Wikipedia talk:Verifiability/Archive 7

Archive 1

←

Archive 5

Archive 6

Archive 7

Archive 8

Archive 9

Archive 10

Really, I find that Light current's view is not so much in opposition with what we are doing. Please read the section #The role of truth in WP policy. --Lumière 07:22, 6 February 2006 (UTC)

Lumiere, perhaps this discussion could be continued on a subpage? I've set one up at Wikipedia talk:Verifiability/truth in case anyone would like to do that. Cheers, SlimVirgin ^(talk) 07:30, 6 February 2006 (UTC)

Sure, but the section #The role of truth in WP policy should stay here because I think it is interesting for every one. I will copy, not cut and paste, the part that I think is useful. Others could do the same. --Lumière 07:33, 6 February 2006 (UTC)

Is anyone who disagrees with the status quo a TROLL?--Light current 05:44, 7 February 2006 (UTC)

The role of truth in WP policy

DO NOT REMOVE THIS SECTION EVEN IF IT DISTURBS YOU.

Light current, I have a few related questions for you. Is it that you carefully looked at the policy itself, i.e., the rules that must be used to resolve disputes and be as much as possible interpreted in the sameway by all kind of editors, and you feel that these rules are not very good? Or is it that you find that the way the policy is presented is not very good? I personally have a problem with both aspects, but I try to work with people to improve the situation. Perhaps, my suggestion would be to separate these two aspects and try to show some appreciation for an attempt to have practical rules to obtain a content of good quality in Wikipedia. I don't think the current attempt is stupid, but it can be improved. If you think that this rule aspect per se, not the presentation in terms of no truth, etc., is not good at all, could you explain why? --Lumière 01:27, 6 February 2006 (UTC)

Sorry Im not clear which policy you are referring to. Is it the WP:CIVIL or something else?--Light current 01:52, 6 February 2006 (UTC)

I am referring to the content policies WP:verifiability, WP:neutral point of view and WP:No original research. Wow, I see that you did not get my paragraph at all. Can you read it again and see if you understand it better?

BTW, I read the paragraph again myself and I can see that your interpretation makes sense, especially in the context of all these issues with persons instead of Wikipedia contents just above. I was not interested in these issues at all. I am interested in the policies that deal with the content of Wikipedia. --Lumière 02:20, 6 February 2006 (UTC)

I have not looked that carefully at the WP:verifiability policy. I didnt need to, as the opening sentence was enough to raise very strong feelings (ie the bit about V NOT T). This to me is just not right that we should blatantly be pushing things that are not true without saying so or at least providing a substantial warning. So, I feel this policy should be changed to say True (as far as we know) AND verifiable. So its the rule thats wrong, not the presentation. Hope that explains my position.--Light current 02:31, 6 February 2006 (UTC)

Ok, but what is the actual rule that you propose for the T part. Is your rule "any one that thinks something is not true can delete it, and no one can oppose to it". That seems no good. Too much valid information can be suppressed. Or perhaps it is "if a consensus amongst the editors consider that something is not true, then they should not include it, even if it is verifiable." You know what, I like this last rule. I see no problem with it in practice. In fact, even if it is not formulated in the policy, I think it is applied in practice. On the other hand, I would not accept the converse: "if a consensus amongst the editors consider that something is true, then they can include it, even if it is not verifiable." The problem with this last rule is that consensus are local amongst a few editors. Was this the only issue that did bother you? --Lumière 02:51, 6 February 2006 (UTC)

ie What is the definition of truth? Working.....please wait

First part answer:

"if a consensus amongst the editors consider that something is not true as far as they have ascertained to the best of their endeavours, then:

• they should not include it, even if it is verifiable OR
• it should be indicated in the article that it is not thought to be true (but not necesarily thought that it is not :true either. ie presented in a reported manner. (It is said by many/some that so&so is a raving...)

What I don't like is that the second option can be used to accept for inclusion about anything inside an article on a given topic. This is not respectful for the topic. To improve the quality of Wikipedia would you accept

"if a consensus amongst the editors consider that something is not true as far as they have ascertained to the best of their endeavours, then:

• they should not include it, even if it respects all other requirements of WP policy OR
• if it respects all other requirements of WP, they can include it, but it should be indicated in the article that it is not thought to be true (but not necesarily thought that it is not :true either. ie presented in a reported manner. (It is said by many/some that so&so is a raving...)

This is more safe to preserve the quality, but it does suppress more information. --Lumière 03:33, 6 February 2006 (UTC)

Yes well that fits in with my exclusionist policy nicely!--Light current 03:34, 6 February 2006 (UTC)

Just realized that there is still a problem with this rule. We are talking about a content that was published in a reputable source. The second option of this rule could be used by a few editors to discredit this publication, just because of their personal bias. Of course, you say in your rule that they must be experts, etc. but who judge that? This is why you were told that we cannot do that because of the no original research policy: what is original research is the critic of the content, even if it is just a short notice. If these experts believe that the content is not true, they only have to publish a paper about it, say in the same journal. Then, they will be able to cite this publication in the article. This is perfectly in accord with WP policy, and even strongly encouraged. --Lumière 03:57, 6 February 2006 (UTC)

In other words, we should apply your excellent suppression principle to the critic of the article as well. Who judge of the truth of this critic? IMO, the safest approach is to remove the second option and let the expert editors use the first option until they have published their critic, and then they can put both the content that they believe is dubious and their published critic as well. --Lumière 04:30, 6 February 2006 (UTC)

Yes your formulation looks promising. I just have some wording changes to suggest.

"if there is a consensus amongst the editors that something is not true as far as they have ascertained to the best of their endeavours, then:

• they should not include it, even if it complies with all other requirements of WP policy

OR

• if it complies with all other requirements of WP, they can include it, but it should be indicated in the article that it is not thought to be true (but not necesarily thought that it is not true either. ie presented in a reported manner. (It is said by many/some that so&so is a raving...)

--Light current 04:52, 6 February 2006 (UTC)

I should have said that I have a critic of my version of your proposal. I did not say I had written a new rule. I only wrote a critic of the rule I had written before. The outcome of that critic is the new rule:

"If there is a consensus amongst the editors that something is not true as far as they have ascertained to the best of their endeavours, then they should not include it, even if it complies with all other requirements of WP policy."

No two options anymore. The second option was a problem (see explanation above).

Now working on definition of Truth...this may take some time!--Light current 03:14, 6 February 2006 (UTC)

--Light current 03:06, 6 February 2006 (UTC)

Absolute truth cannot be known. Truth is a difficult concept (said by Alan Clarke (possibly/probably)).

The definition of truth that must be used is:

:That which is generally accepted to be the correct representation of reality as far as can be known by a collection of editors knowledgeable in the subject in question whether by means of their own knowledge and/or by their researching and agreeing many and varied references which appear to them reliable

This is my first draft of the defn WP should consider using--Light current 03:32, 6 February 2006 (UTC)

It looks nice, but never such a definition can be as useful in practice as a formulation of a rule. For example, I am reading it and I can think of different interpretations in terms of a rule. I am not saying that it is useless. It is less heavy than a rule and will better fit in an introduction. Actually, it looks good. --Lumière 03:41, 6 February 2006 (UTC)

I agree its not perfect at all. But it is a start. No doubt others will find many faults with it. Thanks for your interest in my concerns.

Well, more precisely how wikipedian editors can verify it in practice? If you only give me a definition, this might not answer the question and nobody will know how to modify the policy (the rules) to please you. --Lumière 03:12, 6 February 2006 (UTC)

Verifying that something is true or correct cannot be done with certainty, but it is better to estimate a probability of correctness than to ignore it completely.--Light current 04:33, 6 February 2006 (UTC)

BTW, the addition of the rule

"If a consensus amongst the editors consider that something is not accurate or not fair, then they should not include it, even if it repects all other requirements of WP policy."

requires that we use subjective concepts such as "accurate" and "fair". Nevertheless, it is a rule that can be applied in practice. This shows that those people that argued with you that these concepts could not be part of a policy were simply wrong. --Lumière 03:01, 6 February 2006 (UTC)

If this is a private discussion, please have it in you talk pages, not here. Thanks.≈ jossi ≈ t • @ 05:42, 6 February 2006 (UTC)

No it is not a private discussion. You should actually try to understand it. --Lumière 07:19, 6 February 2006 (UTC)

This is either a private discussion, or a monolog. The only thing to learn from this discussion is the ability that some people have of ignoring the understanding developed by the community over years, and the presumptuousness that they know better. ≈ jossi ≈ t • @ 16:16, 6 February 2006 (UTC)

It is, er.., interesting watching two different incarnations of light bat around the concept of truth with each other. -- Dalbury^(Talk) 18:22, 6 February 2006 (UTC)

The truth is hard to accept sometimes. You cannot really give up. Isn't it. --Lumière 18:32, 6 February 2006 (UTC)

Spoken like a true believer. -- Dalbury^(Talk) 19:21, 6 February 2006 (UTC)

Where is the organization behind the belief? Where are the non negotiable principles? Typically, people with a belief feel challenged when alternatives are presented. --Lumière 20:27, 6 February 2006 (UTC)

Please do not disrupt Wikipedia to make a point. See WP:POINT. ≈ jossi ≈ t • @ 21:15, 6 February 2006 (UTC)

Disrupt Wikipedia? Perhaps you meant to say "Do not disrupt our conviction by expressing your opinion." because I did not even do a single edit in relation to this discussion! I concede that I will not succeed to change your conviction. What is the other issue here? What is your problem? --Lumière 21:33, 6 February 2006 (UTC)

See Wikipedia talk:Verifiability/truth to continue the discussion. --Lumière

Self-published sources as sources on themselves

Can a statement of the form "Mr X has expressed the opinion that President Bush is ..." be accepted for inclusion in an article about Bush and sourced in the website of Mr. X on the basis that telling the opinion of Mr. X is an information about Mr. X? In my opinion, it is safer to clarify this issue. --Lumière 01:58, 7 February 2006 (UTC)

Only if Mr. X is notable, and Mr. X website is a reputable source. ≈ jossi ≈ t • @ 02:09, 7 February 2006 (UTC)

Is there a general consensus on this? Certainly, the concepts of "notable author" and "reputable website" are not mentioned in the section. The concept of website with a notable owner is mentioned in the preceding section, but in this case it is clear that the information must also be sourced in a separate reputable source. So, the simple answer is NO. Right? --Lumière 02:20, 7 February 2006 (UTC)

I, for one, don't know what you mean. Jossi gave you an answer. SlimVirgin ^(talk) 04:17, 7 February 2006 (UTC)

Oh please, if you don't have the answer, just say nothing. Also, I have the right to check with others if I have a serious doubt about Jossi's answer. --Lumière 04:55, 7 February 2006 (UTC)

You shouldn't have serious doubt about Jossi's answer, (S)he is correct.--Sean Black ^(talk) 05:04, 7 February 2006 (UTC)

I also find your question confusing. If you mean, can you give a web site as a source for a given statement, the answer is yes, and it doesn't matter who made the website, so long as the website is a reputable/verified source. ··gracefool |☺ 04:32, 7 February 2006 (UTC)

If you are right, then what exactly is the significance of the sections "Self-published sources as primary sources" and "Self-published sources"? Especially, of the section "Self-published sources"? Give me some sentences that are excluded by these sections (but would be acceptable for inclusion otherwise) and explain why it does not apply to the above sentence. --Lumière 04:55, 7 February 2006 (UTC)

Oups I see that you did not read carefully the question. The question did not say any website, but the website of Mr. X. I think it was clearly mentioned. What you say raises an important issue: What is a reputable website in general? This is not the question here. --Lumière 05:04, 7 February 2006 (UTC)

In case it is needed, I repeat the simple question: Is a sentence like "Mr. X has expressed the opinion that Bush is ..." acceptable for inclusion in an article about Bush, if the sentence is only sourced in the website of Mr.X? --Lumière 04:55, 7 February 2006 (UTC)

Whether a website is "reputable" is up to community consensus. What this project page says is, don't use a source of dubious/controversial reputation as a third-party source. So if Mr. X's website is a dubious/controversial source, it should only be used in articles about Mr. X. ··gracefool |☺ 05:07, 7 February 2006 (UTC)

I think if someone stands on record as saying something, even if self published on his own website., then this should be admissible in a WP article. After all, this would be almost from the horses mouth itself!--Light current 05:10, 7 February 2006 (UTC)

Ok, it seems that we are not reading the same sections "Self-published sources as primary sources" and "Self-published sources". Sorry, I cannot connect anything that has been said above to these sections. However, I will not go as far as saying that you are all wrong. One explanation is that the sections have been rewritten, and you didn't notice. Read them again. Another explanation is that some have expressed what they think should be the rule, but not what is written in the sections. Perhaps also it will be useful to have more inputs here. --Lumière 05:26, 7 February 2006 (UTC)

Lumiere, please read the policy. Self-published material is not acceptable as a third-party (secondary) source, and that includes personal websites, except where "a well-known, professional researcher in a relevant field, or a well-known professional journalist, has produced self-published material. In some cases, these may be acceptable as sources, so long as their work has been previously published by credible, third-party publications." SlimVirgin ^(talk) 05:21, 7 February 2006 (UTC)

So, I conclude that the answer is NO! Note that the question say "only sourced in Mr. X website". Therefore, it was not previously published by any third-party publication. --Lumière 05:26, 7 February 2006 (UTC)

It would depend who Mr X was. Read the policy, please. SlimVirgin ^(talk) 05:28, 7 February 2006 (UTC)

What do you mean? My interpretation is that Mr X cannot be the owner of a third-party publication, and this is by definition of third-party. He is the one that has the opinion. A third-party is an independent publisher. I would say that even if Mr. X was by profession a reputable publisher, he would not count as a third-party publisher. If someone has another interpretation, I am curious to see what it is. --Lumière 05:39, 7 February 2006 (UTC)

I don't know whether there's a language problem here or some other, but you've misunderstood the polic, Luminere. I'm sorry I don't have any more time to explain it, but I'd say it's fairly clear to most people. SlimVirgin ^(talk) 05:42, 7 February 2006 (UTC)

You maybe right about the problem of language. Can someone give me what is the definition of "third-party publication" in the above context? I thought that it meant that the publisher is not connected in anyway with the content that is being sourced. --Lumière 05:52, 7 February 2006 (UTC)

Not self-published. SlimVirgin ^(talk) 05:55, 7 February 2006 (UTC)

Well! Who is the "self" in self-published here? It must be Mr. X. So, we cannot use the website of Mr. X as a source. Right? --Lumière 06:17, 7 February 2006 (UTC)

Yes well this shows the ultimate nonsense of the policy of V over T if you cant even quote someone who has said something and also recorded it (wherever). It is certainly verifyable that he has said it if its on his website with his electrinic sig. So whats the problem?? Notice I am not saying that the person should be allowed to self cite on WP, only that other editors should be allowed to cite him a s source.--Light current 05:31, 7 February 2006 (UTC)

Er. No it doesn't. I've read the above section, and it's very confused, but Wikipedia custom, and general common sense agree about this: If someone says something on a website, and we want to say that they said it, we can use the website as evidence that they said what they said. This is not in any dispute, although it seems to be the subject of a considerable amount of confusion. JesseW, the juggling janitor 05:44, 7 February 2006 (UTC)

Not if it's a personal website. SlimVirgin ^(talk) 05:51, 7 February 2006 (UTC)

Well, I suppose it would be prudent to make sure it was archived in the Wayback Machine first, but AFAIK, if we are writing an article on someone with a personal website, and something they say on it is relevant to the article, we certainly can quote them, and provide the URL to where they said it on the website as a source. If this is not the case, I'd love to see some examples, etc. JesseW, the juggling janitor 06:02, 7 February 2006 (UTC)

That's what the policy says. Self-published sources may be used as sources about themselves i.e. as primary sources. SlimVirgin ^(talk) 06:14, 7 February 2006 (UTC)

I agree. So, I'll assume your comment "Not if it's a personal website." was just a mis-reading, only referring to using a personal website as a secondary source, which we both agree is not OK. (Oh, and Sean, I don't mind that you removed my (attempted) humour.) JesseW, the juggling janitor 06:35, 7 February 2006 (UTC)

I am not sure where was the mis-reading. In any case, I read the sections again, and when I look at the context, it seems very much that "third party publication" means that we have an independent publisher, typically a large independent organization. The policy says that self-published sources are fine if the content is about the owner of the source. The example makes it clear: Mr S website can be used as a source in an article about Mr. S ( but not in an article about Mr. Bush.) This part of the section could not have been more clear! --Lumière 06:49, 7 February 2006 (UTC)

Whether or not we should change the sections is another issue. Right now, it is more important, IMO, that we agree about what these sections say. Then, it will be easier to discuss possible modifications. --Lumière 05:39, 7 February 2006 (UTC)

Ok, when I read the above, I am not sure that we have a consensus! I quit! Just for a while. --Lumière 06:28, 7 February 2006 (UTC)

Subpage about the value of truth (as a complement).

DO NOT REMOVE. It is not because we have a subpage about the role of truth as a complement to verifiability that we cannot say a word about it here. Just a simple question. Was there a consensus to highlight verifiability, not truth in the first sentence? See Wikipedia talk:Verifiability/truth for more details about this issue. Also, this subpage presents an interesting example of a rule in NPOV that is advocating something related to "truth" as a necessary complement, if we consider that "significance" is related to "truth". They are certainly both very subjective. --Lumière 06:58, 7 February 2006 (UTC)

Secreting

Why is SV trying to hide legit posts away on sub pages that no one will see and erasing other legit posts about the format of the discussion. I wonder if this will end up on a sub page or will it just be archived, or will it be deleted like my other posts on this page? --Light current 07:31, 7 February 2006 (UTC)

Quote from WP guidelines:

Talk page vandalism:

Deleting the comments of other users from article Talk pages, or deleting entire sections thereof, is generally considered vandalism. Removing personal attacks is often considered legitimate, and it is considered acceptable to archive an overly long Talk page to a separate file and then remove the text from the main Talk page. The above does not apply to the user's own Talk page, where users generally are permitted to remove outside comments at their discretion.

I wonder if those who, in future, remove/delete posts from this talk page could defend their actions adequately in light of the above guidelines?--Light current 23:31, 7 February 2006 (UTC)

Why negation of truth?

