Wikipedia talk:Verifiability/Archive 9

Archive 5

←

Archive 7

Archive 8

Archive 9

Archive 10

Archive 11

→

Archive 15

Common sense

Yes. Common sense starts with choosing the correct ditionary defined words. Would you agree Zak?--Light current 22:09, 16 June 2006 (UTC

No, common sense involves choosing the best (and least misleading) English we can; and then using judgment in not following it over a cliff. Dictionaries describe English; they do not control it. Septentrionalis 22:35, 16 June 2006 (UTC)

Sorry I was not talking to you!I was talking to Zak. Thanks! So Iwould appreciate it if you did not speak unitl you are spoken to! IOW: KYFGS! Thanks!8-)--Light current 23:22, 16 June 2006 (UTC)

Can you please point me to the guideline that says that only the person addressed may respond to a point? Thank you. Also, can you please translate "KYFGS?" I got zero Google hits for it. Thank you again. Robert A.West (Talk) 13:38, 18 June 2006 (UTC)

No I cant. Can you? But I was talking to Zak no one else!--Light current 16:53, 18 June 2006 (UTC)

Then use his talk page; that's what it's for. Septentrionalis 18:29, 18 June 2006 (UTC)

Yes but Zak was talking to me here!--Light current 18:45, 18 June 2006 (UTC)

Well, if it leads to discussion which resolve the point of information under discussion, then it is productive discussion and satisfies WP:NPOV and would be common sense. Terryeo 00:30, 25 June 2006 (UTC)

Slim and me getting personal

LC, I haven't followed your exact argument here, but as I recall from the last time, you're objecting that the word "verifiability" has nothing to do with truth in this policy, because "to verify" is often used to mean "to check that something is true." If that's your argument, I agree with you. Verifiability was not a good name for this policy, because we're precisely NOT trying to verify that things are true when we refer to a source; rather, we are checking only that the material has already been published by a reliable source. (I think that distinction also deals with the "with reference to" and "by referring to" distinction we're discussing above: material must be verifiable "with reference to X" which indicates this is a special use of the term. If it were "verifiable by referring to," we'd be implying the material could be verified as true, which is not what's meant. But that's another point, and I'm not expressing it very well.)

However, having said that, this is the name of the policy. It has stuck and lots of people are familiar with it. The first two sentences say what the word means in this context, so the fact that it isn't an ideal name doesn't really matter any more. SlimVirgin ^(talk) 23:46, 16 June 2006 (UTC)

OK Slim! Thanks for being frank with me. I will consider your response and reply within 24 hrs!--Light current 23:55, 16 June 2006 (UTC)

Having now read and considered your post, it seems we have similar sentiments. I understand that you have the same view as I do about the inadequacy of the term 'verifiablility'. Now the problem is that, although you and I may understand what we are talking about when we use that term, others may not. In fact a number of editors have demonstrated by their posts that they do not have the same understanding as we do. Some may in fact be inferring the dictionary definition of the term.

I can understand why you are reluctant to change the word after all this time, but if this word is to remain it must be defined properly in the policy to avoid confusion amongst users (like me). A good explanation would be that it means 'referability' which is what Ive been trying to get the policty to say without success! If the word 'verifiability' cannot be defined in the policy then it must be changed! Do you agree that we must make the policy plain and clear so that anyone (however literate) can understand it? BTW your statement wording is very clear!. I wish other editors could be as clear as you!--Light current 21:17, 17 June 2006 (UTC)

I am not so sure that the word is not appropriate. Verifiablity = if an attributed statement in an article can be verified as made by the person/organization to which that statement is attributed to, by checking the sources provided. As in "verify a claim" [1]. ≈ jossi ≈ t • @ 21:28, 21 June 2006 (UTC)

Interruption

(response to SV) Modesty aside, you have explained what you intend "with reference to" to mean quite clearly. If we want to say that, "with respect to" might be better; "with reference to" has come to be a wooly periphrasis for "about" and should be avoided. Septentrionalis 21:33, 17 June 2006 (UTC)

Thanks, Sept. I'm adding below a response to LC that would involve rewriting that sentence, so perhaps we should wait until that's replied to. SlimVirgin ^(talk) 21:41, 17 June 2006 (UTC)

This is exactly right. I think that "by referring to" is best because what is required is that edits be verifiable by going to the source and seeing that it says what we say. We could say something along the lines of "edits must be checkable against reputable sources". I think "checkable" is much more what we are looking for than "verifiable", which does have that note of truthiness about it.Grace Note 08:57, 19 June 2006 (UTC)

Back to me and Slim

Thank you, LC. I don't see any need to change the name at this stage. It has "taken," people are used to it, it's one word and easy to remember. All we need to do is explain how the term is being used. I think the current intro is clear enough, but we could try to make it clearer, for example:

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any editor must be able to verify that material added to Wikipedia has already been published by a reliable source, because Wikipedia does not publish original thought or original research.

Would that be better? SlimVirgin ^(talk) 21:41, 17 June 2006 (UTC)

FWIW, I like it. Of course, by "any editor" we really mean, "Any editor with access to a good research library and whose Internet access is not restricted," but it may be silly to emphasize that distinction. Robert A.West (Talk) 13:18, 18 June 2006 (UTC)

Not wishing to be pedantic 8-), but the definition of verifiable here is circular (ie it uses the word verify). I thought Robert would have picked that one up! I would suggest replacing that occurence of the word by the worm 'confirm'. Then the sentence would expalin what we mean and we could move on!--Light current 17:54, 18 June 2006 (UTC)

I like it, too, and agree with LC re: confirm, but perhaps also substitute editor with reader or perhaps researcher? (I prefer reader). mjb 18 June 2006 (UTC)

What about 'user'? That covers editors and readers! 8-)--Light current 19:03, 18 June 2006 (UTC)

If one were being pedantic, one would note that Slim did not supply a definition, but explained which sense of "verify" we mean, and how the word applies to the specific case. Circularity in such an instance is not always a vice. Would "ascertain" work better for you? As for "reader" versus "editor", I suppose it is true that too many editors seem not to have read the articles they edit, but I hate to encourage the practice. Robert A.West (Talk) 03:02, 19 June 2006 (UTC)

'Ascertain' would suit me perfectly Robert!--Light current 03:22, 19 June 2006 (UTC)

Re: "editor", using that term seems to make editorship a prerequisite for verification. Shouldn't verification be conductable by anyone, not just just those who are editing (or know how to edit) the article in question?—mjb 05:36, 19 June 2006 (UTC)

Youre a little behind the times mjb! THe current proposal is to say 'any reader' instead of 'editor'. Please see posts further down this page.--Light current 21:54, 21 June 2006 (UTC)

Well perhaps it is I who is behind the times. I think mjb has seen the modified proposal--Light current 21:56, 21 June 2006 (UTC)

Interjection

Obvesoily I'm to stupid to red a diktonry so I'll withraw my coment. Gerry Ashton 01:14, 18 June 2006 (UTC)

Yes but the meanings of the words are implied by their roots- are they not? Im afraid you do not understand the meaning of he words 'verify' or 'refer'. Please look them up in a good dictionary! --Light current 22:49, 17 June 2006 (UTC)

Etymology is often a blind guide for many of the same reasons that false cognates exist. Words acquire shades of meaning, and ultimately completely different meanings through usage. Consider the word, "virtue". Among other uses, this can mean "electromotive force" (UK). By reasoning similar to yours, this would be objected to on the grounds that this has nothing to do with a male human, the meaning of the root, "vir." By such reasoning, "hussy" can't mean a woman of low morals, because it is short for "housewife". Words that have gone through a process of folk etymology are landmines in this regard -- they have changed meaning because of a misunderstanding concerning origin. One needs to look at how the word is actually used. Robert A.West (Talk) 13:14, 18 June 2006 (UTC)

Yes Robert, but if you look those words up in a modern dictionary, it will give you the modern meaning. (with its derivation if its an etymological dic). The modern meaning of 'verify' and 'verifiable' is.... well look it up for yourself!--Light current 18:09, 18 June 2006 (UTC)

The OED (second edition) defines "verifiable" as That can be verified or proved to be true, authentic, accurate, or real; capable, admitting, or susceptible of verification. This is also no less circular than the present WP text. Septentrionalis 18:28, 18 June 2006 (UTC)

Yess. This will be a rather circular definition because its an adjective. But if we take out the circular references we have:

That can be proved to be true, authentic, accurate, or real.

But can you look up 'verify' in the OED (I dont have a copy) please and let us know what it says?

--Light current 18:57, 18 June 2006 (UTC)

There are about seventeen different clauses; those which are closest to the narrow meaning on which Light current insists are marked rare. Wikipedia is using 4. To ascertain or test the accuracy or correctness of (something), esp. by examination or or by comparison to known data, an original, or some standard; to check or correct in this way. Septentrionalis 19:17, 18 June 2006 (UTC)

I dont actually insist on any particular meaning except the most common one! Which is.. ?--Light current 20:08, 18 June 2006 (UTC)

Not specified: rare, like obsolete, is a warning. Septentrionalis 20:15, 18 June 2006 (UTC)

I dont actually insist on any particular meaning except the most common one! Which is.. ? But lets take the one you have quoted above.:

To ascertain or test the accuracy or correctness of (something), esp. by examination or or by comparison to known data, an original, or some standard; to check or correct in this way'

This is not the primary use of the term in the policy. THe primary meaning of the term as we use it means to check that a ref exists. This is qualified by stating we must have reliable sources (and seems to be a secondary definition).

I quote a senior admin/burocrat (*SlimVirgin) who has done a great deal of work on this policy:

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any editor must be able to verify that material added to Wikipedia has already been published by a reliable source, because Wikipedia does not publish original thought or original research.

The above is what Slim thinks it means (becuase (s)he has just said so). Now apart from the slightly circular definition, we can deduce that what Slim is saying is that:

• We need 'verifiability', where verifiability is defined in this policy to mean:

1. that material added to Wikipedia has already been published
2. that the material has been published by a reliable source

The above rule does not require that the statement in our article has been confirmed as fact, it just means that some (reliable - whatever that means) person has published something similar in a 'reliable' publication!

Now Im sure anyone can see that the WP defn of 'verifiability' (above) does not agree with the dic defs!.

So, if we are using the word in a nonstandard way, we should define what we mean by it on the policy page. I would of course rather that we did not use this word, but maybe a snowball in hell would have a better chance than that wish! --Light current 20:08, 18 June 2006 (UTC)

/me beats himself over the head for responding to this astonishingly oft-repeated (and rebutted) question), but, nevertheless...

No. Your analysis in the comment above is incorrect. SV (and the policy, and most Wikipedia editors) require that statements made in Wikipedia articles must have been published by reliable sources elswhere, before we say them. For claims where there is insignificant opposition(such as, "the sun goes around the earth"), it is acceptable to include the claims as fact. Where there is opposition, we turn the claims into statements that the claims have been made, i.e. "The company claims 1.2 million dollars in revenue in 2006.". In either case, verification means exactly what it means in the dictionary def given above - checking that something is correct by making sure it matches an external source. It is not nonstandard, and we do define what we mean by it - that's what the policy page is for. JesseW, the juggling janitor 22:04, 18 June 2006 (UTC)

Noted. So could you point out exactly where I have gone wrong in my above logic please? 8-|--Light current 22:07, 18 June 2006 (UTC)

I would start with your own words above: "I dont actually insist on any particular meaning except the most common one!" For a word of multiple shades of meaning, like "verify", the concern should be whether there is a reasonably-common sense that fits our usage and needs, not whether that is the most common sense. That would be hard to measure in any case. Water-cooler snippet: "That really happened -- I verified it on Wikipedia." Which precise sense was meant? Robert A.West (Talk) 03:23, 19 June 2006 (UTC)

Well actually, it doesnt really matter what the word means if we are going to redefine it on the page (does it?) 8-)

--Light current 03:26, 19 June 2006 (UTC)

You think we are redefining it. I think that we are just picking one of the well-established senses. Neither of us will ever convince the other, I suppose. Robert A.West (Talk) 03:55, 19 June 2006 (UTC)

I really am trying to agree with you, you know!. Any way I support your motion- so thats agreement of sorts insnt it?--Light current 04:05, 19 June 2006 (UTC)

Understood. However we got here, I think we got to an improvement in the phrasing. We'll just have to agree to disagree on the pedantic points.  ;-)Robert A.West (Talk) 04:12, 19 June 2006 (UTC)

Yeah we can save that for another day and another word! Actually you've just given me a new word to describe myself - pedantic. THanks! 8-)--Light current 04:18, 19 June 2006 (UTC)

Is there a second to this motion?

