Archives

Index 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 51, 52, 53, 54, 55, 56, 57, 58, 59, 60 61, 62, 63, 64, 65, 66, 67, 68, 69, 70 71, 72, 73, 74, 75, 76, 77, 78, 79, 80

Archives by topic First sentence (Nov 2010–March 2011) First sentence (April–August 2011) 2012 RfC about the lead section

This page has archives. Sections older than 30 days may be automatically archived by Lowercase sigmabot III when more than 4 sections are present.

Recently a talk page discussion on this policy page led to this change.

Q: Was this a good idea? Please express a view below.—S Marshall T/C 23:12, 9 April 2017 (UTC)[reply]

Note: the same discussion also prompted this related change to the lede, also relevant to this Request for Comments. Diego (talk) 13:06, 10 April 2017 (UTC)[reply]

Yes

• Reyk _YO! 14:30, 9 April 2017 (UTC)[reply]
• Yes, for many many reasons delineated in the thorough discussion that led to the change, and also the many reasons delineated in the more recent large discussion discussion. Briefly:

1. WP:ver in essence imposes a requirement for inclusion, nothing more. Nowhere does it say anything about meeting wp:ver being a full or partial reason for inclusion. Some folks have been misreading this. This merely means that once wp:ver has been met, wp:ver "steps aside" and allows all other policies and practices to determine what happens.
2. As detailed at length in the discussion on this, a common problem (rooted in a false urban legend in Wikipedia) is mis-use or mis-reading of this policy to claim that meeting wp:ver is a force or reason for inclusion. For example, every time an editor says the ubiquitous "undo removal of sourced material" in an edit summary they are in essence making this claim. Nobody makes the claim that it is an absolute mandate for inclusion, but it is quite common to make the false claim that it is a strong force or reason for inclusion. This false urban legend is extensively used by wiki-lawyers to POV articles. Flatly saying that meeting wp:ver is not a reason for inclusion in the lede is (unlike "not sufficient") an accurate summary of of the body of the policy, and directly addresses the false urban legend.
3. IMHO many folks (including, going by memory, one or more current or ex Arbcom members) have said that Wikipedia has an overly byzantine set of confusing and overlapping / conflicting policies. This is a perfect example of sitting on the edge of that precipice, and avoiding the abyss. Regarding content, WP:VER sets a requirement for inclusion, nothing more unless we let it be a mess. It should not try to wade into the immensely complicated question of "sufficiency" for inclusion. To start with, "sufficient" requires not being in violation of dozens of policies and core guidelines. If those are complied with, there is an immensely varying additional bar for inclusion. At the low end might be an uncontested inclusion at an obscure article where the additional bar is essentially zero. At the other end of a spectrum, the additional bar for inclusion (on a contested major item) might be getting a strong consensus in an RFC. WP:VER should stick to imposing the verifiability requirement, not start reflecting on or overlapping with the hugely complex area of what constitutes sufficiency for inclusion.
4. For new timid editors, this falsely implies that there is some unknown-to-them wiki-rule about needing a compelling reason to include something. This imposes a false hurdle and also conflicts with new editor objectives (and wp:bold)
5. The RFC just asks for an overall opinion on collectively several different changes made in February, without asking or proposing any specific actions on any of them. If anyone is thinking that it is asking about any particular action, it is malformed with respect to that. But the work below on a compromise action is showing some promise.North8000 (talk) 21:03, 2 May 2017 (UTC)[reply]
6. The current wording was arrived at in a 27,000+ word discussion that ran for 23 days (Feb 7th - March 2nd) with no objections at the end of the process. North8000 (talk) 22:01, 18 April 2017 (UTC)[reply]

To be fair, S Marshall's first objection was raised only 18 days after the discussion you point us to ended (he started to object on March 20th) and we have been debating his objections ever since. Given the short time span between the change and his first objection, this all needs to be seen as one single ongoing discussion that started in Feb, and has not yet ended. Blueboar (talk) 22:23, 18 April 2017 (UTC)[reply]

I don't agree, although if it would resolve that aspect I'd agree to a compromise that considers them to be on equal footing. North8000 (talk) 01:45, 19 April 2017 (UTC)[reply]

North8000 (talk) 14:43, 9 April 2017 (UTC)[reply]

The idea that a fact being covered by a reliable source provides a reason to try to include it in the article is not a "false false urban legend in Wikipedia", it is how many of us see the policies and the core way that Wikipedia is built. If a strongly reliable and high-profile source covers a fact in depth, that's a very good reason for that fact to be included in a relevant article; but the new wording makes this coverage look not-at-all important. Surely there may be other reasons why the content might be excluded, but this doesn't mean that being verifiable provides no reason at all for using it, not even a partial one, as the new text says.

With the new version, if you have two facts where one is non-verifiable and the other is covered in-depth by multiple reliable sources, both would have the same reason for inclusion according to this policy, which is none at all. In fact, now the only remaining reason in policies to include content is editors have consensus to include it; in an attempt to remove from WP:V the weakly verified content, the really strong verifiable has been eliminated with it, thanks to the poorly and overbearing way the new wording is expressed. Diego (talk) 12:40, 10 April 2017 (UTC)[reply]

I point you to NPOV. That's the policy that requires looking into allot of strong sourcing (thus ,expressly making that a non-V reason). The language under discussion says "verifiaability alone", which just reinforces the long-standing introduction that points to other policy and again says, 'you know, you do have to come up with other reasons than you have one RS.' Alanscottwalker (talk) 15:07, 10 April 2017 (UTC)[reply]

Do you mean this wording in the lede? "However, while verifiability is required for including something, it is not a reason for inclusion." It has been there since last february, at the same time that the section was changed; there's no long-standing introduction saying "you have to come up with other reasons". By the way, NPOV introduction says: "It is also one of Wikipedia's three core content policies; the other two are "Verifiability" and "No original research". These policies jointly determine the type and quality of material that is acceptable in Wikipedia articles, and, because they work in harmony, they should not be interpreted in isolation from one another.". Diego (talk) 16:09, 10 April 2017 (UTC)[reply]

I think the lead would best follow the body of the policy,("verifiability alone"), but no I expressly pointing to the links to other policy including NPOV in V's lead, which says you have to look at more than V to determine content. Alanscottwalker (talk) 19:00, 10 April 2017 (UTC)[reply]

@Diego Moya:, I'm with you on your overall view, but what you don't realize is that the February wording is also with you on your overall view. The February wording merely accurately summarizes wp:VER; once wp:ver is met, it steps aside, allow other practices (such as normal article building) and policies take over. In fact, per my last of four points, the February wording actually usually works on the inclusionist side. North8000 (talk) 21:22, 10 April 2017 (UTC)[reply]

• Yes, Verifiability ≠ nobility/inclusion. - Mlpearc (open channel) 14:52, 9 April 2017 (UTC)[reply]
• Yes, per North, but also because "does not guarantee inclusion" implies that it is a reason for inclusion...and it isn't. Nobody should be adding information to an article if their sole reason for doing so is "it's verifiable". DonIago (talk) 15:53, 9 April 2017 (UTC)[reply]

Question: why shouldn't someone add information if their sole reason is "It's verifiable"? (Note: adding information is not the same as objecting to someone else removing it) Blueboar (talk) 16:11, 9 April 2017 (UTC)[reply]

For various reasons outlined in the policies WP:NOT and WP:POV, and in the guidelines like WP:RS and WP:TECHNICAL, and even style guidelines like WP:EMBED. Bright☀ 20:48, 9 April 2017 (UTC)[reply]

Consider this... suppose there is information that you think might improve an article... you are fairly sure it is accurate, but you hold back on adding it because you don't have a source. You search and search... and Then you finally find one. You can now add the information... why? The reason is because you can now show that it is verifiable.

Now, this verifiability does not guarantee that the information will stay in the article... we all agree that there are many reasons why it might get removed... but the verifiability does allow you to add it in the first place.

In other words, verifiability can be a reason to include information... But verifiability is not enough to protect information from removal (i.e. exclusion). It's a subtle, but important distinction. Blueboar (talk) 22:36, 9 April 2017 (UTC)[reply]

"you think might improve an article" -- you already have a reason, and it is not that it is verifiable because (in your scenario) you are not even sure it is verifiable. You should already be able to articulate why you think it improves the article. -- Alanscottwalker (talk) 22:49, 9 April 2017 (UTC)[reply]

That hypothetical person already felt the information improves the article even before they knew it's verifiable, so the information must have had some other value. Then, once it's verified and added, that reason can be considered by other editors, hopefully according to Wikipedia policies and guidelines. Bright☀ 23:57, 9 April 2017 (UTC)[reply]

It must have some other value than verifiable, as you say, and that reason existed before it was added. Alanscottwalker (talk) 12:14, 10 April 2017 (UTC)[reply]

WP:WEIGHT says "Neutrality requires that each article or other page in the mainspace fairly represent all significant viewpoints that have been published by reliable sources (yadda yadda prominence of each viewpoint)", so apparently some other policy does consider verifiability as a reason to include at least some kind of content, in some circumstances. Diego (talk) 13:51, 10 April 2017 (UTC)[reply]

No. That other policy, expressly makes that an NPOV reason not a V reason. And it's actually focusing on, examining and representing the wide literature - that is not V's issue, which is 'we require a single RS'. Alanscottwalker (talk) 14:26, 10 April 2017 (UTC)[reply]

And per NPOV, the reason why a topic is prominent enough to have due weight is ... drum roll ... because it is verifiable in several RSs! The assertion "Verifiability is not a reason for inclusion" at the very least mischaracterizes this part of WP:DUE. Diego (talk) 15:10, 10 April 2017 (UTC)[reply]

The policy says: "Verifiability alone is not a reason for inclusion." Alanscottwalker (talk) 15:31, 10 April 2017 (UTC)[reply]

But the section title says: "Verifiability is not a reason for inclusion." Diego (talk) 15:38, 10 April 2017 (UTC)[reply]

• Yes. I see altogether too many partisan editors who insist that an encyclopedia article has to include today's trivial tempest in a teapot solely "because it's got reliable sources" (and because it advances their agenda). — Malik Shabazz ^Talk/_Stalk 17:01, 9 April 2017 (UTC)[reply]

This is a typical way how articles are expanded:

1. A reliable source describing some fact about a topic provides support towards including the fact in an article.
2. Some editor provides a reason why this fact should not be included, based on another policy.
3. The strength of the source and the other policy are weighted, and a decision is made by consensus whether to include the fact or not.

The blunt recent change to policy amounts to forbidding step 1, i.e. including a fact in an article while giving as reason that you have found a reliable source supporting it; but this is overkill. The right approach to solve your stated problem instead would be to emphasize step 3, i.e. that in case of dispute you need consensus even if the fact is verifiable. Diego (talk) 14:02, 10 April 2017 (UTC)[reply]

• "Does not guarantee" is the wrong idiom. At issue is not whether anything is guaranteed, but whether it is a good idea. "Is not a reason" is much better, but might be amended for clarity to read "is not itself a reason". (Insofar as non-verifiability is a sufficient reason for exclusion, verifiability is a necessary but not sufficient reason for inclusion.) ~ Ningauble (talk) 19:18, 9 April 2017 (UTC)[reply]
• Per above, Malik Shabazz and Ningauble, whom I assume in talking about 'good idea', includes in that the general fulfilling content policies/guidelines (not just V) . I also think this is better, in part, because Verifiability policy sets its minimalist requirement at only "a" (single) RS. It makes little sense to suggest, 'I gotta source' is enough reason. -- Alanscottwalker (talk) 21:53, 9 April 2017 (UTC)[reply]
• Yes, "for many many reasons delineated in the thorough discussion that led to the change" (see #Where verifiability is not an adequate criterion, above). ~ J. Johnson (JJ) (talk) 22:12, 9 April 2017 (UTC)[reply]
• Mostly yes, only that I feel the wording should be "Verifiability is not the only reason for inclusion." which is stressed more in the explaination but could be misread (without "only") as that we don't consider verifiability as one of the reasons for inclusion. --MASEM (t) 01:05, 10 April 2017 (UTC)[reply]

One of the sentiments in the original discussion is that WP:V should not be a reason for inclusion, only a requirement; that once something meets this minimal bar WP:V should have nothing further to say. If something is questioned verification is required. If that can't be done, then it is out, end of story. If it can be verified, that is as far as WP:V goes; inclusion depends on other considerations. ~ J. Johnson (JJ) (talk) 06:53, 10 April 2017 (UTC)[reply]

I recognize you're talking the diff between a requirement and a reason, and I fully appreciate that, but I see the current suggested change as something that an inexperienced editor is going to take out of scope since it does not capture this subtle difference. (And yes, the body text covers it more, but keep in mind how inexperienced editors often take one line of policy to justify a point). It would be better that the sentence include this, even something like "Verifiability is a necessary but not sufficient reason for inclusion" or the like. --MASEM (t) 13:17, 10 April 2017 (UTC)[reply]

• Yes. Many of those who argue for "no" insist on drawing on other reasons why verifiable by high quality sources tends to be worth including in some article. I believe each policy should strictly stick to the goal of the particular policy, and not trespass on the territory of other policies, otherwise we end up with six or seven policies that all say the same thing, and it becomes impossible to change any of them because we have to change all of them at once. If the wording surprises people, good; we should vigorously shove the concept that each policy has its own territory in editors faces. Jc3s5h (talk) 13:56, 10 April 2017 (UTC)[reply]

I believe each policy should strictly stick to the goal of the particular policy, and not trespass on the territory of other policies. I could agree with that sentiment. The problem is that the recent change does not achieve that goal; it makes WP:V step in the territory of WP:WEIGHT, describing reasons why some content should or should not be included in an article. I recognize that the change intended to isolate WP:V from reasons for and against inclusion, but the way it's worded provides some reasons not to include content instead. Diego (talk) 14:11, 10 April 2017 (UTC)[reply]

No. The Weight reference has been in that paragraph forever (and was never changed), and it's function is to point away from V to NPOV (by link), that is expressly not stepping into NPOV. Alanscottwalker (talk) 14:36, 10 April 2017 (UTC)[reply]

(Uh, I don't understand to what paragraph do you refer? "Weight" is mentioned only in the introduction, though there's a See also link to UNDUE at Other issues). With "describing reasons why some content should or should not be included in an article", I refer to the wording "Verifiability is not a reason for inclusion", not any outside link. In any case, Weight links back to WP:V as a way to decide whether some fact has enough weight. Diego (talk) 15:10, 10 April 2017 (UTC)[reply]

The paragraph section under discussion has a big Undue (WEIGHT) link right there. The policy language is Verifiability alone is not a reason for inclusion, in other words, you are going to have to have more reason then just "I gotta RS", so look at these other policy and guidelines. Alanscottwalker (talk) 15:28, 10 April 2017 (UTC)[reply]

In that case, it should rather say "you have to have X reasons to include it", rather than "Verifiability is not a reason for inclusion". The change is very poorly worded. Diego (talk) 15:37, 10 April 2017 (UTC)[reply]

I strongly agree that each policy should stick within it's defined scope. Which is violated if WP:V gets into the other reasons for inclusion (as in "you must have ..."). The intent of the change (aside from any issue of implementation) is to remove WP:V from inclusion debates; that it speaks to (and is used in) only the narrow issue of a minimum requirement for inclusion. ~ J. Johnson (JJ) (talk) 20:06, 11 April 2017 (UTC)[reply]

• Yes: Makes sense for newbies who are confused about verifiability and creating their first articles. KGirlTrucker81 ^{huh? what I've been doing} 23:38, 10 April 2017 (UTC)[reply]
• Yes The language itself may need tweaking, but the previous wording did not stress the distinction between the minimum threshold for inclusion and editorial judgement regarding all other policies concerning inclusion. WP:V is not the be-all and end-all of inclusion. I've seen this mistaken before, particular when trying to include some minor negative material into BLPs of people some editor doesn't like.--v/r - TP 00:25, 14 April 2017 (UTC)[reply]
• Yes When I remove a line from an article for a reason, even if subjective, it is not an appropriate response to re-add it simply because it is verifiable. "Do not remove sourced information", as has been used in reversion of my edits, is not a good explanation to why content should be kept in an article indefinitely. Reywas92^Talk 00:18, 17 April 2017 (UTC)[reply]

But it is appropriate to re-add it when it has been removed without a reason, which is all too common. Reverting a removal of sourced information is an adequate edit when the removal was done without an edit summary nor a comment at the talk page. It is also a good way to prompt the removing editor to explain themselves and explicit the reason they had for deleting the information. Diego (talk) 10:17, 17 April 2017 (UTC)[reply]

So given an addition (inclusion) "without a reason", and a subsequent removal (deletion) "without a reason", you are saying favor the inclusion. Which is contrary to WP:V:

[First four points not applicable. See below. ~JJ]

• "all material must be attributable ..." (nutshell);
• "Any material that needs a source but does not have one may be removed" (lede);
• "The burden to demonstrate verifiability lies with the editor who adds or restores material..." (first section; emphasis in the original);
• "Any material lacking a reliable source directly supporting it may be removed and should not be restored without an inline citation to a reliable source" (following; emphasis added); and not the least:

• "The onus to achieve consensus for inclusion is on those seeking to include disputed content (WP:ONUS)."

Is any of that unclear? ~ J. Johnson (JJ) (talk) 22:21, 17 April 2017 (UTC)[reply]

Reywas92 was talking about sourced (and thus verifiable) information. You are quoting the policy on unsourced information. It is unclear what relevance your comment has over the comments it replies to. Diego (talk) 08:39, 18 April 2017 (UTC)[reply]

Point taken: the first four points hinge on a lack of verifiability. And this discussion is entirely on the inclusion or exclusion of verifiable material. However, the last point still holds: in content disputes consensus has to be obtained for inclusion.

Note that in the particulars you describe – where material has been removed without a reason, which is to say, without an edit summary – I actually favor a revert. But that is on the basis of BRD, not anything to do with WP:V. (I also favor reversion of additions that lack an edit summary.) At any rate, the removal, addition, reversion, or whatever that is done without a reason is irrelevant here, because the issue is whether verifiability IS a "reason" for inclusion. The particulars I envision are after the preliminaries of Boldly adding/removing/whatever verifiable material and Reverting, whether verifiability alone, without any other reason, is sufficient to "guarantee" inclusion. ~ J. Johnson (JJ) (talk) 23:57, 18 April 2017 (UTC)[reply]

whether verifiability alone, without any other reason, is sufficient to "guarantee" inclusion. See here, there's again this essential misunderstanding I see in that statement. No one here is defending that verifiability alone, without any other reason, is sufficient to "guarantee" inclusion (not even S Marshall). Yet the change in the policy wording is a gut punch against our position that verifiability is a sign leaning towards including the content, if there are no reasons to exclude it, which was compatible with the old guideline but not the new one.

The problem comes from the insurmountable ambiguity of the words "not a reason", which can mean either not "a guarantee" but also not "support towards". We'd better off if the change had said "not enough reason" or "not a reason sufficient to...", which only rejects the notion that verifiability gives a free pass for inclusion (which nobody argues), and not the idea that verifiability provides some indication of due weight (which I support). Diego (talk) 10:19, 19 April 2017 (UTC)[reply]

The problem (as has been addressed elsewhere) is that terminology like "not enough reason" or "not sufficient reason" implies that verifiability is a reason for inclusion (in the sense of "support"). As to whether that should be a free pass, that is the sense of guarantee, as well as the plain meaning of Marshall's own words, which I have quoted elsewhere. But that is an argument we are having elsewhere. Whether "verifiability provides some indication of due weight" would be a good discussion to have, but should be started afresh on its own thread. ~ J. Johnson (JJ) (talk) 02:25, 20 April 2017 (UTC)[reply]

Then use a wording that does not endorse that sense of support, but that also don't endorse its nonexistence. At this point, I'd be in favor of removing the sentence as a whole, since now it clearly doesn't express anything that the community as a whole agrees to, nor provides any guidance. I'd rather have the paragraph reduced to what the community actually agrees: Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content. BTW there's a whole world of difference between "providing support" and "being a free pass", and you would do well in recognizing it in order to stop misrepresenting the argument of the other part. Diego (talk) 09:04, 20 April 2017 (UTC)[reply]

• Yes. It helps fix a not uncommon problem in ideologically contentious articles. Earnest editor A adds widely published, verifiable information that may well be important for everyone to know but that relates only indirectly to the subject. B accordingly removes it. A restores it "per WP:V"... --Dervorguilla (talk) 19:52, 18 April 2017 (UTC)[reply]
• Yes, for many reasons, but the main one would be that the current summary text better summarizes the text in the box below. LK (talk) 01:54, 22 May 2017 (UTC)[reply]
• Yes as this seems to be a reoccurring problem, particularly with new editors and POV warriors. Dennis Brown - 2¢ 14:28, 23 May 2017 (UTC)[reply]

No

• —S Marshall T/C 14:19, 9 April 2017 (UTC)[reply]
• While not as adamant about this as S Marshall seems to be, I do think the recent change in language was a mistake. The thing is, verifiability actually can be a reason to add information (ie include it) at least initially... The problem is that verifiability often isn't enough for us to keep added verifiable information in an article. There are lots of reasons why we might remove (i.e. exclude) verifiable information from an article. Some of these reasons are outlined in other policies (for example, mentioning it may give undue weight to a fringe view)... some are not based on policy (for example, editors may simply feel that the information is too trivial to be worth mentioning.) In essence, what we want to tell our editors is this... Both adding and removing verifiable information is allowed... both are part of the normal editing process. I think the original "not a guarantee of inclusion" language expressed this concept better than the current "not a reason for exclusion" does. Blueboar (talk) 15:37, 9 April 2017 (UTC)[reply]

I think you mistake the issue. It is not that [b]oth adding and removing verifiable information is allowed" – that is given. The issue is more in cases where there "are lots of reasons" to remove, and verifiability is not just "a" reason among others for retention, but is the sole reason put forward for retention. Where there are "lots of reasons" for retention, verifiability is not needed. Where there are lots of reasons against, verifiability alone should not be superior ('guaranteed'). Part of the argument against the term "guarantee" is that it seems more formal, and simply denying any "guarantee" would leave plenty of scope for mischief ~ J. Johnson (JJ) (talk) 22:23, 9 April 2017 (UTC)[reply]

That's not what it says. If that's what it's meant to mean, then that's what it should say.—S Marshall T/C 22:29, 9 April 2017 (UTC)[reply]

Let me give a hypothetical to show how verifiability can be a reason for inclusion... Suppose an editor remembers reading a bit of information in a book, but hesitates to add it to a relevant article because he can't remember what the book was (ie he can not show that the information is verifiable). He searches high and low... and finally, after perhaps years of searching, he finds the book... and joy of joys, it qualifies as reliable! Now he can finally add the information to the article. Why can he add it? The reason is solely because he can now show that the information is verifiable.

