UTC)
I would agree in principle about the potential use of guidelines to amplify policies, but in reality RS ends up being a distraction and source of confusion for V. For example, the push to promote scholarly sources over non-scholarly, which I believe violates NPOV, started on RS, whereas something this fundamental should start at the policy level. In general, I believe we really need to get our core policies stabilized before expanding them with guidelines, otherwise we end up going around in circles. If it were up to me, I would redirect RS to V, make sure any missing RS material is properly covered in V, and continue fine tuning V in one place. Doing these things in two (or more) places is very confusing and counterproductive. Crum375 02:00, 30 June 2007 (UTC)
Indeed. Moreoever, I'd suggest the discussion take place on the NPOV talk page, because changing any policy to prioritize scholarly sources is an NPOV and Foundation issue, so we would need the broadest possible input. We need to focus it on one page so that we can alert people to the discussion. SlimVirgin (talk) 02:24, 30 June 2007 (UTC)
First we get directed from RS to V, then from V to NPOV, because somehow using scholarly sources violates NPOV. (Note also that WP:NPOV points to -- guess what? -- WP:RS, so it's hardly fair to berate people for referring to WP:RS.) No offense, but I feel like we're getting jerked around. And I'd like to see where it says that counting newspapers as equal to scientific journals is a Foundation issue. Not that I'm doubting you, but I'd like to be able to read the policies for myself. Raymond Arritt 02:40, 30 June 2007 (UTC)
A better solution may be to return to the earlier version, which did not specify that scholarly and non-scholarly sources were to be accorded equal status. That version left it to editors to work out on individual topics which sources, if any, should be accorded priority. That was actually a better approach; the change made in April (and the proposal to prioritize scholarly sources) both introduce a level of prescriptiveness that may not work well for all articles. A solution would be to go back to the prior version, which was more flexible. The change made in April undoubtedly had consensus at the time, but at present, with wider input, that consensus appears no longer to exist. MastCell Talk 02:34, 30 June 2007 (UTC)

Verifiability vs. Truth

So, is the Jim Wales comment still the rule of law here? I'm referring to:

If your viewpoint is held by an extremely small minority, then whether it's true or not, whether you can prove it or not, it doesn't belong in Wikipedia, except perhaps in some ancilliary article. Wikipedia is not the place for original research.'

Truth is an irrelevancy, only sheer volume of sources matters. It is original research to expect an WP editor to show proof of a source in a book. Is this still the rule?

My concern regarding published sources is the ease to which sources get published. Someone who has written a non-fiction book stating that George Bush is from the planet Zandar and he has come to earth to save 1,934,918 people so he can transport them on his silver surf-board back to Zandar so they can be eaten during the "Festival of O'poza", can get it published by any publishing house that thinks they can make a lot of money with the book.

The standards to get a book published have fallen substantially in recent years so the reputation of the publishing houses are about as good as the reputations of the accounting firms who took money during the late 1990's to inflate the balance sheets of various scandal-ridden corporations.

Also, the growth of publishing companies has exploded in recent years. I know about the rules of self-publishing but anyone who knows how to create their own publishing house can easily get around this rule. But more importantly, the assumption that the publishing houses are doing their job with regard to fact checking is inceasingly becoming a large leap of faith. People who think that publishers are more interested in fact checking than they are in making profits are putting far too much faith in the publishers. If the fact checking of a book that could potentially make multi-million Dollars domestically and internationally would stand in the way of those profits, does anyone seriously think fact checking will prevail in the decision of whether or not to release the book?

My point is that I think the rule regarding Verifiability should be modified. I think that an editor here should be able to reference a book for an article but if challenged on the book, the editor should be required to provide the source of the information used in the book. For example, if we are dealing with an article about Global Warming and an editor posts a book that says that Global Warming is caused by Al Gore's invention of the internet, the editor should be required to point to the scientific source used in proving that Al Gore's invention of the internet is what is causing Global Warming. The way it stands right now, all the editor has to do is point to the page number in the book where the author makes this claim, i.e. the source is irrelevant; only the fact that the author has made that claim is all that matters.

If we require editors to "prove up" their sources, we are going to see a lot less arguing on the Discussion pages of each article. Just the mere fact that people spend so much time arguing about the reliability of a source takes up a serious amount of time.

For anyone interested in reading about the fact checking disaster in the publishing houses, I urge you to please read, "The Factchecking Facts". It's quite an eye-opener and a serious wake up call to the Verifiability vs. Truth discussion. Thank you. Jtpaladin 19:30, 27 June 2007 (UTC)

You are simply way off the mark. You are assuming that WP is seeking truth, and that by going after verifiable sources it believes it will find it. But that is not the case. WP is seeking verifiable sources, period. We assume that information that comes from such sources has been vetted by multiple persons, with specific focus on legal liability in the case of making fraudulent claims. But we do not expect perfection, and in fact by allowing and even requiring multiple points of view, we expect and welcome contradictions. Obviously when there are contradictions, there must be at least one vetted fact that is wrong, so we expect wrong information. Our goal is simply to represent the reliably published information, not to seek the truth. Think of us as a helpful librarian, who helps an interested reader to the best available published sources that cover a given topic. No guarantees are made as to 'correctness' or 'truth', only our best effort to find and present the best published sources neutrally and in a balanced fashion. An equivalent way to state our mission: we are librarians, not scientists. Crum375 20:13, 27 June 2007 (UTC)
No, Crum375, that's not what I'm assuming. I'm clear that WP is after Verifiability and not necessarily truth. My point is the faith in fact-checkers is a mistake. I'm simply questioning the process. I have no interest in changing policy. That's up to you guys. Jtpaladin 23:50, 27 June 2007 (UTC)
Then we'd better advertise that widely to the public. A lot of people think that Wikipedia is concerned with accuracy. It's troubling that there's little effort to disabuse people of that notion -- it's almost like we're willing to let the misperception go on, because it's to our benefit. Raymond Arritt 20:21, 27 June 2007 (UTC)
But we are concerned with accuracy - accurately reflecting the verifiably published literature on a given subject, and presenting it neutrally and in a balanced fashion. Accuracy and truth are different concepts. Crum375 20:29, 27 June 2007 (UTC)
The distinction may be clear to us Wikignomes, but I wonder how many casual readers recognize the difference. Wikipedia:About says we try to avoid "misinformation" (conventionally defined as "false or misleading information") and the like. Why not be clear up front in saying that we may knowingly include false or misleading information in order to fairly represent all points of view? Raymond Arritt 20:49, 27 June 2007 (UTC)
If we're presenting false information, it should be made clear that the information is generally considered false. Unfortunately, this isn't always the case. As to Jtpaladin's initial argument, this appears to be a serious case of forum-shopping regarding a dispute at Talk:Joseph McCarthy, in which Jtpaladin is soldiering on against a consensus. MastCell Talk 21:11, 27 June 2007 (UTC)
MastCell, sorry but that's not the case. I'm trying to gain a better understanding of a policy that puts verifiability over the establishment of truth by examining sources. This has nothing to do with your absurd claim of forum-shopping (I mean, come on) since I'm not even mentioning the article in any other place other than in that particular discussion board. However, I am concerned that an author can make a claim and not have to prove the claim and yet that author's claim is used as a source for a WP article. As the article I posted above "The Factchecking Facts" demonstrates, putting your faith in fact-checkers is bound to lead to disappointment. Jtpaladin 23:50, 27 June 2007 (UTC)


The following language has wide support at WT:RS. It originally derives from this page, and is general; it should, in some form be here as well:

The most reliable material is published by scientists, scholars, and researchers, particularly publications in peer-reviewed journals and academic presses. These sources are preferred in subjects such as medicine, natural sciences, social sciences, and humanities. However, alternative reliable sources are used when scholarly publications are not available, such as in topics related to popular culture or current events.

Septentrionalis PMAnderson 20:37, 27 June 2007 (UTC)

This wording forms part of the proposed new section which is being discussed above. Tim Vickers 20:46, 27 June 2007 (UTC)

Edit war

I've protected the page for a few days. There seems to be an edit war going on. I had actually been hoping to take a closer look at this page some time, but there's so much going on that I wouldn't know where to begin or which side I'm on. Can you come to some agreement without reverting? The protection expires in four days, but if you most of you want it unprotected sooner, just let me know. ElinorD (talk) 20:48, 27 June 2007 (UTC)

Thanks, that was probably a good idea. I expect you will have protected the wrong version, but that's to be expected. Tim Vickers 21:16, 27 June 2007 (UTC)

Request for reversion to a non-contentious version


The current version of this page removes the long-standing exemption that applies to professional journalists. Can it please be reverted to a version that includes such an exemption until the discussions above have been concluded? I would suggest any of the following revisions:

SlimVirgin at 19:11, 27 June 2007
JulesH at 14:24, 27 June 2007
Johan Lont at 12:12, 26 June 2007

My preference (obviously) is for the second. This is a reversion to an earlier state of the article, which to me seems substantially less contentious than any of the recent phrasings, and fixes the issues some editors have had with the policy while not throwing the baby out with the bathwater. The first is the most recent phrasing that does not include any of the recent contentious changes, the last is the state of the article before the current edit war began. JulesH 07:45, 28 June 2007 (UTC)

Does anyone object to this change? --ais523 08:49, 28 June 2007 (UTC)
We don't need "(scholarly or non-scholarly)" as in SlimVirgin's edit, but your version is acceptable since it reflected discussion on the talk page. --Aude (talk) 12:48, 28 June 2007 (UTC)
We might again wring our hands over "expert", but if people believe that it meets some of the examples that have been presented, I won't quibble. It's certainly preferable to the "non-scholarly" innovation. (The example is a bit wordy, though.) Marskell 13:02, 28 June 2007 (UTC)
The JulesH version is OK. The "scholarly or non-scholarly" wording is inappropriate. Raymond Arritt 13:20, 28 June 2007 (UTC)
Agreed, I don't remember anybody arguing seriously against JulesH's version. --Anonymous44 13:54, 28 June 2007 (UTC)
☑Y Done, to the JulesH version. --ais523 14:29, 28 June 2007 (UTC)
I strongly object to what you did, Ais. The journalism thing was removed months ago after an agreement to do so, but was recently restored without agreement, and was therefore removed again. Please leave the page alone until the protection is lifted. SlimVirgin (talk) 16:49, 28 June 2007 (UTC)
Checking through the history, it appears that the (scholarly or non-scholarly) text was the one present a month ago (to pick an arbitrary figure), so the 'original version' is not clear and both versions that have been recently reverted to are 'innovations' in some sense. (Having now actually read the affected versions, rather than merely trying to gauge consensus, it would seem that at least one of the causes of the dispute is a disagreement on whether this text excluded journalists or not; the claim would seem to be that the version SlimVirgin reverted to just now excludes journalists whilst the long-standing version didn't. Note that the removal of the (scholarly or non-scholarly) text is the last edit before protection of the page; the version on which a page is protected during a dispute is often a matter of chance.)
Trying to piece together the chronology of what happened; this edit by SlimVirgin is a revert to what SlimVirgin apparently believes is the old version of the policy (and is above in a shortlist of possible versions to revert to, so this text, which some people think is inappropriate above but does seem to be longstanding, seems to be believed by JulesH to contain an exemption for journalists). The most recent revert by SlimVirgin is to similar text, but doesn't contain the exemption from JulesH's point of view (see the diff between the two versions SlimVirgin reverted to). What is SlimVirgin's objection to the way I fulfilled the editprotected; was it concerns with the content (in which case, there seems to be a misunderstanding about whether the month-old version that I'm using as a reference actually excluded journalists or not, and whether the version I reverted to or the version that SlimVirgin just reverted to reflects the longstanding version)? Was it an attempt to revert to a longstanding version due to no consensus with the version I reverted to, but not noticing that the version reverted to was a slightly modified version just (3 minutes) before the protection, not a longstanding one? Was it a purely procedural revert due to the page being edited while it was protected? (In this case, I'd point out that one of the two main purposes of editprotected is precisely to correct a case when a page has been protected on The Wrong Version and many editors agree that it's the wrong version that's been protected.) Or was it something else? See also my explanation on User talk:Marskell as to why I answered the editprotected request.
However, I would ask that you at least correct the interwiki to nl: (it's going to their equivalent of Wikipedia:Citing sources at the moment); I can't even read the language, but the content of the page can be deduced by looking at its structure and checking the interwikis back to en:. (Probably a minor matter, I just noticed this looking over the diffs.) --ais523 17:24, 28 June 2007 (UTC)
I'm OK with the JulesH version, the "non-scholarly" addition was not an improvement. Tim Vickers 17:15, 28 June 2007 (UTC)
It wasn't an addition; it had been there for some time. SlimVirgin (talk) 18:29, 28 June 2007 (UTC)

(outdent) ais523, I believe your edits of a protected page were totally out of order. We clearly have disputes over various issues here, and the page was protected. Many of us here are admins who can physically edit this page, and of course we all refrain from doing so. The only way a protected page can be modified beyond a trivial edit (e.g. spelling fix) is via a clear and unambiguous unanimous consent of all editors. There is clearly no such consent, and for you to perform edits that are taking part in a content dispute while using your admin tools is absolutely unacceptable. Crum375 17:44, 28 June 2007 (UTC)

Ais, you really shouldn't be editing protected policy pages when you're clearly not familiar with the dispute. There are two main issues: 1. Here is Marskell's removal on June 27, without consensus, of "scholarly or non-scholarly," which has been in the policy for some time. 2. Here is JulesH on June 27 restoring an old version that made an exception for professional journalists; this had been removed in April after agreement on talk.

My guess is that JulesH did this in order to restore the "scholarly and non-scholarly" distinction that Marskell is trying to remove. If that was his motivation, I agree with him, but in so doing he (perhaps inadvertently) restored a version that was problematic for other reasons.

This is the kind of chaos we end up with when people turn up to edit war. I hope everyone will stop, read the archives, read the other policies, see how they hang together, and then discuss intelligently on talk, rather than forcing us to regurgitate discussions that have taken place many times already. SlimVirgin (talk) 17:41, 28 June 2007 (UTC)

You should not be editing protected pages, in which you are involved in the dispute. --Aude (talk) 17:45, 28 June 2007 (UTC)
Let's all just take a step back here and to try to WP:assume good faith and not seek to own this page in any way. I am not here to edit war and I do not think any of the other editors who are trying to improve the deficiencies listed above are here to edit war either. Repeated assertions of authority without any attempt to explain what the problems are not entirely constructive. Tim Vickers 17:49, 28 June 2007 (UTC)
The problems have been explained to you several times on this and other pages, but you're ignoring them and ploughing ahead with your own ideas anyway. It's therefore becoming increasingly difficult to assume good faith. SlimVirgin (talk) 18:29, 28 June 2007 (UTC)
I agree with SlimVirgin's assessment of what the dispute is in their most recent comment. Crum375, I'm not entirely sure I agree with your interpretation of the protection policy. "Admins should not edit pages that are protected due to a content dispute, unless there is consensus for the change, or the change is unrelated to the dispute." There seemed at the time (and still seems to me) that there is consensus that the page was protected in an unusual version that didn't reflect consensus.
You're wrong about that. The version protected isn't the one I would prefer, but it's a lot better than the one from April you restored, and there's nothing "unusual" about it. If you want to continue with this, please consult the protecting admin. SlimVirgin (talk) 18:29, 28 June 2007 (UTC)
There is another dispute, about which version should actually be reverted to, if not the one the article was protected in.
I have previously made exactly one other edit of this nature ([[2]]). In this case, the edit to the contentious part of the template happened to take the template back to the version it was protected on. However, I feel that relying on which version a page happened to be protected is flawed, especially as the protection was made just 3 minutes after an edit away from the longstanding version. One risk in such circumstances is that a user reports to WP:RFPP with the right timing to get the article protected in their favoured version; however, that doesn't seem to have happened here. Another risk is that the protecting admin has a vested interest in the timing of the protection, but again that doesn't seem to have happened here, so the timing of the protection is fortituous. Normally, it's not worth returning a protected page to a longstanding state, but when there is a large outcry on the Talk page, and the disputed text (in this case, the express or implied exemption for journalists, which was in fact in the policy this time last year with different wording) is different from the longstanding state and seems to never have gained consensus, according to the Talk page. (I'm not too clear about whether it ever had gained consensus, though, looking through the arguments here; the relevant sections seem to have drifted off-topic, and it's obscured in other discussion. If it 'had' gained consensus, there is then the problem of wondering whether it had consensus for long enough that it should count as the default version to revert to during a dispute, which strikes me as the sort of question that promotes process wonkery and wikilawyering rather than any sort of sensible discussion, but is likely to come up anyway.)
I made the edit as part of running through CAT:PER for editrequested requests (I tend to stay out of mainspace in such requests, as it's not where my expertise lies, but I do many of the technical CAT:PER requests that happen in my timezone, although I'm not the only admin who covers them.) It seems to have become remarkably contentious for a backlog-clearing.
Again, I'd reiterate that the purpose of editprotected is for edits to be made to a protected page. It seems that there is something of a dispute brewing up now on a higher level than this one about whether protection policy allows editing of a page protected due to a dispute, even by people who aren't involved in the dispute (I previously wasn't, and had no opinion on the matter before making the edit, but was following what appeared to be consensus on the Talk page), to return it to a less contentious state. I asked for input from the regulars here as to which version was the least contentious as a result, and received a pretty-much unanimous answer. That may have been bad luck due to which editors were fastest to respond, in this case, but at the time the edit appeared clear-cut.
I hope this answers any queries about my edit. I will not, of course, revert here again with respect to this dispute (could someone please correct the protection tag from semi'd to full-protection, though, please?), and wouldn't have even if there had been no opposition to my edit (if there was a lot of opposition and consensus had swung the other way, though, I would have tried to be the admin to revert back to avoid a wheel war, as self-reverts to me clearly don't count in terms of wheel warring). I would urge people who oppose my edit to think about whether they're opposing it on content, policy, or process grounds. I hope that clears things up (but I suspect it won't). --ais523 18:21, 28 June 2007 (UTC)

I'd just like to clarify my reasons for requesting this change. I do not believe there has been consensus at any point to exclude journalists or other subject experts from the types of self-published source we may use.

There was consensus; please see the discussion about it from that time. The reasoning was that there's no need to single out journalism as the one profession we make exceptions for, in terms of self-published material, when as a matter of fact we make exceptions for any previously published specialist researcher in a relevant field. SlimVirgin (talk) 18:29, 28 June 2007 (UTC)
I remember the discussion well. There was consensus for not listing journalists separately; there was not consensus for not allowing them. I understand your viewpoint, but I have to say that the most likely reading a user who has not been following this talk page for the last few months will have of this policy is that journalists are not appropriate sources. I don't believe this current phrasing has consensus, because of the fact that it does not make it cleare that journalists are acceptable sources: the change to remove them from the list only gained consensus because of the insertion of "(scholarly or non-scholarly)" as a compromise position.
However, because User:Marskell and others do not seem prepared to accept that compromise, and were repeatedly reverting any attempts to restore it, I felt an older version, from before the journalist text was removed from the exception, might be a more stable version. Other than for the sake of textual clarity, no serious objections have been raised to the inclusion of journalists in the text, and as there does not seem to be any other widely accepted phrasing, it seemed sensible to revert to it. JulesH 19:00, 28 June 2007 (UTC)

Certainly, at the time of the original removal of this part of the exception, it was stated by all those who appeared in favour of the removal that it was not their intent to do so; they were, rather, clarifying the text as these people were examples of researchers. Others (including myself), however, did not consider that such people would generally be interpreted as "researchers", a word which has a rather specific meaning (an academic researcher) in addition to the more general one that was apparently intended (anybody who performs research). As a compromise, the phrase "scholarly or non-scholarly" was added to the article, to clarify which of these interpretations was intended.