Does anyone other than Light current support this complaint, or is it opposed by consensus? (both can be true at once)

• Oppose Light current. Septentrionalis 20:42, 18 June 2006 (UTC)

It is not clear to which complaint you refer. Could you be a bit more specific? Thanks!8-)--Light current 20:46, 18 June 2006 (UTC)

I thought it was clearly the one immediately above; but let's generalize: Is there a second to any of Light cuttent's plaints (and if so, which?) Septentrionalis 21:43, 18 June 2006 (UTC)

Well we do seem to have two people who partially agree with me:

FWIW, I like it. Of course, by "any editor" we really mean, "Any editor with access to a good research library and whose Internet access is not restricted," but it may be silly to emphasize that distinction. Robert A.West (Talk) 13:18, 18 June 2006 (UTC)

I like it, too, and agree with LC re: confirm, but perhaps also substitute editor with reader or perhaps researcher? (I prefer reader). (Anon I think its mjb!)

my bolding Pls see further up page for the context of thee posts.--Light current 21:59, 18 June 2006 (UTC)

Robert A. West was agreeing with SlimVirgin; no-one appears to agree with you, on anything. Jayjg ^(talk) 00:47, 19 June 2006 (UTC)

Im afraid you are not following things Jayjg. Robert A. West agreed with SV, and I mostly agreed with SV. Therefore, Robert agrees with me!! Then there is of course mjb who seems to agee with me. Please read the full story before commenting! Thanks! 8-|--Light current 01:16, 19 June 2006 (UTC)

Motion

To adopt Slim Virgin's text, slightly altered per comments above.

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any reader must be able to ascertain that material added to Wikipedia has already been published by a reliable source, because Wikipedia does not publish original thought or original research.

• Support I made the alteration to "reader" as more inclusive -- readers who never edit may still look things up. I substituted "ascertain" to avoid apparent circularity. Robert A.West (Talk) 03:55, 19 June 2006 (UTC)

• Support (But what about editors who never read anything? sorry!)--Light current 04:03, 19 June 2006 (UTC)
  • Then they won't have read the policy. Robert A.West (Talk) 04:20, 19 June 2006 (UTC)

Ahh good logical thinking Mr Spock!--Light current 04:27, 19 June 2006 (UTC)

• Support, especially with the editor -> reader change.—mjb 05:40, 19 June 2006 (UTC)

I'd much prefer "check" to "ascertain". We mean only to say that one should be able to go look and see that the information is there. Grace Note 08:59, 19 June 2006 (UTC)

I agree with Grace Note; and we should still link to WP:RS, as the definition. Septentrionalis 15:13, 19 June 2006 (UTC)

Why should you have to go to another page to see what is being talked about on this one 8-? Theres a wiki link right there if needed. reliable, published sources.--Light current 15:24, 19 June 2006 (UTC)

But of course, the requirement is not met simply because the reference can be checked; the check must confirm that the source exists, that it is reliable and that the article accurately reproduces the material. That is why I prefer "ascertain." Robert A.West (Talk) 17:19, 19 June 2006 (UTC)

I suppose "ascertain" does mean all that: To find out or learn for a certainty by experiment, examination, or investigation; to make sure of, get to know. (The only current use.) But would "confirm" not be more understanded of the people? Septentrionalis 20:28, 20 June 2006 (UTC)

• Oppose. The current wording, if we don;t split hairs, is perfectly reasonable: The threshold for inclusion in Wikipedia is verifiability, not truth. This means that we only publish material that is verifiable by referring to reliable, published sources.. Verifiable as in "verify a claim", not as an assessment of truth. ≈ jossi ≈ t • @ 21:32, 21 June 2006 (UTC)

Thats what I ve been complaining about: the wrong word 'verifiable' being used to imply only that there is a published source. The word doesnt mean that as I have repeatedly shown by quoting dictionary references. Therefore if we are going to stick with the word, we have to define what we mean by it. I would have though that that was patently obvious to everyone. In policy making Im afraid we DO have to split hairs to make thinks absoultely clear to everyone. Please try to keep up with the discussion if you are going to comment on it. THanks!--Light current 21:49, 21 June 2006 (UTC)

• Oppose per Jossi. But I agree that the word "verifiability" may be somewhat confusing to new editors. In practice it means that information can only be included in Wikipedia if its existence in the real world can be verified. Also note that "material" is not a good word in the context of this proposal: publishing material already published elsewhere = copyvio. I also think the proposal is too verbose. How about something along these lines:

The threshold for inclusion in Wikipedia is verifiability, not truth. This means that we only publish information whose existence can be verified by consulting a reliable source, because Wikipedia does not publish original thought or original research.

AvB ÷ talk 22:56, 21 June 2006 (UTC)

Consensus

Is there consensus now to change the wording to:

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any reader must be able to ascertain that material added to Wikipedia has already been published by a reliable source, because Wikipedia does not publish original thought or original research.

--Light current 14:16, 21 June 2006 (UTC)

No, "sources" don't "publish", bad English, what does "published by a [...] source" mean? Unless by "source" you intend "publisher", which would change the meaning (giving way to the opposite assertion than what was concluded in a recent arbcom case, see wikipedia talk:reliable sources). The source has to be as reliable as possible. Less reliable publishers may publish (copies of) very reliable documents. This has limited influence on the reliability of the source. --Francis Schonken 14:37, 21 June 2006 (UTC)

yes but 'published by a reliable source' is already effectively in the existing wording. If its wrong, you should create a separate argument to correct it.! 8-|--Light current 15:39, 21 June 2006 (UTC)

But, this is a good time to get it right -- Wikipedia is not a race. And the objection is so easily solved -- "published in a reliable source" should do fine. Or, "in a reliable, published source." Robert A.West (Talk) 15:53, 21 June 2006 (UTC)

Ill take your advice on the sematics here Robert! I assume that a publication (any publication) can be taken as a 'source'. 8-|--Light current 16:04, 21 June 2006 (UTC)

I like the new intro, except that I would say "check" instead of "ascertain," because it's more straightforward. Francis, the "published by a reliable source" issue: by "reliable source," we mean the material, and the emphasis can be on the publisher of the material or writer of the material interchangeably. If, for example, a person normally regarded as a good source were to have something published on a neo-Nazi website, many editors would not accept it as a reliable source, because the story may not have been faithfully reproduced, and because it wouldn't have been fact- and libel-checked. Similarly, if the Guardian publishes a story by a completely unknown person, the story is regarded as a reliable source because the Guardian is a reliable publication. Cases where a reliable person publishes on an unreliable medium are dealt with on a case-by-case basis, because we can't legislate for every eventuality. But in general when we say "this is a reliable source," we mean "this piece of text, or this broadcast, can be trusted," because what we're judging is the process of publication: that the material was properly checked and edited. SlimVirgin ^(talk) 21:59, 21 June 2006 (UTC)

Sorry, I meant to add: so it's clear from the intro that we're saying the material must already have been published reliably, and we leave open whether the reliability attaches to a publisher or a person, but in fact most often it will be the publisher. SlimVirgin ^(talk) 22:01, 21 June 2006 (UTC)

Well Slim, if you like it, when do you think we could actually put the new wording into effect on the page without it being reverted?--Light current 22:03, 21 June 2006 (UTC)

Added with the change of "ascertain" to "check" as per SM's. ≈ jossi ≈ t • @ 22:11, 21 June 2006 (UTC)

Thank you Jossi! 8-))--Light current 22:32, 21 June 2006 (UTC)

Too fast

I think this went way too fast. I oppose the new language for reasons given above:

Oppose per Jossi. But I agree that the word "verifiability" may be somewhat confusing to new editors. In practice it means that information can only be included in Wikipedia if its existence in the real world can be verified. Also note that "material" is not a good word in the context of this proposal: publishing material already published elsewhere = copyvio. I also think the proposal is too verbose. How about something along these lines:

The threshold for inclusion in Wikipedia is verifiability, not truth. This means that we only publish information whose existence can be verified by consulting a reliable source, because Wikipedia does not publish original thought or original research.

AvB ÷ talk 23:03, 21 June 2006 (UTC)

I like your word 'existence'. Maybe it can be used in the next version! Its a pity you werent here earlier! 8-( --Light current 23:22, 21 June 2006 (UTC)

"Existence" is ambiguous. I think it's fine as it is. It's obvious what it means. SlimVirgin ^(talk) 00:18, 22 June 2006 (UTC)

I exist, you apparently exist. Why do you say the word 'existence' is ambiguous?--Light current 00:20, 22 June 2006 (UTC)

Light Current: Teehee, too late, not too fast. I'll buy it. Having said that, my main point is that I find the previous consensus version much better with the exception of the vague and ambiguous "referring to"/"with reference to" text. In that sense I fully agree with the work done here to solve the ambiguity. My main change is the use of "by consulting". For future reference: without "existence" my version resolves to:

The threshold for inclusion in Wikipedia is verifiability, not truth. This means that we only publish material that can be verified by consulting one or more reliable sources.

SlimVirgin: I do not agree that "existence (of information)" is ambiguous. In fact it signifies precisely what I think Wikipedians are required to do. Publication in a reliable source is our cue that information is really out there and that we are getting a relatively undistorted version. We can then propagate the information to the rest of the world via the encyclopedia, obviously also avoiding undue weight and citing our sources.

But the main problem I had with this version has been solved by the change from "by reliable sources" to "in reliable sources" (edit by Francis I think). This has the exact same effect intended in my version. The overly verbose stuff can remain as far as I'm concerned. The point put across by "in" or by "by consulting" is what counts. AvB ÷ talk 09:52, 22 June 2006 (UTC)

It has to be published by a reliable source, not necessarily in one. If Steven Hawking were to publish his latest ideas about time in a local gardeners' newsletter, it would have been published by a reliable source (Hawking), but not in one (the newsletter). We can say this even though Hawking is not strictly speaking the publisher. And if an unknown person publishes a pile of nonsense in the New York Times, the pile of nonsense is reliable because it was published in and by the NYT. "By" covers more eventualities than "in" does. In some cases, "in" is plain wrong.

