Wikipedia

Wikipedia:Verifiability/First sentence

< Wikipedia:Verifiability
(Redirected from Wikipedia:V/First sentence)

Earlier polls can be found at Wikipedia:Verifiability/First sentence/Archive 1.

Poll V_FC_P_11 Essay on Verifiability, Not Truth

edit
Closed, open to see valuable polling results
The following discussion has been closed. Please do not modify it.

If the phrase "verifiability, not truth" is to be kept in the main body of the policy, the essay WP:Verifiability, not truth should be added to the "See also" list.  If the phrase "verifiability, not truth" is moved to a footnote, the footnote should reference this essay.

Added link to poll per consensus here and at main WP:V page. -- Avanu (talk) 03:51, 11 September 2011 (UTC)[reply]

Poll V_FC_P_13 Blueboar's compromise - move discussion of truth/untruth out of lede and into new section

edit
Closed, open to see valuable polling results, moved to following sections
The following discussion has been closed. Please do not modify it.

Change the lede paragraph to:

The initial threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source. Note, however, that while verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article.)

Add a new section (right after the lede) to deal with the issue of truth/untruth... as follows:

==Assertions of truth and untruth==

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth"'.

Assertions of untruth (ie an editor's assertion that some bit of information is untrue) are a more complicated issue. If the dubious information is not supported by a source, it should be challenged; but the question of how to challenge (whether to tag the information as needing a citation or to remove it immediately) depends on the nature of the information (see: WP:Burden, below). If the dubious information is supported by a reliable source, the problem should be discussed on the article talk page, with reference to policy concepts such as maintaining a neutral point of view (and especially the sub-concept of due weight). In many situations, a simple rewording to present the information as an opinion rather than as an accepted fact can resolve issues of verifiable but potentially untrue information.

Question... is this an objection to the wording or to the concept? Blueboar (talk) 13:49, 3 September 2011 (UTC)[reply]
  • It's OK. No objections. --SmokeyJoe (talk) 13:45, 3 September 2011 (UTC)[reply]
  • Support  Balances many tradeoffs, not all of which I agree with, but overall a huge step forward.  Unscintillating (talk) 14:24, 3 September 2011 (UTC)[reply]
  • Would generally support if... the section were titled "Verifiability, not truth" (it's the phrase lots of editors use), and that paragraph expanded a little to help people understand and to encourage new editors not to be disheartened if they've been pointed here from a content dispute. After all, we need to address concerns that new users may get discouraged. eg For example, living in a town does not qualify your word as a good source for information that town because our readers would have no way of checking your information is correct. If you cannot find the information you want to include in a reliable source, then you cannot include it, no matter how accurate or true you personally know it to be, or how honest and honourable your intentions. It is also important not to be offended if people remove your material if you have not sourced it; they are simply trying to ensure the accuracy of the encyclopedia. The best response is to find good sourcing. The last paragraph on untrue material is problematic (too much detail on various policies) and needs re-writing if not abandoning.VsevolodKrolikov (talk) 14:26, 3 September 2011 (UTC)[reply]
  • Support. Good solution for the problem that doesn't seem to open any new ones. Hans Adler 15:22, 3 September 2011 (UTC)[reply]
  • Enthusiastic Support. Overall, this is by far the best idea to have come out of the discussions of the past few months. At the level of tweaking the details, I'm receptive to most of the points raised by VsevolodKrolikov. I would also suggest pruning the second sentence of the lead of the words "Note, however, that". Thus, the sentence would start with "While...". --Tryptofish (talk) 16:41, 3 September 2011 (UTC)[reply]
  • Oppose - There is a significant problem with this proposal that people may not be aware of: it's too much a change to be successful. Please see the discussion at Wikipedia talk:Verifiability/Draft. When you go there you will see that I wholeheartedly agree with Blueboar's first message there and that this proposal is inconsistent with the good points made in that message by Blueboar. Please read that message of Blueboar's and my message that follows it which shows how to address the issue of the phrase "verifiability, not truth" in the first sentence, which is the main issue. Adding other issues and a new section now is a recipe for failure. Please save them for later, once the main issue is settled. --Bob K31416 (talk) 17:14, 3 September 2011 (UTC)[reply]
I followed Bob's link, and read Bob's comments there. Honestly, I just don't get them. I don't see the problem. --Tryptofish (talk) 17:22, 3 September 2011 (UTC)[reply]
I think Bob means "The more changes we make (even if we think the change is relatively small), the less likely it will be that this will achieve consensus." -- Avanu (talk) 20:53, 3 September 2011 (UTC)[reply]
Please don't take my words out of context... My comment was specifically directed towards suggested changes that had nothing to do with resolving the "not truth" issue (such as changing "threshold" to "criteria"). It was a plea to deal with one issue at a time. Blueboar (talk) 12:00, 4 September 2011 (UTC)[reply]
Regarding the change from "threshold" to "requirement", I could agree, but worry that it could be a distraction from the overall change, see Threshold again and especially Threshold vs. requirement.  Unscintillating (talk) 15:10, 5 September 2011 (UTC)[reply]

Work area for development of main proposal

edit

Developing this per talk. North8000 (talk) 23:30, 9 September 2011 (UTC)[reply]

Record of Blueboar's proposal ca. Sept 9th

edit

Change the opening paragraph to:

The initial threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source. Note, however, that while verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article.)

Add a new section (as the first section after the lede, after the index box) to deal with the issue of truth/untruth... as follows:

==Assertions of truth and untruth==

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth"'.

Assertions of untruth (ie an editor's assertion that some bit of information is untrue) are a more complicated issue. If the dubious information is not supported by a source, it should be challenged; but the question of how to challenge (whether to tag the information as needing a citation or to remove it immediately) depends on the nature of the information (see: WP:Burden, below). If the dubious information is supported by a reliable source, the problem should be discussed on the article talk page, with reference to policy concepts such as maintaining a neutral point of view (and especially the sub-concept of due weight). In many situations, a simple rewording to present the information as an opinion rather than as an accepted fact can resolve issues of verifiable but potentially untrue information.

It received overall strong support.

Proposal, as evolved from Blueboar's proposal

edit

The next two areas are for development of the final proposal from Blueboar's proposal. Blueboar's proposal received strong support from the the respondents, in the context of which it was proposed. This can conservatively be interpreted as support of the general approach in the highly debated areas. Proposals and discussions regarding potential changes should be crafted and debated with recognition of what has already transpired. Following is the proposal ca. 9/9/11 with any subsequent consensused modifications:

Change the opening paragraph to:

The initial threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source. Note, however, that while verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article.)

Add a new section (as the first section after the lede, after the index box) to deal with the issue of truth/untruth... as follows:

==Assertions of truth and untruth==

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth"'.

Assertions of untruth (ie an editor's assertion that some bit of information is untrue) are a more complicated issue. If the dubious information is not supported by a source, it should be challenged; but the question of how to challenge (whether to tag the information as needing a citation or to remove it immediately) depends on the nature of the information (see: WP:Burden, below). If the dubious information is supported by a reliable source, the problem should be discussed on the article talk page, with reference to policy concepts such as maintaining a neutral point of view (and especially the sub-concept of due weight). In many situations, a simple rewording to present the information as an opinion rather than as an accepted fact can resolve issues of verifiable but potentially untrue information.


Proposed changes to and discussions about main proposal

edit

OK, now it's time for folks to make finalize, specific proposed changes to Blueboar's proposal, and to see which if any have a consensus. Moved previous discussion to discussion page North8000 (talk) 15:44, 20 September 2011 (UTC)[reply]

Updated version

edit

Here is my attempt to synthesize what discussion has pointed to so far. Your mileage may differ. All in the lead, no separate section of the policy page at this time:

The threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source. While verifiability is a requirement for inclusion, it is not a sufficient reason for inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article).

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth". For guidance on the inclusion or exclusion of verifiable but potentially untrue material, see WP:NPOV (and especially WP:UNDUE).

--Tryptofish (talk) 00:11, 14 September 2011 (UTC)[reply]

It seems to me to be progress again. Good. Could it be better? Comments seem to be required upon the following decisions you've implied:-
  • First, you are one of the people saying we should NOT try to change "The threshold...", to "A fundamental requirement...". I am not sure there is consensus on this. Personally I would prefer to try to make the best version possible.
  • You have proposed above that instead of Blueboar's whole new section, you have just one new paragraph. This means two major changes to Blueboar's proposal: (1) no section break and (2) a paragraph gets deleted. I think no one replied to your proposal above, so it is not clear if there is consensus about it. I personally think it is OK. The deleted paragraph was this one: --Andrew Lancaster (talk) 08:33, 14 September 2011 (UTC)[reply]
The paragraph to which Andrew Lancaster refers
The following discussion has been closed. Please do not modify it.
Assertions of untruth (ie an editor's assertion that some bit of information is untrue) are a more complicated issue. If the dubious information is not supported by a source, it should be challenged; but the question of how to challenge (whether to tag the information as needing a citation or to remove it immediately) depends on the nature of the information (see: WP:Burden, below). If the dubious information is supported by a reliable source, the problem should be discussed on the article talk page, with reference to policy concepts such as maintaining a neutral point of view (and especially the sub-concept of due weight). In many situations, a simple rewording to present the information as an opinion rather than as an accepted fact can resolve issues of verifiable but potentially untrue information.
I think you make valid points, and they are, indeed, the kinds of things we should discuss. The reason I collapsed the paragraph is that I don't want people to get confused about what is and isn't proposed here, and it's still easy to click to see it. Where you say no one replied above, I want to point out the discussion on the talk page, where there actually seems to be no substantive objection. About the opening wording (threshold etc.), I based it on what is on the page now (see the discussion below where S Marshall and I agree), but I did so recognizing that it would still be discussed. --Tryptofish (talk) 17:30, 14 September 2011 (UTC)[reply]
My infinitely preferred version (which I think we had pretty much agreement on) is this one:
A fundamental requirement for inclusion in Wikipedia is verifiability— whether readers can check that material in Wikipedia has already been published by a reliable source. While verifiability is a requirement for inclusion, it is not a sufficient reason for inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article).

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth". For guidance on the inclusion or exclusion of verifiable but potentially untrue material, see WP:NPOV (and especially WP:UNDUE).

Howzat? (Yup I added the italics as well, but if anyone had any major objection to the italics it wouldn't break my heart to see them go.) Pesky (talk …stalk!) 08:36, 15 September 2011 (UTC)[reply]

Please correct me if I'm missing something, but it looks to me like you changed 2 things: (1) the opening words, and (2) the italics. Is there anything else? I'm fine with the italics. I'd even be fine with bolding, as is done in the next paragraph. As for "The threshold" versus "A fundamental requirement", I'm personally pretty neutral on the merits, meaning that I just don't think it matters either way insofar as the usefulness as a policy. However, as I have been saying before, I advise against changing wording in the first sentence beyond what seems really necessary at this step. With each additional change from what the policy says now, it's just one more reason for the community to say no to the whole thing. If it's important enough to you to push for "A fundamental requirement" that you would be OK with the community rejecting everything and leaving the page unchanged, then go for it. But if you think that "not truth" is the important issue to address, then you should consider letting it go. --Tryptofish (talk) 20:43, 15 September 2011 (UTC)[reply]
If you want either of those versions to be accepted, you have to get SlimVirgin's support. She can muster all the opposition she needs to defeat it, if she wants to. See my previous message too. --Bob K31416 (talk) 13:05, 16 September 2011 (UTC)[reply]
We can't accept ownership, and slim being at the at the extreme end of the status quo / change needed spectrum would certainly affect chances for her support of a true compromise version. North8000 (talk) 20:36, 20 September 2011 (UTC)[reply]
I'm not convinced that she's at the extreme at all. Many of the editors who are most eager for change are probably more at the margins of what the larger community thinks. But I personally don't think it's a matter of needing any one editor's blessing (including mine, of course!). --Tryptofish (talk) 21:11, 20 September 2011 (UTC)[reply]

As of this time, I think that something along the lines of what Pesky and I are suggesting should be the way to go, especially from the perspective of getting the "opposition" to consider it. --Tryptofish (talk) 20:04, 20 September 2011 (UTC)[reply]

I actually agree with you that it's more of a compromise. That's why I think it's a good idea. Of course, there is an alternative: insist on less compromise, and get nothing at all. A narrow majority of those who responded to the last RfC were perfectly happy to keep the status quo. And I'd be perfectly happy to keep it too. --Tryptofish (talk) 21:07, 20 September 2011 (UTC)[reply]
My quick thumbnail is that in the biggest RFC half the folks wanted the most "extreme" change...taking "not truth" completely out. The other half would be a mix of "no change" folks and "lesser change" folks. I think that that very roughly indicates that something half way between the "extremes" (= Blueboar's proposal) is probably too conservative, if anything, yet the premise of your proposal is that we need to go farther towards the "very little change" end. Sincerely, North8000 (talk) 21:24, 20 September 2011 (UTC)[reply]
Can we propose two options? I think it might be important because we have two ways we can work: one is artificially compromising with imaginary inter-locutors who are not really explaining their position, and the other is just to go for quality as best as we can judge it and hope that when people see the result they will see the improvement too. It is very hard to choose between these two options and I think this is partly what has slowed down discussion here.--Andrew Lancaster (talk) 10:59, 23 September 2011 (UTC)[reply]
Andrew, that's a good question. I suspect that if the small group of editors who are still participating here were to !vote on that, the longer version would be chosen, whereas if instead the much larger population of editors who will eventually respond to the RfC were to decide, the shorter version I proposed here would have a lot more traction. --Tryptofish (talk) 18:24, 23 September 2011 (UTC)[reply]
I think we do not know. And that is a weakness of what we are doing, given that we expect that this will become a wide RFC. It is not just concerning the proposed compression, but also the idea of also changing the first few words. In my opinion, they are part of the problem which make "not truth" an issue to start with. Anyway, if there will be a vote for two proposals for change versus one proposal for no change, the vote for change will be split and won't pass. So unless we develop a system of preference voting, which no one will agree upon to begin with, we would perhaps need to make a proposal for change which gives two examples, but no final version?--Andrew Lancaster (talk) 19:50, 24 September 2011 (UTC)[reply]
From what I've seen of the discussion between Blueboar and SlimVirgin, the eventual RfC absolutely must be a binary choice between a single proposal and the existing wording. I agree with you that what we are currently doing on this page has weaknesses. --Tryptofish (talk) 21:13, 24 September 2011 (UTC)[reply]
It is probably better to ask her directly, but my understanding of SlimVirgin's position is that she would not insist on a strict binary choice... As I understand it, her concern is that the RfC could be seen as being a choice between two changes... and what she insists on is that the existing wording be clearly listed as being an option. For myself, I don't have a problem with a three way choice between a) the current language, b) my proposal and c) Trypto's proposal. To be honest, while I prefer my version (I think it clarifies some points that Trypto's does not), I could easily live with any of these three choices. Blueboar (talk) 21:36, 24 September 2011 (UTC)[reply]
I think Andrew makes a valid point about a split vote, though. --Tryptofish (talk) 21:40, 24 September 2011 (UTC)[reply]
Would it make sense to hold back to back RFCs, or even concurrently, the first for whether a change is needed, and the second for the which version is better? I would not mind having a concurrent !vote, eg. "Support the current wording, but support option A if current wording is to be changed" --Nuujinn (talk) 23:18, 24 September 2011 (UTC)[reply]
I suspect that many people will not support change as an abstract principle, given the past RfCs. There probably has to be a concrete proposal. (By the way, I can no longer tell whether this is a page, or a talk page.) --Tryptofish (talk) 21:25, 25 September 2011 (UTC)[reply]

The word initial (and the word threshold)

edit

Earlier parts of thread are on the talk page.

