Wikipedia talk:Attribution/Archive 12

Archive 5

←

Archive 10

Archive 11

Archive 12

Archive 13

Archive 14

Archive 15

Specific use question, advice please

I was just wondering if anyone could assist me with a question: Is more attribution necessary if the fact in question explicitly refers to the credits of a television show? From Are You Smarter Than a 5th Grader?:

During the credits at the end of the show, a disclaimer states: "Members of the class [the children] were provided with workbooks that covered grade school level material in a variety of subjects. Some of the material could have formed the basis of questions used by producers in the show."

This statement was marked citation needed, which was then replaced by the qualifying statment, "This disclaimer can be viewed by pausing the show at the appropriate point as the credits roll." This obviously does nothing more than repeat that the information is present in the credits sequence, and otherwise is useless.
My first thought is that the statement should be sourced, but the more I think about it ... how can it be? Is the show itself not a primary source? I don't think it'd be easy (if possible at all) to find a different source citing the credits of the show (I've looked). However, this statement shouldn't be discounted if it is true (to which I cannot verify, as I've never paid particular attention to the credits). Any advice? Thanks! –Dvandersluis 19:33, 2 March 2007 (UTC)

No need for the "pause the credits" statement. The show itself is the source. If someone demands a citation try: <ref>''Are You Smarter Than a 5th Grader?'', Fox Television, closing credits, first aired: March 1, 2007</ref> If you really want to be accurate (almost to the point of absurdity), time the show and note when the line appears. Blueboar 20:00, 2 March 2007 (UTC)

Thanks! :) –Dvandersluis 20:03, 2 March 2007 (UTC)

You should include enough information that people can verify that the claim is actually supported by the source. The should be able to seek and find the source and then find in the source where the claim is supported. Exactly what degree of detail is needed for this varies between sources. WAS 4.250 22:13, 2 March 2007 (UTC)

A source is something that someone can look up themselves, to prove that a statement came from somewhere. The source may or may not be convenient (maybe it is in a book at the Library of Congress) but at least the sourced statement does come from a place that can then be further researched.

In the case of the source being a claim that the credits state such-and-such... well, this is also something that can be looked up (by pausing a video or Tivo of the show) to prove that it is accurate. Until proven inaccurate, the claim should be left alone. This would be similar to a source citing a book at the Library of Congress. If you personally don't believe the quoted passage, the burden is on you to prove that the claimed source does not exist or that the quoted material is inaccurate. --Pmurph5 23:26, 2 March 2007 (UTC)

Actually we use common sense in these situations. If a known liar says the source is in some hard to access source, they are often reverted. WAS 4.250 23:38, 2 March 2007 (UTC)

What about well-sourced absurd claims?

I had inserted a well-sourced exceptional claim that keeps getting deleted. The claim is that a certain person can change from a man to a woman instantenously. The source is among others a peer reviewed university article. See Talk:Sathya_Sai_Baba#Request_for_Comment:_Exceptional_Controversial_Claim Andries 19:36, 2 March 2007 (UTC)

Any source that states that someone can change from a man to a woman is not a reliable source. Please provide links or citations of a source that states, scientifically, that someone can change from a man to a woman. Thanks. Hipocrite - «Talk» 19:41, 2 March 2007 (UTC)

Why should I? A relevant peer reviewed university press article remains a reliable source, according to this policy, even if the statements in the article contradict your beliefs. Andries 19:49, 2 March 2007 (UTC)

Please provide a link or a citation to this "peer reviewed university press article" which states, as a fact, that someone can change genders. Thanks. Hipocrite - «Talk» 19:50, 2 March 2007 (UTC)

Okay, I will though the article does not state it as a fact. Andries 19:52, 2 March 2007 (UTC)

A peer reviewed article that states that someone else claimed that someone can change their gender is a reliable source for the fact that someone else claimed that someone can change their gender, not a reliable source for the actual gender change. Hipocrite - «Talk» 19:53, 2 March 2007 (UTC)

Exactly. This is the sort of thing where, if you have good sources, you can pass along the info that those sources made that claim. It just should be presented as "X says he saw Y" and not "Y happened". I've been dealing with something very similar at John Edward. --Milo H Minderbinder 19:55, 2 March 2007 (UTC)~

Hipocrite, I replied on your talk page detailing the sources. Andries 20:02, 2 March 2007 (UTC)

But the current wording of the policy suggests that absurd claims even if they are well sourced should be removed. Andries 20:02, 2 March 2007 (UTC)

Not if reasonably read; both the paragraph that says "absurd" and Jimbo's post in the footnote talk only about unsourced material. But I added another "unsourced" to clarify this.Septentrionalis PMAnderson 20:08, 2 March 2007 (UTC)

Our articles on religion are filled with absurd claims. WAS 4.250 22:17, 2 March 2007 (UTC)

(outdent) We can have 'weird claims' or 'absurd claims' if they are properly attributed to a notable person, for example. If that person is the subject of an article, and is otherwise notable, then those claims, attributed to him/her, are includable. In situations where the source is not the article subject, and yet the views expressed are extreme, then we must decide, per Jimbo's fringe exclusion rule, whether the claims are so extreme and represent such a small minority as to be regarded as a fringe view, in which case they would be excluded. If the article is medical or scientific, for example, a reliable source would normally have to be published in a peer reviewed publication, and the more exceptional the claim, the higher the needed quality and quantity of the supporting sources. Crum375 23:02, 2 March 2007 (UTC)

It should be noted that in the talk pages of articles on a number of topics -- religion and political views, for example -- people regularly suggest that the views of those who disagree with them should be removed as absurd or wierd, despite the fact that views representing both sides are held by a substantial fraction not only of the population but professional philosophers/theologians/commentators. Atheists regularly claim that Wikipedia shouldn't have any articles on religion; religious people claim Wikipedia shouldn't have atheist or critical perspectives, you name it. Any proposed policy language that says that well-sourced views can be removed on grounds of absurdity shoud be very carefully vetted to make sure opinionated editors won't be quoting it to remove material based on their own subjective view of what is "absurd". Regarding the specific application raised, I would think that any article on Satya Sai Baba would need to describe the beliefs Sai Baba and his followers have about his nature and activities. A belief in Sai Baba's personal capacity to perform miracles seems to be a central tenent of that faith. However, as with religion articles generally, consistent with WP:NPOV, Wikipedia would do well to report both what believers believe and skeptics' criticisms, using fairness of tone for each, and leaving readers to make up their own minds. Given that the Satya Sai Baba religious beliefs involved are notable, the fact that some Wikipedia editors might think the beliefs involved absurd simply should not influence editorial decisions. A fair presentation of views offered by skeptics will have to do to present such editors' POV. Otherwise, we'll be hearing from editors saying Wikipedia shouldn't have an article on the Resurrection of Jesus or the article shouldn't include the views of theologically conservative Christians on that subject. Best, --Shirahadasha 03:41, 5 March 2007 (UTC)

The keyword here is "multiple". We need multiple references of the claims. If The New York Times, BBC, CNN and Washington Post state cows can fly, I will be the first to edit cow to add that fact. -- ReyBrujo 03:52, 5 March 2007 (UTC)

I would find it sufficient if Nature had a peer reviewed paper showing cows can fly. I would be careful to attribute the claim to the Nature paper, and probably quote the relevant items verbatim. I agree with Shirahadasha that 'absurd' is subjective, what we need here are proper attributions, not opinions about 'truth' by Wikipedians. Crum375 04:04, 5 March 2007 (UTC)

Have a look at Kent Hovind. True, his claims are absurd, but as long as we're careful to attribute to him ("Hovind states", "Hovind theorizes that"), they're entirely appropriate for his article. Seraphimblade ^{Talk to me Please review me!} 04:10, 5 March 2007 (UTC)

There was a single line that explained what was Wikipedia about: "Verifiability, not truth." Unfortunately it seems it was lost in the policy slaughter. -- ReyBrujo 04:58, 5 March 2007 (UTC)

Role of truth

This point has been raised a few times already and I haven't seen any good answer; I suspect it hasn't been clearly understood. The answers I saw seemed to be addressing a logically different point.

I'm concerned about a major policy shift in moving from "verifiability, not truth" to "attributable, not true". The word "verifiable" implies truth; "atributable" does not, and the meaning of the policy is different.

With the new Attribution policy, an editor can argue: "I admit that this statement is false, but I insist that it remain in Wikipedia, and I cite WP:A to back me up. Furthermore, I won't let you insert 'Agency X reports that...' because it will disrupt the flow of the article. There is already a footnote, and the false but attributed statement meets the criteria of Wikipedian policy perfectly. Truth is not a criterion."

That's not the way Wikipedia works, and I don't want it to start.

Attributability is one necessary criterion for inclusion; it shouldn't be the only criterion. Editors use their judgement to select among all possible attributable statements the ones that seem interesting, relevant, and important and that fit into the organization of the article, and in doing so, their perception of what is likely true rightly influences their judgement of what is worth saying.

Example: A Wikipedian article reported that someone did something in a certain year. The year was before the birthdate of the person. A Wikipedian editor rightly removed the obviously false statement. Another correct remedy might be to change it to "Source X reports that the person did this in such-and-such year." Simply leaving contradictory and obviously false information in the article should not be an option -- but the new Attribtution policy supports exactly that.

"Verifiability, not truth" could be fairly easily interpreted to mean that being true is not enough: they must (also) be verifiable. (Even that was argued about.) But "Attributable, not whether it is true" is clearly stating that truth is not a criterion for inclusion of information in Wikipedia -- opening the door to knowing and admitted inclusion of false statements.

I suggest that the policy be modified to either say something positive about truth, or to say nothing about truth. I.e. either delete "not whether it is true", or change it to something like "not only whether it is true," or insert something like: "The purpose of this policy is to ensure, as far as practical, that information in Wikipedia is accurate and reliable." Or: "Not all attributable statements are worthy of inclusion." Or: "In addition to being attributable, material to be included should be interesting, important, apparently true, and relevant to the article."

I would also like to change "the threshold for inclusion" to "a necessary condition for inclusion." The word "the" before "threshold" implies that it's the only criterion, again opening the door to baldfaced false (but attributable) statements.

I think the intended purpose of this policy was to exclude apparently true but unattributable statements, but that the wording can be taken as explicitly encouraging the inclusion of apparently false but attributable statements, so it needs to be edited to more clearly reflect its actual intended purpose. --Coppertwig 02:20, 3 March 2007 (UTC)

If there is no discussion on this point, then I will soon edit in the word "merely" to make the policy page say "not merely whether it is true". --Coppertwig 19:21, 4 March 2007 (UTC)

I suggested an edit very much like that a few weeks back. I'd be glad to see that clarification. DCB4W 19:26, 4 March 2007 (UTC)

Well stated. That makes perfect sense to me, for one. --Wetman 20:55, 4 March 2007 (UTC)

Thanks. OK, I went ahead and inserted "merely" into the policy; I also inserted it into the FAQ in a similar context; and I've proposed an additional related edit to the FAQ at Wikipedia talk:Attribution/FAQ#Truth. --Coppertwig 13:47, 5 March 2007 (UTC)

Coppertwig, please don't change the policies without clear consensus. Adding "merely" changed the meaning of the sentence completely. The point of it is that we don't do truth. We only do sourcing. SlimVirgin ^(talk) 13:50, 5 March 2007 (UTC)

I agree with slim... adding "merely" changes the meaning significantly. Wikipedia does not care if a theory or statement is "True"... for "Truth" is often in the eye of the beholder. That said, what we should care about is "Accuracy". This is a fine distinction, but a significant one. "Truth" is subjective, "Accuracy" is not. Truth can not be cited ... Accuracy can. Blueboar 14:34, 5 March 2007 (UTC)

Also agree. This merely dilutes the concept we want to put across. ≈ jossi ≈ (talk) 15:24, 5 March 2007 (UTC)

I apologize for not waiting long enough for comments before editing. I feel that a much broader discussion is required before changing the policy from "verifiability, not truth" to "attributable to a reliable published source, not whether it is true"; I see that as a fundamental change in policy.