The removal of this phrase, however, clearly changes the policy to the point where such sources are not acceptable. I do not believe such a change has consensus, so I suggested reverting to the previous phrasing as a compromise between my position, and those who did not accept the non-scholaraly part of that phrase. JulesH 18:07, 28 June 2007 (UTC)

You're right that the removal of "scholarly or non-scholarly" has no consensus, and I intend to restore it when protection is lifted. But restoring just one profession that we accept isn't the answer. It's best to keep it general as professsional researchers in a relevant field, whose work in that field has previously been published by a reliable source. That's quite narrow already; no need to narrow it further by saying only journalists fit it. SlimVirgin (talk) 18:32, 28 June 2007 (UTC)
It seems that the version I reverted to was too recent and didn't have consensus either, and that I was unlucky in my timing when I checked the talk page. Speaking on the content dispute now (which this rather unexpected controversy has given me a chance to finally think about), rather than on my edit itself: the problem is that there are three versions, one which specifically includes professional journalists and researchers but not anyone else (the 'ancient' version, so to speak (I've checked a version of this policy from a year ago); the problem with this is that it makes an arbitrary distinction that presumably isn't really appropriate in such a policy and will exclude too much), one with no specific or implied exclusions, except for researchers (the version protected; the real problem with this is that in one possible reasonable reading, and the one that's most obvious to me, it excludes even more than the version that excludes too much, and if taken literally would lead to half of Wikipedia being deleted, as far as I can tell), and one which excludes 'scholarly and unscholarly' researchers (presumably including journalists, but many people here think that the wording's lousy). In summary, problem with the protected version is that it's ambiguous in a way that makes it seem as though newspaper articles aren't a reliable source in the simplest reasoning; with the version I reverted to that it allows sources from an arbitrary and restrictive (although common) subset of what is desired; and with the month-old version that the wording isn't very good and can be unclear. At the moment, the third (month-old) version seems to have the least problems, but none are perfect. Coming back to my contentious revert, I apologise for not noticing the intended meaning of the protected version; I found it hard to see how anyone who advocated a version that made exemptions for a broad range of unscholarly researchers could prefer a version that made an exemption for some over an exemption for none! (I therefore erroneously mentally counted SlimVirgin as supporting my revert as an improvement but opposing it either for process reasons or because they'd misread it.) --ais523 18:51, 28 June 2007 (UTC)
Ais, the section in question deals with when self-published professional research sources may be used. It therefore has nothing to do with newspaper articles. The current version doesn't say such sources have to be scholarly, or that they have to be journalists, so it's okay. SlimVirgin (talk) 18:55, 28 June 2007 (UTC)
OK, that makes it less of a problem, but still ambiguously worded in my view, with what comes as the obvious interpretation to me being the incorrect one. (The very worrying thing here, unrelated to what wording is chosen, is that we'll still have to delete half of Wikipedia, but that's not really an issue of what's discussed here but of changing a culture of the creation of unsourced articles.) I really ought to be going home, anyway, so I'm unlikely to reply to further messages here for a while (possibly not at all if the topic of discussion moves on), though. --ais523 19:01, 28 June 2007 (UTC)
For many people, "scholarly" will be implied. See, for instance, the American Heritage Dictionary, whose primary definition of "research" is "scholarly or scientific investigation or inquiry". JulesH 19:06, 28 June 2007 (UTC)
"You're right that the removal of "scholarly or non-scholarly" has no consensus, and I intend to restore it when protection is lifted" -- I think it's fairly clear now that its addition has no consensus either, due to the large number of editors who have objected to it on this page. It is for this reason tha I reverted to the older version, which had consensus before and did not seem to be overly contentious. JulesH 19:06, 28 June 2007 (UTC)
Scholarly and non-scholarly has clear and long-standing consensus by the very fact that good editors don't only allow self-published scholarly sources into articles, whereas making an exception for journalists is not something I've seen anyone do, and there were good arguments in favor of its removal at the time. It makes no sense to try to solve one dispute by resurrecting another one. SlimVirgin (talk) 19:19, 28 June 2007 (UTC)
I'm having a little trouble following this; in particular, "self-published scholarly sources" seems like a contradiction in terms. It's also not clear that a change made only a couple of months ago[3] and to which many editors are raising objections now that they know about it can be taken as a "clear and long-standing consensus." Raymond Arritt 19:37, 28 June 2007 (UTC)
Please explain why it would be a contradiction in terms. SlimVirgin (talk) 19:43, 28 June 2007 (UTC)
Because "scholarly literature" carries a strong connotation of peer review; see e.g., [4][5][6]. Raymond Arritt 21:00, 28 June 2007 (UTC)
The page was protected in good faith by Elinor. It shouldn't have been edited at all; Ais, I think, felt s/he was reverting to be helpful in order to resolve a dispute, and that should be treated as good faith as well. But reverting to the exact version at the time of Elinor's edit is preferable. While four days is generally too long, it seems a good time frame in this case, to allow discussion.
"Journalist" has been gone for some time (though it obviously has some support in the history). "Scholarly and non-scholarly" is a recent innovation, which has little support. We should work out these things before the protection expires. Marskell 20:41, 28 June 2007 (UTC)
Scholarly and non-scholarly is not a recent innovation, obviously, because journalists aren't scholarly sources, and you're saying the journalists thing had support several months ago. What you want to do (prioritize the use of scholars, and allow non-scholarly sources only where scholarly ones aren't available) is the recent innovation, and it's a serious violation of NPOV. SlimVirgin (talk) 20:47, 28 June 2007 (UTC)
The words "scholarly or non-scholarly" are, in fact, a recent innovation. Did it exist on V before April? I don't recall it, but I'm willing to be corrected. As it stands, virtually no one commenting agrees with it.
Meanwhile, our NPOV pseudoscience wording has stood for six years (about twice as long as either of us has been editing). "...prioritize the use of scholars, and allow non-scholarly sources only where scholarly ones aren't available..." Absolutely not. "I'm using this because it's the only source available" is one of the worst arguments going, and I'd never, ever sanction it. Marskell 20:58, 28 June 2007 (UTC)
Marskell, are you trying to mislead people deliberately? You e-mailed part of a quote to me yesterday, and I returned the full quote to you, because you were taking words out of context. You wrote: "'The task is to represent the majority (scientific) view as the majority view' is long-standing policy wording." In fact, the full quote from NPOV/FAQ is in a section devoted only to pseudoscience, and it says:

The task before us is not to describe disputes as though, for example, pseudoscience were on a par with science; rather, the task is to represent the majority (scientific) view as the majority view and the minority (sometimes pseudoscientific) view as the minority view; and, moreover, to explain how scientists have received pseudoscientific theories.

And that is correct: the majority view is represented as such, and the minority view is represented as such. Neither is simply left out. Now are you saying the policy says: "prioritize the use of scholars, and allow non-scholarly sources only where scholarly ones aren't available..."? If so, where does it say that? SlimVirgin (talk) 08:15, 29 June 2007 (UTC)
Ya Slim, I'm trying to mislead people deliberately.
Last first, I posted above that I would never sanction "when not available" wording. I've had the "but I have to use the blog because there's nothing else" conversations. In practice, people are going to use X if there is no Y in a variety of situations, but we should not place wording of that sort in.
On the pseudoscience wording—yes, that is correct. It defines majority with science in brackets and asks us "to explain how scientists have received pseudoscientific theories." Why? Because we don't and shouldn't care about how soccer moms and WSJ Republicans have received the theory of evolution when describing the theory of evolution itself. Scientic theory qua scientific theory should be described with scientific sources because "scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such."
To remove that NPOV wording as you attempted and add "scholarly or non-scholarly" without qualification here would lead to people define "majority" and "minority" in the raw—as a matter of popularity—rather than wrt to reception within the discipline in question. Thus, the Evolution article, one might argue, ought to have a large section devoted to Intelligent design and another devoted to Biblical literalism. Most emphatically, it should not. Now, of course, we have pages on Intelligent design and Biblical literalism, socio-religious phenomena of note that are entirely worthy of articles and should take reliable sources just as Evolution itself should.
This is not an inconsistent position; it is, as I've told you, best practice on a number of articles. I have asked you thrice and will ask you again: what is the area of science that you edit where you feel it would be useful to remove the distinction between scholarly and non-scholarly? Marskell 10:48, 29 June 2007 (UTC)
I still don't know what you're saying. You wrote above: "Meanwhile, our NPOV pseudoscience wording has stood for six years (about twice as long as either of us has been editing). '...prioritize the use of scholars, and allow non-scholarly sources only where scholarly ones aren't available...'". Are you saying this quote is in a policy somewhere; if so, where? Or are you referring to this quote: "The task before us is not to describe disputes as though, for example, pseudoscience were on a par with science; rather, the task is to represent the majority (scientific) view as the majority view and the minority (sometimes pseudoscientific) view as the minority view; and, moreover, to explain how scientists have received pseudoscientific theories." Because if the latter, (a) it is about pseudoscience only; and (b) it anyway does not support your interpretation.
It is misleading of you to go around claiming that the NPOV policy has contained your view for six years, when in fact it has never contained your view, and almost certainly never will. SlimVirgin (talk) 02:21, 30 June 2007 (UTC)
This is genuinely confusing. I placed "prioritize the use of scholars, and allow non-scholarly sources only where scholarly ones aren't available" because you had placed it in brackets to (mis)represent my opinion in your post of 20:47, 28 June 2007 (UTC). Do you see that? I won't reply further because I don't think this is being understood. (Though my offer that you provide an example of an area of science you edit that shouldn't have a distinction between scholarly and non-scholarly still stands.) Marskell 06:21, 30 June 2007 (UTC)
It is possible that Factory farming and Talk:Factory farming may answer your question. WAS 4.250 11:57, 29 June 2007 (UTC)
Indeed, the "scholarly or non-scholarly" wording is not supported by the broad consensus of this talk page. Tim Vickers 23:47, 28 June 2007 (UTC)

Self-published sources (online and paper)

A number of users who are editing the article St. Mark's College (University of Adelaide) have used the College Roll to demonstrate the attendance of certain individuals at the college. The roll is where students sign up to the college (actually a residential institution). A particular user claims that this use is not acceptable. Any comments from other users? Ozdaren 12:50, 28 June 2007 (UTC)

Certainly a reliable primary source. Blueboar 14:12, 28 June 2007 (UTC)
Ozdaren, this is a difficult one, and it would depend how the source was being used. First, it may not be notable that the individuals attended that college. Secondly, they may only have registered, then left. Third, they may not be the same individuals you think you're writing about. If there's an objection, it's always better to find a secondary source, particularly if it concerns living persons. SlimVirgin (talk) 17:45, 28 June 2007 (UTC)
Thanks for the comments.Ozdaren 05:32, 29 June 2007 (UTC)
You're welcome. SlimVirgin (talk) 05:38, 29 June 2007 (UTC)


Is accessibility part of the WP:V policy? I am presently trying to collaborate with another editor about the meaning of a word. I am relying upon the definition and etymology provided in the Oxford English Dictionary. Yet I am told that because the online access to the OED is by subscription, or by membership in a library which has a subscription, the OED is therefore not accessible to some editors and therefore not verifiable per WP:V. Also, the OED is not physically available in a convenient local library to some editors. What is the WP:V policy about this? SaltyBoatr 14:13, 28 June 2007 (UTC)

Such a requirement is plainly absurd, of course, and it's not in the policy. BTW, I think it is reasonable to ask an editor to explain exactly what his source says, or possibly to request a precise quotation (there is a handy template for that [need quotation to verify]). It might make sense to mention this in the policy.--Anonymous44 14:41, 28 June 2007 (UTC)
Note that you can cut and paste from the online OED. I would add that much of the online stuff is in draft; it may be better to cite the Second Edition, if it differs. That is, of course, in print; and is an eminently reliable source (and available at many public libraries). Septentrionalis PMAnderson 15:42, 28 June 2007 (UTC)
Thanks I have initially refrained from cutting and pasting from a copyrighted dictionary. I still am not sure this is legal, and have asked for permission over here. SaltyBoatr 15:46, 28 June 2007 (UTC)
Sources must be published; easy accessability is not a requirement but we are not required to believe a contributor's claim that a source says what he says it says if there is evidence to support doubt. Sources must reliable but reliability is relative - an expert in a peer reviewed respected published source trumps a dictionary and a dictionary trumps an offhand comment in a newspaper. Fair use covers short quotes and statements of fact without creativity (regardless of effort expended in compiling the facts) are not covered by copyright in the first place. WAS 4.250 15:53, 28 June 2007 (UTC)
This is interesting. I think a citation should generally be believed true until another editor checks it and fails to verify it; it's our job to check whether the claim is really present in the source. The exception should be cases where the evidence to the contrary is very hard. This would include various striking implausibilties in the citation, and/or a user apparently having lied before on Wikipedia. I did once encounter such a situation (an exceptional claim, supported with technically flawed references to a possibly non-existent journal and a possibly non-existent author, by a user who changed his/her claims several times and was clearly lying that s/he is a scholar) and I insisted on undoing the edits in question. I suppose it might be a good idea for the policy to address such cases, too. --Anonymous44 18:31, 28 June 2007 (UTC)
There is reason to doubt the citation. In the Oxford Dictionary of Current English (published by the same people who right the OED) it says something different from what has been vaguely stated that the online version says. Which begs the question of why they are different if they are published by the same people.--LWF 18:38, 28 June 2007 (UTC)
Well, there can be various reasons for this. This, alone, doesn't sound like a sufficient reason to assume that a user is lying. If you have two different sourced definitions, why not quote (precisely, not "vaguely") both versions? --Anonymous44 18:53, 28 June 2007 (UTC)
The question is what the term, a compound word, does not mean. This involves 'proving a negative'. The Oxford English Dictionary documents 119 unique usages of the key word in the compound word term, none which match. To precisely quote the 119 entries in their entirety, I fear, would violate WP copyright policy. Reading the dictionary should be easy enough. Further, a full text search of the entire 20 volume dictionary and does find any usage of this compound word term. That is very powerful evidence that, per the OED, this compound word term is not observed by them to be standard English. SaltyBoatr 20:26, 28 June 2007 (UTC)
Actually we have quoted (exactly) from other sources which I will list: Oxford Dictionary of Current English, Random House, American Heritage, WordNet, Kernerman, and Webster. These were all dismissed as 'non-Standard English', and then told that the OED is the only proper source. Whereas the other party in this dispute has made only vague references to the OED's definition.--LWF 19:44, 28 June 2007 (UTC)
Pardon me, vague references? I have precisely pointed to a very specific entry in the Oxford English Dictionary. The problem is that you cannot, or will not, read it.
As to the other dictionaries I think they should be considered, but they all obviously carry a lighter weight. For instance, the so-called Oxford Dictionary of Current English which you quoted has only a 20 word entry for the word in question, where the Oxford English Dictionary has a 2,700 word entry. I argue that the more comprehensive dictionary with authority is more reliable. You are arguing that we should ignore the OED because you cannot, or will not, read it. SaltyBoatr 20:12, 28 June 2007 (UTC)

For starters, length is no indicator of higher quality, and when I say vague references I mean that you have not come up with a single quote, sentence, or paraphrase. You just keep talking about warfare and combat. And I still wonder how or why Oxford would not include definitions from all its dictionaries when making a comprehensive dictionary? Especially so when the one it supposedly doesn't contain is a compilation of the current usage of words. I would think that would be one they would make sure to include.--LWF 00:10, 29 June 2007 (UTC)

The OED definition I am looking at is 2,700 words long. Is that 'short' in context of fair use? Also, I am concerned that my cut and paste to the talk page, to save the subscription fee for other editors, violates item #2 of the WP:Fair_Use#Policy. SaltyBoatr 16:11, 28 June 2007 (UTC)
Without going into the detailed Wikipedia:WikiLawyering, or the value of the "powerful" OED in the semantic war, or the details of the slow-burning revert war that the one editor has carried on repeatedly while others try to mediate, it is probably valid to state just what the issue is. Namely, whether or not "Hunting weapon" is a valid expression in the English language. State, Federal, and International usages of the term "hunting weapon" have been offered, but found wanting, relative to the "powerful" OED by the one editor. Comments by requested 3rd-party inputs from the WP:LOCE have also indicated that "hunting weapon" is in common usage, but have been found lacking relative to the "powerful" OED. Also, when presented with "lesser" dictionaries and published regulatory "bad grammar" usages of the term "weapon" that in no way require only combat or warfare relative to the term "weapon", the one editor claims that the OED trumps all these other reliable and verifiable sources, while also stating that only the 2,700 word definition of the OED is acceptable for Wikipedia usage, and that "weapon" implies only combat or warfare. Yet, when asked what part it is of the OED definition that disallows this combined terminology, of "hunting" plus "weapon", no direct quote is provided. I don't think anyone believes the OED is not reliable; however, one editor's interpretation of the OED is certainly at issue, especially when other editors can't check the definition without purchasing a $295 subscription for one year, or for $30 for one month, from the OED online. Wikipedia is not about accepting only the OED as a reference regarding English usage of terminology. There are more sources than just the "powerful" OED that are acceptable for accomplishing verifiability when editing articles on Wikipedia. Yaf 20:38, 28 June 2007 (UTC)
In short, Yaf argues (diff) "The OED is not presently verifiable, and so should not be used." I disagree. My offers[7][8][9][10][11] to help Yaf find free online access to the OED have also been ignored. SaltyBoatr 21:58, 28 June 2007 (UTC)
You have my sympathies in dealing with YAF, who is as partisan as his name suggests. But your sole point is that "Hunting weapon" does not occur under Weapon, which is really not surprising. The OED does not pretend to list every two-word phrase in the language. Septentrionalis PMAnderson 22:25, 28 June 2007 (UTC)
Thanks. Though, it appears you misunderstand me. I see that that the OED is quite clear that in the vast history of English usage of the word 'weapon', almost exclusively, it is used in context of combat or warfare. Hunting is not anything close to combat or warfare. Also, hunting per the OED, is the search or pursuit, not the killing. I do grant that in modern usage, popularized by the Gun culture, the term 'hunting weapon' is acceptable. But that this is a very modern usage and is slang. My true sole point is that considering this evidence I have provided that hunting weapon is a modern slang usage, why not use a more mainstream standard English term like Hunting implement? SaltyBoatr 23:16, 28 June 2007 (UTC)

I notice you say almost. What context is it used under the rest of the time?--LWF 00:10, 29 June 2007 (UTC)

Please read it for yourself, it is enlightening. About 80% of the 119 usages are warfare or combat. A small number are transfer, figurative or vulgar usages, like five usages for the penis being a 'weapon'. There are also transferred usages, where animal parts, like a monkey's teeth, are called weapons. Plus, the figurative usages like 'womans weapons' meaning tears. None of the 119 deal with hunting. SaltyBoatr 00:33, 29 June 2007 (UTC)
Do any deal with injury, harm, damage, or death?--LWF 02:46, 29 June 2007 (UTC)
It's best not to use dictionaries as sources. If you have reliable sources (books, articles) using the term "hunting weapon," then it's a valid term, whether it's in the OED or not. And SaltyBoatr, there's no indication that it's slang. Here's John Bolton, Under Secretary for Arms Control and International Security: "Now, if the AR-15 is viewed as a military weapon, as opposed to a legitimate hunting weapon ...", [12] and thousands of other sources like it. SlimVirgin (talk) 08:01, 29 June 2007 (UTC)
John Bolton is both very recent and quite political; hardly the best counter-example here. This search turns up several 19th-century uses of "hunting weapon", mostly by paleontologists and archaeologists. Septentrionalis PMAnderson 15:16, 29 June 2007 (UTC)
Doing Google searches to 'prove' what is or is not standard English is a bad idea. Google is more of a database, returning all forms of English indiscriminately, including misspellings, slang, and minor varients. The people at Oxford English Dictionary are scholars, methodologically working to document English language usages. They are authorities and carry more weight. Your personal research using searches on Google falls far short in comparison. SaltyBoatr 15:35, 29 June 2007 (UTC)
Perhaps this discussion belongs over on Talk:Hunting weapon, but: I grant that 'hunting weapon' is common in a subset of English that used by the Gun culture. John Bolton is a famous gun rights activist, so it is no surprise he is using this subset of English. My question is, considering that Wikipedia is global: Should Wikipedia use Gun culture English when standard English options exist? Looking to the Oxford English Dictionary seems the obvious thing to do to establish what is standard English. Looking to speeches made by NRA activists, even if they are also government officials, does nothing to prove standard English. SaltyBoatr 15:35, 29 June 2007 (UTC)
Some low minimum bar of accessibility is justifiable and would be beneficial in my opinion. If a source is so obscure and hard to find that verification is inordinately difficult, then theoretical verifiability is not much use. I am reminded of a case where someone cited "The Evening Star of Auckland" - a long defunct New Zealand paper - for a controversial claim. A catalog search indicated that archives of this newspaper existed in only a handful of libraries in the entire world. The paper may have had a splendid reputation for fact-checking in its day, but there was no realistic way to check the citation. (I believe a different source was later provided for the claim.) I think it is sensible to say that long out-of-print materials not generally available in university libraries should not be accepted as the sole sources for a fact. There may be other materials that have extremely limited circulation but would otherwise seem to qualify as reliable sources, but I can't think of any. nadav (talk) 08:25, 29 June 2007 (UTC)
I see that the accessibility issue is indirectly covered here: WP:ATT#Wikipedia articles must be based on reliable sources, with the most reliable sources being 'mainstream' and 'known', therefore I conclude that 'obscure sources' are less reliable. In my specific case, the OED is most reliable. SaltyBoatr 16:18, 29 June 2007 (UTC)
SB, it's almost never appropriate to use a dictionary to determine usage. We look at whether and how reliable sources use a term, not dictionaries. There was an RfAr about this once, when someone tried to use a dictionary to define "capitalism." It's clearly an odd thing to do, when the way the term is used by political theorists and economists is so readily available to us. People who write dictionaries can't be experts on every term they write about. SlimVirgin (talk) 21:57, 29 June 2007 (UTC)
Newspaper articles and dictionaries and peer reviewed expert sources are all reliable sources for some claims. Slim, at WP:ATT we debated the phrase "reliable for the claim" and you said the phrase was not understandable to you. Maybe so. But none-the-less chosing one source over another based on which is more reliable for the claim being sourced is in fact a critical distinction and at the heart of our disagreement at Talk:Factory farming over proper sourcing for the definition of the term "factory farming". An expert in a peer reviewed respected published source trumps a dictionary and a dictionary trumps an offhand comment in a newspaper. WAS 4.250 03:42, 30 June 2007 (UTC)
I wouldn't want to carry the FF discussion here, but as we noted there, and SlimVirgin noted above, dictionaries are rarely considered good sources for usage of terms. The reasons are several, but the most important is that effective definitions of terms is determined by real world usage, in the specific context. Dictionaries are staffed by people who scan the literature and media, and collect information; they are essentially a tertiary source, like us, and in general we stay away from tertiary sources. We need to find the specific articles that relate to our specific topic, and glean the definition from there. So the definition of capitalism should come from relevant papers and books about it, not from dictionaries. Crum375 04:04, 30 June 2007 (UTC)
Crum375 wrote: "...dictionaries are rarely considered good sources for the usage of terms." Astonishing. Isn't that the point of dictionaries, to study and document the usage of terms? See for instance the statement of purpose in the OED 2nd Edition "The aim of this Dictionary is to present in alphabetical series the words that have formed the English vocabulary from the time of the earliest records down to the present day, with all the relevant facts concerning their form, sense-history, pronunciation, and etymology." Clearly, this dictionary is intended as a resource to learn the usage of terms. SaltyBoatr 20:05, 2 July 2007 (UTC)

Web Archive Sources

Can an web archived source be used, when there is a superseding that would appear to contradict it? The case in point is as follows:

The Web Archive source is as follows:

16 April 1999 [13]

Leonard Peltier has now spent twenty-three years in prison. Amnesty International considers Peltier to be a political prisoner whose avenues to legal redress have long been exhausted. The US Government has repeatedly denied requests for a special executive review. Amnesty International recognizes that a retrial is no longer a feasible option and believes that Peltier should be immediately and unconditionally released.