Is "A fundamental requirement..." acceptable to everyone?--Andrew Lancaster (talk) 12:32, 20 September 2011 (UTC) Or:[reply]

My thought is at this later point, it time for folks tomake specific proposed changes (to Blueboar's proposal) to discuss. Sincerely, North8000 (talk) 00:44, 21 September 2011 (UTC)[reply]
"A fundamental requirement" works for me. --Nuujinn (talk) 01:06, 21 September 2011 (UTC)[reply]
It worked for many people. So if we are going to be using it, let's start putting it in drafts. And if we still doubt we are going to be using it, let's have a quick poll?--Andrew Lancaster (talk) 07:20, 21 September 2011 (UTC)[reply]
By all means, be bold. --Nuujinn (talk) 10:44, 21 September 2011 (UTC)[reply]

Wrong. There was discussion of why "fundamental requirement" would create problems, and this discussion has now been moved to the talk page. As for making specific proposed changes to Blueboar's proposal, that's what people have been doing. I'm getting tired of repeating myself. Increasingly, this discussion has petered down to just being an echo chamber amongst a very small number of editors, who are likely to get a rude awakening if they try to take this to the community. --Tryptofish (talk) 19:12, 21 September 2011 (UTC)[reply]

Yes. I think however that this makes sense. We have Blueboar's proposal in hand and it is good we are not rushing to propose it before it has not been considered at length. It is also healthy that the small group you mention are talking about ideas, but making a wikimess out of them by pasting them all together bit by bit and letting it go out of control. That all seems logical.--Andrew Lancaster (talk) 19:40, 22 September 2011 (UTC)[reply]

Proposed change from North8000

edit

Change "Wikipedia has other policies and guidelines that affect inclusion" to "Other policies, guidelines and considerations affect inclusion." North8000 (talk) 15:52, 20 September 2011 (UTC)[reply]

  • I don't think the long version that includes the sentence in question is as likely to get community support. But if we go this way, I'd drop "and considerations", so it would just be "Other policies and guidelines". With those caveats, I agree that shortening the beginning of the sentence would be a good copyedit. --Tryptofish (talk) 20:07, 20 September 2011 (UTC)[reply]
Hardly seems worth debating on this particular forum of discussion?--Andrew Lancaster (talk) 20:10, 20 September 2011 (UTC)[reply]
Just a proposed tweak. My thought was that "considerations" would make it more precise, and things can be left out for reasons that are not policy or guidelines. Suggest we just do a quick thumbs up/down on this idea and move on. I won't jump off a tall building if this idea gets rejected. North8000 (talk) 20:29, 20 September 2011 (UTC)[reply]
I agree with Tryptofish, I like the shorter version, but considerations opens the door for all kinds of lawyering. But do you have any specific notions as for considerations that would not fall under guidelines and policies? --Nuujinn (talk) 01:09, 21 September 2011 (UTC)[reply]
Yes, for example when there is a consensus by the editors at the article to leave it out because it is too far off-topic, too lengthy, or implausible (probably false). And assuming that in each case there is not a "conflict" situation that kicks in the "inclusiveness-for-balancing" provisions of wp:npov.North8000 (talk) 10:24, 21 September 2011 (UTC)[reply]
I would think those might be covered by WP:DUE and WP:LIMIT. --Nuujinn (talk) 10:43, 21 September 2011 (UTC)[reply]
Are you saying that editors can only leave something out of an article if there is a policy or guideline that excludes the material? Don't forget...pick any article, and you will see that about 99.99999999999% of all of the material in the world has been "left out" of it. North8000 (talk) 11:22, 21 September 2011 (UTC)[reply]
No, not at all. If there's consensus to leave something out, there's no problem. "Considerations" opens the door for one to argue outside of policies and guidelines. If something is too far off-topic, too lengthy, or implausible (probably false), they are covered by DUE, LIMIT, and V respectively, so I think leaving "considerations" out works better. --Nuujinn (talk) 21:42, 21 September 2011 (UTC)[reply]
Another time/place I'd debate on a couple of those items, but either way I'm cool with leaving it out.....it was just an idea. North8000 (talk) 21:48, 21 September 2011 (UTC)[reply]
OK, so lets review my proposed changes as two separate questions:
  1. The copy editing (without addition of "considerations" )
  2. Addition of "considerations" North8000 (talk) 10:24, 21 September 2011 (UTC)[reply]
  1. Not really needed, because we shouldn't be discussing this (a separate section that takes "not truth" out of the lead).
  2. Oppose in any case. --Tryptofish (talk) 19:15, 21 September 2011 (UTC)[reply]

Results: "Other considerations" did not receive any support except by me and so I consider that it didn't fly. The copy edit received support and no opposition except one comment that it should be a moot point. North8000 (talk) 11:21, 23 September 2011 (UTC)[reply]

So now it's: Change "Wikipedia has other policies and guidelines that affect inclusion" to "Other policies and guidelines affect inclusion." Should I put it in the revised draft? North8000 (talk) 11:23, 23 September 2011 (UTC)[reply]
I think I'll withdraw this for the sake of simplicity. North8000 (talk) 14:57, 23 September 2011 (UTC)[reply]

Work area for development of rationale for main proposal

edit

Working draft of rationale for main proposal

edit

- - - Beginning of editable section - - -

Intro

edit

There is a long history of intensive discussion surrounding the term "not truth" and that is the central topic of this proposed change. Since August 30th, 2011 this has occurred primarily at Wikipedia_Talk:Verifiability/First sentence and it's project page Wikipedia:Verifiability/First sentence; in the months prior to that it occurred primarily at Wikipedia_talk:Verifiability. The primary objective of the proposed change is to reduce negative impacts associated with the term "not truth" while retaining it's intended meaning (that "truth" is not a substitute for verifiability) This is a compromise proposal which emerged via an extensive process. It is hoped that folks who advocate less or more change can support it as a compromise.

Main rationale presentation

edit

The first line of the policy currently reads: "The threshold for inclusion in Wikipedia is verifiability, not truth—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true." There are substantial concerns about this sentence (in previous RFCs, about 50% of editors asked felt the first sentence was unsatisfactory), and particularly the phrase "Verifiability, not truth". Some editors feel the phrase should be removed from the policy entirely and the wording should focus on what WP:V is, not what it is not. Others feel that the phrase is vital. After months of debate (and numerous polls and RfCs) a new proposal has emerged, to keep this phrase but move it to its own section and clarify it.

  • Background: The concept that truth is not the threshold for inclusion in Wikipedia was added for a specific reason - to support WP:NOR in saying that material should not be included in Wikipedia unless there is a source that directly supports it. At that time, we had a serious problem with editors wishing to add unverifiable material purely because "it's true". This attitude was useful to editors attempting to "prove" their pet fringe theory. However, as WP:V has changed over time, the sentence has been moved earlier and earlier in the policy, and stripped of its original context. It has therefore taken on meanings that were never part of this original intent.
  • Concern: The sentence can be misconstrued to mean that any material that appears in a source must be included... simply because it is verifiable. This misinterpretation is in conflict with several other policy and guideline statements (especially the WP:Undue weight section of WP:NPOV) but several examples of this misinterpretation happening in practice have been provided.
    • How the proposal resolves this concern: The proposal adds an explanation that "while verifiability is a requirement for inclusion, it is not a guarantee of inclusion", and it notes that other policies and guidelines can affect inclusion.
  • Concern: The sentence can be misconstrued to mean that we may not discuss the possibility that verifiable information is untrue (ie that the source may contain an error). That we must accept what a published source says simply because it is published, even if we have reason to believe that the material is inaccurate or that the source is less than reliable. This was never the intent. We often need to make editorial judgment calls about the reliability of specific sources when it comes to specific information. A source may be reliable for one statement, but not reliable for some other statement. We also have to make editorial judgment calls as to the relative reliability of one source when compared to others. As Jimbo Wales puts it, "We are not transcription monkeys." We do want the information we present in Wikipedia to be accurate, as far as possible.
  • Counter Concern: However, as NPOV notes, we can not omit significant viewpoints just because we disagree with them (or even because the majority of sources disagree with them). Sometimes we should discuss facts and opinions that may be untrue, because doing so gives the reader a complete picture of disagreement among the sources.
    • How the proposal resolves these concerns: The proposed language acknowledges that inclusion of potentially untrue information depends on context. We can not make a firm one-size-fits-all rule on this. The proposal points out that the question of whether to include controversial and potentially untrue material is a complex one, that involves applying editorial judgment. It points the reader to other policies and guidelines that may help.
  • Concern: Outside the world of Wikipedia, the natural meaning of "verifiability" concerns truth. Introducing the phrase "not truth" in the lede is distracting and confusing particularly for new editors. The lede should focus purely on explaining the Wikipedian idea of Verifiability without introducing secondary concepts. While it is important for the policy to discuss the issue of truth/untruth, this belongs in the body of the policy.
    • How the proposal resolves the concern: The issue of truth is moved out of the lede and into to its own separate section.

Conclusions: The goal of this proposal is not to change the meaning of the policy in any way... but rather to better clarify existing policy and avoid misinterpretation. The concepts behind the phrase "Verifiability, not truth" are valid and important, and they should remain part of the policy. But they are very complex concepts that need to be more clearly explained.

Extra note on main concerns

edit

There is a spectrum of concerns with the current version, but the main issues are:

  • Philosophy: As Jimbo Wales put it, "We are not transcription monkeys." Editors espousing this view say that Wikipedia is seriously aimed at getting to the truth, and the phrase "not truth" is not just confusing but actually misleading about Wikipedia's aim, because it seems to imply that Wikipedia might be unconcerned about the truth.
  • Ambiguity: This concern observes that the sentence either 1) protects against insertions of material that are alleged to be "true" but not verifiable, or (2) protects against exclusions of material that are alleged to be "not true" but verifiable. The current first sentence is ambiguous which is meant. But discussions appear to show broad consensus that (1) is intended, but (2) is not. Therefore editors are concerned that other editors are misunderstanding, or deliberately misunderstanding, the ambiguity in order to introduce or protect a minority POV version or material which they do not even claim is accurate.

Farther-reaching suggestion by Jimbo Wales

edit

Wikipedia founder Jimbo Wales suggested change on this that was two steps more dramatic; beyond complete removal of "not truth" to actual reversal of it (Copied from wp:verifiability talk page):

"There are cases in which we should removed cited material on the basis that it isn't true. Existing policy allows for this and encourages this but the "not truth" meme leads people to think that we should be simple transcribers, as opposed to editors. I propose that we change it to "and truth" to prevent that misunderstanding but in any event, separating the words is a good start.--Jimbo Wales (talk) 05:18, 31 August 2011 (UTC)"[reply]

However we chose a compromise proposal which represents a smaller change which retained "not truth", but moved it later, and added clarifying wording to it.

Logical dissection

edit

The intent of the first sentence is the same as the policy, which is to firmly establish verifiability as a requirement for inclusion of material. Verifiability, while not causal, is associated with accuracy. The term "not truth" with its current position and lack of intent-clarification is often construed to mean a statement of the exact opposite, i.e. a statement that Wikipedia does not seek accuracy. It is also often misconstrued as meaning that verifiability gives one a mandate to force inclusion of the material. It is also often used to assert that editors may not discuss or take into consideration falsity of (sourced) material as a basis for possibly leaving it out. (recognizing that in certain cases wp:npov trumps this discussion and dictates that it stay in) The proposal moves those two words and provides additional context for them. This preserves the phrase and it's intended meaning (that truth/accuracy is not a substitute for verifiability) while reducing its unintended effects.

- - - End of editable section - - -

Discussion for development of rationale for main proposal

edit

I find it interesting that while we are all happy to propose ideas and debate them endlessly... no one has (yet) come up with a simple statement as to a) why a change is needed, and b) how the proposal actually resolves the issue. Could we please focus on this for a while... we are not going to agree on wording unless we agree on what we are actually trying to achieve here. Blueboar (talk) 17:58, 13 September 2011 (UTC)[reply]

S Marshall has given a first effort which aims at a big picture. OK, on that challenge, I'll try to come from the other direction and list the most widely shared concerns I know of, point by point, with the first part of the current first sentence:-

User:Andrew Lancaster. Draft of first sentence (edited after discussion)
The threshold for inclusion in Wikipedia is verifiability, not truth
  • ...not truth... This is the most complicated discussion. The intended meaning is that discussion about inclusion should not be based on discussion about personal opinions. But even defenders of it normally seem to admit that it does not really say this in any clear way. They defend it because it gives a "jolt" of thoughtfulness which can then, ideally, lead to enlightenment later. Instead of giving the intended message, saying that in Wikipedia truth is not important for inclusion does a few problematic things:
(1) It wrongly gives the impression that Wikipedia editors are actually commanded not to aim at truth, which is obviously inconsistent with Wikipedia editors being intelligent human beings working together to make a reliable, verifiable, quality encyclopedia. Or to put it in Jimbo Wales' words it makes Wikipedia editors seem like "transcription monkeys".
(2) As with some of the above concerns, it leads people to the idea, quite logically given the way it is written, that WP:V trumps all other policies, including the other 2 core content policies.
(3) As seen by some of the twisted defenses of what is after just a wording question, it is leading rapidly to a form of hieratic thinking in Wikipedia, in which normal people do not need to understand the complex and indirectly written traditional rules of the higher order editors. This is against the principle that WP:Anyone can edit.
  • Of secondary importance is the way in which this first sentence seems to try to over-rule other core content policies:-
(1)The... It is the wrong word. There are 3 core content policies and there is a consensus that they work in an interlocking way. It should be made clear
(2)...threshold... Again it is the wrong word. "Requirement" or something like that is obviously what is intended. The idea of a threshold is that if you get over it, you are in. There is consensus that verifiability is not enough to be in.