An alternative is to change those words back to the longstanding formulation, with the advantage of being shorter. What does everybody think of that idea? I would also appreciate comments on how the policy does or should apply in the hypothetical case I mention above of an editor insisting on retaining a statement the editor has admitted is false. --Coppertwig 22:05, 5 March 2007 (UTC)

No, because one of the things achieved in the merge was to get rid of "verifiability," given its implications. There's no appreciable difference between the two sentences, and this was discussed very widely before being put in place. SlimVirgin ^(talk) 22:17, 5 March 2007 (UTC)

Please give links to where there was discussion and consensus to "get rid" of the concept of "verifiability". If the new wording has the same meaning as the old wording, then the old wording should be kept as shorter. If the meanings are different, wider discussion was needed and the change needs to be presented as a change in policy. I await your comment on the concern I raise at the beginning of this section. --Coppertwig 22:58, 5 March 2007 (UTC)

The old 'verifiability, not truth' has the exact same meaning as 'attributability, not truth'. In both cases we are saying that Wikipedians are not here to scientifically or otherwise verify the truth of the statement, but are merely summarizing what was reliably published. One of the reasons for the wording change is specifically because some people misinterpreted 'verifiability' to be related to the search for truth, whereas it simply means that we can provide a reliable published source. Attributability makes that same point unambiguously clear. Crum375 23:06, 5 March 2007 (UTC)

(reset indent) User jossi said in a poll section, "The policy is very clear: We do not make assessments about the truth or falsehood of the assertions made in reliable sources. Period. If editors do not wan t to use a specific source, for whatever reasons, and they reach consensus about not using that source, don't use it. Simple. " Actually, we do make aseessments about the truth or falsehood of the assertions made in reliable sources. Where does it say in policy that we don't? If it says that, it's not true. People almost can't avoid forming opinions of the truth of falsehood of what they read. Editors might choose not to use a source for a variety of reasons; whether it seems to be true is and should be often one of those reasons. If the policy simply says nothing about truth, that's fine: people will naturally assume the encyclopedia is trying to provide true material as much as possible, and will try to do so. But it says "not whether it is true," and the context of this negative statement about the truth is changed significantly by removing the word "verifiability". So it now looks as if it may be telling editors not to use truth as a criterion. Presumably they can still use whether something seems interesting or important, etc. Do writers of this policy actually want it to discourage editors from using apparent truth as a criterion? If someone knowingly inserts false (but attributable) statements into the encyclopedia, are they helping to write the encyclopedia? --Coppertwig 15:19, 6 March 2007 (UTC)

The idea is that we're not in a position to determine whether a statement is true or not. There's a good reason for that: if we do, articles start to reflect the opinion of the authors rather than a neutral representation of the available sources. Reliable sources, however, can make such a determination, and what really matters is the consensus of experts, as expressed in published works. Jakew 15:35, 6 March 2007 (UTC)

I think a basic problem here is that "verifiability, not truth" is pithy, whereas the current version just isn't any more. This happens in writing: sometimes an idea fits well into what you're saying, and sometimes it doesn't. I think to many here, a catch phrase like "verifiability, not truth" is simply less threatening than a more elaborate sentence saying truth doesn't matter. Indeed, further down the spectrum, you could actually say it even more specifically yet: "The question is whether something is attributed. Truth doesn't matter." Would that be ok?

The thing is, also, if not as a good catch phrase, why make this point? Indeed, aren't there an infinite number of things that aren't ultimately the issue, particularly after combining the policies? "Not whether it is true, apt, obvious, important, etc." Particularly after combining the two, singling out truth as not mattering on any issue relating to either seems to take the policies where they didn't go before. I guess the question is: does the immateriality of truth really need to be stated with such prominence? This may have been primary to the idea of "verifiability," but to the combined policy I'm not sure it is. Mackan79 16:02, 6 March 2007 (UTC)

User Jossi said in a section below: This is exactly the stuff we do not want in our policies. Editors' needs to apply their best and collective judgment in these cases. I strongly oppose such wording in the policy. I agree completely! I oppose any wording in policy that is likely to be used to prevent editors from using their best and collective judgement in choosing which attributable statements are or are not worthy of inclusion in an article, and in what order.

I suspect that there may not be much disagreement about what we want the policies to do (or not do). But there is definitely disagreement about the wording.

In reply to SlimVirgin, BlueBoar and Jossi's comments above: As I understand your comments, you're indicating opposition to the insertion of "merely". However, I don't see any comment from you about other possible ways of addressing the concern I raise at the beginning of this section.

"Verifiability, not truth" means, as user tjstrf points out below, approximately the same as "Truth, not truth". The phrase "not truth" is there only to indicate that the part of the word "verifiability" that means "truth" is not what is meant here -- what is meant is attributability. When we move to the word "attributability", we no longer need the "not truth" part -- we can be very concise and just say "attributability". Leaving in the "not truth" part when it doesn't have a truth-word such as "verifiability" balancing it makes it look as if we mean that Wikipedians don't care about the truth, which is of course not true.

Inserting "merely" is one way to move the meaning of the "not truth" part back closer to what it meant when balanced by "verifiability". Deleting "not truth" is another solution. Simply keeping the original words, which are shorter, is yet another solution. There are probably still others.

User Mackan79 makes a very good point above. Think how terrible it would be if the policy were to say, "The threshold for inclusion is whether it is attributable, not whether it is interesting, important or relevant to the subject matter of the article." Then someone could insert information about a famous pet dog into an article that has nothing to do with it, and when people try to revert it, insist on keeping the information in, citing the policy to support them. If the policy says nothing about whether something has to be relevant, then common sense and other policies will prevail; but if it says something negative about relevance, people will use it to keep in irrelevant stuff. Similarly with the truth. The policy should simply say nothing about the truth.

User JulesH said, in another section, that people were confused about the meaning of the word "verifiability". If some interpretations were correct and others were incorrect, then there must be a document or record of a process somewhere, at the level of discussion required for establishment of policy, which clarifies which interpretations are correct. Would someone please tell me where to find this document? If not, then the actual wording "verifiability, not truth" is in itself the policy and all interpretations are equally valid. If these pages are merely the documentation of the policy and there's a real policy somewhere that they're documenting, then where is the real policy? I'd like to read it. --Coppertwig 13:50, 7 March 2007 (UTC)

(I apologize; I think I accidentally cut-and-pasted a user name, changing a signature above, when I meant to merely copy; I think I've fixed it.)

I suggest replacing " The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true." with the longstanding wording, "The threshold for inclusion in Wikipedia is verifiability, not truth.". --Coppertwig 14:24, 7 March 2007 (UTC)

Primary sources examples

Should these be moved to WP:ATT/FAQ? Otherwise that section becomes too verbose. ≈ jossi ≈ (talk) 02:55, 3 March 2007 (UTC)

Right now, "primary source" is chiefly defined by those examples; if we strip them out, we are left with the first sentence. We probably should do this, but we'll need a paragraph to replace this one. Septentrionalis PMAnderson 03:54, 3 March 2007 (UTC)

I liked the simplicity of that paragraph without all the examples. SlimVirgin ^(talk) 02:56, 4 March 2007 (UTC)

Wikilinked to the FAQ. ≈ jossi ≈ (talk) 03:32, 4 March 2007 (UTC)

independent sources

This edit by Jossi does not preserve the point involved: that an article which uses questionable sources must be based, somewhere, on reliable sources, independent of the subject. The revised phrasing would permit an article to use dubious sources if the article is based on other bad sources.

This probably does need to be rephrased; it's been seriously misunderstood twice in the last week; but the point is (and has been reverted to as) consensus. Septentrionalis PMAnderson 04:02, 3 March 2007 (UTC)

This section requires intricate reasoning. It only covers the case of questionable or self-published sources in articles about themselves. In general, sources do not always have to be independent of the subject of the article; in some cases, the concept does not even make sense. --Gerry Ashton 05:14, 3 March 2007 (UTC)

Can a PhD thesis be a reliable source?

Specifically, can this Sociology PhD thesis on the history and social context of Transcendental Meditation be used as an RS for the Wikipedia article on TM? -Sparaig 06:06, 3 March 2007 (UTC)

The short answer is no. Remember what a thesis is: It's someone's opinion on a subject, being submitted for a university examination. Any decent thesis should have reliable sources of its own which can be drawn upon. Use those instead.

Two exceptions come to mind: If you are discussing the thesis itself as part of the article's subject, and the author of the thesis is notable, then it may be suitable as a source. Also, if a thesis gets published in a reputable, reliable scientific joural that is separate from the university, then it's suitable as a source. In both cases, it's important to characterise the thesis as being an opinion, not as an authoritative statement of fact. -/- Warren 06:18, 3 March 2007 (UTC)

The thesis does cite written sources, but it also summarizes interviews by the author of people who are now dead--specifically on the topic of TM and Maharishi Mahesh Yogi and his relationship to his religious tradition. This is central to any discussion of MMY AND TM's history and background since he was secretary to Brahmananda Saraswati, the Shankaracharya of Jyotiermath, and one of the people interviewed served as Shankarachara at the same monastery following their mutual guru's death. The interviews and written sources are probably the best available on the topic. -Sparaig 06:49, 3 March 2007 (UTC)

Use of the interviews rests on evaluation of the author of the thesis as a reliable source or the institution accepting the thesis as an institution that is reliable in verifying interviws in a thesis; and the lack of surprise in the content of the interviews. Something surprising in the interview would need to be corroberated by another source. WAS 4.250 08:42, 3 March 2007 (UTC)

I would say that a thesis is a reliable source. At least in the US, a thesis is reviewed by an academic committee and is published and publicly available in the university's library. A thesis is not an opinion, any more than any academic publication. Kevinp2 08:00, 3 March 2007 (UTC)

I agree with Kevinp2: I don't understand why a Ph.D. thesis should not be considered a reliable source. In addition to the reasons Kevinp2 offered, the work is constantly supervised by at least one experienced professional in the field to ensure quality control and solid presentation. True, there are some flawed (sometimes seriously) Ph.D. theses out there, but given the quality of some published non-fiction, that shouldn't be the reason to deny them for WP citation. By the way, as Kevinp2 stated, a thesis is more than just an opinion: that's why other academics consider it reasonable to cite them in other published works. -- J Readings 08:29, 3 March 2007 (UTC)

I thought a thesis was Original Research? - Peregrine Fisher 08:31, 3 March 2007 (UTC)

We can't put our original research in the article. We can do original research as background to make choices like what is a reliable source and we can use other people's original research as a source if it is reliable and published. WAS 4.250 08:37, 3 March 2007 (UTC)

If a thesis is published, then there's no problem, it's been vetted by another source. We can use that secondary source. If it hasn't been published, then all we can say is so-and-so believes whatever, if we're even doing an article that involves what grad students believe. I think we're talking about non-published thesises, which would be a no-no. Basically just saying what Warren said again. - Peregrine Fisher 08:59, 3 March 2007 (UTC)

But a thesis is also reviewed and approved by an academic committee of the university (to allow the student to graduate) and are then published by the university and made available to the public in its library. I would say that a draft non-approved thesis is probably not yet a reliable source but a thesis that has been reviewed, approved and published by the university is a reliable source.Kevinp2 09:19, 3 March 2007 (UTC)

Exactly. -- J Readings 09:28, 3 March 2007 (UTC)

I agree with Kevinp2. The review of a thesis by an examining board of the university is analogous to peer-review in academic journals; some say that the thesis review is stricter than peer review. If there is a single copy of the thesis always available in the university library for anyone who walks in to look at, it can be considered published as much as the rare-book example discussed earlier. More widely available material is preferred if such exists.