Could I request that people stop making all these little changes without discussing them first? The page is becoming quite unstable and even a small change of wording can have a big change in meaning. SlimVirgin ^(talk) 10:04, 22 June 2006 (UTC)

Actually, I'm thinking we should perhaps return to the version with "verifable by reference to," in order to return stability to the page, because that language covered all the "by" and "in" problems people are now creating, which is why it was written that way in the first place. Would anyone support returning to that version? SlimVirgin ^(talk) 10:07, 22 June 2006 (UTC)

I would support returning to that version. I do not agree "in" creates a problem though; IMHO it solves a problem. Which highlights the differences in interpretation that all had a place in the more ambiguous (and hence more pliable) version. A version that accommodated your interpretation as readily as it did mine. Summary: I do not see an authority figure like Hawking as a reliable source in Wikipedia terms. We agree that unless his statements, research, opinion pieces, interviews etc. are published in a reliable source, we have zilch. Hawking's statements in the lay media etc. don't count because he's a reliable source; they count because he is notable. What he writes in peer-reviewed journals counts because they are considered reliable scientific sources, opening his contributions up to further scientific scrutiny, replication, falsification, the whole shebang. Not especially because it's Hawking (although, I admit, his stature does make a difference).

In a way this is a general problem with recent changes to several policies: they tend to fine-tune beyond a point that is workable, unearthing interpretation differences between veteran Wikipedians that so far had not led to an unworkable situation. There are two ways to solve such problems:

1. Leave somewhat ambiguous language as it is; editors can interpret and apply policy on a case-by-case basis (which, indeed, was the case when I arrived on Wikipedia), advancing/discussing their own interpretation in the consensus process.
2. Create a consensus interpretation of such things as "verifable by reference to" and fine-tune the text to reflect that and only that interpretation.

AvB ÷ talk 11:44, 22 June 2006 (UTC)

That is the problem with this constant tweaking. Often, the wording has been carefully chosen for a particular reason, and if it's ambiguous, it's because it's meant to allow some wriggle room, so that we don't have to turn ourselves inside out thinking about e.g. the in/by problem. "Verifiable by reference to reliable, published sources" covers both, and also gets rid of the copyvio implication that people are seeing in the new version.

I see your distinction between reliable source and notable, but people do become reliable sources (in WP terms) in virtue of notability, in part because they have easier access to the reliable publications, but also in part because they're regarded as authorities regardless of who publishes them. It's something we should deal with on a case-by-case basis, I feel, rather than trying to nail down a more precise definition of what a reliable source is, because we might close good loopholes as well as bad ones. At some point, we have to rely on editors' common sense, as scary a prospect as that is. SlimVirgin ^(talk) 12:48, 22 June 2006 (UTC)

(reducing indentation level)

Which about says it all. I'm wondering if other editors would also agree to go back to the previous version after reading the above. Wikipedia is a BIG place and an interpretation that works for one editor editing a specific type of article may not work for another editor editing a different type of article. Some leeway is essential. AvB ÷ talk 13:57, 22 June 2006 (UTC)

I've restored the last sensible version (though I still have to clean up some stuff). People keep "tweaking" wording for various reasons, but all the tweaks are doing is adding giant loopholes for people to insert dubious information from unreliable sources into articles. Please stop messing around with this page without significant consensus, it is a policy and is supposed to be stable. Jayjg ^(talk) 14:45, 22 June 2006 (UTC)

Good points again. Perhaps we can go back in time to the ancient stable version of paragraph 1 of the lead section found here:

The threshold for inclusion in Wikipedia is verifiability, not truth. This means that we only publish material that is verifiable by referring to reliable, published sources.

AvB ÷ talk 19:35, 22 June 2006 (UTC)

Neologisms

Given that a true neologism is by definition not verifiable, wouldn't it make sense to create a speedy tag for articles about a neologism? I ran across a neologism a few days ago with zero Google hits, and an admin went apoplectic when I tagged it for a speedy delete. It seemed only a matter of common sense to me that an article entirely based upon a policy violation finds itself speedied. It's not like anyone could rewrite such an article in a way that would make it acceptable. Rklawton 00:43, 21 June 2006 (UTC)

If a topic is truly not verifiable, then yes it should be deleted. The issue is in determining if it is not verifiable. I don't think we need a separate speedy tag for it because verifying that something is not verifiable is not something that can be done immediately. - Taxman ^Talk 14:12, 21 June 2006 (UTC)

Self-published sources, again

This section used to have: "Self-published sources may never be used as sources of information about another person or topic, subject to the limited exceptions discussed above."

Somehow this got deleted, then was "restored" on June 10 by SlimVirgin, omitting the phrase noting that exceptions exist. That phrase had been in the text, and I see no evidence of discussion about deleting the phrase by SlimVirgin. This was a quite important phrase, as there are legitimate exceptions mentioned in this very text, among other places. I can't follow how it has become part of the revert-war. Gimmetrow 22:20, 21 June 2006 (UTC)

There are more agreed upon exceptions than the single one mentioned as an example in this policy page, see WP:RS:

• Limited use of references to primary sources as explained in WP:RS#Personal websites as primary sources;
• With caution, expert opinions expressed by experts in their field of expertise, as explained in WP:RS#Reliability of online sources;
• When reporting that an opinion is held by a particular individual or group, the best citation will be to a direct quote, citing the source of the quote, as explained in WP:RS#Reporting about opinions: use of direct quotes.

--Francis Schonken 22:25, 21 June 2006 (UTC)

Yes, I see that the WP:RS, where this text was moved, now has a list of exceptions specified. The point is that without a phrase noting some exceptions, it excludes any exceptions. Now this text is completely absent from WP:V, effectively, reducing the text from a policy to a guideline. I can't find any real discussion of this deletion (of the phrase or the entire section) here. Gimmetrow 23:35, 21 June 2006 (UTC)

I don't recall intending to remove that, Gimmetrow. I'll take a look at the context and if it makes sense (i.e. if there are limited exceptions to the above that it refers to, I'll put it back in. SlimVirgin ^(talk) 00:11, 22 June 2006 (UTC)

Okay, I took a look and I wonder if someone recently added it, and then I removed it because it made no sense. As it stands, I can't see what exceptions it refers to. The sentence is: "Self-published sources may never be used as sources of information about another person or topic." What are the exceptions? SlimVirgin ^(talk) 00:15, 22 June 2006 (UTC)

How about things widely accepted as so in the scientific community? The best example here is the Poincare conjecture, it was considered solved by mathematicians (and there are ample sources for that), yet the formal proof still isn't announced TTBOMK. We can't use Perelman's papers, which being preprints only wouldn't count as credible source by the standards of Wikipedia. That's kind of silly, considering that for the last few year or two, any mathematician you would ask would consider the problem solved and build on the results put forth in the papers. Dr Zak 00:27, 22 June 2006 (UTC)

One exception is a recognized expert in a field with self-published books or a personal web site. This seems common in some fields, like coin collecting or heraldry, as mentioned in this article. Gimmetrow 00:38, 22 June 2006 (UTC)

An expert cannot be used as a source on a third-party if s/he is self-published. That's an important point to retain for BLP. We need to know that the publication has run a fact- and libel- check, which WP can't afford to do. That is the whole point of the restriction. SlimVirgin ^(talk) 02:01, 22 June 2006 (UTC)

Please stop removing that section. This is a policy page. Dr Zak and Francis are removing an important section saying it does not rise to policy level, whatever that means. That is your opinion and you can argue it on talk, but please don't simply remove it. SlimVirgin ^(talk) 02:03, 22 June 2006 (UTC)

Consider trying to understand what "does not rise to policy level" means. Verifiability means that statements made must be backed up by appropiate sources, according to proper bibliographic and scientific practice. What "proper scientific practice" is is explained in the Reliable sources guideline. Dr Zak 03:18, 22 June 2006 (UTC)

Then clearly restrict the sentence to BLP. My concern is mostly with "topics." The sentence, on its own, could be interpreted as not allowing as sources essays on heraldry from websites of heraldry experts. Gimmetrow 02:08, 22 June 2006 (UTC)

Actually, the restriction about third parties is already in the bulleted list, so I'm fine with it as it is. I reverted my restoration of that one sentence. SlimVirgin ^(talk) 02:13, 22 June 2006 (UTC)

I'm sorry, this still has the problem. That sentence looks like it stands on its own. "Self-published sources may never be used as sources of information about another person or topic." So my heraldry essays are out? Gimmetrow 02:35, 22 June 2006 (UTC)

I think the list on its own without the sentence at the end limits it to BLP, with should satisfy my concern. Gimmetrow 04:00, 22 June 2006 (UTC)

A pain in the ass

Im very sorry to appear to be a pain in the ass to everyone, 8-( but I've just noticed that the intro mentions the word 'material'. This implies that we are lifting stuff verbatim from published sources. I suggest changing this word to 'information'.--Light current 00:54, 22 June 2006 (UTC)

Material is better, because information implies that it's true. The sentence doesn't imply that we are plagiarizing. It says we only publish material that has already been published, not that we copy it word for word. SlimVirgin ^(talk) 01:58, 22 June 2006 (UTC)

LC, "verifiability" isn't grammatical in that second sentence. It is verifiable because it can be checked, not verifiability because it can be checked. It really is fine as it is. The constant tweaking is causing deterioration. SlimVirgin ^(talk) 02:06, 22 June 2006 (UTC)

I'm not horribly keen on the wording. It is rather ambiguous and I share the concern that bickering will break out over whether it contradicts the policy on copyrights, even though it's absurd to think that it does. Slim, I think we are all aware that there is no such thing as an argument too absurd for some Wikipedians to make! I agree that "information" is not an improvement but we want a word that is a substitute for "facts" that does not have the implied value judgement of "facts", because what we mean is the "matter" of what is added must have been published elsewhere. I do not believe that the sentence should be edited again hastily, but this should be considered. The problem with this sort of article is that most experienced editors have a good grasp on what we mean here, but it's a good idea not to rely on that and to write policies as tightly as possible. Grace Note 02:18, 22 June 2006 (UTC)

Grace Note, would you mind reverting to my 02:04 June 22 version. A new account added nonsense while I was editing, and I inadvertently saved it, but I'm out of reverts thanks to the constant self-published section removal. He added something about how we go for truth via verifiability, but we don't, and it's horrendous writing in any event. I agree with you that this hasty editing should stop. SlimVirgin ^(talk) 02:22, 22 June 2006 (UTC)

No need. I was bold. After all, the discussion had not finished. -- Alias Flood 02:32, 22 June 2006 (UTC)

Note that WP:BOLD applies in different ways to different namespaces. Perhaps the WP namespace should be a little less bold. This particular page is changing drastically with little or no discussion, and is not setting a good example for the newbies. Gimmetrow 02:40, 22 June 2006 (UTC)

Exactly my reason for restoring the article to its original form and my reference to the unfinished discussion. -- Alias Flood 02:48, 22 June 2006 (UTC)

Well duh I missed this entire section. You folks are fast. But I really have to insist that the current "material" has to change as this is a recepy for copyright violation (as also argued and noted in my contribution above). Using the word "information" is an elegant way to solve the problem. AvB ÷ talk 10:08, 22 June 2006 (UTC)

Or returning to the previous version. SlimVirgin ^(talk) 12:49, 22 June 2006 (UTC)

Yes, "material" as used in the previous version did not have this problem. AvB ÷ talk 13:21, 22 June 2006 (UTC)

Verifiability VS Popular Opinion

We all know that Wikipedia isn't meant to include "fact" (it's not the determining factor), but I've always thought that popular opinion is what determines inclusion. For example, I've read that the article must represent what differing viewpoints in real life proportionately; if majority believes in ghosts, and minority don't, than the article must portray it this way, may stating that majority believes in ghost; while a small section is devoted ti minorities who disagree. Another thing I remember reading is that if Wikipedia existed back when they thought the world is FLAT; than Wikipedia would say that the world is flat; not because it is truth, but because that is the general consensus. So should consesus/popular opinion be the determining factor, or is it verfiability? 24.23.51.27 05:48, 22 June 2006 (UTC)

Verifiability. I believe that in the hypothetical Ancient Wikipedia, just as in today's Wikipedia, the inclusion, characterization, and relative prominence of claims in an article is supposed to be determined not by popular opinion, but rather by the preponderance or dearth of reliable sources making one claim versus another. For better or worse, when the sources are very lopsided in favor of one claim, it is more likely to just be repeated as if it were fact, without any kind of "some" (or many) "say" qualification, and without explicit references (though references must be provided if a serious challenge is made, of course).