Slim: Why is not sufficient to say just verifyibility?. --Light current 08:05, 7 February 2006 (UTC)

Unfortunately, I think I know the answer. It is very simple. It is important for some that the policy does not use truth. They think it is great because in this way the policy relies less on the judgment of the editors. So, they want to inform the readers of this "objective" approach that they consider very fundamental. Of course, the reality is that they are confused. They think that WP policy should not rely on the judgment of the editors, but there exists plenty of WP rules that do rely on the judgement of the editors. For example, in accordance with NPOV, "non significant" views should not be included, and it is explained in the NPOV talk page that the precise meaning of "non significant" depends on the situation and must be determined by the editors. --Lumière 22:39, 7 February 2006 (UTC)

I want to get rid of the impression given in the policy that truth in articles is not at all necessary. Thats why I say: if we cant say verifiable AND true, we should just say verifiable (as I think he policy used to say at one time). At some point in the editing process, the judgement of editors is, of course, going to be required. If not, then a computer could do the whole job of writing the 'pedia!!(eventually)--Light current 23:20, 7 February 2006 (UTC)

I would agree, if it is what you say, that even if the current policy does not explicitly use truth as a criteria, it nevertheless partially remove innaccurate contents, and thus we should not present the opposite impression. I agree that mentioning "not truth" does create the wrong impression and also agree that it is not absolutely necessary to insist on it. It still remains that the sentence is not false: the policy does not use truth. Do you think that it should use truth or is it sufficient that we only remove "not truth" in the formulation? --Lumière 01:17, 8 February 2006 (UTC)

Bon! Ideally I would like to see the word Truth included as a necessariy criterion. However, I would be fairly satisfied if just the not truth was removed from the article opening sentence. It didnt used to have it in, Im sure.--Light current 02:17, 8 February 2006 (UTC)

The purpose of my answer to your question to Slim was to present and discuss the perspective of those who really want to keep "not truth". I tought it could be useful. --Lumière 01:24, 8 February 2006 (UTC)

PolicY

How can this page be WP policy if it changes every 10 mins?--Light current 08:15, 7 February 2006 (UTC)

Light current, you're trolling. If it continues, you may be reported and blocked for disruption. SlimVirgin ^(talk) 08:14, 7 February 2006 (UTC)

Defn of trolling:

In Internet terminology, a troll is a person who posts rude or offensive messages on the Internet, such as on online discussion forums, to disrupt discussion or to upset its participants. "Troll" can also mean the message itself or be a verb meaning to post such messages. "Trolling" is also commonly used to describe the activity

I dont belive I have posts rude or offensive messages. Anyway, the terms rude and offensive are subjective! Im not trolling. Im trying to improve policy (or at least put it back to what it was). Why is it that only 2 people (you and jossi) can edit this page without getting reverted by admins? BTW does 3RR apply to you too or are you immune? --Light current 08:21, 7 February 2006 (UTC)

The current policy does not use truth at all as a threshold, even though it should.

Whether we like it or not, the current policy does not use truth to filter for inclusion. The purpose of the first sentence is to explain the policy as it is. Therefore, I support that we keep the first sentence with the "not truth", even if it stupid, because this is the situation.

However, I also support that we change this policy to add a very simple rule that uses "truth". This proposed rule is very similar to another rule that uses "significance" instead of "truth". This other rule, which is implicit in the current NPOV policy, says that if the editors agree that something is not significant, they should not include it in the article. If you ask the experts behind NPOV what is the precise meaning of "significant", they simply say that it can be determined amongst the editors. It would make perfect sense that we do the samething: If all the editors agree that something is not true (or say innacurate), they should not include it. What is wrong with this? --Lumière 08:25, 7 February 2006 (UTC)

At first sight this looks OK to me. I would rather truth not be mentioned at all rather than in a negative context which is why I tried to remove it from the page. It was reverted of course by SV who now seems to be soley in charge of this page.--Light current 08:45, 7 February 2006 (UTC)

This is trolling

I would urge that the editors who have been asked, multiple times, to cease their excessively volumnious comments on the talk page not be responded to further. There's no need to archive the talk page, but again, there's no need to respond either. The reason that some edits are reverted and others are not, is that some edits have consensus, and others do not. Again, I urge that no one further respond to this. JesseW, the juggling janitor 08:29, 7 February 2006 (UTC)

Copied from Talk User:Jossi

Hi Jossi. I've been watching the WP:V page also, and it's my impression that primarily one user there is being disruptive, Light current. Because of that I'm hesitant to protect the entire page over it. However I did just leave him a message asking him to stop what he's been doing [8] so I'm hopeful he'll listen. · Katefan0(scribble)/poll 16:57, 7 February 2006 (UTC)

Please define 'disruptive'--Light current 22:09, 7 February 2006 (UTC)

What you've been doing: trolling the talk page, and making edits against consensus. I gave you two discrete examples in the message I left for you on your talk page. · Katefan0(scribble)/poll 22:35, 7 February 2006 (UTC)

Disruption is not a WP defined word in the sense in which you use itand therfore cannot be used in disciplinary matters because we have no definition of it!!!--Light current 22:37, 7 February 2006 (UTC)

Retrieved from "http://en.wikipedia.org/wiki/User_talk:Jossi"

This page was last modified 23:37, 7 February 2006. All text is available under the terms of the GNU Free Documentation License (see Copyrights for details).

Wikipedia® is a registered trademark of the Wikimedia Foundation, Inc.

"Disruption" is a word defined in an enforcable wikipedia guideline: Wikipedia:Don't disrupt Wikipedia to illustrate a point. Apart from the explanation of what is meant by disruption in this context, the guideline also contains:

Egregious disruption of any kind is blockable by any administrator — for up to one month in the case of repeat offenses that are highly disruptive.

--Francis Schonken 23:20, 7 February 2006 (UTC)

THank you for that link and your explanatory additions. I shall study them avidly--Light current 23:21, 7 February 2006 (UTC)

Reply to Lumiere

Well any changes we try to make are immediately reverted by the admins and if you try to change them back you get accused of 3RR or trolling. I have tried to change the page for the better a number of times and Im in danger of being blocked again for 2 helpful changes I made (This is called trolling if you put something on the page that an admin doesnt like). So I cant really make any other changes to the page myself today, but others could.

I dont see what can be done except to accept that WP policy is not made by the users but by one or two admins.--Light current 08:30, 7 February 2006 (UTC)

Well, they are just very attached to the policy, like if it was the bible! I feel as if we should not touch it until after we become members of the order! They won't like it otherwise. They barely tolerate us in the talk page. --Lumière 08:50, 7 February 2006 (UTC)

Absolutely!. Its called paranoia!--Light current 08:54, 7 February 2006 (UTC)

Im not sure if they are attached to the policy, because some admins keep changing it on an hourly basis. One thing is certain: they dont want any one else interfering in the policy or even suggesting its partial stupidity - Sacre Bleu! (pardon my French!)--Light current 22:34, 7 February 2006 (UTC)

Tabloids

I ve removed the reference to tabloids because its out of date and is not NPOV. ie it presents the view that tabloids are unreliable- most are reputable these days. Also mentioning a specific newspaper The Sun and casting doubt on its reliability is again POV but Ive left that for the moment.--Light current 14:23, 7 February 2006 (UTC)

I agree. I would not use a newspaper to source a scientific content, but it might be fine for some other contents. The main issue is the last part. A criticism of The Sun should be notable and well sourced. Otherwise, if it is not notable it is POV and if it not well sourced it qualifies as original research, a research on quality of newspapers. Moreover, and perhaps more importantly, a criticism of the sources sounds like totally off topic and inadequate in an encyclopedia article. What about the sources that are used to support this criticism. If these sources are also tabloids, should we also critic them, and so on. --Lumière 22:05, 7 February 2006 (UTC)

So after the recent revert of my edit regarding Tabloid Press, we can safely say that this page continues to breach WP policy on NPOV- (maybe others too- I havent studied it in detail yet) with the full blessing of certain admins/editors--Light current 22:18, 7 February 2006 (UTC)

Moved here from my talk so all can see what admins are up to:

Your actions on Wikipedia:Verifiability

Please stop what you're doing to this policy page. Use the talk page and discuss the issue if you like, but stop making what are essentially vandalizing edits to modify longstanding consensus on this important policy page. You're welcome to participate in a less disruptive manner, particularly on the talk page, but if you don't stop these sorts of edits to the article itself [1] [2], I won't hesitate to block you temporarily for vandalism and disruption. · Katefan0^(scribble)/poll 16:53, 7 February 2006 (UTC)

You have been asked not to feed th Trolls- but you ignore this request and continue to stir up and worsen bad feelings. You assume bad faith on my part. Thats a mistake on your part. These were perfectly valid edits. They are certainly not Vandalism.

It seems any admin can make whatever changes (s)he likes without any consensus or discussion, but a simple user is just blocked or threatened with blocking. Do you like wielding the big stick to batter everyone who disagrees with you into submission? This is how it starts: supression of truth and opinion. It's happened so many times in history.

Even my and other peoples posts on the talk Wikipedia:Verifiability page are being deleted or moved into backwaters. THis could be called vandalism or disruption, but I suppose its called something else when admins do it. So how to get proper discussion on these matters? I can see that both I and User:Lumiere have touched some raw nerves here, I unwittingly. I certainly did not expexct such vicious reaction to my suggestions and edits. As I said before this is a sign of paranoia and this particular group of admins editng this page are showing classic traits of dictatorship. Is that really the intention of WP? These admin actions are now really starting to look foolish.

BTW would you care to define the word 'disruptive' in the WP sense. I cant seem to find a definition here. If im being accused of something, Id like to know what it is. Could you also say wher the policy on disruption is kept and wher the punishments are stated?--Light current 22:02, 7 February 2006 (UTC)

Totally agree. --Lumière 01:52, 8 February 2006 (UTC)

Let's not overgeneralize. I agree, as an admin, that we should be able to discuss reasonable changes to the policy here, and with consensus implement them, and that no such discussion should be censored. As this is a long-standing policy, however, the right of anyone to just Be Bold is much more limited.

I do not personally agree with any change that requires our sources to be highly reputable and trusted, because often an irreputable source can still be used as evidence provided that it is properly qualified (for example, "Senator McCarthy's speeches accused over 500 people of being Communists, allegations later shown to be almost entirely false"). The most important thing is that we have sources, and that we do not misrepresent their level of reputability. Deco 03:21, 8 February 2006 (UTC)

I agree with you Deco. Im talking about the recent bad changes to this page that now emphasise the truth is completely unimportant as acriterion for inclusion in WP>--Light current 03:25, 8 February 2006 (UTC)

Deco, I also agree with you, but only because of the "any" in "any change that requires our sources to be highly reputable". I do think that we need to further discuss ways to improve the quality of Wikipedia and reputable sources are the primary ingredient to achieve this goal. Of course, we should not always require highly reputable sources. It should depend on the content and its context. --Lumière 06:45, 8 February 2006 (UTC)

Deco, can you explain the context of your last comment. It does not fit with the previous paragraph, so I realize that I do not have the context. Was your comment related to a previous discussion in this talk page? --Lumière 07:00, 8 February 2006 (UTC)

Wheel Warring

Of course! Ordinary users must not interfere in wheel wars or they are lible to get blocked. I have just learned this valuable lesson!--Light current 00:10, 10 February 2006 (UTC)

Regarding troll food

I encourage others not supply anymore of it. Watch the page, revert ill-formed, misguided or bad faith edits, and leave it at that. FeloniousMonk 00:12, 8 February 2006 (UTC)

I liked your last album BTW!--Light current 00:15, 8 February 2006 (UTC)

"Local interest" articles

Recently, an item was added to the Cheesesteak article. It's under discussion for various reasons, but I want to present it here regarding questions of verifiability in local-interest articles.

Characteristic	Pat's Steaks	Geno's Steaks	Steve's Steaks
Steak	cut up into small, thin pieces; unseasoned	long, uncut, thick pieces; seasoned	long, thin, fatless pieces; unseasoned
Onions	mixed in with steak; sweet taste	underneath meat; spicy taste	above meat and cheese
Cheese Whiz	Kraft brand; on top of steak	generic brand; under steak	generic brand; above steak
Bread	hard and firm	soft, with unique taste	soft yet firm

I had nothing to do with adding this item. On the one hand, I think it adds something to the article (other editors don't necessarily agree). And I think it's quite likely to be accurate. On the other hand, it's unsourced; it's obviously the personal observation of Wikipedia editors. There's no doubt at all in my mind that under the verifiability policy it should go; unsourced, unpublished, original research.

The question I have is this. A lot of "local-interest" articles are largely comprised of this kind of material. They are basically personal contributions by Wikipedia editors. You see this in descriptions of eateries. An interesting example is Nick Tahou Hots, IMHO a good, readable, informative article with many external links. But. a) individual items of fact are not traceable to sources, and in fact the sources don't necessarily confirm the items in the article; b) the sources are all in the nature of blogs, forums, personal websites, and in one case another Wiki.

The same is often true of non-Rambot material in articles about individual towns. I'm personally guilty of contributing some of it myself.

This sort of material is fun, it's interesting, it's even informative, but there's probably no way most of it can be properly sourced. (A lot of it can be, but I think it's unlikely that information on the brand of cheese used by different cheesesteak restaurants, and whether present it on above or below the meat can be found even in Philadelphia newspapers. I'd be delighted to be proved wrong. I'd accept e.g. a campus newspaper as a source, on the assumption that they do have editors who do vet articles...)

Would it be appropriate to have some sort of category of "local-interest" articles for which the policy would establish a different, and lower standard of verifiability?

This is not at all a rhetorical question. I'm personally quite puzzled as to how this should be handled. Apply the current policy, and gradually tighten the noose on hometowncruft? Create variable standards for different kinds of articles (a genuinely slippery slope IMHO)? Or what? Dpbsmith (talk) 15:27, 10 February 2006 (UTC)

Yes, this sort of material fails WP:V. I believe wikitravel welcomes content based on personal experiences of contributors, tho. We already in practice have different standards for different articles, but I don't think this should be turned into written policy. I personally think that if the only source for a particular bit of info is a school newspaper, that info is too trivial to be included in an encyclopedia. Friday (talk) 15:34, 10 February 2006 (UTC)

That content does not belong in WP as it fails WP:V. There are other wiki projects such as Wikicities in which this type of content belongs. ≈ jossi ≈ t • @ 18:37, 10 February 2006 (UTC)

We should be trying to bring all articles up to standard on references. the fact that we're not there yeat doesn't mean we should soften the standards. If you can't find a reliable published source, it doesn't belong in Wikipedia. -- 68.223.206.135 19:53, 10 February 2006 (UTC) - Argh! Resign while logged in. -- Dalbury^(Talk) 19:54, 10 February 2006 (UTC)

For what it's worth, some people (including many old-time editors, from what I've seen) don't particularly believe in verifiability. Those folks may argue that having this material in the encyclopedia is just fine. Friday (talk) 19:57, 10 February 2006 (UTC)

I know. They just don't understand yet that leaving articles unsourced seriously weakens Wikipedia. After all, Wikipedia is not some giant blog, it's supposed to be an encyclopedia, gathering together what is known about the world (and universe). The best way to insure that we are putting solid information into Wikipedia is to require that it has been previously published in reliable sources. -- Dalbury^(Talk) 20:28, 10 February 2006 (UTC)

All right, but there are reliable published sources on local subjects which do not exist in English and are very probably not available in the United States, e.g. Roth and Straubhaar, z'Bärg - Wege zum Alpkäse, Amt Frutigen, 2004 (any number of facts and figures about Bernese Alp Cheese and the western Bernese alps where it is produced, also excellent resource for articles on the communities there) ISBN 3-909532-11-X. Not available at Amazon.de - but it's sold in most local bookshops in the Bernese Oberland. Or an older one Zürcher Bürger- und Heimatbuch, 1968, first issue 1938, published by the government of Zurich, some 300 pages about geography, history, economy, military and law as present for citizens attaining their majority, authors are mainly professors , definitely a very reputable source - again, lots of useful facts and figures, statistics, developments all with a nice conservative pre-1968 POV which is, by itself, a historical feature. No way to find it at Amazon.de, and there is no chance whatsoever to find a source with the same data in English. I can take a text from those sources, write it in German and English so anyone can verify the translation, no problem (and give the source, of course). But I am not willing to skip the use of books in my shelf due to their not being readily available to average English-speaking verifiability-hawks. ::All right, but there are reliable published sources on local subjects which do not exist in English and are very probably not available in the United States, e.g. Roth and Straubhaar, 'z'Bärg - Wege zum Alpkäse, Amt Frutigen, 2004 (any number of facts and figures about Bernese Alp Cheese and the western Bernese alps where it is produced, also excellent resource for articles on the communities there) ISBN 3-909532-11-X. Not available at Amazon.de - but it's sold in most local bookshops in the Bernese Oberland. Or an older one Zürcher Bürger- und Heimatbuch, 1968, first issue 1938, published by the government of Zurich, some 300 pages about geography, history, economy, military and law as present for citizens attaining their majority, authors are mainly professors , definitely a very reputable source - again, lots of useful facts and figures, statistics, developments all with a nice conservative pre-1968 POV which is, by itself, a historical feature. No way to find it at Amazon.de, and there is no chance whatsoever to find a source with the same data in English. I can take a text from those sources, write it in German and English so anyone can verify the translation, no problem (and give the source, of course). But I am not willing to skip the use of books in my shelf due to their not being readily available to average English-speaking verifiability-hawks. Irmgard 00:06, 11 February 2006 (UTC)

Again, sorry, but no. This is, after all, the English Wikipedia. The only language that we can be sure a reader or editor of the English Wikipedia understands is English. Sources in other languages cannot be checked by most editors and readers, and are therefore unverifiable to those readers. We'll have to just get along with what is available from reliable sources in English. -- Dalbury^(Talk) 00:22, 11 February 2006 (UTC)

• There is no way we can make an unbiased encylopedia if we restrict ourselves to English language sources, to say nothing of what the folks making minority language wikipedias would end up with. WP:V say "English-language sources should be given whenever possible, and should always be used in preference to foreign-language sources, so that readers can easily verify that the source material has been used correctly." - it certainly does not say that foreign language sources are unacceptable. Kappa 00:26, 11 February 2006 (UTC)

I would be very careful about relying on foreign language sources for articles in the English Wikipedia. I personally would want to challenge any article that had only foreign language sources. If there are no sources in English for a topic, then maybe that topic can wait for inclusion. -- Dalbury^(Talk) 00:51, 11 February 2006 (UTC)

No, this interpretation of verifiability is too narrow. If there is a well researched and sourced article on, say, the Spanish Wikipedia, and it is translated into English relying on the original sources, I think it would be utterly unreasonable for someone to come along and say "I don't speak Spanish, this is unverifiable". Verifiable does not mean that anyone can verify it - it means that someone with reasonable and appropriate expertise can verify it. Enchanter 01:16, 11 February 2006 (UTC)

I agree. Some sources are simply not available in English, e.g. most of the opposition to Opus Dei resources are in Spanish or Portuguese, but are sure important for NPOV.

Also, reliable, published and comprehensive English sources on foreign subjects are rather limited: which is, as I've noted in quite some articles, a factor which limits comprehensiveness and also sometimes quality in the English Wikipedia. Compare, e.g. the French to the English article on Édouard Manet (where the English one is not well referenced, as well) or the German and English article on Gelsenkirchen (where the English one probably has also the German article as best resource). It's also seen in the Gelsenkirche talk pages, where, BTW, in the German talk page verifyability plays a much bigger role - sure, the same facts would not be easily verifiable using English sources.