Peer-reviewed publications also contain opinions, and we Wikipedians rely on that: we can't write our own opinions into the articles, but we do need opinions to make the dry facts flow together meaningfully. --Coppertwig 14:31, 3 March 2007 (UTC)

I think I read somewhere (it's hard to tell with the recent reoganization) that we're not supposed to use grad student research papers. Now maybe it's different with a thesis that's accepted, and the student graduates. Is this where it changes? - Peregrine Fisher 09:37, 3 March 2007 (UTC)

It would be nice to be sure. Coplin got his PhD in Sociology from UCSD in 1990. The online version of his thesis is billed as the final version. -Sparaig 09:40, 3 March 2007 (UTC)

You might want to check with the UCSD library to see if what they have matches what he has posted online. You might be able to get the thesis via an interlibrary loan or they may have digitized it (less likely since he graduated in 1990).Kevinp2 10:02, 3 March 2007 (UTC)

it exists. I have no reason to assume he's putting up kilobytes of false info online. -Sparaig 12:28, 3 March 2007 (UTC)

Me neither :-) But it bolsters your case for considering it a reliable source.Kevinp2 16:14, 3 March 2007 (UTC)

I can find it in the library of University of California, San Diego here, so that should help with an inter library loan. The only thing I would place caution on is whether you are giving this source undue weight. What exactly are you looking to substantiate through this source? Hiding Talk 19:41, 3 March 2007 (UTC)

(indent reset) I would tend to agree that the review process for a PhD thesis is analogous to peer-review, so long as it's a thesis for an accredited university. If the university is unaccredited, its fact-checking process is meaningless. Seraphimblade ^{Talk to me Please review me!} 20:04, 3 March 2007 (UTC)

Perhaps we should have an FAQ entry on this subject? JulesH 22:06, 3 March 2007 (UTC)

The university was University California San Diego, for a PhD in Sociology. The thesis covers the early history and historical/social context of the earliest years of the Transcendental Meditation organization, starting in the mid-1950s through the late 1980's, and includes research on the background of Maharishi Mahesh Yogi, as well as summaries of interviews with his earliest students, and a monk who was in the monastery he lived in at the time he lived there who later became head of that particular physical monastery (see here for an explanation of my peculiarly precise wording). At this point, aside from a contention about his date of birth (is he 90 or 95? --even MMY says he doesn't remember, even though someone told him a few years back), I'm not worried too much about including specific bits in the relevant articles, but I wanted to check first. Also, Dr. Coplin only put about 1/3 of his thesis online, and the more meaty bits appear to be the parts he didn't post. If the thesis is considered a RS then I'll bug him more about putting the rest up. I'm also curious for my own interest in the subject. -Sparaig 22:50, 3 March 2007 (UTC)

It's déjà vu all over again. I brought up this question last October; the discussion is at Wikipedia talk:Reliable sources/archive8#Dissertations. —Angr 08:09, 6 March 2007 (UTC)

A doctoral thesis which is accepted by the person's doctoral committee becomes a published piece of scholarly research. It is far from just being like a term paper a student writes for a course. If it were a hoax, biased opinion or shoddy research, it would reflect on the department, the university, the doctoral advisor and the committee, so they have even a stronger incentive to make sure it is accurate and valid than if they were anonymous volunteer editorial consultants for a journal. A scholar has to jump through considerable hoops to get the thing accepted. They are then sometimes cited in books or scholarly journals. They are indexed in Dissertation Abstracts and are available on demand from University Microfilms. They are frequently published in scholarly journals, but we need not wait for that to use them as citable sources in Wikipedia articles. My criterion would be that it had been accepted as part of the requirements for the PhD (not just a draft or preliminary or rejected dissertation) and that the university itself be accredited. That would put it in the same reliable source category as a refereed scientific journal (which is no guarantee that everything in it is absolutely true and accurate and unbiased). But it is far, far more than just someone's opinion, like a blog posting or a letter to the editor. The greater hazard of dissertations is not that they are wrong, but that they are boring unimaginative repetitions or boring empirical studies of subjects that others do not find interesting. Edison 21:44, 6 March 2007 (UTC)

I realise I'm jumping in a little late here but...A thesis is a reliable source, but should not be used in every circumstance. If you are writing in an article, for instance, Some political scientists, such as X (name of author of thesis) (reference), have argued that political power is dominated by a small elite of wealthy corporations, then a thesis is OK as a source. If, however, you are trying to use it as a statement of fact rather than of academic opinion, then it isn't necessarily an appropriate source (unless, for instance, it's a scientific study that generated new experimental data). There's nothing inherently unreliable about a thesis, but it should be used in conjunction with other, more definitive sources. Walton ^{Vivat Regina!} 16:17, 7 March 2007 (UTC)

I would agree that a dissertation in the humanities anyway (I am unfamiliar with dissertations in the sciences) should qualify as a reliable source. The fact that they are overseen by a committee of scholars in the field makes them peer-reviewed sources. In my field, English literature, scholars often rely on dissertations when there are no other sources. Therefore, you do see dissertations quoted and footnoted in both peer-reviewed articles and books when there are no other sources available. I agree that these works are not ideal because they are incompletely argued and contain lots of summary (they are student exercises in many ways), but that does not make them unreliable. Many books and articles also contain the same flaws. I would say that one has to evaluate each dissertation individually; one famous example is of course Derrida's which contained the seeds of deconstruction. One significant problem with dissertations, at least in the United States and Canada, is accessibility; one must obtain them through a well-financed research library. I was under the impression that all dissertations were published through the University of Michigan. They used to have a dissertation microfiche program, although now I think that it is slowly becoming digital. Awadewit 15:28, 8 March 2007 (UTC) - I agree with Sparaig that this should be added to the FAQ on sources so that question doesn't keep coming up.

Press releases as a primary source

I added "press releases" as a type of primary source. It seems appropriate to me, given that Wikipedia links to quite a number of press and news releases for attribution, and we have {{cite press release}} to go along with that sort of usage. Hopefully this is allright. This is a really well-written document, kudos to everyone who's been involved with it. -/- Warren 06:38, 3 March 2007 (UTC)

Thanks, Warren. SlimVirgin ^(talk) 02:55, 4 March 2007 (UTC)

Self-published source restrictions seem too broad

I am a little concerned that the restrictions on using self-published sources will kill off the ability to use a lot of good information that is available on the Internet. I offer the example of Tom's Hardware Guide that is an online site that publishes excellent reviews of computer hardware and accessories. Here is an example of a review of DDR RAM. This review is one that I feel comfortable using to make a purchase worth hundreds of dollars. Yet, we can't use it because it is a self-published source and the author is not a "well-known, professional researcher". Given the fast moving nature of this business, I doubt that this review will ever appear in any printed publication, since it will likely be obsolete by the time it is published. Yet, the review itself seems reasonable and well researched to those who understand the subject matter.

There are numerous such sources of information in many fields of human endeavor. While there is no guarantee that the information is reliable, readers are usually able to read the articles and decide if they are reliable.

At the present time, the reliable sources policy seems biased towards academia, publishing houses and the news media. While these organizations have staff and a peer review process, these processes by no means guarantee reliability, considering some of the news media scandals that have been exposed in recent years. Moreover, they tend to be exclusive in nature, often filtering out authors, subject matters or perspectives that they don't find interesting. The Internet in general is democratically leveling the publishing playing field, whether we like it or not. I certainly agree that some self-published material is unreliable but I also contend that some self-published material is reliable and should be used when appropriate. If we don't come up with some way to do this, Wikipedia will be left behind in future years as more and more good information is directly published by individual citizens.

I would like to invite comment and suggest some criteria that might be feasible:

• The material is signed by a real person using a full name, i.e. no anonymous or pseudonymous postings.
• The real person provides contact information to enable direct communication.
• The material is marked with a special link to indicate that it is self-published i.e. reader beware.
• Restricting sourcing to non-political and non-biographical articles.

Kevinp2 08:48, 3 March 2007 (UTC)

Is Tom's Hardware not a reliable source? I would have thought it was. Do they let just anyone write for them? If they do, then there not reliable. - Peregrine Fisher 09:01, 3 March 2007 (UTC)

They have a list of editors but it is not evident to me about how their authors are selected, or how or if their work is approved. I have never personally heard of any of the authors (which is not surprising in this field). I focus on the subject matter and whether it seems to be reasonably written and well researched. Are you suggesting that having some kind of control process is what you consider important?Kevinp2 09:14, 3 March 2007 (UTC)

(edit conflict)Having an editor who stands in the way of bad info is key. That's why blogs and wikis aren't reliable; anyone can say whatever they want. Basically, there's two kinds of reliable: a writer can be reliable on their own, or an editor can screen writers to make sure they're reliable. In the big picture, there should be no way that a wikipedia editor who wants something added can go and add that info to another source, and then cite it. That's where editors come in. - Peregrine Fisher 09:30, 3 March 2007 (UTC)

There can be cases where blogs can be considered reliable. As an example, Volokh.com is a blog of well know lawyers and law professors. If you read this post by law professor Jonathan Adler, it is a good example of a secondary source, reviewing the primary source of an Ohio Supreme Court ruling. In fact, this is my larger point, that just because something is self-published, does not automatically mean that it is unreliable. I would like to see some high but attainable standard for sourcing self-published material. Perhaps one of the standards could be that the real person who is the source is well known as an expert in the field.Kevinp2 09:53, 3 March 2007 (UTC)

The problem here is that by the rules listed here (and previously at WP:V), these blogs are not reliable sources. I don't dispute that they should be, but the rules are quite clear on the subject. Unless somebody is a "professional researcher", which none of these bloggers appear to me to be, their blogs cannot be cited. JulesH 13:07, 4 March 2007 (UTC)

Which is why I am asking that we change the policy to state that subject to strict criteria, that certain blog posts (as opposed to all posts from certain blogs) and other self-published material can be used, with every use being evaluated individually. The criteria could be all of the following:

• The material is signed by a real person using a full name, i.e. no anonymous or pseudonymous postings.
• The real person provides contact information to enable direct communication and the editor verifies that the real person exists and owns up to the material.
• The real person is acknowledged or reputed to be an expert in the field.
• The material is relevant to the topic at hand and no other secondary source is available that can be used instead.
• The material is marked with a special link to indicate that it is self-published.
• The material meets the other requirements for being a reliable source - i.e. non-partisan, not advancing an agenda.
• The material cannot be used in biographical articles.
• Every use has to be individually evaluated, i.e. if an author's self-published source is used once, that gives it no special credence for future use.
• There is a way to state (perhaps through a < - - comment template ) that the editor has verified that the source meets the above requirements).

The above list sets a high but attainable standard that is reasonable for Wikipedia to accept some of the the increasing amount of good material that is self-published on the Internet and will continue to increase in future years, that will not make its way into the official publication system.Kevinp2 14:29, 4 March 2007 (UTC)

I agree with a lot of the comments you are making, unfortunately blogs and more generally user submitted information is still a new thing to a lot of people who have not grown up with this. But this will only get bigger and bigger, as it becomes more widespread. This is part of "Web_2.0". Consider this, many hundreds of years ago when newspapers were still young we would have heard a lot of similar objections to what we are hearing now. Likewise blogs for instances are increasingly becoming the modern day newspaper, as can be seen by many major newspapers including blogs as part of what they do. It is highly ironic that wikipedia that is itself part of Web 2.0 is not moving with the times and recognising these changes that we too are part of this new improvements of the web. Mathmo ^Talk 03:03, 7 March 2007 (UTC)

Ask youself: why should the reader belive this claim? Oh cause this source says so. Well, why should the reader accept that source for that claim? do others? Is the author an acknowledged expert? By who? was the source reviewed by independent knowledgeable people? and provide in the footnote enough for the reader to judge the nature and quality of the source if there is an issue. WAS 4.250 09:24, 3 March 2007 (UTC)

I would certainly do these things. In this case, I consider Tom's Hardware (the site) to be an acknowledged expert. But the text of the policy (as presently written) still seems to forbid this source. Kevinp2 09:35, 3 March 2007 (UTC)

Are you concerned with "A self-published source is material that has been published by the author, or whose publisher is a vanity press, a web-hosting service, or other organization that provides little or no editorial oversight."? I don't think tomshardware fits this. Their articles are not written by Tom, and I think they have editorial oversight. I looked at there home page and couldn't find where it says what hoops a reviewer needs to jump through, but I think they are significant. - Peregrine Fisher 09:50, 3 March 2007 (UTC)

I agree with you that the web site Tom's Hardware has some kind of editorial oversight, even though the individual authors may not be well known. However (and maybe I am mistaken), I have seen and understood the ban on self-published sourcing to be construed very broadly.Kevinp2 09:57, 3 March 2007 (UTC)

I think the few users talking here today can agree that it's a realiable source. It would be nice if we could get a definitive answer out of this page, that we could use in the future, but we probably cannot. That's the beauty/curse of wikpedia, it's all on the editor to judge. Maybe someday we can use something like the AfD process to decide what sources are reliable; until then it's on a case by case basis. The best you can probably do is provide a rational on the talk page why you think it's a reliable source, mentioning the reasoning we've been talking about. - Peregrine Fisher 10:29, 3 March 2007 (UTC)

Does this all mean that an editorial is self-published? There is not editor for one (the board wrote it). Also, most op-eds are writen in the same atmosphere (no editor can change content) so they should also be considered the same as self published? If so, the page could use some clarification. Pdbailey 02:21, 4 March 2007 (UTC)

By definition, an editorial is not factual, so no, it wouldn't be reliable, no matter what it's printed in. Tom's Hardware, though, I imagine would be-it undergoes fact checking and editorial control, and is accepted as an authority in its field. Seraphimblade ^{Talk to me Please review me!} 02:24, 4 March 2007 (UTC)

"By definition, an editorial is not factual, so no, it wouldn't be reliable, no matter what it's printed in." I completely disagree. Citing editorials from well-known reliable sources is perhaps the best way we have available of showing people's opinions on a subject, and opinions are important in many articles. As to Tom's Hardware, I have no doubt of the site's reliability. The material is not self-published: it is published by Tom's Guide Publishing AG, and is no different in any real respect to most magazines (i.e., it is written on contract for the publisher, which retains editorial control). JulesH 13:07, 4 March 2007 (UTC)

Alright, that I can see-a statement that "The New York Times endorses candidate X for position Y", and citing the op-ed they wrote as an endorsement as a source. I agree on that one. Seraphimblade ^{Talk to me Please review me!} 20:23, 10 March 2007 (UTC)

Proposed new example of synthesis

(I originally posted this comment a few weeks ago, and decided at the time to leave the example and revisit the issue when the Attribution policy was live)

I think the whole plagiarism example is hard to understand and ambiguous. It's meant to be illustrating the idea that given points A and B which have reliable sources, you can't a new point C if this argument has not been made in the sources. However, it's not at all clear from the example what points A, B and C are meant to be, particularly as the plagiarism case it is based on is complicated. As a result, different people have repeatedly been coming to different conclusions about what the example is trying to say.