Am I missing anything?--Andrew Lancaster (talk) 20:30, 13 September 2011 (UTC)[reply]

In my opinion, yes, you are missing the fact that the primary issue is really "not truth", the view by some editors (not necessarily me, not necessarily a majority of the community) that it makes it sound like Wikipedia doesn't care if untrue material is published in the encyclopedia. The stuff about "The" and "threshold" is just not a concern for the vast majority of Wikipedians. The more that editors here weigh down the argument with secondary issues, the more likely it is that the larger community will find reasons to disagree and stick with the status quo. --Tryptofish (talk) 21:04, 13 September 2011 (UTC)[reply]
I agree that it would be best if we separated out "not truth" for the moment and focused on that. "Threshold" will be a sideshow. It needs fixing but I don't think fixing it will be anything like the struggle we had with "not truth".—S Marshall T/C 22:59, 13 September 2011 (UTC)[reply]
Tryptofish and S Marshall agree about something! The rest of you better take this seriously!   --Tryptofish (talk) 23:03, 13 September 2011 (UTC)[reply]
OK. But then it is not that I was missing something, but rather you are saying I did not emphasize something enough, or you are maybe saying I had too much. If we are going to separate out this one first sentence issue for special attention then the draft above should start by saying that there are various issues, but one draws the most discussion. So then looking at my "not truth" summary, am I missing anything?--Andrew Lancaster (talk) 08:25, 14 September 2011 (UTC)[reply]
  • I do think we need to use more neutral phrasing. Words like, for example, "obviously inconsistent with Wikipedia editors being intelligent human beings" will derail an RFC in minutes. Remember, this is Wikipedia, where editors absolutely insist on going on tangents—on Wikipedia, if God said "let there be light", he'd get no further because of all the people going "what colour?" Basically, if we make it possible for people to find ways to be offended by it, they will.—S Marshall T/C 11:12, 14 September 2011 (UTC)[reply]
OK, so we should take out the bit about being intelligent human beings.--Andrew Lancaster (talk) 12:08, 14 September 2011 (UTC)[reply]

I think that another thread of the rationale should be that this is a compromise solution which will (hopefully) resolve a large, difficult and long running debate, and vs. the alternative of not doing so. North8000 (talk) 22:54, 14 September 2011 (UTC)[reply]

Yes. It is and must be a compromise. Should the final version include some summary of the evidence we have for the different community opinions which we tried to take into account?--Andrew Lancaster (talk) 07:27, 15 September 2011 (UTC)[reply]
So, why not just delete the two words "not truth"? Those two words have been criticized by many editors including Jimbo Wales. It's just a two-word style change, not a policy change. The sentence still retains the same meaning.
As far as a compromise is concerned, if you want to do that, you have to make the compromise with SlimVirgin, who is the originator of the phrase and is the leader of those who are against change. --Bob K31416 (talk) 17:14, 15 September 2011 (UTC)[reply]
I was mostly making the point that resolving this vs. it going on forever is another reason to support whatever proposal emerges. North8000 (talk) 17:37, 15 September 2011 (UTC)[reply]
In regard to those last several points, it's worth taking a look at the discussion between Blueboar and SlimVirgin at User talk:SlimVirgin#WP:Verifiability. I think an important take-home message is that it's likely that the community will be presented with a binary choice between one proposal for a new version, and the status quo. Anything the community dislikes about the new version will be reason to say "no". --Tryptofish (talk) 20:52, 15 September 2011 (UTC)[reply]
It is an interesting discussion. SlimVirgin is apparently amongst those who want WP:V to be taking what she knows is a non-obvious philosophical position about the meaning of the word "truth", saying that it needs to be in the first sentence "especially given that we use the word "verifiability," which implies truth-seeking". So what does the community of Wikipedia feel most strongly about? WP:V itself, or taking this minority stand on a philosophical point? There are many of us who think the policy wording should be based on trying to make a clear wording of the consensus ideas about what the policy intends, not taking positions about the meaning of the word truth at all. The two aims are quite different, I would say strictly in opposition to each other, and perhaps we should mention the two aims as background to the proposal.--Andrew Lancaster (talk) 20:19, 16 September 2011 (UTC)[reply]

I think not addressing the question at all is a form of expressing a philosophy on truth by ommission. There is a nearly non-negotiable truth in Wikipedia: WP:NPOV and every thing else flows from that. Even WP:V is saying, "We can sometimes ignore absolute neutrality because the sourcing is overwhelming" - that is verifiability, not truth. It is not a minority philosophical position on truth, it is a succinct, disruptive explanation of NPOV. NPOV itself is non-intuitive, and this is often discussed in both practical and philosophical ways, but it is what this project is about. I am for internal consistency whenever possible, in particular because any exceptions to a any rule is covered by WP:IAR. I think opposition to the "verifiability not truth" is largely based on a philosophical, rather than practical consideration, one that says that there is objective truth, and that this truth is to be help true by the encyclopedia. It is a less extreme version of the argument that leads to Conservapedia: roughly wanting to codify (ie "resolve for good") issues that should better be left for consensus on a case-by-case basis.--Cerejota (talk) 19:08, 20 September 2011 (UTC)[reply]

I don't know about others, but that description bears no resemblance to my concerns about "not truth"......mine centers centers around negative impacts that have nothing to do with verifiability. But we digress. North8000 (talk) 19:26, 20 September 2011 (UTC)[reply]
I am not so clear what North's answer refers to. But Cerejota, I like your response, but only think you are misunderstanding others. I think NPOV is much more important than most newbies and readers of Wikipedia are being taught. The idea that WP:V flows from WP:NPOV is not what they are learning at all. Your intention is good but we need to find a way to achieve it.--Andrew Lancaster (talk) 20:14, 20 September 2011 (UTC)[reply]
Agree that NPOV is very important...in fact it needs strengthening by making it more operative. But don't agree that wp:ver flows from wp:npov. North8000 (talk) 20:40, 20 September 2011 (UTC)[reply]
Well, perhaps not literally enough that it means the poliy pages should say it like that. But I think it is true that WP:V would not be necessary if it were not for the aims of WP:NPOV. It is neutrality which is the hardest thing to define and discuss. Think about this: If neutrality were easy to agree upon then we could just aim at "truth" or accuracy or whatever, in the sense that we would all be agreeing about where to draw the line anyway. Many or most discussions on WP:RSN turn out to be about balance and weighting, with WP:NOR and WP:V being pulled in as part of an argument about what is essentially "neutrality".--Andrew Lancaster (talk) 07:19, 21 September 2011 (UTC)[reply]
Yes, but wp:ver does an immense amount of vetting work where pov is not even a question. North8000 (talk) 11:13, 21 September 2011 (UTC)[reply]
I see what you mean, but I think Cerejota's point goes beyond that. WP as a community has found a way to categorize types of debates, but the original problem with many of them is "favouring of personal opinions". Whether we say this most basic opinion is properly called WP:NPOV or WP:V is therefore a somewhat artificial question in the sense that these are terms invented by us to solve the problem. So I think it is not WRONG to say that WP:NPOV required WP:V and caused it to come into being.--Andrew Lancaster (talk) 11:19, 21 September 2011 (UTC)[reply]

Hybrid version?

edit

The "not truth" phrase in the first sentence has generated almost endless, often heated, and complicated discussions, which have made it clear that quite a number of people, old and new, can see problems with it

  • Philosophy: It implies that Wikipedia is so unconcerned about real facts that editors are actually commanded not to aim at truth; this is inconsistent with Wikipedia's aim of being a reliable, verifiable, quality encyclopedia. As it currently stands, it can be (is?) quoted by those who write Wikipedia off as being "unreliable".
  • Ambiguity: It is unclear, at present, whether our aim is to
    1) protect against insertion of material that is alleged to be "true", but is not verifiable, or
    2) protect against exclusion of material that is alleged to be "not true", but is verifiable

The combination of these points can be "creatively misunderstood" by editors wishing to include inappropriate material. Defence of the "not truth" phrase so far is that it supposedly gives a "jolt" of thoughtfulness which then, ideally, leads to enlightenment later, and / or it's been quoted so often, it's traditional, etc.

We currently have a compromise solution which will (hopefully) resolve a large, difficult and long running debate; without a solution, this debate will continue to take up many hours of editor time. It is not going to go away.

I think / hope I've covered most of the bases so far. Pesky (talk …stalk!) 04:45, 15 September 2011 (UTC)[reply]

Over all, I think that this is a good approach, not too long. --Tryptofish (talk) 20:45, 15 September 2011 (UTC)[reply]
I am not confident of that. I am not sure now what our latest proposal is. It is being argued that this proposal needs a quite polished draft and also a well presented "case".--Andrew Lancaster (talk) 12:30, 20 September 2011 (UTC)[reply]

Hi, the part about "can be used (is?)"... See for example, Simson L. Garfinkel in Wikipedia and the Meaning of Truth. He quotes "Verifiability, not truth" in the context of criticizing Wikipedia for not caring about objective truth. I believe his criticism is not so much that this makes Wikipedia unreliable, but that Wikipedia endorses that awful style of relativistic, pseudo-balanced reporting when the two sides of any issues are treated with equal validity, letting the reader think the truth is somewhere in the middle. I believe the new proposal, which discusses "not truth" in context and also links to WP:DUE, should help avoid such misunderstanding. Regards, Vesal (talk) 18:10, 19 September 2011 (UTC)[reply]

Perhaps yes, the proposal justification should mention things like this, and perhaps also the posting of Jimbo Wales on the subject. Any proposal will face a lot of rhetorical resistance, so it would be important to make sure there is some evidence that this is not just a small group wanting change for its own sake.--Andrew Lancaster (talk) 16:15, 20 September 2011 (UTC)[reply]
Jimbo's just another editor, remember. Doc talk 03:37, 21 September 2011 (UTC)[reply]
Sure, but his talkpage is also a forum of discussion on issues like this which is separate to this one. And one discussion there caught attention of a wider public. The subject under discussion is how to stop this discussion being fillibustered by people who are going to insist on a community wide RFC. It is understood, I think, by everyone in this discussion that the reason for calling for such an RFC is because "voters" wil not be thinking long about it, and therefore they will be open to simple suggestions that this is just the proposal of a small group of editors who hang out on the policy talkpage and want change for no good reason. If we want to avoid rhetoric determining the result of the RFC we have to also think about how rhetoric works in real life, no? Therefore, as was being discussed, any evidence of discussion on this matter in other places is interesting.--Andrew Lancaster (talk) 07:01, 21 September 2011 (UTC)[reply]
It's time to move concretely forward. A few more days proposing/deciding any changes to Blueboar's proposal and finalizing a succinct rationale. North8000 (talk) 20:18, 22 September 2011 (UTC)[reply]
Re Doc's comment "Jimbo's just another editor" - Unlike other editors, Jimbo has been quoted and referred to significantly in policy. See WP:NOR and do an edit/find using the keyword Wales, and you will find 9 hits on that page. So he's not just another editor. --Bob K31416 (talk) 20:54, 22 September 2011 (UTC)[reply]
Agreed. And, of course we know that he does not claim or participate on this with any formal authority or position; that's not what we're talking about. North8000 (talk) 21:07, 22 September 2011 (UTC)[reply]
@ Andrew - how beautifully clearly stated :o) Pesky (talk …stalk!) 04:49, 23 September 2011 (UTC)[reply]
Seeking Jimbo's opinion on this particular issue (as it was sought) could be seen as an attempt to alter normal consensus-building - could it not? Is that completely out in left field? He's the founder of WP: many will agree automatically with him because of that. But it all comes down to consensus among all the editors. His opinion (as he acknowledges) shouldn't be given any more weight than anyone else's in cases like this. Jus' sayin'. Doc talk 04:56, 23 September 2011 (UTC)[reply]
People aren't robots. I think that they would take what Jimbo says into consideration, many would assign a little extra weight to it, and the make their own decision. North8000 (talk) 10:25, 23 September 2011 (UTC)[reply]
It's normal consensus building for editors to be positively influenced by the opinions of others, depending on how trustworthy a person’s opinions have been in the past. Please note that Jimbo’s opinion has been sufficiently valued in the past that it has been quoted in all three major policies.
It is up to each editor to decide how much weight he or she gives to anyone’s opinion. Personally, considering Jimbo’s understanding of Wikipedia and its position in real-life society, I would not discount his opinion as the opinion of just another editor. --Bob K31416 (talk) 10:43, 23 September 2011 (UTC)[reply]
People seem to be far more worried about what SlimVirgin thinks.--Andrew Lancaster (talk) 10:59, 23 September 2011 (UTC)[reply]
I guess it's only human to promote the importance of opinions that agree with one's own, and downplay the importance of those that disagree with one's own. I've seen folks (when all else failed) try to say that creatively interpreted ancient comments by Jimbo should have huge sway. And now here he has pretty directly weighed in on this topic. North8000 (talk) 11:31, 23 September 2011 (UTC)[reply]
And I've seen here the extremes of someone saying that SV should be given a huge amount of weight, and that Jimbo's comments should not even be mentioned. North8000 (talk) 11:57, 23 September 2011 (UTC)[reply]
SV is one of our longest serving editors, and she was instrumental in developing most of our policies and guidelines (including this one)... Jimbo is (of course) the guiding light behind Wikipedia itself. Both should be respected for their contributions, and I think both of their opinions should carry a fair amount of weight. Neither of them can impose their will against consensus (nor do they desire to do so). However, their support or opposition to a proposal does have a profound impact on what the consensus is, and how it develops. We would be wise to listen to their opinions. Blueboar (talk) 12:19, 23 September 2011 (UTC)[reply]
Once again Blueboar... you make a lot of sense. So where exactly does this go from here, anyway? Doc talk 13:14, 23 September 2011 (UTC)[reply]
IMHO none of the proposed changes to Blueboar's proposal (including mine) have gain a consensus or even traction and so my suggestion is that we should consider it to be the proposal. And take a few more days developing the rationale, and then go to RFC (?) North8000 (talk) 15:01, 23 September 2011 (UTC)[reply]
That's up to Blueboar, and I'm not even kidding. He should get a "diplomacy" barnstar, as he's shown himself to be quite able to successfully moderate the debate (and then propose a possible compromise). Not an easy thing to do. Where next, Blueboar? Doc talk 15:10, 23 September 2011 (UTC)[reply]
Good, but Blueboar has not been active in that particular respect. North8000 (talk) 18:31, 23 September 2011 (UTC)[reply]
It's up to me... great: I say we should continue to debate the meta-meaning of every word here on this sub-page for a few more years, until the whole issue slips into obscurity (this will advance my secret agenda of having the policy itself retain the current language while we debate). :>) OK, seriously... it isn't really up to me when we "go live" with an RfC. I would agree that, so far, my proposal is the only one that has achieved some degree of consensus, and I agree that none of the various suggestions as to amendments to my proposal have gained consensus... but whether we have reached the point where we should stop discussion and make my proposal "THE" proposal is not something I can decree by fiat. Blueboar (talk) 13:06, 24 September 2011 (UTC)[reply]
So I propose a week (October 1st, delay if needed) to develop the rationale, and develop any last minute tweaks to Blueboar's proposal, and then go to RFC. North8000 (talk) 13:21, 24 September 2011 (UTC)[reply]
Suggest you propose this in its own section (both here and at the V/FS talk page)... it will get lost buried here in this thread. Blueboar (talk) 14:10, 24 September 2011 (UTC)[reply]
Blueboar, you are right about debating the meta-meaning, etc. I think that has resulted in the suggested amendments not really being examined properly. Your mileage may vary, as may my eventual support, or not, for the proposal that emerges. --Tryptofish (talk) 21:18, 24 September 2011 (UTC)[reply]
  • Philosophy: It implies that Wikipedia is so unconcerned about real facts that editors are actually commanded not to aim at truth; this is inconsistent with Wikipedia's aim of being a reliable, verifiable, quality encyclopedia. As it currently stands, it can be (is?) quoted by those who write Wikipedia off as being "unreliable".
I don't see any such implication. It's a strong reminder that Wikipedians, for all their ego, still need humility. We are not seekers of "truth;" we are faithful recorders of what others "out there" have studied -- we merely claim to understand it enough (certainly more than monkeys), so that we have recorded their (those out there) "truth" faithfully.
  • Ambiguity: It is unclear, at present, whether our aim is to
    1) protect against insertion of material that is alleged to be "true", but is not verifiable, or
  • 2) protect against exclusion of material that is alleged to be "not true", but is verifiable
That's not ambiguous, it's broad. Just as a policy should be.