And then, there are differences between verifiability and verifiability. A statement on quantum mechanics could give as source an academic article which is, though in English, in a language Joe Doe does not understand enough to get to the point which has been referenced. Such stuff can only be verified by someone who is at least a hobby-scientiest. Then Wikipedia says that outlandish statements need strong sources, surely correct - but there is also the corollary that not at all outlandish statements do not need such strong sources, and an example for that would be, e.g., undisputed facts regarding local history, geography or biography. You don't have an English source about the description of the blazon of Gelsenkirchen? So what? No English quote regarding the influence of Gustave Courbet on Edouard Manet? (French quote and source given in the French Wikipedia) - do you seriously think omitting the statement will improve the English Wikipedia? Irmgard 13:11, 11 February 2006 (UTC)

Restricting ourselves only to English-language sources would be a very bad thing. Ausir 14:30, 11 February 2006 (UTC)

IMO, it makes no sense to accept foreign language sources. We are not equiped to check the validity of the interpretation of a wikipedia editor of a source in a foreign language. It is already a challenge to check that a Wikipedia article is faithful to its sources when they are in english. It is almost impossible to check that if we accept sources in all kind of foreign languages. Of course, it is very bad that we cannot use all this information that is not yet available in english, but what can we do? We have to accept it. We are simply not equiped for translation. -Lumière 17:18, 11 February 2006 (UTC)

There are two opposite tendencies in the Wikipedia community. There are those who advocate that we should not suppress information even if it is hard to check, dubious, etc. There are those who are for high-quality standards that can be implemented within our open structure, and thus are very selective about the kind of content that is acceptable for inclusion. The above is just an example of a debate between these two groups. I am on the side of High-quality standards. This is why I say that we should not accept foreign language sources until we adopt some structure to check their translations. -Lumière 17:32, 11 February 2006 (UTC)

General remark: Hard to check and dubious are two very different things. An academic thesis on specifics of quarks might be difficult to get at for the average user and it's doubtful if he would understand it if he got it - so that's hard to check, but at the same time the article is peer reviewed and published in a highly valued scientific journal. Even with academic level books as sources, there is not everyone in position to verify them. Take a theology article using as sources Jaroslav Pelikan's "The Christian Tradition" (5 volumes), Thomas C. Oden's Systematic Theology (3 volumes) and N.T. Wright's "The New Testament and the People of God" (3 volumes so far). All in English, and no one will question them as reputable sources, but could you verify the quotes? How many Wikipedians do have these books on their shelf? --- On the other hand, a convincingly written article on "the only exact interpretation of the Revelation" published on a website of an anonymous group "Followers of the Christian prophet", can be found by anyone who knows how to use Google, so it's easy to verify, but its value as reputable reference is close to zero. So by all means be careful on dubious sources, I am as well - but to not accept a source because it is "hard to check" in someones view is one of the surest ways to limit the quality of Wikipedia to average high school level."

Foreign language sources classify surely as "hard to check" for some people. But Wikipedia is a community effort, there's no one who has to (or could) singlehandedly verify all references - and that community includes lots of people who do know foreign languages: have a look at the Category:user languages: in the English Wikipedia there are thousands of users who understand some Spanisch, German, French, Italian, Latin, Japanese, Russian (minimum 1000 users each language), and you also find some who speak Hebrew, Dutch, Norwegian, Swedish, Italian, Hindi, Classical Greek, or Croatian (minimum 100 users each language) and you also find someone for more exotic languages (and that's only those who did make use of the Babel system. Many of those users are already interested in local articles regarding the area of their language and check them for correctness - and sure most of them would readily help out verifying a source in a local area article, which usually does not need any expert knowledge if asked to. You do not need to know a language perfect to check a reference, you do not have to translate it, just check the translation (I can do this in all languages where I rate myself at basic level, not only in those languages where I can write or translate articles. So if you are not able to verify a foreign language reference, there is a good chance someone else is - the only problem would be a constructed problem of a source in a language which only three users know and all of them are having the same pronounced bias, but that's sure not the model case for the general rule. I plead for inclusion of foreign language references to enhance the quality of Wikipedia - they can be checked by this community. Ok, by all means qualify "foreign language sources" as "sources in languages which are known to at least 100 wikipedians per Category:user languages", this will probably cover 99% of the sources in question. --Irmgard 16:30, 12 February 2006 (UTC)

Irmgard, I acknowledge that "dubious" and "hard to check" are two different things. Your clarification on this issue is very important. It is not understood enough in the Wikipedia community, and this is due to a lack of appreciation of the important role that expert editors must play within Wikipedia. Actually, I have discussed that before with others (see Wikipedia talk:No_original_research#More_appreciative_of_the_role_of_experts). The problem is that Wikipedia has no structure to distinguish an expert editor from a non expert editor. So, unless we adopt a precise structure to identify some kind of expertise, we cannot officially use this kind of expertise. This includes translation. That is all what I am saying. -Lumière 17:01, 12 February 2006 (UTC)

Lumière, instead of experts Wikipedia uses crosschecks by other Wikipedians who know more or less about the subject, without verifying their exact status. If this works for medicine, physics, and astronomy articles (Ok, we agree, in some cases a verification can only be done by an expert or at least someone who has access to a university library - but that's somewhat self-monitoring: could you, e.g. verify the Fermi paradox article? I couldn't.) , why should it not work for foreign sources?T here are Wikipedians who know enough about mathematics or programming languages to check the articles and verify sources, and there are Wikipedians who know Portuguese or even Armenian to check articles and verify sources. There are a lot more people who can verify a translation than who can translate (I can translate into English, German and on some subjects into French, and sure I can verify such translations - but I also can verify a translation into English from Italian, Spanisch, Latin or Old Greek and have often done so (in and out of Wikipedia) where I doubt if the translation is a translation or an rather an interpretation of the original.

And, again, leaving out everything which is not published in English will make a poor encyclopedia - the US is top notch on science, so there you can reasonably argue " if it is not published in English, it's not necessary". But that's not as true e.g. for general history or history of arts, and for sure not true for any non-English literature, non-English places, non-English customs, laws, etc.

And in some cases, foreign sources are indispensible: Imagine an article on a French law without having as sources that French law aand how the French reason about it (no joke, I came upon such an article where almost every statement an English quote - but all the sources quoted either misunderstood the issue or were biased or both, and the article was anything but NPOV until a Wikipedia sysop of French nationality took over and provided the French view (with French sources, but I verified them). Irmgard 18:46, 12 February 2006 (UTC)

Yes, we can do the same for translation as we do for other kind of expertises. You are right. My sentences were "It is already a challenge to check that a Wikipedia article is faithful to its sources when they are in english. It is almost impossible to check that if we accept sources in all kind of foreign languages." In other words, I was just trying to remove an extra level of difficulty, which I felt was not small at all. So, we agree on the principles. We just disagree about where exactly we should draw the line.

To conclude with an agreement, I realize that it is important to distinguish between "hard to check" and "dubious". If we take this angle, we are both working together because I think that too often some editors reject contributions of other editors because it requires "too much" of an understanding of the sources to see that these contributions are actually well supported by these sources. The problem is that sometimes the "too much" is not that much. It just that it is an easy way that opponents can use to filter some contributions for inclusion. The case of a language barrier is one example, but not the most frequent examples in my opinion. -Lumière 19:17, 12 February 2006 (UTC)

I fully agree with you on that - rejecting contributions as unsourced because the rejector does not understand the source (or has not really read it) is not what is intended by the verifiability policy. I even balk a bit a the sentence "so editors should cite reliable sources so that their edits may be verified by readers and other editors" and would like to qualify it by "readers and other editors who know the subject", but that's again getting close to the "expert/no expert" subject. --Irmgard 20:27, 12 February 2006 (UTC)

Are we splitting hairs here? The policy is very clear on this subject. The burden is on the editor adding material. Provice verifiable material from reputable sources. Period. ≈ jossi ≈ t • @ 20:37, 12 February 2006 (UTC)

I agree with you Jossi. The burden to provide sources that are reputable is on the editor adding the new material. All editors must agree that the sources are reputable: Wikipedia works with consensus. The problem that we consider is how to check that the material added is indeed supported by these sources. We can see two steps here: first we see that the sources are reputable and second we see that the material added is well supported by these sources. What is your opinion about the second step? Is the burden for this second step also on the editor adding the material? Note that in any case, at the end all editors must agree that the new material is well supported by the sources: Wikipedia works with consensus. -Lumière 21:08, 12 February 2006 (UTC)

Editors apply policy to the best of their ability. Let's not get into too much instrucntion-creep. We have done consireably well with this policy so far. Let it be. ≈ jossi ≈ t • @ 21:18, 12 February 2006 (UTC)

Sure! The first sentence is perfectly reasonable. Now, to who do you adress the next two sentences. To me? To me and Irmgard? Perhaps you adress these two sentences to every one that works on the policy and try to better explain it with tutorials, with additional clarifications, simplifications, etc. Is this a personal issue? -Lumière 21:26, 12 February 2006 (UTC)

So, I maintain that the writing of the policy should provide a better understanding of what is intended by the Verifiability policy when it says that new material must be supported by reputable sources. This is related to the role of expert editors in Wikipedia. The difficulty with foreign language sources is just one case. -Lumière 21:40, 12 February 2006 (UTC)

As a general response to the question of using foreign language sources as references in articles, I'll quote from Wikipedia:Reliable sources#Sources in languages other than English

Therefore, when the original material is in a language other than English:

* Where sources are directly quoted, published translations are generally preferred over editors performing their own translations directly.

* Where editors use their own English translation of a non-English source as a quote in an article, there should be clear citation of the foreign-language original, so that readers can check what the original source said and the accuracy of the translation.

—Preceding unsigned comment added by [[User:{{{1}}}|{{{1}}}]] ([[User talk:{{{1}}}|talk]] • [[Special:Contributions/{{{1}}}|contribs]])

This only provides rules for quotes. Is that means that we can only quote a foreign language source? In many WP article, it is impossible to faitfully represent the sources just using quotes. We have to be realistic, no original research does not mean that we can only quote the sources. In fact, it is easy to misrepresent an article by quoting some specific parts of the sources and not others. So, interpreting the no original research as if it said that we can only quote the sources is pointless. So, the issue of foreign language sources is still not clear. In general it is still not clear how to determine if a given content is supported by given sources. At the least, we should be clear that it does not mean that the content is found word for word in the sources. Also, it should be explained that it is fine that only those who can understand the sources can determine if the content is well supported by these sources. -Lumière 05:44, 13 February 2006 (UTC)

Provisional Sources

WikiPedia strives to use only published sources, but in many cases where documented sources exist, publishing them may not be feasible due to biases in the publishing industry. The most brilliant prose on a topic may not find a publisher because they are just not interested in the subject, the subject is too short, the subject isn't sexy enough, a publisher doesn't exist for that topic, it's not "important" enough, everyone's too busy to bother with it, etc., all of which may prevent good documented information from being used by WikiPedia editors as verified sources. For cases where verified, documented sources exist but are unpublished and publishable if publishers would let it be, maybe we can establish a "Provisional Source" classification and a Source Commons, so that articles that would stand if the legitimate sources did get published would not be hampered. "Provisional Sources" could be classified as "verified, documented sources that would be published, provided a publisher were available". Provisional Sources can be marked as such. Articles using Provisional Sources can state something like, "This article or section contains information that has been verified but not published. Please help by replacing unpublished sources with published sources where possible." What qualifies as a verified, documented source can be hashed out here, but I think it can include third party corroboration of an article's facts, through affidavits, email requests for confirmation, interviews, and other factfinding methods used by publishers. This would take the sources beyond Original Research by using only third party corroboration and not creating one's own sources. This would be particularly helpful for articles on social phenomena where few if any published sources are available due to publisher bias, but can be verified through third party corroboration, such as local interest, or special interest groups and fandoms outside the mainstream. (Publishers are especially biased about the latter, interested only in yellow-journalistic reporting if they're interested at all, which frustates any well-sourced articles.) Provisional Sources would satisfy the condition of verifiablity, while stressing that published sources are prefered, and quell arguments about well-known facts that have no published sources but can be corroborated. --Coyoty 20:00, 10 February 2006 (UTC)

Sorry, no. There are usually very good reasons why unpublished manuscripts are unpublished. The policy say reliable published sources. That gives us something to point to when an item in an article is questioned. If the source is not one that is generally available to other editors (i.e., in libraries, at bookstores, or on the Internet), it can't be verified by other editors and readers. Also, 'verifiable' does not mean that some person has verified the information before it's put into Wikipedia, it means that anyone else can go to the cited source to verify that the source supports what has been put into the article. -- Dalbury^(Talk) 20:23, 10 February 2006 (UTC)

Exactly right, Dalbury. Coyoty, you're using another meaning of "verify," and what you're describing we call original research. Wikipedia has no established fact-checking or research process in the way newspapers or publishing houses do, which is why we rely on material already published by reliable sources. SlimVirgin ^(talk) 21:34, 10 February 2006 (UTC)

Absolutely. Just think of the massive can of worms we will be opening if allowing these "provisional" sources. Material that is good, will be sooner or later, published. Let's not forget to look at WP as a work in progress! ≈ jossi ≈ t • @ 17:24, 11 February 2006 (UTC)

Radio interview

Some editors are using transcripts of radio interviews and labelling these "reputable published sources". I would argue for a distinction between a published source in which fact checking and editorial review processes are in place, and such interviews in which the person interviewed, can very much say whatever he/she wants. We can assert the verifiability of the interview taking place, but we could not assert the reputability of the source, just because it was in a media outlet. Any comments? ≈ jossi ≈ t • @ 17:28, 11 February 2006 (UTC)

Jossi, I don't know if you care about my opinion, but I totally agree with you that we should distinguish between something that is merely brought to the public attention (even in a way that can be checked later) and something that was published in a reputable source. As your comment suggests, the term "verifiability" was not well chosen at all. Most people interpret it as if it means that we can verify that the author is really the author and that the text (or the video, etc.) was not modified by others. Of course, they did not read the policy to have such an interpretation, but still there are plenty of editors that do have this interpretation. In fact, the policy even goes further in the opposite direction and mentions that what is a reputable source depends on the type of content. For example, for scientific content, peer-reviewed journals are recommanded. Also, the policy refers to guidelines which strongly suggest that we check the reputability of the publishers (to make sure that they only publish accurate material). -Lumière 17:51, 11 February 2006 (UTC)

I would say that if the transcript of the interview is available from a reputable source, then it can be used as a source for what the interviewee said. We do have to be careful that the transcript is trustworthy. We do not want to be putting words in somene's mouth that we cannot prove can from a reputable source. -- Dalbury^(Talk) 21:03, 11 February 2006 (UTC)

You mean "came from" not "can from", right? Lumière 01:48, 12 February 2006 (UTC)

In general, interviews should only be used as a source for the interviewees opinions, because, as User:Jossi points out there is no vetting of their contents. But I fail to see the difference between a print interview and a broadcast interview, other matters aside. If an interview is "published" by posting the contents to a reliable source, such as a radio station website, then I'd be inclined to give it credence. -Will Beback 21:11, 11 February 2006 (UTC)

Agreed.--Sean Black ^(talk) 21:20, 11 February 2006 (UTC)

I don't see why it is any good even for the interviewees opinion. Perhaps you meant "interviewees opinion about himself", because then indeed the interview can be used if it self-published by the interviewee, but only in an article about the interviewee. -Lumière 01:52, 12 February 2006 (UTC)

Imagine George W. Bush gives an interview to "60 Minutes" and says that Vladmir Putin lied about the Russian nuclear forces. Of course we wouldn't use that as a source for Putin being a liar, but we should be able to use that in articles on Putin or the nuclear forces as a source for a notable opinion on those topics, IMO, not simply in the article on Bush himself. -Will Beback 02:04, 12 February 2006 (UTC)

If an interview in a TV program can be considered as a reputable source for some kind of content, tell me what kind of content? As pointed out before, the problem is that no fact checking and editorial review processes are in place. What kind of content should not have this form of quality control? I disagree that we can use an interview as a source to include Bush's opinion about Putin in an article on Putin. I know, it seems like a strong suppression of information, but this rule is necessary to guarantee that only fair and accurate stories appear in Wikipedia. Say if another important man says something bad about the pope in an interview, can we include it in an article about the pope? I don't personally care that much about the pope, but if we want to avoid a bad impression we better be sure that we only include content that was carefully filtered for unfair or innacurate stories. -Lumière 02:29, 12 February 2006 (UTC)

At this point I think the discussion has strayed from the question regarding this policy. I'd say the point above concerns Wikipedia:reliable sources. A blanket ban on using interviews in anything other than the interviewees' biographies is indeed harsh and unnecessary, IMO, but it does not sound like the issue is whether the interivew is verifiable. -Will Beback 22:30, 12 February 2006 (UTC)

Don't worry about my opinion. Just read the policy again and ask yourself if it provides rules to determine what is a valid source. Put some attention to the section WP:V#Sources. Whether or not it is my opinion doesn't reall matter. -Lumière 22:48, 12 February 2006 (UTC)

The question is: how can you give "credence" to an interview, in which the interviewee can say whatever he/she wants. As Dalbury says, given WP:V, we can only assert that the intervieee said A, B and C. Period. ≈ jossi ≈ t • @ 00:51, 12 February 2006 (UTC)

That's what I mean by credence: that we believe that the interview correctly reflects what was said by the interviewee. Since print interviews are sometimes transcribed incorrectly, radio/TV interviews may be more credible than print interviews. Certainly whether the interviewee is correct or not is a problem no matter what format the interview appears in. -Will Beback 01:13, 12 February 2006 (UTC)

Indeed, that's an issue with any personal account, like an autobiography or essay.--Sean Black ^(talk) 01:17, 12 February 2006 (UTC)

I was just asking about making a distinction between what we consider a reputable source, and an interview. A reputable source is described in WP:RS as being reliable because Publications with teams of fact-checkers, reporters, editors, lawyers, and managers — like the New York Times or The Times of London — are likely to be reliable, and are regarded as reputable sources for the purposes of Wikipedia. An interview's content does not have that imprimatur. The interviewee can say A, B and C, but we cannot assert that A, B and C is reliable information. But if the Times of London says A, B, and C, we can assert that as reliable information. ≈ jossi ≈ t • @ 02:34, 12 February 2006 (UTC)

Yes, I agree. It is policy. -Lumière 02:51, 12 February 2006 (UTC)

An interview would be used to source opinion, and per WP:RS, An opinion is a view that someone holds, the content of which may or may not be verifiable. However, that a certain person or group holds a certain opinion is a fact, and it may be included in Wikipedia if it can be verified; that is, if you can cite a good source showing that the person or group holds the opinion. As long as the interview is archived in some form, either in a media file "as broadcast" hosted by a reliable source, or transcribed with the transcription agreed, again hosted or printed by a reliable source, or a quotation has been run by a reliable source, there is no concern. An interview can also be used to source facts the interviewee may give regarding themselves, the interviewee being primary source and the interview itself being the secondary source. You are correct that we cannot use an interview to assert A B and C where they are unrelated to the interviewee: if George Bush claimed there was an alien invasion due tomorrow we could not assert an alien invasion was imminent; we could however, assert George Bush stated on the 12th February that "an alien invasion is due tomorrow". But again, only if the interview was available for verification. I am unclear as to how possible it is to gain access to tapes or transcripts of radio and television transmissions for verification purposes from the broadcasters - that would be a flaw in the verifiability chain that would have to be examined. Hiding talk 22:54, 12 February 2006 (UTC)