I propose replacing the example in the synthesis section with a more straightforward example along the following lines:

Example:

"In Northtown there were 1000 violent crimes last year_{Reliable source 1}, whereas in Southtown there were 2000 violent crimes in the same period_{Reliable source 2}. Therefore there is more violent crime in Southtown than Northtown_{No source}."

Under the attribution policy, the conclusion that there is more violent crime in Southtown than Northtown is not permitted, where it is not backed up by an appropriate source. This is because coming to a conclusion like this requires careful consideration of the source data; for example, the two statistics might have been calculated in different ways, or using different definitions of which crimes are violent or not. Wikipedia is not in a position to verify this background research itself. Instead, a conclusion must be attributed to a reputable source, unless it is very straightforward and uncontroversial.

Lots of alternative examples along similar lines could also be considered. What do you think? Enchanter 11:47, 3 March 2007 (UTC)

Your example seems to fall into the "simple mathematical calculation" exception to me. Can we come up with something that does not involve numbers? Blueboar 13:55, 3 March 2007 (UTC)

If the current is too complicated, then this is too simple. What's good about what we have now is that statement C actually involves research: picking up the manual of style and making an apperantly harmless observation, that is in fact OR. Note it's an "if/then" statement (with "then" elided) and this is central to why it's OR. The suggested lacks the nuance of the current. Marskell 14:16, 3 March 2007 (UTC)

I suggest changing the beginning of the 2nd quoted text from "If Jones's claim that he always consulted the original sources is false, ..." to "If Jones merely copied references, ..." This also has the advantage of shortening it.

Even better: Delete the first sentence, and change the second sentence to: 'The Chicago Manual of Style does not refer to copying references as "plagiarism." ' --Coppertwig 14:16, 3 March 2007 (UTC)

The suggestion is too simplistic, and it's not clear it's an example of an unpublished synthesis that serves to advance a position. The example we have in the policy is a real one, and it's illustrative of the complexity of the examples we often have to deal with. SlimVirgin ^(talk) 22:37, 3 March 2007 (UTC)

Could you explain further? Do you mean that complexity in itself is desirable, to illustrate that these situations are complex? Or, is there some essential aspect of the example which my edit would remove, and if so, what is it?

How about this instead? Source A classifies armed robbery as a violent crime. Source B states that Brown was convicted of armed robbery. Wikipedia should not state, based only on those sources, that "Brown committed a violent crime."

How about a simple fictional example, plus a link to the more complex, real-life Smith/Jones example on a subpage, perhaps in the FAQ? --Coppertwig 02:13, 4 March 2007 (UTC)

It's better to have a real example than a fabricated one. The examples you're giving aren't really examples of an unpublished synthesis serving to advance a position. An armed robbery, for example, is unquestionably a violent crime, and you offer no context for us to judge whether a position is being advanced. The real example is complicated, and it needs to be, otherwise there would be no point in including an example. The simple cases are easy to understand. SlimVirgin ^(talk) 02:53, 4 March 2007 (UTC)

What we need is a simple example which is easy to understand, in order to illustrate the concept of "synthesis". It's not obvious to me that someone who was convicted of armed robbery committed a violent crime; maybe they just had a gun in their pocket during a nonviolent crime or something. Could you come up with a simple example of disallowed synthesis? Concepts are normally illustrated with the simplest possible example that illustrates them. I don't think "synthesis" means "complex". I see nothing wrong with having both a simple, fabricated example and a real, complex one.

Here's another example: Source A says that everybody in a certain town always wears a hat all day on Sunday. Source B says that a resident of the town was interviewed on a Sunday. Wikipedia should not say, based only on these sources, "The resident was wearing a hat during the interview." --Coppertwig 04:04, 4 March 2007 (UTC)

The synthesis in the hat example is a flawed synthesis because the resident of the town might not have been in the town at the time of the interview. If the example stated the residient was in the town during the interview, the synthesis would be allowable. The section of Attribution that justifies rejecting the example is not "Unpublished synthesis" but rather "Exceptional claims require exceptional sources." It is an exceptional claim that everybody in an entire town always wears hats all day on Sundays.

I'm not commenting just for the sake of poking holes in Coppertwig's logic. I think this illustrates that for any synthesis example we use, our objection should be just to the synthesis process, and not because the statements taken from source A or B are difficult to believe.--Gerry Ashton 04:32, 4 March 2007 (UTC)

A couple of notes on SlimVirgin's comment above:

• I don't understand the argument that it is better to have a real example than a fabricated one. It is irrelevant to a reader trying to understand the policy whether the example is real or not; and
• I strongly disagree with the argument that complicated examples are a good thing. The point of this page is to explain policy. If people don't understand the example, they are likely to fail to understand the policy that the example is supposed to be illustrating.

I suggest that, at minimum, the existing example is simplified so that it is clear to the reader what is going on, without having to guess at what the context and background is. (the statement that "The whole point of this paragraph is the conclusion that, given the Chicago Manual of Style's definition of plagiarism, Jones did not commit it." does not logically follow from the text above it, although this is not obvious given the complexity of the example). Enchanter 23:27, 4 March 2007 (UTC)

Perhaps we need two examples: first, an example of a synthesis of the type which is allowed, and secondly, an example of one which is not allowed. Here's another example: Source A states that residents of a certain town are always wearing boots when they're on the street in January. Source B states that a resident was interviewed on the street in that town. Wikipedia should not say (or should it?) based only on these sources, "The resident was wearing boots during the interview." --Coppertwig 01:36, 5 March 2007 (UTC)

lecture not a reliable source

Is a lecture about how a discovery or theory-change was made a "reliable source?" Specifically, I'm in an edit war over this phrase in the article on John Hagelin. The other editor says that Hagelin isn't a reliable source concerning... something... :

• 1986: After a series of discussions with Maharishi Mahesh Yogi on the potential relationship between Physics and Vedic Cosmology (see [1] and [2]), Hagelin made some preliminary modifications in Flipped SU(5) Superstring Theory to make it more in-tune with Vedic philosophy, which also made the theory more robust from a Western scientific perspective. He contacted John Ellis of CERN with this information who then contacted Dimitri Nanopoulos and the three published many papers on the subject over the next several years." -Sparaig 07:51, 4 March 2007 (UTC)

Any claim that connects Vedic Cosmology and Superstring Theory needs to be very very reliable and some lecture is nowhere good enough. It like claiming Columbus not only discovered America but also inspired computers. WAS 4.250 07:56, 4 March 2007 (UTC)

The point is that Hagelin contacted the other two after applying his philosophical beliefs to the theory. This is, in principle, no more controversial than Kekulé using a dream to explain the structure of benzene, though of course Kekulé didn't go on to write long articles on the relationship of snakes to chemistry. -Sparaig 08:05, 4 March 2007 (UTC)

As I've said elsewhere, this doesn't satisfy V at all (a lecture a TM supporter went to?), certainly doesn't satisfy RS, and comes nowhere near satisfying the Acceptable Sources criteria of the Pseudoscience ArbCom decision for sources on science. Both of the Hagelin papers also clearly fail the AS criteria, as well as RS. --Philosophus ^T 08:27, 4 March 2007 (UTC)

Actually, a video of a lecture Hagelin gave, formally distributed by his university (IOW, published). Regardless, it's Hagelin's account of how he came up with his initial changes to the theory, and is certainly as RS as Kekulé's account of how he came up with the structure of benzene in a dream which is mentioned in the wikipedia article on Kekulé:

He wrote that he discovered the ring shape of the benzene molecule after dreaming of a snake seizing its own tail, a common symbol in many ancient cultures known as the Ouroboros. This dream came to him after years of studying the nature of carbon-carbon bonds. Kekulé claimed to have solved the problem of how carbon atoms could bond to up to four other atoms at the same time. While his claims were well publicized and accepted, by the early 1920s Kekulé's own biographer came to the conclusion that Kekulé's understanding of the tetravalent nature of carbon bonding depended on the previous research of Archibald Scott Couper (1831-1892). Furthermore, the Austrian chemist Josef Loschmidt (1821-1895) had earlier posited a cyclic structure for benzene (and many other cyclic systems) as early as 1861, although he had not actually proved this structure to be correct. -Sparaig 09:43, 4 March 2007 (UTC)

OK, as I see the discussion here, we're talking about using a self-published source by the subject of the article to fill in some historical information about how he claims he developed some theory (which I'm not familiar with, and am not about to attempt to understand). I don't really see an issue here -- the guy is, presumably, the best available source on how he developed his own theory. The only thing that strikes me about the text quoted above is that "which also made the theory more robust from a Western scientific perspective" requires an additional third party source that's qualified to make such a statement. JulesH 13:00, 4 March 2007 (UTC)

That's pretty much the gist of it, except he didn't develop the theory originally, Dimitri Nanopoulos did. According to Hagelin, he did his initial tweaks of the theory, contacted Ellis with a fax "Isn't this the sweetest little thing" or similar words, and Ellis contacted Nanopoulos and the three collaborated on a bunch of papers. Flipped SU(5) isn't that important these days, but back in the day, it was a contender--it even got a special mention in the first episode of Sliders with John Rys-Davies scornfully scribbling the name on the blackboard because no-one in the Physics class could remember it. It isn't that important in the article save as insight into how Hagelin went from a hotshot theorist to being a New Age physicist: he made his mainstream Physics rep using the theory in the first place. -Sparaig 13:27, 4 March 2007 (UTC)

More accurately, he made his reputation by applying his philosophy to a mainstream theory. The question of whether or not it was better is unanswerable: it is what he says about it. One assumes that Ellis and Nanopolous agreed with him or they wouldn't have published papers with him on the subject. While Hagelin is bright, I don't think his reputation was such that they would have sought HIM out if he hadn't already made some substantial contribution to the theory--he was head of the Physics department at Maharishi International University (now called Maharishi University of Management) at the time, which was and is hardly a prestigious school in the Physics community. Regardless, I can certainly add a "which Hagelin believed" to the article in order to avoid OR. -Sparaig 19:28, 4 March 2007 (UTC)

Saying that Nanopolous and Ellis agreed about the Transcendental Meditation parts of this is likely a BLP violation without an extremely strong source, considering that if it were true, it would completely destroy their reputations. --Philosophus ^T 00:26, 5 March 2007 (UTC)

Except this source doesn't appear to satisfy WP:V. We can't just rely on Sparaig's word (even disregarding the COI problems with doing so). --Philosophus ^T 00:26, 5 March 2007 (UTC)

I'm sorry. I apparently have paraphrased Hagelin so badly that you have completely misunderstood. At no time did Hagelin say that he told Ellis that "TM made this theory better." He made no mention of including anything about Maharishi Mahesh Yogi, Vedic philosophy or TM in the fax to Ellis. All Hagelin asserts is that following his conversations with MMY about Vedic Cosmology, he examined several theories and decided that Flipped SU(5) was the most amenable to modification along Vedic lines. After his initial modifications, he realized that the modifications made it stronger scientifically,and faxed Ellis with a copy of his initial modifications and the note "isn't this the sweetest little thing?" or similar words. Ellis contacted Nanopolous, the original author of Flipped SU(5), and the three published many papers on the subject over the years. This was the gist of Hagelin's lecture. I have it on video someplace, and I'm quite willing to forgo adding the phrase until I review the video again. I'll even put it on youtube if I can figure out how if you are worried that my COI is coloring my memory so badly. -Sparaig 01:32, 5 March 2007 (UTC)

Important item missing from RS

Plase see this discusion - which was suddenly archived (shortly after I advertised in on WP:RFC and WP:VPP). Since now the entire RS page is being shoved aside, I'd like to raise the point of historigraphic bias here.-- _{Piotr Konieczny aka Prokonsul Piotrus | talk}  18:43, 2 March 2007 (UTC)

My thoughts on the subject of historiography and bias is that bias is not really in the realm of RS (or even ATT... it is more a NPOV issue). A source can be reliable under our rules and still be biased. The key is how you use it... a history written with a distinct Nazi or Soviet bias is certainly reliable for what it says (as a statement of the opinion of the author if nothing else)... however, when using such sources in an article, it would be helpful to have some context, such as a mention of the potential bias of the source, what current authors say about that bias and contrasting what it says with other sources. Blueboar 19:41, 2 March 2007 (UTC)

Quite right; what do we have on the proper identification of sources? Septentrionalis PMAnderson 19:50, 2 March 2007 (UTC)

(edit conflict)As far as I can see the point being made here is that Communist historiography was blatantly Marxist, and dropped politically inconvenient facts down the memory hole. Both these things are of course the case; but Piotrus' conclusion seems to be that we should declare all such sources inherently not-reliable and ignore them completely.