HOWEVER... Once added (due to verifiability), the information might still end up being removed. There are a host of reasons to remove information (even verifiable information). So, while verifiability can be a reason to add (include) information, verifiability does not "protect" information from being removed.

You know, the more I think about this, the more I am thinking that the underlying problem isn't with the words "guarantee" or "reason"... the problem is actually with the word "inclusion". The section under discussion isn't really about inclusion... it's about removal. Perhaps it would be more accurate to rephrase: "Verfifiability on its own does not protect information from being removed." Blueboar (talk) 23:49, 9 April 2017 (UTC)[reply]

In your second scenario, it's still not the only reason or even the main reason -- they have presumably read countless verifiable things in their life, they must have some overweening reason other than verifiability for wanting to add that there. -- Alanscottwalker (talk) 00:49, 10 April 2017 (UTC)[reply]

Nope... in my scenario the editor read it, thinks the article should include it, but waited until he found a source to verify it. That's it. Lack of Verifiability, was the only reason for the wait and verifiability was the only reason for the addition. I hope you are not trying to argue that he can't add it. Are you saying he shouldn't add it? Blueboar (talk) 01:10, 10 April 2017 (UTC)[reply]

No. I am saying that is not the reason - why did your hypothetical editors think it belonged there? That's the reason. Alanscottwalker (talk) 03:09, 10 April 2017 (UTC)[reply]

I think you may be assuming a motivation that is not there. In my scenario, the editor reads an article... sees that information he remembers reading about is missing from the article, and thinks the article should mention it ... no more, no less. Blueboar (talk) 10:59, 10 April 2017 (UTC)[reply]

Why does the editor think that should go there? Why does that information improve the article? It's not just that a source exists (out of the multitudes of RS or verifiable information, they have known) and because it is not just that a source exists, you are beyond V, as reason. Is it relevant (or due)? Does a knowledge of the subject suggest that it is needed for context? Why? Does the fact that it is missing, render the article somewhat more confusing, or unclear about the topic? Why and how? If it is not added, is the article skewed? How? And it goes on. V alone - a source exists - (or often, at all) is not the answer for reason(s) for it going there. Alanscottwalker (talk) 11:43, 10 April 2017 (UTC)[reply]

In your scenario the editor reads it, but waits until he found a source? Ah, surely you have not forgotten WP:SAYWHEREYOUGOTIT. Where he reads it is the source. Perhap he doesn't know how to cite it in proper form, but if he can't point to where he got it, then there is a keen question of just where this comes from. Of course, it appears that sometimes editors see really interesting stuff in non-reliable sources, so they hunt for some barely adequate source to give them cover, but that is going about this backwards. Material for which an editor cannot provide both a reliable source and a reason for inclusion does not belong in the 'pedia. ~ J. Johnson (JJ) (talk) 07:28, 10 April 2017 (UTC)[reply]

Read the scenario again... the editor remembers something he read some time ago, but can't remember the source he read it in... so he does exactly the right thing - he searches for the source, and does not add anything until he finds it (in other words, he is following SAYWHEREYOUGOTIT). Blueboar (talk) 10:59, 10 April 2017 (UTC)[reply]

Blueboar, I think you are worrying about a non-problem, and missing the real problem. This merely means that wp:ver "steps" aside after wp:ver has been met. Nobody uses wp:ver to exclude wp:verifiable material. But many mis-use the urban legend to try to us meeting wp:ver to force inclusion, usually in POV maneuvers. North8000 (talk) 12:23, 10 April 2017 (UTC)[reply]

But this is not what the new wording implies; had the text been something like "after it is satisfied as a minimum requirement, verifiability has no bearing on whether something should be included or not", you might have a point. However, the new wording strongly implies that it doesn't matter that a fact is verifiable, even by very strong references, it should not be included, or at least not until you jump through hoops to explain how it matches every other policy under the sun. Diego (talk) 13:00, 10 April 2017 (UTC)[reply]

No. "Verifiability alone" does not imply that. "Very strong sources" is rather meaningless, without answering the 'for what?' The 'for what' answer goes into other reasons (This policy expressly already says in the intro: "Verifiability" is no reason for adding sourced text that makes the article OR, that renders it POV, that makes it CVIO, that makes it a BLP problem by giving 'false light/undue, etc.) "Strong sources" is an NPOV or OR or BLP reason, V requires a single source. - Alanscottwalker (talk) 13:48, 10 April 2017 (UTC)[reply]

But the recent change forbids using WP:V as a reason for inclusion even if there are many in-depth reliable sources. If the change had said ""Verifiability is no reason for adding sourced text that makes the article OR, that renders it POV, that makes it CVIO, that makes it a BLP problem by giving 'false light/undue", we wouldn't be having this conversation. Diego (talk) 14:06, 10 April 2017 (UTC)[reply]

Plainly false, one of the things in that very paragraph says, among others, is look to weight(due) and points to NPOV policy, making it an NPOV problem and reason (not V).Alanscottwalker (talk) 14:17, 10 April 2017 (UTC)[reply]

And per NPOV, the reason why a topic is prominent enough to have due weight is because it is verifiable in several RSs. V points to NPOV, but WEIGHT in NPOV refers back to V. The two policies collaborate, NPOV does not take the full responsibility. Diego (talk) 15:08, 10 April 2017 (UTC)[reply]

Thus, you have just stated, "variability alone" is not enough - and that's what the the policy says. Alanscottwalker (talk) 15:17, 10 April 2017 (UTC)[reply]

I've said elsewhere that I could live with the "verifiability alone" wording. But the section title says "Verifiability is not a reason for inclusion", not "verifiability alone". I tried to change it to make both appearances consistent with each other, but it was reverted. Diego (talk) 15:22, 10 April 2017 (UTC)[reply]

Oh? I see. I agree that the wording of the policy would be better in the title. Alanscottwalker (talk) 15:37, 10 April 2017 (UTC)[reply]

• I agree with Blueboar. Verifiability certainly is a reason for inclusion, but other policies often override it for exclusion. Sławomir Biały (talk) 11:17, 10 April 2017 (UTC)[reply]

And I say it's not a "reason" for inclusion. (Merely a requirement.) Which goes to show that we're all arguing about terms, but we don't yet have consensus on the underlying principles. Even if we could agree to specific terms, we would still have different theories of how it interpret them, so the confusion and arguments would continue unabated. That is why this RfC is fundamentally flawed: it doesn't have an agreed on basis such that we are all starting from the same place. ~ J. Johnson (JJ) (talk) 20:47, 10 April 2017 (UTC)[reply]

I don't dispute that you do not find it a reason for inclusion. However, my perspective is that, if something is "verifiable", then all other things being equal, it would be better to include the thing in an encyclopedia than not to. More precisely: if our prior is that the other policies and guidelines are neutral, then I believe it is marginally better to include a verifiable fact than not to include that fact. From my point of view, being verifiable does create a reason for inclusion (although that is a necessary but not sufficient condition of inclusion), and this is in conflict with the way the guideline is currently written. Of course, it is often not the case that other things are equal. But I think that for most mundane facts in the encyclopedia, other things are, in fact, equal. Determining how each individual edit sits with respect to the entire edifice of WP:PAG is simply not a reasonable requirement. Sławomir Biały (talk) 00:17, 11 April 2017 (UTC)[reply]

Most verifiable information does not belong in this, or any other, encyclopedia. Every entry in every phone book, the information from every birth, marriage, and death certificate for everyone who is not living, the details of every benchmark maintained by the U.S. National Geodetic Survey, etc., do not belong. The fact that we seldom encounter editors anxious to add such trivia is a credit to our pool of editors, but nonetheless, the vast majority of available verifiable information does not belong. Jc3s5h (talk) 02:28, 11 April 2017 (UTC)[reply]

I did not say that "most verifiable information ... belong[s] in this ... encyclopedia". This seems to neglect the very specific prior that other policies and guidelines should be neutral. In particular, the considerations of WP:NOTDIRECTORY already exclude the kind of content that you mistakenly believe that my view condones. Sławomir Biały (talk) 09:27, 11 April 2017 (UTC)[reply]

That is one of the key issues here: does "being verifiable ... create a reason for inclusion"? Consider this: in the universe of ALL reliably sourced material, a great deal (such as phone directories) is excluded by various polices (WP:TRIVIA, WP:BLP, etc.). But there is an immense amount of material NOT excluded, much more than could be included in the entirety of Wikipedia. What we are concerned about is where some editor, WITHOUT ANY OTHER REASON, insists that because something is verifiable it MUST be included in THIS (some) article. That WP hasn't burst with such information is because editors do have reasons for not including such stuff. The argument here is that inclusion should be based on thoseDoe reasons, not on mere verifiability alone. ~ J. Johnson (JJ) (talk) 20:49, 11 April 2017 (UTC)[reply]

Why not? Wikipedia is not paper. See "#Inc vs Del". Staszek Lem (talk) 21:08, 11 April 2017 (UTC)[reply]

Yes, it does create a reason for inclusion. When an editor adds a small fact here or there, for example what is the vapor point of mercury, the primary consideration is not and should not be whether it meets NOT, NPOV, NOR, BLP, or a host of other policies and guidelines. For most of the routine facts that appear in an encyclopedia, the verifiability condition is both necessary and sufficient, without having routine facts legalistically dissected noticeboard style. I am not saying here that other policies do not create a reason for excluding some kinds of content, but verifiability itself most definitely creates a reason for including something. And for the record, I think the Inc vs Del argument is a red herring. I am not an "inclusionist", nor do I believe that any verifiable content should be added to the encyclopedia. But I do think that, if something is not excluded under some other policy or guideline, then verifiability creates a reason for inclusion, absent any reasons for exclusion. It seems like some folks aren't getting that. And if people are saying that WP:V doesn't create a reason for inclusion, I challenge them to find any clear "reason for inclusion" in any policy. It seems to me that our background position is one of inclusion, with exclusion usually being dictated by one or more PAGs. Sławomir Biały (talk) 21:14, 11 April 2017 (UTC)[reply]

re red herring - yes you are an inclusionist ("centrist-leaning"), because for you everything is to be included unless there are reasons for exclusion. Staszek Lem (talk) 21:45, 11 April 2017 (UTC)[reply]

No, I am not. I actually prefer to delete things that are poorly sourced, and insist on high quality references for things that remain. However, there quite simply are no "criteria for inclusion". In order to have an encyclopedia at all, it is necessary to operate on the presumption that information is worth cataloging and including. If the default assumption is one of exclusion, our guidelines would need to be rewritten entirely to provide guidance on what we should include, but I read them as primarily exclusion. So if our default position were not to include information, and our guidelines further stated certain kinds of things cannot be included, then that leaves no content suitable for an encyclopedia. That is not what deletionists believe: we don't wish to destroy the encyclopedia, we just want our guidelines to be interpreted strictly (e.g., only scholarly sources of the highest quality should be allowed for an "encyclopedia", although that has become quite an unpopular position). Sławomir Biały (talk) 22:22, 11 April 2017 (UTC)[reply]

Yes you are. Because you include everything unless it must be excluded because it is poorly sourced. The presumption "that information is worth cataloging and including" is inherently wrong: who decides what is worthy? Personally, I would wipe all pornstars from wikipedia. Of course, there is a wide spectrum of inclusionists, from "extreme left" ("include it because my dad told me it was so") via "far left" ("include every person and every pizza hut outlet into wikipedia") to "moderate left" ("include everything verifiable") to "centrist" ("include all verifiable and notable"). Staszek Lem (talk) 23:42, 11 April 2017 (UTC)[reply]

"Yes you are. Because you include everything unless it must be excluded because it is poorly sourced." Wrong. Please check your reading comprehension. If something is well-sourced to scholarly literature (for example, a fact sourced to a standard textbook on quantum chromodynamics), then of course it should be included. I defy you to say otherwise, and if you do I will seriously begin to question your good faith in this discussion, and indeed whether you are here to build an encyclopedia at all. Sławomir Biały (talk) 00:30, 12 April 2017 (UTC)[reply]

Slawomir: Seemingly, your "presumption" could only begin to work if we had one single article entitled, "Everything." We do not, we have millions of articles. "Cataloging", requires more reason than, 'I have a source', it also requires reasons on where and when to include and exclude, and if it's worthy of inclusion at all. So: 1) There are millions of articles we will undoubtedly exclude a particular 'fact', even though it is verifiable. 2) There are a comparative handful of articles where we may include that 'fact' but not just because 'a source exists' (ie. not just, it is Verifiable). 3) There are some number of articles where someone might try to include that 'fact' but in the process create, POV, OR, UNDUE, in-context-trivia, or otherwise skew the article; and 5) there are millions of verifiable facts which will NOT be included, at all. Thus, there is not a single article where variability alone, works, and saying it works contradicts all the Wikipedia content policies, including WP:V itself, which in its introduction says you cannot just consider WP:V in isolation for determining article content. -- Alanscottwalker (talk) 00:17, 12 April 2017 (UTC)[reply]

I will reiterate that, if the prior is that other policies and guidelines are neutral, then verifiability does create a reason for inclusion. So, if information is not excluded under WP:NOT, WP:NPOV, WP:OR, WP:BLP, then of course verifiability creates a reason for inclusion. To rephrase, if there is no reason in policy at all not to exclude the information, then of course being verifiable creates a reason for inclusion. I am astonished that anyone would seriously suggest otherwise. For example, if a fact is printed in a standard textbook on Quantum Electrodynamics, you are of the opinion that that information should not be included, because we don't have a policy that says that facts from standard physics textbooks are the kinds of things that should be printed in an encyclopedia? Nonsense. The presumption is that such facts do belong in an encyclopedia. And I defy anyone here to seriously suggest otherwise. Sławomir Biały (talk)

Other policies and guidelines are never "neutral", they are all in operation simultaneously. What you mean by 'neutral' appears to be, all the other policies support inclusion, and thus you are far beyond Verfiability alone supporting inclusion. As for that fact on Quantum Dynamics, there is no doubt that fact should not be included in millions of our articles (almost all of our articles, in fact). Alanscottwalker (talk) 00:37, 12 April 2017 (UTC)[reply]

I don't think there is anything in conflict here with what I am saying. Verifiability creates a "reason for inclusion", along with neutral point of view, BLP, OR, etc. No one has suggested a binary decision here. But the current wording of the policy is not acceptable: verifiability most certainly is a reason for inclusion. There may be other reasons too, either for inclusion or exclusion. The purpose of the sentence should be to emphasize the need for the policies to work together, not to exclude verifiability from that particular calculus. Furthermore, I still maintain that most of the boring facts that are added to our encyclopedia (e.g., the vapor point of mercury), do not require a complicated calculus of policies to justify their inclusion: verifiability is usually enough. Sławomir Biały (talk) 01:02, 12 April 2017 (UTC)[reply]

• No, or at least certainly not this way. Regardless the merits for a change, the new wording is a terrible one. For a start, the new section title (not a reason for inclusion) doesn't match the new body description (not alone a reason for inclusion), in a way that makes for different and in some way conflicting meanings; I for one could live with the text in the body, but the section title is unacceptable.

Procedurally, the new wording was achieved with a short discussion that was not publicized, therefore achieving limited feedback that could have got us a better text. Diego (talk) 12:19, 10 April 2017 (UTC)[reply]

P.S. Policies and guidelines are descriptive, not prescriptive, and their purpose is to explain the principles and best-agreed practices of the community; the new wording imposes constraints based on a principle that a large amount of editors don't agree to and don't regularly follow, so it is not a description of agreed upon practices. We'd be better off reverting to the previous version, unless we can agree to a new improved wording thanks to this RfC. Diego (talk) 12:59, 18 April 2017 (UTC)[reply]

P.P.S. The original discusion was a local consensus, without an attempt to advertise the changes beyond this page or open a formal process. WP:CONLEVEL says that "Wikipedia has a higher standard of participation and consensus for changes to policies and guidelines than to other types of pages. [...] Changes may be made without prior discussion, but they are subject to a high level of scrutiny. The community is more likely to accept edits to policy if they are made slowly and conservatively, with active efforts to seek out input and agreement from others."

This high level of scrutiny didn't happen, therefore the limited consensus achieved for the change is not enough to gain the "stability and consistency" required at policy text. Diego (talk) 12:17, 20 April 2017 (UTC)[reply]

• No While it is certainly not the reason, it is a reason. While likely unintentional, the new wording suggests a false equivalence between verifiable and unverifiable information. – Philosopher ^{Let us reason together.} 12:47, 10 April 2017 (UTC)[reply]

Very much ^this. YY The people that suggested the new wording so far seem to fail to acknowledge this important point; while I don't see anyone rejecting most of the reasons for the change, the new wording has so many unintended and problematic implications that it should be changed to something more agreeable. Diego (talk) 12:56, 10 April 2017 (UTC)[reply]

• I agree with user:Sławomir Biały's statement, and think that the unaltered wording was clearer. -- PBS (talk) 12:54, 10 April 2017 (UTC)[reply]
• No. The suggested section title may be confusing, at least for people with limited command of English an/or logic. The function/purpose of "Verifiablitity" may be may be described variously; e.g., IMO the descriptor "confirmation" IMO is more appropriate, rather than "reason", because "WO:V" is nearly always a secondary step: Someone adds a piece of info not because they think it is verifiable, but because they think it is useful for encyclopedia. That said, from the discussion I have an impression that the actual purpose of this statement is 'Verifiablity cannot be the only reason for inclusion", to which is the change in the text body actually approaches: "verifiability alone is not a reason for inclusion". And this is why I once reverted this change for being self-controversial: the crucial word here is "alone", which is what many 'ay'-!voters say, but they fail to see the text body contradicts the title. Staszek Lem (talk) 16:32, 10 April 2017 (UTC)[reply]
• No. This change would affect a lot of articles. Also, this would lead to more endless debates on any content. Worse would be more AfDs, especially some failed ones. George Ho (talk) 18:21, 10 April 2017 (UTC)[reply]

George, the result that seems to concern you depends very much on how you interpret WP:V, and the details of the situations it is applied to. This is where examples are needed, but in the preceding discussion no examples were presented where deletion or retention came down to verifiability alone. How this could affect an entire article is quite unclear (examples?), and I think quite unlikely. ~ J. Johnson (JJ) (talk) 21:46, 10 April 2017 (UTC)[reply]

Sorry, I just got emotional about the AfD, so I'll strike that out. Examples I can think are Cold War II (whose topic is very contentious) and some other biographies and other political topics. --George Ho (talk) 21:55, 10 April 2017 (UTC)[reply]

I understand. Do you perchance have any examples where verifiability alone, lacking any other consideration, was the issue in whether to include or exclude any material? ~ J. Johnson (JJ) (talk) 19:44, 11 April 2017 (UTC)[reply]

I can think current events, like United Airlines Flight 3411 (not an accident or crash actually), which is currently AfD-ed, which NOT#NEWS is cited for deletion. Also, Vince (rhinoceros) was nominated for deletion but then kept due to improvements made. George Ho (talk) 02:20, 12 April 2017 (UTC)[reply]

At the Flight 3411 AfD verifiability was not raised at all, and WP:V mentioned only as a qualification passed by numerous articles (thus establishing notability). There was no issue of WP:V at Articles_for_deletion/Vince_. Both articles were "keep" without needing any help form WP:V. ~ J. Johnson (JJ) (talk) 21:10, 12 April 2017 (UTC)[reply]

Other than notability concerns, I can't think of any other issues concerning inclusion based on verifiability alone. That's all. George Ho (talk) 21:27, 12 April 2017 (UTC)[reply]

Which is par for the course. Several editors have expressed concern that broad swaths of text, even whole articles, might be deleted on the basis of the current language, but examples have not panned out. Cases where people vaguely recall some element of verifiability turn out to hinge on other issues, and it appears this change does not "affect a lot of articles." ~ J. Johnson (JJ) (talk) 21:28, 13 April 2017 (UTC)[reply]

It applies to every article where an unexplained deletion has been reverted with the comment "don't remove verifiable content without explanation", which North8000 above agreed is a common occurence. (While North8000 calls this a "false claim", many of us look at it as a perfectly valid practice that the new wording forbids, thus the disagreement). Diego (talk) 13:04, 18 April 2017 (UTC)[reply]

George said "[t]his change would affect a lot of articles", I asked for examples, and it turns out that the ones he had in mind did not actually involve verifiability. It is not clear what you are referring to in "a perfectly valid practice that the new wording forbids," If you are talking about using verifiability for retaining material for which there is no other reason for retaining, well, we have been over that elsewhere. This sub-thread (commenting on George's "no") is not the proper place for zinging off into another direction. ~ J. Johnson (JJ) (talk) 00:50, 19 April 2017 (UTC)[reply]

I was replying mostly to your assertion "it appears this change does not affect a lot of articles", with an example of how it does. I agree it is tangential to your request for examples of large discussions where verifiability alone has played a primary role, though this illustrates that those examples are not the only possible way by which articles are affected by the policy change. Diego (talk) 10:23, 19 April 2017 (UTC) P.S By "a perfectly valid practice" I refer to reverting unexplained removals with a comment like "restore verifiable information", which North8000 dislikes, and with the new wording is "not a reason" for the revert. Diego (talk) 10:29, 19 April 2017 (UTC)[reply]

As to other possible ways "by which articles are affected", I don't recall that anyone has suggested any, even hypothetically. As to the reversion of unexplained deletions, WP:BRD would be the better basis. And I would note that one implication of WP:V is that ALL material added (or re-added) is implicitly verifiable, so a rationale like "restore verifiable information" is actually redundant. As I have been saying, it is not necessary to invoke verifiability for restoring "unexplained removals". ~ J. Johnson (JJ) (talk) 02:37, 20 April 2017 (UTC)[reply]

• The old wording was better than the current wording. Verifiability is a reason for inclusion. Several examples have been given above, and I do not find the attempted refutations of those examples convincing. Verifiability is not the only reason for inclusion, and it can be trumped by reasons for exclusion, but it remains a reason for inclusion. The policy should not flatly say that it is not a reason for inclusion in the section header, nor in the lead. “Verifiability does not guarantee inclusion” is probably the best wording in terms of accuracy, striking the right balance for weighing verifiability, and elegance, but I am open to other qualified statements such as “verifiability is not sufficient for inclusion”.--Wikimedes (talk) 22:17, 19 April 2017 (UTC)[reply]

• No -- the prior version is less wordy and clearer. K.e.coffman (talk) 02:42, 20 April 2017 (UTC)[reply]
• No - The phrase "is not a reason for inclusion" is inaccurate and misleading and counter to what WP:V actually is about. It not only is a reason to include content, but it's actually the first, most fundamental criterion for inclusion we have. Changing it was a horrible idea. Swarm ♠ 03:27, 20 April 2017 (UTC)[reply]

This is why a number of editors feel that the correct term is requirement, not reason. DonIago (talk) 03:47, 20 April 2017 (UTC)[reply]

I think some of the argument is whether that should be correct term. But either way, that is a critical distinction that seems not entirely appreciated. ~ J. Johnson (JJ) (talk) 20:00, 20 April 2017 (UTC)[reply]

No I'll admit it took me a while to parse out what this discussion was about. That said, I think the prior wording is clearer, though I think I largely am okay with the ideas behind either. Hobit (talk) 01:34, 27 April 2017 (UTC)[reply]

• No. Per Bluboar, and the wording itself in the change is simply put, less precise. Guarantee does mean the same as no reason. There is reason to include verifiable content that's why we have a verifiable guide, but there may be not be a guarantee that content qualifies further for inclusion per other policies and guidelines. Our policies and guidelines work in consort with all other policies and guidelines so there is never(almost) a definitive no. I prefer to have agreement before changes are made to policies.(Littleolive oil (talk) 18:04, 1 May 2017 (UTC))[reply]

Making the change was not a good idea

Making the change was not a good idea. No consensus existed for the change: The proposal for wording change on the table in Wikipedia_talk:Verifiability/Archive_65#Proposal was:

Original Proposal: following the third sentence of the lead ("Even if you're sure something is true, it must be verifiable before you can add it."), this text: "However, being verifiable is not sufficient for including something."