You are saying that the transformation of an opinion into a fact, as explained in the NPOV policy and as you reexplain above, together with a weak notion of verifiability without a reputable source, is sufficient to transform anything that is not acceptable for inclusion into something that is acceptable for inclusion. For example, a crackpot theory about the universe is not acceptable in a respectable WP article on this topic, but if you transform it into the fact that Joe proposed this crackpot theory and this is sourced in some interview, it is OK to include that crackpot theory into this WP article. This is in conflict with the WP verifiability requirement that says that the fact that Joe proposed this crackpot theory must be supported in a reputable source before it is acceptable for inclusion. You are in conflict with an essential part of the Wikipedia policy. --Lumière 02:20, 13 February 2006 (UTC)

If the policy is like Hiding describes it, there are parts of the formulation of the policy that are completely misleading because they say the opposite. For example, there is the part of the policy that says that the role of a reputable publisher is to provide fact checking, some form of peer-review, etc. and also the part that says that what is a reputable source depends on what is being sourced. Somehow it is like Hiding did not read the same policy than me. Should we delete these parts? I think we should not delete these parts because they are important, but it is more important for me that we all agree on a policy. If there is a consensus that they are useless, let us delete these parts. -Lumière 05:22, 13 February 2006 (UTC)

• Lumière, I am totally unclear as to how both your statements above in any shape or form relate to my statement. You seem to have misunderstood me somemwhere, since everything you state my opinions to say is not even a misrepresentation; there is no resemblance between my words and your reading of them. If you wish to discuss the issue with me further, please use my own words rather than your understanding of them. Thanks. Hiding talk 18:06, 13 February 2006 (UTC)

If you say that I misunderstood you, I say great! I say it is great because I was in total disagreement with what I understood, and I would prefer to be in agreement. What may have confused me is that you never referred to the concept of reputable source, which is basic to the verifiability policy. The part that I like to hear about the verifiability policy is that a reputable source must provide some kind of review process to guarantee fairness and accuracy. -Lumière 18:29, 13 February 2006 (UTC)

Sorry, I may miss something, but is the gist of this discussion that we consider sources only verifiable if they still can be obtained from the original publisher? Because if they are not available from the original publisher anymore then who knows what may have been done with the radio interview or the magazine or the article by somebody for propaganda purposes. Does this mean that we cannot use newspapers as a source older than, say 30 years? Does this mean that articles that were once verifiable become unverifiable after a few years? Do we also have to check whether books are not sold out? Do we have to make a regular check for this? I bought the last copy of a certain academic book and used it for Wikipedia. Is this unallowed? Please note that the existence of this particular radio interview and certain assertions voiced in them are undisputed. Audio files of a radio interview that seems authentic and are claimed by various contributors of Wikipedia to be a copy of the original radio interview are available on the internet, so it is possible to check whether the audio file corresponds with transcript. Andries 00:32, 15 February 2006 (UTC)

The gist of the discussion is making a distinction between the ability to verify that an interview took place, and the ability to verify the assertions made in an interview. Being able to verify that an interview took place, and being able to verify that the transcription of the interview is correct, does not give this interview the imprimatur of "reputable/reliable source". A reputable source is described in WP:RS as being reliable because "Publications with teams of fact-checkers, reporters, editors, lawyers, and managers — like the New York Times or The Times of London — are likely to be reliable, and are regarded as reputable sources for the purposes of Wikipedia". That is clearly not the case with an interview, in which the interviewee can say pretty much whatever he/she wants. Clear now? A ≈ jossi ≈ t • @ 00:43, 15 February 2006 (UTC)

Thanks Jossi! Very clear! -Lumière 00:57, 15 February 2006 (UTC)

Jossi, I think you're slightly confused here. Yes an interviewee can say pretty much what they want in an interview. That doesn't negate it as a reliable source; it merely entails us to remember that what we are sourcing is generally opinion. However, were George Bush, for example, to announce in an interview that he is resigning as President, that is a fact, is sourcable and is verifiable. Note what needs to be verified with interviews is that the words were actually said, and by the person whom they are presented as being spoken by. Many broadcasters are reputable sources, consider for example the BBC. Hiding talk 09:21, 15 February 2006 (UTC)

No, I don't think Jossi is confused. What might be confusing is the term "reliable" instead of "reputable". Again, the main points of the policy are stated in terms of "reputable sources", not in terms of "verifiable sources" (see WP:V#The policy). So, I assume you meant "reputable source" when you wrote "verifiable source". The policy is clear that a "reputable source" is a source from a publisher that provides fact checking, etc., which is obviously more than only checking the authenticity of the authors and protecting the material against tampering to make sure that the attribution is correct. -Lumière 18:19, 15 February 2006 (UTC)

No, you're confused too. You have missed the part where I ask you to consider the BBC as a reliable or reputable source. However, since you state that what we're arguing about now though, is not verifiability but reliable sources. Are an interviewees own words opinion? Yes. Therefore, per WP:RS, If there is text, audio, or video available of someone expressing the opinion directly, it is preferable to include or transcribe an excerpt. Hiding talk 21:53, 15 February 2006 (UTC)

question about well known (by some) facts

Just to say from the outset, I agree that verifiability is an issue of the utmost importance to make Wikipedia a credible source. I have two specific questions about how it should be applied:

1. If a "top-level" article refers to another "lower-level" article in a section, through, say the Main template, and that article is thoroughly referenced, does the top-level article need to repeat the referencing of the lower level article? Can Wikipedia draw on Wikipedia as a source, in this limited circumstance? This is not an issue of verifiability, if the lower-level article is carefully referenced.
2. Academic referencing has two main purposes. The first is to reference statements the reader may not have heard, so he knows where to look for more information. The second is to properly attribute the work of others. Therefore, it is usually not necessary to attribute very widely known information. Can the same be said for Wikipedia?

This has come up recently. I've mentioned it on Wikipedia talk:Featured article candidates where it occured to me with respect to a comment on Talk:Big bang. Here is a scenario I've been thinking about. Hypothetically, if Big Bang were up on FAC (in reality, it's already featured) and someone mentioned the statement

Huge advances in Big Bang cosmology were made in the late 1990s and the early 21st century as a result of major advances in telescope technology in combination with large amounts of satellite data such as that from COBE, the Hubble Space Telescope and WMAP

is unverified, the nominator would probably Google some relevant words. It would be trivial to find a reference, say to CNN.com. It would accordingly be added, an oppose vote would change to support and everybody would be happy. But does adding random references really benefit Wikipedia? The statement about "huge advances" is known by virtually every physicist and astronomer, and could be verified by any number of sources:

• The popular press
• An article in Physics Today, Science or Nature
• Citing the original papers (HST, COBE, WMAP...)
• Textbooks or any recent history book
• Something more clever, like an of the number of academic papers mentioning cosmology over time

But the fact is that it is well known and accepted, and hopefully the reader that continues on and reads the whole article will come to accept this fact. From the standpoint of verifiability, it doesn't seem to matter which is used, but from the standpoint of the reader who actually looks at the footnotes, it probably does matter. I had always thought that statements that were common knowledge, at least amongst a certain (large-ish set, like physicists, Italians or squash players) don't really need a reference (at least, an in-line one). It is already becoming somewhat confusing for editors not familiar to Wikipedia to edit highly-referenced articles, because of the profusion of Wiki markup, and the necessity of referencing every sentence seems absurd. But what is the rule of thumb to decide what needs a reference? –Joke 01:32, 13 February 2006 (UTC)

It's simple, if someone challenges the statement and no supporting reference is supplied, the statement can be removed. This has nothing to do with whether or not something is 'well-known'. If something is indeed 'well-known', it will be easy for someone to provide an appropriate reference. It's even better if someone provides the reference up front. It can be embarassing if you put in something you think is 'well-known', have someone challenge it, and not be able to find an appropriate reference. I've had the experience of challenging something because it did not agree with the source I was consulting, have another editor supply their source, and then discover that the source they used does not agree with the source I have. In that case, we have to cite both sources and note the disagreement. I've also had the experience of wanting to add information about an important public event I personally witnessed, and not be able to find any reference source to support what I remember. -- Dalbury^(Talk) 02:24, 13 February 2006 (UTC)

I agree. There are no "discounts" as it pertains to Verifiability. What may be "well known" and obvious to someone, may be totally unknown to another. By providing a reference, we are safe and able to deter a possible challenge down the line. ≈ jossi ≈ t • @ 02:37, 13 February 2006 (UTC)

Ok great. So we then need to reference the following statements:

• "George W. Bush is president of the United States of America."
• "Germany and France share a border."
• "The colors of the American flag are red, white and blue."

Surely the line can be drawn somewhere and not every sentence needs a footnote. And please address my first, simple question (regarding "top-level" and "lower-level" articles) also. –Joke 02:58, 13 February 2006 (UTC)

I must agree with Joke, it seems to me that there is a line to be drawn with respect to refs. Some statements (like the ones he mentioned) clearly do not need references. However, I disagree with him about where this line ought to be drawn - I think, for example, that the statement

Huge advances in Big Bang cosmology were made in the late 1990s and the early 21st century as a result of major advances in telescope technology in combination with large amounts of satellite data such as that from COBE, the Hubble Space Telescope and WMAP

does need a ref. That said, reasonable ppl disagree here as it is a grey area. Debate on the issue is needed. Specifically, I think we need to discuss 1 (top level vs. lower level articles needing refs). Mikker ^... 03:18, 13 February 2006 (UTC)

Again, if someone challenges something in an article, supply a reference. There may well be readers of Wikipedia that do not know that George W. Bush is president of the United States, and I'b willing to be that there are Americans that do not know that France and Germany share a border. As for citing other Wikipedia articles as references, I have read in one of the policies or guidelines a stricture not to do it, but I can't find it now. One problem is that the other article may not be well referenced, or may deleted, merged, or otherwise made unsuitable as a referenc at some point in the future. Even if the article is well referenced, it is easy to bring the relevant references into the current article, and it is a courtesy to the reader to not make them go to another article to see the references. - Dalbury^(Talk) 03:27, 13 February 2006 (UTC)

So you would be ok with:

George Walker Bush (born July 6, 1946)<ref here> is the 43rd and current President of the United States.<ref here> Prior to his political career, he was a businessman in the oil industry<ref here> and served as the managing general partner of the Texas Rangers baseball team.<ref here>

Bush, a member of the Republican Party,<ref here> was elected 46th Governor of Texas in 1994<ref here> and was re-elected in 1998.<ref here> From there, he moved on to win the nomination of the Republican Party for the 2000 presidential race<ref here> and ultimately defeated Democratic Vice President Al Gore in a particularly close and controversial [1] general election. In 2004, Bush was elected to a second term, defeating Democratic Senator John Kerry.<ref here> This term will expire January 20, 2009.<ref here>

Bush's presidency has been defined by the terrorist September 11, 2001 attacks on the United States,<ref here> and his administration's response to the attacks. In the aftermath the U.S. and a multinational force took military action in Iraq, overthrowing and capturing Iraqi President Saddam Hussein.<ref here> The war proved controversial both in the United States and internationally.<ref here> In response to the September 11th attacks, Bush created the Department of Homeland Security and increased the powers of law enforcement agencies with the passage of the USA PATRIOT Act.<ref here>

? Mikker ^... 03:34, 13 February 2006 (UTC)

It depends. Quoting from Wikipedia:Citing sources#How to cite sources:

Sometimes — for example, when the article treats an uncontroversial or simple topic, and draws on a few, widely accepted general sources — it is sufficient to provide a "References" section at the end of the article, containing an alphabetized list of general references and authoritative overviews of a subject (such as textbooks and review articles). In other cases this is not enough, and in addition you should use in-line citations such as the Harvard references or footnotes described below.

I generally just use a References section, and only use in-line notes when it seems necessary to me. Often I cite two or three sources for the same fact, which is kind of hard to handle in-line. If I have multiple sources, then what I write is a blend of those sources, and in-line citing would be quite complicated. So a citation of a single comprehensive article or book about Bush could be cited in the References section. It is only when challenged about individual statements that a more detailed citations are needed. The simple fact is that 99.999% of items in Wikipedia are not going to be as well-known as who the presidnet of the US is, and need proper citations. Due to the inherenet confusion of where the to draw the line of what is 'well-known' enough to not need referencing, lets just reference everything. -- Dalbury^(Talk) 04:05, 13 February 2006 (UTC)

Exactly. ≈ jossi ≈ t • @ 04:35, 13 February 2006 (UTC)

There is the potential here for "source spam" used as a denial-of-service method, either as a protest of excruciatingly strict verifiability policies, or just for the sake of vandalism. If spammers insist on every single sentence being sourced, as in the examples above, then technically no one can do anything about them under the policies as stated, which can be interpreted as telling them to do it. To avoid this, the policies can state:

It is not practical to reference every sentence on WikiPedia. Please use common sense and provide references where they would likely to be required by a publication editor or college professor. Be aware that any fact, depending on the degree of a challenge, may be tagged with a source request, or moved to an article's Talk page for review, or removed entirely if it is thought to be obviously false, until a reliable source is provided.

--Coyoty 04:51, 13 February 2006 (UTC)

I don't think that the above paragraph entirely avoid the problem because it says that any fact may be tagged with a source request (or worst). Having a source request at every sentence is not very nice. The only solution is to draw a line with respect to refs as suggested by Joke, a line that would ultimately be determined by a consensus. Something like "... any fact that is doutful, ..." would work. The term "doutful" is subject to interpretation, but this is part of the solution because then in case of dispute a consensus can be used to determine if a fact is doutful or not. -Lumière 06:34, 13 February 2006 (UTC)

Thanks. I meant "doubtful facts", but it got lost in the translation to my fingers. --Coyoty 23:22, 13 February 2006 (UTC)

Sure, an editor could create trouble if we need to call a consensus all the times because of him, but this is not a new problem. Other WP rules can be used to implement this kind of attacks. For example, one can keep saying that the view that is proposed for inclusion is not notable. Or one can keep saying that the view that is proposed is not significant (NPOV requires that only significant views should be included). One could also dispute each fact and requires that it is attributed. For example, when the statement is "Bush is the president of the United State", a malicious editor could insist that this information is disputable and thus must be replaced by a statement of the form "USA Today reports that Bush is the president of the United State." An article in which every single fact is attributed will not look very good. The principle behind all these attacks is simply that a single person can force a call to a consensus all the times. Fortunately, WP:Consensus has been written with this kind of attacks in view and explains what to do in these cases. -Lumière 06:34, 13 February 2006 (UTC)

I mostly agree with Lumière, the drawing of the line should be determined in context by people who are familiar with the topic. Having marked my fair share of undergrad essays I can attest to this being a huge problem even in academia (an essay can have too many refs or too few refs; getting students to realise where the boundary lies is extremely difficult.) However, I must note that FA candidates (which is where the argument started - see the talk of WP:FAC) should be help up to quite a high standard... (but certainly still far away from the Bush article I over-referenced to make an earlier point). Mikker ^... 06:50, 13 February 2006 (UTC)

There are some great comments above. I think we all pretty much agree that a reasonable line needs to be drawn between what needs an in-line reference and what doesn't. I guess I think it is a bit of a moving target, based on what the article is and the audience that is most likely to read it, because the absurd level of referencing in the GWB article above can't suddenly become the norm for more technical articles, when some specialized knowledge is very widely known within a field but is not widely known by laymen. Generally, the rule of thumb I use is that if you provide general references that each independently confirm the basic facts of an article (say, in the case of GWB, a couple of books and newspaper articles about him) then these basic facts shouldn't each need an in-line reference. However, very specific statements (e.g. "On February 13th, GWB took his dog for a walk") should be referenced in-line, as should more controversial statements that someone is likely to dispute (e.g. "GWB is the hardest-working president, ever."). –Joke 16:46, 13 February 2006 (UTC)

I think the standard should be specificity. In particular, we need to have some idea about common knowledge. Most newspapers, for example, keep detailed rules about common knowledge (for example, there is a list of New York Times cities that do not require context -- Minneapolis doesn't while "Saint Paul, Minnesota" must always have the state name after it). There are facts referenced, for example the fact that Joke pointed out in the Big Bang article, which have numerous sources and no one authoritative source. In that case, it hardly makes sense to reference the point since a judicious reader would use the references section at the bottom of the article and find a few places where this summary fact could be culled. --ScienceApologist 17:46, 13 February 2006 (UTC)

There is another more problematic point. When writing in Wikipedia:Summary style sometimes statements rely on the summary of many sources. These don't lend themselves well to inline citing. What to do? --ScienceApologist 17:46, 13 February 2006 (UTC)

Good point. I guess one thing that disturbs me about providing a reference for "common knowledge" is the arbitrariness of it. With many references to choose from, why would you pick just one? For example, Lumière pointed out that a malicious editor could insist on the attribution "USA Today reports that Bush is the president of the United States." Or maybe, if you wanted to make it sound like he's probably really the president, you could say, "Most major newspapers – including USA Today, the New York Times, the Washington Post, the Guardian and the Süddeutsche Zeitung – report that Bush is the president of the United States." But why were these sources cited and not others? I mean, it's a bit of a straw man, but the argument carries over to other things, like the statement above about the Big Bang, or that the atomic number of oxygen is 8, or... You can push the problem into footnotes, but it doesn't obviate the central fact that you are providing references that are chosen almost entirely at random from among an umanageably large number of possibilities. –Joke 18:00, 13 February 2006 (UTC)

I don't see why this is a problem at all. A reference is a reference. It's not a certification of truth. You say, "here's this fact, here's where it's been published before, you can verify that it's been published there, and you can judge whether you trust that publication." Only one reference is needed. It's important that it be a good one, where good means a) one that's easily checked, and b) one that's well-known (so the reader can easily judge whether or not to trust it). It's not important that it be the best one. Here, the choice can be fairly arbitrary.