I dispute this, and I see signs that Piotrus would qualify his remarks upon reconsideration. For one thing, the Stalinist PoV is a substantial PoV on many questions, and we should include it (together with refutations, often) under WP:NPOV. Again, work published under the Communists may well be the best current work on, say, Kievan Rus', although we should take the circumstances into account when judging its reliability against other sources. Septentrionalis PMAnderson 19:50, 2 March 2007 (UTC)

I don't claim we declare such sources not usable for Wikipedia. Certainly they have many real facts, and their POV is usually notable. However I see some editors ignore such POV and claim for example that if a prominent Soviet researcher said A, and prominent modern Western scholar(s) say B, views of both sides should be presented as equal, even when the issue concerns an area that would very likely be affected by Soviet ideology. After some thought I agree it is closely related to WP:NPOV#Undue_weight, but I don't think Stalinist (Soviet POV) should be give the same weight as modern research. Particulary when 'only' Soviet sources support such claims, can we claim they fail the current part of RS claims not supported or claims that are contradicted by the prevailing view in the relevant academic community.? For a case study, see Polonization, where a certain editor inserted a POVed view by a Soviet-era scholar, and keeps removing contradictory modern research...-- _{Piotr Konieczny aka Prokonsul Piotrus | talk}  05:10, 3 March 2007 (UTC)

RfC in a Dispute concerning Original Research

Wikipedia:Requests for comment/Language and linguistics (the request as of 07:19, 1 March 2007).
07:20, 1 March 2007 (UTC)

Thanks, JulesH, PMAnderson and WAS 4.250 (in alphabetical order), for answering the RfC. Of the three, JulesH and Was 4.250 sided with my position (like Grouse before you), while PMAnderson suggested some compromise, which Yom accepted and promptly incorporated into the Semitic languages article. I consider PMAnderson's suggestion an improvement over the previous version, so i'm pleased with this development. Nevertheless, i think this suggestion is still not entirely satisfactory in terms of complying with the attribution policy. I'd appreciate it if interested parties (the three above or others) spared some time to express their opinions in this matter. PManserson described his suggestion in his comment as of 18:59, 2 March 2007 on Talk:Semitic languages, and my critique of it can be found in my comment as of 07:59, 3 March 2007 on the same page.

I'd also be grateful for your participation in the related discussion that's been going on on the Talk:Tigrinya language page between JulesH, MikeG and myself.

Thanks. Itayb 07:47, 4 March 2007 (UTC)

I've indicated in the article whatever criticism i had. I'm pleased with the modified compromise. It seems to be stable (has lasted three days now). I've therefore struck the RfC request off the list of RfC's. I thank everyone who's stepped in to help. :) Itayb 08:43, 9 March 2007 (UTC)

Isn't there a policy saying Wikipedia itself is not a reliable source?

I asked this at Wikipedia talk:Reliable sources and was advised to ask here... Didn't the reliable sources guideline once spell out explicitly that Wikipedia articles themselves are not considered to be reliable sources? Am I mistaken? Or is this now addressed somewhere else, and if so, where? (In another venue, someone has cited Wikipedia as if it were a reliable source, and I'd like to point out that Wikipedia itself does not accept Wikipedia as a reliable source... but can't find what to point to). Dpbsmith (talk) 21:31, 26 February 2007 (UTC)

The point of wikipedia providing published reliable sources is to provide a way to verify our content, to establish credibility, and to provide a resource for further learning. See also and other internal links provide a source of further learning but pointing to ourselves does not add to legitimate credibility nor provide an independent way to verify our content so those aims must be met by linking to reliable published sources other than ourselves regardless of whether we are or are not ourselves reliable. We are actually at the point that we are more reliable than newpapers but less reliable than Britannica and have been cited by courts of law and numerous newspapers as well as scholarly papers using us as a primary source. Once you fully understand that even the most reliable sources have errors, it all becomes a matter of careful intelligent use and not bright-line distinctions. WAS 4.250 00:52, 27 February 2007 (UTC)

Whether we are more reliable than newspapers depends on the paper and on the article. We are less reliable than the NYTimes, but whether we are less reliable than News of the World is debatable. DGG 03:02, 27 February 2007 (UTC)

I believe there is a sentence to this effect in WP:ATT/FAQ. "Wikis, including Wikipedia and other wikis sponsored by the Wikimedia Foundation are not regarded as reliable sources." JulesH 09:22, 27 February 2007 (UTC)

When this was in RS, it was explained that Wikipedia articles were not reliable for several reasons... one, it would be a self reference (never a good idea)... two, since wikipedia is a work in progress, any information cited could well change or even be deleted at any time (which would affect the verifiability of any statement that was cited to a Wikipedia article)... three, because anyone can edit Wikipedia, we do not know if the person who wrote the article knows what they are talking about or not.... and four, there is no editorial oversight, peer review, etc. In other words, Wikipedia does not meet a number of our requirements for reliability. Blueboar 14:48, 27 February 2007 (UTC)

That's nonsense.

The first reason doesn't apply. A self-reference is saying something like "in this article..." Using Wikipedia as a source isn't a self-reference, as long as if you copied the article somewhere else outside of Wikipedia, the reference would still make sense.

The second of the four reasons given above is wrong. Wikipedia has the ability to link to a version made as of a specific date. This can't change or be deleted at any time; it's no more subject to change than any ordinary web-based source.

Reasons three and four amount to "it's a self-published source". While we *usually* can't use self-published sources, there are circumstances under which self-published sources are allowed.

This should be fixed, in the FAQ too. WP:RS has a lot of problems with this, among which is this problem. Ken Arromdee 17:09, 27 February 2007 (UTC)

• For the first point, see WP:SELF for the relevant sense of self-reference: a statement which implies that "this is a Wikipedia article".
• For the rest: Wikipedia is not a reliable source; if any source were appealled to which was as inaccurate, PoV-pushing and self-serving as the average WP article, I would insist, at a minumum, on a prose attribution. Septentrionalis PMAnderson 21:52, 27 February 2007 (UTC)

1) Using Wikipedia as a source does *not* imply "this is a Wikipedia article".

2) Saying that Wikipedia is inaccurate, POV-pushing, and self-serving is the same thing as saying "it has the problems of a self-published source". Nevertheless, self-published sources are allowed under some circumstances. Ken Arromdee 14:40, 28 February 2007 (UTC)

Using Wikipedia in Wikipedia is allowable, but should be avoided where possible; because we're almost all writing that article from a strong PoV. Beyond that, I see nothing to (2) at all. Septentrionalis PMAnderson 22:22, 28 February 2007 (UTC)

You're forgetting that this isn't just about Wikipedia articles. I can't think offhand of any case where we'd want to use Wikipedia articles as sources, but we might want to use articles from other Wikis or even Wikipedia talk pages. Consider the case where someone writes a statement about themselves or their own activities on a talk page (like the one where Eric Raymond says a particular accusation is false and the person who made it apologized). This is a self-published primary source. If Eric Raymond wrote that on his own web page, it would be clearly permitted as a self-published source in an article about the author, where there is no reasonable doubt who wrote it. It should be permitted if he writes it on a Wikipedia talk page instead. Ken Arromdee 14:59, 2 March 2007 (UTC)

How about linking to Wikipedia in another language? Somethig like: See the French version of the article for more info. --Brian Wiseman 00:30, 10 March 2007 (UTC)

The reason is that we are a tertiary source, but we use secondary sources. What works for one may not work for the other.

We strive to use mainly secondary sources. Our sourcing guidelines are geared mainly towards secondary sources; there is much doubt as to whether or not other wikis or sources of similar nature can be authoritative in that role. As secondary sources are by definition introducing new ideas to the world--ideas which can't easily be verified by following a citation--other means, such as peer review and author reputation, must be used to vouch for their reliability. Wikipedia is not a reliable secondary source (and per WP:NOR, not a secondary source at all), and shouldn't be used as such.

The Wikipedia project, on the other hand, is based on the assumption that the wiki model can produce a reliable tertiary source, which is what Wikipedia strives to be. With tertiary sources--especially ones which insist as strongly as this one does that the provenance of information be documented--you don't have to trust the author to know what he/she is talking about (at least not as much); in theory, everything is checkable. (In practice, there's a lot of work to do). So as a tertiary source, we are reliable. We hope.

However, we prefer to cite seconary sources. If tertiary sources cite each other (or themselves), there is the danger of tautologies being introduced. We avoid that issue by citing sources which are not likely to cite us for anything.

--EngineerScotty 22:36, 27 February 2007 (UTC)

If we are using another Wikipedia article as a source, then why aren't we cutting out the middleman and using the sources cited in that article? Oh, wait a minute, you say the Wikipedia article doesn't have sources for the information we want to cite the article for? Oh dear! -- Donald Albury 00:36, 28 February 2007 (UTC)

Exactly. Of course, tertiary sources have tremendous value--a well-written science textbook is probably a better source for information on general relativity than are the papers of Albert Einstein, as the former aggregates and arranges the information into a cohesive whole--but for many topics, well-written tertiary sources don't exist. Or the only well-written tertiary source is Wikipedia. (Or a Wikipedia article exists, but it isn't well-written). --EngineerScotty 01:02, 28 February 2007 (UTC)

A Wikipedia talk page, for example, is a reliable primary source about itself. Usually this is useless, because the page isn't notable. But if it is notable: if, for example, an inflammatory comment on a Wikipedia talk page is widely quoted in newspapers and therefore causes an international incident, someone later writing a Wikipedia article about the original cause of the incident could (in my opinion) use the original comment on the Wikipedia talk page as a reliable reference -- not as a reference to verify claims made on that talk page, but to verify the wording, date and time of the notable comment. It's reliable because we assume nobody tinkers with the page histories. --Coppertwig 02:24, 4 March 2007 (UTC)

Re the original question above: Wikipedia articles are not to use other Wikipedia articles are sources, in order to avoid circularity. This does not mean that Wikipedia is not a "reliable source" to people who consult Wikipedia for information. We should not give out the impression that Wikipedia does not consider itself a reliable source. For the more well-developed articles, Wikipedia may be a better source (in some ways) than any other, because it is based on several other sources and has been extensively discussed. It still should not be used as a source for writing other Wikipedia articles. --Coppertwig 02:30, 4 March 2007 (UTC)

There may be rare exceptions, but I would think even in this case it would still be a good idea to require using the media sources as primary information. If enough information didn't get into secondary sources to be worth writing an article, this would be some indication the phenomenon is ephemeral and not encyclopedic. The requirement that there be enough information in reliable secondary sources to support and verify an article also offers a check on media self-centeredness -- the tendency of a medium, such as Wikipedia, to focus unduly on itself. The requirement of secondary sources helps us avoid unduly encouraging personalities who might use Wikipedia to create rather than merely report events. --Shirahadasha 13:03, 4 March 2007 (UTC)

The existence of the media articles is needed to establish notability, but the verification of the facts (what was the wording of the actual original talk page comment that started the whole thing?) would be best done in this case by referring to the original talk page article. --Coppertwig 17:41, 4 March 2007 (UTC)

Someone should then check Essjay, as it is quoting user talk pages. -- ReyBrujo 17:43, 4 March 2007 (UTC)

No problem. That quote from a user talk page is a self-published comment (published on a Wikipedia talk page) by a notable person and presumably can be verified to have been written by that person. That's equally as valid as using a notable person's personal webspace blog as a reliable source: that is, it can be used as a reliable attribution of what the words were that the person said, not necessarily as a verification that what they said was true. --Coppertwig 01:16, 5 March 2007 (UTC)

Yes problem. The problem is that the attribution policy and FAQ explicitly rules out all uses of Wikis as sources. No exceptions are given such as "wikis are okay as self-published sources when other self-published sources would be accepted". When I tried changing it so that it did say that, it got reverted, and people above are fiercely arguing against it for what seems to me to be spurious reasons. Ken Arromdee 16:26, 5 March 2007 (UTC)

Skeptic Magazine not a reliable source?