Now, from a AI Bulletin on April 30, 1999 [14]

USA: Amnesty International calls for release of Leonard Peltier [News Service 080/99, 27 April 1999]. Amnesty International is not able to take a position regarding Leonard Peltier's guilt or innocence, and does not consider him a prisoner of conscience, but has decided to call for his release from prison because of continuing doubts about the fairness of the legal proceedings against him. A retrial no longer seems a feasible option.

Now, an AI July 15, 2007 statement: [15]

Amnesty International has investigated this case for many years. Although Amnesty International has not adopted Leonard Peltier as a prisoner of conscience, the organization remains concerned about the fairness of the proceedings leading to his conviction and believes that political factors may have influenced the way in which the case was prosecuted.

I suppose my question would be, can we use a dead link as a source stating that AI believes Leonard Peltier to be a political prisoner, even when the source is no longer active, and material published at a later date would indicate that AI takes the opposite position?

Thanks. Torturous Devastating Cudgel 15:54, 28 June 2007 (UTC)

Your sources indicate that AI's position on Leonard Peltier status as a political prisoner is unclear. Say so and use all three sources as your reference. WAS 4.250 16:02, 28 June 2007 (UTC)
Would not the two more recent statements, where they do not take the position that he is a PP or PoC, be the current position of AI considering that the first link has been taken down by AI? Torturous Devastating Cudgel 16:13, 28 June 2007 (UTC)
I'd have to say that when using the same source, the most recent info/opinion that source has regarding a person or issue is the one to go with.--MONGO 20:06, 28 June 2007 (UTC)
Only if there is clear evidence that the source is not divided against itself. The differences in opinion may not be temporal in origin. WAS 4.250 22:47, 28 June 2007 (UTC)
I would probably refer to all three statements. It would be worth looking around to see whether a secondary source has written about how Amnesty changed its mind within the space of two weeks. SlimVirgin (talk) 08:04, 29 June 2007 (UTC)
Note that it is quite possible that "political prisoner" and "prisoner of conscience" are not the same thing. AI affirms the first and denies the second. Septentrionalis PMAnderson 15:09, 29 June 2007 (UTC)
Yes, good point. I suppose that's the danger of relying on primary sources: we don't really know what they meant. Could be the first statement was a mistake by an intern or something. SlimVirgin (talk) 21:00, 29 June 2007 (UTC)

I searched AmnestyUSA; according to their FAQ, these are both technical terms.

  • Prisoners of Conscience (POCs) are people detained solely because of their beliefs or because of their ethnic origin, sex, color, language, national or social origin, economic status, birth, or other status who have not used or advocated violence.
  • A political prisoner is anyone whose imprisonment is politically motivated. Perhaps the "crime" is politically motivated; perhaps the arrest, trial, or sentence has political undertones.

It is perfectly reasonable to hold that Peltier is one but not the other; he is not detained solely because of his beliefs etc. (Whether he used violence is part of the controversy.) The end of the third quote above virtually repeats the definition of political prisoner. Since the source is not inconsistent, we need not analyze further. Septentrionalis PMAnderson 22:30, 29 June 2007 (UTC)

This is ridiculous

All of you know better than to continue edit warring on a locked page. Stop doing it. WilyD 20:50, 29 June 2007 (UTC)

Please see User talk:David Levy#WP:SOURCE. I'm baffled as to why SlimVirgin objects to a dummy edit and believes that we're required to retain an incorrect Dutch interwiki link (and no protection tag) for the protection's duration. —David Levy 21:06, 29 June 2007 (UTC)
Because, a) it is protected, and b) it is always protected on the wrong version. Please note that SlimVirgin notes that the protected version is not her preferred version. I will request the protecting Admin to apply the tag. LessHeard vanU 21:18, 29 June 2007 (UTC)
Huh? I'm referring to a minor correction that has absolutely nothing to do with the dispute. Per the protection policy, those are allowed. Why in the world wouldn't they be?
There was a protection tag in place, and SlimVirgin inexplicably responded to my dummy edit by removing it and once again reverting to the incorrect Dutch interwiki link. —David Levy 21:31, 29 June 2007 (UTC)
(edit conflict)It is now tagged, by the protecting Admin. While there may be legitimate edits outside of the disputed areas it may be considered a more chilled attitude to permit these indiscetions to remain while the disputed matter is sorted out. If there is an absolute need to edit something which does not impinge on the dispute then why not get consensus on the talkpage before performing it - it would certainly clarify if it is not part of the disputed content for a start - and not potentially inflame the matter. Cheers. LessHeard vanU 21:56, 29 June 2007 (UTC)
Yes, it's absurd. I understand that SV was probably under the influence of a WP:MASTODON effect (the previous admin who edited the page, mistakenly but mostly due to a coincidence, also suffered her wrath); but I don't understand LessHeard vanU's reaction.--Anonymous44 21:47, 29 June 2007 (UTC)
Uncontraversial changes may sometimes be made to protected pages. When these changes result in edit warring, they're not uncontraversial, and they have to stop immeadiately. Edit warring is bad. Edit warring on a protected page is double plus bad - it's wheel warring. Fixing shitty grammer or correcting an interwiki link are not noble purposes worth wheel warring over - and the world won't end if those tasks aren't done today. WilyD 21:52, 29 June 2007 (UTC)
There was no edit warring. It would have been edit warring if somebody had been reverting his edit on purpose, but the impression made was that it was a side-effect of the other reverts. When it became clear that SV was doing it on purpose, he stopped. --Anonymous44 22:12, 29 June 2007 (UTC)
That's exactly right. :-) —David Levy 22:19, 29 June 2007 (UTC)
I agree, and I didn't change the interwiki link since it became apparent that SlimVirgin objected to its correction. I don't, however, understand why she objected to its correction, let alone to a dummy edit (to which she responded, for some reason, by reverting to the incorrect interwiki link and removing the protection tag). —David Levy 22:19, 29 June 2007 (UTC)

(Resetting indent)This may be partly my fault. I forgot to add the tag. That is now rectified. The tag says "until disputes have been resolved". It doesn't look as if disputes are any closer to being resolved right now, so I have made it indefinite. That doesn't mean infinite. I don't think it's a good idea for pages to remain locked for long, long periods, and I'm more than happy to unprotect as soon as the people here feel they've come to some kind of consensus — even a grudging one. Or request unprotection at WP:RfPP. It would be nice if the page could be unprotected as soon as I originally envisaged, or sooner.

I don't think admins should edit over protection, except to correct things like obvious spelling mistakes — but not to make edits that other editors will disagree with. Regarding the Dutch interwiki link, the article, in the version I locked it in, has nl:Wikipedia:Bronvermelding. If you go there, and click on the English interwiki link, it will take you to Wikipedia:Citing sources. I checked the interwiki link that David was changing to, which is nl:Wikipedia:Verifieerbaarheid, and that links back to this Verifiability page. Therefore, I think that's an uncontroversial change. If people agree, I'm happy to make that change, as an uninvolved administrator, who protected the page, but only if I'm sure that there's no dispute over that. (Actually, I am fairly sure.)

Regarding the shortcut, I haven't figured out what's going on. If it's an uncontroversial change, I'd be happy to make it, but I really think it should come from the talk page first, rather than an admin deciding to make a change, and other admins objecting. Again, my apologies for forgetting to add the tag. ElinorD (talk) 21:57, 29 June 2007 (UTC)

Elinor, I have no problem with interwiki links or tags being fixed. My concern is that David kept changing what the shortcut should direct to, which is something that has been in dispute. Admins must not edit protected pages if the issue is one that someone is objecting to; otherwise there's no point in protecting it. The current version is not one that I want either, but that doesn't give me the right to change it. SlimVirgin (talk) 22:02, 29 June 2007 (UTC)
I'm going to assume that there's no controversy over the interwiki tag, and I'll change it myself. To avoid a repeat of the flurry of edits after the protection, I ask that if anyone thinks I shouldn't have done it, please let me know, rather than reverting, an I'll revert myself. There does seem to be a dispute over the tag, so I won't touch that. If there's anything else that there's agreement about, please let me know. If it's something small, I'll do it; and if you've all come to some kind of agreement over the main dispute, I'll unprotect. ElinorD (talk) 22:10, 29 June 2007 (UTC)
Please see User talk:David Levy#WP:SOURCE and [[Talk:WP:SOURCE]] for an explanation of the shortcut issue. I was correcting documentation that became outdated when SlimVirgin mass-reverted the page, not changing the actual shortcut. SlimVigin, however, has since changed the actual shortcut and convinced me that her setup is preferable. —David Levy 22:19, 29 June 2007 (UTC)
Again, I merely edited the page to correctly indicate the shortcut's purpose. I DID NOT EDIT THE SHORTCUT, which had led to that section without complaint from you or anyone else for the past week. Until you complained and edited the shortcut (which I made no attempt to revert), there was absolutely no indication that this was a disputed matter.
If you didn't object to the correction of the interwiki link, why did you revert it (and remove the protection tag)? —David Levy 22:19, 29 June 2007 (UTC)

May be its they work toghter. I dont think its good to do that to much because sometimes it can be like admins are trying to get arount the 3RR rule.[16][17] —The preceding unsigned comment was added by Practical Pig (talkcontribs) 21:54, 29 June 2007.

Popular culture

I'd like to open up a discussion about including a section in the self-published sources section that allows them to be used for pop culture articles. This was something that editors wanted to add to WP:ATT, but it was vetoed by others. I always felt that was a mistake. How would people feel about adding it to this page?

Some articles about popular culture rely on self-published posts on bulletin boards, blogs, and Usenet, because few other sources exist for them. In such cases, the material used must have been posted by named, or well-known pseudonymous, individuals with a known expertise in the area, although the individual need not be a professional in a relevant field. Anonymous posts should never be used. Some areas of popular culture also rely on self-published secondary sources, such as well-known fansites. If in doubt about how to use a source in this area, consult the relevant WikiProjects for advice. This provision does not apply to material about living persons, or to topics in other areas, such as history, science, religion, literature, or current affairs.

SlimVirgin (talk) 04:02, 30 June 2007 (UTC)

Sounds good to me. JulesH 07:07, 30 June 2007 (UTC)
If we caveat both aspects of topics, as well as topics themselves, I can live with inclusion. "This provision does not apply to material about living persons, or to topics in other areas, such as history, science, religion, literature, or current affairs." OK, and perhaps add something like "even within popular culture articles, quotations, statistics, and judgements of quality or popularity, should take reliable sources as otherwise defined by policy. If sources cannot be found, such information should not be added." This would allow pop cult editors to describe "in-universe" aspects of topics with less than ideal sources, as they are often wont to do, but not allow an editor to declare such-and-such the greatest video game ever with a dubious source. Marskell 07:43, 30 June 2007 (UTC)
It's often on these specialist websites that you'll find reviews and comparisons. If we say they can't use quotations, statistics, or judgements of quality or popularity, the resulting articles will be deadly dull. The proposal is to allow appropriate sources for popular culture as we do in every other area, so that those editors can create good articles, not shallow ones. Well, they do that already. They just ignore this policy. SlimVirgin (talk) 08:29, 30 June 2007 (UTC)
What I worry about is the dentist problem—two out of three agree that every toothpaste on the market is the best one available. Fansites, by definition, laud their subject and often make exaggerated claims; it's " of the greatest ever..."-type wording that I find deadly dull on pop culture.
While I was busy ensuring ATT would fail, I did in fact present "aspects of" wording that you found amendable, here. I do think it better to focus on the type of information involved, rather than make an exception for a broad category of article (the definition of which is far from clear-cut).
But I rarely edit in this area. I know some of our best pop cult editors don't ignore policy but wind up agreeing with it because they want to improve their articles and actively seek out reliable sources that are neither commercially tied to their subject nor weak fan writing. Marskell 12:06, 30 June 2007 (UTC)
  • Sounds good to me. I used a couple of sources where I had to give some pretty good rationale as to why I'm using them, but they have improved the article(s) twentyfold. As long as we make the playing field clear, it should be fine. — Deckiller 13:59, 30 June 2007 (UTC)
  • I have some problem with the wording "Some articles about popular culture". What does "some" mean? pop culture topics that have few "reliable" sources about them? There are some pop culture articles where there are plenty of newspaper, magazine, and other fairly reliable sources. If at all possible, we should strive to to use these sources. If a topic really lacks such reliable sources (enough sources to demonstrate notability of the topic), then it may simply not be notable enough or worth including in Wikipedia? In some cases that might be the better alternative to going out and relying on bulletin board posts, usenet posts, and blogs. --Aude (talk) 14:54, 30 June 2007 (UTC)
  • I assume SlimVirgin is referring to this discussion that involved a relatively small number of editors. I'm rather puzzled by the suggested wording and where it might fit into the policy page without sounding totally contradictory. It reads more like an admission of failure than a guideline. To mockingly paraphrase:
Some articles on WP break our fundamental and long cherished policies on using reliable sources. But that's ok. We know how hard it can be to stick to the rules. Just stay away from the articles read by adults, please.
(please read this with a sense of humour)
SlimVirgin asks how we'd feel about this: very worried indeed. WP's sources are its foundation. Pop culture is a huge part of WP. This proposal undermines that foundation in a big way. BTW: I'm a newbie wrt to policy but hang around WP:FLC so I've seen plenty good pop culture articles that are based on reliable sources. If this exclusion is allowed, I'm going to end up seeing:
This list meets WP:V because it is based on Derek's geocities web site. His site rocks and is totally respected.
Sigh. Colin°Talk 19:25, 30 June 2007 (UTC)
I think that formulation is pretty good, SlimVirgin. One suggestion would be that for any "well-known" fansites or blogs to be used, these should have been mentioned in reliable sources, such as magazines or newspapers. This would put a crimp on the "Well we all know about Derek's geocities web site" argument. Tim Vickers 19:47, 30 June 2007 (UTC)
The phrase "According to Wikipedia..." appears in the press with depressing regularity. If such a well-known unreliable source can be mentioned so frequently, what makes you think they can be used as a judge of internet sites? Colin°Talk 22:36, 30 June 2007 (UTC)
  • The meme problem mentioned by Marskell in the previous discussion worth repeating. These sites, keen to show how in-depth their knowledge is, feed off each other. A "fact" then gains currency by being mentioned in multiple places, when actually there was only one dodgy source. The most serious version of this is the tendency for amateur sites to use Wikipedia as a source. Not only is it considered reliable, but you can nick the text with a clean conscience. I've come across numerous personal web sites that use Wikipedia text. Sometimes credited, sometimes not. Let's add an obscure fact to Wikipedia: "XYZ was released on DVD in France in October 2005 with subtitles and a 4:3 aspect ratio." A few months later, I can probably source this fact to a "well-known fan site". Colin°Talk 22:36, 30 June 2007 (UTC)

If it is possible to take something that is self-published, but reliable and reasonably trust worthy, as a source, then I'm all for it. There are "fan websites" that are professionally run, and are anything but Joeshmoe's geocities website. They are technically self-published, but are as respectable and reliable as a published source. Website such as , where a good many anime reviews are found. One way or another, we'll need to find a way to use these sources in an acceptable fashion for Wikipedia. Soon it won't be just pop-culture that is effected by such things, but many other topics we cover.

Maybe we need to really look at this at the fundamental level. Given the reasons we stay away from self-published sources, what can we do to validate the ones we need, and how do we determine which ones are even respectable, etc. -- Ned Scott 06:05, 1 July 2007 (UTC)

Such as, a somewhat strict criteria for said websites. -- Ned Scott 06:07, 1 July 2007 (UTC)

Verifying the individual

I'm just going to pull out one sentence of Tim's to unpack it: "One suggestion would be that for any "well-known" fansites or blogs to be used, these should have been mentioned in reliable sources, such as magazines or newspapers." I had three principal problems with the exception (noted in part in Colin's link). One was the circular logic involved: if the whole reason that Usenet is not reliable is because we cannot ascertain who posted what where, how can we ascertain that "named, or well-known pseudonymous, individuals with a known expertise in the area" are behind the material? The qualification continues to beg the question. Thus I think wording similar to Tim's would be appropriate. Everybody's favourite example was Spoo, a FA which uses Usenet as a source. Thus the requirement would not be that you have to source the particular fact to a reliable magazine or newspaper, only the general fact that the individual involved did in fact use the medium. If "everybody knows" that one of the developers of Spoo used Usenet, source that to a reliable magazine etc., and then you can use Usenet for the specifics in the article.

That's one of three. Two of three: how are we defining pop cult? This is obviously of importance. Three of three: this does create dual standards, however you look at it. While we need to be sensitive to the editors who desire the exception, the reverse problem will likely occur: threads here to the effect of "when the heck did this happen?" Marskell 13:50, 1 July 2007 (UTC)

Isn't the answer that you are going to find it hard (impossible) to define a general rule, it needs to be evaluated on a case by case basis. In one area a fan site will be a respectable tome, another far more unreliable. Yet within either, some stuff will be good quality, other stuff poor. Perhaps that is the only wording: Where reference is made to non-scholarly sources, especially informal publications such as fansites or blogs, a judgement needs to be made as to the quality of the site and therefore the value of the citation on a case by case basis. It is not just a problem for blogs, it is for newspapers, USA Today is probably a less worthy source than the Herald Tribune, The Sunday Sport less than The Independent. Spenny 16:20, 1 July 2007 (UTC)


Articles about such sources should not repeat any potentially libelous claims the source has made about third parties, unless those claims have also been published by reliable sources.

I'm a bit (a lot!) concerned about that wording. It has the implication that if you can find someone in a reliable source making a libellous claim then it is quite OK to spread the libel through Wiki. If it is truly libellous then you simply cannot repeat the claim, you are party to that libel. Newspapers and TV do not use the fact that another paper has published a claim to allow them to use that claim. They will couch the reporting of the problem in the most careful language. Worse, in UK law the burden is on the one committing the libel to prove that it is not libellous. I think this needs some serious review - I'd go as far as to suggest a real Wiki-Lawyer review that text.

I suggest:

Articles about such sources should not repeat any potentially libelous claims the source has made about third parties, even if those claims have been published by reliable sources. Spenny 10:59, 30 June 2007 (UTC)

Yes, that's a bad sentence. I have a memory of rewording it, or perhaps I only intended to. The problem with the one you suggest in its place is that editors aren't in a position to judge what's potentially libelous: if the NYT sees fit to publish something that is potentially libelous after extensive discussion with some of the most expensive libel lawyers in the world, it would be odd for us to assume we knew better. Also, BLP has moved on since that sentence was added, so it's no longer appropriate anyway (bad sources may never be used for living persons, whether the claim is contentious/libelous or not). I suggest the wording we had in an early version of WP:ATT:

A questionable source is one with no editorial oversight or fact-checking policy or with a poor reputation for fact-checking. Such sources include publications that are widely acknowledged as extremist; or that rely heavily on rumors, satire, or personal opinions. Questionable sources may only be used in articles about themselves; these articles should not repeat claims the source has made about third parties unless those claims have also been published by reliable sources, which must be cited.