And it's likely to be stable. There's no reason to edit-war about it, because if something is generally thought to be true, there will be many references that all say about the same thing. (The only reason to use a different reference is would be if it said something utterly different—but in that case, it would mean there's a different point of view to be stated separately and supported by that reference. Dpbsmith (talk) 00:03, 14 February 2006 (UTC)

Yes, if reliable references disagree, that should be covered in the article. I often 'blend' more than one reference because they overlap, so that, say, all three references cover some points, and other points are covered by only one or two of the references. So, unless I think something is going to be controversial, or I'm adding a specific point from a reference that is not otherwise relevant to the article, or the article already uses in-line refs, I just put my refs in the References section. -- Dalbury^(Talk) 00:56, 14 February 2006 (UTC)

What is the difference between reliable and reputable? The main points of the policy (see WP:V#The policy ) are stated in terms of "reputable sources", not in terms of "reliable sources". Why is it that Dpbsmith and Dalbury do the opposite and only uses "reliable" and not "reputable"? Are they only referring to the details of the policy that consider reliability? Or is it that reputable and reliable are synonymous? -Lumière 01:35, 14 February 2006 (UTC)

Intuitively, a source that is not reliable cannot be reputable, but a source can be reliable and not reputable. There is more to reputability than just reliability, but this is just the situation in general. In the specific context of the policy, the situation might be different. -Lumière 02:00, 14 February 2006 (UTC)

I guess one difference, for better or worse, about referencing on Wikipedia as opposed to say, Brittanica, or academic referencing, is that a large component of it is simply about facilitating easy fact checking. The references don't have to be good, authoritative, memorable or even worth reading. They only have to be reputable. This leads to some strange situations, where totally random sources are used simply because they are convenient (e.g. see reference [3] of the Wayne Gretzky article, which is to Cigar Aficionado). –Joke 02:24, 14 February 2006 (UTC)

There is perhaps some sarcasms in your comment. It's true that for people in general "reputable" is more superficial than "reliable". This explains why you say "only have to be reputable". I was wrong when I proposed that in general, "reputability" is more than "reliability". I guess these two concepts are simply not comparable -- they are about different things. However, sarcasms put aside, the expression "reputable source" has a very specific meaning in the context of the policy that is not the ordinary meaning. For example, it is said that a reputable publisher (i.e. one that publishes reputable sources) is expected to provide some form of review process to guarantee accuracy and fairness. Also, it is said that for scientific content it is better to use peer-reviewed scientific journals. This means that, if you don't use peer-reviewed journal, you run a chance that your sources will be disputed. So, the purpose of referencing is more than only allowing easy fact checking, but also to benefit from some quality control that is provided by the publisher. -Lumière 02:47, 14 February 2006 (UTC)

No, I assure you there was no sarcasm intended! Particularly with my use of reputable, I just used it because you mentioned WP:V uses it, but I could equally well have said reliable. My main point is that the source can be perfectly valid, but it could nonetheless be unlikely that any reader would actually use the source, except as a way to fact check the article. This is quite different from most kinds of referencing. Although Cigar Aficionado may be reputable (OK, that's a bit comical, but the same could be said for the New York Times), I certainly wouldn't read it if I was looking for detailed, thoughtful articles about hockey. –Joke 03:15, 14 February 2006 (UTC)

Ok, I think I understand what you are saying. You are saying that references in Wikipedia are mainly used for easy fact checking, that is, to support or endorse contents already in the article, not to provide additional information that is not provided in the article, which is a typical use of references in other media. I agree, except for the "easy" part (see #Articles should contain only material that has been published by reputable sources.), but my main concern here is that you don't say enough that these "easy" fact checking are very important in Wikipedia and should be done with reputable sources. This is because Wikipedian editors are not equipped to directly check the validity of contributions from other Wikipedian editors. In other words, a wikipedia article must rely more on its sources for its fairness and accuracy than other encyclopedia articles do. So, the reputability of the sources is even more important for Wikipedia articles than for other type of articles. This is why I could not understand your statements "The references don't have to be good, authoritative, memorable or even worth reading" and "They only have to be reputable." In particular, in the spirit of the WP policy, "accurate", "authoritative" and even "worth reading" are considered a part of what "reputable" means. This is also why I do not understand why some seem to be saying, but maybe I misinterpret them, that the role of sources in Wikipedia is only to correctly attribute statements or views before we accept them for inclusion in an article. If this was the case, wikipedia articles would be full of contradictory views such as Professor Ingrid said X, Professor Joe said not X, etc. and the readers would be so often left in a state of confusion because it is hard to know which scientists is right. WP policy attempts to avoid this state of confusion as much as possible: it says that a scientific content is preferrably sourced in a peer-reviewed journal (or otherwise, as usual, the reputability of the source can be disputed and the contribution removed until a consensus is reached.) Of course, it is possible that two contradictory views are each supported in their respective reputable sources. In this case, they would have to be both included in the article, but usually the peer-review process guarantees that some explanation for the conflicting views is provided in the sources, and the associated state of confusion can be avoided. At the least, inclusion of views that are in conflict happens less often and only with scientific views that have been carefully considered in their respective peer-review process. -Lumière 06:39, 14 February 2006 (UTC)

Here is another way to understand the need for reputable sources. We all know that it is not appropriate to repeat every thing that some person said, even if it is a fact that this person said it. In normal life, this is what we would call gossips or propagation of (sometimes unpleasant) rumours. A policy that would accept any correct attribution could be used to support exactly that kind of gossips or propagation of rumours. It is innaceptable to have an interpretation of Wikipedia policy that supports gossips or propagation of rumours. The issue here is not only accuracy, but also fairness because these rumours can be totally unfair. A tabloid might propagate a lot of these rumours, but an encyclopedia should have an higher level of standards. The requirement for reputable sources is to avoid these innacurate or unfair stories. -Lumière 06:39, 14 February 2006 (UTC)

The conclusion is that we should have very high standards in our definition of reputable sources, and to achieve these high standards, what is a reputable source should depend upon the content that is sourced. We must have high standards for scientific contents and different high standards for religious content, etc. -Lumière 06:39, 14 February 2006 (UTC)

So where are we at

I've just removed a couple of inappropriate notices from Wikipedia:Verifiability/Jguk's version. But at Wikipedia talk:Verifiability/Jguk's version#Big Bang, User:Jguk says I'm therefore being BOLD and, after tweaking the wording of this proposal to state explicitly that there is no carte blanche for mass deletions, making this version live.

That was on 31 January. It doesn't seem to have happened, and I can't find further discussion of this proposal. Maybe I've missed it. Or as User talk:Jguk reads If you wish to send me a message, please email me here. Otherwise assume I have not seen and will not see your message. Thanks, jguk, maybe the discussion continued on email, where nobody can find it. Or, maybe they merely meant this edit by going live.

I don't think we can have two different versions both belonging to Category:Wikipedia official policy, and bearing a notice claiming to be official Wikipedia policy (let alone a notice saying WP:V redirects here... no, it doesn't). It may be obvious to those in the loop what is happening, but it's not all that obvious to the rest of us.

Personally I think this big bang is a very bad idea, and I also have my doubts about Jguk's experiment with their user talk page.

(And this talk page is also badly in need of archiving.) Andrewa 06:13, 14 February 2006 (UTC)

Articles should contain only material that has been published by reputable sources.

I believe that sentence is a bit ambiguous, and I think it should be clarified.

The way I read it, it seems to prohibit even the minimum of aggregation and reformulation that separates an encyclopedia from a mere selection of quotes and summaries. I'm worried about what Wikipedia would look like if all the content were indeed written by people who have fun solely summarising what they read in external sources, without any creative process whatsoever.

I think it might be better if this said that WP articles should only use such information, or "contain material" were replaced by some clearer term -- one, I would think, that made it clear that it is okay for articles to include sentences that I might not be able to find a single source for, by combining several sources. If I can find a source saying that world population in the 1940s was around 2.3 billion, it should be okay for me to write that world population during World War II was around 2.3 billion, without having to find a source that includes this specific statement. (Of course, as the policy currently stands, it seems to suggest it would be perfectly okay to remove that statement if it didn't provide a source for the time World War II happened. I'm glad most people seem to apply more common sense than required by a strict reading of the policy.)

RandomP 15:31, 14 February 2006 (UTC)

I totally agree! I would use the expression "supported by reputable sources" instead of "published in reputable sources", but it could be something else. Whatever is the expression that we use, the concept should be clarified in the policy. To put this issue into its context, I would add that this is only one half of what we need to clarify. The second half is what we mean by "reputable sources". -Lumière 17:00, 14 February 2006 (UTC)

Logic and verifiability

Earlier this month, there was a small discussion on Talk:Brian Peppers about whether or not deductive reasoning based on facts from verifiable sources counts itself as verifiable information. (The discussion was since removed when the article itself was deleted. I have no interest in whether or not the Brian Peppers page exists; I just want to get to the bottom of the logic issue.)

Somebody wanted to point out in the article that Brian Peppers lives in a nursing home. Somebody else pointed out that there is no reliable source that states this. In a literal sense, this was true, but there were in fact two sources through which one can deduce this. One reliable source listed Peppers' address. Googling for that address resulted in a hit for a nursing home, with exactly the same address listed. If it happens that the nursing home is itself a reliable source, then simple logic dictates that Brian Peppers does in fact live in a nursing home.

To put the question simply: is deducing something (using pure logic and not conjecture) from two reliable sources to come to a single conclusion verifiable information, or instead original research?

I argue that it is verifiable information. If the reasoning used to make the conclusion is fallacious, then obviously such a conclusion should not be considered verifiable information, but suppose the reasoning is sound. Then what? It technically is original research in a sense, because we did arrive at a conclusion that wasn't stated anywhere else. Yet it is technically verifiable information as well because it all "checks out". The individual facts are verifiable and irrefutably work together to produce that conclusion. What is a fact but the sum of its parts?

Here's most of the relevant discussion, verbatim:

---

• Nobody wants to list his address, they simply want to say he lives in a nursing home. There is a big difference between the two. VegaDark 06:06, 2 February 2006 (UTC)

• Once it is published by a reliable source that he lives in a nursing home, we may include it. Otherwise, you're talking about original research. What parts of WP:NOR and WP:RS and WP:V do people not understand? FCYTravis 06:08, 2 February 2006 (UTC)

• I think you're being kind of unrealistic here. We have a link saying A=B, and a source that B=C, but you think we need a source saying A=C before we can add it. VegaDark 06:25, 2 February 2006 (UTC)

• Facts about a person's life are not abstract letters and numbers to perform mathematical gymnastics with. Either we have a verifiable reliable source that says he lives in a nursing home, or we don't. FCYTravis 06:32, 2 February 2006 (UTC)

---

Is there any consensus on this matter? - furrykef (Talk at me) 10:10, 15 February 2006 (UTC)

Hmm, it's a thorny issue. It appears to be Original Research, but then does that make it wrong to look at a newspaper to see what date it was published? I suppose the way you do it is cite both sources; A is declared by X as living at B, which declares itself to be C in Y. Then people can argue over the sources; it isn't original research. Some people would also consider it acceptable to move that text to the footnote and allow the assertion that he lives in a nursing home in the article. Hiding talk 15:15, 15 February 2006 (UTC)

A point to take into consideration is that the purpose of a reputable source is not only accuracy, but also fairness. It could be that no reputable publisher would accept to publish the fact that Brian Peppers leaves in a nursing home, even though a reputable source publishes his address and another unrelated reputable source publishes the fact that it is the adress of a nursing home. Similarly, stating these two facts one after the other in the WP article is just an indirect way to state that he lives in a nursing home. Combining two or more sourced facts with the intent to communicate something that is not sourced count as original research.

However, in general, this does not mean that every sentence of a WP article must be found word for word in its sources. As long as a WP article only describes a debate or a content that is already provided in these sources, it is not original research. It does not have to be exact quotes. This is obviously the kind of situations that might require a discussion amongst the editors. However, the policy should explain that a good understanding of the sources might be necessary to verify that the WP article is well supported by these sources. The policy should encourage the participation of experts that can understand the sources and then report on them. -Lumière 16:15, 15 February 2006 (UTC)

I don't think it's enough to say, "From this evidence, it's clear that this conclusion is true." It may not be as clear as you think, and if your conclusion is false, you open yourself and Wikipedia to liability issues. In this case, having the same address as a nursing home does not necessarily mean Peppers is living in a nursing home. He may be living in the same building, but a different suite. I've seen that happen, with the address being the same, even a suite number being the same, because someone got it wrong. He may be using the nursing home address as a mail drop. Some businesses will take mail for people whose schedules or other circumstances keep them from collecting it at home. There may be some other reason for the addresses being the same without him living there. The only true thing you can say is that his address matches that of the nursing home. You would need other sources to establish that he lives there. --Coyoty 16:23, 15 February 2006 (UTC)

Yes this is another justification to remove a statement that would directly say that he lives in a nursing home. However, your argument would still allow someone to state both facts one beside the other in the WP article, so that the readers can conclude by themselve. This is also not acceptable. It still counts as original research. We can also say that it is an insinuation which is against NPOV, as clearly explained in the NPOV tutorial. -Lumière 16:43, 15 February 2006 (UTC)

Well, you see, Brian Peppers was a bad example, but I used it because this was the discussion that actually happened. Of course, the counterpoints you two raised are all valid, and would warrant the removal of the information from the article until they could be addressed by other verifiable sources. Let's go back to the original discussion from the talk page. VegaDark pointed out that if A=B, and B=C, then A=C, which is a logical axiom (and therefore irrefutable). In the Brian Peppers example, as pointed out, the conclusion doesn't necessarily follow the premise. So it was fallacious reasoning. However, I make my question on the assumption that the conclusion does follow the premise. That is, would it be acceptable then? Simple peer review can help determine if the conclusion really does follow the premises -- as demonstrated by your refutation of the argument. I don't see a problem with allowing an argument that passes that test. It could indeed be a violation of Wikipedia:No original research, although I don't think it violates the intent of the "no original research" rule. Perhaps a discussion should be started at Wikipedia talk:No original research as well. - furrykef (Talk at me) 16:51, 15 February 2006 (UTC)

Coyoty and Lumiere reiterate a point I was making; that the argument isn't about original research, it's about the sources. I don't think amalgamating two sources is original research: if we follow that line then if something states it was published on the 15th February 2006 and another source states that 15th February 2006 was a Wednesday, people are claiming it is original research to state the work was published on a Wednesday. The argument isn't about original research; it's about the verifiability of the sources. Note that sometimes the information isn't going to be accurate, that's not important, only that we have reproduced a verifiable source. This allows, as per discussion above, the challenging and removal of the information on verifiability grounds. The argument regarding the nursing home address isn't that it is original research, it is that we can't verify it is correct to state Brian Peppers lives in a nursing home. We can verify the conflation of the two sources though, but as arguments above show, that information may not be encyclopedic or may be falsely assumed. To my mind this is still not an original research issue but a verifiability one. Hiding talk 17:25, 15 February 2006 (UTC)

Honestly, I never really understood the difference between verifiability and no original research: they both states that reputable sources are required. I think that one small difference is that No original research provides a context, a motivation, for verifiability, but it does not provide new rules. Similarly, I don't see that verifiability provides fundamental new rules that are not already stated in No original research. Verifiability goes into more details about the importance that attributions are correct. In fact, this creates confusion because some tend to believe that the views that are attributed are not important in themselve, but only their attribution needs to be correct. To the contrary, the role of a reputable publisher is also to check that these views are accurate, fair, etc. Verifiability is more than only checking that we correctly attribute views. -Lumière 17:52, 15 February 2006 (UTC)

Again I'm not sure that we're misreading each other. However, verifiability does create extra provisions outside protecting against original research: it also minds us that information may not be of worth even though it is not original research, and that information that is false may be acceptable if it is correctly attributed and verified. I agree however that it is not only the attribution that needs to be checked but the content as well. Hiding talk 18:30, 15 February 2006 (UTC)

The provision against non worth it information is satisfied as soon as we require reputable sources. You miss the point that the No original policy goes beyond the original purpose that is reflected in its name. It is clearly explained in the No original research policy that it does more than only filtering out new unpublished theories, but also works toward the goal of Neutral point of view and the goal of Verifiability.

Also, I don't understand what you mean when you say that false information maybe acceptable if it is correctly attributed. If, as should be expected, the false statement is not found in a reputable source, at the least its attribution, say "Joe said <the false statement>", must be found in a reputable source. Then it is clear that the fact that Joe said this false statement is something worth it, fair to report, etc. Also, the spirit of the reputable source must be respected. For example, if it clear from the reputable source that the statement is false, this must be reported as well.

You cannot source "Joe said <a false statement>" from an interview where Joe was the interviewee because the interview is not a reputable source. Well, there may be some kind of contents for which an interview can be used as a reputable source, but this would have to be discussed before it is inserted in the policy. Just to present an extreme case, you cannot use any interview to source a new scientific theory in a WP article on a related scientific topic, even if you attribute the new scientific theory to its source. - Lumière

Again you misread me. I believe part of the problem is this: you are arguing from a perspective of preventing people from inserting scientific theory based upon an interview. I am arguing from the perspective of allowing people to source a resignation statement from a world leader from an interview. Let us clarify our positions. Interviewees express opinions in interviews. Those opinions are sourcable, and useable if of note. The interviewee is a primary source, and if the interview is broadcast by a broadcaster of repute then my reading of policy is that it is sourcable. You seem to acknowledge that. However, note we can source false statements from interviews, again following the proviso's above. For example, Jeffrey Archer has made several false claims regarding his background in interviews. These have become notable and are therefore sourcable. Also, interviews are already covered in policy, as I pointed out above. ALso note, facts an interviewee reveals about him or herself in an interview are allowable, in this instance the interviewee is a primary source, and is therefore sourceable, if the interview is with a reputable broadcaster, per WP:RS: We may not use primary sources whose information has not been made available by a credible publication. For example, an interview broadcast by the BBC is one I would consider a reputable source, and fair game for sourcing facts regarding or opinions held by the interviewee. However, you are completely correct to state that new scientific theory may not be sourced in such a way, a scientific theory is not opinion, nor is it a fact relating to the interviewee. You may however, if it is notable, source the fact that Joe has stated he is working on a particular theory, or that he described a particular theory. This would depend upon the notability of Joe, and the notability of the topic. I would consider Einstein commenting in an interview that he was working on a theory pertaining to unifying the fields may be of note in providing historical datings of his advancing of and working upon the theory. Hiding talk 22:16, 15 February 2006 (UTC)

Point by point

Again you misread me. I believe part of the problem is this: you are arguing from a perspective of preventing people from inserting scientific theory based upon an interview. I am arguing from the perspective of allowing people to source a resignation statement from a world leader from an interview. Let us clarify our positions. Interviewees express opinions in interviews. Those opinions are sourcable, and useable if of note. The interviewee is a primary source, and if the interview is broadcast by a broadcaster of repute then my reading of policy is that it is sourcable. You seem to acknowledge that. Hiding

Yes I did agree that an interview can be used as a reputable source, but only for some type of content. In particular, since the interview is a primary source, it should not be used as if it was a secondary source. Lumière

• That's something we agree on then. Hiding talk 22:08, 16 February 2006 (UTC)

However, note we can source false statements from interviews, again following the proviso's above. For example, Jeffrey Archer has made several false claims regarding his background in interviews. These have become notable and are therefore sourcable. Hiding

I agree, but I need to clarify the context. I assume that you are not suggesting that one can say "Jeffrey Archer said <some false claim¨>" without making clear that it is a false claim. An interview does not normally say that the claims that it contains are false. Therefore, if it is notable that the claim is false, it must exist some reputable source other than the interview about this notable fact. This reputable source must be cited together with the interview. This is not a disagreement, just a clarification of the context. Lumière

• I'd taken that as read, apologies. Hiding talk 22:08, 16 February 2006 (UTC)

Also, interviews are already covered in policy, as I pointed out above. ALso note, facts an interviewee reveals about him or herself in an interview are allowable, in this instance the interviewee is a primary source, and is therefore sourceable, if the interview is with a reputable broadcaster, per WP:RS: We may not use primary sources whose information has not been made available by a credible publication. For example, an interview broadcast by the BBC is one I would consider a reputable source, and fair game for sourcing facts regarding or opinions held by the interviewee. Hiding

Which part of the policy says that an interview is sourcable if it is about the interviewee. A self-published source such as a personal website can be used to source information about the website or its owner, but an interview is usually is not self-published. A source of dubious reliability can be used as a source for information about itself. The key ingredient is the self-reference. The interviewee is different from the interview source. In particular, the interviewee does not control whether or not the interview will be kept public. So, this rule does not apply. I only accept that the interview can be used as a reputable source for some kind of content, which need to be defined, and only as a primary source. Lumière

• I have no idea what you mean here. There appears to be a huge communication issue between you and me on some level. However, addressing what I think you mean, policy at WP:RS states If there is text, audio, or video available of someone expressing the opinion directly, it is preferable to include or transcribe an excerpt. Therefore there is policy that states an interview is sourcable if it is an opinion. However, in this instance there is an issue regarding verifying whether an interview is conducted with the person whom it is purported to be conducted with. For example, do we source opinions from interviews conducted with notable people by websites which we would not personally hold to be reputable/reliable sources? We have to ask ourselves if we trust the site which hosts the audio file to be presenting the audio file as is; i.e. that the audio file is not doctored or hoaxed or made up. A fact regarding an interviwee, expressed by said interviewee in an interview, is a fact drawn from a primary source, therefore it is covered by policy regarding primary sourced material.