I have been "informed" by two non-admin editors that Skeptic Magazine is an "Unreliable source", "not a reliable source" in this article [3]. Any opinions pro or con? The dispute is about this article: When Scholars Know Sin: Alternative Religions and Their Academic Supporters by Prof. Stephen A. Kent and Teresa Krebs. --Tilman 12:15, 4 March 2007 (UTC)

I'd say it is reliable, but you have to be careful about potential bias, as it has a stated and all-guiding political goal. I'd make sure I attributed opinions explicitly to the magazine, but see no reason it cannot be used. JulesH 12:48, 4 March 2007 (UTC)

These articles are clearly in the nature of editorials, so even in the absence of other issues they would be reliable for purposes of offering the opinions of its authors. --Shirahadasha 12:56, 4 March 2007 (UTC)

Correct, they are the opinions of their authors. However, in the article they are treated to support factual claims. Anyways, Skeptic Magazine is by itself not a relevant source, since its readership is confined to a narrow portion of the population. Fossa_?! 18:13, 4 March 2007 (UTC)

So is every magazine. There is no magazine that is read by a large portion of the population. The more scientific a publication is, the less read it. This argument is silly to the extreme. --Tilman 19:11, 4 March 2007 (UTC)

I have now been informed that the Skeptics Society is an activist organisation, like the KKK: [4]. Life is full of surprises. --Tilman 17:11, 4 March 2007 (UTC)

Well, undoubtedly Skeptics Society is an activist organization, they are an interest group, just like the KKK or NAACP is. That's, frankly, a banality. Fossa_?! 18:17, 4 March 2007 (UTC)

I would argue that it is an organization that clearly has as specific viewpoint. As such, we should not create a blanket statement such as "this is a reputable source" or this is not one. Each edit needs to be evaluated within its context. Please discuss in that article's talk page and find common ground, and if you can't, please pursue dispute resolution. ≈ jossi ≈ (talk) 18:46, 4 March 2007 (UTC)

I'm not experienced enough to try this, and it would probably cost more weeks of work. If the skeptic society is activist about something, then it is to be pro-science. It could as well be said that the APA is an activist organisation (pro psychiatry), thus the "APA monitor" would no longer be a reliable source per WP:Fossa. This is simply a bad faith attempt to keep out something by Wikilawyering, instead of doing research to find articles with a different point of view.

I also consider the comparison with the KKK very offensive, because this also associates me, a longtime occasional reader of their publications, with the KKK. --Tilman 19:08, 4 March 2007 (UTC)

As said many times, and as clearly stated in this policy: How reliable a source is depends on context; what is reliable in one topic may not be in another. Rather than engage in edit wars there, or bring your dispute here, please discuss in the talk page of article you are editing rather than here, and follow dispute resolution if you are unable to find common ground. ≈ jossi ≈ (talk) 19:33, 4 March 2007 (UTC)

Ok, understood, EOD. I'm outahere. --Tilman 19:43, 4 March 2007 (UTC)

The publishers of Science (journal) are the American Association for the Advancement of Science, which "defends scientific freedom, encourages scientific responsibility and supports scientific education for the betterment of all humanity." They are thus an activist organization, and many of its presidents have been controversial advocates of various viewpoints, such as evolution, medical experimentation on animals, eugenics or the development and use of nuclear weapons. Yet the journal is a highly reliable source. Nature (journal) is a similar respected source, of which its article says "The relatively progressive, controversial nature of the journal’s first articles and writers may have contributed to its success, as many early publications included evolutionary theory and Darwinism, at the time a divisive issue due to its radical nature and its religious implications." Its mission statement shows its activist viewpoint: "It is intended, FIRST, to place before the general public the grand results of Scientific Work and Scientific Discovery; and to urge the claims of Science to a more general recognition in Education and in Daily Life; and, SECONDLY, to aid Scientific men themselves, by giving early information of all advances made in any branch of Natural knowledge throughout the world, and by affording them an opportunity of discussing the various Scientific questions which arise from time to time." Lancet (journal) is another highly respected reliable source, whose article says "The Lancet takes a stand on several important medical issues - recent examples include criticism of the World Health Organization, rejecting the efficacy of homeopathy as a therapeutic option and its disapproval of Reed Elsevier's links with the arms industry." These and other respected scientific journals dating back to the 17th century publications of the Royal Society would probably endorse the stated viewpoint of the Skeptics Society: "The Skeptics Society is a nonprofit, member-supported organization devoted to promoting scientific skepticism and resisting the spread of pseudoscience, superstition, and irrational beliefs." Thus there is no basis for rejecting the Skeptical Inquirer as a reliable source due to its organization having a viewpoint favoring science, unless our mission is seen as spreading and promoting pseudoscience, superstition and irrational beliefs. Edison 22:20, 6 March 2007 (UTC)

My compliments

Anything that creates fewer, more easily understood and less "secret" rules is just wonderful. Wikipedia needs a small number of policies that are as solid as a rock, well known and easily understood, not a proliferation of policies, guidelines and essays with varying degrees of adherence and enforcement. This consolidation is a huge step in the right direction. Thank you to all who worked on it for the past four months. Dino 18:26, 4 March 2007 (UTC)

Personal Experience and Knowledge as a Primary Source

The idea that I know something personally and am able and willing to share it with the world obviously does not make that knowledge acceptable under the attribution guidelines, especially if I am not an acknowledged expert in that field and may not have previously published the knowledge in question. But doesn't that very policy exclude huge bodies of knowledge that might otherwise be lost? I can state with absolute certainty the name of the High School and College that I attended. This information could be attributed to me, but would it be verifiable and therefore acceptable? Probably not, on the basis that I could just as easily lie about the school names. But what about useful information. and maybe just some trivia, that has no source other than a person's memories? Memories (even mine) are sometimes faulty, but more often (I believe) they are real and reliable, and as such, should not be lost to the world. I've contributed to a few articles with items that I know to be true, though I can't necessarily come up with a source. Some survive, others have been taken down by editors. This stings a bit personally, and while I laud the goals of Wikipedia, I would still like to be able to contribute what I KNOW, and not just what I can cite. I think one of the great strengths of Wikipedia is that people who KNOW things about a given topic can share that knowledge easily and freely. To tap into that huge treasure of knowledege requires a certain lowering of the bar with regard to academic credentials, as some here have already suggested, and some degree of trust. -- Esjones 22:36, 4 March 2007 (UTC)

If the knowledge is that important, a reliable source will report it. If there is no source, we can not and should not include it. Blueboar 22:46, 4 March 2007 (UTC)

That is the very crux of why we need the attribution policy - "probably" can't cut it because it'd be impossible to separate fact from fiction, without having in-depth knowledge about the subject. But as a general encyclopedia, readers shouldn't need that in-depth knowledge. While it seems easier to just add what you know, at the end of the day it's actually more work, in terms of maintenance and the separation of fact from fiction - but it isn't done by whomever adds what they know, so it seems easier to them. If we didn't have mechanisms for separating fact from fiction, we'd just be another message board. ColourBurst 03:53, 5 March 2007 (UTC)

What you say makes enough sense that that is indeed how wikipedia was run at first. But at some point what one person knew to be true conflicted whith what another person knew to be true and revert wars started. OK so revert wars are outlawed. But people knew they were right so they brought in friends and battled. what to do? what to do? They tried to convince each other. With sources. And so here we are. You can add stuff without a source. But if someone wants to remove it, what can you say? That you know the truth and they don't? WAS 4.250 03:57, 5 March 2007 (UTC)

There are other wikis that have more lenient policies related to sources, such as Wikiinfo. Maybe that wiki would be more suitable for your knowledge. ≈ jossi ≈ (talk)

image manipulation

Existing text:

Manipulated images should be prominently noted as such. If they are noted as manipulated, they should be posted to Wikipedia:Images for deletion if the manipulation materially affects the encyclopedic value of the image.

This seems to be too strong. Some image manipulation, such as brightening an image to make it easier to view (without in any way altering the information provided by the image), or cropping an image to remove things honestly irrelevant to the article, should be allowed and should not automatically need to be noted. Things like that are a normal part of moving a photo from a camera to a computer (and in fact no raw image from a camera is suitable for Wikipedia without some manipulation such a scaling). I also question "materially affects" (should be "materially degrades"?) and finally "encyclopedic value" is a weasel phrase. I think the purpose of this part of the policy is to prevent image manipulation that constitutes original research or bad faith, such as adding or removing objects from the image. Here is a proposed replacement, based slightly on Reuters' policy.
Proposed text:

Minor adjustment of photographs, such as scaling, cropping, and tweaks to brightness and hue are permitted provided the resulting image does not materially differ in content from the original. That is, the new image must convey the same relevant information to the reader as the old image. Greater manipulation, such as addition or deletion of objects, unrealistic alterations of color or lighting, or blurring to disguise something in the image, are not generally permitted. All such manipulations must be noted by the uploader, who has the onus of justifying them.

--06:05, 5 March 2007 (UTC) As an example of a plausible exception: if I take a photo of my dog to illustrate dog, then I might use Photoshop to remove the table leg that sticks into the edge of the frame. That should be ok so long as I don't alter the dog, but I should still have to mention the alteration when I upload the image. (Major news organisations would not allow this example, but I think we should. Only the dog in the photo is actually relevant to the article.) --Zero^talk 06:05, 5 March 2007 (UTC)

I'd be happy with this change. It's a nice and simple statement, and seems to achieve everything required of it. JulesH 14:40, 5 March 2007 (UTC)

Let's put that detail in the FAQ. Let's keep the wording in the policy as simple and as succinct as possible. ≈ jossi ≈ (talk) 17:03, 5 March 2007 (UTC)

text needs clarification

"Deductions of this nature should not be made if they serve to advance a position, or if they are based on source material published about a topic other than the one at hand." --- What does the last part of that mean? What scenario is being envisaged here? I honestly don't know. I asked about this once before, but nobody offered an explanation; can someone explain it now? Better still, can someone reword it so that the meaning and motivation is clear? --Zero^talk 06:18, 5 March 2007 (UTC)

Does Wikipedia talk:Attribution/Archive 11#Suggestion for OR example to replace A + B = C example help? "Someone adds sourced data that the witness said it was a moonless night. Someone else adds a statement sourced from the almanac that there was a quarter moon. Someone else adds that that there was heavy cloud cover blocking the moon sourced from a local newspaper. Someone else adds a statement sourced from a rival newspaper that the first newspaper's weather reports are unreliable. NOR shuts this nonsense down." WAS 4.250 16:22, 5 March 2007 (UTC)

That example certainly breaches the OR barrier, but how is an ordinary editor going to read this policy sentence and know what things are supposed to be excluded by it? Policy which nobody can understand is worse than no policy. I propose to delete the words "or if they are based on source material published about a topic other than the one at hand". --Zero^talk

That's been there a long time, Zero, and is an important part of the policy. I think it's fairly clear what it means. It means don't use sources to make a point about X unless the sources are talking about X. SlimVirgin ^(talk) 23:01, 5 March 2007 (UTC)

Unfortunately, it doesn't really make sense though. What it seems to unintentionally say is that there are two absolute violations: 1. Where you're advancing a position, or 2. Where you're taking from something which is talking about something else. The second one can't have been intended, though. Why would that be worse than any other synthesis, if it isn't to advance a position? Would you solve the problem then by simply desourcing the statement? If that rule needs to be there, it should simply be placed in the general discussion, and probably better explained. Mackan79 23:39, 5 March 2007 (UTC)

Of course it makes sense, and the example in the article is a great illustration (the plagiarism example). As soon as you allow people to bring in sources that aren't dealing specifically with the topic of the article, they'll pull in sources from all over to build their arguments - the very definition of OR. Jayjg ^(talk) 00:22, 6 March 2007 (UTC)