SlimVirgin (talk) 20:41, 30 June 2007 (UTC)

Good points, in fact most critical statements about living people could be interpreted as "potentially" libellous. Asking editors to consider this is particularly complicated in that they would have to check if the material might be objectionable under the libel laws of any country in which Wikipedia might be prosecuted. This is an impossible task to ask of an editor, and an extremely difficult one even for a experienced lawyer. Tim Vickers 21:20, 30 June 2007 (UTC)

I would suggest (after a couple of beers) that the real libellous stuff is fairly obvious. The main point is to say: don't believe you can repeat a libel just because someone else said it. (colon intended!) Spenny 23:10, 30 June 2007 (UTC)
...but more to the point, I would assume a Wiki policy that says "users must not repeat libel" would be more defensible in court for the Foundation than a policy that says "users may repeat libel if they've read it somewhere reputable." That may not ultimately be a useful defence, but it is worth running past the real lawyers. With regard to SV's prior version, it still has the implication that if someone else says it then we can say it. That is not the way UK law works. We must at least strive for plausible deniability in policy, reliable sources will sometimes make libellous claims. I think there is some wording that can accept good faith repetition of the "slightly possibly" libellous without the assertion that relying on a third party absolves us of being libellous. Seriously, having being threatened with libel over a trivial matter in the real world, I suggest a proper legal wording is required here. Spenny 00:07, 1 July 2007 (UTC)

In the end, though, the problem is that to write informative articles on the subject of questionable sources, we must repeat libellous claims, because if we don't it is impossible to characterise the kinds of claims made by the source. See, e.g. The Sun, which repeats libellous claims made by that newspaper because they are part of what makes the source notable. Perhaps the section needs to make clear that such claims should not be phrased in such a way that implies they are true, but that stating they have been made by the source is acceptable. JulesH 00:15, 1 July 2007 (UTC)

Obviously, we do have to be able to repeat in good faith negative assertions made by reputable sources. For example, if the New York Times says "Jane Crummy is a known drug addict and prostitute who sold blasting caps to terrorists", then we should be able to say so. By US standards, for a factual claim to be libelous, it must be both false and either made with malicious intent (i.e. by some who knew in advance it was false) or with reckless disregard for the truth. As long as the claim is traceable to identified reliable sources, one will generally avoid any risk of being found to have been reckless with the truth. In the US at least, as long as it is prefixed with "According to the New York Times", you can pretty much repeat anything they say with zero risk of libel liability. Dragons flight 00:55, 1 July 2007 (UTC)

OK, perhaps we have a split here between US and UK law, but recently there has been an indication that the UK courts have been used by US citizens to pursue libel under the legal framework here which is more conducive to a successful action. I think that it is important to see that there is an element of newspaper reporting to come in here where the urge to be timely undermines the useful passage of time to resolve things. I think a good example is the George Michael incident of last year, which we can talk about freely now as the facts have been discussed in court. It was widely reported that he had been found slumped over the steering wheel. The assumption was that it was drug related (a valid assumption later admitted and apologised for). At that time, it would have been inappropriate for Wiki to comment on the speculative cause of his lapse being drugs, and inappropriate to report what the newspapers were reporting as the speculative cause as it had the potential to be defamatory. There could have been an innocent explanation, some undiscovered medical condition of a private nature for example. Wiki is not a newspaper and can and should wait for clarity on such issues.
So in terms of policy, I see it as very important to have the guideline of no libel, regardless of source, especially as Wiki has such prominence these days. If people use good judgement to place the more informed speculation with solid reputable sources, then it is a judgement call, but Wiki should set itself the highest standards. Spenny 09:36, 1 July 2007 (UTC)
I don't see it as inappropriate to have included the speculation as long as it was clearly identified as speculation from mainstream sources. I also don't see what a guideline of "no libel" would even mean. Intentionally malicious libel (i.e. that were Wikipedians knowingly lie) is certainly already excluded, so that only leaves reckless libel. And if you have firm sources, it isn't reckless. Wikipedia, as an encyclopedia, doesn't anticipate having "truth" per se, but rather is only able to discuss what other reliable and verifiable sources describe as the truth. Adhering to our existing content policies should be sufficient to eliminate any threat of libel. Dragons flight 10:15, 2 July 2007 (UTC)
You say "I also don't see what a guideline of "no libel" would even mean." Wikipedia:Libel is already policy. WAS 4.250 10:43, 2 July 2007 (UTC)
Yes, but that basically says that we firmly apply existing content policies to prevent defamation. I get the impression that Spenny wants something more than that (or else why discuss it?), and that's where I don't understand what he wants "no libel" to mean. I assume it is blatantly obviously that no one wants to intentionally libel others, so it is all a question of working around potential libel, for which our existing content policies seem more than adequate with respect to US law. I am realizing though that there are apparently important differences between US and UK standards for defamation. Respectfully though, I see no point in using the UK standard on a US based site, any more than I would using a German, French or Iranian standard. Dragons flight 11:09, 2 July 2007 (UTC)
I'm not that wedded to this to see it worth pushing further, perhaps I am being too nitpicky. Simply to repeat that my point was that we have a guideline that undermines the principle of do no evil as it suggests that if someone else says something nasty and we can prove that someone else said it, even if the circumstances are not resolved, then we can say something nasty too. In a firefight, that wording will be unhelpful, it is not really an encyclopaedic approach and is not consistent with WP:BLP. Spenny 10:55, 2 July 2007 (UTC)
If a mainstream, reputable organization (e.g. the New York Times) has already made the assertion, I don't see how repeating it is significantly "evil", since it is already out there and prominent. I'd agree that we wouldn't want to report claims from third tier rumor mills, but that should already be covered by the requirement of good sourcing and exceptional evidence for exceptional claims. Again, I don't think I understand what change it is you are looking for? Dragons flight 11:09, 2 July 2007 (UTC)
I'll give you a current example. An important Ferrari manager has been accused of various things, including sabotage. ITV did a small presentation on their F1 coverage where it was quite clear that, whatever the facts were, it was all so murky as to be very unsafe waters to be wading in, I'll bet the lawyers were sweating over every word. Wikipedia has no obligation to be there however well based you can make that speculation. In this case it might be appropriate to say that there was a legal dispute, but it might not be appropriate to repeat the specific allegations, especially as one of the comments about the case was that the allegations were firmly denied and claimed to be part of a character assassination. Given that context, being party to continuing the disputed allegations is inappropriate.
The article Nigel Stepney covers that ground. The article as it stands is as fair reflection of the press, but given the dispute, it probably would be helpful not to repeat the disputed allegations. A newspaper article disappears in a few days, this is a semi-permanent record. A real encyclopaedia would not contemplate that kind of material. Bother, I said I was going to let it drop!! Spenny 12:00, 2 July 2007 (UTC)

Scholarly and non-scholarly sources

One of the editors who wanted to add to V and RS that scholarly sources should be prioritized over non-scholarly ones posted on my talk page that reliable non-scholarly sources are being "actively suppressed" in certain science subjects. An example he gave was Global warming, and examples of reliable sources being suppressed were The Wall Street Journal, the Telegraph, and the National Post. The comment in full [18]:

Non-scholarly opinion is actively suppressed on a number of our well-hit science articles, even when particularly widespread and even when noted in reliable but non-scientific sources. Global Warming has remained an FA precisely because editors have gamely stuck to that principle. With the WSJ, the Telegraph, and the National Post—all de jure reliable—you could create a fine (but generic) skepticism section. And it would have no place in the article. Thankfully, it won't wind up in the article while our better science editors are watching it.

I have a lot of sympathy for editors on pages like Global warming given the number of cranks that swarm around them, but I feel the protection of these articles has gone too far if dissenting views published by reliable sources are being "actively suppressed," just because the publications aren't peer-reviewed.

My experience of scholars is that they want to educate people, and that includes teaching them about material that might make them wince, but that needs to be known about so their students can judge the field for themselves. That's what we're here to do as Wikipedians: let our readers know about all reasonable published opinion, so that people can make up their own minds.

I would therefore like to add something to the beginning of our Sources section to help editors who find their material being challenged because they're not using peer-reviewed sources. Suggestion:

Articles should be supported by reliable, third-party published sources with a reputation for fact-checking and accuracy. All articles must adhere to Wikipedia's neutrality policy, which requires the inclusion of all majority and significant-minority views that have been published by a reliable source.[1]

Wikipedia offers no definition of what constitutes a "reliable source." Sources should be appropriate for the topic and the claims made, and appropriateness is a matter of editorial judgment. In general, the most reliable sources are academic books and journals; university-level textbooks; magazines and journals published by respected publishing houses; and mainstream newspapers. As a rule of thumb, the more people engaged in checking facts, analyzing legal issues, and scrutinizing the writing, the more reliable the publication.

Scholarly and peer-reviewed sources are highly valued and are likely to be the sources of choice in areas such as history and science, but material from non-scholarly reliable sources is also welcome in these areas, particularly if they are mainstream sources with good reputations. The appropriateness of such sources will depend on the context. Where there is disagreement between scholarly and non-scholarly sources, the views should be clearly attributed in the text: the scholarly view is presented as the scholarly view, and a strong and reliable dissenting view as a strong and reliable dissenting view. Views should be presented in rough proportion to their coverage by reliable sources; that is, the majority view should be presented as such, in accordance with WP:UNDUE.

  1. ^ Tiny-minority views and fringe theories need not be included, except perhaps in articles devoted to them.

Any thoughts? SlimVirgin (talk) 23:09, 30 June 2007 (UTC)

Yes. This makes no sense, so people will ignore it. There's no other way to characterize the idea that on a scientific topic, a blurb in USA Today (a "mainstream newspaper") deserves the same regard as a paper in the Proceedings of the National Academy of Sciences. A rule that is bound to be widely ignored as nonsensical will be counterproductive. Raymond Arritt 23:32, 30 June 2007 (UTC)
You are aware of course that the articles referred to are op-ed's/editorials and not written by journalists under the normal strong fact-checking and standards that those news outlets may normally have? For instance the National Post article series, that i assume are referred, is a profile series that has so far been objected to by at least 4 of the persons that have been the focus of the series. (Good (and extreme) example [19] - where the NP had to write an excuse on top of the online article). I'm interested in what consensus there is here on the reliability of an op-ed/editorial as a source though.--Kim D. Petersen 23:44, 30 June 2007 (UTC)
Kim, what do you think the difference is between the kind of analysis piece you linked to and articles "written by journalists under the normal strong fact-checking and standards"? There is no difference. Articles and editorials in newspapers undergo varying degrees of peer-review prior to publication, which may be rigorous or not depending on various factors, but rarely if ever depending on what kind of piece it is. The point here is this: if you want to keep that material out, you must change the NPOV policy. If you want to do that, you must do it via the Foundation, not by trying to change this policy. Wikipedia has never promoted scholarly point of view. Scholarly views are highly valued, yes, and are often de facto prioritized, but the kind of de jure prioritization that people were trying to add to V and RS is wrong-headed. SlimVirgin (talk) 00:19, 1 July 2007 (UTC)
Thats actually rather simple - first look up under op-ed and editorial. Both are reflections of a single persons opinion and are clearly marked as such by being in the op-ed or editorial columns. The National Post pieces are not written by a journalist, they are written by a director of a think-tank/lobby group. And are not the result of (as one would expect from the style) interviews and investigative journalism. Lets take another example from the same series - one in which the author actually gets the opinion of the subject reasonably right - but where the subject protests anyways: article - subjects comment. What i focused on was this: "I presume that the writer just surfed my website, picked whatever he liked, glued it together and voilà, got an original article. I presume it is legitimate, and it even saves a long distance phone call." - To me this doesn't give me a good impression of the Journalistic "peer-reviews" nor of any "reliability" to the source. Its a simple op-ed/editorial - which by definition is pure POV. --Kim D. Petersen 10:34, 1 July 2007 (UTC)
It may be useful to note that what these NP articles do is explain the views of scholars who supposedly more or less deny global warming. So even these media seem to assume that it is the scholars' views that count. This is hardly a good example for using non-scholarly views/sources. As far as the journalist has done some significant original scientific analysis himself, an analysis that no scholar has done before, it's clear that his competence to do it is somewhat questionable.--Anonymous44 20:05, 1 July 2007 (UTC)
That's a very constructive suggestion SlimVirgin. Thank you. I would suggest that you specifically cast it as a way of assessing relative weight:

Draft proposal, please feel free to edit

All articles must adhere to Wikipedia's neutrality policy, with their writing fairly representing all significant viewpoints that have been published by reliable published sources, in proportion to the prominence of each view.[1]

Articles should be referenced to reliable published sources that have a reputation for accuracy. Sources should be appropriate for the topic and the claims made, with appropriateness determined by editorial judgment and common sense, guided by consensus. In general, the most reliable sources are peer-reviewed journals and books published in university presses; university-level textbooks; magazines, journals, and books published by respected publishing houses; and mainstream newspapers. As a rule of thumb, the greater the degree of scrutiny involved in checking facts, analyzing legal issues, and scrutinizing the evidence and arguments of a particular work, the more reliable it is.

Academic and peer-reviewed sources are highly valued and should usually be given most weight in areas such as history, medicine, and science. Material from reliable non-academic sources can also be used in these areas, if they are respected mainstream sources with good reputations for accuracy. The appropriateness of sources always depends on context and non-academic sources should not be used as the sole support for bizarre or extraordinary claims.

Discussion of draft

I have also removed the instructions on dividing scholarly/non-scholarly sources in the text. This seems an unnecessary diversion into telling editors how to lay out their articles, rather then talking about sources. Tim Vickers 00:00, 1 July 2007 (UTC)

Also most academic journals are published by socieites or are commercial, rather than being published by universities directly. Tim Vickers 00:26, 1 July 2007 (UTC)
What's a "verifiable source"? SlimVirgin (talk) 00:21, 1 July 2007 (UTC)
I think that means a reliable source, I copied that part directly from WP:NPOV in the hope that wouldn't be controversial! :) I'll change it to reliable source. Maybe we should chage that in the NPOV policy as well? Tim Vickers 00:26, 1 July 2007 (UTC)
What's a "known publishing house"? Would "reputable" be better? Tim Vickers 00:28, 1 July 2007 (UTC)
Reputable would be fine. Regarding "verifiable source," the problem with these policies (NOR, V, and NPOV) is that they're not well written, so copying from one to another is usually not a good idea, because it just transfers problems. A few of us spent several months revamping NOR and V for WP:ATT, which got rid of some of the writing issues (though not all), but it was rejected, so we're stuck with these two pages. The word "verifiability" is one of the biggest problems, because it's meaningless as we use it here, but I've argued this until I'm blue in the face and can't argue it anymore. This is why I have to laugh when accused of WP:OWN. I strongly dislike the way these policies are written: all I try to do is stop them from getting worse. SlimVirgin (talk) 00:37, 1 July 2007 (UTC)
Actually, best to hold off on "reputable," because I can see people using it as a battering ram. SlimVirgin (talk) 00:38, 1 July 2007 (UTC)
We seem to have been operating completely at cross-purposes SlimVirgin, I agree entirely that these policies aren't well-written and all I was trying to do on the other policies was improve their clarity, without changing their meaning - but perhaps more quickly than people here were used to! Anyway, let's put all that behind us. How should we deal with the fringe publishing houses putting out stuff like neo-nazi literature or creationist tracts? These are certainly "known" (if only for the wrong reasons :) ) but edit-warriors could argue that they fit this new proposal as reliable sources. What about just "mainstream magazines, newspapers and journals" that would leave us with just one vagueish term. Tim Vickers 00:48, 1 July 2007 (UTC)
I'm confused, do you want editors to develop several different different proposals and then try to merge them later? Or to all edit one proposal and try to reach a single consensus version? What do you think is the best way to proceed? Tim Vickers 01:57, 1 July 2007 (UTC)

I would be concerned about adding a statement such as Wikipedia offers no definition of what constitutes a "reliable source." While Wikipedia need not offer a definition per se, it does not need to state it so categorically since in the next sentence, sources are qualified based on their degree of reliability. I am generally sympathetic to the inclusion of minority views as long as they don't veer too far off from the mainstream. Since the threshold for inclusion is verifiability, almost any sourced statement can be included if properly weighed. I am glad, however, that there does appear to be consensus regarding the re-inclusion of this qualification: In general, the most reliable sources are books published by universities and peer-reviewed journals. As someone who edits mostly mainspace and runs into similar issues, including all manner of sourced nonsense, I understand completely why it is so important to add this qualification. I think it would be helpful also to make direct reference to undue weight. — Zerida 04:52, 1 July 2007 (UTC)

I've added a bit about UNDUE. We don't want to get too much more into that, though, because it's from the NPOV policy. But you're right that it's worth mentioning. SlimVirgin (talk) 08:28, 1 July 2007 (UTC)
Slim, would you like to take out a banner with "User:Marskell has used the words 'actively suppressed'" and fly it in a major city somewhere? Perhaps it will make you feel better.
I can live with the suggested if the caveat scientific fact as such, should only be challenged with sources of a scientific nature is added, which we can then link to the arbcom. Marskell 08:43, 1 July 2007 (UTC)
No, definitely not. That's exactly the attitude that we need this section to defeat. We present material published by reliable sources. If a serious dissenting view is published about a scientific theory, we publish it, attributing it to its sources, and representing it in relation to its representation by reliable sources.
We are not here to "challenge." We are not here as researchers. We are here as educators, as librarians. As another editor said, someone walks into our library and asks what we have on Global warming. We say we have this by scientist A, and this by scientist B, and oh, we have this Wall Street Journal journalist who's written a dissenting view: would you like to see his work too? That's all we do. We don't hide certain books because the scientists don't like them — or because some of our librarians believe the scientists might not like them. To imagine we might do that is to completely misunderstand this project. SlimVirgin (talk) 08:50, 1 July 2007 (UTC)
OK, then I oppose the wording. A journalist cannot, properly speaking, write a "dissenting view" scientific theory if they do not publish it in a scientific source.
I also agree with Zerida regarding "Wikipedia offers no definition of what constitutes a 'reliable source.'" While that may be true, I don't think it's something we should actually say given potential misuse. Marskell 08:57, 1 July 2007 (UTC)
Well, I know you oppose the wording, Marskell. It's because of your comments about reliable sources being suppressed, and your attempt to add SPOV to this page, that I'm suggesting this addition in the first place. SlimVirgin (talk) 09:01, 1 July 2007 (UTC)
I have pointed out repeatedly the context in which such material is inappropriate and relevant wording from elsewhere on Wiki that supports the fact. "Scientific theories promulgated outside [scientific] media are not properly verifiable as scientific theories and should not be represented as such." That is a fact. Here, there, and everywhere, whatever you tell me.
Perhaps you've seen our page, or have at least heard the phrase, Teach the controversy. Because what neo-creationists peddle is not actually science they cannot actually raise "dissent" in the academic sense. Rather, through various media (say a top-ten website) they want to repeat the claim that there's controversy over Evolution because repetition will lead lay people to believe it, even if there is no substance to the claim. There's controversy within the discipline, of course (e.g. Evolutionary psychology), but there's no controversy about the salience and veracity of Darwinism itself. This has clear parallels with conservative media reporting vis-a-vis Global warming.
Would you like to enable those efforts? By not prioritizing scholarly sources, that's exactly what you are doing. I'm not saying we should not report on the repugnant view—not at all. As I've said, Intelligent Design is a socio-religious phenomenon of some note and deserves a good, reliably sourced page. It's categorized as Neo-Creationism and Pseudoscience because that's what it is, and it need not be described on a page categorized in Biology. Marskell 09:21, 1 July 2007 (UTC)

Marskell, are you saying nonscience should not be characterized as science? SlimVirgin, are you saying Wikipedia does not censor nonscientific claims? Can you both agree with those two statements? WAS 4.250 09:51, 1 July 2007 (UTC)

I think we would both agree with those two claims. But I would further add that non-scientific sources should not be used to characterize scientific claims, which is the point of disagreement. Marskell 09:56, 1 July 2007 (UTC)
How about: "Non-scientific sources should not be used to characterize the science (scientific validity, scientific process, scientific value, etc) of scientific claims"? WAS 4.250 10:26, 1 July 2007 (UTC)
Yes, something like that, but Slim is objecting under the weird assumption it violates NPOV. Marskell 10:35, 1 July 2007 (UTC)
I have not suggested presenting non-scientific views as scientific views. I suggested: "Where there is disagreement between scholarly and non-scholarly sources, the views should be clearly attributed in the text: the scholarly view is presented as the scholarly view, and a strong and reliable dissenting view as a strong and reliable dissenting view." SlimVirgin (talk) 10:39, 1 July 2007 (UTC)
Updated thoughts on a second read. The first sentence of the last paragraph is a "yes, but" and I'd suggest two declaratives.
"Scholarly and peer-reviewed sources are highly valued and should usually be given most weight in areas such as history and science. Material from non-scholarly reliable sources may also be appropriate in these areas, particularly if they are mainstream sources with good reputations. The appropriateness of such sources will depend on the nature of the claims being made."
Note the bold is not the explicit wording I've advocated, but for the sake of policy peace it satisfies my point and leaves people latitude in applying the wording to different areas. This, coupled with the pseudoscience wording and other things, should be enough if (actually, when) I wind up on another time-wasting thread with a Ufologist.
I would then kill the rest as problematic. In the case you've cited Slim, of survivor stories, it may be highly appropriate to contrast scholarly and non-scholarly sources. But this may give people a license to try to challenge scholarly literature where it's not appropriate at all.
And again, I would remove "Wikipedia offers no definition of what constitutes a 'reliable source.'" It's contradictory to begin the paragraph that way. Marskell 13:23, 1 July 2007 (UTC)
I'm still confused as to how you think the best way of proceeding here is SlimVirgin. Do you want each editor to develop a different draft, or do you think it would be better for us to all edit the same draft? Tim Vickers 16:01, 1 July 2007 (UTC)
No, don't edit the initial draft SV posted or someone else's because it will confuse people coming to the thread for the first time. Edit your own and note the change made, if need be. Marskell 16:49, 1 July 2007 (UTC)

This thread is becoming rather long and difficult to follow, but I wanted to add in my $0.02 regarding "reliable scientific sources". Some above have made the assertion that newspapers and journalistic publications are "peer reviewed" in the same manner as academic journals... that is absolutely false. They may be fact checked (less so in OpEd pieces), but they are by no means peer reviewed. Academic journal articles are written by subject matter experts and reviewed by at least three other experts in the field, and who face the possibility of losing all credibility if the information contained therein is incorrect. Newspapers and other sources published by journalists may be fact checked, but fact-checkers are generally not experts in a given subject and journalists don't have nearly as much invested in the content. Academics spend years doing research and writing single articles -- journalists do not. It would be a severe disservice to Wikipedia if we give as much credibility to blurbs in USA Today as we do to truely peer-reviewed sources. /Blaxthos 17:00, 1 July 2007 (UTC)

Rather than Verifiability, this seems more of an issue related to good article criteria and featured article criteria. As alluded to in this discussion, Sparknotes may be reliable for a certain fact in the article Romeo and Juliet, but does it really represent a professional standard to cite to Sparknotes in such an article when T. J. B Spencer (ed.) The New Penguin Shakespeare "Romeo and Juliet" (Penguin, London, 1967) is reliable and may be cited for that same fact? -- Jreferee (Talk) 17:29, 1 July 2007 (UTC)

I am most concerned by the idea, repeated throughout this discussion, that we are librarians. We are not. We are editors and Wikipedia is not a library. If you want access to unsifted information, try Google. As editors we have to uphold NPOV for sure, but we mustn't disengage our critical faculties. Separating the wheat from the chaff, choosing reliable sources and using them fairly is our job. I don't think anyone here thinks the Daily Mail is a suitable source for science or medical articles. I fear the wording of WP:V will be weakened in an attempt to achieve harmony with NPOV. Complete harmony is not necessarily good. Some tension is fine. A balance can be found on a case-by-case basis by intelligent editors. Colin°Talk 19:16, 1 July 2007 (UTC)

Well the current draft proposal does say "Scholarly and peer-reviewed sources are highly valued and should usually to be given most weight in areas such as history and science," I think that covers this point as a general guidelime that is true in most cases, but without making an absolute statement that scholarly sources must be given more weight. Do you think this is a fair summary? Tim Vickers 19:58, 1 July 2007 (UTC)
I've reworded the "rule of thumb" from "As a rule of thumb, the more people engaged in checking facts, analyzing legal issues, and scrutinizing the writing, the more reliable the publication." to "As a rule of thumb, the more people engaged in checking facts, analyzing legal issues, and scrutinizing the writing of a particular article, the more reliable this article is as a source." to avoid the problem of op-eds in newspapers (where there are 100s of writers involved in the publication) being given more weight than peer-reviewed papers or books written by multiple authors, (with, for example, five authors, two editors and three external reviewers - fewer in total than in the newspaper, but more then the single op-ed writer if you consider just the particular article.) Tim Vickers 21:16, 1 July 2007 (UTC)
I've changed "article" to "work" since books are important in many humanities disciplines. semper fictilis 21:36, 1 July 2007 (UTC)
Excellent points. Thank you. That new organisation is indeed much clearer. Tim Vickers 21:40, 1 July 2007 (UTC)