This had support and was also considered to be minor enough not to require broader community input.

Supporting: J. Johnson (support inferred as proposer), BrigtRoundCircle, Kmhkmh (with minor alteration), Staszek Lem, North8000, Blueboar, Alanscottwalker, K.e.coffman

Opposing: Someguy1221 (with comment specifically opposing another part of the proposal not related to wording change),

Uneasy about the change: WhatamIdoing.

Two amended changes were then put forward:

Proposal 2a: Change title: "Verifiability does not guarantee alone is not sufficient for inclusion

Proposal 2b: Change first sentence: "While information content must be verifiable in order to be included in an article, verifiability alone is not sufficient for inclusion, and does not guarantee that any content this does not mean that all verifiable information must be included in an article."

Supporting: J. Johnson (support inferred as proposer), BrightRoundCircle, North8000.

J.J. Johnson then made [1] this proposal (in a threaded conversation):

               Okay. How about if my proposed addition is replaced with: "However, while verifiability is required for including something, it is not a reason for inclusion."

               For my additional proposals we can replace "not sufficient for" with "not a reason for". (Though in the text I'm still inclined towards "not a sufficient reason for".) I think most everyone here is generally agreeable, and not going to sweat the details. If you and I agree on the details I'll make those changes.

North8000 strongly supported. [2]

J. Johnson made the change.[3]

Total number of supporters out of ten participants: 2. Total time elapsed from proposal to change: about 13 hours. There was both retroactive support for and opposition to the change. However, making a change under discussion without consensus was not a good idea. The change will not break the encyclopedia, but User:J._Johnson: please put it back the way it was until consensus exists for the change.--Wikimedes (talk) 22:46, 19 April 2017 (UTC)[reply]

Excellent analysis of this fly-by-night change to the lead of Wikipedia's most fundamental policy. Sławomir Biały (talk) 22:52, 19 April 2017 (UTC)[reply]

Bullshit. And a gross misrepresentation of the situation. The discussion leading to the change grew out of a comment by Kmhkmh on 02:10, 8 Feb. ("Verifiability alone is not (and never was) a sufficient criteria for mandatory coverage ....") (see #Where verifiability is not an adequate criterion, above), and went through various proposals, discussion, and amendments over 20 days. There being consensus for the change in the general sense, North8000 and I worked out the final details with the tacit approval of the rest. Your characterization of this as a quick trick by only two editors is not only flat out wrong, it manifests bad faith. ~ J. Johnson (JJ) (talk) 03:25, 20 April 2017 (UTC)[reply]

I did not characterize, nor do I think of it as a quick trick. The fact remains that the change you made was not the change that had consensus. You might read again the post by Kmhkmh that you linked; it does not say that verifiability is not a reason for inclusion, it says that verifiability alone is not a sufficient reason for inclusion. I have read most of your posts above (not all below) and I believe that you believe that this is the same thing as saying that verifiability is not at all a reason for inclusion, but I hope the discussions over the past several weeks have convinced you that many editors disagree with this equivalence.

As for tacit approval: 13 hrs is not enough time to gauge tacit approval. The fact that several participants in the discussion have since disagreed with your change indicates that the approval was misjudged. Such a misjudgement is easy to make after a long discussion on a topic one feels strongly about, especially if some of the other editors don't follow the same logic that you do.

I'm OK with tacit approval in principle: In the "original proposal" I mentioned above you did the right thing and asked if a formal RfC was necessary. One person said that it was not, and the other respondents did not mention it. You had tacit approval for making a change without a formal RfC. Latecomers might have disagreed with the change, or the fact that it was made without wider community input, and I might have been one of them, but I would not have faulted you for making the change.--Wikimedes (talk) 07:31, 20 April 2017 (UTC)[reply]

What a fine piece of wikilawyering. If we had had a full-on RfC, running a couple of months with dozens of editors, resulting in a specific block of text, and if someone being directed to make that change noticed there was some kind of trivial error, and said: "Oh, it looks like a small detail needs to be corrected; it looks uncontroversial, but I'll wait a couple of hours in case anyone objects", and then, no objections being raised, and making the change: would you then object to that change as being made by one person with only a "couple" of hours of notice? In fact we had 20 days of discussion that considered and then narrowed the scope of what we were trying to achieve, with tacit approval for North8000 and I to work out the final details. No one has complained, either then or since, that we went beyond that scope. For you to characterize the final result as "13 hrs" is totally superficial, and such a blatant mischaracterization as amounts to a knowing and willful deceit. Ten editors participated in the original discussion, out of some 2,000 page watchers who presumably were aware of that discussion, and my recollection is that there were only two dissents. That sure looks like consensus.

Which all is actually rather beside the point. While Marshall originally complained of the process by which this change was done, in this RfC he asks only if the idea is good (or not), not if making the change was good or not. So it appears you are off-topic. ~ J. Johnson (JJ) (talk) 19:55, 20 April 2017 (UTC)[reply]

Ouch. Well, you've stated your position twice and I've stated mine twice, and we can leave it there. We still have one unaddressed point of disagreement over whether the change procedure is within the scope of this RfC; I think that although of secondary importance to the content of the change, it is in scope. But we're unlikely to convince each other, and I'm OK with your refactoring, so I'll leave it at that.--Wikimedes (talk) 21:45, 20 April 2017 (UTC)[reply]

I am not okay with your gross misrepresentation. Your "13 hours" characterization is not a mere difference of viewpoint, nor an "alternative fact", it is flat out wrong.

As to the general question of process: note that I am not saying that it cannot be raised; I am saying only that it is not within the scope of this RfC as formulated. Which could be taken as a fault of the RfC for having inadequate scope. ~ J. Johnson (JJ) (talk) 21:36, 21 April 2017 (UTC)[reply]

Discussion

This diff is germane.—S Marshall T/C 22:17, 9 April 2017 (UTC)[reply]

Hardly. That diff is your comment at WP:ANI#User:Spacecowboy420 blanking articles [now archived], where you make a distorted and faulty interpretation that no one else buys into. It seems like you are trying to generate an issue there (where I don't believe there was any issue of WP:V) which you can cite here. It rather seems like you are citing yourself as an example of how people can misinterpret WP:V. ~ J. Johnson (JJ) (talk) 23:28, 9 April 2017 (UTC)[reply]

• I have noticed something... the policy uses the term "inclusion" (and thus, through omission, its opposite: "exclusion")... but in our discussions most of us keep using the terms "retention" and "removal". What is the difference? ... retention/removal is a more specific sub-set of inclusion/exclusion that occurs only after information has already been included. Is that, perhaps, the root of the disconnect here? We are not actually focused on inclusion, but on removal? Blueboar (talk) 00:51, 10 April 2017 (UTC)[reply]

It is unclear what we are focused on here because a certain editor has generally failed to make those distinctions. ~ J. Johnson (JJ) (talk) 07:56, 10 April 2017 (UTC)[reply]

My observation was based on the comments by multiple editors (including my own)... not those of any one particular editor. Blueboar (talk) 10:32, 10 April 2017 (UTC)[reply]

Likewise. None of us are perfect in using terms and concepts precisely and consistently. It's when these are left ambiguous, or even unacknowledged, from the start that we keep going around and around and .... ~ J. Johnson (JJ) (talk) 20:35, 10 April 2017 (UTC)[reply]

• The title of the RfC section should be changed. The "yes" and "no" sections are inverted relative to the title question. Sławomir Biały (talk) 11:19, 10 April 2017 (UTC)[reply]

I've changed the title to avoid this problem, it no longer is a yes/no question itself. Diego (talk) 13:33, 10 April 2017 (UTC)[reply]

RfC design discussion

Gotta say I'm surprised at the neutrality of the RfC, I was expecting it to be biased and was pleasantly surprised. However, "was this a good idea" is the canonical bad way to present an RfC. The all-or-nothing approach is the very reason the discussion above is stymied. A better approach would be tackling the issues one at a time (like the original discussion that led to the change) instead of swallowing it whole as "good idea"/"bad idea". Bright☀ 14:27, 9 April 2017 (UTC)[reply]

If you can come up with a third option that would satisfy everyone, please suggest it. Blueboar (talk) 15:39, 9 April 2017 (UTC)[reply]

"Satisfy everyone" is a tall order, but I've already suggested tackling the changes one by one (as they were originally tackled) instead of all-or-nothing. Anyway this RfC is already rolling so let's just roll with it. Bright☀ 20:51, 9 April 2017 (UTC)[reply]

The problem is not just in being presented in a simplistic either-or form, but (independent of the number of options) in the failure to provide any background or other information to give the respondents any basis for an informed decision. There are reasons why this change was made (and it was both well-considered and consensual), but Marshall has not even provided a link to that prior discussion. There is also question of whether this is a good time for this RfC, as the last month's contentious and unproductive discussion on this has led to discussion fatigue, so any extended discussion on this now will likely be of sub-standard caliber. ~ J. Johnson (JJ) (talk) 22:15, 9 April 2017 (UTC)[reply]

Well, It is fine SMarshall raises an RfC, and feel free to link to that prior discussion. Alanscottwalker (talk) 22:19, 9 April 2017 (UTC)[reply]

Let him clean up his own mess. ~ J. Johnson (JJ) (talk) 07:05, 10 April 2017 (UTC)[reply]

S Marshall, since you initiated this RfC, please could you fix the RfC opening statement in two ways: (i) to not include that table; (ii) so that a signature (or a valid timestamp, at the very least) appears before the "Yes" section. To see why, have a look at how the RfC appears at Wikipedia:Requests for comment/Wikipedia policies and guidelines (Permalink). Both of these are also explained at WP:RFC. --Redrose64 🌹 (talk) 22:48, 9 April 2017 (UTC)[reply]

I've added the timestamp. I don't see a provision prohibiting tables at WP:RFC, and I note that it's one of the recommended options at Wikipedia:Requests for comment/Example formatting. However, you're welcome to fix it so that it displays correctly at Wikipedia:Requests for comment/Wikipedia policies and guidelines.—S Marshall T/C 23:17, 9 April 2017 (UTC)[reply]

It's in WP:RFC#Statement should be neutral and brief (which was linked by BrightR (talk · contribs) above): "For technical reasons, statements may not contain tables or complex formatting, although these may be added after the initial statement (i.e., after the first date stamp)." If you follow the permalink that I provided, you'll see that nothing is displayed after the opening brace that starts the table. --Redrose64 🌹 (talk) 23:22, 9 April 2017 (UTC)[reply]

I've moved the signature above the table as requested. If this doesn't fix it, feel free to make whatever change you need to resolve the transclusion issue. My wikimarkup-fu is weak, and you admins are supposed to be good at this.  :)—S Marshall T/C 23:29, 9 April 2017 (UTC)[reply]

While verifiability is required for inclusion it is not the only thing required for inclusion (content must also be notable). Doc James (talk · contribs · email) 23:35, 9 April 2017 (UTC)[reply]

Actually, Doc... according to our WP:NOTABILITY guideline, content does not need to be notable... article topics need to be notable. Perhaps what you mean is "content must also be worth noting". (it is a subtle, but important distinction). Blueboar (talk) 01:00, 10 April 2017 (UTC)[reply]

@S Marshall: that this novel wording requires community consensus and in the absence of consensus, should be reverted to the stable version. There was no absence of consensus, you are wording it backwards. There was consensus for the current version, and you objected on the basis "I missed the discussion", which is not a valid reason. This has already been pointed out to you. This RfC is about achieving a new consensus, as the current consensus already favors the change, even if you weren't there for the discussion. As it stands, from what I see, there is no consensus for the revert, and the current version is preferred anyway. Bright☀ 19:28, 19 April 2017 (UTC)[reply]

I propose that the guideline should say what I think people mean when they say "Verifiability is not a reason for inclusion" in this discussion. What they apparently mean is that "Verifiability alone is not a sufficient reason for inclusion." Is that an acceptable alternative to the current (unclear and literally wrong) wording? Sławomir Biały (talk) 19:55, 19 April 2017 (UTC)[reply]

Of course, I entirely disagree with BrightR, and of course, I join issue with him. I understand why his position is that I'm challenging a consensus that was established after due process, but I see his position as not closely aligned with the facts. My position is that this was a major edit to a policy; that was made without the RfC that should normally accompany a major edit to the policy; and I challenged it reasonably promptly after it was made. I think custom and practice is on my side here. I also concur with Blueboar when he says we need to let some other people have their say, as we're mostly repeating ourselves now.—S Marshall T/C 20:26, 19 April 2017 (UTC)[reply]

It was felt that the change was not "a major edit to policy", but a clarification. The perceived significance of the change depends on various and differing interpretations of the wording, the policy, and the effects. A major defect of this RfC (and why it has become such a muddle) is in failing to identify the caveats on which it is based, to the result that we still don't have any common understanding of just what we are discussing. ~ J. Johnson (JJ) (talk) 03:30, 20 April 2017 (UTC)[reply]

While I feel J. has allowed personal feelings to interfere with some of his postings on this matter (no offense intended), I strongly agree with the last part of this sentence. This RfC appears to be tackling multiple issues with multiple interpretations simultaneously, to the point where I don't even know how it could reasonably be closed other than "no consensus for any change to current wording" unless we rein things in and more clearly elucidate both terminology and intent. DonIago (talk) 03:56, 20 April 2017 (UTC)[reply]

On the contrary, the RfC is a very simple Yes/No question. There was a change made to a core policy through a WP:LOCALCONSENSUS, that was challenged just after. The process here should determine whether there's community-wide support for that change in the form of approval through a RfC. While it's true that a "no consensus" outcome would not be able to change the wording, it is clear that it would make it unenforceable and useless as guidance (since "editing a policy/guideline/essay page does not in itself imply an immediate change to accepted practice").

If a single thing is obvious at all in this discussion, it is that the change does not have the "high level of consensus" required by WP:PROPOSAL for new rules in policies and guidelines. The proponents of the change failed to follow the "Good practice for proposals", which is to "create a proposal that has a solid chance of success with the broader community, [then] start an RfC for your policy", and at WP:PGCHANGE "major changes should also be publicized to the community in general"; and now are fighting tooth and nail to keep their entrenched version against legitimate opposition. There is nothing good coming from that attitude. Diego (talk) 09:58, 20 April 2017 (UTC)[reply]

I disagree on several levels. First, the wording of the RFC is what it is (flaws and all), not new things created in this talk area. Next, the current wording (instituted March 2nd) merely accurately summarizes the actual policy, it does not change it. Third, the current wording was discussed with 27,000+ words for 23 days (Feb 27th - March 2nd) and instituted March second (at the time with no objections) and is the current wording; phrasing it as a "proposed change" is not correct. Sincerely, North8000 (talk) 10:21, 20 April 2017 (UTC)[reply]

the current wording (instituted March 2nd) merely accurately summarizes the actual policy The idea that the wording is an accurate summary of the policy has direct opposition in this RfC.

it does not change it The idea that the wording is merely a minor change or no change at all has direct opposition in this RfC.

the current wording was discussed with 27,000+ words for 23 days - by a small amount of editors, with no announcement to the wider community; and as described by Wikimedes above, the current wording was supported by a total of two editors, with several other editors agreeing to different wordings, not this particular one. Therefore, the idea that the previous discussion was enough level of consensus to satisfy the requirements for a change in policy has direct opposition in this RfC, when seen in the light of WP:PGCHANGE.

A honest good-faith attitude would be to acknowledge that the changes have failed to gain the wide support required by policy, agree to change to a neutral version (either the previous long-standing one or a full removal of the disputed sentence), and work towards finding a new wording acceptable to roughly all editors participating in the RfC. Diego (talk) 11:09, 20 April 2017 (UTC)[reply]

It does not really matter whether the February change had a proper consensus or not... because WP:Consensus can change... Even if we say that there was a consensus for the "Reason for inclusion" wording back in February, what is clear now (in April) is that the "Reason for inclusion" wording no longer has consensus. What is not clear yet is whether we should return to the old "not a guarantee of inclusion" wording, or go with some other (new) wording. That may take a follow up RFC. Blueboar (talk) 13:19, 20 April 2017 (UTC)[reply]

Having not commented here for almost a week, it is diappointing there has not been movement toward a compromise, such as the "Alternative" below. There is much agreement among the 'sides', but it keeps coming back to extended disagreement, unfortunately. Alanscottwalker (talk) 13:37, 20 April 2017 (UTC)[reply]

I agree that consensus can change. But note the second sentence there: "On the other hand, proposing to change a recent consensus can be disruptive.". (Especially when poorly thought-out.) Note especially that this RfC does not propose to change any consensus, nor any language; it asks only if the change was a good idea. If it was, then presumably there is no need to undo anything. If not, then presumably we might proceed with what to do, but that is not actually addressed here.

My deep objection here is not to possibly changing anything, but to any attempt to force a change based on such wretched discussion, where we still do not have any definition of terms, identification of concerns (let alone any examples of those concerns) or relevant sub-issues, and certainly not any statement of how the issue is to be resolved and what should ensue. This RfC is defective, which is the main reason why the discussion has been so disruptive; it would be an extremely poor justification for any result. ~ J. Johnson (JJ) (talk) 20:36, 20 April 2017 (UTC)[reply]

Blueboar: This RfC is worded in the canonical bad way to present an RfC. "Was this a good idea? Please discuss" and all-or-nothing approaches almost never lead to consensus. The lack of consensus on this RfC does not nullify the consensus of the previous discussion. A new consensus needs to be reached, not "no consensus". If and when this RfC reaches its "no consensus" conclusion, it should not lead to a revert, but to further discussion, hopefully after a break. Bright☀ 11:55, 21 April 2017 (UTC)[reply]

Which is why I said that a follow up RFC will be necessary. My take on S Marshall's question is that he is essentially asking "Does the current language (still) have consensus?" And looking at the replies, it is clear to me that the answer is "no, the current language does not have consensus"... However, having answered that yes/no question, we are left with another question: What next? I agree that we have yet to reach consensus on what to do next. Should we simply revert to the old text? Should we reword to something new (and if so, what)? Should we remain silent on the issue? The question of what to do next still needs to be addressed, but it will take a follow up RFC to address it. Blueboar (talk) 12:21, 21 April 2017 (UTC)[reply]

That is undoubtedly Marshall's view, but it is not what he asked. And to grant credence to that which is not stated is to blind us as to what the issue is. It would be like asking a jury for verdict, and only afterwards telling them what their decision applies to.

As to any lack of consensus: the thrust of all these discussions is to show a loss of support for the prior language. At least not as some people implicitly interpret it. To resolve that will take a broader (and better structured) discussion. Unfortunately, the ground is so chewed up now that any discussion that is simply an extension of this one is just going to get mired. What is needed next is break – a rhetorical fire break — so that the next discussion can proceed on its own terms, without carrying over the deep flaws and bad feelings of this discussion. ~ J. Johnson (JJ) (talk) 21:45, 21 April 2017 (UTC).[reply]

I'll be gone and off the grid & wiki for 4 days. Everyone here is simply doing what they think is best for the good of Wikipedia. I think that the current wording is clearly the best of all considered but am open to better ideas or a compromise.North8000 (talk) 12:34, 20 April 2017 (UTC)[reply]

Moving forward

If this RfC ends with "no consensus", what then?

• There was previous consensus for a change (in a discussion spanning about 20 days, which three weeks later was contested).
• The policy WP:PGCHANGE suggests (but doesn't explicitly state) that the existing consensus was reached correctly, since the result of the discussion was clear.
• The policy WP:NOCON suggests (but doesn't explicitly state) that when there's no consensus, no action should be taken except in the specified cases.
• The policy WP:EDIT favors "existing consensus" or "present consensus", and does not mention "static version"/"long-standing version"/"stable version".