However, you are completely correct to state that new scientific theory may not be sourced in such a way, a scientific theory is not opinion, nor is it a fact relating to the interviewee. Hiding

I agree with the conclusion, but do not follow your argument. My argument is simply that a published scientific theory is normally a secondary source, not a primary source. So, when you need to source a scientific theory, a reputable secondary source is expected. If the theory is only mentioned or explained in a primary source, this primary source cannot be used in replacement to the expected secondary source. Lumière

• Our line of argument is identical, merely couched in different terms. Hiding talk 22:08, 16 February 2006 (UTC)

You may however, if it is notable, source the fact that Joe has stated he is working on a particular theory, or that he described a particular theory. This would depend upon the notability of Joe, and the notability of the topic. I would consider Einstein commenting in an interview that he was working on a theory pertaining to unifying the fields may be of note in providing historical datings of his advancing of and working upon the theory. Hiding talk 22:16, 15 February 2006 (UTC)

I agree if you are simply giving an example where an interview, which turns out to mention a scientific theory, can be used as a primary source. For example, this interview might be acceptable as a primary source for a WP article about the history or the philosophy of science, but I am not even sure -- remember Wikipedia is not equipped to include original research. Certainly, it cannot be used as a source for a WP article that is directly about the scientific topic where the scientific theory would normally fit. Lumière

• Again I am not sure I understand your point here. It would help if you address the examples I give, as I would then have a better understanding of your issues. As I state, I see no problem with inserting into an article on, for example, Einstein's unified field theory, a fact revealed in an interview regarding his working on the theory, if that revelation pre-dates any other reference to his first mention of the theory in the article. If the article said Einstein started work on the theory in 1911 for example, and an audio interview exists from 1910 where Einstein declares himself as working on it, then I would not believe the insertion of such information was original research. If the audio is from a reliable/reputable source, I would have no problem with moving back the date based on the source. Hiding talk 22:08, 16 February 2006 (UTC)

The important point here is that I only agree that an interview can be used as a primary source for some kind of content and only where this kind of content is acceptable. The fact that the interviewee is allowed to say about anything he/she wants, severely limits the type of content that can be sourced with an interview. Certainly, the fact that it cannot be used to normally source a scientific theory is just an example. Another example where an interview cannot be used as a source is to critic some notable person inside an article about this person, unless this critic is notable and separately sourced in a reputable secondary source. Otherwise, it means that Wikipedia is supporting propagation of rumours and gossips.

The general principle is that what is a reputable source must depend upon the content that is sourced. The policy would be more useful if it provided more specific information about what is a reputable source for different type of content or, similarly, what type of contents can be sourced with different type of reputable sources. I would say that this would be particularly useful in the case of interview. -Lumière 03:24, 16 February 2006 (UTC)

• I broadly agree with your first point, but I think policy covers your second very well as is. Broadcasters which employ fact checkers, researchers and the like would be reputable/reliable sources per policy, podcasters working from their own bedroom would not. Hiding talk 22:08, 16 February 2006 (UTC)

I did not read Hiding comment when I wrote this comment. It is a reply to Furrykef. The example Furrykef provides is too trivial to allow us to appreciate the general situation that I think he/she has in mind. If I nevertheless stick to this example, the criteria should be whether or not A = C is going beyond the essential content of the sources, which separately state A = B and B = C. In other words, the editors should agree that A = C is not original research, but is just part of a different way to communicate the content that is sourced. Again, this is the kind of situations where a discussion amongst the editors might be necessary. We should not require that every sentence appears word for word in the sources. The role of expert editors should be mentioned here: they do not have authority, but their understanding of the sources should be useful in the discussion. It is not the first time that the WP policy requires interpretation. It would be a mistake to try to have a policy that does not require interpretation. -Lumière 17:32, 15 February 2006 (UTC)

This is a trivially easy case. "State records show that Brian Peppers lives at *INSERT ADDRESS*. According to *INSERT OTHER SOURCE*, *NAME OF NURSING HOME* is located at *SAME ADDRESS*." State the facts, let the reader interpret. Hipocrite - «Talk» 18:20, 15 February 2006 (UTC)

You exactly say the opposite of what I said. The difference between me and you is that I referred to specific parts of the policy to support what I said, and I can be more explicit about these parts if you want. A useful reply to what I said would consider these parts of the policy or ask for more precision. It does not help the discussion just to say the opposite of what I said. -Lumière 18:49, 15 February 2006 (UTC)

One could also consider the possibility that I don't read what you write, ever. Hipocrite - «Talk» 19:29, 15 February 2006 (UTC)

What do you have against me? What did I do to you? -Lumière 19:38, 15 February 2006 (UTC)

[3]. Stop spamming the policy talk page. Hipocrite - «Talk» 19:39, 15 February 2006 (UTC)

I reply to others and it turns out that I work on different aspects that go on at the same time on this talk page. Also, many of my edits are just small corrections/additions. This is not spamming. -Lumière 19:51, 15 February 2006 (UTC)

No ofense, but please read Wikipedia:What is a troll, because that is the way that many editors are seeing you as, given the innordinate amount of comments to policy pages in the last few days. ≈ jossi ≈ t • @ 20:00, 15 February 2006 (UTC)

Thank you for the information, but where did you take it? Are people gossiping about me somewhere? Now, read this

"Trolling is deliberate and intentional attempts to disrupt the usability of Wikipedia for its editors..."

So, I am not a troll because I only have good intentions. As far as the numbers of comments that I write is concerned, there are certainly not enough of them to prevent other editors to discuss. -Lumière 20:07, 15 February 2006 (UTC)

Lumière could be more careful about making sure his replies are complete before posting them, so he doesn't keep going back to rewrite them, but that doesn't make him a troll or a spammer. I've found everything he's written to be on topic and rational, and if he has a difference of opinion, that doesn't make him a troll, either. --Coyoty 20:46, 16 February 2006 (UTC)

Coyoty, thank you. I acknowledge that I should be more careful as you said. It does make the history less easy to use to determine what are my contributions. If it is what Jossi and Hipocrite had in mind, but just did not express themselve with the proper terms, then I accept the critic. Still, I am definitively not a troll or a spammer. -Lumière 21:20, 16 February 2006 (UTC)

I'm glad you brought this discussion up, furrykef. This kind of got lost in the midst of the whole Brian Peppers debate. I was quite pleased to find this when looking at the "what links here" section of my page, as I would like to get this settled. Yes, Brian Peppers is a bad example for this, as there is so much controversy over that. Let me give another example: Source 1 says that a soldier died on March 5th. Source 2 states that the only soldier that died on march 5th died of a gunshot wound to the head, but doesn't state the specific soldier. Are we allowed to make the connection on the soldier's page and say he died by a gunshot wound to the head? Or would that be considered OR? I like the suggestions of simply stating what both sources said, but I think it would detract from the article in practice. Also, for something as specific as an address, we aren't going to list that on the Wikipedia page, so it would make that type of thing harder. VegaDark 05:52, 3 March 2006 (UTC)

logic, NOR, verifiability: retake

Sorry, this section went on for a bit; I thought it would be good to start fresh. let me admit my bias: to my mind, the purpose of an encyclopoedia is to inform, and this is more than the presentation of mere data. any time you put two sentences about some topic near each other, the human mind will naturally and inevitably try to draw those two sentences into some over-arching relationship, and I don't think a good editor should ignore that. I think he or she should try to aid the reader in making balanced, informed conceptual structures. this thought is already built into wikipedia, mind you, in disambiguation pages (which structure valid content domains).

I think one issue I have with the verifiability policy is that it's designed with scientific theories in mind, and that model doesn't necessarily extend itself well to other kinds of information. for instance, the insistance on publication by reputable sources is fine for excluding untested and unproven theories in physics or chemistry, but it becomes much more difficult to apply if you shift to the social sciences, and impossibly congested when you enter into history or the humanities or out into the broader social world. I can think of numerous instances in psychology, anthropology, political science, etc., where mutually contradictory theories are held and published concurrently, even in the same journals, the situation is more difficult when you get to cultural and literary criticism, and it's impossible in the real world. really, what holds knowledege together is logical induction, not logical positivism; if you disallow logic as a valid source, you might as well disallow discussing anything you can't poke with a stick.

don't get me wrong, I understand the issues. we want to:

• exclude unproven and untested scientific theories
• exclude misinformation (intentional or not)
• exclude the presentation of distinct opinion as fact
• exclude material that is intentionally offensive, argumentatitve, or overly-narrow

• (noting that it's the intention at issue, not the material)

all of these things are undesirable because they misdirect and generate disagreement and confusion. but if the goal is not to misdirect, well... one can misdirect just as easily through what isn't said as what is. There are times when you just can't make sense of a topic without weaving a logical basket to put all the separate facts in, and I don't think that should be excluded as a matter of policy. (whoops, forgot to sign) Ted 05:27, 27 February 2006 (UTC)

Reputable sources

My interpretation of "reputable sources" is that they are sources with good reputations, i.e. "trusted". Would it be less confusing to say "trusted sources"? --Coyoty 20:55, 16 February 2006 (UTC)

I prefer reliable, the phrase used for the page itself, Wikipedia:Reliable sources. They are sources which have been proved to be reliable. Hiding talk 23:09, 18 February 2006 (UTC)

Propaganda websites as sources

Is there a way to prevent editors from adding material from propaganda websites without resorting to a heated debate every time? Do we have a "blacklist" of websites that are known to be POV/propagandist in nature? What is an easy way to argue out with someone that the source is not "trustable"? A huge number of conflicts arise on wikipedia due to these kinds of issues. Sometimes I am not able to correct blatantly false and misleading information, because I don't feel like wasting time fighting over the trustability of sources. deeptrivia (talk) 00:10, 19 February 2006 (UTC)

IMO, the fact that a source must be published (i.e. sent to a reputable publisher) prevent sources that are strongly POV. In particular, this rule alone prevents the use of websites (of organizations or individuals) as sources. In addition, the policy explicitly prohibits personal websites as sources (but it is redundant). In some cases, organization might be big enough to have their own reputable publication. In this case, the "third-party" part in "credible third-party publication" in the section WP:V#Sources should do the job. Note that it is OK that some information that is self-published in some personal website is used as a source, but only when the information is about the person (or the website) and is used in a WP article about the person (or the website) -- and this freedom also applies to organization as well. Perhaps, if you gave one example of what you mean, we could see how the policy applies. -Lumière 00:51, 19 February 2006 (UTC)

My comment from Wikipedia talk:WikiProject Fact and Reference Check:

"My particular concern (that motivated me to write this) is with http://www.dalitstan.org/ and http://www.ambedkar.org, both notorious websites, being used as sources on wikipedia. Some people just want to believe blatant POV as the Truth. Simply telling the editor that your source is untrustable will produce ego hurts, edit- and revert-wars, and all kinds of tensions. Is there a nice way to go about it, that retains goodwill? Other such sites include http://www.hinduunity.com/. deeptrivia (talk) 00:32, 19 February 2006 (UTC)" deeptrivia (talk) 01:02, 19 February 2006 (UTC)

The three examples that you give are obviously not valid sources, except for information about the organizations themselve and only in a WP article about themselve. The policy even adds that not all information that they might have about themselve is acceptable for inclusion. The problem that you have is that it seems that people don't understand what "publisher" means. The organizations behind the websites that you mention aren't publishers! I noticed that one of them had its own journal, so it plays the role of a publisher. Still, the organization cannot use this journal as a source because it is not a third-party source. The policy is clear that only credible third-party sources can be used, but again the problem is that it is not understood! Sorry, but I cannot help you there. -Lumière 01:50, 19 February 2006 (UTC)

Thanks, Lumière,this was helpful. BTW, your username reminds me of the fantabulous Fête des lumières. Cheers :) deeptrivia (talk) 01:57, 19 February 2006 (UTC)

Over-referencing

The number of references in articles which are making it through FAC has been rising, seemingly inexorably, for some time now. I often object to articles with more than about 40 references because I think it's almost always unnecessary to cite so many sources, and very distracting for readers to see so many footnotes in the text. In every case I object to, people are citing things which really don't need citing, like uncontroversial facts and things which are just common knowledge, and they commonly respond to my objections by saying they're only following guidelines here. So, I think it's time to include guidelines on when citations aren't necessary, and possibly a guideline on how many citations are likely to be appropriate for articles of a given length. What does anyone think? Worldtraveller 17:22, 21 February 2006 (UTC)

We've been over this in here before. What you think is common knowledge may not be known by someone else. Remember, Wikipedia is being written for everyone in the world (the English version for everyone in the world that can read English), and many potential readers do not share the "world knowledge" of North Americans, Britons or Australians. Take a look at something like Upanishad and see how much is "common knowledge" to you. -- Donald Albury (Dalbury)^(Talk) 00:08, 22 February 2006 (UTC)

Upanishad has 7 references so is not one I would complain is over-referenced. I think it gives me enough information to verify what's in the article, should I need to. Of course, what should be considered common knowledge is very subjective; I suppose what I am saying is that there are things which don't need citing because they are so easy to verify, such as 'the sky is blue' or 'the earth is round' at the extreme. Worldtraveller 00:44, 22 February 2006 (UTC)

I have an impression of "déjà vue". It seems that Wikipedia policy asks editors to provide references that other encyclopedia (with the same kind of audience) do not provide. IMO, if the request for references in Wikipedia was in all cases about the same as in other encyclopedia, there would be no impression that some Wikipedia articles, such as those seen by Worldtraveller, are over referenced. -Lumière 01:16, 22 February 2006 (UTC)

Wikipedia requires a lot of easily-verified references because it is not an academic paper or published encyclopedia. Articles here are rarely fact-checked and anyone can edit them at anytime. People get things wrong. I have fact-checked articles at WP:FAC and found many mistakes. It is the ones who do not provide online alternatives to the references that are the bigger problems. Those are the ones that do not get fact-checked (I did it once, found more than average errors, and it was not appreciated). Being liberal with the amount of references to related material may be annoying but providing the minimum amount of references required (especially on technical or obscure topics...like Upanishad) and making them 300-page books (not referenced to any particular page) not available at many libraries is a liability to Wikipedia's reputation. That is why I think it is best to err on the side of over-referencing than under-referencing. --maclean25 03:42, 22 February 2006 (UTC)

Maclean25, you took the words right out of mouth! My previous comment was just an invitation to a comment like yours. This was the next point I wanted to say, but I am glad you did. Now, maybe it is impossible to have the best of both worlds: (1) the type of fact-checking that we need amongst editors in Wikipedia together with (2) the type of not over-referenced articles that readers expect from an encyclopedia. -Lumière 04:03, 22 February 2006 (UTC)

I disagree that the implication that "heavy sourcing" is mainly just for editors. School students should be able to use Wikipedia to do research. At lower levels they're learning about what is generally considered "common knowledge" that "everybody knows". I don't want a kid to think just because they read something here, its true. Before relying on it (e.g. using in an essay for school), they should definately check the external source (which means we had better have cited them). This applies to the most utterly obvious and unconestable items. The problem we have is a lot of highly educated people work on articles here, and detest the notion of having to "prove" things "everybody knows". They don't realize Wikipedia is being used by people still getting their education, who can't tell the difference between an article written by a PhD and a fellow kid in their grade. --Rob 04:35, 22 February 2006 (UTC)

Rob mentions an important point in the first part about school students. In Wikipedia we might not want to make a distinction between a user that is a reader and a user that is an editor. Every user can become an editor and check for accuracy of simple facts, etc. We should not distinguish between two roles, reader and editor, as far as the use of references is concerned. Also, one key point to consider is that it must be clear to the readers that every fact in Wikipedia, unlike in other encyclopedia, should not be trusted and needs to be checked all the times. This is the purpose of over-referencing. -Lumière 10:03, 22 February 2006 (UTC)

The second part about the highly educated people is where the real problem is. It is not sufficient to request them to "prove" things "everybody knows" in their role of editors, but they should also see all these references when they are just readers so that they are invited to check them. This is because Wikipedia would not trust an editorial process that would separate the reader-role from the editor-role and provide an over-referencing that is only seen in the editor-role. The over-referencing must also be seen in the reader-role to invite all readers to participate in fact checking. -Lumière 11:50, 22 February 2006 (UTC)

Under-referencing: Painting the fence white

Part of the problem of getting people to provide adequate sources is that it's seen as a chore. If sourcing were to be promoted as part of the fun of editing, or even reading, I think people would be more inclined to do it. We can make sourcing out to be a kind of scavenger hunt, and a way to build up articles instead of ripping them up for lack of sources. A good example of how this can be done is the Al Lewis article, where the editors are gleefully trying to sort out Grandpa's tall tales. As for myself, I've been trying to find sources where needed to improve articles, instead of automatically dismissing things as bunk. (I don't always succeed, but I'm trying.) --Coyoty 20:34, 22 February 2006 (UTC)

Note that the problem that was originally mentioned is not that editors refuse to provide references, but only that too much of them were provided because the policy imposes it. So, yes it was confusing to say that the main problem is that some editors don't want to provide references. I think the problem was well captured by a few amongst us in a previous discussion: there is a need to formulate the policy so that not every thing needs to be sourced. As it was said previously, a line must be drawn somewhere and it is OK if it is ultimately decided amongst the editors when there is a dispute or when the question is raised. What is too strong is to say that, a source can be requested for every single statement in the article, irrespectively of any consensus. -Lumière 21:37, 22 February 2006 (UTC)

This is a different topic than oversourcing. This is a topic on undersourcing, where sources are needed. It is not supposed to be continuing the previous section's discussion. --Coyoty 21:52, 22 February 2006 (UTC)

Ok, so we can have a problem on both sides! -Lumière 22:29, 22 February 2006 (UTC)

Verifying assertions about the content of movies?