I removed the special exclusion of non-academic sources and tweaked overall. In general, exceptional claims require exceptional sourcing, but the NYT, Wash Post, and WSJ all reporting something would constitute as good a source as some obscure little scientific journal. Crum375 17:03, 4 July 2007 (UTC)

...but not nearly as good a source as Nature, Science, or Proceedings of the Royal Society. I've added compromise wording that "the appropriateness of sources always depends on context" without specifically mentioning non-scholarly sources. We're slowly approaching a compromise between those who prefer scholarly sources on scholarly topics, and those who think scholarly and non-scholarly sources are equivalent even for scientific topics. Raymond Arritt 17:14, 4 July 2007 (UTC)
Can anybody think of an example of when the NYT, Wash Post, and WSJ would all make a statement on the topics of science, medicine or history that would be an "extraordinary claim", but not based on any academic publication? What might such a claim be? A real-world example would help me see this from your point of view. Tim Vickers 17:33, 4 July 2007 (UTC)
Just off the top of my head, both National Geographic and Canada's National Post lent credibility to this bit of patent nonsense [20][21]. Raymond Arritt 17:44, 4 July 2007 (UTC)
No, that's not an example, as the scientist involved has published their theory in at least one journal link. This is a very restricted statement - if an extraordinary claim has even just a single reference in academic publications then it can be included. "The appropriateness of non-academic sources always depends on context and they should not be used as the sole support for extraordinary claims." This provision is intended to apply only to magnetic motors, flat earth geography and anti-gravity machines. Tim Vickers 18:13, 4 July 2007 (UTC)
To raise the bar even higher, I've suggested "The appropriateness of sources always depends on context and non-academic sources should not be used as the sole support for bizarre and extraordinary claims." Saying that variations in solar output might change Earth's climate is not "bizarre and extraordinary", however, saying that an alignment of planets will knock Earth off its axis is a "bizarre and extraordinary" claim. As an example of a "bizarre and extraordinary claim" that should be covered is this Nature paper from 2005. Tim Vickers 18:22, 4 July 2007 (UTC)
I think we need to assume that the respectable and reputable mainstream press will stay away from reporting wild and unsubstantiated "bizarre and extraordinary claims", which are not supported by any academic sources. If you have an example where that was not the case, let's examine it. Otherwise, I see no reason to discriminate against any source that is reputable, respectable, vetted by multiple layers of fact checkers, etc. Crum375 18:34, 4 July 2007 (UTC)
There's fact checking, and there's fact checking. Newspapers do check that people aren't misquoted, their professional affiliations are correctly reported, and so on. I'm not aware that most newspapers have staff members with expertise in molecular biology, radiative transfer, and so on who would be able to fact-check scientific claims. If you know otherwise I'd be glad for pointers. Raymond Arritt 18:52, 4 July 2007 (UTC)
(ec) To clarify, I wasn't referring to the possibility that solar output can change Earth's climate (which is widely acknowledged), but to his dismissal of a significant greenhouse effect on Earth. (Even global warming skeptics don't deny the presence of a large natural greenhouse effect.) To wit, as noted in the National Post article: "Ascribing 'greenhouse' effect properties to the Earth's atmosphere is not scientifically substantiated," he maintains. "Heated greenhouse gases, which become lighter as a result of expansion, ascend to the atmosphere only to give the absorbed heat away." This is so impossibly in violation of the basic laws of physics that it's hard to even read it without wincing. But it may well sound logical to someone with no scientific training (a reporter, say?) who vaguely recalls that warm air rises. Raymond Arritt 18:44, 4 July 2007 (UTC)
Well, if you don't think reliable sources will publish "bizarre and extraordinary claims" that are not supported by any academic sources, then a source that published such claims is, by this argument, not a reliable source. Therefore saying that such sources should not be used reinforces and does not contradict the injunction to use reliable sources. However, if there are examples of genuinely encyclopedic claims that will be excluded by this wording, we need to hear them. Tim Vickers 19:36, 4 July 2007 (UTC)
I suspect that if Einstein had elected to first talk to a newspaper before publishing his theory of relativity, many would have considered his claims (math aside) to be 'bizarre and extraordinary'. But my point remains: why single out any type of source? We can just say, as we do elsewhere, that exceptional claims require exceptional sources. Sources need not be discriminated against – any reputable and respected source that has multiple independent layers of fact checking, is a good source. The more reputable and respected news outlets have science correspondents on board, and they all have access to outside experts. Some academic sources have been known to screw up. Let's just focus on the core criteria, which make a source a reliable source for us, not on any labels. Crum375 21:10, 4 July 2007 (UTC)
The policy is WP:Reliable Sources not WP:Sources. Once the question is what is reliable, we have no choice but to discriminate--or, rather, to give editors principles that they can use to discriminate. As for news outlets and their science correspondents, their authority is ultimately based on how well they understand the science that they are reading and the scientists they are interviewing. Since their authority is derivative from the scientist, the better source is the scientist. In many ways, of course, this a false dilemna. Many science articles will indeed be citing NYT, etc. It is only when the popular and professional diverge that the policy is likely to be invoked, and it is precisely then that it will be needed. semper fictilis 03:23, 5 July 2007 (UTC)
Yes, we do discriminate, but based on merits, not labels. That would mean we would give the larger publishers with more independent vetting layers, reputability, respectability, mainstream-ness, etc. more weight. We don't discriminate based on an automatic bias that news media are always trash, and academic papers are always gold. This is what NPOV is all about, and as long as we focus on the merits, we'd be just fine. Crum375 03:36, 5 July 2007 (UTC)
(ec)In an ideal world, you would be correct. But Wikipedia isn't an ideal world. In my experience, few will apply the careful distinctions you propose; people will merely say "look at the policy, it says we can use both news sites and journals." They won't interpret it as you do here, i.e., that the best mainstream media are preferable the worst academic journals. Raymond Arritt 03:54, 5 July 2007 (UTC)
This is indeed a false dilemma, because, as Crum375 pointed out earlier, mainstream, reliable sources do not publish "bizarre and extraordinary claims" that have no backing in academic sources. As a compromise wording, what about "The appropriateness of sources always depends on context and multiple, independent, and highly reliable sources are needed to support extraordinary claims." This focusses on reliability, as you suggest Crum375, and does not use the "labels" you object to in any way. This is also a good reflection of how editors actually work and the discussions that occur on talk pages. Tim Vickers 03:48, 5 July 2007 (UTC)
Well, no one believes news media are "always trash"; I don't get that impression from the last sentence in the draft. But I think putting non-academic sources on par with academic ones regardless of the situation at hand contradicts the first sentence of the paragraph. It suggests that editors are free to include extraordinary claims (original research) that may not have received scientific consensus, based *solely* on a newspaper article. "Some academic sources have been known to screw up"—true, they are not infallible, but newspapers regardless of notability tend to do it on a more regular basis because their own criteria for inclusion is far less strict, as both they and peer-reviewed journals have entirely different objectives. This simply helps us decide which scientific stories to take more seriously and which ones we should meet with caution and a healthy dose of skepticism. It doesn't mean that the baby should be thrown out with the bath water. It does mean that we can report the findings/conclusions with proper qualification, and wait until they have gained scientific consensus.
Tim, doesn't the draft define "a highly reliable" source in science essentially as an academic one? — Zerida 04:30, 5 July 2007 (UTC)

(outdent) As far as unusual claims go, I like simplicity: "Exceptional claims require exceptional sources." Then we let the editors hash it out, based on the context. If we force them to have multiple sources, they wouldn't be able to use an exclusive article from Nature about some bizarre new creature discovered at 10,000' below sea level. Or from the Washington Post, about some president who hired thugs to break into his rivals' office. We need flexibility, and a merit based system, not a lot of specific and complex rules. If we say that we insist on respectable, reputable, mainstream, large publishers, with multiple layers of independent fact checkers, we'd be on safe territory. Certainly no less so than by having an academic source. This is what NPOV tells us - we must entertain all significant view points, but we may discriminate based on their intrinsic quality, i.e. merit - not an external label. Crum375 04:44, 5 July 2007 (UTC)

The proposed version does not require multiple sources, nor would it exclude material from Nature or the Washington Post. As I see it, the question is what we do when sources clash. When this happens, editors have to make a choice: we either write that x says this and y says that, or we reject x in favor of y (or vice versa). How do we make that decision? This is what this rule helps us with. If a scientific finding says something that is mischaracterized by the popular press (as has been the case recently with the warming of Mars), it would be profoundly negligent for wikipedia to give the misreporting any attention, except perhaps to include that interpretation x was a misreportage. semper fictilis 12:55, 5 July 2007 (UTC)

"Bizarre and extraordinary claims"

To help focus the discussion, the policy now states:

"...non-academic sources should not be used as the sole support for bizarre or extraordinary claims."

What claims are we talking about? Please feel free to add examples to the list below.

Bizarre and extraordinary

  • Machines that break the laws of thermodynamics.
  • Alien lifeforms regularly visiting Earth in secrecy
  • Non-relativistic time-travel
  • Faith healing bringing people back from the dead.

Unusual, but not bizarre and extraordinary

  • Political corruption and dirty tricks
  • Criminal activity
  • Minority scientific views

Having a requirement that any of these really strange ideas need very solid sources to support them reinforces the NPOV policy in that it requires that for a bizarre idea (in medicine/science/history) to be included, it has to have reached the level of prominence of being mentioned in academic sources. If people can think of real-world examples where this exclude material that needs to be in Wikipedia, please list them here. Tim Vickers 16:29, 5 July 2007 (UTC)

I don't know if I like the wording. "The appropriateness of such sources always depends on context and non-academic sources should not be used as the sole support for significant or extraordinary claims when describing an academic subject." This would allow us to use newspapers for the President ordering a burglary, which is significant and extraordinary but not an academic subject (that's exactly where we should be giving newspapers priority) but not allow their use in the troublesome examples mentioned. Would you like "In biology, evolution is the change in the inherited traits of a population from generation to generation" changed to "In biology, evolution is the controversial idea that there is change in the inherited traits of a population from generation to generation" with "controversial idea" sourced to a non-academic article about the Kansas School Board? Back on the circle (though surprised I'm on it with you). Marskell 18:01, 5 July 2007 (UTC)
The proposed sentence states "..non-academic sources should not be used as the sole support for bizarre and extraordinary claims." I'm not sure where the word "significant" came from in your example. Such creationist/evolution discussion is covered by the NPOV policy, this proposal does not attempt to address such controversial issues. Instead it focuses on clear examples of absolutely off-the-wall additions so that it will be generally accepted. Tim Vickers 18:49, 5 July 2007 (UTC)
I think the second sentence in Marskell's post is actually another proposed sentence: ...non-academic sources should not be used as the sole support for significant or extraordinary claims when describing an academic subject. This makes sense to me. I would prefer using specific wording to the effect that using at least one academic source is required to substantiate claims on academic/scientific topics (claims that do not yet enjoy consensus in the academic community). The current version says "the highest quality sources should be used..."; this makes sense too, but is a little vague. If not specifying that one of the sources should be academic, then it should at least provide specific examples of the type of sources that can be used. And if none of the sources are scholarly, then I think multiple sources are needed. However, at the end, I still think that such claims should never be *solely* based on non-academic sources. — Zerida 20:59, 5 July 2007 (UTC)

Bizarre and extraordinary claims that would be unfairly excluded

  • List examples here

Slight reorganization

The organization of the first three sentences seems to me a bit disordered. Wouldn't something like this be better

All articles must adhere to Wikipedia's neutrality policy, which states that articles should fairly represent all significant viewpoints that have been published by a reliable source, and should do so in proportion to the prominence of each.

Articles should be referenced to reliable sources, that is, sources that have a reputation for fact-checking and accuracy. Wikipedia offers no exact definition of what constitutes a "reliable source", and editors should use their common sense. etc.

It seems to me that it's better to treat NPOV in one paragraph and what constitutes a reliable source in another. (If you agree, perhaps go ahead a change the draft version above.) semper fictilis 21:22, 1 July 2007 (UTC)


The word "scholarly" seems to me unhelpful in this context. To me, scholarly means sober, intelligent, and informed, and thus there is in my experience much that I regard as "scholarly" in works that this policy would describe as "non-scholarly", and much that is not in works that might get the "scholarly" label. Equally, to me, the idea that anyone would use a source that is "non-scholarly" is a bit of a shock--"non-scholarly" (to me) means unreliable. Now, those may be my own bug-bears and shared by no one else. But wouldn't it be better if the words "scholarly" and "non-scholarly" were replaced by (e.g.) "academic" and "non-academic"? Maybe one of the scientists can chip on "scientific" vs "popular", or such. semper fictilis 22:40, 1 July 2007 (UTC)

My OED (1972 ed) defines "Scholarly" as - Pertaining to or characterising a scholar, and a "scholar" as either a member of a university (synonym of student) or as a learned or erudite person. Under this definition (and the OED is the final word in everything! :) ) the suggestion that we should use "non-scholarly sources" is, under this definition a very strange idea, as these would be produced by somebody with no learning on the subject in question. I have therefore changed the draft to use "non-academic", as you suggest. Tim Vickers 23:25, 1 July 2007 (UTC)

UNDUE and definition of RS

I would like to take SlimVirgin's addition regarding UNDUE and add it here ... the inclusion of all majority and significant-minority views that have been published by a reliable source in accordance with WP:UNDUE; or to this version: ... and should do so in proportion to the prominence of each in accordance with WP:UNDUE. Since the opening statement mentions NPOV and the representation of different views, I think it makes sense to also include direct reference to the part of policy that discusses this point in greater detail.

I don't want to harp on the second point, but I am still concerned about the wording in However, Wikipedia offers no exact definition of what constitutes a "reliable source"... I'd like to suggest something like this:

While Wikipedia offers no strict definition of what constitutes a 'reliable source', in general, the most reliable sources are books published by universities and peer-reviewed journals; university-level textbooks; magazines and journals published by respected publishing houses; and mainstream newspapers. Editors should use sources that are appropriate for the topic and the claims made. Appropriateness is a matter of editorial judgement and consensus. As a rule of thumb, the more people engaged in checking facts, analyzing legal issues, and scrutinizing its evidence and argument of a particular work, the more reliable it is as a source.

I think it's appropriate to refer to WP:CON when in doubt about whether or not to include a particular source or its level of reliability. — Zerida 23:28, 1 July 2007 (UTC)

In academia, the reliability of sourcing goes something like this (rank order):
  1. Peer reviewed journals (University & Professional) ("Journal of Correctional Education")
  2. University-level textbooks and University Press books (which usually contain plenty of citations)
  3. Magazines and newspapers
  4. Non-University books (publishing houses and the like)
  5. Mainstream media
  6. Self-published media (the interwebs)
I would have been strung up from the rafters in school if I had used anything other than #1's and maybe a sparingly few #2's. I guess we're not setting our standards that high, but I would again strongly caution giving #5 and #6 nearly as much credibility as #1 and #2. /Blaxthos 23:49, 1 July 2007 (UTC)
Thanks, good ideas. As the proposal says, the uses of sources depends on context. For any of the thousand Pokèmon articles, #1 and #2 are not an option! I've worked in your suggestions, but feel we need to keep the injunction to use common sense, I think that's is one of the most important parts of the policy. Tim Vickers 00:23, 2 July 2007 (UTC)
I took out "common sense" from the suggestion I made because I was hoping to see more emphasis put on the use of reliable sources, and I thought it went without saying that editors would exercise common sense. Now that I think about it, I run into situations everyday where this is not the case at all, so you're absolutely right. Is it possible though to just take out the dreaded "However, ..."-sentence and focus on what the sources should be? That is, go from Articles should be referenced to reliable sources, that is, sources that have a reputation for neutrality and accuracy immediately to The sources should be appropriate for the topic and the claims made, and appropriateness is a matter of common sense, editorial judgement and consensus.?
Or perhaps combine them: Articles should be referenced to reliable sources, that is, sources that are neutral, accurate and appropriate for the topic and the claims made. Appropriateness is a matter of common sense, editorial judgement and consensus.? — Zerida 01:03, 2 July 2007 (UTC)
The "common sense" clause was an excellent suggestion from Crum375. I agree that all we need do is offer the limited definition we have. There is no need to note specifically how limited and general this definition is, people can read it for themselves. Tim Vickers 01:41, 2 July 2007 (UTC)

I'm a bit puzzled about the distinction between 3 (newspapers and magazines) and 5 (mainstream media). I'd also place 4 (books) higher than newspapers and magazines. But that's getting away from the point... the difficulty on choosing an order reflects how important the context is (appropriate for the topic). So I certainly support that idea. Newspapers, for example, are great for many topics and appalling for others. The draft looks good at the moment. Colin°Talk 08:03, 2 July 2007 (UTC)

There's no need to make a strict distinction of reliability like this. Sometimes newspapers are better than books (in fact, many books from university presses are highly unreliable). Furthermore, many of the lower quality journals in many areas are about as good as newspapers, sometimes worse. On the other hand, adding in a clause about common sense will lead to a complete crap-out. We don't need people labeling things as not sufficiently reliable based on their personal notion of what constitutes common sense. JoshuaZ 16:18, 2 July 2007 (UTC)

This sort of guideline on reliable sources belongs on the WP:RS guideline page. It would be useful to talk about reliability of sources in context of various topic areas, maybe have an FAQ there, and summary of policies on sources (spread across numerous pages - WP:NOR, WP:NPOV, WP:BLP, and others) as a place that we can refer users. --Aude (talk) 16:26, 2 July 2007 (UTC)
Not just common sense (which would indeed be too open to individual interpretation) but common sense and consensus, which allows flexibility enough to cover unusual cases, but carries the requirement that other views are listened to and POV warriors cannot claim that their "common sense" overrides other editors judgment. There is a delicate balance here. We do not want a policy that removes editorial flexibility and must be applied rigidly in all cases - that would be impossible to write and counterproductive. Instead this policy states the general rule and then provides guidance that will be true in most cases, but allows for occasional exceptions, based on consensus and common sense. Tim Vickers 17:02, 2 July 2007 (UTC)
Speaking of WP:RS, in formulating this policy, it would be helpful to decide what will happen of that guideline. I was surprised to find out from the discussions here that it is now largely discredited, though I continue to see it cited by editors (including myself) more often than V. However, as it stands, I don't find it very useful, particularly where it states: "Examples of reliable sources are well known university-level textbooks and major encyclopedias." I find this quite inadequate, but I fear that editors will continue to be misled into thinking that WP:RS by virtue of its name is the primary gauge with which to measure the level of reliability. Also, note that WP:NOR states: "In general, the most reliable sources are books, journals, magazines, and mainstream newspapers; published by university presses or known publishing houses." It goes without saying that all policy and guideline pages will need to agree with one another at the end. — Zerida 22:03, 2 July 2007 (UTC)
My view is that once consensus exists on the statement here, it should be copied to WP:RS and expanded to help give a little more guidance--perhaps include some well chosen examples, some danger signs, etc. But we should agree on the policy version here, first, and then allow the guideline version to follow. semper fictilis 00:38, 3 July 2007 (UTC)

Reading age

Not that I can or do deal with this myself, but it strikes me that the reading age of the prose is quite high. I guess there is a consensus forming amongst the educated, has anyone got a spare hoodie to test it against? I guess it is a general policy issue and it is for the Village Pump, but the principle should be to aim writing at the general public, not acedemia.

For the first sentence, I had a stab, trying to avoid some of the hard words.:

Wikipedia has a policy of neutrality which says that all writing should be fair to all reasonable viewpoints. In Wikipedia, reasonable means looking at where the information in the writing comes from and deciding how reliable the information is. Writing should favour better sources of information, which policy WP:UNDUE talks about further.

What did pop out as I did that is that article is the wrong unit of measure for assessing POV. I thought about concept but that failed the readability test: writing is a nice vague collective term. Spenny 07:42, 2 July 2007 (UTC)

Anyone who does not understand can ask others. The article is the right unit of measure for some aspects of this and the claim being sourced is the right measure for other aspects. WAS 4.250 09:35, 2 July 2007 (UTC)
Maybe, but then that stance begs the question of why create the policy page in the first place. Given that getting the right wording is painful, then take the pain of considering the audience too. I think also that if a more relaxed writing style was encouraged, then the problems of the detailed implications of particular turns of phrases might disappear as there would be the sense of getting the general point across above tight definition. It does go back to the core intent of there are no rules.
With regard to the UOM, POV is vaguer than an article, a single statement can be contentious, as can a sub-section, or the article; I simply suggest there is unhelpful precision in the definition contrary to the intent, good NPOV writing is more subtle than writing a balanced article. Spenny 10:02, 2 July 2007 (UTC)
I've simplified the first sentence, which wasn't particularly well-written and put your suggestion of "writing" in instead of "article". I also removed the statement that common sense should only be applied if the source is contested, saying this is seemed unnecessary to me, common sense and editorial judgment should always be applied. Tim Vickers 15:45, 2 July 2007 (UTC)

if<>iff (the concept of if and the concept of If and only if are not identical). WAS 4.250 16:44, 2 July 2007 (UTC)

Yes, that is very true. However if we say this approach should be applied in all cases, this is a broad and simple statement that still covers the situation if the sources are contested. I just think that is a simple and effective way of putting it. Tim Vickers 16:54, 2 July 2007 (UTC)

"Neutrality and accuracy"

I'm fine with just "accuracy", my old wording of requiring both "neutrality and accuracy" might be seen as excluding some sources such as newspapers with a definite political leaning, which still print factual material, just mixed with slanted editorials and however much I dislike it, FOX news is still arguably a reliable source in some contexts. I have also removed the second use of this phrase. Tim Vickers 19:01, 2 July 2007 (UTC)

Are we moving towards consensus?