The RfC is interpreted by its proposer in a manner inconsistent with these policies. Even if the previous consensus does not reflect current consensus (it's certainly possible), a "no consensus" result does not mean a revert to the "previous-previous" consensus. Bright☀ 12:55, 21 April 2017 (UTC)[reply]

This is the most fundamental policy on Wikipedia. It is important that every aspect of it have solid consensus. As already observed, the consensus for the original edit in question was not particularly solid. It was not advertised to the wider community in an RfC, and it changed this policy in a significant and fundamental way. The old consensus wording should be restored. If there truly is solid consensus for the new wording, then that consensus should be demonstrated in a request for comment in which the entire Wikipedia community is invited to participate. Sławomir Biały (talk) 16:34, 21 April 2017 (UTC)[reply]

• "What then" is not up to any of us. We need to let the closer deal with that. I have suggested before that this RfC should be closed by a triumvirate of experienced admins, and I would like to repeat that suggestion now.—S Marshall T/C 17:12, 21 April 2017 (UTC)[reply]

Sławomir Biały, I've noted above that there's at least one fundamental aspect of this policy ("Consensus may determine [...] include disputed content") which was never discussed, before or after it was put in the policy. It's been inserted, reverted, and then inserted again, with consensus through silence. Claiming that the 20-day discussion is any less "solid" than the discussion for most other parts of this policy is disingenuous. There was more discussion about it than many other parts of the policy, including the wording of that section before that discussion (which lasted for less time and included less participants). By your reasoning, the old consensus needs to be rejected in favor of something that was RfCd... but this section was never RfCd, the most thorough discussion it had was the one you're contesting. Feels... disingenuous. Bright☀ 13:19, 22 April 2017 (UTC)[reply]

As S Marshall correctly points out, what to do next will depend on what the closers have to say to us in their closure. Obviously, those of us who support the current language want them to say "Yes, the current language had consensus, and thus should be kept", while those who support the older language want them to say "No, the current language did not have consensus, and thus the policy should return to the older language"... but despite the yes/no wording of the RFC, the closers have more options available to them than just "yes" or "no". They might tell us that neither has consensus, and that we all need to compromise and find some new consensus language (which will require a follow up RFC). Or, they may tell us to do something that none of us has thought of yet. In other words... it is premature to ask "what next". We need to have some patience... we won't know what to do next until we hear what the closers have to say.

That said, given the heated nature of the exchanges so far, I fully endorse S Marshall's call for a three admin closure. Blueboar (talk) 14:06, 22 April 2017 (UTC)[reply]

The proposed wording of the new second sentence of the guideline was made in this post to WT:V, and then it was implemented 13 hours later. There was a discussion that preceded that edit, with a number of perhaps better options, but given the lack of community support now that this has been subjected to an RfC, the only solution is to go back to the previous stable RfC version. The lead of the WP:V was the subject of this formal RfC. If the community wants to change the WP:V policy in a fundamental way as to exclude verifiability as a possible reason for including information in an article, then it should also be through the RfC process, not an apparent local consensus among various options only one of which is the current wording. Indeed, it is actually difficult to find any discussion of the present wording of the guideline, let alone find a place where to !vote against its inclusion. In particular, in that discussion I do not see much that I would claim is "consensus" for the wording that "Verifiability is not a reason for inclusion", which is what the current policy says. Perhaps this is all obvious to those few editors who were involved in that discussion, but if an outsider can't even find it, I find it hard to swallow that it represents any kind of Wikipedia-wide consensus in the way that the result of the previous RfC can claim to be. So, in conclusion, without a similarly strong consensus for this radical change to the policy, I do not see how a 13 hour period between a proposal and edit can possibly be placed on equal footing with the result of a widely-publicized RfC that led to the former stable version of the guideline, particularly not now that the changes are contested by quite a few editors. I think the only reasonable solution is to go back to the stable version. If there is consensus for a change to the policy from the one established at that RfC, then it should demonstrate a similarly high level of support in which the whole community is invited to participate, not a 13 hour period between proposal and implementation based on a (fairly weak) local consensus only. Hopefully this clarifies my position, and I have no doubt that a panel of administrators will unanimously agree with this assessment of the situation. Sławomir Biały (talk) 14:34, 22 April 2017 (UTC)[reply]

The RfC you point to is about the lead, not WP:ONUS. There is no "previous stable RfC version" for WP:ONUS. The wording in WP:ONUS was created in this discussion, not an RfC, where the wording was this, which slowly changed to this, without an RfC, and the last sentence without even as much as discussion, only consensus through silence. If you seriously suggest all consensus other than RfC is invalid, there is no reason to revert to that particular version, but to delete the section entirely. However RfC is not the only valid consensus, and a discussion among ten people is larger than many discussions for any previous similar alterations (some of them without discussion at all!). The claim that this is a "previous stable RfC version" is disingenuous, to put it mildly. Bright☀ 15:19, 22 April 2017 (UTC)[reply]

I am not being disingenuous, and remind you please to assume good faith. This RFC discussion concerns changes made to the lead of the verifiability policy. There is a very clear diff of that change in the statement of the RfC. I do not believe that there was adequate consensus for that change, to override a previous RfC on the lead. The fact that this RfC fails to demonstrate any consensus for it merely shows that there was none to begin with. Our default position should therefore be to revert to the previous stable version if the lead until a new (and much clearer consensus) forms, probably as a result of a new RfC, since stating that verifiability is not a factor in determining whether something is suitable for Wikipedia seems to be a fundamental change that would require broad community-wide support if it to be implemented. Sławomir Biały (talk) 15:55, 22 April 2017 (UTC)[reply]

{{hat}}

You are being disingenuous. And also rather one-sided where you ask others to assume good faith even though you do not do so your self (e.g., here). I will remind you that under Behavior that is unacceptable in the WP:Talk page guidelines "No personal attacks" is followed by "Misrepresentation of other people, which you also continue to do. Your comments do not further any resolution of this matter, and violate WP:TPG.

I note in passing that your questioning of the process by which the text was changed is actually off-topic. As I noted above, this RfC only asks if the idea is good, not if making the change was good. ~ J. Johnson (JJ) (talk) 20:57, 22 April 2017 (UTC)[reply]

Do you claim to be special, that you can not only insinuate bad faith in others and misrepresent them, but you can freely judge the relevance of, and hat, comments you don't like? (As you just did to mine.) Are the rest of us allowed to play by these new standards you seem intent on setting? If so, then I should be allowed to hat some of your comments, right? ~ J. Johnson (JJ) (talk) 23:12, 22 April 2017 (UTC)[reply]

I don't think this even merits a reply. But I will say that I would not expect to continue to receive aggressive replies from someone who already wrote me off as too incompetent to be worth spending any more time on. I shouldn't think that what such a person writes would matter in the least to you. Anyway, I think from the beginning of your post in this section, this discussion should be hatted as completely off topic. I know I'm certainly done caring what you think. Sławomir Biały (talk) 01:32, 23 April 2017 (UTC)[reply]

I did indeed wonder (20 Apr.) if it might be beyond your competence to "just line up the original with what I quoted and show us the difference" (this in response to your accusation that I had misquoted you). A simple demonstration could easily settle that, perhaps even the allegation of misquotation; the lack of same speaks for itself. As to taking offense: you set the standard with your imputation on the 14th of "either a WP:CIR failure on your part, or is just bad faith trolling." If you play rough expect to get bumped. ~ J. Johnson (JJ) (talk) 18:47, 23 April 2017 (UTC)[reply]

I'm unclear why you persist in talking about quoting me out of context, since it seems to bear no relation to the subject under discussion. Sławomir Biały (talk) 21:35, 23 April 2017 (UTC)[reply]

The discussion was about the disingenuousness of your argument. Following your lead we are now talking about your tendency to question the good faith and competence of other editors when you don't like their comments. Is that clear enough? ~ J. Johnson (JJ) (talk) 20:43, 24 April 2017 (UTC)[reply]

I did not question the competence of BrightR. I did question your competence after repeatedly bringing up the same refuted misrepresentation, and quoting me out of context. But thanks for the clarification anyway. Sławomir Biały (talk) 01:45, 25 April 2017 (UTC)[reply]

You are such a sly, slippery fellow. Of course you did not question BrightR's competence, you only reminded him "please to assume good faith", as if he were the one that needed reminding. And you have slipped around the little fact that you introduced "competency" into this discussion (at 01:32, 23 Apr.). And I see that you, having failed to demonstrate in the previous discussion any actual misquotation, are now sliding into "quoting me out of context." Such adroitness feels (as BrightR said) ... disingenuous. ~ J. Johnson (JJ) (talk) 23:38, 25 April 2017 (UTC)[reply]

{{hab}}

For the second time Sławomir Biały has hatted my comment (with the reason "Offtopic personal attacks"), which I am unhatting. Sławomir, it is not for you to censor someone else's remarks when you don't like them. Your style of argument was criticized, to which your response was to impugn the good faith of another editor, just as you have done previously. We can hat this section when you acknowledge the criticism and abandon your incivility. ~ J. Johnson (JJ) (talk) 21:14, 26 April 2017 (UTC)[reply]

For the record, this not just whining of J.Jonson. I was a subject of several personal attacks of Sławomir in this page as well. This really poisons the discussion. Slawomir, please refresh your grasp of WP:NPA ASAP. Staszek Lem (talk) 17:10, 27 April 2017 (UTC)[reply]

Slawomir, IMHO the February change was not a policy change; it was aligning summation type wording with the actual policy. North8000 (talk) 20:59, 24 April 2017 (UTC)[reply]

Yes, aside from a tiny matter that there seems to be differing views on what the policy actually is. Or should be. ~ J. Johnson (JJ) (talk) 21:38, 24 April 2017 (UTC)[reply]

To return to the question of what next: the exact and entire question asked in this RfC (in regard of the contested change) is: "Was this a good idea? Please express a view below."

Nothing more is asked. While some participants here have an anticipation that disapproval should result in removal, there is no basis for that in the RfC. The arguments about a supposed lack of consensus for the change are outside the scope of this RfC.

What the views expressed have shown are some differences in interpretation and even fundamental assumptions (which certainly predate the change, and do not follow from it). As these issues evoke strong reactons, it would be a useful (if possible) next step to sort out and clarify them. Even more so, examples of the rather speculative concerns that have been raised should be collected and examined to see just what the problems are, and if their derivation or remediation depends on or requires any particular interpretation of WP:V.

What we do not need is another poorly formulated RfC. I think what we need most is no RfC for several months, lest it become just a continuation of the current debacle. (Serveral months might be needed just to prepare, as described above, while lack of preparation would make it senseless.)

For those who insist the current wording will have all kinds of dire consequences: if so, then six months should provide plenty of examples, which would greatly strengthen your argument. ~ J. Johnson (JJ) (talk) 21:44, 24 April 2017 (UTC)[reply]

What I see moving forward: two things are very clear:

• One, there are existing Wikipedia policies that exclude some verifiable material—WP:NOT. Those who think all verifiable information belongs on Wikipedia are at odds with current policy.

Information should not be included in this encyclopedia solely because it is true or useful. A Wikipedia article should not be a complete exposition of all possible details, but a summary of accepted knowledge regarding its subject. Verifiable and sourced statements should be treated with appropriate weight.

• Two, in case of an editing dispute, consensus is needed for the inclusion of the disputed material—the (not-currently-disputed) last sentence of WP:ONUS. Those who think all verifiable information should be included until there is consensus to remove it are at odds with current policy.

Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content.

These two ideas can be summarized as verifiability does not guarantee inclusion or verifiability is not a reason for inclusion or verifiability is necessary but not sufficient for inclusion. The last one, "necessary but not sufficient", has been suggested and supported by people on both sides of the RfC, so it seems like the best WP:CONSENSUS to use it. Bright☀ 20:31, 25 April 2017 (UTC)[reply]

I could go with "necessary, but not sufficient". Perhaps better if that could be offered as a clean RfC. ~ J. Johnson (JJ) (talk) 21:17, 26 April 2017 (UTC)[reply]

Alternative suggestion

IMO a better way is to combine the two versions (as explained in my 'nay'!vote above):

Title : Verifiability does not guarantee inclusion

Body: While content must be verifiable in order to be included in an article, verifiability alone is not a reason for inclusion, and does not guarantee that any content must be included in an article.

Staszek Lem (talk) 16:44, 10 April 2017 (UTC)[reply]

1. Support. Per my prior comments. Alanscottwalker (talk) 18:49, 10 April 2017 (UTC)[reply]
2. The current version is better but I'm worn out and in the mood to cave on a compromise if that would settle it. North8000 (talk) 19:08, 10 April 2017 (UTC)[reply]
3. Oppose. Partly because I'm getting cranky and disinclined to yield anything to a poorly-organized and ill-advised RfC. But largely because the "verifiability alone" phrasing suggests that (with other arguments) verifiability is a reason for inclusion. The point here is that verifiability should only open the possibility of inclusion. ~ J. Johnson (JJ) (talk) 20:18, 10 April 2017 (UTC)[reply]

   You are precisely correct.North8000 (talk) 20:28, 10 April 2017 (UTC)[reply]

4. I could live with it... but would prefer it to say: "Verifiability is not a guarantee of retention", and "verifiability alone is not a reason to retain problematic material". That's really what we have been discussing. Blueboar (talk) 20:47, 10 April 2017 (UTC)[reply]

   In my understanding of English "inclusion" covers "retention"; in any case both must be covered, so that people thingk twice before dumping trivia into articles. Staszek Lem (talk) 19:48, 11 April 2017 (UTC)[reply]

   for those of us who are getting fatigued (and I include myself in that)... perhaps it is time to stop trying to convince each other, and to instead sit back and let others comment for a while.Blueboar (talk) 20:41, 10 April 2017 (UTC)[reply]

   The discussion that led to the current language wasn't this fatiguing. And faulty rhetoric is pernicious, it should be resisted at every step. ~ J. Johnson (JJ) (talk) 19:38, 11 April 2017 (UTC)[reply]

re: verifiability should only open the possibility of inclusion - please explain the difference between "opening a possibility" and "being a reason", so that we can arrive at a better mutual understanding and hence a consensus. Also explain why you think this distinction is important for the policy (eg what bad consequences may come out of their confusion). Staszek Lem (talk) 19:42, 11 April 2017 (UTC)[reply]

P.S. It dawns upon me that this issue is a watershed between "inclusionists" and "deletionists": For an "inclusionist" , verifiability is a reason for inclusion, and the OP has to work hard to prove why something must be excluded. For a deletionist it is vice versa: you have to have really solid reasons to include something, so that when it is in, no policy can kick it out. Staszek Lem (talk)

I'm an "inclusionist" and I believe verifiability is not a sufficient reason for inclusion. Bright☀ 01:34, 12 April 2017 (UTC)[reply]

That could use some explaining. ~ J. Johnson (JJ) (talk) 01:54, 12 April 2017 (UTC)[reply]

I'm also an "inclusionist" and I believe verifiability is not a sufficient reason for inclusion. But that is not what this about. The current wording simply says that after wp:ver has been met, wp:ver "steps aside" on the include/exclude question. Sufficient reason for inclusion can be something as simple as "it's a plus for the article" or "we're building an article"

Not quite. Your characterization is spoiled by your caricaturization of the alternatives. On one side, it is the nature of a guaranteed inclusion that "no policy can kick it out." (Roughly speaking. "Policy" is not the right word here.) On the other side, no one is claiming that "you have to have really solid reasons to include something". But you miss the point: what we are saying is that all the argument about why something should be included (or excluded!), the reasons for doing so, and how "strong" those reasons have to be, should be entirely outside of mere verifiability.

I don't understand why you seem unable to grasp something so simple, but I will try one more time. First, if something is not verifiable, it does not belong in the encyclopedia at all, and lack of verification is therefore sufficient grounds for removal. Which is to say, verifiability is required for inclusion, and sufficient for exclusion. Okay? The issue is whether any material, having met this minimum requirement of verifiability (and not excluded by any other policies), is an automatic home run. I am saying that verifiability only puts you on first base, where you then have a possibility of scoring a home run. Or if you want a deeper analogy: you have to be on the eligibility list before you can swing at the ball. Is that clear enough? ~ J. Johnson (JJ) (talk) 02:02, 12 April 2017 (UTC)[reply]

• Yes, I agree that "The issue is whether...". However you are not answering my question on the difference between "opening a possibility" and "being a reason" (notice the indefinite article; your "home run" analogy is synonymous with "the reason", not "a reason"). Staszek Lem (talk) 17:43, 12 April 2017 (UTC)[reply]
• I think what you describe as "something so simple" is perfectly understood by all in this discussion, and is in fact accepted by us opposing the change. No one here is saying that a verifiable fact must be included for being verifiable.

What we oppose is the idea that, when a fact has been found verifiable, that has no bearing at all in our weighting of its merits for inclusion. It's all a matter of degree; you're saying here that verifiability should not put the dial for inclusion up to 100%, to which we all agree (in fact, the old wording "not a guarantee for inclusion" described this much better). But the change has instead moved the dial all the way down to 0%, which is the source of the disagreement; many of us think that finding a fact supported by a reliable source provides some value for inclusion in between 0 and 100%, with the amount being larger depending on the higher quality of the source.

That recent change does not reflect how many editors work, and thus is failing to describe the actual practices of the community - which is what the policy should be doing. It should be worded in a way that reflects this disagreement, or at least remaing ambiguous about it, not taking any position. Diego (talk) 06:25, 12 April 2017 (UTC)[reply]

• No one is suggesting that WP:V creates an "automatic home run". But the current guideline does not say that verifiability does not create an "automatic home run". Instead it says that verifiability is not a reason for inclusion, which is to say that it is not a reason (among others) for including material, and instead can only be used to exclude it. There is a huge gap between "a reason" and "a home run". In court, lawyers on both sides of an argument will discuss reasons for things, but surely you're not suggesting that anything that has a reason behind it is always therefore a home run. This is, in fact, the same kind of fallacious black-and-white thinking of which Stanisław is guilty. Sławomir Biały (talk) 12:25, 12 April 2017 (UTC)[reply]

of which Stanisław is guilty I strongly protest against personal attacks and misrepresenting my position. Please strike it out. Here I am, starting a subsection in an attempt to find a middle ground, and suddenly I am an extremist. Staszek Lem (talk) 16:27, 12 April 2017 (UTC)[reply]

It may help to discuss not only what we think the policy should say... but why we are saying it. Looking back to when we first added the paragraph, the purpose was this:

WP:V is very often the first policy that new editors are pointed to (usually because they didn't realize that we want citations, and they added a fact that needed a citation). The new editor quickly discovers that if you don't provide a source, someone might come along and challenge your contribution. It might be removed. They assume (incorrectly) that if "Don't provide a source and your contribution might be removed" is true, that the opposite must also be true - "provide a source and your contribution won't be removed". They don't realize (at this point in their wiki-history) that there are other reasons, besides verifiability that might cause their contribution to be removed. And they often get upset when (having provided a source) they find that a contribution is still being challenged and removed.

So... we added the "not a guarantee" language ... simply to say - WARNING: Verifiability is just the beginning of the conversation when it comes to deciding whether to keep or remove a contribution... it isn't the end of the conversation. There are other policies and guidelines that might cause your contribution to be removed. Blueboar (talk) 12:46, 12 April 2017 (UTC)[reply]

So ... why don't we just say that? " Verifiability is just the beginning of the conversation when it comes to deciding whether to keep or remove a contribution... it isn't the end of the conversation. There are other policies and guidelines that might cause your contribution to be removed., followed by the consensus line. – Philosopher ^{Let us reason together.} 14:53, 12 April 2017 (UTC)[reply]

well... it's a bit clunky... and when we first added the "guarantee" wording we thought we had come up with a way to get the same warning across using phrasing that sounded a bit snappier. But if everyone prefers the clunkier (but perhaps more direct) language, you will get no objection from me. Blueboar (talk) 15:47, 12 April 2017 (UTC)[reply]

Maybe we have something here, but first we should clarify exactly what the idea is...what (if anything) would come out from where, and what would go in where.North8000 (talk) 16:10, 12 April 2017 (UTC)[reply]

Do we all agree about the inherent non-symmetry: non-verifiablity is a sufficient condition for exclusion, but verifiability is not a sufficient condition for inclusion? Staszek Lem (talk) 18:05, 12 April 2017 (UTC)[reply]

I don't think any compromise is possible. We're at RfC because some users want the policy to say that verifiability isn't a reason for inclusion and will not accept any change that removes that wording. In fact it's the sine qua non for inclusion. If there are any reasons to include things in Wikipedia at all, then verifiability is the very topmost item on the list. If there are no reasons to include things in Wikipedia then why are we here? We might as well just switch off the servers. "Verifiability isn't a reason for inclusion" is a novel interpretation that can't remain in this policy unless there's a community consensus that this really is what editors want.—S Marshall T/C 18:10, 12 April 2017 (UTC)[reply]

• I think that compromise is possible once that you realize that you are working against your own goals. :-) I think that you and I are both inclusionists. The main reason for inclusion is simply to build good, informative articles, wp:ver is merely a constraint on inclusion. Isn't building good informative articles enough of a reason? Why would you need to say that compliance with one of many constraints is a reason for inclusion? You are supporting wording that (among st other things) implies than more than that is required to include something. Sincerely, North8000 (talk) 18:31, 12 April 2017 (UTC)[reply]

• It doesn't say that, though. If it said any of that then my objections would be much reduced, although I remain firmly of the opinion that collecting verifiable information is what Wikipedia is for, and therefore, in case of dispute, verifiability is the single most important reason to include a disputed thing.—S Marshall T/C 19:10, 12 April 2017 (UTC)[reply]

• Most of the policies that are mentioned when there is a disagreement about whether to include something, such as WP:V, WP:OR, WP:NOT, and WP:NOTE mostly describe constraints on what can be included. The policy that seems to say more about what should be included is WP:Editing policy, which says "Wikipedia is here to provide summaries of accepted knowledge to the public...." (Emphasis added). The word "summaries" tell us the majority of accepted knowledge should not be included, because if we try to include everything, it isn't a summary, and it isn't an encyclopedia. Jc3s5h (talk) 19:20, 12 April 2017 (UTC)[reply]
• @S Marshall: I agree, but let's dissect that sentence. Even by your own wording in that sentence, "verifiable" (in its position as an adjective that narrows the term "information") is a restriction on inclusion, not a reason for inclusion. Implicitly (and I agree that it should be said explicitly in some prominent place) the reason for inclusion is to build good informative articles.North8000 (talk) 19:30, 12 April 2017 (UTC)[reply]
• verifiability is the single most important reason to include a disputed thing At the first glance "collecting verifiable information is what Wikipedia is" is a reasonable stance. However please clarify you position to me: the fact "Lil Nigga Ho used the word 'f...' 11 times in his new album, including 'f... you' 5 times and 'f... him' 3 times", if true, belongs to wikipedia, yes or no? Staszek Lem (talk) 19:36, 12 April 2017 (UTC)[reply]