I think we need some kind of specific policy statement as to what constitutes verifiability for a statement about the contents of a movie, e.g. List of movie appearances of the Statue of Liberty. Personally I think a movie can be considered a source, but that a source citation should be required that references a specific edition of the movie (e.g. a particular DVD release) and the number of minutes into the movie.

Please discuss at Wikipedia talk:Citing sources#Movies, etc. Dpbsmith (talk) 15:49, 24 February 2006 (UTC)

Citing sources

How would I cite something I found in a book? Evan Robidoux 03:18, 26 February 2006 (UTC)

You can follow the examples at Wikipedia:Cite sources/example style#Books. Don't worry about getting it perfect, someone else will come along soon and copy edit any glitches. The main thing is to get all the pertinent information into the citation. Anyone can fix the format, but others might have problems finding the source information if you don't include it. -- Donald Albury (Dalbury)^(Talk) 03:49, 26 February 2006 (UTC)

Evan. I will respond with a quick "how to" guide, on your talk page. ≈ jossi ≈ t • @ 04:36, 26 February 2006 (UTC)

yes or no question

Is Wikipedia a valid resource for historical investigation?

Is it a good resource for historical inquiry? Yes, it's excellent resource as a starting point. SlimVirgin ^(talk) 10:11, 26 February 2006 (UTC)

I do not see how Wikipedia can be given to students as a verifiable source for information. How can Wikipedia justify itself as a formal source for historical reserach (even at the high school level)? It's fun but it is not valid. —Preceding unsigned comment added by Bobtine (talk • contribs)

Evaluate a Wikipedia article much as you would a book or other source. There are plenty of books that are neither fun nor "valid". Hyacinth 13:15, 26 February 2006 (UTC)

IMO, it is not a valid secondary source. It has not achieved this status. It is highly used, but in the sameway as the Internet (Google search, etc.) is also highly used -- people don't necessarily trust it. Studies indicate that Wikipedia seems to do fine as far as accuracy is concerned for standard encyclopeadic topics. However, a lot of material find its way in Wikipedia under POV pushing that would not normally be found in an encyclopedia, and to my knowledge the level of accuracy of Wikipedia is not known in these cases. A more serious problem is that the balance amongst views in a Wikipedia article can be biased by a small community of people that put a lot of attention on this article. Even if every statement in the article was true, the article can yet be completely biased. This issue is serious because it means that, even as a starting point, Wikipedia could mislead the readers with regard to what is important, etc. -Étincelle (formerly Lumiere) 13:26, 26 February 2006 (UTC)

It's true we're not a valid secondary source. But, we never try to be. Articles that cite good secondary sources, can be a good starting point in research. Wikipedia will never be an "end point" for research. At best, we're a good starting point. On POV, I disagree. We're probably better than most sources in terms of POV. Most publications have POV pushers. But, Wikipedia is one of the few that let everybody "push back". --Rob 07:27, 27 February 2006 (UTC)

Yes, as you say, everbody as the opportunity, but especially those with a specific agenda will do so intensively. Perhaps some see it as a good thing because then Wikipedia is a platform to allow these people to pass material or criticisms that would not pass in other credible third-party publications. Wikipedia can then become the voice of these people. In some cases, it is perhaps a good thing, but how can we evaluate that? It is a very difficult problem. Wikipedia uses WP:verifiability, WP:No original research and WP:Neutral point of view to address this problem (and also Wikipedia:What Wikipedia is not.) This global content policy is used to define the level of filtering and quality that Wikipedia attempts to achieve. I invite you to consider what is the basic approach of this policy. It basically tries to achieve its goal by only allowing in Wikipedia what was already published in standard reputable publications and only in the same proportion as in these publications. The details are more complicated than that because there are many different kind of contents, but this is the basic idea. The implication of this approach is that Wikipedia should not be a platform to allow editors to pass a view to counterbalance for what is not said enough in other reputable publications. In other words, what some perhaps think is a good thing is exactly what Wikipedia policy tries to avoid. My point is that Wikipedia did not really succeed to achieve its goal. The problem is partially due to the fact the policy is simply not applied. The people with a strong agenda ignore the policy or understand it in a way that allow them to go ahead with their agenda. The consequence is, as you say, 'Wikipedia is one of the few that let everybody "push back"' . So, what some consider a good thing is exactly what Wikipedia policy tries to avoid, but without success. Of course, Neutral point of view says that the article should fairly represent all significant viewpoints, in proportion to the prominence of each. However, this does not mean at all that we should let everybody "push back". There is also the opposite problem, which is that some articles do not receive the attention of all parties, and therefore is biased because some interested people have pushed their view, without the other views being fairly represented. In other words, Wikipedia let everybody "push back", but sometimes, even if it would be acceptable in accordance with policy, it simply does not happen enough. -Étincelle (formerly Lumiere) 14:08, 27 February 2006 (UTC)

To Thivierr on Citing sources

I'm not going to revert you for now, but I would note that this is policy and WP:CITE is a guideline, and as I understand it, guidelines should normally be changed to conform to policy, not the other way around. -- Donald Albury (Dalbury)^(Talk) 11:51, 26 February 2006 (UTC)

Bootstrapping

One thing I've noticed that should be opened for discussion: individuals who cite other Wikipedia articles as verifiable sources when they have, in fact, been primary editors on those articles. It allows for some rather unsubtle POV pushing. Wikipedia articles are themselves webpages, subject to the same limitation as as any other webpage. I suggest the following modification to the 'self-published sources' section:

Substantive points that make reference to other Wikipedia articles rely on the verifiability of those articles. In no case should a chain of evidence cross through more than one Wikipedia article before reaching a verifiable external source. Supplementary links (see also links, descriptive links, secondary topic links, etc.) may rely on the looser presumption that the linked page itself satisfies verifiability standards.

Ted 04:10, 27 February 2006 (UTC)

I'd go further and expand it to any collaboratively compiled document. We shouldn't limit it to merely wikipedia articles but any citing of articles on wikimedia based or styled projects. Hiding talk 14:22, 27 February 2006 (UTC)

Strongly Disagree with Jguk's version

I strongly disagree with the current formulation of the policy. It strongly fails to Assume good faith. In particular, these two are troubling:

2. Editors adding new material to an article should cite a reputable source, or it may be removed by any editor.
3. The obligation to provide a reputable source lies with the editors wishing to include the material, not on those seeking to remove it.

The policy should, instead, be that *all* editors have a resonsibility to cite a source (what's "reputable", really?) whether or not they were the one adding material. Removing valid information just because it wasn't sourced smacks of distrust of fellow editors and paranoia.

Further, those who feel that this policy is acceptable really should write a guide on how to use the ref tag / template, because that sure doesn't exist right now. If we are to demand sourcing, we damn well make it easier than it is now to do so.

For the record, let it be known that I applaud the efforts to improve the sourcing of Wikipedia articles. It's especially necessary in the unfortunate context of overly large agglomerated articles. --The Cunctator 07:08, 27 February 2006 (UTC)

On point 1, requiring sources has nothing to do with assuming or not assuming good faith. Editors can in good faith add wildly contradictory claims to the same article. Providing sources insures that the material is taken from a reliable source, and not dependent on often faulty memories. It also gives readers the opportunity to check the sources for themselves.

On points 2 and 3, it is much easier for the editor adding material to provide the sources, than it is for a reader who is not familiar with the subject to track down sources. Not providing sources when you add material is being rude to other editors and readers. -- Donald Albury (Dalbury)^(Talk) 13:14, 27 February 2006 (UTC)

Indeed, the policy as it stands precisely assumes good faith. If an editor adds something controversial (and let's be frank: it is only in controversial cases - where one editor challenges another - that this is even an issua), but in "good faith," then it would be easy for that editor to provide a citation. If he or she cannot, thehn - and only then - do we have evidence of bad faith. I agree that there are collective responsibilities. In fact, I added something to the Jesus article can though I was able to provide verifiable sources, several other editors started providing other good sources. To me this is collective editing at its best. Be that as it may, if someone adds something to an article, and one or more editors believes it is baseless, it is utterly reasonable for them to ask for a verifiable source. Asking is the sign of good faith, because the alternative is to revert the questionable edit. Let us remember that the main job here is to produce a quality encyclopedia. We must have standards. Slrubenstein | Talk 13:50, 27 February 2006 (UTC)

As for your last comment, let editors provide sources without being concerned about the proper formatting of the cite. The latter is the easier thing, really. Other editors familiar with the ref tags will surely come along and fix he formatting. ≈ jossi ≈ t • @ 15:56, 27 February 2006 (UTC)

Slrubenstein, I don't think this is quite correct. for instance, I've just started trying to mediate a dispute on an article about problems with the 2004 US presidential election: one problem is that almost all 'verifiable' sources are partisan, and so each side claims that the other side's sources are biased: who has bad faith? might be the person making the cite (if he knows the source is biased); might be the person disputing the cite (if he knows there are no objective referents and wants to push POV); might be neither; might be both. if we insisted on your approach, no one could publish anything on this page. as a temporary (and maybe permanent?) aid, I went and made an in-line template - looks like this: {{Disputed passage|nobody likes me, everybody hates me, I guess I better go eat worms.|Ted}}. it's a low profile way of calling attention to issues that might avert edit wars. if someone could drop a note on my talk page and tell me how to add this to the list of common templates, I'll do that. Ted

Ted: all sources are biased. As editors, we do not make judgments about the neutrality of a source, instead we present competing views fairly by describing them and providing sources that reflect these views. In the case of the 2004 US presidential election article, if there are two opposite views, describe them both, in relative proportion to their prominence. It works. ≈ jossi ≈ t • @ 16:01, 27 February 2006 (UTC)

As per your proposal, it is indeed a can of worms: all controversial articles will end up plastered with that "disputed passage" tag. That is why we have a discussion page, to engage in a civil discussion and hopefully rech consensus. Some editors believe that these few official policies, WP:NPOV, WP:V,WP:NOR and WP:NOT are there to limit us. IMO, there are there to protect the edting process, and eventually the possibility of success of this project. ≈ jossi ≈ t • @ 16:05, 27 February 2006 (UTC)

(after an edit conflict with Jossi, with whom I agree) Ted, with all due respect, I think the issue you raise is not one of verifiability but of NPOV. NPOV does not require that verifiable sources be NPOV, it requires our own articles to be NPOV. When all verifiable sources are partisan, the only solution - and a good one, that will improve the article - is to say that "Most people writing on this topic have taken one of five (or fifteen or fifty) distinct views" and then provide a non-judgemental description of each view, with the verifiable sources. Slrubenstein | Talk 16:14, 27 February 2006 (UTC)

• wikipedia:footnotes, guide on how to use the ref tag
• wikipedia:footnote3, guide on how to use the ref template --Francis Schonken 16:24, 27 February 2006 (UTC)

I would suggest you use only wikipedia:footnotes, as it makes it much easier when moving text around, keeping the numbering consistent. ≈ jossi ≈ t • @ 16:32, 27 February 2006 (UTC)

Jossi, Sl: I don't think we can make these distinctions quite this way. NPOV is an attitude that we hope editors will take towards their editing; verifiability is a method that we use to ensure NPOV when we feel that other editors don't have the right attitude. they are intertwined unto death. when editors are being responsible there won't be much of a problem any way you look at it - they will resolve things via discussion and show some care and concern for the appearance of the page. when editors get caught up in their own head-trips, well... would you rather have a page that is plastered with dispute tags, or one that undergoing constant edit wars? all I was thinking with the dispute tags is that they allow all sides to have their say on the main page, thus reducing the urge to chop and revert to keep some 'wrong' opinion from holding the high ground.

Jossi: thanks, by the way, for the advice. that's what I was thinking myself with this particular page (thought they'd already gotten tangled in the question of relative proportion and prominance before I arrived on the scene, and have started questioning each others' claims to prominence on verifiability grounds... can o' worms indeed...). I was just using it as an example of the kinds of problems that can arise. Ted 16:43, 27 February 2006 (UTC)

Ted, NPOV and Verifiability are two different policies and I advise you to keep them distinct. Hopefully, all our policies fit together and support one another. But they are different. Verifiability is not primarily a means to achieving NPOV (although it certainly can provide an avenue for resolving NPOV conflicts). If anything, it is a way of ensuring compliance with NOR. Slrubenstein | Talk 17:03, 27 February 2006 (UTC)

I am trying to keep them distinct, but I am trying to make sense of them at the same time. I mean, I could easily counter what you say by pointing out that the purpose of NOR is to prevent unreliable POV presentations. I don't think it's wise to consider each policy as a separate, stand-alone issue. we need these three poles (bad analogy alert!) to hold up the Wikipedia teepee, but that means we need to give some thought to the way the poles interact to make the teepee stand. Ted 17:35, 27 February 2006 (UTC)

We do need them all - no argument. But neither NOR nor Verifiability alone or even together ensure compliance with NPOV. Moreover, an article can violate NOR and still be NPOV. In the case you describe (if I understand you correctly), NOR/verifiability is simply not an issue: you imply that there are verifiable sources for the various claims. It is only NPOV that is an issue. And the way to handle this is to provide as many major (i.e. not held by two or three cranks) points of view in a non-judgemental way, and provide some context about each POV so people understand where the people who hold those views are coming from, and so people can assess the sources (is it a document published by a government? A corporation? A political party? A religious organization? A lobbying group or NGO? All of these are verifiable sources. If you chose only one to present in the article, or if you presented all of them and stated that one of them is right, another is wrong, then you would be violating NPOV. But if you present all of them without making any judgements, you are complying with NPOV. Slrubenstein | Talk 17:44, 27 February 2006 (UTC)

Clarification of "we need the three policies"

It may be a surprise to some, but I have seen editors interpreting the statement that the policies should not be used in isolation as if it meant that it is OK to violate one policy, say verifiability, if another policy, say NPOV, suggested that. An example will be that one side of the dispute has one or two reputable sources, but the other side only has a few personal websites, which are not valid sources in accordance with verifiability. In such a case, some editors can argue that, we should give priority to the NPOV policy and accept the websites because otherwise the article would entirely be the view of one side. When you see it that way, this argument to violate the verifiability policy starts to make sense. I still disagree though, but I mention this because I think it needs to be clarified. -Étincelle (formerly Lumiere) 18:22, 27 February 2006 (UTC)

Show me an example of that interpretation. Diffs, please. ≈ jossi ≈ t • @ 20:21, 27 February 2006 (UTC)

Please review [[4]] Hipocrite - «Talk» 21:54, 27 February 2006 (UTC)

a close, if not perfect, example here, from the page I was discussing earlier (***). as I understand it (haven't fully digested everything yet) the balance debate comes because one side has cites pointing to a large number of cases of Republican malfeasance, while the other side has cites pointing to ten or so cases of Democratic malfeasance. all cites are from politicized sources... the argument is over whether to simply mention Dem and Rep badness equally (without mention of differences in scale) or whether to minimize the Dem badness and focus on the Rep badness (because of differences in scale). I believe there are questions raised about the reliability of the anti-Dem sources as well, though I haven't finished reading yet. —Preceding unsigned comment added by Twrigley (talk • contribs)

Jossi, I will try to find where I saw such an interpretation. Meanwhile, can you clarify what is your position. -Étincelle (formerly Lumiere) 21:44, 27 February 2006 (UTC)

I would rather follow Hipocrite's suggestion. ≈ jossi ≈ t • @ 22:47, 27 February 2006 (UTC)

NPOV is not about an article saying (meaning, the editors of the article saying) that Democrats and Republicans are equally bad or equally good. It is about representing all views (meaning, views found in verifiable sources) in an accurate and non-judgemental way. This isn't "Crossfire." Slrubenstein | Talk 12:14, 28 February 2006 (UTC)

My goal here is to have a policy that cannot be misinterpreted, especially the sentences:

Jointly, these policies determine the type and quality of material that is acceptable in the main namespace. They should therefore not be interpreted in isolation from one other, and editors should try to familiarize themselves with all three. The three policies are non-negotiable and cannot be superseded by any other guidelines or by editors' consensus.

IMO, the expressions "Jointly" and "not be interpreted in isolation from one other" strongly suggest that even if a policy, say the verifiability policy, is quite clear that some material is not acceptable for inclusion, we can still refer to another policy, say the NPOV policy, to argue that the material should be included. This is not my understanding of the policies. My understanding is that the policies have rules to exclude material, and to be acceptable for inclusion a material must respect all these rules. Each rule can be understood separately. NPOV also have rules to determine the proportion of space attributed to each view that is acceptable for inclusion. None of these rules can be used to violate another rule. I have one example of an editor that is arguing with me that the policy criteria are not individual ground for exclusion. Originally this editor extended his argument to verifiability, which meant that verifiability could not be used as an individual ground for exclusion, but after my criticisms he limited his argument to notability and other criteria in NPOV. His argument (see this edit) is to ask why they should be individual ground for exclusion and not individual ground for inclusion -- IMO, a weak argument that has nothing to do with the actual policy. I have seen this used in practice, but it was long ago and I cannot find it anymore. -Étincelle (formerly Lumiere) 13:33, 28 February 2006 (UTC)

Again you're making a mountain out of a molehill. The statement is very clear as it is. Content has to meet all of the major content policies to be able to be included. That's what they "can't be interpreted in isolation of each other" means. Think about it, the only logical conclusion you can make from "can't be interpreted in isolation" is that content has to meet all three. You can't meet one and violate the other two and be included. - Taxman ^Talk 15:30, 28 February 2006 (UTC)

Taxman, please review [[5]]. Thanks. ≈ jossi ≈ t • @ 16:22, 28 February 2006 (UTC)

Ahh, now I get it. I had pretty much ignored the flood on this page before. But it seems silence doesn't work either. - Taxman ^Talk 22:39, 28 February 2006 (UTC)

Thank you Taxman! I agree with your interpretation. This is the most important. However, I am not sure that it is the only possible interpretation, and this is the problem. Really, the interpretation depends upon the context. To provide an appropriate context we should add a sentence like "These policies provide criteria that are individually valid ground for exclusion." Otherwise, the phrase "can't be interpreted in isolation of each other" could mean that the criteria are not individually valid ground for exclusion and therefore must be considered together, which is the opposite of what we want to say. At the best, without the context, this phrase is useless, except to add an uneccessary complication in a sentence that only says that the three policies are important. -Étincelle (formerly Lumiere) 18:08, 28 February 2006 (UTC)

NY Times Example

I'm very uncomfortable with the NY Times example in the current article, which reads:

"Verifiability" in this context does not mean that editors are expected to verify whether, for example, the contents of a New York Times article are true. In fact, editors are strongly discouraged from conducting this kind of research, because original research may not be published in Wikipedia. Articles should contain only material that has been published by reliable sources, regardless of whether individual editors view that material as true or false. As counter-intuitive as it may seem, the threshold for inclusion in Wikipedia is verifiability, not truth.