Thanks for all the comments and edits, the draft is looking pretty good thanks to all your input. Are people generally happy with the current formulation or are there points of view that have not been addressed? Tim Vickers 16:25, 3 July 2007 (UTC)

  • I think we're getting there. There is something about the logical transition into the third paragraph that still strikes me a bit odd. And I think we need to find a way to include the sources that are best for issues of contemporary politics and culture: that we use x, y, and z for a government scandal and a, b, and c for popular culture. semper fictilis 23:56, 3 July 2007 (UTC)
How about something like "In areas that are prone to controversy, the use of sources that have reputations for neutrality is strongly recommended." Tim Vickers 00:13, 4 July 2007 (UTC)
That may not always be available. We may just need to assume that we'll get POV sources (which are acceptable), and we need to present them properly per NPOV and UNDUE. Crum375 00:30, 4 July 2007 (UTC)
A problem with that formulation is that for controversial topics, sources that one side may view as neutral are often rejected by the other side as representing crank views or institutional conspiracies. In many of the India-related articles there are frequent edit wars in which the work of Western academics, published by University publishing houses, are subject to attack as "biased" simply because they do not confirm the views of some editors. I do not agree with these criticisms, but my point is that "neutrality" is in the eye of the beholder. If you are convinced that invisible aliens walk among us, any source that does not confirm that or at least mention it as a plausible theory is clearly not a neutral source. Personally I wish there was a more clear bias in favor of certain types of publishing houses, specifically those affiliated with universities. When I have tried to make a case for using academic materials in some articles pertaining to religion, it has been an uphill battle. Buddhipriya 00:27, 4 July 2007 (UTC)
I think neutrality is probably best left to the WP:NPOV page.
— Preceding unsigned comment added by Semperf (talkcontribs) 00:34, July 4, 2007 (UTC)
I agree. Crum375 01:19, 4 July 2007 (UTC)
On reflection, I agree as well. Tim Vickers 02:44, 4 July 2007 (UTC)
To Buddhipriya, we do give more weight to respected publishing houses and media outlets with more independent layers of scrutiny and fact checking. In general, that should favor the mainstream press and peer reviewed academic publications. Crum375 01:35, 4 July 2007 (UTC)
Hmmm Tim, with your most recent wording re. neutrality, it almost sounds like we're asking editors to maintain neutrality under special circumstances, but of course it's assumed that it will be maintained at all times. To me, the most basic definition of neutrality for our purposes here is to present two sides of an argument within the limits of logic, reason, and well, reliable sources (per Crum, I think).
The current draft has at least my support. It opens with the important mention of NPOV/UNDUE, moves on to describe what the policy means by reliable sources and what kind, and finally pointing out that peer-reviewed and academic publications are usually the most reliable sources in history and science. I hope that most of us are at least in agreement with the essence of this proposal, yes? — Zerida 02:31, 4 July 2007 (UTC)
This proposal seems to be being greeted with general approval, mainly I suspect because it tries to describe how normal editing practice operates, rather than dictating any new rules that are not already followed. Tim Vickers 04:22, 4 July 2007 (UTC)
The one way in which it doesn't reflect what really goes on is with regards to the web. Truth be told, my guess is that at least half (and maybe more) of the referencing that is done on wikipedia is to websites. I think we should add something about that, with a link to WP:EL. semper fictilis 19:51, 4 July 2007 (UTC)

Verifiability for images?

I've been wondering about how verifiability for images is established. For example, how do we know that Image:Nc3.jpg is an image of Elijah Shaw and Naomi Campbell, as the uploader claims? And that they're performing the activity that the uploader claims they are? The description of the image ends up being used in articles, too, as a caption. Not to pick on this example specifically -- there are hundreds of images with no real verifiability. Some are inconsequential; some could be mis-representative or defamatory. Are images at all verifiable? -- Mikeblas 22:50, 2 July 2007 (UTC)

WP:OR specifically note that images enjoy exceptions to the usual enforcement of no original research. A similar philosophy applies here. WilyD 23:02, 2 July 2007 (UTC)
Thanks for the link! Is it just me, or is that incredibly dangerous? -- Mikeblas 23:13, 2 July 2007 (UTC)
I see your point, but I'm not sure its that dangerous. Pragmatically, it would be difficult if not impossible to obtain "free-but-citeable" images (whether photographic, diagrams, maps, etc) for Wikipedia. From a "but is it correct?" standpoint, images have to go through the same community scrutiny - "bad" images don't tend to last long. -- MarcoTolo 23:52, 2 July 2007 (UTC)
Problem is, we don't know what images are bad. Look at the image at the Prostitution page, for example. Is that really a prostitute? Is she really from Germany? Or is it just a picture of an exgirlfriend that somebody's vengefully mad at? This seems like a big liability for Wikipedia, and that's a big danger to the project. It also seems very dangerous to allow images that "seem to be correct", without verifiability, to be attached to a site that has a lot of credibility. After all, very many people blindly trust Wikipedia, despite the disclaimers. -- Mikeblas 01:19, 3 July 2007 (UTC)
We are at the point now with images where we were with text in the early years of wikipedia when people just wrote what they thought was true without feeling the need to prove it with a cite. We'll get there with images sooner or later. This is all a process. WAS 4.250 14:35, 3 July 2007 (UTC)
Is there a way to make the process for images less painful and more expeditious than the process was for prose? -- Mikeblas 19:08, 5 July 2007 (UTC)

Just looking in

Hi, the discussion is too much for me to wade through without being frightfully interested in it, but I've scanned this page briefly, and I think the atmosphere seems friendlier than it did when I protected the policy page. How near are you all to reaching a consensus? Any signs of a compromise version being accepted by most editors? Forgive my laziness in not reading though the whole discussion in order to answer my own question. ElinorD (talk) 17:00, 3 July 2007 (UTC)

If you are asking, is it safe to unprotect the page, I'd say yes. The warring is over, it appears that consensus is forming. I don't get the impression that people are going to go awol if you unlock. Spenny 17:29, 3 July 2007 (UTC)
I'm agreeable to the draft above under "feel free to edit." Perhaps we should unprotect, add, and then reprotect, to allow people to express opinion on the addition without rv'ing over it? Marskell 14:50, 4 July 2007 (UTC)
As nobody has objected to the proposal above, it might be better to apply WP:BRD and just add the consensus wording to an unprotected page. If anybody reverts, they can explain their objections on this page and we can see if they can be incorporated. Tim Vickers 16:44, 4 July 2007 (UTC)
Yes. Let's unprotect; then add the draft above under "feel free to edit" to the policy (after the first paragraph in the "sources" section). WAS 4.250 08:09, 5 July 2007 (UTC)

Proposed changes to self-published sources section

I've been looking at discussions above regarding self-published sources, and trying to come up with a proposal that will hopefully not be too painful for anyone. What I've come up with is replacing the current subsection entitled "Self-published sources (online and paper)" with the following text:

Self-published sources (online and paper)

Anyone can create a website or pay to have a book published, then claim to be an expert in a certain field. For that reason, self-published books, personal websites, and blogs are not usually acceptable as sources.

Self-published material may, in some circumstances, be acceptable when produced by an established expert on the topic of the article whose work in the relevant field has previously been published by reliable third-party publications. However, caution should be exercised when using such sources: if the information in question is really worth reporting, someone else is likely to have done so, and information in self-published sources has not been subject to independent fact checking and may therefore be less reliable than professionally-published work by the same author. For this reason, it is usually necessary to provide a prose attribution for material sourced to self-published works. Self-published works that make controversial claims are not usually considered reliable sources: if a self-published source is contradicted by more recent professional sources, it should not normally be used.

Self-published sources should never be used as third-party sources about living persons, even if the author is a well-known professional researcher or writer; see WP:BLP.

Self-published sources in articles on popular culture

Some articles about popular culture rely on self-published posts on bulletin boards, blogs, and Usenet, because few other sources exist for them. In such cases, the material used must have been posted by named or well-known pseudonymous individuals with verifiable expertise in the area, although the individual need not be a professional in a relevant field. Anonymous posts should never be used. Some areas of popular culture also rely on self-published secondary sources, such as well-known fansites. If in doubt about how to use a source in this area, consult the relevant WikiProjects for advice. This provision does not apply to material about living persons, or to topics in other areas, such as history, science, religion, literature, or current affairs.

When self-published sources are used in such an article to provide a subjective opinion on the subject (particularly for assessments of quality), the source should always be explicitly stated in the text of the article.

I've put the above together from previous proposals that haven't been summarily rejected and from comments to those proposals that seemed to be generally agreed upon. The specific changes to previous proposals are:

  • Unified the types of acceptable self-published source proposal with the pop culture proposal
  • Incorporated Marskell's suggestions for changes to the acceptable sources paragraph that followed that
  • Removed unnecessary and confusing commas in first paragraph of pop culture section
  • Changed "individuals with well-known expertise" to "individuals with verifiable expertise", with the implication that there must be a reliable source that shows the individual is qualified to speak on the topic (e.g. a reliable review of the site, verifiable credentials of being involved with a production, etc...). See section #Verifying the individual above.
  • Addition of last paragraph, with the intention of addressing Marskell's concern about fansites being overly enthusiastic about the quality of their topic -- change the problem from making an inaccurate assessment to one of "why should I care that they think this?", which will hopefully lead to the result that such useless statements will not be added.

Suggestions I haven't included:

  • Any more detailed requirements on the credentials of sources than that they are "verifiable". I think this single word is enough.
  • A definition of what exactly pop culture is, because nobody has provided one.
  • Marskell's wording that limits the use of such sources to particular topics within articles: I'm hoping that he will accept the requirement for a prose attribution for subjective judgements as a less limiting way of addressing the same concern (I can certainly think of occasions where using quality judgements made by fans might be useful, so I don't want to rule it out entirely).
  • I haven't include any of the proposals in the #Scholarly and non-scholarly sources section, because that seems to be moving towards consensus independently, so it would probably be a bad idea to lump it together with another independent proposal.

Any comments? JulesH 13:03, 4 July 2007 (UTC)

JulesH, this is great work but my biggest issue remains that anybody can claim to be a biographer, get published by a reputable publishing house, and still the book can be full of garbage and without any footnotes or false footnotes. As I pointed out in this article "The Factchecking Facts", fact-checking is undependable. So what happens here is some editor inevitably will use this author as a source for an article here and because there is no requirement (at least that I can find that is obvious), that author can be used as a verifiable source. This issue may speak more to WP:CITING and/or WP:VERIFY but it certainly should be clarified here. What are your thoughts? Jtpaladin 14:04, 4 July 2007 (UTC)
This has hopefully been addressed by SlimVirgin's "reliability rule of thumb" in the scholarly and non-scholarly proposal - As a rule of thumb, the more people engaged in checking facts, analyzing legal issues, and scrutinizing the evidence and arguments of a particular work, the more reliable it is. If a source has no fact-checking, then it isn't at all reliable. Tim Vickers 14:53, 4 July 2007 (UTC)
I'm a little uncomfortable with "the more people..." For a science article I'd rather have sources vetted by three peer reviewers and two technical editors than 100 newspaper editors who satisfied their university science requirement with the easiest rocks-for-jocks course they could find. But I doubt any requirement for disciplinary competence would be allowed. Raymond Arritt 14:59, 4 July 2007 (UTC)
Tim, I have to agree with Raymond. Although SlimVirgin alludes to a better fact-checking, it's not a fast and hard rule. You know how things go around here. If they aren't set in stone, people will weasel their way around them. I also addressed this very issue here. In that section called, "Say where you got it", it seems to state that an intermediate source like an author who is writing about a topic that requires a footnote to another author or source requires that the WP editor be able to know the quality of the primary source, i.e. Author A writes about something Author B says so WP editor needs to verify Author B's information. This is the only real way of providing first quality sources because we simply can not depend on the fact-checking of Author A's publisher. I have seen situations where one author will quote a second author who then quotes a third author and when you get to the primary source, it turns out to be nothing more than a rumor. If we want to create solid encyclopedia articles, we have to be able to cut through the intermediate sources and go directly to the originating source. I believe that this is what WP:CITING in the "Say where you got it" section is saying and we should back that up in this WP rule as well. Also, as WP:CITING also states, if there are likely to be challenges to a particular quote (and we all know how common this is), an inline quote is required. That is an excellent rule that I would like to see implemented more often. In the meantime, we need to back up these Citing rules in the Verifiability guideline as well. Thoughts? Jtpaladin 20:32, 4 July 2007 (UTC)

I disagree with the proposed re-write. Policy pages cannot, by design, rule on every little aspect of sources. It needs to provide sufficient information about the spirit or principles upon which the policy rests, and nothing more. The current, long standing formulation does exactly that. ≈ jossi ≈ (talk) 19:52, 4 July 2007 (UTC)

I'm not sure I can agree with that. The current phrasing of this section has been the subject of a lot of contention, and my experience of editing a variety of articles suggests that there are a lot of editors out there who don't understand it. The problems that need to be addressed in the currently existing section are that "researcher" is the wrong word (it carries connotations that aren't intended: most people understand it as meaning "academic researcher"), and people are concerned that self-published sources may be used to inappropriately contradict mainstream knowledge. There is no principle in this policy that suggests they shouldn't at the moment.
As for the pop culture exception, again, there is no principle currently in the policy that would allow this kind of use.
What, specifically, about this change do you object to? JulesH 20:10, 4 July 2007 (UTC)
  • Question: Is there a precise definition of "popular culture"? In my experience, editors will find any loophole they can in the policies and guidelines to justify adding crap to articles. Just about any subject could be argued to be "pop culture". I think that in an effort to be more clear, we may be opening up a pandoras box. There have been instances in the past where I would have liked to be able to use self-published sources, such as showing online usage of neologisms, but I have learned to live without them, and have grown to see that it isn't worth opening up a hole in the sourcing rules just for the benefit of a few articles. The overspray just isn't worth it in the long run. If something is truly notably and encyclopedic, there will be reliable secondary sources that can be used eventually, you just might have to wait a while, but what's the hurry? Wikipedia isn't going anywhere soon. - Crockspot 20:29, 4 July 2007 (UTC)
I agree. The pop culture "exception" is not needed in policy. This is an encyclopedia, after all. ≈ jossi ≈ (talk) 05:42, 5 July 2007 (UTC)
Crockspot is spot-on about popular culture. This loophole will surely come back to bite us. Popular culture will be harder to write about because of the high standard we set but that seems to be a better outcome than leaving the loop hole. Of course the issue of self published works transcends popular culture. Somehow, a judgment call is going to have to be made about which self-pubs are reliable and which are not. JodyB talk 12:36, 5 July 2007 (UTC)
I concur... The desision of when to allow an exception to the limits on self-published sources should be made at the individual article level, and not in Wikipedia-wide policy. We all understand that there are occational exceptions to every rule... there is no need to write such exceptions into the rules. If it is vital to cite a self-published source in a specific article, the editors can always invoke WP:IAR. Blueboar 12:45, 5 July 2007 (UTC)
relying on IAR is relying on quicksand. Editors will do anything they think will work, and there will not be a basis for resolution. We need guidelines as a starting point, or we can not have discussions based on policy. But we need them flexible, so we can reach appropriate conclusions. The point is to avoid having every use of such references end up at AfD. DGG 05:01, 6 July 2007 (UTC)


Okay, I've unlocked the article, I hope you can all work it out without edit warring again. Remember, one bold edit, one reversion, and then lots of discussion to consensus, and then the next bold edit, the next reversion, the next discussion and the next consensus, and then the next...

Good luck. Steve block Talk 09:02, 5 July 2007 (UTC)

Verifiability in lists

User: Until(1 == 2) started decimating several lists, such as List of ethnic group names used as insults, which are annotated lists of wikipedia articles. His position is: "please provide citations to addition to the article, Wikipedia cannot be a reference to itself". I find it inapplicable for the cases of lists. I asked for Third Opinion (pointing to Talk:List of ethnic slurs by ethnicity#WP:V), but no one hurries to join the discussion.

It occurs to me that User: Until(1 == 2) will not be the last one with this attitude, and I even don't want to think about th emoment when someone will start chopping lists like List of Romanian actors demanding provide reverences (in the list) taht they are Romanians and are actors. Therefore I think that the phrasing of the policy must have more common sense as to where it is reasonable to request citations (now it says "in the article"). `'Miikka 18:58, 5 July 2007 (UTC)

I have no problem with demanding citations in lists... I know of quite a few where doing so is a requirment for adding a new entry. In the case of your specific list, I could see the requirement for citations as cutting out potential neologisms. Blueboar 19:05, 5 July 2007 (UTC)
Yes it is a requirement, if it is a list of entries that do not have their own wikipedia article. If there is an article, all references must be there, where a NPOV picture of the term is to be presented. If an article is a "potential neologism", then the article is a subject of AfD and then delisting. If there is an alive and well wikipedia article, it is a valid entry for the list. `'Miikka 20:12, 5 July 2007 (UTC)
When did Wikipedia become an acceptable source for list entries? Each article/list must support itself wrt references. Lists are not exempt from WP:V, nor are they a special case. Colin°Talk 22:10, 5 July 2007 (UTC)
Then why don't you go and put List of Romanian actors for deletion? It does not contain a SINGLE reference! Wikipedia has always been a source of reference for its lists. Lists are lists are lists. They are navigation/summary tools. Next thing you start requiring references for categories and navboxes. `'Miikka 23:15, 5 July 2007 (UTC)
This has been a contentious issue for some time. When I last checked, the consensus was that citations must be included if the material is challenged, per this policy, but as you say there are many lists that have no references at all. SlimVirgin (talk)(contribs) 01:29, 6 July 2007 (UTC)
Miika, you seem to be the only one supporting your position. I looked through Wikipedia:Verifiability and could not find anything that said Lists are the single exception to the core content policy Verifiability. Can you propose an argument on why it would be in the best interest of Wikipedia not to require references on lists? P.S. I have seen categories with {{unreferenced}} on them. Jeepday (talk) 01:50, 6 July 2007 (UTC)
Miikka, I agree completely: such lists should be deleted en masse, and the sooner, the better. Lacking references guarantees that we are misinforming the public at least some of the time, which is completely unacceptable.Proabivouac 01:57, 6 July 2007 (UTC)
Agree, AfD all these. ≈ jossi ≈ (talk) 02:40, 6 July 2007 (UTC)
Agreed. SlimVirgin (talk)(contribs) 02:41, 6 July 2007 (UTC)
  • Articles don't tend to fail AfD just for being unreferenced, and the editor who brings such suggestions is generally chastised for it, Jeepday (talk) 02:49, 6 July 2007 (UTC)
That is because much of the community lacks a sense of scholarly responsibility appropriate to the gravity of our mission.Proabivouac 03:27, 6 July 2007 (UTC)

OK I am totally confused now. Are you all suggesting to delete lists like List of Romanian actors, List of Roman laws, List of pornography industry personalities, List of Awards presented by the Governor General of Canada, and several thou others whic don't have a single reference? Because they are LISTS OF WIKIPEDIA ARTICLES (some with summaries, for convenience), why would they need references? The verifiability is not violated: references are INSIDE articles. If there are no references INSIDE listed articles, then it is a problem of the article, not of the list. `'Miikka 03:34, 6 July 2007 (UTC)

Wikipedia articles aren't reliable sources, for exactly the shortcomings we're discussing. If and when the day comes when all Wikipedia articles (and lists) are well-referenced and well-reviewed, then we can cite them.Proabivouac 04:18, 6 July 2007 (UTC)
Yes Mikkalai the suggestion is to delete the unreferenced lists, as that would be keeping with policy. If you attempt to create NEW ARTICLE you will notice this statement in bold print "Articles that do not cite reliable published sources are likely to be deleted." immediately above the edit box. The reality is that if if the article (or list) appears to be attempting to cite reference and is moving towards improvement then it is generally not considered for deletion but is an article in progress. Take for example List of rail trails which is partially referenced, which has references and is improving (slowly). Another reality is that articles with content are not generally deleted for lacking references, but everything unreferenced in the article may be removed as unreferenced per WP:V leaving a blank or nearly blank article which is then subject to {{db-nocontext}} or {{db-empty}} then it is gone. Jeepday (talk) 13:05, 6 July 2007 (UTC)
The reason citations need to be in the list article is that when you just link to an article it can look like there is a citation when there is not. See Talk:List_of_ethnic_slurs_by_ethnicity#Finding_sources_for_removed_content, it looked like all the entries were cited through wikilinks, but when I actually looked through it I found over half were not cited. Just as the policy says now, all challenged facts in any type of article needs to be cited in the article. This is good. Until(1 == 2) 13:38, 6 July 2007 (UTC)
Oh, and you did get a response to your third opinion request here. Until(1 == 2) 13:42, 6 July 2007 (UTC)
Here is another example of what happens when you audit a page that is "referenced" with wikilinks: Talk:List_of_Romanian_actors#Work_log. Over 2 thirds are not cited at all in their linked articles. Wikilinks used in place of references gives the false impression of citation that is more often than not absent from the article. The reason for this is Wikipedia is not a reliable source. Until(1 == 2) 18:39, 7 July 2007 (UTC)

List guideline

I point out Wikipedia:List guideline, pointed at by Category:Lists. It looks like it will need updating. (SEWilco 04:10, 6 July 2007 (UTC))

How does it need updating? Per Wikipedia:List_guideline#Lists_content The verifiability policy states that "articles should contain only material that has been published by reputable sources. Editors should therefore provide references." The responsibility for providing a citation rests "with the editors who have made an edit or wish an edit to remain." Inclusion on the list should be based on what reliable sources say, not on what the editor interprets the source to be saying. In the case of edits lacking citations, according to Wikipedia:Verifiability: Signed Jeepday (talk) 12:48, 6 July 2007 (UTC)

Non-scholarly sources, again

I see scholarly and non-scholarly sources is back, after I thought we had a consensus to not use it. This construction makes me very uneasy, and I oppose including it. I am sure it will be used to support the inclusion of nonsensical conspiracy theories, pseudoscientific nonsense, and fringe opinion. I wish I understood what those who want to include it hope the positive result will be. Tom Harrison Talk 01:16, 6 July 2007 (UTC)