Alanscottwalker (talk) 01:52, 24 September 2011 (UTC)[reply]

We are seekers of truth, and that includes you. What the present policy of WP:V is trying to say is that we shouldn't include material that has not already been published in a reliable source, based only on our assertion that it is true. The slogan "verifiability, not truth" alone does not say this but implies that we are not interested in the truth. Please note that the last phrase in the first sentence of this policy is much closer to what we want to express in policy, i.e. "not whether editors think it is true". Unfortunately, it is the nonsensical slogan "verifiability, not truth" that has been used to represent the policy WP:Verifiability. May I suggest that you try asking someone you know who doesn't edit Wikipedia, what they think of the slogan "verifiability, not truth". --Bob K31416 (talk) 12:32, 24 September 2011 (UTC)[reply]
I've said it before and I'll say it again: the very simplest solution is to put the word truth in "scare quotes". This would further illustrate that one person's "truth" (without verifiability) is not necessarily the actual truth (which we could never discourage, naturally), and therefore not fit for inclusion. No one wants to hear it, however. Doc talk 13:04, 24 September 2011 (UTC)[reply]
I put the quotes on and expect to be reverted in 3... 2... 1... Doc talk 13:17, 24 September 2011 (UTC)[reply]
No objection from me to your edit. Alanscottwalker (talk) 13:42, 24 September 2011 (UTC)[reply]


"We are seekers of truth, and that includes you."

Please, get over yourself. You want to seek truth, write your own book. We are not seekers of truth, here. We are recorders and relators of the knowledge of others. Verifiability not truth, is the exact right goal in this policy. And 'brevity is the soul of wit.' Alanscottwalker (talk) 13:42, 24 September 2011 (UTC)[reply]

Alanscottwalker, There seems to be some miscommunication. Please carefully reread my previous message and you will see that I agree with you that editors should not include their own notions of what is true if it hasn't already been published in a reliable source.
Regarding "seeking truth", at least in your case it appears that the phrase "verifiability, not truth" has caused you to think that "seeking truth" means something bad and undesirable. I think that what you are against is Wikipedia editors putting their own unverifiable notions of truth into Wikipedia, and I agree with you. This is not just seeking truth but rather claiming something to be the truth that is not verifiable, and I agree with you that this is not what we want. Please reread my last message with this in mind. --Bob K31416 (talk) 14:58, 24 September 2011 (UTC)[reply]
Yes, truth telling is bad, where it gets in the way of verification, in Wikipedia. The more common error, I see, is when Wikipedians, out of mistake, or their regard for there own truth, misrepresent or deny the source. Alanscottwalker (talk) 15:15, 24 September 2011 (UTC)[reply]
Now that it appears we agree, please look at the first sentence of WP:V and see if you would still agree with it if just the two words "not truth" were removed, i.e. "The threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true." --Bob K31416 (talk) 15:42, 24 September 2011 (UTC)[reply]
I would agree that your proposal is not as effective, communicative, or efficient, in getting across the point that what we are interested in is verifiability not truth. Alanscottwalker (talk) 16:32, 24 September 2011 (UTC)[reply]
It looks like we agree that the meaning of the first sentence wouldn't be changed by removing just the two words "not truth" but we disagree about whether it would be as effective, communicative, or efficient. 1) I think it would be more effective and communicative because it eliminates a nonsensical phrase and defines first what is meant by verifiability. 2) It would be more efficient because it has fewer words. I suspect that you have the opinion of others here that the nonsensical phrase is desirable because it gives readers a "jolt". I agree that a nonsensical phrase would give someone a jolt, as would other types of bad writing. --Bob K31416 (talk) 17:43, 24 September 2011 (UTC)[reply]
It's only nonsensical, if you don't realize that the two different words have two different connotations, but that's how language works. Such thinking as yours is a hallmark of bad writing. Alanscottwalker (talk) 18:08, 24 September 2011 (UTC)[reply]
When one reads the phrase "verifiability, not truth", one has no way of knowing what its connotations are at that point. It isn't until one reads the last phrase "not whether editors think it is true" does one have any idea what the phrase "verifiability, not truth" might mean. One may then go back and reread the sentence to try to understand what the first phrase means, but may still be uncertain. Wouldn't it be simpler for the reader if just the two words "not truth" were left out, so that the reader doesn't have to needlessly reread the sentence and deal with a special meaning of "not truth" which deprecates the meaning of the word "truth"? --Bob K31416 (talk) 19:52, 24 September 2011 (UTC)[reply]
Two quick comments: 1. The other alternative is to explain the two words at more more length like Blueboar's compromise. 2. Part of the problem, let's not forget, is the way the two words are set up by the beginning of the sentence (The threshold for inclusion in Wikipedia is...) which proclaims the words about to come as being all powerful (trumping WP:NPOV for example, when I think it should be the other way around). I am amongst those who think that they are also important. Or, to put it another way, it is the combination which creates a bad impression. WP:NPOV represents a real aim we all have: balance. WP:V, is a practical tool to achieve it. But the current wording is out of kilter, telling us "not truth" is our true aim, and giving us wording which implies WP:NPOV and other policies are less important.--Andrew Lancaster (talk) 20:31, 24 September 2011 (UTC)[reply]
The phrase is in the middle of a sentence; its generally required of readers that they remember the beginning, middle and end of a sentence. That's how written English works. Andrew, I don't get the process argument. First you have to have something verifiable (whether or not you know its true) in order to add it to any article, then you consider POV and relevance. You won't be able to judge the last two without the first. Alanscottwalker (talk) 21:27, 24 September 2011 (UTC)[reply]
That would be reasonable to expect if the sentence didn't have any parts that weren't clear. Unfortunately it has the the nonsense phrase "verifiability, not truth" which is deliberately "counter-intuitive and paradoxical", according to its originator/lead supporter, and others. Seems like our discussion has run its course. Regards, --Bob K31416 (talk) 00:59, 25 September 2011 (UTC)[reply]
The claim that it's nonsense fails because the sentence defines it. That's how language works. Words have multiple meanings, connotations, and shades; so, words cannot be taken in the abstract, they can only be read in context. Even, if it were counter-intuitive and paradoxical, in the abstract (which I don't see -- they are two different words) that claim is meaningless, without knowing the context. Alanscottwalker (talk) 14:22, 25 September 2011 (UTC)[reply]

(The following is a copy of a relevant message from another discussion, for easy reference.)

The slogan "verifiability, not truth" is contradictory and nonsensical. It's used out of the context of WP:V. Note that one could similarly have a context which defines the words in the contradictory nonsensical slogans of Orwell's fictional work 1984: "war is peace", "freedom is slavery", "ignorance is strength". It's better writing to use the common meanings of words instead of changing them. The two words "not truth" are not needed in the first sentence. The sentence means the same thing without those two words and wouldn't foster the nonsensical slogan "verifiability, not truth" if they were removed. --Bob K31416 (talk) 14:49, 25 September 2011 (UTC)[reply]

(I suggest, we not discuss this in two different places (because they have been two different conversations, with different points made, and it takes up space), but sinse you have, to complete the comments . . .
Do you read your own overheated examples? If you did, you would know how absurd your argument is. The phrase is not, "verifiability is truth," the phrase is, "verifiability, not truth." They are two different words. Alanscottwalker (talk) 15:04, 25 September 2011 (UTC)
I suggest you reread my message and note what the slogans all have in common: They are nonsensical and contradictory. --Bob K31416 (talk) 15:20, 25 September 2011 (UTC)
I reread your message. It is still absurd because you claim unlike things are the same. They are not; just as, verifiability and truth are not the same, as the phrase says -- they mean different things. Alanscottwalker (talk) 15:37, 25 September 2011 (UTC)) )
Alanscottwalker (talk) 15:58, 25 September 2011 (UTC)[reply]


There are always the little loopholes, of course. For instance, no one could write a FA on a classic film or novel without what would be normally be considered extensive original research comprising the plot summary. In most cases, however, verifiable sourced material trumps unreferenced "truth" when it comes to surviving inclusion. Doc talk 14:01, 24 September 2011 (UTC)[reply]

I'm afraid, I don't understand your example, a plot summary is verifiable and should be referenced. Alanscottwalker (talk) 14:09, 24 September 2011 (UTC)[reply]
This is just the first example I found.[1] There are no inline citations in the plot summary - so how is any of that information really verifiable? But that's not an indictment of how those types of articles are any sort of exception: I was more pointing out that there are always some exceptions to any rule, but that the general language is not so confusing for the vast majority of cases. Doc talk 14:16, 24 September 2011 (UTC)[reply]
OK, I generally agree, but difficult to verify is not the same as unverifiable, and that article would certainly be more informative (improved), if some references were included. Alanscottwalker (talk) 14:53, 24 September 2011 (UTC)[reply]

Proposed next step

edit

Please see Wikipedia_talk:Verifiability/First_sentence#Proposed next step. North8000 (talk) 11:11, 26 September 2011 (UTC)[reply]

I think the way of working on this discussion is becoming problematic, and making progress very difficult. I really don't like the current proposal for the rationale, and many of the issues have already been discussed. Problem is: where are we supposed to discuss ideas when things are being moved around so that things are not near each other, leading to discussion being ignored all the time?--Andrew Lancaster (talk) 12:12, 26 September 2011 (UTC)[reply]
Are you talking about the rationale or the overall change proposal? On the rationale my idea was that once an idea had a few folks behind it in the discussion to put it into the editable draft section. But not much was happening there and so I just put a few ideas straight in, knowing that the section is editable which means that folks can edit them or take them out. Maybe my bad. North8000 (talk) 12:43, 26 September 2011 (UTC)[reply]
I have the rationale in mind, but not just the rationale. But discussion about most things seems kind of stuck. I think it is an on-going problem, that as these discussions go on it becomes more and more difficult to contribute. I am not blaming individuals or suggesting bad intentions of course.--Andrew Lancaster (talk) 12:49, 26 September 2011 (UTC)[reply]