This example seems to blur the distinction between original research and checking primary sources. For example, if the NY Times published a demonstrably false article, and the falseness could be trivially proven by linking to primary sources, it would be perfectly acceptable to link to those sources in order to disprove the contents of the article. It only becomes original research when new primary sources are being created; a distinction that the above fails to make. - O^O

The distinction between primary and secondary sources is important in the no original research policy, but only to better explain what kind of material cannot be used to source an article. The no original research policy says that an article must be entirely supported by secondary sources:

"That is, we report what other reliable secondary sources have published, whether or not we regard the material as accurate."

The emphasis is mine. This is not to say that primary sources cannot be cited if they provide useful information, but the material in the WP article must be separately sourced in reliable secondary sources.

My understanding of policy is that checking primary sources to obtain a conclusion to contradict a secondary source (or for any other purpose) is original research. Any material based on an analytical or evaluative process must be sourced in a reputable secondary source before it can be included in Wikipedia. Editors cannot just include such a material in Wikipedia whether it comes directly from them or from a friend or through an interview with an expert, unless it is also sourced in a reputable publication. Therefore, editors cannot argue inside the WP article against the material that is sourced in the NY times, unless their argument is also sourced in a reputable secondary source. However, if the verifiability policy had a rule such as:

"If something is not true as far as the editors have ascertained to the best of their endeavours, then they should not include it, even if it complies with all other requirements of WP policy."

then at the least the editors would be justify not to include the NY times material. However, there is no such a rule in the policy, and I do not understand why. See http://en.wikipedia.org/wiki/Wikipedia_talk:Verifiability/truth for more details, and also previous discussions in this talk page. The current policy says that the NY times material must be included (if it is relevant and respect all other policies) even if it is known to be false, and you are not even allowed to include your personal criticisms of this material because it would be original research. The only thing you can do is to clearly attribute it to the NY times so that the readers know from where it comes from. IMO, this is a ridiculous situation. -Étincelle (formerly Lumiere) 10:01, 1 March 2006 (UTC)

Actually, WP:NOR (just before your quote) says that part of an article, and in some cases all of an article, may be based on primary sources. This is very badly written. Editors should not write false information, under the excuse a secondary source said so. Requiring verifiability, means we wish things to be true *and* that truth to be verifiable. It doesn't mean we can publish falsehoods. We can't write about our personal criticism of a NY Times piece. But, if somethings known to be false, it probably shouldn't be written about. --Rob 17:17, 1 March 2006 (UTC)

I would say, if two reliable sources disagree about the facts, cite both, but do not try to determine which is true. Leave that to the reader. Trying to determine "truth" gets us into POV territory. We don't analyze the evidence, we just try to present a balanced account of what is offered by reliable sources. There are plenty of other people out there eager to prove the 'falsity' of stories in the New York Times. If they 'discover' proof that an article is false, AND, that story gets picked up by other reliable surces, then we cite those sources as well as the original. -- Donald Albury (Dalbury)^(Talk) 17:39, 1 March 2006 (UTC)

I think the assertion that we have to use only secondary sources is completely wrong. The WP:NOR says we are strongly encouraged to collect and organize primary sources. I do completely agree that the NPOV way of handling this situation would be to cite the NYT, and then cite the primary source that conflicts with the NYT. However, we do not need to wait for some other secondary source to link to the primary source, we can do so ourselves. I'm going to make some clarifying edits to try to clear this up. - O^O 18:47, 1 March 2006 (UTC)

Hey, wait a minute O^O. To claim that an important part of the policy is completely wrong, you refer to:

"In some cases, where an article (1) makes descriptive claims the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge, and (2) makes no analytic, synthetic, interpretive, or evaluative claims, a Wikipedia article may be based entirely on primary sources (examples would include apple pie or current events), but these are exceptions."

Fine, because of these special exceptions, it becomes a question of degree whether or not some material requires a secondary source. However, the examples that are given try to convey the idea that these exceptions are for non controversial materials. There is a big difference between "some exceptions" and "completely wrong". Moreover, these exceptions do not apply to a case where you contradict the NY Times. It is difficult to contradict the NY Times and argue that it is a [well known] current event or an apple pie example. I don't think you have a case to change the policy here.

In this second part of your comment, you consider the other option: if we cannot contradict it, let us not write about it. To support this second option, you say that the policy requests verifiability and truth. You do not read the same policy as me. What I read is that it requests verifiability, not truth. My point is exactly that the policy should request verifiability and truth, as you seem to agree. User:Light current and myself, mainly Light current, tried a lot to change the wording so that it reads "verifiablity and truth". As ridiculous as it may seem, the policy says that truth is not a consideration and that we should include what is well sourced, "whether or not we regard the material as accurate."

Perhaps your argument is that in practice it will not happen that editors will include something that they believe is not true. Why would they do so, you may think. This is a naïve evaluation of the situation that assumes that decisions are taken in a simple manner by editors that are all thinking along in a coherent way. The reality is that most decisions require that we obtain a consensus out of opposing opinions. For example, one editor might be a proponent of a false statement. All editors, except perhaps this one, know that the statement is false. However, all editors agree that the statement is verifiable. Because, this single editor requests that we include the statement, it will have to be included according to policy. You might say that the other editors will argue with the dissensious editor. This single editor could immediatly put an end to this discussion and say "the threshold for inclusion is verifiability, not truth. The problem is that the policy does not even give room for a discussion, like the one that you suggested. -Étincelle (formerly Lumiere) 19:13, 1 March 2006 (UTC)

Regarding your latter question; why would an editor cite something that is verifiable but "wrong" - it happens all the time when a good editor tries to be NPOV on a controversial subject. If a pro-choice editor writes "the X church holds that abortion is murder", then that editor is citing a position that is verifiable, whether or not it is "true" that abortion is murder. Moving back to primary references; primary references from reputable sources have always been acceptable on Wikipedia. I would argue that in many cases it is better to reference the primary source than a secondary source. If you want to quote a sentence from a speech, better to cite a transcript of the speech, instead of an article in a newspaper that happens to include the speech. - O^O 19:28, 1 March 2006 (UTC)

No offense O^O, but you do not seem to be aware of the many discussions that we had in this talk page. Whether, a primary source such as an interview or a speech in some media can be used as a source was debated before. The argument against the use of such a primary source is that its content was not filtered by a review process that includes fact checking, etc. by a team that might include lawyers, etc. You might have in mind innocent situations, when you propose that we accept primary sources to support material for inclusion, but the policy exists for the situations that are problematic. I am personnaly concerned by contents that are used to discredit some individuals or their work. For example, the content of an interview can be used either to discredit the interviewee himself or some person that is mentioned in the interview. If you use the content of a reputable secondary source, which refer to this primary source, then you have some guarantee that the way the primary source was used has been checked for accuracy and fairness. It is very easy to misinterpret the content of primary source by using it out of context. This is not to say that the primary source cannot be used, but the significant content of an article (except in special exceptions) [including how to refer to primary sources] should be found in reputable secondary sources. This what the policy says. I do not invent anything here. -Étincelle (formerly Lumiere) 19:46, 1 March 2006 (UTC)

Clarifying what is a primary source

This following sentence was just added in the no original research policy based on a discussion on the Verifiability talk page!

"As an example, results of research published in the Journal_of_the_American_Medical_Association may be a primary source, while an article in the New York Times that summarizes this research would be a secondary source."

This sentence proves that we do not understand what is a primary source. Research published in the Journal_of_the_American_Medical_Association is typically used as a secondary source. The above is an example of a secondary source that refer to another secondary source. It is true that the Journal_of_the_American_Medical_Association may in some special cases be used as a primary source. For example, if we make a comparative study of the kind of research that is published in different journals, then every journal in that research becomes a primary source. The point is that the journals become the primary ingredients in that research. In such a comparative study, the validity of the research published in the journals might not be an issue at all. However, when a journal is used as a secondary source, then we use it because it is a reputable source that provides some guarantee with regard to its accuracy, fairness, etc. This important difference between a primary source and a secondary source explains why sources such as dubious websites can only be used as primary sources, never as secondary sources. -Étincelle (formerly Lumiere) 19:29, 1 March 2006 (UTC)

I'm sorry, but I'm struggling to understand the exact distinction you are making. I will try harder. When research is published in a journal, and is published no other place, then that journal is the primary source. If you don't think the JAMA example is appropriate, then perhaps we should use another journal as an example. With regards to your "dubious website" example, I think we already stress that the quality of the source matters. But if we were discussing (for example) the safety features of a new car, we would not have to wait for some car magazine to write about it, if we could simply link to the website of the car maker. I need to cut my end of this talk short for now, but I will check back in later. - O^O 19:39, 1 March 2006 (UTC)

Please review [[6]] Hipocrite - «Talk» 19:57, 1 March 2006 (UTC)

??? It seems to me that most scientific journal papers are mixtures of primary and secondary material. That is, every statement made in a paper is either original research, based on observations and logical reasoning personally made by the authors—and hence unreferenced—or is referenced to someone else's work.

However, except for "review articles," the whole point of a scientific paper is the portions which are the original research—for which it is a primary source.

So, when a scientific paper is cited, it would almost always be functioning as a primary source. If you wanted to refer to the material it references, you wouldn't use that paper, you'd use the papers it cites. The only exception I can think of would be if the papers it cites were so rare or obscure that it would be impractical to obtain them. Dpbsmith (talk) 20:06, 1 March 2006 (UTC)

O^O, you do not give your definitin of primary source, but only an example. Therefore, I cannot be sure what is your definition. However, it does not seem to be the correct definition. My guess is that you say that only sources, such as the NY Times, that report on already published research are secondary sources. According to this definition, all technical research papers that report new research would be primary sources. This is certainly not true. If it was true, you are right that it would be ridiculous to require that only secondary sources can be used to support material for inclusion. Of course, technical papers which are published in reputable scientific journals are normally secondary sources. These technical papers report on research that collect, analyse, etc. primary and secondary source materials, and therefore respect the definition for a secondary source. Perhaps you should read the definition of primary and secondary sources that are given when we follow their respective wikilink in WP:NOR. These definitions are given in the context of the work of historians, but they provide a very good idea about how the concept of primary/secondary source is used in Wikipedia. -Étincelle (formerly Lumiere) 20:21, 1 March 2006 (UTC)

Dpbsmith, I think that my reply to O^O apply to you as well to some degree. You say that in Wikipedia we use scientific papers, except review papers, as primary sources. This cannot be true. It would be in complete contradiction with the requirement that we should essentially only report what is published in secondary sources. What is this? Of course, the policy allows us to report on the content of scientific papers that are published in reputable journals, including what is original in these papers. The No original research policy is fine (it was fine even before O^O modified it). It is your interpretation of it that you have here that is incorrect. -Étincelle (formerly Lumiere) 20:21, 1 March 2006 (UTC)

Perhaps my example of a possible use of scientific papers as primary sources was confusing. This example was intended to illustrate a very special case. To simplify the situation, as a first approximation, you could forget about it. The normal use of a scientific paper that reports on a research is as a primary secondary source, not as a primary source. Let me give another example of a special (not usual at all) use of a scientific paper as a primary source, which I hope will clarify the situation. Take a paper that was published 200 years ago on geometry. An historian uses that paper to evaluate who were the scientifics at the time, how they communicate, etc. The scientific value of the paper is irrelevant. This is an example of a scientific paper used as a primary source. The key point is that we are not interested at all in the scientific content of this paper except as a testimony of how the scientifics communicated at the time, etc. In the sameway, an interview can be used as a primary source, and is usually used as a primary source. In principle it could also be used as a secondary source, but not in accordance with Wikipedia policy. -Étincelle (formerly Lumiere) 20:37, 1 March 2006 (UTC)

No. Wikipedia is a secondary source (or a ternary, quaternary, etc. source). Wikipedia cites other sources. Those other sources need only be reputable; they certainly can be primary sources, secondary sources, whatever. The only restriction on the kind of source Wikipedia cites is that it be reputable. Dpbsmith (talk) 21:28, 1 March 2006 (UTC)

I was just referring to the policy before O^O modified it. The previous version made perfect sense to me, and it was the policy. Now, because you insisted, I made some research to see what was the policy before. I found this edit, which I suspect was not really discussed in the talk page. However, I still maintain that my understanding of what is a primary source is correct, and scientific papers are not normally used as primary sources. The distinction is very important because materials that qualify as primary sources do not necessarily qualify as secondary sources. (The converse is perhaps also true, but is less problematic.) The basic point that I made remain valid. The content of a WP article that can be controversial is essentially the content that requires a secondary source. The logic is that if something is controversial it is because there is some evaluative claim or some analysis in it, and therefore it requires a secondary source. Therefore, on the essential, what I explained remain valid. -Étincelle (formerly Lumiere) 22:14, 1 March 2006 (UTC)

This was also inserted at WP:NOR, with absolutely no discussion over there and reference to some discussion here which looks, frankly, rather thin. · Katefan0^(scribble)/poll 22:29, 1 March 2006 (UTC)

Hello Étincelle; since we seem to be both struggling to completely understand each other, I will try to write as directly as possible without the use of examples or analogies. Here is my position:

1. The wikipedia definitions of primary source and secondary source are the ones I am using. Another good definition is at this web site.
2. A given source can sometimes be used as either a primary or secondary source. The distinction comes from how the source is used, not from the source itself.
3. Wikipedia has always allowed references to reputable primary sources. There has never been a requirement that we use exclusively secondary sources.

I think, but am not completely certain, that you disagree with my final point. Is this correct? - O^O

I cannot say anything about the first point because your definition of primary and secondary sources may have changed as a result of our discussions. I can only say that I use the definitions that you cite and that the view that you expressed that scientific papers are normally used as primary sources is not consistent with these definitions. Scientific papers are not direct accounts of events, etc. I agree with the second point, which is clear from my previous comments. The third point can be interpreted in a way that is consistent with my understanding of the policy, therefore it is not a point of disagreement, even though it can also be interpreted in a way that is opposed to my understanding. Here are the points that you made to which I disagree:

1. Scientific papers are normally used as primary sources.
2. The Wikipedia policy that we had for several months (before you modified it) is completely wrong when it says that an article should report what is published in reputable secondary sources.

The first point means that you did not have the correct definition of primary sources, which made it impossible for you to appreciate the policy as it was before you modified it. The second point shows that indeed you did not appreciate this policy. This policy was perfectly fine to say that articles should usually report what is published in reputable secondary sources, the exceptions being articles such as apple pie and current events. This is not in contradiction with the final point in your list because, while reporting what is published in secondary sources, it is very natural to cite primary sources. The subtle point is that the support for the analytical, evaluative, etc. content of the article, which is its essential content, are the secondary sources, not the primary sources. Therefore, the sentence that you did not appreciate actually made perfect sense. The fact that you said that it is completely wrong together with the fact that you said that scientific papers are usually used as primary sources are two clear indications that you did not appreciate a policy that was actually perfectly fine. I would prefer that we keep this previous version because, when properly interpreted, it conveys an important understanding that is missing in your version. -Étincelle (formerly Lumiere) 02:35, 2 March 2006 (UTC)

Hello again and thank you for your response. I think that by changing my style of writing to be very direct, we are more quickly understanding exactly where we disagree. First, I will comment on the definition of "Primary Source".

1a: I maintain that scientific papers are often appropriate for use as primary sources.

1b: The wikipedia definition Primary source only directly addresses history. However, it references two sites [7] & [8] that give examples of primary sources in the sciences. These examples include "research reports" and "Journal articles".

Second, I will comment on the WP:NOR policy.

2a: The WP:NOR policy has always allowed us to site both primary and secondary sources.

2b: I said it was "completely wrong" to interpret this policy as requiring only secondary sources.

2c: I did not say it was "completely wrong" to cite "secondary sources" at all.

In particular, thank you for pointing out this edit. It appears that this is where confusing language was inserted into the article. However, as the article has always made clear, editors are encouraged to use both primary and secondary sources. - O^O 05:09, 2 March 2006 (UTC)

It seems that we do not really progress. You still found a way to present your story in a way that is not in direct opposition to my points, but yet avoid the main issues. Therefore, instead of saying what you maintain, could you address what I maintain.

1. I maintain that the normal intended use of scientific paper is as a secondary source, not as a primary source. However, it is possible to use a scientific paper as a primary source. For example. if you use a scientific paper about a physics discovery in the area of sociology or history, then it is usually a primary source. The idea is that this paper is a testimony of a scientific event. I only maintain that whenever a scientific paper is used by another scientific paper in the normal way, in the way it was intended to be used, for its scientific content, it is used as a secondary source. This obviously remains true even if the scientific paper is the first paper to report a scientific research. The only difference is that such a paper has an additional value as a testimony of a scientific event, which makes it also a candidate as a primary source.

2. I maintain that the following sentence is perfectly fine:

"That is, we report what other reliable secondary sources have published, whether or not we regard the material as accurate."

The emphasis is mine. I maintain that this sentence, especially the fact that it only refers to secondary sources, convey an important understanding. Therefore, we should keep it. It is certainly not completely wrong. If you meant something else, we are not in disagreement. Yes, we can cite both primary sources and secondary sources. Again, as I explained in my previous postings, this is not in contradiction with the above sentence. -Étincelle (formerly Lumiere) 15:30, 2 March 2006 (UTC)

Guilt by association

I am currently in a dispute that is undergoing mediation. The process has bogged down over the interpretation of criteria. Specifically, [WP:BLP] states Negative information related to a person's notability should be mentioned if solidly verifiable, e.g. plagiarism by an artist, fraud by a scientist, doping use by a sports person, etc. Remember that verifiability requires direct evidence from reliable sources regarding the subject of the article specifically. Beware of claims that rely on guilt by association, or other generalizations. To me this says that you cannot place information in an article about a living person if it infers guilt by association, ie. if it links two people so that the reader may be lead to assume that information that he has about one person will apply to the other. My opponent won't acknowledge this.

There used to be a section on guilt by association in the Wikipedia:Verifiabilityguideline, but it was removed a month ago. I cannot find a reason for its removal. You can read it here [9]. Was there a reason for this removal? Has guilt by association become acceptabe.

If you feel yourself knowledgeable about the guidelines and want to put your two cents in I would appreciate it. The article is here [10] and the mediation subpage here [11]. Engjs 13:17, 10 March 2006 (UTC)

This policy was refactored to shorten it, and to eliminate overlap with other policies and guidelines. It is my opinion that there was no consensus to change what is allowed or not allowed in Wikipedia. -- Donald Albury (Dalbury)^(Talk) 16:18, 10 March 2006 (UTC)

I also wanted to use this section once, and was disapointed that I could not find it anymore. I could not find the equivalent elsewhere in the policy. Therefore, IMO, removing it was actually a change in the policy. -Étincelle (formerly Lumiere) 16:37, 10 March 2006 (UTC)