Indeed, in the discussion above, nobody supported this wording. Tim Vickers 01:21, 6 July 2007 (UTC)
The wording says that scholarly sources are likely to be the sources of choice in areas such as history and science, and that non-scholarly reliable sources are welcome too, particularly if mainstream with a reputation for accuracy. There is nothing in there that implies the proliferation of conspiracy theories or nonsense. The bottom line is that nothing in this policy may contradict the NPOV policy, which stresses that all majority and significant-minority published opinion must be included. SlimVirgin (talk)(contribs) 01:26, 6 July 2007 (UTC)
We know an opinion is significant because we find it in a reliable source. We do not on our own determine an opinion is significant and then find the best source we can. Tom Harrison Talk 01:30, 6 July 2007 (UTC)
Sorry, Tom, I don't follow your point. SlimVirgin (talk)(contribs) 01:38, 6 July 2007 (UTC)
The wording implies that the point of view comes first, and the best sources available are then found to support it. Instead of this, we should be only be including a point of view that exists in reliable sources. Being in a reliable source is how we know a point of view exists. Tom Harrison Talk 02:13, 6 July 2007 (UTC)
The wording "Academic and peer-reviewed sources are highly valued and should usually be given most weight in areas such as history, medicine, and science." is entirely consistent with the NPOV policy as this wording does not state that non-academic points of view cannot be included. Tim Vickers 01:34, 6 July 2007 (UTC)
Nothing in NPOV requires us to encourage people to use non-scholarly sources. It would be better to say nothing than to say 'scholarly or non-scholarly'. Tom Harrison Talk 01:34, 6 July 2007 (UTC)
Of course it does. All maj and sig-min published opinon must be included. ALL. Not only scholarly. SlimVirgin (talk)(contribs) 01:37, 6 July 2007 (UTC)
Indeed, all sig-min opinion can be included. This wording does not preclude that. This is not a statement about content, it is a statement about relative weighting that should be given, in most cases, to different types of sources. Tim Vickers 01:40, 6 July 2007 (UTC)
It's not at all necessary to say 'non-scholarly.' If I didn't have so much respect for your work I would think you were deliberately misconstruing what I say. I know the down side as well as you. We both deal with cranks every day. What good edit are we loosing by not saying 'scholarly or non-scholarly'? I'm happy to listen to your argument however you choose to make it. Tom Harrison Talk 01:45, 6 July 2007 (UTC)
The "non-scholarly" clause, whatever its goal, amounts to a green light for cranks; if it stays, we will never hear the end of it.Proabivouac 01:51, 6 July 2007 (UTC)
Tom, I'm trying to balance two things: (a) to establish wording that will help to keep out crank theories, fringe theories, and conspiracy theories; and (b) to make sure that the wording isn't misused to keep out non-scholarly ideas that are not crank, fringe, or conspiracy theories. I've seen both happen, and both are violations of NPOV. Therefore, I want to make clear that non-scholarly and scholarly sources are welcome, though the latter are likely to be the sources of choice in subjects such as history and science. SlimVirgin (talk)(contribs) 01:52, 6 July 2007 (UTC)
Surely we can work out on the talk page a way to express that without tipping the balance too far. I thought we had. I'll do no more tonight. Tom Harrison Talk 01:56, 6 July 2007 (UTC)
I'm sure we can. The current wording looks fine to me. SlimVirgin (talk)(contribs) 01:58, 6 July 2007 (UTC)
This non-consensual edit is flaccid and unclear - "likely to be the sources of choice". What does this phrase mean? This policy does not aim to describe likely attributes of people's writing, it gives specific guidance on how to select and assess sources for reliability. Tim Vickers 02:02, 6 July 2007 (UTC)
The current version seems to have consensus. Non-scholarly has been removed, which is not my preference, but I can live with the current wording so long as it's not watered down any further. SlimVirgin (talk)(contribs) 02:17, 6 July 2007 (UTC)
Could you explain exactly what you mean by "likely to be the sources of choice" is this just a statement about probability? If so, why is it needed in the policy? Tim Vickers 02:38, 6 July 2007 (UTC)


A recent edit added "per NPOV" after the words "material from non-scholarly reliable sources is also welcome in these areas." WP:NPOV says nothing about a blanket requirement to "welcome" non-scholarly sources. Indeed, it advises us to use "best and most reputable sources you can", which in scientific topics will often be scholarly sources. Raymond Arritt 02:04, 6 July 2007 (UTC)

The point of NPOV is not to exclude sources simply because editors have personal views about the issues. So we say we welcome other sources, to emphasize that per NPOV we don't exclude any significant view that is well sourced. Crum375 02:06, 6 July 2007 (UTC)
Your second sentence does not logically follow from your first. Suppressing the personal views of editors does not require us to use non-scholarly sources on scientific topics. Raymond Arritt 02:12, 6 July 2007 (UTC)
It does, if those sources are majority or significant minority views. SlimVirgin (talk)(contribs) 02:17, 6 July 2007 (UTC)
So, on a topic such as genetically modified organisms we're to "welcome" the word of a reporter from the Daily Telegraph or The Sun exactly as we would a plant pathologist? Two uninformed reporters beat one scientist, becoming the "majority view"? Raymond Arritt 02:28, 6 July 2007 (UTC)
Majority in reliable sources I assume. Take a hypothetical example of where there is disagreement on the molecular weight of hydrogen. Ten Op-ed writers state in their columns that this is 1.008(7) g·mol−1, while two peer-reviewed papers in Nature state that it is 1.00794(7) g·mol−1. Should the majority view in non-academic sources be given more weight here than the majority view in academic sources? Tim Vickers 02:25, 6 July 2007 (UTC)
Tim, let's not use nonsensical examples. We know that the mainstream respected sources will have multiple fact checkers, and access to technical experts. Let's see where there is a real example where such a mainstream source insists on contradicting well established scientific facts, and then we can deal with it. Crum375 02:30, 6 July 2007 (UTC)
And by the way, we already say that sources must be appropriate to the topic - so for example a respected scientific database would trump a popular journal in accuracy, if that's the issue. Crum375 02:33, 6 July 2007 (UTC)
It is indeed true that if you assume that all mainstream sources never make mistakes, then they are all obviously of equal weight. However, that isn't true, so we can't make this statement. Tim Vickers 02:36, 6 July 2007 (UTC)

Disputed paragraph

Proposal, removing "most weight" and making a simple statement about what is usually the case in terms of accuracy. Tim Vickers 02:10, 6 July 2007 (UTC)

It's already removed. SlimVirgin (talk)(contribs) 02:17, 6 July 2007 (UTC)

Academic and peer-reviewed publications are highly valued and usually the most reliable sources in areas such as history, medicine, and science. Material from reliable non-academic sources can also be used in these areas, if they are respected mainstream sources with good reputations for accuracy. The appropriateness of sources always depends on context and only the highest quality sources are acceptable as support for extraordinary claims.

No, the writing is ambiguous. SlimVirgin (talk)(contribs) 02:17, 6 July 2007 (UTC)

Here is the current wording, which I think is a reasonable compromise:

Academic and peer-reviewed sources are highly valued and are likely to be the sources of choice in areas such as history and science, but material from other reliable sources is also welcome in these areas, per NPOV, particularly if they are mainstream sources with good reputations. The appropriateness of the sources will depend on the context, and is a matter of editorial judgment. Where there is disagreement between sources, the views should be clearly attributed in the text. Only the highest quality sources are acceptable as support for extraordinary claims.

I think we should let it stay for a while, and gather more comments. Crum375 02:14, 6 July 2007 (UTC)

Why not roll it back to the version before the contentious edits began? That would be here. The present wording clearly is disputed. Raymond Arritt 02:18, 6 July 2007 (UTC)
Which part of it do you dispute, Raymond? SlimVirgin (talk)(contribs) 02:20, 6 July 2007 (UTC)
The discussion had settled down to an apparent consensus before the page was unprotected. The point was to work this out on the talk page yesterday, not to see who could shoe-horn a version onto the page and then demand consensus to change it. With an approach like that we would risk losing the generous benefit of the doubt we should be able to accord each other. Tom Harrison Talk 02:26, 6 July 2007 (UTC)
No need for comments like that, Tom. SlimVirgin was very busy with editing other pages and may not have been following the discussion here. She might therefore have been unaware of the level of support that the text had when she made the bold edits we are discussing. Tim Vickers 02:33, 6 July 2007 (UTC)
See Tom's response below. Raymond Arritt 02:30, 6 July 2007 (UTC)
No, I meant which part of the current text do you dispute, Raymond. SlimVirgin (talk)(contribs) 02:31, 6 July 2007 (UTC)
(edit conflict-o-rama) Several things, but the unfounded reference to WP:NPOV is especially galling. Raymond Arritt 02:38, 6 July 2007 (UTC)
It has been discussed endlessly. At some point it has to be added to the article, the writing has to be good, there shouldn't be repetition, and it must be consistent with the other core policies. We now have that, and Tom's removal of "non-scholarly" while retaining the rest of the edit seems to have met with approval. SlimVirgin (talk)(contribs) 02:28, 6 July 2007 (UTC)
Except for 'particularly', which should be 'only', or should be taken out. Tom Harrison Talk 02:36, 6 July 2007 (UTC)
"Particularly" is important. The meaning of the sentence changes without it. SlimVirgin (talk)(contribs) 02:43, 6 July 2007 (UTC)
Certainly it does change the meaning. The problem with 'particularly' is it implies 'but even if not'. Why would we want sources that are not mainstream, or do not have a good reputation? Tom Harrison Talk 02:45, 6 July 2007 (UTC)
Sources don't have to be mainstream. They need only be reliable, and must represent majority or significant-minority POVs. Trying to introduce that they must also be "mainstream" would be a major policy change. SlimVirgin (talk)(contribs) 02:49, 6 July 2007 (UTC)
I can live with this if we remove the terrible phrase "likely to be the sources of choice", the single word "preferred" is much better. Tim Vickers 02:41, 6 July 2007 (UTC)
It's not a terrible phrase. It's more accurate than "preferred," because we can't say that WP prefers them without knowing the context, but we can say that editors are likely to choose them. SlimVirgin (talk)(contribs) 02:42, 6 July 2007 (UTC)
You must have experience with a vastly different pool of editors than I do! My experience is that people are going to use whatever sources they can get away with to push their POV. We're constantly having to deal with sources like this. Raymond Arritt 02:45, 6 July 2007 (UTC)
Is that a 'mainstream source with a good reputation'? Crum375 02:50, 6 July 2007 (UTC)
It doesn't have to be. It would just be particularly welcome if it were. Tom Harrison Talk 02:52, 6 July 2007 (UTC)
But that's not a reliable source, Raymond. You don't need to add special words to keep that kind of thing out. SlimVirgin (talk)(contribs) 02:58, 6 July 2007 (UTC)
Really? You don't think so? I've had to go through long arguments with people who do. Granted it's not a "mainstream source" but the present wording doesn't say it has to be -- only that it's "particularly" welcome if it is. Raymond Arritt 03:20, 6 July 2007 (UTC)
The current version (and every version before it for a long time) says sources must be reliable and appropriate, and that source is neither. I'm getting the impression you're trying to change the policy because you've not been emphasizing the current wording enough in arguments with sources. The thing to do is to rely more on what we already have, rather than trying to strengthen it, because the danger is you'll go too far in the opposite direction, which will lead to a weakening of the policy over time, because of backlash. I've seen it happen many times. The key is to strike a balance, then leave it stable. If you have problems with people trying to use poor sources and claiming to rely on this policy, let me know and I'll help you out. SlimVirgin (talk)(contribs) 03:42, 6 July 2007 (UTC)

Academic and peer-reviewed sources are highly valued and preferred in areas such as medicine, history and science, but material from other reliable sources is also welcome in these areas, if they are mainstream sources with good reputations. The appropriateness of the sources will depend on the context, and is a matter of editorial judgment. Where there is disagreement between sources, the views should be clearly attributed in the text. Only the highest quality sources are acceptable as support for extraordinary claims.

I suggest we let the current version sit, as it seems to have agreement (though it may not be the version each one of us would choose first), and if it causes the problems Tom thinks it might (letting in conspiracy theories etc), bring it back to the table with real examples that show the problem. Then we can tweak the wording further. SlimVirgin (talk)(contribs) 02:46, 6 July 2007 (UTC)

This is absolutely stupifying. Has it always been like this here and I am just now noticing? Tom Harrison Talk 02:49, 6 July 2007 (UTC)
I have removed that paragraph that is disputed. Revert warring on policies is unacceptable. Tim Vickers 02:50, 6 July 2007 (UTC)
Then don't do it yourself. SlimVirgin (talk)(contribs) 02:52, 6 July 2007 (UTC)
"Treason doth never prosper..." Tom Harrison Talk 02:54, 6 July 2007 (UTC)
I think it is sensible. Indeed, some of us were under the impression that there was consensus on previous version discussed above. — Zerida 02:57, 6 July 2007 (UTC)
I think we all agree that edit warring is not a good idea, so I hope removing the disputed paragraph will stop any need for re-protection of the page. Tim Vickers 02:58, 6 July 2007 (UTC)
Too late it seems. Tim Vickers 03:01, 6 July 2007 (UTC)
It's sad how this is degenerating into yet another petty edit war after thorough discussion over days, particularly by users who did not take part in it at all. — Zerida 03:02, 6 July 2007 (UTC)
I took part in the discussion and expressed my view many times, but we can't have discussion-terrorism, where people are expected to say the same thing a thousand times or else be told they're not taking part. Just because something is decided on this page by a very small number of editors doesn't mean it has consensus, particularly if the regular page editors object. SlimVirgin (talk)(contribs) 03:05, 6 July 2007 (UTC)
Removing the paragraph was a rather absurd response; clearly the paragraph is needed, though you appear to have some sort of unclear issue with the exact wording. To entirely remove it would obviously leave the policy in far worse shape. Jayjg (talk) 03:10, 6 July 2007 (UTC)
Removing the paragraph was a rather absurd response Exactly. — Zerida 03:13, 6 July 2007 (UTC)
Since this section consisted only of a single sentence last time the page was protected Link, moving up to two uncontroversial paragraphs with this latest protection could be seen by an optimist as reasonable progress. Tim Vickers 03:19, 6 July 2007 (UTC)
I'm sure if we all contribute to the discussion it will eventually produce something that everybody can be happy with. Your major concern seems to be a NPOV concern, SV, unfortunately that wasn't discussed at all in the section above where we produced what appeared at the time to be a consensus wording. Is this the only issue you have with the paragraph we produced? Tim Vickers 03:15, 6 July 2007 (UTC)
It's still not clear what your objection to the current paragraph is. Jayjg (talk) 03:17, 6 July 2007 (UTC)
I can live with the box above if "but" is removed from the first sentence and we have two declaratives. (And Tom is exactly right about "particularly".) Marskell 10:29, 6 July 2007 (UTC)

Few notes: 1) this seem more an issue of reliability, not verifiability 2) I am opposed to the fragment "in areas such as medicine, history and science"; i.e. the sentence should read "Academic and peer-reviewed sources are highly valued and preferred in all areas". We also need to define what are the "the highest quality sources".-- Piotr Konieczny aka Prokonsul Piotrus | talk  12:41, 6 July 2007 (UTC)

Drafts of disputed paragraph

Draft 1

Academic and peer-reviewed publications are highly valued and usually the most reliable sources in areas such as history, medicine, and science. Material from reliable non-academic sources may also be used in these areas if they are respected mainstream sources with good reputations for accuracy. The appropriateness of sources always depends on context, and only the highest quality sources are acceptable as support for extraordinary claims.

Draft 2

Academic and peer-reviewed sources are highly valued and are likely to be the sources of choice in areas such as history and science, but material from other reliable sources is also welcome in these areas, per NPOV, particularly if they are mainstream sources with good reputations. The appropriateness of the sources will depend on the context, and is a matter of editorial judgment. Where there is disagreement between sources, the views should be clearly attributed in the text. Only the highest quality sources are acceptable as support for extraordinary claims.

Draft 3

Academic and peer-reviewed sources are highly valued and preferred in areas such as medicine, history and science, but material from other reliable sources is also welcome in these areas, if they are mainstream sources with good reputations for accuracy. The appropriateness of the sources will depend on the context, and is a matter of editorial judgment. Where there is disagreement between sources the views should be clearly attributed in the text. Only the highest quality sources are acceptable as support for extraordinary claims.

Discussion of drafts

I would be happy with either draft 1 or 3. - I feel draft 2 suffers from two severe shortcomings:

  • Saying that an attribute of articles is "likely to be" the case has no bearing on policy. This policy does not describe probabilities about articles, it gives guidance on reliable sources.
  • Saying that sources are welcome "particularly" if they have good reputations, must mean that sources may be welcome if they do not have good reputations. This contradicts the earlier paragraph that states "Articles should be referenced to reliable published sources that have a reputation for accuracy." Tim Vickers 03:28, 6 July 2007 (UTC)

Draft #1 is better than #3 in my opinion; draft #2 is much less helpful than either. -- MarcoTolo 03:40, 6 July 2007 (UTC)

Draft #1 or #3 in this order for me as well. Furthermore, no explanation has been given for removing this sentence: "Appropriateness is determined by editorial judgment and common sense, guided by consensus." — Zerida 03:43, 6 July 2007 (UTC)
I agree with Zerida. WAS 4.250 16:41, 7 July 2007 (UTC)
The first sentences of drafts 2 and 3 are clearly strawmen and don't belong if for no other reason than that. Marskell 10:21, 6 July 2007 (UTC)
Yes, draft 1 is preferable. Draft 2 is plainly unacceptable. Draft 3 is not quite as bad as draft 2. Raymond Arritt 14:01, 6 July 2007 (UTC)

The wording "but material from other reliable sources is also welcome in these areas" in draft #3 is problematic. Adding "per NPOV, particularly" is even more problematic, as Tim says. Saying "can also be used" is much better wording, and the rest of the paragraph is also good in draft #1. So, I suggest we use draft #1. --Aude (talk) 11:11, 6 July 2007 (UTC)

I think there is some talking at cross-purposes here. Even in academic topics it can be appropriate to discuss the non-academic view. Public perception, for example, is valid and is properly described in non-academic sources even in scientific articles. That draft 1 misses out the important qualification of editorial judgement is a fatal flaw in that version for this reason. Context is all, and context is not at the high level the article which is the implication of these drafts as I read them, it is finer grained than that.
In history, it may be necessary to refer to popularly held misconceptions where there is absolute consensus in the academic world, the best citation for that may be a quality newspaper article. To pick an example, 1421: The Year China Discovered the World there is academic agreement on history in this context, yet the book exists, and there is a need to present the hypothesis and the public reception (check the article as it stands, and there is no citation for the sales, and no citation for the synthesised suggestion that the theory has popular public support due to the sales - but that is for another section). Clearly this is history, yet to describe the controversy well we probably need non-academic sources, even though we may rightly present the academic conclusion that it is all tosh with some force with those citations. I'm confused as to where to go because of this. I understand the perspective of describing scientific and historical concepts clearly and concisely, but an encyclopaedia is not a text book, and I think that makes a difference to this text.
If we test 1421 against the above draft paragraphs, yes, academic sources are required for Wiki to assert that the theory is rubbish, but non-academic sources are entirely appropriate for use in giving the context of the article. In trying to introduce History and Science qualifications we undermine the need for context, not of the subject area but of each individual claim being made. I do understand the point that is being made, but I don't think the wording makes the point clearly (not that I have a solution as yet). We haven't got the right words to cope with perspective. Spenny 11:16, 6 July 2007 (UTC)
That is why the first proposal does not state that academic sources are the only ones that should be used, or that they must be given more weight in all situations. Instead, draft 1 states that they are usually more reliable. Hence, if you do need to cite a newspaper for context and discussion there is nothing in draft 1 that would prevent that. However, if you wanted to cite a newspaper source that contradicted a peer-reviewed paper, although the newspaper opinion could still be included, the peer-reviewed paper would usually be given more weight due to it being more likely to be accurate. Tim Vickers 18:42, 6 July 2007 (UTC)
I think I could get nearly comfortable with that, I think the niggly pedantic point is the selection of weight by area rather than, erm, argument, which the wording doesn't have the nuance to cover. Still, all in all, very pedantic and if I sit back a bit, the basic flavour is there. Spenny 18:57, 6 July 2007 (UTC)
As I'm trying to assess people's opinions here, Ian, which of the three drafts are you most happy with and which least happy with? Tim Vickers 19:01, 6 July 2007 (UTC)
Uncharacteristic decisive moment: #3, but changing the first but the full stop as in #1. Spenny 19:50, 6 July 2007 (UTC)
Value judgements as to which of two or more sources meeting WP:RS is better for the article should be based on Good article criteria and Featured article criteria. Using WP:V to as a basis for deciding such value judgments is instruction creep. -- Jreferee (Talk) 16:52, 7 July 2007 (UTC)
At present the criteria for GA and FA point right back here (and to the "deprecated" WP:RS). Raymond Arritt 17:03, 7 July 2007 (UTC)
Unfortunately, it's a self-defeating circle all around. For a discussion of reliable sources, this page points to WP:NPOV and WP:RS, WP:RS points here and to WP:NPOV, and WP:NPOV points to WP:RS. Yet, the only page so far that actually discusses sources in more detail is WP:NOR, which itself points to WP:RS and WP:V as being the Main articles on the use of reliable sources! At the end, an effective remedy will entail working step-by-step on each relevant policy and guideline to keep them in line with one another. — Zerida 18:28, 7 July 2007 (UTC)

I prefer draft 1, mainly because the word "welcome" will cause more trouble than the neutral "can also be used". I accept some of the points made above about the usefulness of non-academic sources, particularly when discussing popular beliefs concerning science and history. In the latter two drafts, the "is a matter of editorial judgement" clause is redundant IMO. The disagreement => "clearly attributed in the text" sentence sounds reasonable at first but isn't suitable for policy. I can think of circumstances were it would be tedious to have to name the sides. Colin°Talk 17:19, 7 July 2007 (UTC)

I would support draft 1, and oppose both 2 and 3, simply because 1 is not 'weasel-worded'; it says what it needs to say, clearly, and leaves little room to debate the intent or meaning of the paragraph, only its implementation (as is proper). SamBC 17:39, 7 July 2007 (UTC)

Draft 1 is my preference, draft 3 is reasonable but not ideal. Draft 2 is unacceptable (imho), since it essentially says that public opinion is as valid (or more - depending on prevalence of articles) as a scientific assessment. While that may be fine on subjects outside of science and history. Science is not about public opinion. On science subjects an article in Nature has more weight than an article in the Washington Post - despite both being reliable sources. While on policy subjects they may be much the same (or even the reverse). --Kim D. Petersen 17:57, 7 July 2007 (UTC)

I did some slight grammar fixing in all three proposals without changing content or context. I don't particularly care for the wordiness and diction used in any of them, and the sentence structures in #2 & #3 are particularly bad (no offense to anyone intended). I also don't know how good of an idea it is to use subjective/nonspecific terms like "highest quality" in such an important wiki policy. I know we intentionally avoid specifically saying what exactly constitutes a reliable source, however I can see all of these proposals being (mis)used or misrepresented, or at the very least being interpreted a variety of different ways and providing traction for more argument later. Be that as it may, #1 is definitely the best choice of the three presented. /Blaxthos 19:12, 7 July 2007 (UTC)
Fundamental predicament: The (apparent) consensus behind Draft #1 seems justifiable, but there is a fundamental predicament here that Zerida and User:IanMSpencer already touched upon previously. The problem is, how do you make express allowance in WP policy for sources that are "non-mainstream" and "non-academic" and yet arguably appropriate given the context and circumstances.[2]
Although I think most of us here can agree that such circumstances should be narrowly construed and considered extremely rare ... the fact is, a good-faith reading of policy suggests that such circumstances are indeed accepted. (See e.g., WP:NPOV#Undue_weight "Minority views can receive attention on pages specifically devoted to them").
Even the (entirely sensible) importance given to medicine, history and science is not without its problems (example provided in footnote).[3]
Suggestion: Draft #1 seems appropriate, with one caveat: I'd propose changing the final sentence from this:
   The appropriateness of sources always depends on context, 
   and only the highest quality sources are acceptable as support for extraordinary claims.
to this (or something similar, but with better wording):
   The appropriateness of sources always depends on context, 
   but sources claiming to support extraordinary claims are subject to strict scrutiny before 
   they may be considered acceptable for inclusion.
Note: I myself don't particularly like this wording, but internal consistency across core WP policy seems to call for recognition of this (undesirable, and rare, but yet permissible) scenario. dr.ef.tymac 02:04, 8 July 2007 (UTC)
  1. ^ Tiny-minority views and fringe theories need not be included, except in articles devoted to them.
  2. ^ Note the term "arguably" ... a good-faith reading of WP:NPOV suggests that "opinion" and "debate" (if well-referenced and correctly handled) may be appropriate in some articles. If we are to accept this, then we have to admit that there will be questions about which sources are appropriate for inclusion to begin with. It is therefore unlikely that all sources will be "unambiguously" appropriate for contentious topics.
  3. ^ There are some respected physicians, for example, who lend credibility to the Vaccine controversy debate. There are certainly physician expert witnesses who could be hired to provide admissible testimony that there are some unanswered questions regarding the risks and complications of vaccination. Would a record of such testimony be suitable as substantiation in a "minority view" WP article? Perhaps, but none of the draft proposals here seem to suggest a "yes" answer is possible, or that close judgment calls such as this are recognized.