It might help to consider that when Wikipedia says "The world is round", it means "A preponderance of reliable sources claim the world is round", with reliability being measured by today's criteria. The Ancient Wikipedia would say "the world is flat" because the criteria for reliable sources would've been different back then; a preponderance of 'reliable' sources back then would've said the world is flat. Similarly, when the modern Wikipedia says "A small minority claims the world is flat", it means "Reliable sources claim that a small minority claims the world is flat" …which is much different than saying "The world is flat" / "(A preponderance of) reliable sources claim that the world is flat."

Wikipedia 500 B.C.

The world is flat.[1] A small minority claims the world is round.[2]

References:

1. Mayan King Tlecpthuoxcptectl, in a statement at a sacrifice of heretics who say the world is round.

2. Tlecpthuoxcptectl, in the same statement, summarizing the claims of the doomed heretics.

Wikipedia 2006 C.E.

The world is round.[1][2][3] A small minority claims the world is flat.[4]

References:

1. Pliny the Elder, Natural History 2.64 (77 C.E.)

2. Bishop Isidore of Seville, Etymologiae (1470 C.E.)

3. Physics Today feature, summarizing the published research of 150 notable astrophysicists (1987)

4. New York Times article on the Tlecpthuoxcptectl Had It Right All Along Society (2006)

—mjb 10:28, 22 June 2006 (UTC)

Verifiable Truth

I changed it to "Verifiable truth" because it retains the essential element of truth, but only that subset which is verifiable, otherwise people could create articles which they know are false, just because there is some published source for it somewhere.

Thank you andrewrutherford

Already reverted. Pay close attention to the mention of truth already in the policy. Someone else can probably provide a better explanation, but basically, a publication containing untrue statements about a subject is not likely to meet more general criteria that would allow one to conclude it is reliable as an authority on that subject. There's no need to ascertain the truthfulness of specific claims. For example, just by the type of publication it is, the amount of editorial oversight it has, the level of peer review it gets, the credentials of its authors, and so on.—mjb 09:02, 22 June 2006 (UTC)

Truth is inherently subjective, not objective. KillerChihuahua^?!? 09:09, 22 June 2006 (UTC)

Andrew, please don't change this again. We publish material that is well referenced, that's all, not "verifiable truth," whatever that means. SlimVirgin ^(talk) 10:37, 22 June 2006 (UTC)

This from Wikipedia: Wales has been a passionate adherent of the Objectivist philosophy of Ayn Rand. — Preceding unsigned comment added by Andrewrutherford (talk • contribs) 12:37, 22 June 2006

That has nothing to do with the topic at hand. KillerChihuahua^?!? 12:40, 22 June 2006 (UTC)

(Sorry I just had some editing difficulties and got cut off half way through) continued...

This from Wikipedia: Wales has been a passionate adherent of the Objectivist philosophy of Ayn Rand. wiki.riteme.site/wiki/Jimmy_Wales Objectivism derives its name from its conception of knowledge and values as "objective", rather than as "intrinsic" or "subjective". http://wiki.riteme.site/wiki/Objectivist_philosophy I don't agree with Objectivist philosophy on all points, but that one I do, and if you don't with the founder of the website on such a basic point, then why quote him on the project page?Andrewrutherford 12:57, 22 June 2006 (UTC)

Andrew, please don't change this again. We publish material that is well referenced, that's all, not "verifiable truth," whatever that means. SlimVirgin (talk) 10:37, 22 June 2006 (UTC)

Verifiable truth would be information that you believe to be true and which you can justify with evidence. There are the philosophical issues to consider when publishing false information, but there are also the legal ones: In English and American law, and systems based on them, libel and slander are two forms of defamation (or defamation of character), the tort or delict of publishing (to a third party) a false statement that negatively affects someone's reputation. http://wiki.riteme.site/wiki/Libel This is discouraged in another Wikipedia policy: "For this reason, all contributors should recognize that it is their responsibility to ensure that material posted on Wikipedia is not defamatory." http://wiki.riteme.site/wiki/Wikipedia:Libel So why not just say that the information has to be true and therefore avoid any possibility of libel?Andrewrutherford 12:57, 22 June 2006 (UTC)

I can only repeat that Wikipedia policy is not to focus on truth, but on what good sources say. We say: "The New York Times says that the sun will not rise tomorrow," and we link to the NYT. It is true that they said it, but what they said was nevertheless nonsense.

Most people disagree about what is true about any given topic, so if we were only to publish what everyone agrees is true (not just true that X said it), we would be left with a very small encyclopedia. SlimVirgin ^(talk) 13:06, 22 June 2006 (UTC)

I would add that if the NYT were in the habit of making claims like that (claims that were otherwise unverifiable), WP:RS would soon be adjusted to sharply reduce the assumed reliability of that publication, and Wikipedians would be obligated to review whether the claims should be mentioned at all.—mjb 18:43, 22 June 2006 (UTC)

Truth is a difficult concept! I think we should certainly believe that stuff is true before we add it. We should not put something in that we believe is false just because we have a (possibly very good) reference for it (unless we point out the unreliability or dubiousness of the ref). --Light current 13:18, 22 June 2006 (UTC)

I often add material, in the interests of NPOV, that I strongly disagree with, and that I would argue is false if I had to argue it. What we have to do is choose high-quality sources and then just report what they say. The better the source, the more intelligent the contribution, so even if we disagree, we should still be able to recognize the quality of the argument/research/opinion, whatever it is. The issue of "truth" rarely comes into it. SlimVirgin ^(talk) 14:05, 22 June 2006 (UTC)

Perhaps it helps if Andrew and Light current (re-)read WP:NPOV in the light of this discussion. This problem can (and on WP it must) be avoided by attributing assertions to specific people/organizations/whatever - as explained by SlimVirgin. This turns viewpoints into truths regardless of our personal POV. AvB ÷ talk 14:23, 22 June 2006 (UTC)

So you're Ok with me creating deliberately false articles, so long as I can find some sources? I'm sure I can create an entire alternative universe using North Korean and Stalinist sources. Most material on Wikipedia isn't aggressively fact checked anyway, so you could probably get away with quite alot via that method. Of course you could have people do it anyway, but why have policy to encourage it?

All I want is for the standard of knowledge in Wikipedia to reflect mainstream philosophical consensus of justified true belief. http://wiki.riteme.site/wiki/Knowledge

At the moment, no one has to actually believe it, not even the person posting it on Wikipedia. You might just find an old book in your attic and think "that would make a great Wikipedia article". So you resurrect some madness that no one believes as if it were a live theory! That would be just as bad as posting original research, in fact it would be worse because it would look more credible.

Now I certainly agree that you should be encouraged to post information that you don't personally believe in order to provide a balanced point of view within a topic, but at the moment there is no difference between this and inventing your own fantasy theory for fun, and then seeking out some crank books in order to provide apparent justification.

Therefore, I propose at least the following threshold of knowledge that is being recorded for reasons other than historical reasons: Someone should believe it to be true! —The preceding unsigned comment was added by 59.17.183.228 (talk • contribs) 00:54, 23 June 2006 (UTC)

These problems have been noticed and solved a long time ago. You may want to study the relevant policies and familiarize yourself with concepts such as notability, the avoidance of undue weight, and the Wikipedia consensus process. AvB ÷ talk 13:02, 23 June 2006 (UTC)

So basically you're part of a system which directs me to web pages in general, rather than answering me. OK! In that case I'll make it easy for you. Here's a question: Is it in accordance with Wikipedia policy for me to create articles which I know are completely fictitious, or not? and (if you can be bothered) why not? Andrewrutherford 16:48, 23 June 2006 (UTC)

Wikipedia would allow an article on Flat Earth. If there is publication by reliable, reputable sources about a subject, an article about the subject would probably work. Aliens invading earth, the pyramids of mars, if it has been published by reliable, reputable sources, there's the meat of it. Just take what has been written create an article with it. Terryeo 17:06, 23 June 2006 (UTC)

The threshold of inclusion is verifiability is good and I like it, but attribution is important too and not included in the threshold statement. I'm finding editors defending themselves because their quotation is verifiable, but not attributable. For example, a newsgroup's past posting, stored on a server, is quoted. Well, that's verifiable but not attributable. Who knows the identity of newsgroup postings? I do understand a newsgroup is neither reliable nor refutable, but a statement like, "verifiable and attributable" would prevent some editor difficulty, I think. Terryeo 15:52, 23 June 2006 (UTC)

I'm sorry, but his has nothing to do with what's being talked about here. You might want to start your own section. Andrewrutherford 16:50, 23 June 2006 (UTC)

Self-published sources in articles about themselves

One of the bullet points in this section of policy describing which material may be used as sources of information about themselves in articles about themselves states that to be included, that material is not contradicted by reliable, third-party published sources;. I would argue that this needs to be removed. Both the self-published material and the contradictory material from reliable, third-party published sources should be allowed to be included as per WP:NPOV side by side. After all, for NPOV, we would like to have to have the POV of the subject and the POV of reliable sources in in these type of articles, wouldn't we? ≈ jossi ≈ t • @ 15:10, 22 June 2006 (UTC)

Hi Jossi, I half agree with you and half disagree. Much of the time, it's fine to say, "according to the NYT, John Doe spends six months every year overseas to avoid paying tax, although Doe on his website disputes this account." However, I've run into situations where borderline notable people put material up on their websites in order to contradict everything that has ever been published about them, which makes the article tedious and turns Wikipedia into an extension of their websites. I therefore feel there should be some restriction along the lines of: "so long as the material is not contradicted by reliable, third-party published sources." But I agree we could water it down a lot from the current wording. Something along the lines of: "Proceed with caution if the self-published information is contradicted by reliable, third-party published sources." Would that work? SlimVirgin ^(talk) 15:18, 22 June 2006 (UTC)

Yes. That would work as it puts the burden on editors to proceed with caution in these borderline situations. Otherwise, I see the other side of the equation in which each statement made by a person on his website gets automatically removed citing this policy, because one author wrote a contradictory statement in a book. ≈ jossi ≈ t • @ 16:23, 22 June 2006 (UTC)

That's a good point. Okay, I'm fine with removing it and adding instead something along the "proceed with caution" line. SlimVirgin ^(talk) 16:51, 22 June 2006 (UTC)

I'm always liked the "not contradicted" wording, myself. --Elonka 18:09, 22 June 2006 (UTC)

Elonka, the problem is that appliying this wording can result in some strange consequences. Imagine a published auto-biography of prominent politician XYZ, in which he makes a particular claim. Now imagine author ABC that is critical of politician XYZ that writes a pasage in a book that contradicts XYZ's claim. As per current wording you could not include XYZ's statement on XYZ's article, only ABC's. And that, is a very strange consequence, and in contradiction with WP:NPOV. ≈ jossi ≈ t • @ 19:16, 22 June 2006 (UTC)

And more. By the current definitions all works from antiquity are "self-published" ("peer review" didn't exist in those days, and according to the current wording of WP:RS all publications that are not peer-reviewed are "self-published"). Then you have, for instance, Tacitus' article. Part of it is about his critical attitude regarding Ancient Romans. Much of that is "contentious", by definition it is about "third parties" and some of it is "contradicted by reliable, third-party published sources". Sorry, in Tacitus' article it is best to quote Tacitus for what he said about Ancient Romans. Not quoting Tacitus on these, but only his critics is the world on its head. How are you going to write about Martin Luther without mentioning what he wrote about his contemporaries, e.g. what on earth would be non-wikipedian about:

Luther issued his To the Christian Nobility of the German Nation (Aug. 1520), committing to the laity, as spiritual priests, the reformation required by God but neglected by the pope and the clergy. For the first time of many, Luther here publicly referred to the pope as the Antichrist.^[1]
[1]Martin Luther, An Open Letter to The Christian Nobility of the German Nation Concerning the Reform of the Christian Estate, 1520, trans. C. M. Jacobs, in Works of Martin Luther: With Introductions and Notes, Volume 2 (Philadelphia: A. J. Holman Company, 1915; Fort Wayne, IN: Project Wittenberg, 2006) http://www.projectwittenberg.org/pub/resources/text/wittenberg/luther/web/nblty-01.html.