• No, it doesn't belong to Wikipedia. It probably doesn't belong in Wikipedia either, which is why the policy wording I want to restore says verifiability is not a guarantee of inclusion. The place to talk about verifiable information that doesn't belong on Wikipedia, though, is in the relevant policy, which is WP:NOT.—S Marshall T/C 20:04, 12 April 2017 (UTC)[reply]

• Well, IMHO the statements "verifiability is the single most important reason to include" and "verifiability is not a guarantee of inclusion" kinda contradict each other, unless one recognizes that the words "reason" and "guarantee" are not exactly synonymous (which does not always happen, and people keep miscommunicating due to speaking about different things). Therefore I suggest to change the wording to less ambiguous a bit below. Staszek Lem (talk) 20:16, 12 April 2017 (UTC)[reply]

• Please notice that I attempted to slightly change the terminology, using a less ambiguous term "condition" instead of "reason". My, now bolded, statement above is logically equivalent to: (A) "verifiability" is a necessary condition for inclusion; (B) it is not an unconditionally sufficient condition for inclusion. Are the statements (A) and (B) correct? Staszek Lem (talk) 18:45, 12 April 2017 (UTC)[reply]

I would agree with both of those. DonIago (talk) 20:07, 12 April 2017 (UTC)[reply]

Not as good as the current version, but looks good to me if that would settle it. A mouthful but logically sound with respect to this discussion. The proposal should be made clear if it goes further. E.G what comes out from where, what goes in where. North8000 (talk) 20:40, 12 April 2017 (UTC)[reply]

I think there is general acceptance of A being correct, but not B. While B might be strictly accurate, the double-negative is confusing, and it allows the implication that verifiability can be sufficient, which is the core issue here. Also, I think it misses a key point: once A is met, WP:V should be no longer applicable. ~ J. Johnson (JJ) (talk) 20:53, 12 April 2017 (UTC)[reply]

P.S. I am more inclined to a more succinct variation of your previous bolded phrase: verifiability is required for inclusion, non-verifiability is sufficient for removal. ~ J. Johnson (JJ) (talk) 21:04, 12 April 2017 (UTC)[reply]

Actually, the statements "verifiability is required for inclusion" and "non-verifiability is sufficient for removal" are logically equivalent. Staszek Lem (talk) 21:21, 12 April 2017 (UTC)[reply]

They would be logically equivalent in a binary world where the only outcomes are true or false. But in Wikipedia, a passage may be necessary to satisfy policy (for example, inline citation associated with a direct quotation), policy may say a passage is required to be removed (controversial statement about a living person not supported by citation to a reliable source), or allowable but not required (most passages). To say "non-verifiability is sufficient for removal" means it would be unfair to apply disciplinary measures for removing the statement, but it does not say that all those editors who read it and left it there are negligent. Jc3s5h (talk) 21:31, 12 April 2017 (UTC)[reply]

• Yes, I tried to formulate as "binary" as possible, because several people already noticed miscommunication issues due to different understanding of natural language, so essentially people do not hear each other. As for your argument, the policy is not intended to be a basis for reprimanding or otherwise slap-tagging wikipedians (comment on text, not on editor). We are volunteers and demands to be perfect are forbidden. Therefore your "unfair" argument is irrelevant. Staszek Lem (talk) 02:16, 13 April 2017 (UTC)[reply]

• It depends on how one squints. "Inclusion" could refer to the act of inserting material (insertion), or it could refer to the state of retaining such material (retention). One could be permitted but not the other. There are additional considerations, such as process. E.g.: given some case where both insertion and retention are "wrong", and therefore exclusion (as both act and state) is proper, there may be issues in how that is done/achieved. Part of the problem we have here is that absence of a statement regarding the flip-side – even though it is tautologically equivalent and perfectly obvious – is taken by many editors as silence, and silence as permission. ~ J. Johnson (JJ) (talk) 22:29, 12 April 2017 (UTC)[reply]

• No. We are here because one user (you), having missed the original discussion leading to the change and not liking the result, re-opened it. But not getting any satisfaction there (for various reasons, broadly attributable to your non-cooperativeness), you insisted on an RfC. Which, as agonizingly evident from all the directions this RfC keeps darting off into, is so inadequately formulated that we still don't have consensus of the underlying issues and terms. Undoubtedly you think the "yes or no" is simple enough, but that is entirely superficial. ~ J. Johnson (JJ) (talk) 20:57, 12 April 2017 (UTC)[reply]

• Sure, the whole problem here is me and my uncooperative, muddleheaded behaviour. I'm a terrible, terrible person for upsetting you in this way. We do know that you think this.  :)—S Marshall T/C 22:03, 12 April 2017 (UTC)[reply]

Histrionics aside, the problem is your opinion that collecting verifiable information is what Wikipedia is for. WP:NOT clearly states that this is not the case: "merely being true, or even verifiable, does not automatically make something suitable for inclusion in the encyclopedia." Even your preferred version of WP:V clearly states that this is not the case: "The onus to achieve consensus for inclusion is on those seeking to include disputed content." Both these statements very clearly show that while verifiability is required, it is not sufficient for inclusion. Bright☀ 16:50, 16 April 2017 (UTC)[reply]

And that's quite consistent with what I've said. "Almost all content that's verifiable to a genuinely reliable source belongs somewhere on the encyclopaedia" does allow for exceptions. Can we now please refocus on restoring "verifiability is not a guarantee of inclusion".—S Marshall T/C 10:26, 17 April 2017 (UTC)[reply]

A large number of editors expressed to you time and again why it's not consistent with what you said and why your wording is less desirable. So no, at the very least this RfC is going to close with "no consensus", mostly because it was never about working towards consensus, it was an all-or-nothing revert attempt. Bright☀ 15:01, 17 April 2017 (UTC)[reply]

Sorry, false. Please take a look at the headers of the subsections of this RFC. Staszek Lem (talk) 17:39, 17 April 2017 (UTC)[reply]

Well, let's set expectations correctly. The purpose of this RfC is to determine whether there is community consensus for the recent, drastic change to the longstanding stable version of the policy. That's a yes or no question, so it's a binary RfC with two possible outcomes. If there is consensus, then okay, that's silly but I will accept the result. If there isn't, then the recent change will be reverted, the longstanding stable policy wording will be restored, and then we can talk about whether there's a rewording that will gain community consensus. But let's let the RfC play out first.—S Marshall [f [User talk:S Marshall|T]]/C 22:02, 17 April 2017 (UTC)[reply]

Well, the wording of the RFC is just what it is, not something written down here. What you just wrote I would have objected to as malformed because it implies that a consensus is needed to retain the current version. Sincerely, North8000 (talk) 01:30, 18 April 2017 (UTC)[reply]

To be fair, a consensus *is* needed to keep and apply the currently disputed version. The text won't change until we reach a new consensus, but it will keep having the "this part of the policy is being discussed for lack of agreement on its wording" tag unless we can agree on a better wording. In addition, there are several editors now around here that don't like the wording achieved in the previous discussion, not just S Marshall. The above comments centered in blaming Marshall for the existence of the current discussion do not lead to solve the disagreement. Diego (talk) 12:48, 18 April 2017 (UTC)[reply]

To be fair, I'm certainly to blame for all this discussion.  ;-) From these editors' point of view, I've forced an RfC about something they would much have preferred to slip quietly through based on a talk page discussion. My position is very clearly and unambiguously that this novel wording requires community consensus and in the absence of consensus, should be reverted to the stable version. I certainly don't expect these editors to agree with my position on this! The point is for the closer to decide and, based on the course of the RfC so far, it's likely to be the point that determines the final outcome. In order to ensure the point is fairly and correctly decided, I would suggest that when the RfC is nearing its end, we post on WP:ANRFC asking for a triumvirate of experienced admins to perform the close.—S Marshall T/C 20:34, 18 April 2017 (UTC)[reply]

You misstate the matter, and insinuate lack of good-faith. The strong suggestion that you should not immediately start an RfC was because the rest of us wanted to take a break and come back fresh, during which break someone might work with you to clarify your concerns and better formulate an RfC. You were not making any progress in the preceding discussion (in part because you refused to clarify your ambiguous statements), and proceeding straight to this RfC without resolving any of the underlying issues has only made for a bigger muddle. ~ J. Johnson (JJ) (talk) 01:15, 19 April 2017 (UTC)[reply]

The whole point of RfCs is to gather opinions of people previously not involved... Telling someone not to start a RfC because one will not be able to participate is opposite to the spirit of WP:Dispute resolution. Diego (talk) 09:25, 20 April 2017 (UTC)[reply]

You misunderstand the situation and intent. Marshall has shown a certain ineptitude in even explaining what his concerns are, let alone advancing the measures he thinks necessary. It was entirely in the spirit of WP:Dispute resolution that he was urged to not proceed on his own, but to seek the assistance of another editor. Not to censor him, but to keep him from mucking up his own best interests. ~ J. Johnson (JJ) (talk) 22:20, 21 April 2017 (UTC)[reply]

Bright* is exactly right. In the preceding discussion Marshall's statements did not could not unambiguously state his objections, and he refused to clarify his statements when requested (while demanding a universal disproof from the other editors). Quite unsurprisingly he could did not obtain any concurrence. His response was simply to bring in more editors via an RfC, without remedying any of the defects that have plagued this discussion since he joined it. Others have tried to help him, but he seems adamant to any suggestions. And it seems to me he may have a memory problem, as for reasons unclear to me we keep having to go over the same points with him. ~ J. Johnson (JJ) (talk) 22:26, 17 April 2017 (UTC)[reply]

[Revised language that might be taken as touching on another editor's abilities. ~ J. Johnson (JJ) (talk) 01:32, 19 April 2017 (UTC)][reply]

• STOP - comment on the proposals, not the other editors. Blueboar (talk) 20:41, 18 April 2017 (UTC)[reply]

I think several of us (I include myself) have reached the point where we are suffering from "last word disease"... where we are so wrapped up in the debate that we simply can not let the "other guy" get in the last word. We end up either repeating what we have already said (multiple times) or resort to personal attacks. May I suggest that those of us who have been the most outspoken in this dispute sit back and let others comment for a while. Blueboar (talk) 02:24, 19 April 2017 (UTC)[reply]

Sorting out a few fundamantals

Regarding a possible compromise, there are a few areas which might be helpful to sort out. I would like to ask this question of the folks that don't like the current wording. Let's talk about an item for potential inclusion. Let's say that it met all "9" of the clear cut policy constraints (requirements), including the "biggest" one,, #9, which is wp:verifiability. Now comes step #10, the question comes whether or not to include it in a particular article. And, practically speaking, this means that there is some sort of debate or dispute about its inclusion. You might imagine a "scale" during step #10 where the reasons for and against inclusion are piled on the two sides of the scale to see which way it tips.

1. One school of thought is that it had to meet the "9" requirements to even be considered for inclusion, and so all of those requirements (such as wp:ver) already had their effect, and that having met any of those requirements (such as wp:ver) should not be usable a second time, not be pile-able on the "scale" of arguments during step #10.
2. Another school of thought is that meeting other requirement should be so usable, usable a second time and be pile-able on the scale during step #10.

Number 2 could be a reason why you oppose the current wording, but it is not the only possible reason for doing so. Would you say that #1 or #2 matches your thoughts? (PS: I'll be gone for 3 1/2 day days) Sincerely, North8000 (talk) 14:49, 20 April 2017 (UTC)[reply]

• My position is that verifiability is both a prerequisite for inclusion, and also a factor in the "scale" in the balancing exercise you envisage. I think after reading all this verbiage we might be looking for wording that says: (1) that only verifiable information can be included on Wikipedia; (2) verifiability is one of several considerations that affects the decision on whether to include information; (3) information that's backed up by an inline citation to a reliable source is more includable than information that isn't (or in other words, information that's actually verified right now is more includable than information that editors think is verifiable from somewhere); (4) verifiability doesn't trump WP:NOR or WP:NOT, so information that fails either of those policies should be removed even if it is verifiable; and (5) Flyer22's law: removing verifiable information from established articles requires care and thought, and should not be done recklessly or against consensus.—S Marshall T/C 23:28, 24 April 2017 (UTC)[reply]

OK thanks. That does mean that we have a fundamental difference and not just a matter of differing impressions from reading it.  :-( It would also be useful to hear the answer from other folks who do not like the current wording. Sincerely, North8000 (talk) 02:29, 25 April 2017 (UTC)[reply]

What if we remove the word "inclusion?

I am getting the feeling that many of us are getting hung up on the word "inclusion"... mentally investing that word with all sorts of different (nuanced) significances and meanings. Hell, I'm not sure if we even agree on why we are using the word "inclusion" in the section in the first place. If I am right, then the only way forward may be to rephrase in a way that omits that word. So, I am curious to see if we can do so. let's try to come up with a few formulations that do not use the word "Inclusion" at all. Blueboar (talk) 22:30, 12 April 2017 (UTC)[reply]

Rather than removing, as I understand, we must try to split it into several meanings encountered here. So far I see the 'trinity' of distinct actions: addition, retention and removal (and a nit-picking symmetrization with a 'non-action' of "non-addition"). Now, the question is do we need separate rules/flavors to govern these? Staszek Lem (talk) 02:21, 13 April 2017 (UTC)[reply]

I think you are making this more complex than it needs to be... rephrasing without the word may be easier. Blueboar (talk) 11:01, 13 April 2017 (UTC)[reply]

I am just diagnosing a cause of disagreement. This even may help removing the offending word. Otherwise there we may try to insert even more "offending" one. Staszek Lem (talk) 16:01, 13 April 2017 (UTC)[reply]

I tried analyzing it and IMHO at the root root of it there is no actual underlying conflict. The problem happens at the next level up which is wording. I may have this all wrong but here is how I dissect the differences:

• First there exists what I've been calling an urban legend, but basically a (lets call it) "floating concept" that, when there is a dispute over inclusion / exclusion of material, that meeting wp:ver is a force for inclusion, or an argument for inclusion. This concept is not specified by any policy, but there are words here and there which implicitly might have supported that concept a bit. This includes the wording that we got rid of a few months ago.
• We tend to go by our experiences. I think @S Marshall: is mostly concerned about / focused on people trying to keep out good material. I think that many of the rest of us are also concerned about wiki-lawyering to force in bad material.
• I don't know whether or not S Marshall is making the logical distinction (regarding meeting wp:ver) between such fulfilling the main requirement for inclusion and such being a force or reason for inclusion (e.g. during disputes). If they are not, IMHO that is a logical error. Even if he is in favor of both of them, he should acknowledge that they are two different things. And, respectfully, that logical error would conflict with those of us who are logically correct on this point.  :-) :-) But if even they did acknowledge this distinction, that would only remove 1/2 of their motivation. While he would then have to admit that the "floating concept" has no solid basis in wp:ver, he still might prefer supporting that "floating concept".
• In the change made (after a long discussion) a few months ago, we did two things regarding this "floating concept" We took out the wording that provided a tiny amount of implicit support for the "floating concept"; we also put in wording which clearly says that the "floating concept" does not exist. That is probably the second area of "conflict".
• This might relate to the broader feeling that we're missing the prominent policy which says "our mission is to build good articles with good material, and that's a plenty good reason to put a piece of good policy-compliant material in". (as a sidebar, so no "floating concept" is needed) S Marshall would probably like wp:ver to do that a little. Most us feel that wp:ver and the wording under discussion is not the place to do that.

Respectfully, North8000 (talk) 20:31, 13 April 2017 (UTC)[reply]

I'm looking at real world disputes about what to include in our encyclopaedia, and the first thing I note is that they're usually, at heart, disputes about what to cut out of our encyclopaedia. I feel that when we're deciding whether to cut something out or not, the very first question is, always, what do the sources say. If the disputed information doesn't appear in sources that are genuinely reliable, then I think that's an extremely strong argument for cutting it. Conversely, if the disputed information does appear in genuinely reliable sources, then yes, that's a factor supporting the case for not cutting it. I think that North's "floating concept" has always been implicit in WP:V. If the information is genuinely based on truly reliable sources, then it should normally be kept, although it might be moved. There's a presumption that verifiable, reliably-sourced information belongs on Wikipedia. A presumption is not a guarantee, and there are still situations in which verifiable information should be cut or moved. All of these situations are fully covered in WP:NOT or WP:NPOV. Those are the appropriate policies to discuss that.—S Marshall T/C 21:49, 13 April 2017 (UTC)[reply]

I agree strongly with the above statement. If content is verifiable, and supported by high quality reliable sources, then that by itself is indeed a very clear indicator that it belongs somewhere in an encyclopedia. The examples I gave above, of standard facts from textbooks, do not normally require any WP:PAG gymnastics to justify their inclusion: in practice, verifiability is sufficient. Sławomir Biały (talk) 22:05, 13 April 2017 (UTC)[reply]

That any and all verifiable content "belongs somewhere in an encyclopedia" is one of the fundamental opinions here, and really deserves its own discussion. As it is, we have covered this, but it has been so disorganized and scattered about that nothing is retained, so it seems we must cycle through it again. ~ J. Johnson (JJ) (talk) 22:18, 13 April 2017 (UTC)[reply]

I will point out one last time, that no one anywhere in this discussion has said that "any and all verifiable content 'belongs somewhere in an encyclopedia'." This is either a WP:CIR failure on your part, or is just bad faith trolling. But it stops now, since we are now absolutely clear that no one in this discussion is saying that. Not me, not S Marshall, not anyone. You can continue to believe that, but please keep these beliefs to yourself in the future. And please don't ever misquote me again. Sławomir Biały (talk) 00:04, 14 April 2017 (UTC)[reply]

It is one of the fundamental opinions which was flatly rejected since early days of wikipedia, as covered in multiple WP:NOT bullets. Staszek Lem (talk) 22:53, 13 April 2017 (UTC)[reply]

I agree that it is not the case that "Any and all verifiable content 'belongs somewhere in an encyclopedia'." In fact, there already is a Wikipedia policy about that very question. So, we are apparently all violently in agreement about this point. Perhaps it is time to stop attacking strawmen that have nothing to do with the discussion? Sławomir Biały (talk) 00:10, 14 April 2017 (UTC)[reply]

Bławomir Biały: Cool your jets; you are getting unnecessarily intemperate. Your insinuation that I lack either competence or good-faith is personal attack, and an apology is in order.

Nor have I misquoted you. That is a false charge, and even a misrepresentaton of what I have said. As to the matter at hand, I quote exactly what you said just above (copied from the diff): "If content is verifiable, and supported by high quality reliable sources, then that by itself is indeed a very clear indicator that it belongs somewhere in an encyclopedia." (22:05, 13 Apr). Your objection seems to be to my qualifying this "verifiable content" term we are all waving at with "any and all". In fact, no one has qualified that term beyond "reliable sourced and not excluded by other considerations". If you really meant something more specific you should have said so. But you didn't, so it is a fair inference that whatever you meant does not exclude anything less than any and all.

What IS being said by some one – notably the chief architect of this RfC, S Marshall – is that "verifiable content", without further qualification other than "probably" and "almost always", "belongs somewhere" in the encyclopedia. E.g.:

• 01:47, 23 Mar: "it probably belongs somewhere in the encyclopaedia". And: "Everything verifiable should be somewhere in the encyclopaedia".

• 00:49, 25 Mar: " I did say that verifiability should be a virtual guarantee of inclusion somewhere in the encyclopaedia ...."

• 12:11, 27 Mar: "almost always belongs somewhere".
• 15:35, 28 Mar: "almost always belongs somewhere".

With a strong inference that "all" is the qualifier meant. Now you might object that these comments occurred not in this discussion (meaning the RfC, opened 10 Apr.), but in a prior discussion. But this discussion is a straight-forward continuation of the prior discussion, without any break, and this discussion is informed by the past discussions.

So as I said before: whether "verifiable content" (however qualified) "belongs somewhere in the encyclopedia" is one of the fundamental opinions here, and really deserves its own discussion. ~ J. Johnson (JJ) (talk) 23:08, 15 April 2017 (UTC)[reply]

• I find it slightly creepy that you've collected the diffs so carefully. I'm taken aback. I realise that my views on this are heretical, blasphemous and beyond the pale, so any objection I might have to J Johnson's damning condemnation is mere quibbling to be dismissed out of hand; but if I can interrupt the thundering from the pulpit just for a moment, I would like to point out that all this is in fact a distortion of my position. I've consistently said "almost all content that's verifiable to a genuinely reliable source belongs somewhere in the encyclopaedia", which means that:- (1) In logic, "All" cannot possibly be the qualifier meant; and (2) One or two hypothetical counterexamples do not make me wrong.—S Marshall T/C 10:21, 17 April 2017 (UTC)[reply]

Your repeated histrionics aren't helpful toward building consensus. Additionally, there's not "one or two hypothetical counterexamples", there's an entire policy called WP:NOT, and a process called WP:CONSENSUS, for when verifiable content should not be included. The wording that you support gives this caveat ("Consensus may determine [...] include disputed content") so really if you want to know what "makes [you] wrong" it's those policies. Bright☀ 15:14, 17 April 2017 (UTC)[reply]

@S Marshall:, I don't think that your views are any of those nasty things, and I have a lot of respect for you as someone who simply wants Wikipedia to be the best that it can be. And, at the root of it, I thin that we're both inclusionists for good material. But I also think that you are mistaken in this case, meaning that what you argue for would be a net minus to Wikipedia. Respectfully, I think that it comes from one perspective error (you are overly thinking about some rare relevant cases, and not about the common ones) and a couple of logical errors. Respectfully, North8000 (talk) 16:34, 17 April 2017 (UTC)[reply]

Oh, were those statements "taken back"? Marshall, is it really necessary to have to explain to an experienced editor the use of strike-out text to indicate retracted statements?