Further discussion of "rationale" drafts so far

edit

Here are some concerns:-

  • We seem to now have the idea fixed of breaking the rational into "philosophy" and "ambiguity".
  • I find the "ambiguity" proposals all rather unconvincing and personally if I read those rationales too quickly as someone who knew none of the background, it would put me off. It has already been said a few times that allowing for ambiguity is sometimes the intention when writing policy: we do not want policy to decide everything.
  • In practice the most important ambiguity is about whether verifiable material should be included, because it is verifiable, or more generally (I think it is the same thing) about whether this policy somehow trumps WP:NPOV, or whether it serves it. I believe it is the latter.
  • (The first words of the policy are to me a critical part of the problem. Not fixing them will lead any proposal open to the criticism that it actually fails to succeed according to its own aims.)
  • The "philosophy" sections are on the other hand more important but the rationales are very watery and unconvincing (I think pesky's version better than North's), and again do not reflect what most people seem to have agreed upon during discussions.
  • One of the key points everyone seems to agree upon is that this re-wording is aimed at keeping the SAME policy, and it is a fact, I think, that most experienced editors agree on how that policy should work, even if they find it hard to compress into a few words.--Andrew Lancaster (talk) 12:23, 26 September 2011 (UTC)[reply]
Furthermore:
  • I think the main SINGLE point of the rationale is one single point: more clear wording, in order to explain the existing policy as understood by the community. Trying to divide this into philosophy and ambiguity is a little misleading and comes across as negative I think.
  • The main reason the wording needs improvement are perhaps these two things:-
  • Making sure that "not truth" does not become an aim in itself. Being un-interested "transcription monkeys", as Jimbo Wales put it, would be inconsistent with Wikipedia's aim of being a reliable, verifiable, quality encyclopedia. The "not truth" formulation was intended to tell Wikipedians not to use unverifiable personal knowledge as a justification for putting something in Wikipedia, but rather to limit our goals for practical reasons, and only put in information in Wikipedia which can be verified by others. In other words "not truth" describes a limit we put on ourselves for practical reasons, not an aim in itself. So the words "not truth", should be given a more clear context, if they are kept.
  • Making sure the strong wording of the first sentence is not used as if it trumps other policies which do describe real positive goals, such as especially WP:NPOV. Concerning this particular point, the first words "the threshold for inclusion" are felt to give a wrong impression sometimes, because it implies that WP:V is the only policy describing a rule about what should be included in WP.
Comments welcome.--Andrew Lancaster (talk) 12:38, 26 September 2011 (UTC)[reply]
Overall looks good. Except I disagree (on structural grounds)with one point. While wp:ver and wp:npov may "work together" in a vague motherhood-and-apple pie statement type of way, and they often interact with each other, they basically cover very different things. My main thought is that policies should cover the areas that they cover, and avoid doing damage in areas that they don't cover. So, on "whether this policy somehow trumps WP:NPOV, or whether it serves it" I would say, in an operative/structural sense, "neither". Also. I do think that ambiguity (of those two words) is the core issue; it's current form it is too open to misinterpretation/mis-use. North8000 (talk) 12:49, 26 September 2011 (UTC)[reply]
Is the proposal a change in the meaning of policy? If not, perhaps that should be mentioned. However, even if it isn't a change in the meaning of policy, people may not believe that because it is large change. BTW, if only the two words "not truth" were deleted, the fact that this is not a change in the meaning of policy would be easier to see, and the fact that it was not a change in meaning was acknowledged by the originator of the phrase "verifiability, not truth". In any case, if the proposal is not a change in the meaning of policy, then that should be mentioned. --Bob K31416 (talk) 13:50, 26 September 2011 (UTC)[reply]
I suggest that any change in the policy is a change in the meaning of the policy, even if it is not a change in the intention we as a community are trying to express in the policy. Removing "not truth" is a change in the policy that many of us do not support--that was established months ago. Andrew Lancaster sums up the main reason for the changes in wording very well, especially in regard to "not truth". The phrase does not lose all value just because it is sometimes misunderstood or misused, or because it could be explained better. I do agree that being explicit that the proposed changes are not an attempt to change the direction or intent of the policy, but rather to improve the explanation would be a good approach. I am not at all sure that everyone will believe that is the case, however. --Nuujinn (talk) 14:06, 26 September 2011 (UTC)[reply]
I guess even a spelling correction could be called a "change" in the policy. But I think that by "not a change in policy" folks meant that it is not a change in what the operative statements of the policy dictate. North8000 (talk) 15:00, 26 September 2011 (UTC)[reply]
I agree with that assessment, but the problem is that the devils is in the details. If one argues that removing "not truth", for example, doesn't really change the policy's intent, I could accept that notion. But to argue that it doesn't change the policy, or the meaning of the policy, or the application of the policy is something that I myself could not accept. Policies are not simply statements of intent, they are tools we use to help guide the behaviour of ourselves and others, so fine distinctions in wording are, as these epic discussion attest, very important. Here's the thing that's become very clear to me over the last few months--there may not be strong consensus to keep "not truth", but thus far, there's clearly no consensus, weak or strong, on what to replace it with, which is what would be required to remove it. Now, you and Blueboar and Andrew Lancaster and others have done yeoman's work to get us this far in developing a compromise that seeks to address the concerns on all sides, and the summary above seems very accurate to me. My suggestion would be to move post haste to get the proposal out in front of the community sooner, rather than later, as the discussions here seem to be fragmenting the last few weeks. But if I can help in whatever direction you all would like to take this, please let me know. --Nuujinn (talk) 15:26, 26 September 2011 (UTC)[reply]
I just made up the "1 week" idea as a compromise between what would be viewed as moving too fast and too slow. Also the rationale still needs work. Sincerely, North8000 (talk) 15:32, 26 September 2011 (UTC)[reply]
It might be helpful in this discussion to look at the definition of "meaning". --Bob K31416 (talk) 15:55, 26 September 2011 (UTC)[reply]

OK, then....

edit

Do we agree on the following rationales:

  • That we are aiming to keep the policy (as it is currently widely interpreted and explained by experienced editors when they have time)
  • That we are aiming to make the wording a slightly more clear expression of that intention
  • That it is not the aim to make this policy less ambiguous in the sense of being prescriptive in more areas than intended do this by trying to cover all possibilities
  • That one major aim is to make sure WP:V wording does not encourage the use of WP:V to trump WP:NPOV considerations.
  • That one more specific aim, within that major aim, is that WP:V wording does not encourage the use of WP:V to include materials only because they are verifiable.--Andrew Lancaster (talk) 16:55, 27 September 2011 (UTC)[reply]
  • First item: Agree
  • Second item: Agree
  • Third item: It will take me a few days to figure out what that says :-)
  • Fourth item: IMHO a non-issue and thus not particularly an aim. (?)
  • Fifth item: Strongly agree
Sincerely, North8000 (talk) 17:43, 27 September 2011 (UTC)[reply]
Concerning the third point, see re-write in blue. Fourth point's aim is to generalize, so in other words it should include five as one case.--Andrew Lancaster (talk) 19:00, 27 September 2011 (UTC)[reply]
OK, so I not only agree with #1, #2 and #5, but I agree that they are core topics and rationales for the proposed change. I have no disagreement with #3 and #4 but IMHO they seem more like sidebar notes than core topics. But that's just my opinion. North8000 (talk) 10:42, 28 September 2011 (UTC)[reply]

Recent edits in rationale draft

edit

I think that we must acknowledge that a strong theme of wp:ver is to not only state that verifiability is a requirement for inclusion, but to be very emphatic and categorical about it. Indeed, I think that the "emphatic and categorical" part is the main point of the folks supporting the status quo. I think that my word "absolute" which Andrew removed may not have been a good way to say this, but I think that the point must be made. Sincerely, North8000 (talk) 11:27, 28 September 2011 (UTC)[reply]

Diff of my edit: [2]--Andrew Lancaster (talk) 12:05, 28 September 2011 (UTC)[reply]
This concern is why my proposal includes the line: "Note, however, that while verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article.)" The language may not be perfect, but I think it adresses the concern clearly and succinctly. Blueboar (talk) 12:16, 28 September 2011 (UTC)[reply]
Yes, and the rationale should match what we are proposing, which is indeed going to be your draft, or something very close to it.--Andrew Lancaster (talk) 17:11, 28 September 2011 (UTC)[reply]
Blueboar, Whether or not the proposal is accepted as a whole, would you say that at least this part would be an improvement to the current version of the policy page if it was proposed alone? --Bob K31416 (talk) 18:05, 28 September 2011 (UTC)[reply]

A different approach

edit

I am going to suggest a different approach to the rational. I think it should be a much more concrete statement... laying out a) what are the concerns, and b) how the proposal attempts to resolve those concerns. so... I would suggest something along the following lines (again this is an approach, not a proposal for exact wording... call it a very rough draft):

The first line of the policy currently reads: "The threshold for inclusion in Wikipedia is verifiability, not truth—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true." Several editors have raised concerns about this sentence, and especially the phrase "Verifiability, not truth". Some editors strongly feel the phrase should be removed from the policy entirely. Others feel just as strongly that the phrase is vital to the policy and must remain. After months of debate (and numerous polls and RfCs) a compromise proposal has emerged that takes into account the concerns of both sides.
  • Background: The concept that truth is not the threshold for inclusion in Wikipedia was added for a specific reason - to support WP:NOR in saying that material should not be included in Wikipedia unless there is a source that directly supports it. At the time that this language was added, we had a serious problem with editors wishing to add unverifiable material purely on the grounds that "it's true". This was especially a problem when it came to editors attempting to "prove" their pet fringe theory. However, over time, the sentence has taken on meanings that were not part of this original intent.
  • Concern: The sentence can be misconstrued to mean that any material that appears in a source must be included... simply because it is verifiable. This misinterpretation is in conflict with several other policy and guideline statements (especially the WP:Undue weight section of WP:NPOV)
    • How the proposal resolves this concern: The proposal adds an explanation that "while verifiability is a requirement for inclusion, it is not a guarantee of inclusion", and it notes that other policies and guidelines can affect inclusion.
  • Concern: The sentence can be misconstrued to mean that we may not discuss the possibility that verifiable information is untrue (ie that the source may contain an error). That we must accept what a published source says simply because it is published, even if we have reason to believe that the material is inaccurate or that the source is less than reliable. This was never the intent of the sentence. We are frequently called upon to make editorial judgment calls as to the reliability of specific sources when it comes to specific information. A source may be reliable for one statement, but not reliable for some other statement. We also have to make editorial judgment calls as to the relative reliability of one source when compared to others. We do want the information we present in Wikipedia to be accurate.
  • Counter Concern: However, as NPOV notes, we can not simply omit significant viewpoints simply because we disagree with them (or even because the majority of sources disagree with them). Sometimes we should discuss facts and opinions that may be untrue, because doing so gives the reader a complete picture of disagreement among the sources.
    • How the proposal resolves these concerns: The proposed language acknowledges that inclusion of potentially untrue information depends on context. That we can not make a firm one-size-fits-all rule on this. The proposal points out that the question of whether to include controversial and potentially untrue material is a complex one, that involves applying multiple policies and guidelines at the same time. It points the reader to those other policies and guidelines.
  • Concern: Introducing the concept of "truth" in the lede is a distraction and can confuse new editors. The lede paragraphs should focus purely on explaining concept of Verifiability without introducing secondary concepts. While it is important for the policy to discuss the issue of truth/untruth, this should be done in the body of the policy and not in the very first sentence.
    • How the proposal resolves the concern: the issue of truth is moved out of the lede and into to its own separate section.

Conclusions: The goal of this proposal is not to change the meaning of the policy in any way... but rather to better clarify existing policy and avoid misinterpretation. The concepts behind the phrase "Verifiability, not truth" are valid and important, and they should remain part of the policy. But they are very complex concepts that need to be more clearly explained.

  • Blueboar, I agree with most of what you say and I think this approach will be productive. However, I think the "conclusions" section needs quite a fundamental rethink. I'm sure that editors will say that this proposal does change the meaning of the policy. And they would be quite right to say so. To my mind, the point of your proposal is that it transforms what the policy currently says into what it was always meant to say. But I believe that we must accept that the new version with the clarifications does change the meaning of the policy. It changes its emphasis as well, and there will be editors who confuse this change of emphasis with a change of meaning.

    I think the "conclusions" section should say that the goal of the proposal is to reduce the potential for accidental or tendentious misapprehension of the policy.—S Marshall T/C 15:25, 28 September 2011 (UTC)[reply]

Hmmm... your point about the proposal "transforming the policy into what it was always meant to say" is, I think, valid. In many ways the proposal is an attempt to refocus the V not T phrase back its original intent while removing or clarifying the misconceptions that have grown up around it since it was added. In that context, my intent was not to "change the meaning", but I could see how others might think it does. To tell the truth, I thought the "Conclusions" section was poorly worded, even as I wrote it... but I felt that some sort of "summation" was needed to close the whole thing off, so I simply wrote what came to mind. I am also sure that I have left out some of the concerns and resolutions that could be added. Again... I was focused on formulating a different way to organize and lay out the rational... and was not at all focused on exact wording. Blueboar (talk) 15:45, 28 September 2011 (UTC)[reply]
Re S Marshall's comment "I'm sure that editors will say that this proposal does change the meaning of the policy. And they would be quite right to say so." - How does it specifically change the meaning of policy? --Bob K31416 (talk) 15:49, 28 September 2011 (UTC)[reply]
The current version says that sourced material may be added irrespective of whether it's true, and follows that with a full stop. Blueboar's version says that sourced material may be added subject to editorial judgment.—S Marshall T/C 16:46, 28 September 2011 (UTC)[reply]

Blueboar, that is a very good presentation. My thought would be to make that the centerpiece of the rationale, but not the entire rationale. I think that several things in the current draft are sort of "outside" of it and useful for retention. But there is one thing that I suggest adding to / is a natural part of the centerpiece, and that is pointing out the new context explanation that you wrote for "not truth" which is in the proposal.....I think that it is a crucial part of the proposal, perhaps the most important part of it. North8000 (talk) 16:11, 28 September 2011 (UTC)[reply]

I copied it into the rationale draft, as a section. Still think it needs to note the new context phrase for "not truth" North8000 (talk) 10:18, 29 September 2011 (UTC)[reply]
Nutshell: The goal of this proposal is to reduce the scope for good faith or tendentious editors to experience real or feigned confusion about what WP:V means, what it allows, and what it prevents.

The first line of the policy currently reads: "The threshold for inclusion in Wikipedia is verifiability, not truth—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true." There are substantial concerns about this sentence (in previous RFCs, about 50% of editors asked felt the first sentence was unsatisfactory), and particularly the phrase "Verifiability, not truth". Some editors feel the phrase should be removed from the policy entirely and the wording should focus on what WP:V is, not what it is not. Others feel that the phrase is vital. After months of debate (and numerous polls and RfCs) a new proposal has emerged, to keep this phrase but move it to its own section and clarify it.