?... Yes he called the pope "antichrist", which in his day was not really perceived as a compliment. It was "contentious", "about a third party", "self-published", "contradicted by other (at least as reliable) sources". So according to the present WP:V policy it should be thrown out of Wikipedia. No, this simply doesn't work this way. This is bad policy writing.

The "self published" section may be OK for biographies of living persons, in view of avoidance of libel charges. It is not OK for most of the other topics. According to Jimbo "WP:LIVING is a decent start, although it needs some attention I think so that we can bring it up to the standard of a full policy." WikiEN-l mailing list, May 7 19:44:05 UTC 2006. The "self published" section of WP:V is a part of WP:LIVING imported here. This is not the way how to move a "guideline that is a decent start" up to "non-negotiable policy". First go improve WP:LIVING. If then its quality is good enough to promote it to policy: fine, I'd be the first to support that. But as long as that doesn't happen, the piece should not be in the WP:V policy unless it is improved, and/or unless that section is strictly tied to BLP (as the best we have on the matter currently). But no way applicable to all wikipedia articles, or to all "self-published sources" --Francis Schonken 22:07, 22 June 2006 (UTC)

Luther is not currently publishing his works. Instead, reputable publishing houses do so. Jayjg ^(talk) 01:20, 23 June 2006 (UTC)

When he published it, it was "self-published, which means it has not been subject to any independent form of fact-checking." Not a letter was changed to these publications since that moment (I hope, that's why we call it a source). So, according to the current definition of "self-published" in wikipedia:reliable sources, these original sources are "self-published" and will remain so. That's why the "self published" section in WP:V doesn't work the way it is now. --Francis Schonken 02:17, 23 June 2006 (UTC)

Don't use the versions he self-published; instead, use the more recent editions which have been published by reputable sources. Oh, wait! That's what everyone does already! Problem solved. Jayjg ^(talk) 02:19, 23 June 2006 (UTC)

irrelevant remark. Wikipedia guidelines & policies mark certain publications as "self-published" (and Luther's works are part of that), and then WP:V tries to apply a rule to "self published" works, that doesn't work. --Francis Schonken 02:34, 23 June 2006 (UTC)

We wouldn't use Luther as a general source. We might say "Luther called the pope the "anti-Christ," as a matter of historical interest, but we would never say: "The pope is the anti-Christ" (Martin Luther 1517).

True, but neither "general source" nor "source for a matter of historical interest" are allowed by the current wording of the "self published" section in WP:V. So, it's wrong. And FYI, for living people it's the individual statements of opinion from unreliable sources we want to prevent too, if you hadn't noticed. The things we today call "libel", and which in Luther's time were sanctioned by "ex-communication". --Francis Schonken 02:34, 23 June 2006 (UTC)

You keep claiming that Luther is "self-published", but he is not (or not any more). His works are published by quite scholarly sources, so the issue you keep raising is not relevant. Jayjg ^(talk) 17:23, 23 June 2006 (UTC)

Francis is confusing a person's expressing their views with a person's publishing their views. There's no bar on including what people said, about themselves, about others, about the Pope even, so long as they get the NYT to publish it. Grace Note 00:42, 25 June 2006 (UTC)

There is a better response here. It has two points:

• WP:V doesn't compel anybody to do anything if nobody challenges a statement. This covers most articles with sane editors, most of the time.
  • Therefore if you show someone the quote from Luther, and they withdraw the {{Citation needed}}, you're done.
• If not, don't argue about sixteenth-century German, find a modern scholar who says "Luther said X", and cite him in the article (or footnote).

This applies to all primary sources, self-published or not. Septentrionalis 18:22, 23 June 2006 (UTC)

I will go farther. If the assertion at issue is whether the Pope was actually the Antichrist, then Luther's statement would be as useless as self-published statement by a living person: say, Jerry Falwell. If the assertion is that Luther thought that the Pope was the Antichruist, then we should have a secondary source to tell us not only that he made the statement, but that he meant it, as opposed to using hyperbole. So for a similarly contentious statement by Falwell, unless he has made unmistakable comments to the effect that he meant it literally. All this is clearly covered by the policy, provided we keep in mind what the assertion actually is. Robert A.West (Talk) 19:20, 23 June 2006 (UTC)

Stability

Francis, please stop reverting and removing content. It is becoming disruptive and making the page unstable. The material you are removing is important and is part of policy. If you object to it, please say why here. SlimVirgin ^(talk) 15:11, 22 June 2006 (UTC)

Sorry, I didn't "remove" any content - where d'you get that false accusation? I added content, which was removed by Jayjg. Was anything wrong with my additions? --Francis Schonken 15:15, 22 June 2006 (UTC)

You've been removing the self-published section, you added something about BLP even though this is not BLP, and thre were several other complex, partial reverts. The page has to be stable because it is a policy page, so could you please discuss your edits here first? SlimVirgin ^(talk) 15:20, 22 June 2006 (UTC)

Sorry, again? "You've been removing the self-published section"... Not in my last six edits. In the mean while Jayjg did 10 edits, repeatedly removing stuff; in the same period you did 4 edits, one of which was the removal of a paragraph from the self-published section.

Apologies, please. I'm wanting to discuss my edits, but maybe better get rid of your false accusations first, in order to show you assume good faith on me. --Francis Schonken 16:37, 22 June 2006 (UTC)

Wow, 10 whole edits, and I was removing stuff too? You mean like these edits [2] [3] [4] [5] where I removed blank lines? Or these edits, [6] [7] [8], where I added links and references? That's terrible, isn't it! Jayjg ^(talk) 01:29, 23 June 2006 (UTC)

You removed a reference regarding the status of Wikipedia:Biographies of living persons I had inserted [9]. Yes, you removed stuff. --Francis Schonken 02:25, 23 June 2006 (UTC)

Um, yes, but you claimed I had made "10 edits, repeatedly removing stuff;" - the "repeatedly removed stuff" was a bunch of blank lines, and your comment was entirely deceptive. Jayjg ^(talk) 02:33, 23 June 2006 (UTC)

You removed the same a second time [10]. So, changing "repeatedly" in my assertion above to "more than once". Hope you accept my apologies. --Francis Schonken 02:47, 23 June 2006 (UTC)

I'm talking about your removal of this [11] but there were other edits, adding things, moving things. People are losing track and the page is becoming unstable. Even small changes in wording can make a big difference, so please discuss the changes first. SlimVirgin ^(talk) 16:49, 22 June 2006 (UTC)

Sorry, that was yesterday, and not one of my last six edits, so clearly I had "stopped" removing anything regarding the self-published section. But you still removed a paragraph from it after that.

Also you *reverted* my tweaks to the first paragraph, that were agreed upon here on talk at the time.

So, again, I'm asking that you would offer your apologies for distributing faulty & damaging information. --Francis Schonken 17:01, 22 June 2006 (UTC)

(whitespace for SlimVirgin's apologies, I insist, per WP:CIVIL - Francis Schonken 02:47, 23 June 2006 (UTC)):

....

....

....

Hey, let us do this by showing some restrain in our comments, shall we? We are killing the fun, otherwise. Thanks. ≈ jossi ≈ t • @ 17:04, 22 June 2006 (UTC)

OK, as far as I'm concerned this section can be removed from this talk page, per "Zero information is preferred to misleading or false information" [12] --Francis Schonken 17:09, 22 June 2006 (UTC)

I do not see that interpretation. To me the quote is useful in explaining that, as well as the consensus, the views of individuals who formulated that opinion will be recorded along with their names. In short, the accountability of the whole truth. -- Alias Flood 01:58, 23 June 2006 (UTC)

Francis, that quote was regarding articles, not Talk: pages, and it's not up to you to decide if other editors are making "misleading or false" statements on Talk: pages. Please do not abuse policy and guidelines in this way. Jayjg ^(talk) 02:17, 23 June 2006 (UTC)

No reason to give "false information" on talk pages. Also in that case "Zero information is preferred to misleading or false information". Failing that, at least apologies should be produced for "misleading or false information". --Francis Schonken 02:47, 23 June 2006 (UTC)

More important is WP:CIVIL. Are you entirely finished claiming that others are making "misleading or false information" when you disagree with what they say? Because Talk: pages are for discussing article content. Jayjg ^(talk) 02:49, 23 June 2006 (UTC)

Am I missing something here? I was having a somewhat tense conversation with SlimVirgin. Then someone suggested to stop it. I moved it to SlimVirgin's talk page. She deleted it. I would have left it at that. Then you brought SlimVirgin's insulting verbiage back to this page, making clear I shouldn't remove it. I apologized to you for something that was implied by what I had said to SlimVirgin (and that otherwise would have been covered by oblivition by now). But SlimVirgin's insulting verbiage is still here. Waiting for an apology. WP:CIVIL --Francis Schonken 03:39, 23 June 2006 (UTC)

Hey, let us do this by showing some restrain in our comments, shall we? We are killing the fun, otherwise. Thanks. ≈ jossi ≈ t • @ 03:51, 23 June 2006 (UTC)

Francis, I apologize for anything I said that you saw as insulting. My only plea and concern is that sections should not be removed from the page without agreement, and that new material shouldn't be added without agreement unless it is obviously in keeping with the letter and spirit of the policy. The content policies and guidelines need stability. That is my only concern. I hope we can concentrate on discussing content from now on. SlimVirgin ^(talk) 05:25, 23 June 2006 (UTC)

Tx. I suppose that by now it is clear that grouping the BLP-related material on the WP:V policy page, and linking it to the BLP guideline, like I did, is obviously in keeping with the letter and spirit of the policy:

• Per #Self-published sources, again above: I see broad agreement that the "self published" section is BLP-related (even if Zak's agreement is maybe not too clear above on this page, he later expressed agreement, on the WP:RS talk page);
• Per #Self-published sources in articles about themselves: there seem to be no further objections to group the "self published" section with the BLP section;
• Per this section (#Stability): performing the operation does not involve any "deletions". --Francis Schonken 13:17, 23 June 2006 (UTC)

How far does this policy go?

I'm trying to revert changes on the telekinesis page. Another user had deleted the entire cultural references section because each fact was not verifiable. Through discussing this with him on the talk page, he appears to beleive that the policy mandates every fact have a reference outside of Wikipedia.