But it is not at all evident that you have actually "taken back" that sentiment when just this week you stated (21:49, 13 Apr):

if the disputed information does appear in genuinely reliable sources, then yes, that's a factor supporting the case for not cutting it. ... If the information is genuinely based on truly reliable sources, then it should normally be kept ....There's a presumption that verifiable, reliably-sourced information belongs on Wikipedia.

And then just a day later (just below here, at 14:19, 14 Apr), you said:

Yes. "If it's verifiable to a genuinely reliable source that meets WP:RS then it probably belongs on Wikipedia somewhere" happens to be my opinion ....

Which you top off with your statement above: "almost[emphasis added]all content that's verifiable to a genuinely reliable source belongs somewhere in the encyclopaedia".

These are your own, unretracted, words, and if they be distortions of your views the fault is your own. ~ J. Johnson (JJ) (talk) 23:20, 17 April 2017 (UTC)[reply]

And I don't retract them. Our sister project Wiktionary has a definition of "taken aback" which, apparently, it would help you to read. It's not my fault you don't understand my position; you've made no effort to do so at all, and as you've just very neatly demonstrated, you skim what I've said at very high speed and take wild guesses at its meaning. I can tell you what I think but I can't force you to read it attentively or with your brain engaged. When your only reaction to everything I've said is to scour it for things you can attack or take umbrage at, it's hardly surprising that your understanding of it is so limited.—S Marshall T/C 19:42, 18 April 2017 (UTC)[reply]

I have made great effort to understand your position (but how would you know?), and I find the greatest difficulty to be: your own statements of your position and concerns. Impugning my competence does not explain your position, nor address your concerns. ~ J. Johnson (JJ) (talk) 01:54, 19 April 2017 (UTC)[reply]

Regarding J. Johnson's quoting out of context, I suppose it is possible that it was a honest mistake. However, I will not apologize for pointing out that he is misrepresenting things, when he continues to do so with this completely false statement: In fact, no one has qualified that term beyond "reliable sourced and not excluded by other considerations". If you really meant something more specific you should have said so. But you didn't, so it is a fair inference that whatever you meant does not exclude anything less than "any and all". I have made it very clear from the beginning of the discussion that I am talking about standard, routine facts, that are sourced to standard scholarly literature ([4], [5], [6], [7], [8], [9]. I gave as examples the vapour point of mercury, and other kinds of routine facts sourced to standard physics textbooks. My opinion is that for routine facts of this kind, with high quality scholarly sources, verifiability alone does create a reason for inclusion somewhere in the encyclopedia. Since I have been very clear about this throughout the discussion, I can only conclude that User:J. Johnson does not believe that standard facts sourced to reliable scholarly literature ipso facto belong somewhere in an encyclopedia. Is that an accurate summary of J. Johnson's position? Sławomir Biały (talk) 13:01, 19 April 2017 (UTC)[reply]

You misinterpreted what I said, proceeded to misrepresent it as a misquotation, and on that basis then questioned here my competence and good-faith. If that does not warrant an apology then you have abandoned the "civil" usually required of "discourse". (Incidentally, you have not shown that I misquoted anything – if I have, just line up the original with what I quoted and show us the difference – nor quoted out of context, but as this may be beyond your competence I am disinclined to spend any more time on you.) ~ J. Johnson (JJ) (talk) 04:02, 20 April 2017 (UTC)[reply]

This now seems offtopic. Your quotation was obviously out of context, since you managed to completely mischaracterize my position as that of 'any and all verifiable content "belongs somewhere in an encyclopedia"'. This was, I should add, after I explained to you the difference "a reason" for something and "an automatic home run", which you still don't seem to grok. Obviously, if you continue to raise the same points that have already been demolished, then it will lead to questions of competency or WP:IDHT. Still, we are now very clear that no one in this discussion is saying that. You can choose to disengage with me, but if you continue to attack strawmen, I will continue to point out your errors. Sławomir Biały (talk) 11:37, 20 April 2017 (UTC)[reply]

I agree, there are certain facts or information bits, that are "obviously" relevant and hence verifiability alone would justify their inclusion. "Obviously" means, that it is fair to assume the community considers them as worth of inclusion and/or so did traditionally other encyclopedias.--Kmhkmh (talk) 13:30, 19 April 2017 (UTC)[reply]

That the community would consider them "obviously" relevant and be able to point to other encyclopedias that include these bits makes them irrelevant to this discussion, as they're not being added solely because they're verifiable. DonIago (talk) 13:43, 19 April 2017 (UTC)[reply]

This debate is not about facts being "added solely because they're verifiable", but about "verifiability not being a reason for inclusion" which is the wording of the policy. Being verifiable is a factor in the community considering the facts relevant and worthy of coverage, and thus a reason for inclusion, despite what WP:V currently says. Besides, you can't cut out WP:V from the debate of content inclusion: content policies themselves say that you need to take all of them into account, not in insolation; that "added solely because WP:V" is a strawman. Diego (talk) 14:48, 19 April 2017 (UTC)[reply]

And my feeling is that verifiability is a requirement for inclusion, not a reason for inclusion. "This is relevant information" is a reason; "this is verifiable information" is not, and editors should never be defending additions to articles with, "it's verifiable so it belongs here". Whether a fact is verifiable should be immaterial to its inclusion beyond meeting the minimum threshold of "we can verify it...but is it relevant, significant, etc...". "Added solely because WP:V" may or may not be a strawman, but it's a frequent defense for retaining information in articles. It's that kind of reasoning that led to, for instance, WP:IPCV, which raised the bar on "In popular culture" material to specify that sources must demonstrate not just the existence of a pop culture reference but also it's significance, precisely because people would defend all kinds of trivial inclusions on the grounds that they could be verified. DonIago (talk) 17:11, 19 April 2017 (UTC)[reply]

Being verifiable in a reliable scholarly source implies significance, contrary to the current wording of the policy. Sławomir Biały (talk) 17:30, 19 April 2017 (UTC)[reply]

[Added] "because WP:V ... is a frequent defense for retaining information in articles". Because it is a good defense. Appearing in a reliable source discussing the article topic makes the fact relevant, so as long as no policy opposes it (such as WP:NOT or WP:UNDUE or WP:SIZE or WP:SYNTH), there is no reason *not* to include it. WP:IPCV is an example of something that the community decided not to do; but lacking such prior consensus, why impose arbitrary restrictions to inclusion on facts that reliable sources have decided are relevant? Wikipedia is WP:NOTPAPER.

Moreover, "verifiability is no guarantee for inclusion" or "not a reason for inclusion" is one of the most used arguments used to try to WP:CENSOR verifiable information that someone just doesn't like, and the new wording is giving fodder to people trying to game the system this way.

my feeling is that verifiability is a requirement for inclusion, not a reason for inclusion. And given that several of us feel that verifiability is a reason for inclusion (maybe not a definitive reason, but a reason nonetheless), the policy should take a neutral wording on the matter, rather than the current taking a stance. Diego (talk) 18:05, 19 April 2017 (UTC)[reply]

And here we have the crux of at least two points of contention between the multiple parties who've contributed to this discussion (smile). DonIago (talk) 18:29, 19 April 2017 (UTC)[reply]

At least from my perspective, I am beginning to see part of the divide. Smarshall has described a "context" issue, imo, but verfiabiliy is not about context, it is about "proving-up" a discrete piece of information. Forgive me for stressing again, V's minimalist requirement is 'a single RS', not multiple sources, as SMarshall's comment discusses. Our policies on the various issues concerning context are WP:NPOV, WP:OR, WP:PAGEDECIDE. I am looking at this from another end then SMarshall, not his macro view, but as I see it, the real, in the act of editing a particular article, where the practical inclusion question, "does this belong, here, in this context, now?" is answered. (For those who love imperfect analogies: we are building a city of various brick-structures, and all verifiability is, is how to make good bricks, not about how to build the many different structures. And as every builder knows, some of the bricks will never be used.) I think, it's best to keep the discrete and practical focus for this policy - the bricks. (To extend my analogy, NOT is all that brush we have to clear away - someone else can use those sticks to build their city. :)) -- Alanscottwalker (talk) 00:18, 14 April 2017 (UTC)[reply]

Whereas I've always thought that verifiability is a matter of principle. For many years this policy said that verifiability rather than truth was the threshold for inclusion on Wikipedia --- which language I, and North8000, both went to a great deal of trouble to change; the two of us are in fact longstanding allies and partners on this point. It's novel for me to be on a different side from North8000. The case we both made in 2010-2011, and which ultimately prevailed at RfC in 2012, is that verifiability is in fact one of several requirements for inclusion. Blueboar enshrined this in the stable wording, "Verifiability is not a guarantee of inclusion", which I endorse.

But if you would prefer to focus on how this new wording will work at the "brick" level then I'm eager to play. You've inserted wording into a policy that greenlights removing verifiable information from the encyclopaedia. Can you think of any verifiable information that, for example, climate change deniers, intelligent design advocates, scientologists, moon landing conspiracy theorists, or pro-Palestinians might want to remove?—S Marshall T/C 01:11, 14 April 2017 (UTC)[reply]

I can think of multiple NPOV and OR reasons why they just can't remove whatever they want. But, it is precisely those types you describe who will say 'We gotta RS. It's V', so it has to be here. --Alanscottwalker (talk) 01:19, 14 April 2017 (UTC)[reply]

• Sure, Alan, and if the only situation you envisage is when the white hat good guy is trying to stop the black hat from adding wrongbad information to the article, then the new wording is obviously ideal. I do get that. But can we please look at some of the other situations that this change affects?—S Marshall T/C 14:19, 14 April 2017 (UTC)[reply]

@Alanscottwalker:, the only situations I can think of where the OR policy would prevent the removal of information would be the removal of citations, while leaving the claim in the article, or removing part of the claim in the article so it no longer agrees with the supporting citations. Except for this, the OR policy is about excluding and removing information, not adding or retaining it. Jc3s5h (talk) 14:33, 14 April 2017 (UTC)[reply]

OR is also about the correct juxtaposition of facts, that creates a non-original article presentation of the subject - if you remove a critical fact, leaving an original impression or original implication regarding the subject, you violate OR. Alanscottwalker (talk) 14:45, 14 April 2017 (UTC)[reply]

• Getting back on-topic: I think "inclusion" is the best, most appropriate word, and should be retained. I agree with Staszek overall (sorry Blueboar: it is complex) but quibble a bit with his wording that "we must try to split it into several meanings" – I think the usage is already split, and we need to clarify those usages. I think WP:V is (and should be) essentially one rule, but may need explicit explanation in how it applies in these different cases. ~ J. Johnson (JJ) (talk) 22:25, 13 April 2017 (UTC)[reply]
  • Yes, that's what I meant under "split": explicitly cover use cases which people may distinguish/confuse. Staszek Lem (talk) 22:53, 13 April 2017 (UTC)[reply]

S Marshall's examples "Can you think of any verifiable information that, for example, climate change deniers, intelligent design advocates, scientologists, moon landing conspiracy theorists, or pro-Palestinians might want to remove?" are about accurate, sourced material that should be in the article, but which POV warriors will want to remove. Just brainstorming, but @S Marshall: if we could come up with something that helped in that area but didn't address your "if it's verifiable it belongs in Wikipedia somewhere" assertion, might that be a suitable compromise? North8000 (talk) 13:05, 14 April 2017 (UTC)[reply]

• Yes. "If it's verifiable to a genuinely reliable source that meets WP:RS then it probably belongs on Wikipedia somewhere" happens to be my opinion, but it was never in the policy in the first place. From my point of view what's important is to restore "verifiability is not a guarantee of inclusion", which was removed in a drastic change to the policy that would require community consensus.—S Marshall T/C 14:19, 14 April 2017 (UTC)[reply]

I was thinking of something that is more than you asked for in one area and less than you asked for in another. My first lame attempt: Current wording plus: "Editorial practices and the objective of building good articles also affect inclusion or exclusion of material, but compliance with wp:verifiability and other core policies and guidelines means that the material has completed important requirements for inclusion." Just brainstorming. I realize that others might have objections to this, but right now I'm just starting with the toughest guy in the room.  :-)  :-) North8000 (talk) 15:51, 14 April 2017 (UTC)[reply]

Unfortunately, that leaves "verifiability is not a reason for inclusion" in the policy.—S Marshall T/C 17:45, 14 April 2017 (UTC)[reply]

No, that is (in some senses) the whole point of the exercise. ~ J. Johnson (JJ) (talk) 23:10, 15 April 2017 (UTC)[reply]

I concur with Diego's observations, passim. I do not concur that the change was a good idea and agree that it is for the reasons S Marshall and J. Johnson (JJ) concur (amazingly) upon: failure to resolve whether "verifiable info" is, ipso facto, a reason for inclusion. Regretting that I did not participate in the earlier, apparently exhaustive debate, I come down in favor of the argument that verifiability is a reason for inclusion. Pragmatically, that should mean that if an insertion is challenged and restored, a new consensus is needed to exclude it. My reasoning is that the fact that the datum was included in a publication meeting our standard for RS means that it has already been deemed worthy of being publicly shared by a respectable source, so that is an implicit "!vote" in support of retention, that can be overcome by at least two objections -- but not by one. FactStraight (talk) 04:19, 25 April 2017 (UTC)[reply]

But that is the requirement for everything to be included in any article. IMHO why should simply meeting the universal requirement be an argument (e.g. in a dispute) for inclusion in a particular article? Sincerely North8000 (talk) 12:38, 25 April 2017 (UTC)[reply]

With the current wording, Verifiability wouldn't be an argument for inclusion in any article.

Meeting the universal requirement of being verifiable is a good reason for inclusion in articles where the information is relevant, provided it isn't undue weight, anything in WP:NOT, or other reasons for excluding it. Diego (talk) 13:25, 25 April 2017 (UTC)[reply]

I think that my argument in that area goes like this: A rule is "Only human beings are allowed to win a Nobel prize." When they meet to decide on whether or not a person gets the prize, should "he's a human being" be usable as an argument for giving them the prize? North8000 (talk) 15:04, 25 April 2017 (UTC)[reply]

Following along with the metaphor: if the Intergalactic Council is making the choice among a pool of humans and aliens being taken into consideration, and it isn't a formal requirement that winners show proof of their humanity, and there are hundreds of thousands of categories of Nobel prizes, then "he's got an official certificate showing he's a human" would indeed be a good reason to award the prize for a category where that human shows some particular talent and no other human is making a competing claim. Diego (talk) 15:37, 25 April 2017 (UTC)[reply]

@Diego Moya:Besides being a pleasure to work with everyone here (because everyone here is working for the good of Wikipedia), it's a particular pleasure to have this exchange with you because you have been elucidating precise thoughtful logical arguments in the area of dispute. I think I see your point. One is that "thousands of" is a better analogy to Wikipedia than the "few Nobel prizes".....agreed. However, I think that I can note some important relevant flaws in your analogy. This is not to pick apart your good analogy, it is because I think it deals with core issues. First, your analogy is to sourcED material in Wikipedia. This whole debate is about material which complies with wp:VER which merely requires sourcABLE material, not sourcED material. So the material under discussion does not (necessarily) have that extra imprimatur (certificate). Second in in your inclusion discussion, you "needed" to give a (another) good reason for inclusion ("shows some particular talent") which somewhat indicates that an actual :-) reason for inclusion is needed. :-). But most importantly, the situation that you describe (in essence undisputed editing) is the situation where this whole concept (old or new wording) never comes into play. Overly-briefly, the situations where it comes into play are all "disputes"; S Marshall's are where people are trying to keep out good material, and "ours" are where meeting wp:ver is used as a way to force in bad material. "Our" side feels that the latter is a much more common problem, and also that the wording is more directly relevant to and used in the latter. Respectfully North8000 (talk) 16:39, 25 April 2017 (UTC)[reply]

Well, my analogy wasn't really meant as a completely accurate description of the situation, with all the hyperbole over aliens and galactic empires :-) It was more an emotional appeal as a reply to your vivid "Nobel prizes are for humans" image.

I admit that I hadn't though about the difference between sourceable and sourced content; but now that you point it out, I still don't think it's really relevant. Verifiable content is one for which is known that reliable sources exist, even if they are not included in the article right now. I agree that content with a "presumption of verifiability", which is the standard for WP:Notability, wouldn't be enough; but when WP:ONUS is invoked as a reason for inclusion by someone arguing "the content is verifiable (and I don't see reasons for exclusion)", I assume we are the point at which at least one RS is known. At this point, the "talent" of being noticed by a valid reference has been met.

WP:ONUS already defaults to excluding content when there is no consensus, so "your side" is at an advantage; recognizing (or at least not rejecting) verifiability as a (possible) reason towards inclusion (but not a guarantee) should entice the editors wanting to exclude the content to provide themselves a good reason why such content, reported by a reliable source, is nevertheless not a good fit for the article. Right now, they could merely reject it with a poor argument like "verifiability is not a reason for inclusion, and you have not achieved consensus for it" (i.e. the current wording of WP:ONUS). Diego (talk) 16:15, 26 April 2017 (UTC)[reply]

I don't have much of a direct response except to point out that the crux of the argument in your last sentence is wp:onus, the "not a reason" part is just a sidebar. FYI probably the most common of the types of situations that I'm thinking about are when there is fast "hit and run" type editing in order to POV an article. And there an edit summary of "undo removal of sourced material" is considered to be a strong argument to force inclusion and sufficient to undo a removal. The others are where a single source was used on a statement (in the voice of Wikipedia) that was either highly biased or clearly factually in error, where that particular statement tilted the article in an area that was under dispute. And in those cases meeting wp:ver is given as a force for inclusion. North8000 (talk) 17:57, 26 April 2017 (UTC)[reply]

Diego: Your "WP:ONUS already defaults to excluding content when there is no consensus" is ambiguous in not recognizing an important caveat. Note that consensus might be consdered to have three states: consensus FOR inclusion, consensus AGAINST, and the indeterminate state of NO consensus. WP:ONUS is really about the last state. But the implication of such a limited view is that no material can included without an explicit consensus for inclusion. This would be unworkable. (And possibly this is the essence of your objection?) Resolution of this can be had by recognizing that WP:ONUS applies where inclusion/exclusion has been contested.

Content is typically added (or removed) Boldly, without discussion (or even comments) on specific items, without any reason given. (Though "editorial discretion" is implied.) And sometimes such content is equally boldly Reverted, per WP:BRD. Up to that point (and subject to the recognized restrictions) I say that editors are pretty much free to add or delete as they consider fit, and no reasons need be given. It is only when an edit is questioned that we get into discussion, argument, and consensus, with WP:ONUS favoring exclusion where consensus is not reached.

I suspect that where "verifiability is not a reason for inclusion" (alone) is given is a reason for exclusion it is likely because someone argued that "verifiability IS a reason for inclusion". The intent of the current language is to clarify that verifiability is not "a reason", whether for inclusion or exclusion, but only a minimal requirement. Any argument that cannot come up with better reasons than "verifiability" is a poor argument. ~ J. Johnson (JJ) (talk) 23:21, 26 April 2017 (UTC)[reply]

I understand the intent of the current language ("verifiability is merely a minimum requirement"), and I would agree with it if that was all it did. The problem with the new wording is that it also interacts with WP:DUE WEIGHT, by undermining the following argument (for which "verifiability is a reason for inclusion" is merely a shorthand):

1. "This fact has been noted by a reliable source, connected to the topic of the article."
2. "When a reliable source notices some fact, it provides some support for it having due weight."
3. "Facts regarding a topic should be covered in an article with prominence relative to their due weight."
4. "Therefore, that this fact was noted by a reliable source provides some support for covering it in the article."

You are right that this is an argument from WEIGHT more than VERIFIABILITY; however, it can be legitimately made every time some fact is verifiable by a reliable source. Thus, the new wording is having effects on a different policy, which I think we all agreed is not a good idea.

If we could find a wording without this side effect, I wouldn't oppose this position of defining verifiability in this page merely as a binary requirement; as long as it didn't interfere with the idea that "verifiability on its own isn't a reason for inclusion, but it provides some due weight, which might be a reason itself". Diego (talk) 09:38, 27 April 2017 (UTC)[reply]

Possibly the answer lies in what you just wrote. You'll note that those arguments for inclusion weren't for something that met wp:ver (=sourcABLE). they were that several aspects of the sourcING were also factors on the side of inclusion. One idea would be for wp:ver to remain silent on that topic. Another might be along the lines of "the nature and contents of the sourcing may be very relevant to any inclusion discussion" North8000 (talk) 11:45, 27 April 2017 (UTC)[reply]

Yes, those both seem very valid options. Do you think the wordings proposed in the Inclusion and exclusion section below can be shaped into something that follows them, and is agreeable to "your side"? Diego (talk) 12:02, 27 April 2017 (UTC)[reply]

Its worth a try.North8000 (talk) 13:07, 27 April 2017 (UTC)[reply]

I think that the one that is garnering some support fulfills the "remain silent" approach? North8000 (talk) 17:30, 27 April 2017 (UTC)[reply]

I think you're right about "verifiability is a reason for inclusion" ("v-is-arfi"? as opposed to "v-not-arfi"?) as being shorthand for this WP:WEIGHT argument. But that is actually the problem: "verifiability" is being used for purposes other than WP:V. I think the argument you point to can be quite valid (assuming the premises are correct), and I have no issue with that. The issue is a "rectification of names" matter, in calling some issue a matter of verificiability when it is really about due weight or notability or such. Saying that this argument does not belong in WP:V does not mean that the argument is no longer valid, only that it should be called by its rightful name, which is not "verifiability". The problem is not that the current wording affects another policy, but that another policy is getting sucked into WP:V. Distinguishing them does not alter any other policy. ~ J. Johnson (JJ) (talk) 21:36, 27 April 2017 (UTC)[reply]

@J. Johnson:I think that your logic is the soundest of all. My only quibble is not relevant here which is that there are (rightfully) more things in play than policies and guidelines in determining what gets into an article. My discussion is more of a pragmatic one. The previous wording did harm in the area, if only by inference. The current wording makes positive statements to fix the problem. I was exploring possibilities for a compromise which merely does no harm in this area. The "remaining silent" would be the simplest variant of this. Wording that acknowledges that other factors relating to sourcing could carry weight in the inclusion/exclusion discussion is IMO accurate but out of place in this policy, but does does no harm and could be a compromise to settle this. North8000 (talk) 23:45, 27 April 2017 (UTC)[reply]

I'm not quite clear on what you are saying here. I think it is not out of place to expressly mention that other (and I do emphasize other) "things", factors, considerations, and policies are pertinent to inclusion/exclusion discussions, and that there is no need to misuse this policy. To the extent we are denying a certain usage some editors feel is essential we really need to indicate the alternatives. To be silent on this leads to the current situation of ambiguity, where we have a clash of interpretations. ~ J. Johnson (JJ) (talk) 23:28, 28 April 2017 (UTC)[reply]

Inclusion and exclusion

I am wondering if the problem we are having reaching consensus stems from the fact that we mention the term "inclusion", without mentioning its flip side: "Exclusion". Perhaps doing so might help us to find common ground. So... I offer the following for thought, and as a (potential) alternative...