NPOV concern

An e-mail by Jimbo cited as part of the NOR policy that discusses a question in Physics might be helpful in this discussion:

What do mainstream physics texts say on the matter? What do the majority of prominent physicists say on the matter? Is there significant debate one way or the other within the mainstream scientific community on this point?

If your viewpoint is in the majority, then it should be easy to substantiate it with reference to commonly accepted reference texts.

If your viewpoint is held by a significant scientific minority, then it should be easy to name prominent adherents, and the article should certainly address the controversy without taking sides. Link

Note that Jimbo states that the assessment of this new theory rests entirely on the opinions of the scientific community - the "mainstream scientific community" and "significant scientific minority", not a just "significant minority", but the majority of scientists - the people who are qualified to judge the theory. If the argument that stating that scientific sources are generally most reliable violates NPOV is true, then it follows that giving more weight to scientific opinion must certainly violate NPOV. This cannot be the case, as this is the very action that Jimbo recommends here. Therefore, stating that scientific sources are "generally most reliable" does not violate NPOV. Tim Vickers 17:20, 6 July 2007 (UTC)
I do not think that anyone is arguing that "giving more weight to scientific opinion must violates NPOV". Scientific opinion is scientific opinion, there are significant opinions that may not be scientific, and may be scientific and non-scientific opinions that are not significant. We report on the formers but avoid the latter. ≈ jossi ≈ (talk) 22:32, 6 July 2007 (UTC)
That does not seem to be the impression that I have been getting from the couple of editors who oppose that academic/scholarly sources rightly be given more weight in topics of scientific/scholarly nature. It has been pointed out unconvincingly, and repeatedly edit-warred over, sometimes by editors who did not at all participate in any one of the discussions, despite what is so far undeniable, overwhelming consensus on this talk page. It appears now that this is also backed up even by Jimbo. Of course, the fact that this has been making one of the most fundamentally important policies of Wikipedia so unstable is not helping, not to mention the process with which Wikipedia is meant to operate is rendered essentially useless. — Zerida 02:16, 7 July 2007 (UTC)

Wikipedia:Reliable sources/Noticeboard

For a time this noticeboard was moved to Wikipedia:Verifiability/Noticeboard, without any explanation and despite the facts that 1) it is and was linked from RS, not V and 2) it was intended to discuss reliability of particular sources, not verifiability. I don't see the need to discuss whether particular sources can or cannot be verified, as in my experience this matter has almost never arised, unlike the issues of reliability of individual sources, which come up often enough. If a coherent argument is presented, we may perhaps consider renaming the board to Wikipedia:Reliable sources and verifiability/Noticeboard as a consensus, but first the need for Verifiability-dedicated noticeboard needs to be argued for.-- Piotr Konieczny aka Prokonsul Piotrus | talk  12:36, 6 July 2007 (UTC)

I am speculating, but the confusion may have arisen from the fact that WP:RS is only a guideline, while WP:V is the controling Policy... The page may have been moved so as to highlight this. Blueboar 12:49, 6 July 2007 (UTC)
Well, the noticeboard discusses reliability, not verifiability, so I found the renaming confusing. It is my understanding that verifiability means that any fact added needs to be easily confirmed as indeed originating from source X, which should also be reliable - but it is WP:RS which goes into details of what is reliable and what is not. The noticeboard is not designed to question WP:V (which, being a policy, is mostly unquestionable anyway, and rightly so) but is intended for discussion of whether a specific source (book, article, webpage, etc.) is reliable or not.-- Piotr Konieczny aka Prokonsul Piotrus | talk  13:17, 6 July 2007 (UTC)
I'm not disagreeing ... although, if you look at the discussions above, the question of what is reliable or not seems to be a major issue here as well. That's the problem with having so many different policies and guidelines dealing with overlaping subject matter. Blueboar 13:47, 6 July 2007 (UTC)

Blanking unsourced articles and tagging them for speedy deletion as empty?

Is it appropriate to blank unsourced articles and then tag them for speedy deletion per Wikipedia:Criteria for speedy deletion#A3? User:Until(1 == 2) seems to think that the Wikipedia:Verifiability gives him the right to do so, and I disagree. I don't think this policy allows that at all. Can we get a third opinion? Picaroon (Talk) 00:29, 7 July 2007 (UTC)

This is based on WP:V "The burden of evidence lies with the editor who adds or restores material. Material that is challenged or likely to be challenged needs a reliable source, which should be cited in the article"
Simply put, if an article has no verifiable content, it has no content, worse than no content it has only potentially wrong content. A blank article is better than a wrong one, especially when it is about living people, blank articles are speedyable. This would not be something one does willy nilly, but the prerogative to remove uncited info is always there. Of course the decision to apply CSD#A3 would be the admins decision, but it would be inappropriate to restore the uncited content once challenged. The alternatives to deletion per CSD#A3 would be to leave the article blank or find citable information. Until(1 == 2) 00:33, 7 July 2007 (UTC)
I am wondering how other people's interpretation of policy meshes with mine? Until(1 == 2) 00:42, 7 July 2007 (UTC)
Properly mark the unreferenced sections long enough for people to fix the article. Particularly in case there is a mistake; I recently had to remove a citation request because the citation already existed in the article. (SEWilco 01:25, 7 July 2007 (UTC))
I was speaking mostly of article with 0 references, and nothing even looking like a reference. But I agree fully an attempt should be made on the talk page of the article to solicit sources and warn of your plans to challenge unsourced info. Until(1 == 2) 01:44, 7 July 2007 (UTC)
It's not a one-two punch, though; you have to give people time to come up with sources, unless it's a BLP violation. This is why admins should always check the page history before speedy deletion. Consecutive blanking/tagging is not kosher in my opinion. -- nae'blis 02:35, 7 July 2007 (UTC)
User Until(1 == 2) is technically correct, but keep in mind three thoughts; Wikipedia:Assume good faith, Wikipedia:There is no deadline and Wikipedia:The Most Important Thing Possible. Jeepday (talk) 02:57, 7 July 2007 (UTC)
I have long thought that we should have a statement along the lines of: "Editors should have sources for their articles at hand before they start writing" in one of our policies or guidelines... it would save a lot of hastle. However, the fact is that people will write articles, create lists, etc., without sources. I feel that standard opperating procedure for dealing with such articles and lists should be... 1) Tag the article, section or sentence that needs sources and raise the issue on the talk page. 2) Wait a while to give editors a chance to respond... to either locate and include sources or at least indicate that they are trying to do so (at minimum I wait at least a week... usually longer). 3) Raise the issue on the talk page again... stating that if no sources can be found soon you will start to delete. 4) wait a little while longer again. and finally... 5) cut the unsourced material... leaving the article as a bare stub in case some future editor can create a properly referenced article... or, if the topic is not really notable, nominate at XfD. Blueboar 13:09, 7 July 2007 (UTC)
The problem with this is, it's a good idea, but the practical reality is that WP would be a fraction of its current size if this approach were formally instituted and enforced. I've participated in numerous articles where I was the only contributor to add any citations or references, even after repeated requests for others to do so as well. It is not unreasonable to assume that most contributors simply do not like adding references, even when it is not particularly difficult to find substantiation for the content itself. This seems to be especially true for articles on topics that appeal to people without training in formal research methodology, nor credentials in a specific area.
The bottom line is, if you really care about improving WP, and you're not just aiming your cross-hairs at content you personally disagree with, you should have considered not blanking an article until you've first made a good-faith effort to add some good citations yourself. If the topic is beyond the scope of your expertise or research skills, tag it as unreferenced and give someone else an opportunity to substantiate it. Unless the content is *clearly inappropriate* (e.g., fraud, BLP violation, slander, attack, advertising, vanity, copyvio, etc.) then you may want to check motives for wanting to remove, instead of improve. dr.ef.tymac 15:15, 7 July 2007 (UTC)
Just to toss out a bit of argumentum ad Jimbonem, we have from the text of WP:V itself: "I can NOT emphasize this enough. There seems to be a terrible bias among some editors that some sort of random speculative 'I heard it somewhere' pseudo information is to be tagged with a 'needs a cite' tag. Wrong. It should be removed, aggressively, unless it can be sourced." Your last sentence implies you "may want to check the motives" of Jimbo Wales! Raymond Arritt 17:55, 7 July 2007 (UTC)
Why did you have to cite Jimbo Wales to make that point? You could have cited the very sentence you seem to be criticizing! It says: "Unless the content is *clearly inappropriate*" ... random speculative crap definitely meets the definition of *clearly inappropriate*. Relevant and unbiased content that can be substantiated from reliable sources; if only someone bothered to try doing so, however, does not. If you can find a cite from Jimbo that disputes this very basic point, I'd love to see it. My bet is Jimbo, and many others, agree. dr.ef.tymac 23:10, 7 July 2007 (UTC)
If the ultimate goal is to make a better and more reliable encyclopedia, then I have no problem with WP being (temporarily) a fraction of its current size. Sometimes you have to take a step back to move two steps forward. But I do agree with the statement that, before rushing to delete, you should see if you can fix the problem by searching for and adding citations yourself. Blueboar 15:44, 7 July 2007 (UTC)
Until(1 == 2) is disrupting Wikipedia to prove a point. --Kaypoh 15:46, 7 July 2007 (UTC)
Please provide diffs when you toss around accusations. I am not proving a point, or disrupting Wikipedia. I am removing unsourced dubious claims. Until(1 == 2) 18:30, 7 July 2007 (UTC)
  • Several times in the past, there have been proposals for speedy criteria allowing the deletion of articles (or particular categories of articles) if they lacked references. These have always failed to achieve consensus, generally by wide margins. The general thought has been that it is better to allow unsourced content to be added in the hope and expectation that future sources will be added, either by the original contributor or by another editor. I think that this is still a sound idea (BLP articles are now an exception, at least as far as negative or controversial content goes, but even there uncontroversial things like birth years are often added without a source). While it is technically permitted to remove unsourced statements, in general i think it is bad form to do so unless either a) the editor actually doubts the accuracy of the statements, or b) there has been a request for sources and significant time has gone by and none have been provided. That is why we have {{unreferenced}}, {{refimprove}}, {{fact}}, and {{who}}. IMO removing significant unsourced content (about which there is no serious question of accuracy) and then promptly tagging for speedy deletion is disruptive, and an attempt to evade the lack of consensus for speedy deletion of articles lacking sources. I would remove any such speedy tags that I encountered. DES (talk) 16:07, 7 July 2007 (UTC)
Listen, I am not just walking up and blanking articles then CSD'ng them right away. I am just saying it is within policy to do so if time to repair it is given before removal, and more time after removal and before CSD. I never said that it all happened in 5 minutes. My general tactic is to mark a page with {{unreferenced}}(or {{fact}} if it is not the whole article that is uncited) and post a talk page message saying I will be removing unsourced data, even if all data is unsourced. Then after a few days I will removed unsourced content, even if all data is unsourced. I will post on the talk page yet again that without content this page will most likely be deleted, I include a link to the unsourced version and give ample time for citations to be found. But if nobody even tries to cite the information then nobody is bringing the article up to our standards of verifiability and my removal is entirely justified, even if it means removing everything. If time passes and nobody fixes it then yes, CSD for having little or no content applies. But in my experience it does not go as far as CSD, people generally fix it when you enforce WP:V.
I entirely agree that the article being repaired is the ideal outcome, and I give plenty of opportunity for that to happen. I have yet to have an article deleted due to this tactic, only repaired. Often I find myself doing the job that WP:V says belongs to the person seeking to include the information, and I gather the citations because the person seeking to include the information stubbornly returns it without citation and won't bother to look one up(even if it is a wikilink away). Until(1 == 2) 18:33, 7 July 2007 (UTC)
See List_of_Romanian_actors Before and After, List of ethnic slurs by ethnicity Before and After. I would call those positive outcomes. It is about being academically responsible, I am just trying to keep Wikipedia honest to its own policies. Until(1 == 2) 18:45, 7 July 2007 (UTC)
User Until(1==2) good work referencing those lists. When ever you are looking for something to do here is a project Wikipedia:Unreferenced articles. Jeepday (talk) 19:02, 7 July 2007 (UTC)
I have a list as long as my arm, mostly of lists that have 0 references. I hope future articles yield less resistance from the community to cleanup than the first two did(see their talk pages). Until(1 == 2) 19:04, 7 July 2007 (UTC)
I have found that it is the nature of some Wikipedian's to attack clean up attempts see Wikipedia talk:Unreferenced articles, Maybe that is why it is so hard to keep volunteers on projects. The bigger the mess you are working on, the more you find that really needs to be deleted, and the more you get attacked. Jeepday (talk) 00:18, 8 July 2007 (UTC)
This is true, I have been made most unwelcome in my cleanup attempts. I have been called "lazy" by one admin and a "troll" by another. Until(1 == 2) 00:25, 8 July 2007 (UTC)
Waiting "a few days" is too quick for people who aren't active daily, such as those who edit on weekends or are on vacation for a week or two. (SEWilco 00:39, 8 July 2007 (UTC))
They delay is a courtesy. If a consensus forms as to how long someone should wait before removing uncited information fine, but until then a few days seems reasonable to me. As for the CSD, I have yet to reach that point, but I imagine it would be after a longer wait and a basic attempt by me to find some citations. These are lists that have been unsourced for years. Until(1 == 2) 01:17, 8 July 2007 (UTC)
There is no such a thing. You place a {{unreferenced}}, add a date parameter and let it be for a couple of months. Better still, do some research and find some sources. If no sources are forthcoming, place in WP:AFD. Blanking and adding a speedy tag is not appropriate. ≈ jossi ≈ (talk) 01:32, 8 July 2007 (UTC)
Well perhaps you should propose a change to this policy so it better reflects your opinion. Until(1 == 2) 02:22, 8 July 2007 (UTC)
That is no my opinion. That is the current process as established in Wikipedia. ≈ jossi ≈ (talk) 03:34, 8 July 2007 (UTC)
Why doesn't policy describe this? Until(1 == 2) 04:12, 8 July 2007 (UTC)

Verifiable information

I am being told by an admin, User:Jossi, that he can re-introduce challenged facts without providing a citation as long as the facts are "verifiable", as in can be verified. I moved the info to the talk page at Talk:List_of_Romanian_actors#Work_log for others to find citations for them. But he insists on keeping the uncited information in the article itself saying "Moving it to talk, simply removes any possibility from the occasional editor to come and add a source". My opinion is that moving it to talk prevents readers from reading potentially false information.

he says to me "You are misinterpreting WP:V". Is this true? Why doesn't the policy say this? Where have I gone wrong in interpreting this policy? Are uncited facts really allowed to stay even after being challenged because they are "verifiable"? I want to follow the rules here, but what are the rules? Until(1 == 2) 04:18, 8 July 2007 (UTC)

First, I am not a she, I am a he. I have invited you to contribute to the project by finding sources rather than deleting material because there are no sources. It took me less than 20 minutes to find sources for the most of the entries in List of Romanian actors that you deleted. We are here to build an encyclopedia, and we do that by collaborating: one editor adds material, another editor adds more, and another finds a source. If the material is not sourced, you can add a {{fact}} tag, and wait. If no source is forthcoming, you may delete, if you wish, or better do some basic research if you are inclined and find a source. ≈ jossi ≈ (talk) 04:44, 8 July 2007 (UTC)

First, I didn't delete anything, I moved it to the talk page where things that need to meet inclusion standards belong. Secondly I found the first 19 references so don't imply that I never did any research. Third, removing unsourced information is an important contribution that stops Wikipedia from turning into a gossip mill. I would appreciate it if you did not diminish my contributions. And finally, {{fact}} tags are not a requirement before removing unsourced data. I gave plenty of notice on the talk page before during and after. Everything I removed I re-posted on the talk page to be addressed. I am not just tossing stuff out. But now we are just repeating ourselves, I am interested in others opinions on this matter. Until(1 == 2) 04:47, 8 July 2007 (UTC)

You say "I am interested in others opinions on this matter." My opinion is that User:Jossi is 100% right on this. Read all his comments in this thread (section). WAS 4.250 12:19, 8 July 2007 (UTC)
On the contrary. I do appreciate your contributions. Moving content to talk is deleting content. The application of policy needs to be done in context. An unsourced controversial claim in a biography of a living person can be deleted on sight, for example. For non-contentious material such as a list of actors, we give some more leeway. If an entry is not sourced, hopefully someone will add it at a later point. Wikipedia articles are work in progress and there are tens of thousands articles without sources. You may want to join one of the Wikipedia projects to help with finding and adding sources to such articles. ≈ jossi ≈ (talk) 04:52, 8 July 2007 (UTC)
You may want to read WP:CITE#Dealing_with_citation_problems where this is explained. ≈ jossi ≈ (talk) 04:57, 8 July 2007 (UTC)
The key here is whether the information is harmful to the article (as outlined in WP:CITE)... since listing someone as a Rumanian Actor is hardly harmful (to the article or the person), the correct approach is to add a citation request (a {{fact}} tag) and leave it in the article. If, after a reasonable period (I would suggest at least a month), a citation has not been provided, then you can delete it or move it to talk. Blueboar 11:33, 8 July 2007 (UTC)
If I look up Romanian actors, and add some to my report that are not Romanian actors because WP put uncited information back in(after another editor challenged it veracity), that is harmful. Unreliable information is a detriment to an article. This is basic academic responsibility, show your work. Until(1 == 2) 13:48, 8 July 2007 (UTC)

Unsourced data

Sourcing is new to wikipedia. Articles written for wikipedia in the early years were typically not sourced. There remain many thousands of excellent article in wikipedia from those years that remain to be properly sourced. Do not delete any data from wikipedia that you believe to be both true and not harmful regardless of whether or not it is sourced. Sourcing is an improvement and only a requrement if someone honestly believes it is either false or harmful (per WP:BLP or WP:LIBEL for example). The reqirement is that it is capable of being sourced. It only fails this requirement if people actually look online and in libraries for a source and can't find one. It does not fail this requirement just because a question on a talk page goes unanswered. WAS 4.250 12:34, 8 July 2007 (UTC)

All I can do is follow policy. I think any incorrect information is harmful to the article. I would appreciate it if this vague advice I am being given is explained in more clear terms in policy. Because it seems that there is a set of rules, but when I follow them I am told there is another unwritten set of rules. Perhaps an addition to the policy would clear this up. I for one am confused. I have no reason to believe that the entries on the list really are actors, really are Romanian, and really have a significants that justifies an entry in an encyclopedia. How can I if all this is true without a citation? How do I know we are not just passing on false data that some guy added one day after school? Why does it say I can challenge uncited facts I am unsure about when it seems I get in trouble when I do? I am not trashing the place, I am citing what I can and moving what is not to the talk page. The article is for the readers, the talk page is for the editors. I don't see why editors should keep notes about "what might be citable" in the article space. Until(1 == 2) 13:45, 8 July 2007 (UTC)