Mackan, if I weren't such a good-natured creature, I might suspect you were following me. SlimVirgin ^(talk) 02:12, 6 March 2007 (UTC)

I know how you feel :P No, I was actually looking at this earlier, but held off on commenting so as not to attract attention... foiled twice, unfortunately. Further context, if you like, is here. Still, I'm not saying this point shouldn't be made, simply like Zero, that it seems to have been misplaced. Mackan79 21:32, 6 March 2007 (UTC)

I don't have any problem with the section containing the plagiarism example, but the text I quoted is from a different section. It is in reference to "straightforward mathematical calculations or logical deductions". The plagiarism example doesn't work there. What would be an example of an otherwise-acceptable "straightforward mathematical calculation or logical deduction" that is excluded by this rule? Suppose I'm editing Mormons. If I had a source that says "over 10% of New York's 18 million people are Mormons", then I think I'm allowed to write "there are more than a million Mormons in New York". Now suppose my source only says "10% of the people in NY are Mormons", and I also have a book on New York politics that gives the population as 18 million. Am I not then allowed to write what I want because the book is not about Mormons? What I'm questioning is whether this clause is required in this section. --Zero^talk 09:30, 6 March 2007 (UTC)

I think you're on shaky ground there; you have no idea if the author of that book was looking at similar figures for New York's population as you found when they made that claim. Jayjg ^(talk) 22:34, 9 March 2007 (UTC)

The Register

I'm sure I have seen the answer around somewhere but can someone confirm whether or not The Register is a reliable source? Thanks Spartaz ^Humbug! 20:55, 5 March 2007 (UTC)

Be cautious with them; they often publish unsubstantiated rumours, but they do meet the absolute bare minimum requirements of not being self published and not falling into the 'questionable' category. JulesH 21:08, 5 March 2007 (UTC)

Thanks for that JulesH. Would you agree that they would be a dubious single source for an subject? Talk:Irish passport the section on Shielding is what I'm wondering about. The other sources don't address the point specifically and most of the bit therefore looks like original research. Thanks. --Spartaz ^Humbug! 21:32, 5 March 2007 (UTC)

I have replied on the article's talk page. JulesH 21:56, 5 March 2007 (UTC)

Copying of information from Village Pump

The Village Pump archive says that material is deleted after 9 days, and that before that time material can be moved to the relevant talk pages. I copied to here some discussion which relates to whether there is consensus on replacing WP:V and WP:NOR with WP:ATT. Someone has deleted it. Please put it back. We need a record of what happened. It contains significant comments related to the discussion. It can be moved to the archive page of this talk page, not deleted. I do not see anywhere a consensus for the replacement of the old policies with the new policy. I see disagreement. --Coppertwig 23:04, 5 March 2007 (UTC)

I was the one who removed it, see my edit summary, feel free to replace it though, but adding this discussion seems kind of pointless. Garion96 (talk) 23:26, 5 March 2007 (UTC)

There is a definite point in copying over that discussion: it relates to the question of whether there is or was ever consensus for getting rid of "verifiability". In that discussion, a user comments on that very issue. --Coppertwig 13:21, 6 March 2007 (UTC)

Clarification of "in relation to the topic of that article"

In the "What is original research?" section, there is the following definition: "introduces an argument without citing a reliable source who has made that argument in relation to the topic of the article". The "in relation to the topic of the article" is unclear to me, and could be a point of considerable acrimony in many pseudoscientific topics. Does the statement require that the source make the argument in direct and explicit relation to the topic of the article, or does it only require that the argument be in relation to the topic of the article implicitly or as a member of a general class?

A very good example of this distinction may be found in the topic of perpetual motion devices. There are some such devices which are notable, but no one in the scientific community bothers to write a source calling each one pseudoscientific and rejected by the scientific community: perpetual motion machines are considered to be such by default, as can be supported by many sources. If the definition covers implicit relation, then it would be acceptable to include sources that label all such devices as pseudoscience on pages about particular devices. If, on the other hand, it requires explicit relation to the specific topic, then most articles on perpetual motion devices would not be able to include any criticism, despite that criticism being obvious.

If I am correct in believing that only implicit or class-based relation is required, then the definition needs to be clarified, or else I will have to argue this point with every new nut who comes to Wikipedia. And of course, if explicit relation were required, then the policy would likely conflict with the Pseudoscience ArbCom decision, particularly the Serious Encyclopedia finding, and there would be no point in me or many other editors continuing to edit, since making pseudoscience articles follow NPOV would be impossible. --Philosophus ^T 23:14, 5 March 2007 (UTC)

Here is my take. If Mr. X invents a device Y, and we have a reliable source classifying Y as 'perpetual motion machine', then we are allowed to wiki-link to perpetual motion, as that is a standard well-defined concept. Crum375 23:20, 5 March 2007 (UTC)

But can we also cite reliable sources that argue that perpetual motion devices violate basic laws of physics? Those sources would not be related to the specific topic, but would be related by the perpetual motion source. --Philosophus ^T 23:22, 5 March 2007 (UTC)

No source that alledges they have built a perpetual motion machine is reliable except with respect to article about that specific source, or about the author of that specific source. Hipocrite - «Talk» 23:25, 5 March 2007 (UTC)

IMO all we would need to do is to link to perpetual motion (PM), where the reader can read more about it. I would stay away from further addressing and citing sources about PM in the article about X, as we would then be creating a mini-PM article inside X's entry, thereby implying that we as Wikipedia editors agree with the PM classification, which would be OR-ish. Crum375 23:31, 5 March 2007 (UTC)

What about having something like "According to X, the device is a perpetual motion machine, which if true would mean that it violates commonly accepted physical laws"? My current application of this is with Transcendental Meditation, which includes a bunch of mind-qft type pseudoscience. The material was discussed in What the bleep do we know, and there are many reliable sources calling everything in that movie nonsense scientifically, but no one has bothered to explicitly address the theories espoused by TM supporters, since the general material has already been addressed many times. If such sources cannot be used, then the paid TM supporters editing the page are essentially free to portray the scientific community as having no objections to the ideas. --Philosophus ^T 23:37, 5 March 2007 (UTC)

My own personal inclination, without having examined the TM article, is to stay way from making the connections. IOW, I think that saying "According to X, the device is a perpetual motion machine, which if true would mean that it violates commonly accepted physical laws" is OR-ish, whereas "According to X, the device is a perpetual motion machine" is not. Now odds are high (I'm guessing here) that X also said something else besides classifying the device as PM, and that would of course be citable and most useful. So for example, "According to X, the device is a perpetual motion machine and violates commonly accepted physical laws[1]" would be fine if X said it. I think that we need to take a step back, and as much as we personally believe something (and I'll admit here that I personally, along with the USPTO, don't believe in PM), we need to stick to our rules and stay as neutral and OR-free as possible. I think that if done properly we can send a message to our readers that we as WP are bending over backwards to be neutral in all matters, and that everything we say is based on reliable published sources. Crum375 01:00, 6 March 2007 (UTC)

Is an expert posting to a moderated mailing list reliable?

If an expert in the field of radioactive airborne hazards faced by people in the Navy posts an email message to a moderated mailing list on radiation protection claiming that someone else was given an award by the Navy for making a decision about protection from airborne radioactive hazards, isn't the poster a reliable source? His signature is certainly verifiable. James S. 23:16, 5 March 2007 (UTC)

While I strongly oppose the use of such, the main reason that there are big gaping holes in our policy with respect to exeperts posting on unreliable sources is because the creator of Babylon 5 likes to write to usenet, and the fans of that series had the clout to water down the guideline for the entire encyclopedia. As such, I would argue that if you can be absolutly certain the poster is who he says he is without even the slightest doubt, then it would be a reliable source. I have not reviewed the link in question for questionability of identitiy. Hipocrite - «Talk» 23:21, 5 March 2007 (UTC)

In principle, no. Email messages are not considered reliable sources, unless described in a third-party published source that meets the requirements as presented in this policy. ≈ jossi ≈ (talk) 23:28, 5 March 2007 (UTC)

As the policies current stand, it boils down to relying on IAR. If you think it makes wikipedia better, you can do it, but if challenged you must rely on evidence, logic and the ability of others to see your point rather than blindly follow non-IAR policies. This is actually sort of standard in much of our pop culture area. WAS 4.250 00:49, 6 March 2007 (UTC)

I'd like to modify this issue a bit by stating that there are *some* "moderated email lists" which are archived in a permanent location. For example I happen to belong to a genealogy list where experts in medieval genealogy post messages to each other about questions or new things they've found. All the email messages are archived online, and available for anyone to peruse, in their own time. So in effect the "list" becomes a bulletin board of a sort, although you can't reply directly through the archive. In a case like this I believe we'd treat the archive as the source and link to it directly, rather than saying "Email list so-and-so" we'd have to say "as archived at link". Email lists have a verification problem, but if they are archived, that problem disappears and we're left with the lower bar of reliable. Wjhonson 00:56, 6 March 2007 (UTC)

Surely this comes down to being a document self-published by a professional researcher, i.e. it is an accepted exception? JulesH 06:40, 6 March 2007 (UTC)

James & I have an ongoing disagreement about this. IMHO, the researcher exception doesn't apply. If you follow the link, the researcher (Riel) isn't being offered for information within his field, he's being offered to establish that a third party (Taschner) recieved an award. This use is specifically forbidden by WP:ATT, in bold face, no less. Specifically, the relevant section states: "Self-published sources, such as personal websites and blogs, must '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." As a general matter, I don't think the professional researcher exception applies to personal facts about other third parties. Let's say that Riel, a nuclear physicist, writes an e-mail saying that Taschner, another nuclear physicist, once won an otherwise unverifiable award. Why is that statement any more reliable than the same statement on Taschner's wife's blog or his son's myspace page? TheronJ 15:00, 6 March 2007 (UTC)

The RADSAFE mailing list is moderated, and there are specific exemptions for moderated mailing lists in the policies you cite. Why do you say that Taschner's award is not information in Riel's field? Earlier you implied that you thought Riel's field was history of science -- why? If you would please phone Riel as I have asked, then you could ask him directly. James S. 18:22, 6 March 2007 (UTC)

If the poster's identity were not verifiable, a mailing list posting would be an anonymous source. If the mailing list is archived and the poster's identity is verifiable, I would think that would fall under the cautions/restrictions on self-published, non-fact-checked sources. (Of course, if the email is later republished in a fact-checked, reliable publication, it's a reliable source.) Seraphimblade ^{Talk to me Please review me!} 18:39, 6 March 2007 (UTC)

When did this happen?

And can I still cite "No Original Research" in argument for an action? Blueaster 23:50, 5 March 2007 (UTC)

Sure. Use WP:NOR. Jayjg ^(talk) 00:27, 6 March 2007 (UTC)

Web-based fora

I'm inclined to use a web forum as a primary source for a language-usage article. While I would understand an objection if I was attempting to use a forum post to justify the addition of a "fact" such as "the world's biggest ball of twine is located in Joe Franklin's basement", since the issue is one of common language usage, a forum should serve as a primary source, as it records actual conversations. Any thoughts? --Dante Alighieri | Talk 01:24, 6 March 2007 (UTC)

It's definitely a primary source. The problem is "for what?" I think the only thing that a forum can be used for is a primary source about the speech habits of that forum - you couldn't generalise it, internet fora are too diverse. ColourBurst 01:54, 6 March 2007 (UTC)

Conducting surveys is original research. In most instances, it would be necessary to survey many posts to many web forums to extablish language usage, and such a survey would be original research. --Gerry Ashton 02:09, 6 March 2007 (UTC)

Hmm, you're right, as soon as you move out of quoting a singular instance to generalizing speech patterns of a forum, that would be original research. I still stand my assertion that you couldn't generalise it for any forum even if you had the data available in the forum somehow. ColourBurst 02:54, 6 March 2007 (UTC)

In many cases it is necessary to see how things are referred to in many newspapers and books in order to establish use. Looking a web sites and reading are similar. This is the reasoning that would have us describe novels from book reviews. Drawing obvious conclusions from data that is in front of our eyes is not OR. DGG 06:17, 6 March 2007 (UTC)

I understand the hesitancy, but it's not as if I'm actually generalizing from the forum to society at large. I happen to "know" about the usage of the terms (as do many people, they are common enough terms), I'm just looking for something that I can cite to "prove" that the terms are in use. In that limited sense, wouldn't a forum be appropriate? I think it's a bit much to ASSUME that the usage is unique to a given forum. --Dante Alighieri | Talk 06:26, 6 March 2007 (UTC)