• While a lack of Verifiability (on its own) is enough to justify the exclusion of material, the opposite is not always true. Verifiablitity (on its own) may not be enough to justify inclusion. This is because Wikipedia has other policies and guidelines that also play a role in determining whether some bit of information is included or excluded.

The exact wording may need additional tweaking, but I think something similar to this (mentioning both inclusion and exclusion) might clarify what the policy is actually trying to say. Thoughts? Blueboar (talk) 14:45, 25 April 2017 (UTC)[reply]

I like the effort. Here's with a tweak: "While a lack of Verifiability (on its own) is enough to justify the exclusion of material, the opposite is not always true. Verifiablitity (on its own) may not be enough to justify inclusion. This is because Wikipedia has other policies, guidelines and article development practices that also play a role in determining whether some bit of information is included or excluded. North8000 (talk) 14:55, 25 April 2017 (UTC)[reply]

"Verifiablitity (on its own) may not be enough to justify inclusion" has the same problems as "Verifiability is not a reason for inclusion". I agree with Blueboar that focusing the message in "there may be other reasons to exclude the content" is more likely to get us a consensus than "verifiability can't be used" to justify/support inclusion. I would summarize it removing that part: "While a lack of Verifiability (on its own) is enough to justify the exclusion of material, the opposite is not always true. Wikipedia has other policies, guidelines and article development practices that also play a role in determining whether some bit of information is included or excluded." Diego (talk) 15:44, 25 April 2017 (UTC)[reply]

I support this wording. It respects the two unspoken realities I most often observe: 1. The verifiability standard is not a reason for inclusion itself, rather, the fact that an editor has inserted a datum meeting the verifiability standard -- before or after being reverted -- is a reason for inclusion because it is an implicit assertion of the editor's assessment that it is relevant. 2. An insertion is reverted citing lack of verifiability often because that's the easiest way to get edits deemed violative or trivial prima facie out of the article summarily. If the factoid can't be verified, often that's the end of it -- as intended. Implicitly, however, the takeaway for the serious editor, especially if a newbie, is that proven verifiability will constitute a ground for retention, otherwise the revert would have been accompanied by a different justification. FactStraight (talk) 00:04, 26 April 2017 (UTC)[reply]

I support Diego's wording. Sławomir Biały (talk) 00:17, 26 April 2017 (UTC)[reply]

I think the sentence "Verifiablitity (on its own) may not be enough to justify inclusion." is important. I think it is the heart of the warning that we want to give editors (especially new editors). WP:V is usually the first policy a new editor encounters. A new editor quickly learns that he/she needs to add sources to support additions... but then new editors often feel blindsided and bitten when that verifiable information continues to be challenged (even though they have added a source). I think it vital to state upfront (and in blunt language) that Verifiability isn't always enough. Blueboar (talk) 11:55, 26 April 2017 (UTC)[reply]

I like Diego's better than Blueboar's(/mine). I think that that sentence does more harm than good.North8000 (talk) 13:02, 26 April 2017 (UTC)[reply]

I think that the current wording is better but could support Diego's as a final compromise to settle this. If this goes any further, we should be more precise about it. What would come out from exactly where and what would go in exactly where. North8000 (talk) 10:52, 26 April 2017 (UTC)[reply]

FWIW, Diego's looks good to me. (Another drive-by comment from – Philosopher ^{Let us reason together.} 14:43, 26 April 2017 (UTC))[reply]

Remove the words in parentheses. Spell out what's "the opposite". Replace "play a role" with "may determine". Remove "some bit of" and "While a". Change "is included or excluded" to "should be included or excluded". Use "information" because "material" appears to be contested (although I prefer "material"). And keep the last sentence of WP:ONUS of course: Lack of verifiability is enough to justify the exclusion of information, but verifiability is not sufficient to justify inclusion. Other Wikipedia policies, guidelines and practices may determine whether information should be included or excluded. The onus to achieve consensus for inclusion is on those seeking to include disputed content. Bright☀ 15:32, 26 April 2017 (UTC)[reply]

I like the shortened parts; but I'm with North8000 that it's a mistake spelling out an absolute "is not sufficient". If the goal is to warn newcomers that sometimes verifiable content will be further challenged, then such "sometimes" should be added to the wording.

Also, I'd like to add a Wikilink to WP:Consensus, given that it's a relevant part of the final sentence in ONUS. I think it would even make sense to include the initial sentence at Determining consensus, which clarifies how policies and guidelines apply to that consensus-building.

So this is how I'd have it: Lack of verifiability is enough to justify the exclusion of information, but verifiability may not always be sufficient to justify inclusion. Other Wikipedia policies, guidelines and practices may determine whether information should be included or excluded. The onus to achieve consensus for inclusion is on those seeking to include disputed content. Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy". Diego (talk) 15:55, 26 April 2017 (UTC)[reply]

I like this as a way forward. I'm concerned that the proposed phrasing may be inaccessible to some editors, particularly if English isn't their first language. I suggest using simple words and focusing on what editors can do or should do. Here's my draft:

Hope this helps—S Marshall T/C 16:30, 26 April 2017 (UTC)[reply]

I think that "practices" or even better "article development practices" should stay in. North8000 (talk) 16:37, 26 April 2017 (UTC)[reply]

The more accurate version is better. There are no "rules", there are policies and guidelines, and we shouldn't use second person. The sentence explaining consensus is detrimental, consensus should be explained on WP:CONSENSUS so there's no conflict between policies. Bright☀ 16:41, 26 April 2017 (UTC)[reply]

• Yes, my more accurate version is better.  ;) Seriously -- Wikipedia is full of rules. Some of them are called policies and some of them are called guidelines, and some of the guidelines (such as WP:N) are treated like policies, and some of the policies (such as WP:PRESERVE) are treated like guidelines; and there are occasional essays such as WP:ATA that are actually bright-line rules. Wikipedia is overflowing with rules, and it's okay to say so. Using the second person is certainly appropriate -- whyever would you think it isn't? Your position makes no sense to me at all. Since we're trying to be clear, it is in fact a great deal more appropriate to write in simple declarative sentences than needlessly wordy hortative circumlocutions such as "editors may". This is guidance for editors trying to work their way through difficult judgment calls, not a set of rules put up on an office wall. Someone reading it needs to know what they can and can't do, what they should and shouldn't do, and what they need to think about. (It wouldn't hurt us to say "please", either. These are volunteers and they're entitled to respect.) You and J Johnson will fight this, of course, because it's me saying it, but there are others I might just convince.  :)—S Marshall T/C 20:38, 26 April 2017 (UTC)[reply]

S Marshall is it a coincidence that when (not) working towards consensus you use the wording that's given as examples of what not to do? WP:BEANS... Bright☀ 08:31, 27 April 2017 (UTC)[reply]

The sentence explaining consensus is copied verbatim from Determining consensus (the "Original" version at least). Diego (talk) 17:17, 26 April 2017 (UTC)[reply]

I think it better to refer to other policies / guidelines than to copy them to here. North8000 (talk) 17:30, 26 April 2017 (UTC)[reply]

True for whole sections, but I think copying a snippet or sentence here and there may clarify what part of the other policy is the most relevant, more than a single unqualified link. Diego (talk) 17:37, 26 April 2017 (UTC)[reply]

I agree that simplifying the language may be a good idea. Taking into account the above feedback, we have:

Editors may remove information that isn't verifiable. Even if information is verifiable, it can still be removed because of other Wikipedia policies, guidelines and article development practices.

The onus to achieve consensus for inclusion is on those seeking to include disputed content. Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy.

I like that "Even if information is verifiable, it can still be removed" is different from "verifiability is not sufficient to justify inclusion" and completely avoids "not a reason". Diego (talk) 17:32, 26 April 2017 (UTC)[reply]

I support this without the line break and without the last sentence. Put a wikilink for consensus in the previous sentence instead. Bright☀ 08:31, 27 April 2017 (UTC)[reply]

Support, with BrightR's fix. Also, IMO "isn't" -> "is not". I like that the focus is shifted to already written article text rather than to "preemptive hypothesizing" on what to include and what not. Writing and article is akin to brainstorming. A person finds a piece of info they think worth including. Some other thinks this is a worthless bit. Clearly a case of WP:CONSENSUS to work. IMO nothing shall prevent editors from reasonable, in their opinion, additions. Staszek Lem (talk) 17:21, 27 April 2017 (UTC)[reply]

@Staszek Lem: DO NOT EDIT MY COMMENTS. Bright☀ 13:36, 28 April 2017 (UTC)[reply]

@BrightR: DO NOT PANIC AND SHOUT and AGF. It is customary to highlight !ivotes, for clarity. Didn't you notice everybody else is doing so? If you think my edit distorted your intentions, I apologize. I will keep in mind that in the future. Staszek Lem (talk) 18:14, 28 April 2017 (UTC)[reply]

Tired of all these editors who should know better editing my comment to what I "really" mean. Don't edit other people's comments ever, with these exceptions. Bright☀ 18:25, 28 April 2017 (UTC)[reply]

OK. Sorry again. Staszek Lem (talk) 18:57, 28 April 2017 (UTC)[reply]

Agree with both BrighteR and Staszek Lem's fixes. A question about "article development practices", though: Is this a new term of art on Wikipedia, similar to "policy" and "guideline"? It sounds clunky and I would rephrase it unless it's widely used in other policies and guidelines. – Philosopher ^{Let us reason together.} 20:32, 27 April 2017 (UTC)[reply]

I made it up, it was an attempt to generally refer to the normal editorial processes, normal editorial decision-making processes...basically everything else that goes into building articles aside from / in addition to the constraints placed by policies and guidelines. North8000 (talk) 21:21, 27 April 2017 (UTC)[reply]

Support, for "Even if information is verifiable, it can still be removed because of other Wikipedia policies, guidelines and article development practices. The onus to achieve consensus for inclusion is on those seeking to include disputed content." I consider that "article development practices" include the to-and-fro process whereby specific facts, word choices, grammar and style are agreed upon, i.e. the collaborative editing which resolves disagreements not addressed by or too trivial to invoke a policy or guideline, but about which concurrence is necessary to stabilize an article's content. FactStraight (talk) 01:42, 28 April 2017 (UTC)[reply]

Support Without the last sentence. North8000 (talk) 11:37, 28 April 2017 (UTC)[reply]

Needs a fix of the issue raised by TransporterMan. North8000 (talk) 14:28, 28 April 2017 (UTC)[reply]

• Weak support - Not really happy that this is being phrased as a permission focused on removers ("Editors may remove...") rather than as a warning to adders that removal might happen. But, I can live with it, if it will move us to consensus. Blueboar (talk) 12:04, 28 April 2017 (UTC)[reply]
• Support my proposal, removing the last sentence as long as the wikilink to "achieve consensus" points to the section Determining consensus of WP:CONSENSUS (which focuses on explaining how to achieve consensus, rather than what is the concept of consensus). Diego (talk) 12:07, 28 April 2017 (UTC)[reply]
• Oppose This can be read, because of the absolutism of "is not", to reverse the standard set in BURDEN that information can be removed simply because it is unsourced, with the only requirement (if that) being that you express a "concern" that it may not be verifiable. May not, not is not. I acknowledge that the link to BURDEN ought to fix that, but the words will be cited to the opposite effect and a complex argument will have to be made to refute it. This cures a minor drama by creating a bigger one. Regards, TransporterMan (TALK) 13:17, 28 April 2017 (UTC)[reply]

What if the first sentence is changed to "Editors may remove information if they think that it may(P.S: might?) not be verifiable"? (or "if they have the concern that...", to keep it closest to the wording of CHALLENGE). Diego (talk) 14:12, 28 April 2017 (UTC)[reply]

Good point TransporterMan, that could really cause a problem.North8000 (talk) 14:28, 28 April 2017 (UTC)[reply]

perhaps "might" is a better word than "may"?... just a thought. Blueboar (talk) 14:31, 28 April 2017 (UTC)[reply]

Would that make a significant difference? The text at WP:CHALLENGE uses "may"... Diego (talk) 14:48, 28 April 2017 (UTC)[reply]

The problem is that it can be read to place a higher requirement for removal of material based on wp:ver grounds....putting the burden on the remover to show that it isn't verifiable. One idea "verifiability is a requirement for inclusion" ...I know this is retro but I think not in any disputed area? North8000 (talk) 15:03, 28 April 2017 (UTC)[reply]

I'm afraid that "Editors may remove information if they think that it may/might not be verifiable", while accurate, may result in an rebellion from the preservationist point of view here. The current hard-fought language in BURDEN strikes a subtle balance supporting the removalist point of view, but the suggested language can be read to tilt that balance even more in that direction. While I, as a removalist, am down with that, I even more do not want to argue about it here — again — for another six months. On another point, this language also suggests that existing sourced material in which the source is clearly not reliable or in which the source is challenged as being unreliable can simply be removed in a single step, rather than removing the bad or challenged source as a first step and removing the material only if a new reliable source is not provided. We've talked here about whether that should be a one-step or two-step process without, as far as I can recall, coming to a firm consensus. Again, as a removalist I'm down with a one-step and this language would support that being the base standard (but having said that I also agree that two-step is the better practice, though it should be as it is now optional, not required). Once again, however, I don't think that the preservationists are going to care much for that hardening of position. I've stayed out of this discussion, while watching it, because I thought we could live with either the pre-change language or the post-change language (though my choice would be the old title with the new text), but we need to be careful not to mess other stuff up. Best regards, TransporterMan (TALK) 16:01, 28 April 2017 (UTC)[reply]

I'm guessing not, but is "Editors may remove information they believe is not verifiable" any better? DonIago (talk) 16:59, 28 April 2017 (UTC)[reply]

Maybe we are getting off track. I think we're talking about an area that is not under debate, a sentence or phrase that is just a preface to the wording that is under debate. If so, perhaps we should just the old or current version preface sentence, e.g. "verifiability is a requirement for inclusion". North8000 (talk) 17:24, 28 April 2017 (UTC)[reply]

I'm with DonIago. "You may remove information if you don't believe it is verifiable." Or, if you have some fundamental objection to writing a clear procedure, I'd agree to "Editors may remove information which they don't believe is verifiable" as a less-good alternative.—S Marshall T/C 00:41, 29 April 2017 (UTC)[reply]

Now we're starting to discuss wording in an area that is not even in dispute. And the overall RFC being malformed (with no explicit resolution proposed) is not going to navigate us out of this. And so I think we're really getting mired down. But, on the positive side, I think that 90% or more of us are evolving towards something that might fly in the core areas of disagreement. I'm going to create a new section that that tries to gel something to move forward. North8000 (talk) 22:04, 29 April 2017 (UTC)[reply]

I thougt (a month ago) that we all understood the issue here to be solely about the inclusion (or exclusion) of verifiable material. As to unverifiable material: the lede says it may be removed. And in the case of contentious BLP material it asks for immediate removal. If an editor has a question of whether material is verifiable it can be tagged, and perhaps this should be mentioned even in the lede. But that's not what we are here for. The issue here is about the addition or removal of verifiable material, and please let's not get off-track. ~ J. Johnson (JJ) (talk) 22:03, 29 April 2017 (UTC)[reply]

Looks to me like the "off-tracking" can be resolved by distinguishing between potentially and actually sourced edits thus, ""Even if information is verified, it can still be removed because of other Wikipedia policies, guidelines and article development practices. The onus to achieve consensus for inclusion is on those seeking to include disputed content." FactStraight (talk) 14:46, 1 May 2017 (UTC)[reply]

I incorporated suggestions expressed so far in #Diego's version II. Staszek Lem (talk) 18:50, 1 May 2017 (UTC)[reply]

Try to gel and organize a compromise and move forward

First, on an organizational side, let's say that step one is trying to gel the "main statement" in the body of the policy. Then, once that is accomplished, we could discuss the details on the three places (lead, subheading and policy-body statement) involved. Second, if we could get an OK from 90% of the active participants, I think we could go slightly-bold and edit it.

How's this?:

Verifiability is a requirement for inclusion but not a reason for inclusion. Inclusion is subject to other policies, core guidelines and other article development processes. However, in inclusion/exclusion discussions, particulars related to the sourcing of material may be considerations or reasons for inclusion.

Sincerely, North8000 (talk) 00:07, 30 April 2017 (UTC)[reply]

You have an extra "for" in the first sentence, it appears. I also don't really care for the last sentence, though I'd be willing to defer to others on that one. DonIago (talk) 02:29, 30 April 2017 (UTC)[reply]

Thanks, I fixed it.North8000 (talk) 12:51, 30 April 2017 (UTC)[reply]

Nope. This is a step backwards, as it returns the word "reason". Multiple comments have objected to that word, saiding that verifiability can be "a reason". We are not going to find consensus if we use that word... so try again. Blueboar (talk) 01:28, 1 May 2017 (UTC)[reply]

I think the problem is not so much with the use of word itself, but with the differing views on whether verifiability is, or is not, a "reason". Avoiding the word leaves the differences hidden, and any supposed consensus is illusory if it only papers-over the differences. ~ J. Johnson (JJ) (talk) 23:44, 1 May 2017 (UTC)[reply]

No. It was noticed here the the term "inclusion" does not have a unique meaning, and people may be not aware during a discussion that they are speaking in same terms. Same with "reason". Staszek Lem (talk) 02:59, 1 May 2017 (UTC)[reply]

I suspect you mean "not aware ... that they are not speaking in same terms". In that case, yes, that is a good part of the problem, and why any discussion on this needs explicit qualification of terms. ~ J. Johnson (JJ) (talk) 23:50, 1 May 2017 (UTC)[reply]

I'm trying to wear a second hat,removing my "one side" hat and just trying to facilitate something. To save time and words I'm withdrawing the above. Based on one exchange above I mistakenly thought that having wording that says that sourcing can provide reasons for inclusion would assuage the concerns of person who don't like the current wording, but as elucidated by Blueboar and Staszek Lem I was wrong. Blueboar and Staszek Lem, I was wondering if you could give your opinion on the "Sorting out a few fundamantals" section above? This would help clarify if there is a an underlying fundamental disagreement or whether its mostly a matter of wording or other considerations. North8000 (talk) 11:56, 1 May 2017 (UTC)[reply]

I believe that #Diego's version is on the right track. However' judging from "oppose" arguments there (and the "opposes" in onther places) people want to squeeze all our major content policies into a single phrase. (Which is probably good to have, but quite hard). Staszek Lem (talk) 16:09, 1 May 2017 (UTC)[reply]

I also like Diego's version. I thought that the last sentence was a little awkward and possibly unnecessary, but had no specific objection to it. Also TransporterMan pointed out a specific problem which I missed, and which probably conflicts with wp:ver and so that will probably need to be fixed. But then if we try to write a more accurate of "may remove" we'll end up having write half of the whole policy in here. My draft above was a (botched) attempt to follow the principles of Diego's version. I also thought about about total removal of those type statements as a (compromise) possibility. I'll noodle on it. North8000 (talk) 17:21, 1 May 2017 (UTC)[reply]

OK so here is Staszek Lem's version (copied from above and then tweaked by them) which they evolved from Diego's version:

Information that is not verifiable may be contested and eventually removed. Even if information is verified, it can still be removed or moved to a different article because of other Wikipedia policies, guidelines and article development practices. The onus to achieve consensus for inclusion is on those seeking to include the disputed content. North8000 (talk) 18:48, 1 May 2017 (UTC)[reply]

• Support as a compromise I think that the current version is better but would support this as a compromise if that would settle it. North8000 (talk) 18:51, 1 May 2017 (UTC)[reply]
• Support as the editor of modifications. Staszek Lem (talk) 19:09, 1 May 2017 (UTC)[reply]
• Oppose due to inclusion of word "eventually", which implies that unverifiable information may not be immediately removed, which it can be (though there are better, though optional, practices). Best regards, TransporterMan (TALK) 19:41, 2 May 2017 (UTC)[reply]
  • Our policies are to describe better practices, not what is actually and sloppily and aggressively and stupidly done. A better practice is to discuss text rather than slash and cut, unless it is outright nonsense. We don't have to squeeze all our guidelines about challenging the content into this small paragraph, which focus is role of WP:V. Staszek Lem (talk) 21:21, 2 May 2017 (UTC)[reply]
  • P.S. Originally I did think of writing "contested or outright deleted", but I realized this this will be immediately objected by "preservationists", who will interpret as a direct invitation to delete everything not referenced. Staszek Lem (talk) 21:53, 2 May 2017 (UTC)[reply]
    • I concur with your PS, but over repeated, hard-fought discussions here we've created a series of acceptable and better practices and "eventually" can be easily argued to change that consensus. I continue to object. Best regards, TransporterMan (TALK) 15:05, 3 May 2017 (UTC)[reply]

Maybe our problem is that we're unwittingly trying to write a a policy summary instead of dealing with the core issue. The pre-February wording hinted that meeting wp:ver is a force for inclusion. The current wording states clearly that it is not. Maybe a simple middle ground compromise is to have wp:ver be silent on that topic. And so completely remove the disputed material, (plus one related 5th FAQ) and don't put anything in it's place. North8000 (talk) 11:30, 3 May 2017 (UTC)[reply]

Just want to repeat that the intent of the pre-February wording was definitely not to make a policy statement about inclusion (or exclusion)... the intent was simply to alert editors (especially new editors) to the fact that there is more to the whole mishmash of determining inclusion/exclusion than just WP:V. I have a feeling that some of us who have been involved in this debate (on both sides) have been over-interpreting the section (treating it, intentionally or unintentionally, as if it were some sort of WP:INCLUSION policy statement). Blueboar (talk) 12:21, 3 May 2017 (UTC)[reply]

I'm kind of worn out. Most of the attempts at compromise fatally waded into unintended other more complex areas. Of course all versions include that verifiability is a requirement for inclusion. The question is whether or not meeting verfiability carries additional weight for inclusion. The changes in February to the current version basically boil down to changing "not a guarantee" (which does not say but sort of implies that meeting wp:ver carries some additional weight) to "not a reason" which clearly says that it doesn't. Note that saying what it is "not" implies that it might be everything less "extreme" (lower on the below list) than that. This range could be (in order):

1. Guarantee
2. Force, strong argument for
3. Reason

Two areas of possible "middle of the road" compromise are

1. Use "force" (or something equivalent at that level) instead of "reason"
2. Keep "reason" but give it less prominence. Maybe take it out of the subsection title (but don't go back to the the other "not a guarantee" extreme) and/or take the new summary completely out of the lead.