Now, I think that goes a little too far. It would mean that statements such as "The sun is a star in the center of the galaxy" and "the sky appears blue in color" have links to scientific documents coroborrating those claims. Liu Bei 17:16, 22 June 2006 (UTC)

Every edit that he is challenging must be sourced to a publication outside Wikipedia. He shouldn't make challenges in bad faith, but misusing the policy to be disruptive is quite unusual. Mostly, if people request a source, it does mean the claim is contentious, and it's often faster to find a source that to argue about it. SlimVirgin ^(talk) 17:56, 22 June 2006 (UTC)

I looked at the edit [13] and it's about movies and video games, all pretty standard stuff. It would be easy to find sources to back up these claims, but the editor making the challenge could do it too. I wouldn't myself ask for citations. We often have this problem with movies and tv shows: people watch them and write down what they've seen, and while there probably are sources out there that could confirm what they're writing, it's not clear there's a benefit to tracking them down. SlimVirgin ^(talk) 18:02, 22 June 2006 (UTC)

So what's the solution here, just ignore the frequent blanking of popular culture sections from articles accompanied by chants of "unencyclopedic"? Even if I had the information on-hand, adding it to the references section would be retarded. It would end up stretching 3 or 4 screens worth of one-line links to pages that are likely less permanent that wikipedia. Liu Bei 18:09, 22 June 2006 (UTC)

As the section you want to keep is basically a trivia section, why not let it go? Edit something more substantial. KillerChihuahua^?!? 18:16, 22 June 2006 (UTC)

Because policies like this are making me realize being active on Wikipedia is a futile endeavor. If that amount of work can be erased because someone does not like minutae or pop culture references to "sully" the article, then it's just not worth it.

I have to agree that it's not particularly encyclopedic. Sorry, Lui Bei. SlimVirgin ^(talk) 18:18, 22 June 2006 (UTC)

I somewhat disagree. What is trivia to one person is a valuable research resource for someone else, and it's not our job to place artificial boundaries on what people can draw from the sources they're pointed to from Wikipedia. Consider the Flat Earth article, which devotes a lot of text to debunking the myths and misconceptions perpetuated by pop culture, as captured in the statement During the 19th century, the Romantic conception of a European "Dark Age" gave much more prominence to the Flat Earth model than it ever possessed historically. Perhaps that statement slips a bit into original research, but regardless, the pop culture references there do provide significant context for / contrast to the relatively scholarly sources about the topic's history. I can't imagine a reference librarian, whose job is basically to point people to reliable sources, would stop at only scholarly references and would withhold pop culture references from a researcher who specifically wanted to sort out which aspects of a topic's history were fact and which were fiction. And when the topic is a pop culture phenomenon itself, often those references are the only sources of information and are as reliable as one can get (until something better comes along). Sometimes they amount to clutter, but I see no harm in pointing people to them, and I see no reason they must first be mentioned in the Washington Post — saying "The 1977 film Star Wars contains a depiction of a form of telekinesis" seems benign, verifiable, and potentially important to me, whereas saying "Star Wars got telekinesis all wrong and here's why" is not.—mjb 23:04, 22 June 2006 (UTC)

If it were that kind of list (here's how Star Wars got it wrong), it would be useful, but when I last looked it was mostly edits like: "In the game Second Sight, the hero, some bad guys, and several children display various psycokinetic abilities," and it was a very long list. SlimVirgin ^(talk) 23:32, 22 June 2006 (UTC)

Recently, I found myself drawn into an article where I had intended not to go. It made an unsubstantiated claim that was open to misinterpretation. I explained my reservations but later I expressed my appreciation to an editor who had found a verifiable source. That article is now stronger for that dialogue. What we are doing here is not easy. Nothing worth doing is ever easy. On top of this, we do not know if it will work or if it will last. Each edit that we make, we make in good faith and in the belief that we are creating something rather good. We might find some claims and challenges uncomfortable or unrewarding, but they are part of what makes Wikipedia uniquely special and worthwhile. -- Alias Flood 00:03, 23 June 2006 (UTC)

Yes, indeed. :-) SlimVirgin ^(talk) 02:29, 23 June 2006 (UTC)

Way to go, Alias ! heh. Terryeo 15:54, 23 June 2006 (UTC)

Self-published sources and BLP

There seemed to be some confusion about what a self-published source is, with some people thinking it was necessarily a living person writing about him- or herself. It needn't be, but might also be a small organization without staff writing about itself. I've therefore made a few changes to clarify that. [14] SlimVirgin ^(talk) 20:58, 24 June 2006 (UTC)

Anyway I tried to find the roots of that section:

Created by Slimvirgin 04:30, 5 February 2006

Erroneously called "fleshed out this section a little" by SlimVirgin. That edit created 3 new sections, about quadrupling the size of the so-called fleshed out section. The new "Self-published sources as primary sources" section did not contain any material previously in the so-called fleshed-out section. No indication that this was copied from WP:RS at the time (as SlimVirgin now purports). I don't think much time was spent on talk that day: this was created amids edit-warring and page moves, the day after jguk's version had gone life. Here's the content:

=====Self-published sources as primary sources=====

Self-published sources and other sources of dubious reliability may be used as primary-source material in articles about that source. For example, the Stormfront website may be used as a source about itself in an article about Stormfront, so long the information is appropriate, not unduly self-aggrandizing, and not contradicted by third-party sources. Subject to the exceptions above, such a source should never be used as a third-party source about anyone or anything else.

A Wikipedia article about an unreliable newspaper should not — on the grounds of needing to give examples of their published stories — repeat any claims the newspaper has made about third parties, unless the stories have been repeated by credible third-party sources.

Subsection level modified by SlimVirgin 01:27, 6 February 2006

Section removed by an anon 01:46, 7 February 2006

Rv by SlimVirgin 01:48, 7 February 2006

Next edit - section rewritten by SlimVirgin 06:39, 7 February 2006

Here the reference to "primary sources" in the title is removed and replaced by "in articles about themselves". There is no indication these modifications resulted from WP:RS or from talk page discussions (edit summary: " deleted references to "primary" sources as it may be confusing"). Here's the content of that version:

===Self-published sources in articles about themselves===

Self-published sources and other published sources of dubious reliability may be used as sources in articles about themselves. For example, the Stormfront website may be used as a source about itself in an article about Stormfront, so long as the information is notable, not unduly self-aggrandizing, and not contradicted by other published sources.

A Wikipedia article about an unreliable newspaper should not — on the grounds of needing to give examples of their published stories — repeat any claims the newspaper has made about third parties, unless the stories have been repeated by credible third-party sources.

-later on in February there is some general vandalism (& reverts) not listed here; first half or March pretty much the same sight-

Minor addition by Zero0000 07:27, 18 March 2006

SlimVirgin reverts to older formulation 07:29, 18 March 2006

again "not contradicted by reliable, third-party published sources." instead of "by other published sources."

Immediately followed by another slight rewording by SlimVirgin 07:32, 18 March 2006

-early April again some vandalism & rv involving that section, not listed here-

New rewrite of the first paragraph of the section by SlimVirgin 02:01, 18 April 2006

Here is where the "Self-published sources may never be used as sources of information on another person or topic" sentence gets added. Rather deceptive edit summary: "tidied English". This was a change of the policy, as far as I can see not resulting from talk page discussion, nor from import of material from WP:RS.

-early May again vandalism & rv-

Rearrangement by SlimVirgin 12:32, 23 May 2006

Edit summary: "added bullets" (no talk page discussion indicated; no indication this had anything to do wit WP:RM). Nonetheless the rewrite involved indicating all "self-published" sources as "sources of dubious reliability". Further, "Stormfront" example removed, and three bullets with new material added, "not contentious", "no claims unrelated to te subject" and "no doubt about author":

Material from self-published sources, and other published sources of dubious reliability, may be used as sources of information about themselves in articles about themselves, so long as:

• It is relevant to the person's notability;
• It is not contentious;
• It is not unduly self-serving;
• It is not contradicted by reliable, third-party published sources;
• It does not involve claims about third parties, or about events not directly related to the subject;
• There is no reasonable doubt about who wrote it.

Self-published sources may never be used as sources of information about another person or topic.

A Wikipedia article about an unreliable newspaper should not — on the grounds of needing to give examples of their published stories — repeat any claims the newspaper has made about third parties, unless the stories have been published by other credible third-party sources.

Theronj adds caveat for consistency 22:35, 30 May 2006

"subject to the limited exceptions discussed above" added

Rearrangement by Mjb 21:23, 31 May 2006

involves subsection level change and "for example..." insertion in first paragraph of the section.

Revert to an earlier version by Francis 22:34, 31 May 2006

overlooking the Theronj caveat (...my fault)

-The Mjb changes reoccur on 3 june, again reverted by Francis-

Section moved to WP:RS by Francis 12:25, 3 June 2006

with a clear link to ther relevant section in WP:RS, edit summary: "move one of the two sections on "self-published sources" to WP:RS, and deep-link to that WP:RS section. See talk." Note, also discussed on talk page.

Section reinserted by Gimmetrow 20:40, 3 June 2006

Cut of one paragraph by Percher 21:21, 3 June 2006

the "may never be used as sources of information about another person or topic" paragraph

Reinserted by Gimmetrow 21:29, 3 June 2006

Whole section again removed by Francis 07:43, 10 June 2006

Edit summary: "[...] rv inclusion of guideline-level description that is included in WP:RS"

Reinserted by SlimVirgin 18:26, 10 June 2006

SlimVirgin moves "unreliable newspaper" paragraph to other section 18:35, 10 June 2006

-later in June general vandalism and reversions-

Francis again removes section 9:08, 19 June 2006

edit summary: "rem. "Self-published sources in articles about themselves": guideline-level (WP:RS); further: incomplete, in present form not universally applicable"

rv by Jayjg 16:24, 21 June 2006

Marskell adds sentence 16:56, 21 June 2006

added "As with dubious sources generally, self-published sources should be specifically attributed." Edit summary: "agreed this is needed. reiterate point about attribution"

Again removed by Francis 21:21, 21 June 2006

Edit summary "rv to last version by Francis Schonken: "Self-published sources in articles about themselves" subsection not part of non-negotiable policy, doesn't even mention valid exceptions, see WP:RS"

Jossi reverts to Jayjg 21:23, 21 June 2006

Francis adds full list of caveats 21:30, 21 June 2006

includes the Theronj caveat mentioned above, and two others derived from WP:RS, with links to relevant sections. So the end of the section now reads:

Self-published sources may never be used as sources of information about another person or topic, subject to the limited exceptions discussed above and in Wikipedia:Reliable sources. These exceptions are:

• Limited use of references to primary sources as explained in the Personal websites as primary sources section of that guideline;
• With caution, expert opinions expressed by experts in their field of expertise, as explained above and in Reliability of online sources of the Reliable sources guideline;
• When reporting that an opinion is held by a particular individual or group, the best citation will be to a direct quote, citing the source of the quote, as explained in Reporting about opinions: use of direct quotes of the Reliable sources guideline.

Additions deleted by SlimVirgin 21:46, 21 June 2006

edit summary: "rv to last version by Jossi; please discuss these proposed additions on talk first" Note that at this point the section again contains *exclusively* material written by SlimVirgin, without an indication of prior talk page discussion, and without anything that could link this to prior content on WP:RS, and neither to WP:BLP.

Section again deleted by Francis 22:17, 21 June 2006

Edit summary: "remove no-consensus version of "Self-published sources in articles about themselves". Only content of "non-negotiable" level should be in this policy."

-From this point there are no further deletions by Francis, only deletions of others, as described above -

Six edits of refactoring by SlimVirgin 20:45 to 20:53, 24 June 2006

Again, only content by SlimVirgin

Indeed, there is no link to WP:BLP. I don't see a link to WP:RS either. I don't see SlimVirgin's text and refactorings resulting from talk page discussions. Maybe I'm wrong. SlimVirgin, could you indicate where you derived this from WP:RS, as you now contend on the WP:RS talk page? Could you show me there was enough talk page consensus for your wording to be inserted in this policy page? For example, above I see Septentrionalis/Pmanderson saying we should go back to *primary source* wording. I didn't see you reply to that, only pressing your own wording directly on the policy page, without prior talk. The note you left here (the beginning of this section) was only written after your last series of six edits.