I've certainly done this myself before, although I wasn't comfortable with it at the time. Yes, a web forum or any other self-published document is a primary source that a particular word is in use (and, in many cases, that it was in use at some particular time). There are almost always better ways of achieving this, however. For instance there are a number of web sites that catalogue new language usage which are self-published by professional researchers in the field (e.g. languagelog, double-tongued word wrester) which may be more appropriate sources. JulesH 06:38, 6 March 2007 (UTC)

Case in point: WP:ATT

(Copied from Wikipedia:Village pump (policy)/Archive where apparently stuff is deleted after 9 days. Please do not delete this information. You may move it to the archive page of Wikipedia talk:Attribution. See Wikipedia talk:Attribution#Copying of information from Village Pump --Coppertwig 13:26, 6 March 2007 (UTC))

Talking about mergers... I'm sure many of you have seen WP:ATT, but it's now ready to replace WP:V and WP:NOR. It has already been tagged as policy, so please have a look at the talk page for the deployment plans. --Merzul 04:28, 16 February 2007 (UTC)

Of course I absorbed a major shock when I viewed the page, but it's wearing off, and I kind of like the idea. Just make sure to note, on that page, that Wikipedia is a tertiary source, and perhaps give the definition of a tertiary source. That's all; good job, people who worked on it. Gracenotes^T § 05:47, 16 February 2007 (UTC)

And perhaps if WP:V redirects to it, mention the word "verifiability" at least once in the policy? Gracenotes^T § 05:48, 16 February 2007 (UTC)

NOTE WP:V and WP:NOR ARE NO LONGER POLICY

(Copied from Wikipedia:Village pump (policy)/Archive. Please do not delete this information. You may move it to the archive page of Wikipedia talk:Attribution. See Wikipedia talk:Attribution#Copying of information from Village Pump --Coppertwig 13:26, 6 March 2007 (UTC))

PLEASE NOTE that Wikipedia:Verification and Wikipedia:No original research have been merged and replaced by Wikipedia:Attribution. This may come as a surprise to some, as this merger was not well advertized when in the proposal stage. From what I can tell, the folks who worked on this new Policy did a fairly good job (although I do wish they had announced their intentions sooner and more frequently). I do not see any substantive changes from what WP:V and WP:NOR had to say. Editors may also wish to note that several of the editors working on that page have expressed the intent to subsume the guideline: Wikipedia:Reliable sources into this new Policy and replaced with a FAQ.

While I actually approve of this merger (one Policy instead of two eliminates the potential for contraditary statements where they overlaped), I am not happy about how it was done. Policy changes should have broad community consensus, and I don't see how the editors who created WP:ATT can claim to have this when it comes as a bit of a surprise. The intent should have been shouted from the roof tops several weeks ago to give everyone LOTS of notice.

So... just so everyone understands, let me repeat this in loud voice: WP:V and WP:NOR ARE NO LONGER POLICY. They have been replaced by WP:ATT. ALSO - IT IS PROPOSED THAT WP:RS BE REPLACED with a combination of WP:ATT and a FAQ. (please don't kill the messenger.) Blueboar 17:51, 24 February 2007 (UTC)

None of these pages has ever had "broad" consensus. They've been written and decided upon by a small number of users who care a lot about the issues, and those who aren't interested in constant fighting and wikilawyering are outvoiced. WP:RS isn't that bad anymore, but I have yet to see it used for anything but abuse. — Omegatron 18:00, 24 February 2007 (UTC)

can't say I agree with that. Without WP:V or WP:NOR, Wikipedia would fall apart in a matter of weeks. It really all boils down to WP:ENC, but a lot of people do need this spelled out in greater detail. dab (����) 18:06, 24 February 2007 (UTC)

well, it is misleading to say that WP:V and WP:NOR are "no longer policy". They still are. They were just merged into a single page. It is an extremely bad idea to keep these pages around as "inactive", since they are not. If people feel they must be merged into a single page, make them redirects along the lines of WP:UNDUE. dab (����) 18:04, 24 February 2007 (UTC)

I've tagged both of the pages back to policy status. Consensus indicates that they should be merged, but this doesn't mean they should be "rejected" as policies and no longer be in use. I concur with Dbachmann, please make them redirects instead. Michaelas10 (Talk) 20:26, 24 February 2007 (UTC)

FYI, I did not mean to say that the concepts behind V and NOR were being rejected... Dbachmann is correct to say that is misleading. The point of this was really to draw attention to the merge, and to let editors know that the merge was a done deal. For instance, if an editor is in the midst of a content dispute or an AfD that centered on V or NOR, they need to know that they should point to ATT now instead of V or NOR. This post was really a "shout from the roof tops" to advertize the merge rather than a complaint about it. Blueboar 03:05, 25 February 2007 (UTC)

• Please note that I supported this policy and the merger. My only issue was with timing and how much notice was given. Looking back, I have to say that there was consensus for it - and that consensus is growing daily as more and more people come by and actually read what the policy says. Is it perfect, no... but it is a damn sight better than having multiple contradictary policies and guidelines. Blueboar 14:03, 6 March 2007 (UTC)

Poll re "verifiability, not truth" versus "attributable ... not whether it is true".

Please state whether you support or oppose changing the wording of the policy from "The threshold for inclusion in Wikipedia is verifiability, not truth" to "The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true." --Coppertwig 13:37, 6 March 2007 (UTC)

1. Oppose for the reason explained in "Role of truth" above. --Coppertwig 13:37, 6 March 2007 (UTC)

• No problems with the "attributable" language. It means the same thing to me. Blueboar 13:52, 6 March 2007 (UTC)
• Comment The wording of this poll is somewhat confusing. The change described has already occurred, so to oppose would presumably be interpreted as to support changing it back.
• Support The policy name has change from verifiability to attribution because many editors were confused about what verifiability meant as a term of art on Wikipedia and were interpreting it in its natural meaning. The original intended meaning of the phrase is more clearly evident in the current phrasing than it was in the old one. JulesH 14:01, 6 March 2007 (UTC)
• Oppose: "Verifiability, not truth," has a long history on Wikipedia, and the new language doesn't present much improvement, if any. TheronJ 15:12, 6 March 2007 (UTC)
• Support the change that already gathered consensus Yes, virginia, this was already decided upon. No, I'm not going to go dig through the archives to look for specific discussion; I had this page watchlisted from almost the first day it was drafted months ago. And I continue to believe that this is a superior wording for presenting the old meaning. GRBerry 18:08, 6 March 2007 (UTC)
• Support. This already has broad consensus. SlimVirgin ^(talk) 01:09, 8 March 2007 (UTC)

Poll re handling of apparently false, but attributable, statements

Suppose a Wikipedia article makes an assertion which is attributable to a published source normally considered reliable, but which the editors of the page are convinced is false, and suppose they can't find an acceptable published source to support the other position. Do you support the following:

It is better to either delete the statement (if it is a relatively minor point) or change it to indirect speech than to leave a Wikipedia article making an assertion the editors believe is false.

An example of changing to indirect speech is to insert "The Daily Times reports that..." --Coppertwig 13:49, 6 March 2007 (UTC)

1. Support The purpose of Wikipedia is to provide a useful reference, that is, to attempt to provide information which conforms to reality as often as possible. --Coppertwig 13:49, 6 March 2007 (UTC)

Coppertwig, this page is getting hit by six threads a day. This is silly. Marskell 13:51, 6 March 2007 (UTC)

• Oppose this is best dealt with by the editors themselves, rather than being prescribed in policy, IMO. Note that there is no policy that requires the inclusion of all material which is attributable. JulesH 14:03, 6 March 2007 (UTC)

• Clarification This poll is not proposing adding that sentence into policy. --Coppertwig 14:17, 6 March 2007 (UTC)

Then what is the Point of it? Oh... never mind, I think I answered my own question. Blueboar 14:26, 6 March 2007 (UTC)

• Support the statement, but if it's not going to be added to the policy, fail to see the point of the poll, per Blueboar. --AnonEMouse ^(squeak) 14:52, 6 March 2007 (UTC)

• Comment For the reason why this poll is happening, see section "Role of truth" above. --Coppertwig 14:57, 6 March 2007 (UTC)

Polls are evil. Do we really need to run a poll on this? I don't think so. ≈ jossi ≈ (talk) 14:59, 6 March 2007 (UTC)

If you want to discuss this subject, let's discuss, but please do not use polls. They are not helpful. ≈ jossi ≈ (talk) 15:00, 6 March 2007 (UTC)

I want to discuss it. Please. In "Role of truth", above. Poll continues, however. --Coppertwig 15:08, 6 March 2007 (UTC)

Discussions are the way to go. Polls in these cases are not useful. ≈ jossi ≈ (talk) 23:33, 6 March 2007 (UTC)

The policy is very clear: We do not make assessments about the truth or falsehood of the assertions made in reliable sources. Period. If editors do not wan t to use a specific source, for whatever reasons, and they reach consensus about not using that source, don't use it. Simple. ≈ jossi ≈ (talk) 15:03, 6 March 2007 (UTC)

• IMHO, it's a terrible idea to suggest that editors should delete attibutable content that they personally believe to be false. It will lead to edit wars like you have never seen, and to POV warriors scouring the encyclopedia of everything that offends them. I would be open to a policy addition that it is acceptable to delete everything that TheronJ believes to be false, but otherwise, it wouldn't be good for the encyclopedia. TheronJ 15:11, 6 March 2007 (UTC)

Some common sense is called for. If the statement is really Fringe, I see no problem removing it if that is the clear consensus of those editing the article. However, if there is the slightest doubt, it is much better to keep the statement - but make it a statement of opinion as opposed to a statement of fact ("According to the author Ima Loony...."). Blueboar 15:22, 6 March 2007 (UTC)

Support. It is sometimes appropriate to move something which is sourced but silly to the discussion page rather than leave it on the project page. No one objected when in the article Talpiot Tomb which refers to the "lost tomb of Jesus" I removed from the article to the talk page the sentence from the Discovery Channel's Discovery News ""Frank Moore Cross, a professor emeritus in the Department of Near Eastern Languages and Civilizations at Harvard University, told Discovery News, ”The inscriptions are from the Herodian Period (which occurred from around1 B.C. to 1 A.D.). " This is silly and somehow garbled or a typo in the source, because there were two "King Herods" and our article on them gives neither such a short and amazingly coincidental time span, The professor likely said first century to second century. If a Nobel Prize chemists says that water generally boils at sea level at 10 degrees celsius and the sun rises in the west, it does us no good to leave such a misstatement of fact in an article. Move it to the discussion page and get more input. Leaving a silly misstatement on the article page and then "balancing" it with a statement from a science book that water boils at 100 celsius and the sun rises in the east still leaves a silly article to no point. Typos and verbal flufs do not add to the knowledge base expected of an encyclopedia unless the article is intended to be a humorous collection of missprints and misstatements. Edison 22:30, 6 March 2007 (UTC)

Comment - With the caveat that such a restriction is *only* applicable to articles *other* than the article about the source itself. It is *always* acceptable for example to quote the National Enquirer on it's own article, no matter what the quote says. For example: "The National Enquirer is known for claims such as: 'Three-headed baby born to Jessica Simpson!'" We should not put any restrictions on claims made by a source, in the article *about* the source. "Adolf Hitler claimed that Jews eat babies for breakfast" would be entirely acceptable in his article. It would not be in an article about breakfast foods. Wjhonson 23:24, 6 March 2007 (UTC)

This is exactly the stuff we do not want in our policies. Editors' needs to apply their best and collective judgment in these cases. I strongly oppose such wording in the policy. ≈ jossi ≈ (talk) 23:32, 6 March 2007 (UTC)

A poll is not going to decide policy. This is a waste of time. —Centrx→talk • 23:49, 6 March 2007 (UTC)

Agree. We don't change policy via polls. SlimVirgin ^(talk) 01:10, 8 March 2007 (UTC)

The problem with allowing editors to use best judgment is that the rules don't make it clear that best judgment is allowed. A number of editors *right here on this page* have suggested that one may *never* remove material that is sourced but which one believes to be false, with no judgment allowed. Ken Arromdee 20:50, 8 March 2007 (UTC)

I still think a paragraph in the key principles section that tells editors to use their judgement when evaluating sources would be helpful. Basically stating that a source that's appropriate for one claim on one article may not be appropriate for another claim on a different article, and to consider the type of claim made and the type of article (i.e. whether it's a biography, scientific article or whatever) when evaluating sources. JulesH 11:16, 9 March 2007 (UTC)