  • Background: The concept that truth is not the threshold for inclusion in Wikipedia was added for a specific reason - to support WP:NOR in saying that material should not be included in Wikipedia unless there is a source that directly supports it. At that time, we had a serious problem with editors wishing to add unverifiable material purely because "it's true". This attitude was useful to editors attempting to "prove" their pet fringe theory. However, as WP:V has changed over time, the sentence has been moved earlier and earlier in the policy, and stripped of its original context. It has therefore taken on meanings that were never part of this original intent.
  • Concern: The sentence can be misconstrued to mean that any material that appears in a source must be included... simply because it is verifiable. This misinterpretation is in conflict with several other policy and guideline statements (especially the WP:Undue weight section of WP:NPOV) but several examples of this misinterpretation happening in practice have been provided.
    • How the proposal resolves this concern: The proposal adds an explanation that "while verifiability is a requirement for inclusion, it is not a guarantee of inclusion", and it notes that other policies and guidelines can affect inclusion.
  • Concern: The sentence can be misconstrued to mean that we may not discuss the possibility that verifiable information is untrue (ie that the source may contain an error). That we must accept what a published source says simply because it is published, even if we have reason to believe that the material is inaccurate or that the source is less than reliable. This was never the intent. We often need to make editorial judgment calls about the reliability of specific sources when it comes to specific information. A source may be reliable for one statement, but not reliable for some other statement. We also have to make editorial judgment calls as to the relative reliability of one source when compared to others. As Jimbo Wales puts it, "We are not transcription monkeys." We do want the information we present in Wikipedia to be accurate, as far as possible.
  • Counter Concern: However, as NPOV notes, we can not omit significant viewpoints just because we disagree with them (or even because the majority of sources disagree with them). Sometimes we should discuss facts and opinions that may be untrue, because doing so gives the reader a complete picture of disagreement among the sources.
    • How the proposal resolves these concerns: The proposed language acknowledges that inclusion of potentially untrue information depends on context. We can not make a firm one-size-fits-all rule on this. The proposal points out that the question of whether to include controversial and potentially untrue material is a complex one, that involves applying editorial judgment. It points the reader to other policies and guidelines that may help.
  • Concern: Outside the world of Wikipedia, the natural meaning of "verifiability" concerns truth. Introducing the phrase "not truth" in the lede is distracting and confusing particularly for new editors. The lede should focus purely on explaining the Wikipedian idea of Verifiability without introducing secondary concepts. While it is important for the policy to discuss the issue of truth/untruth, this belongs in the body of the policy.
    • How the proposal resolves the concern: The issue of truth is moved out of the lede and into to its own separate section.

Being a product of two editors (yourself and Blueboar) let's plan on swapping this one into the draft. But, comments: While "feigned" in the first sentence is spot-on, I'm thinking that starting with a criticisim word like that might be too negative sounding. Also I think that this one could use that "context" note that I added to Blueboars. North8000 (talk) 11:44, 29 September 2011 (UTC)[reply]

Not bad... but I have a few comments... a) "...but several examples of this misinterpretation happening in practice have been provided." If we are going to say this, we should probably link to at least one or two of these examples. b) Before the "transcription monkeys" quote, we should probably talk about a few other forms of editorial judgment that can affect inclusion... such as deciding whether a bit of information is relevant to the topic, or trivial. c) I suggest we not raise the concern about the meaning of "verifiability" in the outside world... to be honest and blunt about this, most of our policies have developed unique Wikipedia specific definitions of terms (for example we have developed our own definitions for the terms "reliability", "original research", "primary/secondary source", "notability" and "neutrality".) The editors who routinely work on policy pages (ie the editors who are likely to respond and comment on our RfC) know this, and are content with this. The fact that Wikipedia has developed its own meaning of common words may or may not be a flaw... but if so, it is a flaw that goes back to the earliest days of Wikipedia and is not about to change now. Raising the issue is unlikely to be seen as a valid concern. Also, I don't see how moving "V not T" out of the lede and into its own section actually resolves that particular concern. I can certainly say that it was not a concern I was trying to address when I crafted my compromise. I think raising the issue will simply be a distraction from more convincing concerns and resolutions. Blueboar (talk) 12:34, 29 September 2011 (UTC)[reply]
So, in this editing, do we start with Blueboar's version, or the Blueboar-edited-By-S Marshall version? Might be simpler to start with Blueboar's version. And edit it in place at the draft rationale. Either way, just do it...BRD. Or more like BDBDBDBD to keep this rolling. North8000 (talk) 13:55, 29 September 2011 (UTC)[reply]

Arbitrary break - side by side listing

edit

I hope no-one minds, but can we lay this out side by side?--Andrew Lancaster (talk) 15:00, 29 September 2011 (UTC)[reply]

Cool North8000 (talk) 15:34, 29 September 2011 (UTC)[reply]
Blueboar S Marshall
[no nutshell] Nutshell: The goal of this proposal is to reduce the scope for good faith or tendentious editors to experience real or feigned confusion about what WP:V means, what it allows, and what it prevents.
The first line of the policy currently reads: "The threshold for inclusion in Wikipedia is verifiability, not truth—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true." Several editors have raised concerns about this sentence, and especially the phrase "Verifiability, not truth". Some editors strongly feel the phrase should be removed from the policy entirely. Others feel just as strongly that the phrase is vital to the policy and must remain. After months of debate (and numerous polls and RfCs) a compromise proposal has emerged that takes into account the concerns of both sides. The first line of the policy currently reads: "The threshold for inclusion in Wikipedia is verifiability, not truth—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true." There are substantial concerns about this sentence (in previous RFCs, about 50% of editors asked felt the first sentence was unsatisfactory), and particularly the phrase "Verifiability, not truth". Some editors feel the phrase should be removed from the policy entirely and the wording should focus on what WP:V is, not what it is not. Others feel that the phrase is vital. After months of debate (and numerous polls and RfCs) a new proposal has emerged, to keep this phrase but move it to its own section and clarify it.
*Background: The concept that truth is not the threshold for inclusion in Wikipedia was added for a specific reason - to support WP:NOR in saying that material should not be included in Wikipedia unless there is a source that directly supports it. At the time that this language was added, we had a serious problem with editors wishing to add unverifiable material purely on the grounds that "it's true". This was especially a problem when it came to editors attempting to "prove" their pet fringe theory. However, over time, the sentence has taken on meanings that were not part of this original intent.
  • Concern: The sentence can be misconstrued to mean that any material that appears in a source must be included... simply because it is verifiable. This misinterpretation is in conflict with several other policy and guideline statements (especially the WP:Undue weight section of WP:NPOV)
    • How the proposal resolves this concern: The proposal adds an explanation that "while verifiability is a requirement for inclusion, it is not a guarantee of inclusion", and it notes that other policies and guidelines can affect inclusion.
  • Concern: The sentence can be misconstrued to mean that we may not discuss the possibility that verifiable information is untrue (ie that the source may contain an error). That we must accept what a published source says simply because it is published, even if we have reason to believe that the material is inaccurate or that the source is less than reliable. This was never the intent of the sentence. We are frequently called upon to make editorial judgment calls as to the reliability of specific sources when it comes to specific information. A source may be reliable for one statement, but not reliable for some other statement. We also have to make editorial judgment calls as to the relative reliability of one source when compared to others. We do want the information we present in Wikipedia to be accurate.
  • Counter Concern: However, as NPOV notes, we can not simply omit significant viewpoints simply because we disagree with them (or even because the majority of sources disagree with them). Sometimes we should discuss facts and opinions that may be untrue, because doing so gives the reader a complete picture of disagreement among the sources.
    • How the proposal resolves these concerns: The proposed language acknowledges that inclusion of potentially untrue information depends on context. That we can not make a firm one-size-fits-all rule on this. The proposal points out that the question of whether to include controversial and potentially untrue material is a complex one, that involves applying multiple policies and guidelines at the same time. It points the reader to those other policies and guidelines.
  • Concern: Introducing the concept of "truth" in the lede is a distraction and can confuse new editors. The lede paragraphs should focus purely on explaining concept of Verifiability without introducing secondary concepts. While it is important for the policy to discuss the issue of truth/untruth, this should be done in the body of the policy and not in the very first sentence.
    • How the proposal resolves the concern: the issue of truth is moved out of the lede and into to its own separate section.
 *Background: The concept that truth is not the threshold for inclusion in Wikipedia was added for a specific reason - to support WP:NOR in saying that material should not be included in Wikipedia unless there is a source that directly supports it. At that time, we had a serious problem with editors wishing to add unverifiable material purely because "it's true". This attitude was useful to editors attempting to "prove" their pet fringe theory. However, as WP:V has changed over time, the sentence has been moved earlier and earlier in the policy, and stripped of its original context. It has therefore taken on meanings that were never part of this original intent.
  • Concern: The sentence can be misconstrued to mean that any material that appears in a source must be included... simply because it is verifiable. This misinterpretation is in conflict with several other policy and guideline statements (especially the WP:Undue weight section of WP:NPOV) but several examples of this misinterpretation happening in practice have been provided.
    • How the proposal resolves this concern: The proposal adds an explanation that "while verifiability is a requirement for inclusion, it is not a guarantee of inclusion", and it notes that other policies and guidelines can affect inclusion.
  • Concern: The sentence can be misconstrued to mean that we may not discuss the possibility that verifiable information is untrue (ie that the source may contain an error). That we must accept what a published source says simply because it is published, even if we have reason to believe that the material is inaccurate or that the source is less than reliable. This was never the intent. We often need to make editorial judgment calls about the reliability of specific sources when it comes to specific information. A source may be reliable for one statement, but not reliable for some other statement. We also have to make editorial judgment calls as to the relative reliability of one source when compared to others. As Jimbo Wales puts it, "We are not transcription monkeys." We do want the information we present in Wikipedia to be accurate, as far as possible.
  • Counter Concern: However, as NPOV notes, we can not omit significant viewpoints just because we disagree with them (or even because the majority of sources disagree with them). Sometimes we should discuss facts and opinions that may be untrue, because doing so gives the reader a complete picture of disagreement among the sources.
    • How the proposal resolves these concerns: The proposed language acknowledges that inclusion of potentially untrue information depends on context. We can not make a firm one-size-fits-all rule on this. The proposal points out that the question of whether to include controversial and potentially untrue material is a complex one, that involves applying editorial judgment. It points the reader to other policies and guidelines that may help.
  • Concern: Outside the world of Wikipedia, the natural meaning of "verifiability" concerns truth. Introducing the phrase "not truth" in the lede is distracting and confusing particularly for new editors. The lede should focus purely on explaining the Wikipedian idea of Verifiability without introducing secondary concepts. While it is important for the policy to discuss the issue of truth/untruth, this belongs in the body of the policy.
    • How the proposal resolves the concern: The issue of truth is moved out of the lede and into to its own separate section.
Conclusions: The goal of this proposal is not to change the meaning of the policy in any way... but rather to better clarify existing policy and avoid misinterpretation. The concepts behind the phrase "Verifiability, not truth" are valid and important, and they should remain part of the policy. But they are very complex concepts that need to be more clearly explained. [no conclusions]

Arbitrary break (Opinions section?)

edit

My opinions (using section numbers that follow the table divisions:

  • Section 1: Stick with Blueboar's original (no nutshell) Nutshell is too hard hitting, could be counterproductive.
  • Section 2 Go with Andrew'sS Marshall's revisions Except one correction: the approx 50% wanted "not truth" completely gone.
  • Section 3 Go with Andrew's S Marshall's revisions
  • Section 4 Stick with Blueboar's original. It is correct and reassuring, and thus effective.

North8000 (talk) 17:09, 29 September 2011 (UTC)[reply]

  • I'm anxious that there should be a one-sentence "nutshell". That doesn't mean it has to be the one I suggested above, although I do think that thought needs to be expressed somewhere in the proposal we put to RFC.—S Marshall T/C 17:58, 29 September 2011 (UTC)[reply]
Cool. My concern was only the wording. IMHO it needs to be upbeat rather than talking about bad faith behavior in the first sentence. North8000 (talk) 18:20, 29 September 2011 (UTC)[reply]
I should mention that the proposed changes were not mine. I am thinking I prefer Blueboar's proposal, despite sympathies for the change proposals. The idea that a nutshell might be good if shorter seems logical to me also.--Andrew Lancaster (talk) 19:37, 29 September 2011 (UTC)[reply]
Ooops, I credited the wrong person. North8000 (talk) 19
48, 29 September 2011 (UTC)
I like the idea of a re-worded nutshell; it has to be one that soothes readers before they begin, rather than getting their hackles up, though. I'm pretty much in agreement with North on this / the rest. Pesky (talk …stalk!) 03:03, 30 September 2011 (UTC)[reply]
So I'll modify the draft per the above. Not pushing any idea, just trying to keep things moving. So now we need somebody to write a nutshell per above feedback. ? North8000 (talk) 13:56, 30 September 2011 (UTC)[reply]

A live example of how "verifiability, not truth" is used to argue for inclusion of content

edit

See Talk:Metrication in the United Kingdom#ASDA. Two editors want to remove the first half of the following sentence from the article: "Following the results of a survey of their customers in early 2011 - which concluded that 70% of them would prefer products to be labelled in imperial units - the Asda supermarket chain are experimenting with selling produce in round imperial measures again." Their reason is that surveys often have a poor quality, and without knowing the methodology and the questions asked it's impossible to assess one. A single editor uses "verifiability, not truth" as his key argument why the sentence must be kept. (The user has now also added information about a similar survey conducted by Tesco.) Hans Adler 16:30, 28 September 2011 (UTC)[reply]