Could you give some more clarity about your mode of operation? --Francis Schonken 23:44, 24 June 2006 (UTC)

Francis, sometimes less is more. What (briefly) is your point? SlimVirgin ^(talk) 23:51, 24 June 2006 (UTC)

Could you give any information how the (former) "Self-published sources in articles about themselves" section derived from WP:RS? In my searches above I couldn't find a shred of evidence of that, despite your contention in that sense. Did I miss something, please point me there then, tx. --Francis Schonken 23:59, 24 June 2006 (UTC)

Also your last set of six refactorings are not supported by prior talk page discussion, as far as I can see. Maybe, again I missed something, could you then point me there. But for avoidance of destabilisation of the policy I'm going to revert until there is agreement. --Francis Schonken 23:59, 24 June 2006 (UTC)

I have no idea what you're talking about. These policies evolve over many months and years of experienced editors and admins with tens of thousands of edits each discussing on policy and guideline talk pages, on the mailing list, and on article talk pages, frequently editing and watching the policy pages, learning how they hang together in theory and in practise, developing them, tidying them, sometimes tightening them, sometimes loosening them. It's an organic process, and as the editors gain more experience, more loopholes are closed, or opened if they're good ones. But someone does have to write them down at some point, yes. You seem to think they can be intuited onto the page. I can't argue with you about this anymore, because I don't know what your substantive point is. You are the one who keeps trying to remove material, and you can tell by the fact that people keep restoring it that there is consensus for it to be here. This page is watched by a lot of people. SlimVirgin ^(talk) 00:11, 25 June 2006 (UTC)

"I can't argue with you about this anymore" is not taken well by me. --Francis Schonken 00:26, 25 June 2006 (UTC)

I just wish to note some historical revisionism around June 3 in Francis' account, when I was involved. There was no "discussion" in talk here - Francis made a bold edit and simply mentioned it in talk at the same time. I restored the deletion. Gimmetrow 00:43, 25 June 2006 (UTC)

What is this all about? Why so many diffs persented as "evidence". Evidence of what, exactly? It is about time that editors be more candid and direct. If an editor has a grudge against another, it would be much easier if these are resolved candindly rather than by "evidence" of diffs. This is not an ArbCom case. It is becoming quite annoying and tedious to have to deal with so much animosity, rather than amicably discuss how this very important policy can be improved, if at all. ≈ jossi ≈ t • @ 01:58, 25 June 2006 (UTC)

I have no idea what the point is either, I just don't appreciate being misrepresented. Gimmetrow 03:23, 25 June 2006 (UTC)

Definite versus indefinite article

Terry and Kenosis, can you say why you prefer "a threshold" to "the threshold." It seems to me that the latter is accurate. There is one threshold for inclusion, and that is "verifiability by reference to a reliable, published source." SlimVirgin ^(talk) 01:43, 25 June 2006 (UTC)

Must be The and not A, otherwise we need another policy to describe all other thresholds... ≈ jossi ≈ t • @ 01:46, 25 June 2006 (UTC)

And one is trouble enough. :-D SlimVirgin ^(talk) 01:47, 25 June 2006 (UTC)

I have no preference. There were two issues going on ([15]), one a revert of Francis Schonken's last edit, the other a proposed change by Terryeo. If it was Crum375's intent to revert both, he did not say so. Best that Terryeo's edit was dealt with on its substantive merits, or lack thereof. ... Kenosis 01:56, 25 June 2006 (UTC)

Introduction, defining verifiability

First of all, changes to such a major policy should not go from proposal to implementation in only a week. Consensus requires the opportunity for people to discuss, and is not reached by rushing. The matter at hand:

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that aAny reader must be able to check that material added to Wikipedia has already been published byin a reliable source, because. Wikipedia does not publish original thought or original research.

This policy does not need to redundantly define verifiability. The meaning of verifiability used here is quite common; it is in fact the dominant meaning of the word. The following sentence clearly and unequivocally states what is meant by this policy, without needing circumlocutions. —Centrx→talk • 03:44, 25 June 2006 (UTC)

Verified normally means confirmed as accurate, whereas we simply mean confirmed as sourced, so it's a particular usage of the term, and it does need to be explained. SlimVirgin ^(talk) 05:28, 25 June 2006 (UTC)

It is explained. The sentence that immediately follows the first describes exactly what is meant by this policy. My objection is to the excess at the beginning of the sentence, which is similar to the poor article format like "Bed is the name of a piece of furniture" rather than "A bed is a piece of furniture".

Note also that the meaning of "verifiable" is "That can be verified or proved to be true, authentic, accurate, or real; capable, admitting, or susceptible of verification" (OED), which is exactly what is meant here. Verifiability is to allow authentication, to substantiate a statement as real, and the Verifiability policy establishes how that is to be done on Wikipedia. —Centrx→talk • 05:52, 25 June 2006 (UTC)

But, as you note above, "Proved to be true" is part of the OED definition. Now, most of us have no problem realizing that what is to be proved is that an assertion fairly represents what is said by a reliable source, rather than the truth-value of the assertion itself. (Or, equivalently, we believe that the two are the same in the context of Wikipedia.) On the other hand, the issue has been raised at least twice here, and so some harmless clarification seemed reasonable. What, other than a sense of "well, duh!" bothers you about the attempt to avoid this misunderstanding? Robert A.West (Talk) 06:33, 25 June 2006 (UTC)

First, proved in this case means tested or trustworthy, not demonstrated irrefutably to be true. Verifiability is a threshhold that ideally exceeds truth. Reliable sources are the way we test for truth on Wikipedia, our interface for verifying accuracy.

The statements should be as direct as possible, and I do not see any possibility of confusion with the more direct "Any reader must be able to check...". Anyone who gets a wrong impression by only reading the first sentence is not going to read the "Verifiable in this context" part any more than they will read the "Any reader must be able to check..." —Centrx→talk • 06:59, 25 June 2006 (UTC)

I agree, but no matter how straightforward is all seems, there's always someone who finds an issue to question. I feel we should go back to the version before the recent changes to the intro, and I agree with those who say there's too much editing of this page at the moment. The bottom line is that we can't legislate for every eventuality. At some point, common sense has to kick it. SlimVirgin ^(talk) 07:04, 25 June 2006 (UTC)

Common sense? If we could rely on that, we would need nothing beyond the policy-in-a-nutshell. :-( Robert A.West (Talk) 17:16, 25 June 2006 (UTC)

SlimVirgin, several days ago we informally proposed to backtrack to the last stable version - I think it's this one:

The threshold for inclusion in Wikipedia is verifiability, not truth. This means that we only publish material that is verifiable by referring to reliable, published sources.

So far, I have seen no opposition. Do you think this needs to be a formal proposal? AvB ÷ talk 19:38, 25 June 2006 (UTC)

I see no opposition to changing it back. By only concern would be that it should be "verifiable by reference to" because "by referring to" has a different meaning, but I'm prepared to discuss that quietly and at length, rather than hold up restoring the sentence because of it. :-) I'm also thinking of requesting page protection to allow the page to calm down.

Does someone else want to restore the old sentence or shall I? SlimVirgin ^(talk) 00:12, 26 June 2006 (UTC)

I am opposed to changing it back. I contest Centrx's claim that "The meaning of verifiability used here is quite common; it is in fact the dominant meaning of the word." I would like to see some evidence of that! There was a great deal of discussion here about the need for the change, and I thought that even SlimVirgin agreed that Wikipedia's use of the term did not carry the same connotations with truth and correctness as the term usually has when applied in the real world.

It is nearly nonsense to say "The threshold for inclusion is foo, not commonly assumed aspect of foo" with no clarification. It's like saying "I want to date someone who is a little bit crazy, not a little bit suffering from some kind of psychosis". It would make more sense to say "I want to date someone who is a little bit crazy; I mean, crazy in the spontaneously fun sense, not crazy in the severely mentally disturbed sense".

Further, "one week isn't enough time for consensus" is ridiculous; the people currently interested in that part of the policy all voted on it, and what few objections were raised were addressed. That level of attention exceeds what usually passes for due process around here, so I disagree with this part of Centrx's argument as well. —mjb 22:47, 26 June 2006 (UTC)

Actually, the time between the first appearance of the current wording on Talk and its implementation on the policy page was two (2) days and, only a couple of hours after it was implemented, we find an opposition to it and further discussion about it. That is, it was still in formation.

The meaning of verifiability is still clarified in what I propose, it just doesn't go through the gymnastics of saying "Verifiability in this context means". The proper analogy would be saying "I want to date someone who is a little bit crazy, someone who is spontaneously fun." without the excess of saying "I mean...in this sense, not that other sense."

The following dictionary meanings are not necessary for this argument to hold. "Verifiable" means 'able to be verified'. The way in which statements are verified on Wikipedia is by the rules of evidence on Wikipedia, both here and at Wikipedia:Reliable sources. Here are the meanings in the Oxford English Dictionary (2d ed. 1989) with relevant quotations:

• Verifiable: That can be verified or proved to be true, authentic, accurate, or real; capable, admitting, or susceptible of verification.

This is the only definition. Note that prove means to demonstrate or test by argument or evidence, not the mathematical or tautological or perfect meaning.

• "Neither Homer nor Hesiod mentioned any verifiable present persons or circumstances.", that is, persons or circumstances which can be checked in some historical record.
• "While the abstract possibility of a revelation is admitted, its verifiableness is in effect denied.", that is, while revelations may be possible, we have no way to check that they actually happen.
• "Pending an agreement that equitably and verifiably limits bombers, the B-1 is indispensable to help maintain the balance of strategic forces.", that is, we can check that there are a certain limited numbers of bombers.
• "If the audits are undertaken, verifiably accurate reports could be made public in September.", that is, we can check ourselves that the reports are accurate by going to the source.

Note that "verifiability" means only "the fact of being verifiable". There are also the related words, verify and verification. I will not list quotations for these because it would be too lengthy and tiresome. Suffice it to say that the OED derives its definitions from carefully selected quotations.

• Verify:

• Law. To prove by good evidence or valid testimony; to testify or affirm formally or upon oath.

That is, good, valid evidence by a certain procedure. The law also does not pretend access to perfect, eternal truth. Though this is primarily legal, if we are to accept the argument in the above discussion that etymology is important, this is the original English meaning from which the others are derived.

• To show to be true by demonstration or evidence; to confirm the truth or authenticity of; to substantiate.
• To ascertain or test the accuracy or correctness of (something), esp. by examination or by comparison with known data, an original, or some standard; to check or correct in this way.

• Verification:

• Demonstration of truth or correctness by facts or circumstances.
• The action of establishing or testing the truth or correctness of a fact, theory, statement, etc., by means of special investigation or comparison of data.

Webster 1913 and Webster's New World College Dictionary, 1999, based on the principle International, also concur with these meanings. Webster 1913 more strongly concurs, with its only meaning being "Capable of being verified; confirmable." —Centrx→talk • 04:16, 27 June 2006 (UTC)

I made essentially the same argument against Light Current for quite some time, so I generally agree. On the other hand, I see nothing wrong with leading people by the nose -- it seems to be necessary sometimes. I would rather have a phraseology that can be understood by 98.98% of readers than one understood by 98.66%. Robert A.West (Talk) 04:43, 27 June 2006 (UTC)

Point taken re: the number of days, although I suspect you wouldn't object to a two-day turnaround on your proposal. :)

Of course verifiability, verifiable, and verification are derived from verify, and their relation to that word does not need to be explained. Fortunately, no one is proposing that such an explanation be made, and the current wording does not contain one.

I do agree that the clarification of which sense of 'verify' is meant is preserved in your version. The problem is that AvB and SlimVirgin are declaring that there is "no opposition" to discarding both the current version and your trimmed version, and reverting all the way back to the "last stable version", which I feel was only "stable" because a small number of editors aggressively reverted attempts to rephrase it and were dismissive of discussion, not because newcomers consulting the policy found it to be crystal clear and agreed with its phrasing.

If I can take a step back, I think my beef is not with your trimming of the revised text, but with the total reversion and the "verifiability, not truth" phrase which is present in all versions. That's what initially tripped me up when I came here, because, as it is repeatedly pointed out, "to verify" can and often does involve ascertaining truth. I'd wager that most editors who have had WP:V thrown at them, or who want to throw it at someone who didn't cite reliable sources, assume that it usually or always has such a correlation, and the recent discussion here does seem to indicate that perception is accurate. So when the policy says, essentially, that readers need to be able to verify statements, not ascertain their truthfulness, it sounds contradictory, and thus is not very well phrased.

The "any reader must be able to…" appositive, along with the "Verifiability, not truth" section that it ostensibly summarizes, do help to clarify, but that initial phrase is still awkward and counterintuitive, despite being accurate in that truth is not the threshold for inclusion, (a certain kind of) verifiability (specifically, the truth-agnostic, published-in-reliable-sources type of verifiability) is. I feel that reverting to the old version is a step backward. There must be a way to phrase it without implying that verifiability and ascertaining truth are mutually exclusive. —mjb 08:50, 27 June 2006 (UTC)

This is the first (and so far only) response opposing the proposed reversion to the last stable consensus version of the first paragraph of the lead section. Your opposition could do with some fleshing out, such as addressing the reasons given above for the proposal. In summary, the departure from the older text (language backed by long years of consensus) brought about a chain reaction of repairs that ended in a dispute that could not be resolved or repaired in the text (about interpretations by veteran Wikipedians of what constitutes a reliable source). I originally supported the changes to some extent, including the one by Francis (from "by" to "in"). After some discussion I noticed that the old text was sufficiently fuzzy to accommodate both interpretations and became convinced we need that amount of fuzziness here because the community consensus is fuzzy on the subject. This means that we must backtrack or otherwise solve the problem that there are two acceptable interpretations doing the rounds. From there we can once again discuss the problems editors have with the old text. I should add that I have not changed my opinion about the speed with which those changes were adopted. In fact I think they should be reverted for that reason alone. Too fast folks. Thanks. AvB ÷ talk 09:29, 27 June 2006 (UTC)

Personally, I think the wording in December [16] was much better, without trying to sound like a clever proverb. —Centrx→talk • 18:46, 27 June 2006 (UTC)

I happen to know that version well: it's the one I studied as a newbie editor. I think I need to review the history to get a better understanding of what has happened here. For now, I still prefer this more recent version. AvB ÷ talk 09:17, 28 June 2006 (UTC)

Wikipedia:Attribution

I almost hate to suggest this because I think we all agree there has been too much change recently. But I've been toying with the idea of suggesting we rename the policy to Wikipedia:Attribution. There would be two advantages:

(1) It would get rid of the confusion over verifiable, which is related to truth, whereas our policy is not; and

(2) It would answer a question often asked by newbies (and some oldies) viz. does verifiability mean that all edits should in theory be sourced. If we think in terms of "attribution," then we can make clear that all edits must indeed be "attributable," but not all edits must actually be "attributed."

I know that "A not T" doesn't sound as good as "V not T," but apart from the loss of our catchphrase, I think the change would be beneficial. However, I know there has been too much back and forth, so I'm mentioning this here only to give people something to mull over for the future. SlimVirgin ^(talk) 08:59, 27 June 2006 (UTC)

Only after saving this did I see that title already exists, but I think we could ask to be allowed to use it. SlimVirgin ^(talk) 09:01, 27 June 2006 (UTC)

To weigh in here, the reason we are asked to supply sources is to enable the reader to verify that any claims made in an article are indeed truly supported by the sources they are derived from. Note that de.wikipedia calls its equivalent to WP:V de:Wikipedia:Quellenangaben, a suitable translation would be "Adequate sourcing". Dr Zak 12:41, 27 June 2006 (UTC)

I know what you're thinking Slim, but I believe this change would leave the policy open to exploitation:

• "You've attributed this to a bogus, ridiculous source which does no peer review/editorial oversight!"
• "Yes, but I've still attributed it."

All that a person will need to prove is that the source exists, not that it's a source that does its own (or can be reasonably assumed do its own) verification. Verifiability is higher bar than simple attribution. Marskell 22:03, 27 June 2006 (UTC)

• "You've verified this in a bogus, ridiculous source which does no peer review/editorial oversight!"
• "Yes, but I've still verified it."

Same problem I guess. AvB ÷ talk 09:01, 28 June 2006 (UTC)

I know it would be "attributable to a reliable source" (if my example seemed a little extreme). I just wonder if it wouldn't create another point to argue over. Marskell 07:20, 28 June 2006 (UTC)

I think "attributable to a reliable source" does not fit current practice where statements/etc. published in a reliable source are attributed to the person/organization/etc. that has made the statement. It takes the "in/by" problem one step further since it exclusively matches the special case described by SlimVirgin where statements/etc. are attributed to the source because the person/etc. is authoritative enough to be considered a reliable source and any publication will do.

The text would be more along the lines of the fuzzier "attributable with reference to a reliable source" or "attributable by citing a reliable published source", or the more explicit/limited "attributable to a person/group by citing a reliable publication". AvB ÷ talk 09:01, 28 June 2006 (UTC)

• "You've verified this in a bogus, ridiculous source which does no peer review/editorial oversight!"
• "Yes, but I've still verified it."
• "It's not possible to verify something with such a source."

But it is possible to attribute something to such a source--thus I think we should leave it with the current title. Marskell 10:47, 28 June 2006 (UTC)

Yes, I see what you mean, Marskell. AvB, I'm not sure I see how the in/by ambiguity would be made worse. But I'm not pushing for changing the title anyway. It was just a suggestion. Consider it withdrawn. :-) SlimVirgin ^(talk) 11:31, 28 June 2006 (UTC)

Regarding the in/by/etc. ambiguity: I mainly mentioned this because Marskell's "attributable to a reliable source" reminded me of these changes to the NPOV Tutorial that were reverted with this rationale.

Marskell, this clarifies your point. It pretty much kills the "attribution" proposal. However, it also does a splendid job showing why replacing "verifiability" with a better word remains important. Your example rests on the understanding that "verifiability" implies "reliability" and suggests "Truth". Exactly the mainstream variety of "verifiability" not meant by the policy. AvB ÷ talk 13:31, 28 June 2006 (UTC)

I don't agree with you there, AvB. I think the version you wanted to revert was better. You're right that, if the NYT reports: "Israeli troops entered Gaza on Wednesday," and no one is disputing it, we can write "Israeli troops entered Gaza on Wednesday" (followed by footnote to NYT), but without an in-sentence attribution. But if there is a dispute, or if the source is not as authoritative as the NYT, we should say: "According to X ..." and still give a full citation in a footnote." (But that doesn't really touch on the by/in ambiguity.) SlimVirgin ^(talk) 13:44, 28 June 2006 (UTC)

I'll think this over when I'm feeling less fuzzy and ambiguous myself ;-) Thanks. AvB ÷ talk 19:01, 28 June 2006 (UTC)

Just to make sure I understand you correctly - when considering these versions:

Attribution and citation An attribution specifies who stands behind a claim. In this example: "According to [Rock and Roll Survey 1998] most Australians think The Beatles are the best rock music group ever." the sentence attributes to "[Rock and Roll Survey 1998]" the claim that most Australians think the Beatles are the best group ever.

and:

Attribution and citation An attribution specifies who stands behind a claim. In this example: "According to most Australians, The Beatles are the best rock music group ever [Rock and Roll Survey 1998]" the sentence attributes to "most Australians" the claim that the Beatles are the best group ever.

- which one would you prefer? Thanks. AvB ÷ talk 13:39, 30 June 2006 (UTC)

PS SlimVirgin's "attribution", although probably unusable, did come very close. We don't verify truth, material or information. We verify attributions. The NPOV tutorial puts it very clearly: "A citation tells readers where they can look to verify that the attribution is accurate." AvB ÷ talk 13:40, 28 June 2006 (UTC)

Yes, exactly. SlimVirgin ^(talk) 13:44, 28 June 2006 (UTC)

Sourceable? Just a thought. When it's used on article talk pages, editors usually mean, or take it to mean, "verifiable," Wikipedia style. AvB ÷ talk 12:38, 30 June 2006 (UTC)

Ten changes a day?

That is what this policy has been averaging for the past week. Isn't this far too many edits? And some that are marked "minor" don't strike me as minor. AGF and all, but this observer thinks that slowing down would be a very good idea. Robert A.West (Talk) 05:07, 25 June 2006 (UTC)

I agree. Even obscure MOS pages have more discussion and consensus than this page, and this is major policy. —Centrx→talk • 05:25, 25 June 2006 (UTC)

Agreed. SlimVirgin ^(talk) 05:25, 25 June 2006 (UTC)

Hear, hear! AvB ÷ talk

I agree. Like most good things, those that are worthwhile and longlasting take time. Consensus is the lifeblood of Wikipedia and a commodity far too precious to be spilled through haste. -- Alias Flood 23:38, 25 June 2006 (UTC)

I would argue that there needs to be a specific need, raised by a specific concern, in a specific situation presented by an editor, with a clear proposal for modifying policy that has served us well for several years and for which there is undeniable consensus, before we contemplate such modification. These haphazard edits and word-smithing are a royal waste of editors' time. We have an encyclopedia to write. ≈ jossi ≈ t • @ 23:51, 25 June 2006 (UTC)

• Hear, hear! Robert A.West (Talk) 00:27, 26 June 2006 (UTC)

• Thirded! SlimVirgin ^(talk) 00:43, 26 June 2006 (UTC)

• Fouthed. Jayjg ^(talk) 00:54, 26 June 2006 (UTC)

• Agreed. --Elonka 06:20, 26 June 2006 (UTC)

• Agreed, sort of. I am not strongly opposed to protection, but the reasoning you give for it is bogus. The fact that there are so many edits and so much debate, explanation and clarification having to be given on the talk page indicates that this consensus is not as "undeniable" as you'd like to believe. Assuming good faith, people keep attempting to change the policy and how it is expressed because it is not perfect and it is not unambiguously expressed with perfect clarity. And it has served us well for "several years"? Please! Look at what the policy was two years ago. How about one year ago! The policy has changed substantially since those versions. Or are you trying to tell us that

"The goal of Wikipedia is to become a complete and accurate encyclopedia. Verifiability is an important tool to achieve accuracy, so we strongly encourage you to check your facts. We also aim to be informative and neutral."

has always meant exactly the same thing as

"The threshold for inclusion in Wikipedia is verifiability, not truth."

? Regardless, protecting the page is a good idea just because newcomers don't realize that the current editors insist on prior discussion of changes due to the sensitivity of the content and ramifications of changes. —mjb 01:11, 27 June 2006 (UTC)

Good point re policy text evolution(s) Mjb. Not quite as stable as I seemed to remember. AvB ÷ talk 09:04, 28 June 2006 (UTC)