Hans, are you claiming that the editor misapplied the policy by including why the super markets took the actions regarding experiments with standard measures in Great Britain, in an article about the use of the measurements systems in Great Britain? If you are, how so? Alanscottwalker (talk) 18:23, 28 September 2011 (UTC)[reply]
Yes. The details of the survey results (not the fact that there was a survey that motivated ASDA) are a borderline case of relevance to the article. If you claim that you can force random barely related stuff into an article just because it's 'verifiable', then, even if it's true, you are abusing the policy. Hans Adler 22:11, 28 September 2011 (UTC)[reply]
This is a perfect example of why the change is needed. The mis-application was essentially saying that the flaws in the material (and potential exclusion based on those flaws) can't be discussed because of "not truth". North8000 (talk) 18:29, 28 September 2011 (UTC)[reply]
No wonder you keep failing at trying to change this. What should be discussed about them? The supermarket made a measurement experiment based on a survey, according to the source. In what universe, without a contradictory reliable source, could such a thing be considered untrue. Alanscottwalker (talk) 18:39, 28 September 2011 (UTC)[reply]
Moreover, what exactly is wrong with the application of the policy: Editor one: I know this is not (my not be) true because of what I know. Editor two: that's nice but irrelevant, see WP:V. Alanscottwalker (talk) 18:54, 28 September 2011 (UTC)[reply]
What's relevant here is that the discussion of the alleged flaws in the material and potentially excluding it should be allowed to occur without someone improperly using "not truth" to block or trump that discussion. It is not to suggest or dictate the the outcome of that discussion. North8000 (talk) 18:56, 28 September 2011 (UTC)[reply]
Your final comment gives another example of the same thing and need for change/clarification. You are basically saying that verifiability gives one a mandate to force inclusion of the material, which is not what wp:ver says. It says that verifiability is a requirement (not a mandate) for inclusion. North8000 (talk) 19:00, 28 September 2011 (UTC)[reply]
How does it "block, the discussion." Your claim is that someone should be barred from arguing WP:V bars certain content changes? Such an argument is absurd on its face. That's its purpose. As for your "forced inclusion" point. Obviously, that's not what I wrote. But editors do need to start somewhere in deciding on inclusion and that point is at verifiability, not truth. Alanscottwalker (talk) 19:36, 28 September 2011 (UTC)[reply]
I understand that that is not what you wrote, nor did you intend to imply or do anything that sounds like the words I used. But my point is that it is nevertheless a premise of your statements and included in your statements. In essence what you are saying is: "If I want it in, and it is properly sourced, per wp:ver it stays in, end of story....no other consideration or discussion matters." This is not a negative reflection on you, it is a reflection on the policy wording problem that we are trying to solve. North8000 (talk) 20:29, 28 September 2011 (UTC)[reply]
It sounds to me like the other editor is correct. Excluding material solely because some Wikipedia editor thinks the survey involved a flawed sample is not authorized by any policy. There's not one word in the policies that says sources have to describe their methodology so that some Wikipedia editor can decide whether the source's conclusions are warranted, or that authorizes Wikipedians to reject sources because the Wikipedian disagrees with their methodology. It's not okay to cherry-pick sources, and that's the only possible outcome here. Either the source is reliable for the statement, or it's not—and the definition of "reliable" is over at WP:RS, where "making it possible for an editor to dissect the methodology and decide whether the editor agrees with the source" is not mentioned.
North, you're missing the point. The material meets all of the content policies, and HiLo is saying "Whether we [=Wikipedia editors] believe or trust the survey is completely relevant. At this point, I don't trust the survey."—in other words, HiLo doesn't care whether it's not only verifiable but also DUE and so forth, and he doesn't care whether his rejection of verifiable material is based strictly on a direct violation of WP:NOR: he personally disagrees with the independently reported survey results, and therefore he refuses to have the information in the article. Since when are we writing an encyclopedia that is filtered according certain editors' opinions? WhatamIdoing (talk) 20:41, 28 September 2011 (UTC)[reply]
Evaluating sources is exactly what encyclopaedists should be doing. If the source is unreliable, and I mean "unreliable" in its natural meaning, then it shouldn't be used.—S Marshall T/C 21:54, 28 September 2011 (UTC)[reply]
Just because one source says something is not even nearly enough sometimes (let alone getting into the whole "truth" thing). In working on a GA recently, other editors and I often had to compare many sources that were equally reliable to determine what the consensus among them was when it came to deciding what to include. Some sources completely contradicted the others on certain details, and agreed sometimes as well; but all were certainly reliable. So something backed by a consensus among reliable sources is even more worthy of inclusion. That's when "cherry-picking" certain details actually must be done. Doc talk 22:04, 28 September 2011 (UTC)[reply]
First: The fact that ASDA is experimenting with a return to pre-metric units is verifiable and clearly extremely relevant to the article, so it should be included. It is also verifiable that ASDA made a survey and how it reported the results of this survey, but it is not clear how relevant it is to the article. This, and only this, is what this dispute is about. I do not have a strong opinion on this.
For all we know, this could be a survey of extremely poor quality, or the results of this survey might be misrepresented in how ASDA reports them. This kind of thing happens all the time. In fact, ASDA clearly decided to return to pre-metric as a publicity stunt. It's entirely possible that this decision was taken first, and then ASDA decided it needed a survey proving that a majority of its customers want that -- in order to gain sympathy from older customers without losing sympathy of younger customers. So maybe they ordered a survery, giving more or less clear instructions what the survey was supposed to say.
Many readers reading about this survey will take its result as a proxy for what the UK population as a whole want. But as a general rule, we simply do not mention poor or dubious surveys just for their results, nor do we mention surveys about which we know so little that it's not at all clear whether the reporting is at least vaguely accurate. There are way too many such surveys, of contradicting or appearing to contradict each other, for the inclusion of random ones among them to make any sense.
This leaves the survey's relevance to the noteworthy ASDA decision. That's the strongest reason for mentioning the survey, but for this the middle part ("- which concluded that 70% of them would prefer products to be labelled in imperial units -") can be removed without any loss, and should be removed due to hyperprecision. Hyperprecision is when you state something so as to create the illusion of a considerably higher confidence than is warranted. E.g. when you say that New York City had a population of "19,378,102 in 2010", then there is clearly something wrong. (Whereas the statement that the city had a population of "19,378,102 according to the 2010 census" is merely more precise than necessary, but not hyperprecise.)
This is just a normal case of crap gatekeeping. A lot of 'verifiable' information is irrelevant crap, and we have no obligation whatsoever to dump it into our articles and turn them into unreadable messes. Crap gatekeeping is a matter of editorial judgement. As a general principle, when some editors think they can get around a consensus to exclude something as not sufficiently relevant, merely by pounding on the fact that it's 'verifiable', then they are unlikely to succeed with that in the long run, but in the short run a lot of disruption is caused because they miscalculate the odds. (I am not saying that is happening here: there are too few editors involved so far, and it's a borderline case that is unlikely to lead to a very strong consensus either way.) Hans Adler 22:11, 28 September 2011 (UTC)[reply]

This discussion confirms that the policy should not change - the proposal is not even a thinly veiled disguise for editors to claim they know the truth, so they don't like what a source says, so it should be ignored. Alanscottwalker (talk) 22:18, 28 September 2011 (UTC)[reply]

Can I get an Amen? Doc talk 22:22, 28 September 2011 (UTC)[reply]

Amen. As for the claim of it being irrelevantL Beside the fact that objection was not even made by the people who wanted to hide the reason why the supermarket was experimenting with imperial measures, that's dealt with by other policies. Alanscottwalker (talk) 22:30, 28 September 2011 (UTC)[reply]

I think that the fact that there is a particular dispute involved, and one of the folks involved is active here is confusing the matter. Alanscottwalker, you keep ignoring the main point being made, and incorrectly restating it as something completely different. NOBODY is saying that wp:ver should (in any current of future form) dictate that the material be removed, but you keep misstating the argument (in a straw man form) as being that. All that we are saying is that wp:ver should be worded so as to not be mis-usable to terminate the discussion on potential exclusion of the material. North8000 (talk) 22:56, 28 September 2011 (UTC)[reply]
Your the one deliberately mistating what other people say, as you admitted, earlier in this thread. Nothing of which, I said. You don't want people to use WP:V, to tell other people they are violating WP:V, to keep out or include content. You therefore want to change the policy. You've made that clear. The problem is, it's a bad idea and completely contrary to the current "core" policy. Alanscottwalker (talk) 23:09, 28 September 2011 (UTC)[reply]
WP:V was originally never intended as an argument to include material just because it's 'verifiable'. That's just an abuse of the policy that sometimes does no harm, and sometimes causes disruption because some editors think once they have proved something is 'verifiable', good-faith serious concerns about accuracy, relevance or NPOV are simply irrelevant. When abuse of a policy becomes rampant and starts hurting the project, then of course it must be rephrased to stop the abuse. That the abusers of the policy claim such a clarification is a substantial change to the policy is hardly surprising. Hans Adler 23:38, 28 September 2011 (UTC)[reply]
Hans. Given your statement above, its clear you intend greater harm to the project. You claim to know how the general public will read: "Many readers reading about this survey will take its result as a proxy for what the UK population as a whole want." Not only is such a claim unsupported, its directly inimical to the project. Alanscottwalker (talk) 23:50, 28 September 2011 (UTC)[reply]
As for the claim that the rule doesn't provide a reason to include something, that's wrong on its face. It's one of the core requirements for inclusion (although not the only one).Alanscottwalker (talk) 23:57, 28 September 2011 (UTC)[reply]
  • I agree with Hans (without prejudice to the content question at issue in the above article). Whether or not something verifiable should be included or not is a matter for WP:DUE (which is a part of WP:NPOV), WP:RS/WP:RSN, and, in the case of living people, WP:BLP. WP:V has nothing to say on that, and should not have anything to say on that. WP:V defines basic requirements for what may be included. WP:NPOV defines what should be included to arrive at a neutral article. Every time someone is using "verifiability, not truth" as an argument why something should be included, or should not be deleted, they are misusing it. --JN466 02:16, 29 September 2011 (UTC)[reply]
  • Would all the editors here agree that, before Alan came on the scene of the example, the editor who was using "verifiability, not truth", was using it with the incorrect belief that verifiability was all that was needed for inclusion? --Bob K31416 (talk) 02:33, 29 September 2011 (UTC)[reply]
No. The fact that ASDA (and by extension Tesco who are bigger) is a very large retailer was mentioned as making the material DUE. tbh, they're so big and Tesco received national and international coverage (and ASDA featured in Which?), it's pretty obvious the information is DUE, if it counts as verifiable. The dispute is over what the sources can support.VsevolodKrolikov (talk) 02:53, 29 September 2011 (UTC)[reply]
Having engaged one of the editors, I think this is a really odd dispute. The editor is questioning the methods of the surveys, not that they were done or influenced policy. An equivalent case would be refusing to mention any details about the Zinoviev letter in an article on Ramsay MacDonald on the grounds that the letter was a fake. It's surprising, because it's an experienced editor.VsevolodKrolikov (talk) 03:11, 29 September 2011 (UTC)[reply]
It's not as if the surveys themselves were notable or anything. That kind of thing is mass-produced. To state a survey's results in an article on metrication in the UK it must surely be among the top surveys on the topic in terms of reliability and/or relevance. (Unless it's independently notable because for whatever reason the survey itself is discussed extensively in the news.) That's clearly not the case here. I have proposed a brief mention of the survey as what ASDA said motivated it to its step, but without indicating the result. The gist of the result is implicit from ASDA's action, and the details fall under WP:UNDUE for the article because the survey is neither particularly reliable nor particularly relevant (as it's only about ASDA customers). There is nothing odd about removing undue weight material from an article, especially when it tends to support one POV on a contentious topic. Hans Adler 08:45, 29 September 2011 (UTC)[reply]
To Bob's question. No. for the reasons more fully stated by WhatamIdoing, above and Vsevolod. The editor did not argue that his only reason for inclusion was verifiability. Alanscottwalker (talk) 03:24, 29 September 2011 (UTC)[reply]
I've had a look over other sections of that talkpage. It really isn't worth bringing the dispute here as an example. There are two editors with open (userpage infobox) POVs regarding metrication, pushing some rather odd stuff, such as London hosting the Olympic games in 2012 being evidence of British metrication (seriously!). Bad cases make bad law etc....VsevolodKrolikov (talk) 17:55, 29 September 2011 (UTC)[reply]


Using the proposal for this situation

edit

From the above debate, even for the editors here who are familiar with the current policy, the present policy is not able to settle their differences regarding the example. How would the proposal do in this situation? --Bob K31416 (talk) 23:19, 28 September 2011 (UTC)[reply]

The proposal would say that as long as sourcing requirements are met, wp:ver would not further weigh in on or be usable to weigh in on this dispute. Discussions could progress to settle it by other means, such as discussions and wp:npov. North8000 (talk) 23:27, 28 September 2011 (UTC)[reply]
WP:NPOV is one of the three core policies, along with WP:V and WP:NOR. "These three core policies jointly determine the type and quality of material that is acceptable in Wikipedia articles. Because these policies work in harmony, they should not be interpreted in isolation from one another, and editors should try to familiarize themselves with all three." Doc talk 23:32, 28 September 2011 (UTC)[reply]
Quite. The current rule is required to weigh in at every step, just like the other core policies are.Alanscottwalker (talk) 00:11, 29 September 2011 (UTC)[reply]
But I don't think the answer was actually given to the question. What "sourcing requirement" was met or unmet in the situation asked about? Alanscottwalker (talk) 00:11, 29 September 2011 (UTC)[reply]
Who cares? If something is likely to be challenged as "untrue", back it up with reliable sources that agree with each other. Go nuts. We don't determine what is true or not, we just report what is accepted amongst the greater consensus of reliable (third-party, ideally) sources that are already out there. Doc talk 00:22, 29 September 2011 (UTC)[reply]
And sometimes not, but we must care because that still doesn't answer the situation, where a reliable source says something and the objection is not that it disagrees with another reliable source but that it is untrue because an editor says it is untrue. Alanscottwalker (talk) 01:00, 29 September 2011 (UTC)[reply]
This is going in circles. Folks seem to be repeating the questions but not listening to the answers which admittedly take a bit a logical head scratching to understand. But briefly, Alanscottwalker, your viewpoint material was sourced, so the wp:ver does not cause it to be removed. After that, wp:ver is and should be silent / not operative on the issue. North8000 (talk) 00:40, 29 September 2011 (UTC)[reply]
So, its not operative when someone comes along and says, "We should remove this because I know its not true." Is that what you are saying?Alanscottwalker (talk) 00:48, 29 September 2011 (UTC)[reply]
Yes, if you are talking about wp:ver. Plenty of other policies and considerations would be operative on that, but not wp:ver. Plus, if the only basis they gave was them saying just "because I know" per your example, that argument has a wp:snow chance of carrying weight in the discussion. North8000 (talk) 01:16, 29 September 2011 (UTC)[reply]
Agree that should be given no weight because the core policy WP:V says so. Alanscottwalker (talk) 01:23, 29 September 2011 (UTC)[reply]

I think this is not an example of WP:V being misused at all. It looks like editors talking at cross-purposes (Were the surveys done and influenced company policy, vs. were the surveys done properly). The real issue for one side is WP:RS, and trying to demonstrate that using the surveys as evidence for support for imperial is unwarranted because of the quality of the source - but it's not clear that that is what the other editors are trying to do at all.

I sense there is still some confusion on the part of some editors here about the difference between criteria for inclusion and criteria for exclusion. "Include it because it's true" is not the same as "exclude it because it's false". Exclusion is not mentioning, rather than saying something is not true. If the source looks dodgy, we don't mention, or we attribute. Exclusion is a matter of WP:RS and WP:RSN.VsevolodKrolikov (talk) 02:06, 29 September 2011 (UTC)[reply]

Well said. North8000 (talk) 02:15, 29 September 2011 (UTC)[reply]
Agree that we exclude if the source is unreliable, but definitely not because it is either true or false; it may well be true, we don't know. More importantly, we don't have a reliable source. That's what WP:V requires. Alanscottwalker (talk) 02:21, 29 September 2011 (UTC)[reply]
Re VsevolodKrolikov's comment "I sense there is still some confusion on the part of some editors here about the difference between criteria for inclusion and criteria for exclusion." - How would the present version of WP:V compare to the proposed version, as far as helping relieve some of this confusion? --Bob K31416 (talk) 02:26, 29 September 2011 (UTC)[reply]
I'm not sure it does. I don't much like the "Assertions of truth and untruth" paragraph, as I think it really should only deal with verifiability, not truth (and have that as its title), with a couple of illustrations of how/why personal knowledge (eg of a place or a person) isn't good enough for inclusion. At the moment the section puts truth and untruth together. I suggested a while back that the policy should contain a section on "What to do if you think inaccurate information has been included". It would explain about asking for citations, and if you think a given source is mistaken, take it up on the talkpage and at RSN.VsevolodKrolikov (talk) 02:46, 29 September 2011 (UTC)[reply]
Most editors (names withheld) should just really never even entertain the idea of editing policy. When you see how all three of these policies intertwine almost effortlessly to support each other: you'll know it. You will be assimilated. Failing that, you'll most likely be informed by someone who is assimilated on how things work for inclusion of material on Wikipedia. Doc talk 02:59, 29 September 2011 (UTC)[reply]
Doc, That's an interesting message and link (assimilated) re The Borg. Do you consider yourself part of a Wikipedia Borg, and is there a Borg Queen? --Bob K31416 (talk) 04:06, 29 September 2011 (UTC)[reply]
We must protect the Queen. We are merely expendable drones. Doc talk 04:13, 29 September 2011 (UTC)[reply]
I agree entirely. We haven't had much evidence at all that current phrasing is causing inaccurate content to be included. I'm persuaded that there may be a case for having the verifiability, not truth section the first one after the lede and thus expanded with a few examples, but merely as an easier way in for new users. Verifiability is one of the harder policies to grasp for a new user.VsevolodKrolikov (talk) 03:10, 29 September 2011 (UTC)[reply]
You and I were once "new users". Did you not freakin' figure it out, at least eventually, without getting bitten? This whole thing is just so absurd to me... but it will continue! How about a straight vote: 1) Remove "not truth" -or- 2) Retain "not truth". So where you think that would go? We probably need yet another poll for that... Doc talk 03:21, 29 September 2011 (UTC)[reply]
Sampling editors who are still here as a means of finding out which new editors were driven away by being bitten isn't the most appropriate methodology ;-). No, I don't favour a straight vote, as it would ignore the good work done in achieving common ground between editors who disagreed with each other for ages.VsevolodKrolikov (talk) 03:32, 29 September 2011 (UTC)[reply]
That straight vote would end this issue once and for all. But that still wouldn't be good enough for those seeking to eliminate the "not truth" phrase. Just admit it. Doc talk 03:39, 29 September 2011 (UTC)[reply]
Doc, That's a good idea to have a vote on just eliminating the two words "not truth" from the first sentence. From the discussion here, it doesn't seem that the rest of the proposal is helping win over anyone who is opposed to changing the first sentence. Following up on your previous allusion to the Borg, even the Borg of Star Trek had a slogan that made sense, "resistance is futile", unlike the Borg of Wikipedia that has the slogan "verifiability, not truth". Maybe it's from a difference in quality between each queen. --Bob K31416 (talk) 04:43, 29 September 2011 (UTC)[reply]
We don't need to get into who's the Borg or who isn't: it was an "analogy". It all boils down to either removing two words from the lede or not removing those two words. Roll the dice with one vote on that alone. Who knows who will "win"? Doc talk 04:51, 29 September 2011 (UTC)[reply]
Protect the queen! Protect the queen! --Bob K31416 (talk) 04:58, 29 September 2011 (UTC)[reply]
Whatever. Stay on topic please ;> Doc talk 05:07, 29 September 2011 (UTC)[reply]
Admit it? Again, you're asking the wrong person. I want to keep, "not truth", as a reading of the discussions over the proposals might have shown you (cough).VsevolodKrolikov (talk) 03:51, 29 September 2011 (UTC) Sorry - I keep calling you Bob in the edit summaries. Brainfart.VsevolodKrolikov (talk) 03:53, 29 September 2011 (UTC)[reply]
Heh - noprob on the "Bob" thing ;> There are so many proposals that it is really very unclear what the #1 proposal to replace "not truth" really is. Again. So I haven't carefully read through the voluminous content of the many proposals and side proposals, especially as I feel that no change is necessary either. Cheers :) Doc talk 03:58, 29 September 2011 (UTC)[reply]
I don't think this situation is covered in either the current version or my proposed changes... because it isn't a verifiability issue. What we have here isn't a case of an "assertion of truth or untruth" (no one is questioning that the survey took place and gave the results it did), instead what we have is an "assertion of unreliability" (ie someone is arguing that the survey is unreliable). That is an issue for WP:RSN. In other words, this is a reliability issue, not a verifiability issue. WP:V is the wrong policy to be pointing to. Blueboar (talk) 03:24, 29 September 2011 (UTC)[reply]
How do they agree on whether its unreliable, without first agreeing that a verifiable claim has been made? They clearly can't agree initially what verifiable claim is being made. One says its shows why a corporation did something, the other says it (deliberately) provides false data. The only way to begin to do this is by first agreeing on what the source claims (verfiability). Not what one editor or the other misrepresents or misconstrues the source to say. Alanscottwalker (talk) 03:50, 29 September 2011 (UTC)[reply]
Alanscottwalker, may I respectfully suggest that some of what you have written above is wandering into the grey area between assertive and aggressive? It's not a clear violation of WP:CIVIL, but it would be much more comfortable to read through if it were toned down just a little. :o) Pesky (talk …stalk!) 07:38, 29 September 2011 (UTC)[reply]

With full respect to all editors involved in this discussion and the example immediately above, I question the usefulness of spending too much time on any one example. Wanting to improve wording clarity is a good thing. We've had enough example discussions to show that the nature of this particular issue is one where it is hard to point to the results of any policy wording. The discussion needs to be based on editors reading different alternative texts and SEEING which one is more clear or less clear. All attempts to avoid this method keep bogging down discussion.--Andrew Lancaster (talk) 09:04, 29 September 2011 (UTC)[reply]

I feel that Blueboar's proposal only complicates what is already an overly verbose set of paragraphs. I still think taking what we have and condensing it down to this is the better solution:
The threshold for the inclusion of information in Wikipedia mainspace (articles, lists, sections of articles, and captions) without exception is verifiability— ie being attributable to a reliable, published source appropriate for the content in question. This requires that all quotations and anything challenged or likely to be challenged be attributed in the form of an inline citation that directly supports the material.(See the discussion about sources in WP:NOR that describes summarizing materials in your own words, leaving nothing implied that goes beyond the sources.) For how to write citations, see Citing sources. Anything that requires but lacks a source may be removed, and unsourced contentious material about living persons must be removed immediately.
This not only eliminates the whole "not truth" distraction but massively streamlines what IMHO is a collection of disjointed paragraph sentences.--BruceGrubb (talk) 14:19, 29 September 2011 (UTC)[reply]
Bruce, barring one or two editors, I think "not truth" has been accepted as having a place in policy. It's clear that there is no way a proposal without "not truth" would be acceptable to the clear majority of editors. The compromise solution is to put "not truth" in the first section after the lede. The compromise reasoning is that (a) we need "verifiability, not truth" but (b) it may be better not to confront people with this in the first sentence of the policy.VsevolodKrolikov (talk) 15:43, 29 September 2011 (UTC)[reply]
I don't agree with that at all. I think the evidence we have is that a majority of editors active on this page want "not truth" and are mainly here for the purpose of protecting it at all costs. I also think there's evidence at previous RFCs that editors who are not active on this page rate "not truth" rather less favourably. I've moved position to accepting some version of "not truth" in the policy for the moment, but I'm not really very happy about having made that concession. The fact is that having a verifiability policy that disregards truth is like having an economic policy that disregards money, or an equestrian policy that disregards horses. The policy as written clearly has nothing to do with verifiability in its natural meaning:- it's about the need to cite sources. And also, I'm highly suspicious of anyone who wants to write an encyclopaedia but doesn't feel any urge towards telling the truth. Encyclopaedia writers are educators, and should be accordingly intolerant of lies or liars.—S Marshall T/C 16:47, 29 September 2011 (UTC)[reply]
You're wrong if you think you'll ever get "not truth" ever taken out (the last vote was stunningly against taking it out, involving large numbers of editors). All you can hope for is these negotiations to accommodate concerns you can persuade people of. Vanishingly few of those who have taken an interest at some point think "not truth" creates content problems. Some are prepared to work on softening the presentation of policy, but if your sole aim is to get rid of "not truth", just walk away now.VsevolodKrolikov (talk) 17:09, 29 September 2011 (UTC)[reply]
Oh please. What the vote was "stunningly against" was Jrincayc's proposal. That one poll is very far from rendering the words "not truth" totally unchallengeable for all time in the way that you suggest.

The truth is that policies tend to react to crisis. We have Mr Seigenthaler to thank for our BLP policy. Thanks to Rlevse, WP:V now explains that verifiability is not an excuse for serial copyright violations. And WP:V will be clarified again in response to crisis in future. I think the issue of adding sourced untruths to the encyclopaedia may well be one of the future crises. But as I said, I've allowed myself to be persuaded in favour of Blueboar's proposal for the moment, and I'll see how that works out in practice once it's been implemented. If we can ever get as far as the RFC...

I have various other aims for this page, including altering certain other wordings (particularly "the threshold"), mitigating the onerous burdens that monoglots can currently place on their more linguistically educated colleagues, simplifying the unnecessarily complicated constructions into the simple imperative (and it wouldn't hurt us to say "please" to our volunteers either), and generally trying to turn it into a coherent and well-expressed chain of thought.—S Marshall T/C 17:48, 29 September 2011 (UTC)[reply]

Besides any other disagreements, your reference to non-English sources intrigues me. I haven't met any difficult in foreign language sourcing so long as I provide a translation. (Well, except for one nutcase who thought Japanese and Chinese were basically the same). How does WP:V not deal with this now?VsevolodKrolikov (talk) 18:12, 29 September 2011 (UTC)[reply]
As usual with WP:V, "nutcases" are the problem.  :) But if no editors were nutcases then we wouldn't need policies.

As presently written the "non-English sources" section is a free gift to a griefer, who could ask me to transcribe quite long sections of books and websites, in both the source language and in English, onto talk pages. In many cases the source language section would be much longer than the single fact I wish to cite. (There's self-interest in this. I'm an editor active in controversial discussion areas and I source a lot of the content I write to non-English sources. I view reform of the "non-English sources" section of this policy as risk management.)

I want the "non-English sources" to be rewritten such that the monoglot can challenge a foreign-language source by asking a neutral, third-party editor who speaks the language to read it and confirm it, rather than by making the editor adding the material jump through all these hoops.—S Marshall T/C 18:54, 29 September 2011 (UTC)[reply]

As I said "not truth" is a 'distraction--it makes no distinction between Verifiability truth and believed truth. As a result you have nonsense like Talk:Weston_Price/Archive_1#Weston_Price_and_Stephen_Barrett_in_their_own_words which was taken over to Wikipedia:Reliable_sources/Noticeboard/Archive_79#Is_a_paper_.28possible_blog.29_by_a_psychiatrist_valid_regarding_old_claims_regarding_dentistry.3F and degenerated into the barrel of utter insanity that was Wikipedia:Administrators'_noticeboard/Incidents/Problem_on_BLP_noticeboard.
While the consensus was Stephen Barrett's comments about Weston Price did NOT meet RS in retrospect the situation would have been better handled over at Wikipedia talk:Neutral point of view/Noticeboard but thank to the "not truth" we had a few editors believing what amounted to a self published blog was MORE RS than a 1939 book published by Paul B. Hoeber, Inc; Medical Book Department of Harper & Brothers because the peer reviewed source was so old and only way to shut that nonsense down was to argue Stephen Barrett didn't meet RS.
The whole "not truth" thing has resulted in editors developing what seems to be an 'accuracy be damned--as long as it meets Verifiability it is good' mentality. This in turn results in the view that anything that contradicts what a particular source states regardless of Verifiability is somehow OR (Talk:Conspiracy_theory#The_first_recorded_use_of_the_phrase_.22conspiracy_theory.22_dates_from_1909.3F_WRONG.21 is the most recent poster child for some of that nonsense)
Much the same problem made the Jesus myth theory a disjointed train wreck for nearly two years as many editors could not understand there was NO set definition for "Jesus myth theory" and much of the information was (and still is) a confused mess.
For some editors Verifiability itself has become truth and when THAT happens "not truth" doesn't mean squat as to them Verifiability is truth and anything that contradicts this truth is OR.--BruceGrubb (talk) 18:36, 29 September 2011 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Verifiability/First_sentence&oldid=1142561652"