If this looks to be of interest, I could ask the question in a way conducive to getting organized feedback. Or we could just take a rest and sit back and see what happens. With no action items proposed in the RFC (and thus none discussed) , my guess is that it will end up staying as the current version, and then on to a new RFC and another 100,00 words. Maybe that is a motivation for a compromise :-) :-) Sincerely North8000 (talk) 19:47, 4 May 2017 (UTC)[reply]

I don't know what kind of concessions could be made either way to make a compromise. The current state of this issue seems to be akin to which side of the road to drive on, and "down the middle" is not a viable option.

I think the best result we could hope for is that someone (with strong analytical skills, and lot of spare time) works out some statements of the underlying concerns, the philosophical differences, and examples showing actual problems, all in language that we all interpret in the same way. Then we would have a basis for starting a discussion. ~ J. Johnson (JJ) (talk) 23:35, 5 May 2017 (UTC)[reply]

Support current article language. The old "WP:V Classic" language can be reasonably misconstrued even by a seasoned admin. Many conscientious editors have understandably interpreted the old formulation to mean that providing an RS citation for a claim warrants that it "bears repeating in an encyclopedic article".

This misconception may then force the editor to discount the WP:UCS essay recommendation that you add (or remove) a claim only if you thereby "better inform your readers".

Illustration: A (formally) reliable source says that Syria dropped 12,960 barrel bombs last year, killing 635 civilians. You add that claim per WP:V Classic. You're thereby leading unwary readers to think that it takes 20 barrel bombs to kill 1 civilian; yet you reasonably believe you're just observing the policy. --Dervorguilla (talk) 07:26, 6 May 2017 (UTC)[reply]

I fear your example is somewhat spoiled in that the inference (calculation?) of the number of barrel bombs per civilian is likely covered by other policies, such as OR. A better example might be the addition of "Syria dropped 12,960 barrel bombs last year" (reliably sourced material!!) into, say, an article on Sex determination factors of some exotic butterfly. Lacking some showing of relevance (perhaps those bombs leave a residue in the main wintering grounds of that butterfly?) I would say verifiability does not warrant inclusion. Perhaps that is more to your point? ~ J. Johnson (JJ) (talk) 23:40, 6 May 2017 (UTC)[reply]

It doesn't even have to be on the wrong subject. We could probably find a reliable source for the location of certain highly public figures on every day. But does any article need a complete list of which days Donald Trump spent in which cities? Is any article going to be improved with a sentence that says "Donald Trump was in the White House last Tuesday?" WhatamIdoing (talk) 16:48, 17 May 2017 (UTC)[reply]

Disputed section header

I looked into the current version once again and I see we forgot the highly contested change of section header.

• Now:
  • Verifiability is not a reason for inclusion
• Suggested:
  • Verifiability does not guarantee inclusion

Staszek Lem (talk) 19:05, 1 May 2017 (UTC)[reply]

I opened with suggesting just getting the main sentences agreed on and then moving onto specific changes in three areas. I just realized that the above main body text is already ready-to-be specific, we just need to say "replace the body of the xxx sub-section with:". But that still leaves the sub-section title and the summary in the lead. Maybe it would be best to leave that go? I would oppose the above. While the body compromise wording sort of goes to the middle ground, this title IMHO doesn't, it goes back to implying that it is a force for inclusion. . Maybe a neutral one something like "Inclusion and exclusion" or "Verifiabiity, inclusion and exclusion"?North8000 (talk) 20:41, 1 May 2017 (UTC)[reply]

• Call it "Removing verifiable information".—S Marshall T/C 23:33, 1 May 2017 (UTC)[reply]

It has been discussed previously that "guarantee" is a stronger word than "reason". And that the lack of a guranteee is implicit, as the alternative – that verifiability guarantees inclusion – is just what some editors have vigorously insisted no one is arguing for. Denying that verifiability is guaranteed (a strawman?) still leaves open its use as a reason, which is the essence of the issue. ~ J. Johnson (JJ) (talk) 00:06, 2 May 2017 (UTC)[reply]

It's not necessary to split that hair. A compromise is when we use language all sides can agree on. We won't agree on "reason" or "guarantee".—S Marshall T/C 19:38, 3 May 2017 (UTC)[reply]

It's not a "hair" when a certain editor (you) has called it a "drastic change" 22:02, 17 Apr, and "a novel interpretation that can't remain in this policy unless there's a community consensus" 18:10, 12 Apr. And it is already split, as reflected in the strong division of opinion on this specific point. As to using "language all sides can agree on": exactly. But we don't agree on "guarantee" either, so by your reasoning that is also out.

What you seem to not understand is that no words will suffice where we have fundamental differences in the concepts they are expected to express, or where we have different interpretations of the words. But possibly you were in a serial stream-of-consciousness editing mode and hadn't seen my prior comment just below here. ~ J. Johnson (JJ) (talk) 21:25, 3 May 2017 (UTC)[reply]

It seems that one side objects to the word "reason", and the other side objects to the word "guarantee"... may I suggest that the only way to find consensus is to find another word. One which neither side objects to. Blueboar (talk) 00:37, 2 May 2017 (UTC)[reply]

Since we're just talking about the section header here- "Verifiability as a basis for inclusion"? It makes no claims as to whether V is or is not actually a basis for inclusion, just lays the groundwork for the detailed information to follow...at least, that's how I read it. DonIago (talk) 05:44, 2 May 2017 (UTC)[reply]

That's getting close to a neutral ground. I think making it a question (if that is allowed in a title) would be even better. "Is verifiability a basis for inclusion?"North8000 (talk) 11:51, 2 May 2017 (UTC)[reply]

The words represent the different views (interpretations), which need to be resolved to a point of shared understanding, not avoided. If the policy is interpreted differently by different editors we have, effectively, different policies. And (quite importantly) the resolution has to be in (or at least mappable to) the language and terms each "side" uses, and address the issues raised. Without that, whatever words are used have no connection to the issues. ~ J. Johnson (JJ) (talk) 22:59, 2 May 2017 (UTC)[reply]

The header must succinctly represent the content. If "guarantee" is bad word then "basis" is just the same. A "non-prescriptive" variant would be:

"Verifiability of text and its inclusion".

(BTW, we habitually use this "and" trick in article titles to make them NPOV: e.g. Islam and violence") Staszek Lem (talk) 19:17, 2 May 2017 (UTC)[reply]

Looks OK with me. North8000 (talk) 19:31, 2 May 2017 (UTC)[reply]

• Call it what it's about: "Removing verifiable information".—S Marshall T/C 19:38, 3 May 2017 (UTC)[reply]

• Alternately: "Non-removal of verifiable information. "

Stuart, you are looking at only one facet of the issue. Which was explained to you back in February or so, but apparently to little effect. ~ J. Johnson (JJ) (talk) 21:29, 3 May 2017 (UTC)[reply]

• Just so everyone knows... "Verifiability does not guarantee inclusion" was just quoted at RSN. Blueboar (talk) 18:16, 5 May 2017 (UTC)[reply]

I couldn't find it there. Do you mean someone saying "Reliability does not guarantee inclusion"? North8000 (talk) 18:51, 5 May 2017 (UTC)[reply]

See the last paragraph of the section on "Conflict between Primary and secondary sources" Blueboar (talk) 21:12, 5 May 2017 (UTC)[reply]

Comment by Kautilya3 (11:48, 5 May) at WP:RSN#Conflict between secondary and primary sources: "Not everything that appears in RS needs to be included in a Wikipedia article." ~ J. Johnson (JJ) (talk) 22:56, 5 May 2017 (UTC)[reply]

How to best close this discussion?

I discussed closing this discussion at User talk:S Marshall#Who to close the RFC discussion at Wikipedia talk:Verifiability and then requested teamwork closure at the WP:AN. Winged Blades of Godric said he'll volunteer as part of teamwork closure, so I'm awaiting one or two more closers. --George Ho (talk) 15:30, 3 May 2017 (UTC)[reply]

I think that our best hope is to try to work out a compromise. As I indicated above, the RFC just asks for an overall opinion on collectively several different changes made in February, without asking about or proposing any specific actions on any of them. If anyone is thinking that it is asking about any particular action, it is malformed with respect to that. And so it's hard to see any resolution come out of a close of the RFC as worded. North8000 (talk) 15:56, 3 May 2017 (UTC)[reply]

The fact that I didn't specify any actions in the question certainly does not prevent any actions in the close. It's for the closers to determine the actions that arise from this RfC. If there's a consensus, they will determine the appropriate actions based on the consensus they find; and if there's no consensus, they will decide what the status quo ante is, and restore it.—S Marshall T/C 17:04, 3 May 2017 (UTC)[reply]

My point is that they are unlikely going to extrapolate to actions from the RFC when it did not specify any actions for discussion. An so it would be a good idea to keep trying on a compromise. Sincerely North8000 (talk) 17:53, 3 May 2017 (UTC)[reply]

The problem with taking any kind of action that was not proposed in the RfC is the lack of proper consideration. That a possible action was raised in any subsequent discussion is inadequate, as many respondents will comment on the RfC as stated, and then move on, with no regard to any subsequent proposals.

Certain editors have argued that the consensus reached for the existing language was inadequate for something as important as WP:V because it had only eight or so editors. If that is so, then having only three admins vote on something that possibly wasn't on the menu would be even less adequate. The underlying issue needs resolution, but this RfC doesn't do that, doesn't even provide a basis for doing that. ~ J. Johnson (JJ) (talk) 21:36, 3 May 2017 (UTC)[reply]

Rather than pre-emptively challenging what the closers might or (might not) say... Let's wait, and see what the closers actually say. Blueboar (talk) 11:27, 5 May 2017 (UTC)[reply]

I wasn't meaning to pre-emptively challenge, only to show that any closure done on the basis of (say) three admins would be subject to the same "inadequate basis" objection raised against the current language, and thus a basis for challenge which ever way the result might go. ~ J. Johnson (JJ) (talk) 22:54, 5 May 2017 (UTC)[reply]

• J. Johnson, the above is too much to read, but I'm having difficulty seeing what the substantive difference is between the versions. Both mean that verifiability is a necessary but not a sufficient condition for inclusion. What is the difference between the versions such that it has sparked so much discussion? SarahSV ^(talk) 21:40, 16 May 2017 (UTC)[reply]

Yes. On all accounts. With the key differences being deeper than the words.

It is difficult to see anything in this RfC. This is due to a number of reasons, starting with poor formulation and total lack of definition of key concepts and terms. Discussion has gone into the ditch several times because of different interpretations of various terms, failure to identify key caveats and assumptions, failure to observe those terms (etc.) that have been identified, and editorial malfeasance. In the discussion last February the intent was not to change the policy, but to clarify it. But it seems that unclear language has allowed ambiguous interpretations; clarifying it has exposed those fundamental differences. It's not so much the words themselves, but the strong feelings as to what they mean.

There is also a deep divide in fundamental philosophy, such as whether certain material should be "guaranteed" a place "somewhere in the encyclopedia". As it is, a bunch of editors have been arguing at cross-purposes, and about supposed cases that don't seem to exist. Without resolving – which would start by identifying – all these differences, I think the only compromise possible is one where the language (whether the older form, or something new) is again so ambiguous everyone think it supports their interpretation. Which only puts the bickering off to individual cases where the same language is invoked for different ends. ~ J. Johnson (JJ) (talk) 23:59, 16 May 2017 (UTC)[reply]

Several responses, in both the yes and no columns, could apply to either version. SarahSV ^(talk) 00:48, 17 May 2017 (UTC)[reply]

If the whole discussion is very difficult to summarize and close, is closing separate subsections, including the original discussion (at the very top), possible? If so, which others besides the very top one need separate closures? George Ho (talk) 03:07, 17 May 2017 (UTC); struck. 19:28, 17 May 2017 (UTC)[reply]

It's not in any way difficult to summarise or close. It's bloated but not complicated at all. Despite J Johnson's continuing refusal to understand this point, nobody at all thinks verifiability should guarantee inclusion and nobody at all has said so. During the discussion, I said that material that's genuinely attributable to a truly reliable source almost always belongs somewhere in the encyclopaedia; but the language I and several others advocate in the policy is that verifiability is not a guarantee of inclusion.

I, and certain editors who agree with me, object to the phrase "verifiability is not a reason for inclusion", mainly because we feel that verifiable information is more includable than information that isn't verifiable. The other side disagree.

Nothing about this discussion is hard to understand or complicated to close. The reason for requesting a triumvirate close is to put the conclusion beyond doubt and enable this discussion to come to an end.—S Marshall T/C 18:35, 17 May 2017 (UTC)[reply]

Got it. Thanks. ...Um, I already know that the original question is simple, but thanks anyway. :) George Ho (talk) 19:28, 17 May 2017 (UTC)[reply]

What is bloated is this discussion. As to ending it, that is very simple: just stop. As for a "conclusion beyond doubt" – hah. There is no basis for that, as there has been no resolution of anything. As I noted just a few paragraphs above, ANY result here regarding any wording will be challengable. And the strong and substantial objections ON BOTH SIDES shows there is no consensus, either way.

Marshall, I am not "refusing to understand" any point. At nearly every point where our "understandings" seem to differ I have tried to sort them out. If my understanding is in any way incorrect or deficient, it is likely attributable to your failure to explain. For the benefit of any newcomers I will note that in the discussion leading up to this RfC (now archived) you refused to clarify your ambiguous statements. Particularly, at 23:57, 25 Mar you said: "I decided not to get sidetracked into a semi-relevant discussion and didn't answer some questions."

That you say "[n]othing [here] is hard to understand or complicated to close" doesn't make it true. Of course, it may seem that way if you understand only one point of view. If you think it would be "not difficult" to summarize all that has been argued here, please do so. But it would be no basis for closing if it is incomplete or one-sided. ~ J. Johnson (JJ) (talk) 23:19, 17 May 2017 (UTC)[reply]

I think the (unstated, but obvious) question that lies at the heart of this RFC is: Should the policy a) go back to the pre-February language, b) continue with the post-February language, or c) use some other (yet to be determined) language? That is essentially what we have been debating all this time. If the closers can give us an answer to that question, then we will know how to move forward. Blueboar (talk) 00:16, 18 May 2017 (UTC)[reply]

The challenge is "unstated", plus only the non-specific is "obvious". Sincerely, North8000 (talk) 13:03, 18 May 2017 (UTC)[reply]

Yet a consensus of "no" to the RfC question would clearly reject option b), the text after the post-February change. Diego (talk) 13:21, 18 May 2017 (UTC)[reply]

Sorry, no, but "what we have been debating all this time" is NOT "obvious". Sure, we could even all agree that "it's obvious", but are we even talking about the same thing? Marshall certainly wants to revert the previous language, but that is not what he asked. He asked (regarding the change): "Q: Was this a good idea?" Whether it "was a good idea" and whether it "should be reverted" are two different questions. And it is quite conceivable that one might think "it was a good idea, but should be reverted", or that "it was not a good idea, but should not be reverted". To ask one question, then apply the answers to a different question, is your basic switch-and-bait tactic. It's saying one thing, but meaning something else. Yet another reason why the past three months of discussion here is worthless. ~ J. Johnson (JJ) (talk) 08:47, 19 May 2017 (UTC)[reply]

You're right that, "the change is a bad idea" is a different question than "the change should be reverted". But it is also obvious that if the change was a bad idea, it can't work as the guideline and can't stand there, as it wouldn't represent the community consensus. This leaves us with only options a) and c) above, either reverting or changing it to something else. Diego (talk) 09:49, 19 May 2017 (UTC)[reply]

Yeah, the RfC question, the responses, and what to do if there is consensus are perfectly clear. The !votes are pretty evenly split, so consensus is not immediately clear, nor is it immediately clear what to do if there is no consensus (though I have my own opinion), but that's for the team of admins to decide.--Wikimedes (talk) 10:12, 19 May 2017 (UTC)[reply]

I think that J. Johnson's post goes to the core of the formation/malformation of the RFC. If some were advocating the closers dictate action on the project page, they are basically advocating a new RFC with participation limited to the three closers,(presumably with input from this RFC) not a closing of this RFC. This gets stretched even further when one notes that there is really no consensus either way in this RFC. Looking at it from another angle, the question would become whether what was developed in the February process has standing vs. being considered to be a "new" proposal. North8000 (talk) 10:47, 19 May 2017 (UTC)[reply]

The admins who have been asked to close this RFC can read the discussion and reach their own conclusions as to how to close it. It may be that the original question was malformed... but the closers are intelligent people... they can take that malformation in stride, and answer the unstated question that lies behind the stated one. To argue that they are limited to only answering the stated question is pure Wikilawyering. It is natural that each of us wants them to close this RFC in a way that supports our own opinions... and we have all repeatedly stated what those opinions are. Now, I suggest that we back off, and let the closers do their job. Blueboar (talk) 12:58, 19 May 2017 (UTC)[reply]

I don't think that anyone has argued the extreme they they are so limited, and I think that nobody has sunk to the normal meaning of wiki-lawyering, and so I would take issue with the use of that term here. And BTW my highest hope is that it ends up with none of the nice & great people who have been participating here being hurt or upset. For me the result is secondary to that. North8000 (talk) 17:16, 19 May 2017 (UTC)[reply]

(ec)Has the closure discussion among admins begun? So far I see Winged Blades of Godric volunteering at Wikipedia:Administrators'_noticeboard#Needing_more_than_one_to_close_RfC_discussion_at_WT:V, but no one else. I'm worried that the request for closure has been forgotten, or that potential closers are waiting for the RfC participants to do something while the RfC participants are waiting for the closers to do something. On a related note, although the RfC discussion is sporadically and occasionally continuing, and there are probably many wording compromises that could gain clear consensus support, I don't see any compromise wording getting enough support within this RfC - there's just too much of a wall of text for enough people to be bothered to find a newly proposed wording and express an opinion on it.--Wikimedes (talk) 17:55, 19 May 2017 (UTC)[reply]

Diego's comment reminds me (and my apologies for letting this slide earlier) of a further ambiguity: one can object to the change in respect of its substance, or its form or process. E.g., one of Marshall's objections is that the February change was not done with an RfC. That objection could be made even if no one objected to the content or effect of the change. To object to some supposed defect in the process does not disqualify the change if the defect is resolved.

Blueboar: I did not say the closers are limited in what they can do. I'm saying that the discussion is inapplicable to any question but the one presented. Otherwise it's like passing a law without an official version, and everyone has their own private version of what it means. It's bait-and-switch, and it's hardly wikilawyering to object to such nonsense. ~ J. Johnson (JJ) (talk) 23:05, 19 May 2017 (UTC)[reply]

I'll let the closers decide how they will close. Nuff said. Blueboar (talk) 23:22, 19 May 2017 (UTC)[reply]

New proposal

A small group of editors — S Marshall, North8000, and I, TransporterMan — representing, I believe, most if not all sides of this discussion — have been working together to try to come up with a version of the section which satisfies all of our concerns. We can all support the following and here propose it to the larger group for consideration. Alanscottwalker also participated but neither supports nor rejects the proposal.

The entire current "Verifiability is not a reason for inclusion", including the title, would be replaced with the following:

"==Verifiability as eligibility=="

"Only verifiable information may be in Wikipedia. Even if information is verifiable, this does not mean it must be included or kept. Whether information that is verifiable will be included, excluded, or removed is determined by other policies and guidelines and consensus. The onus to achieve consensus for inclusion is on those seeking to include disputed content."

The current "see also's" and shortcuts would be retained

Offered for your consideration. Best regards, TransporterMan (TALK) 01:15, 20 May 2017 (UTC)[reply]

Support:

• S Marshall (talk · contribs), North8000 (talk · contribs), TransporterMan (talk · contribs) — As proposers.
• This is an improvement over the current wording. Sławomir Biały (talk) 12:20, 20 May 2017 (UTC)[reply]
• Do I think this suggestion is perfect? Nope. But I do think it is acceptable. More importantly, it will give us way to move forward while we continue to discus those small tweaks that might make it even better. In other words... it is good enough for me to support. Blueboar (talk) 14:12, 20 May 2017 (UTC)[reply]
• While concurring that the language can still be improved, I find it summarizes the degree to which our discussion yields agreement substantial (though not unanimity). I note valuable observations in the opposition arguments about improving process, pursuing a more definitive and/or useful consensus, but not, to my mind, objections that render the proposed wording inaccurate, unuseful or lacking in apparent consensus among current discussants. I dissent from tossing out the fruit of this thoughtful, wide-ranging effort toward resolution. I'm fine with saying "This is what we substantially agree upon in this matter thus far." FactStraight (talk) 18:43, 26 May 2017 (UTC)[reply]

Neutral:

• Alanscottwalker (talk · contribs)
• Doniago (talk · contribs) - I wanted to be a Support, but I'm not a huge fan of "eligibility" in the header as-is...I'd personally prefer "eligibility for inclusion". I also think "should" would be a better word-choice than "will" in the third sentence. Also in the last sentence, I think "burden" might be slightly more readable than "onus". Overall, I'm alright with this, but I just think there's too much room for improvement for me to be comfortable listing myself under Support. Noble effort though!
  • We're stuck with that word, because WP:ONUS and WP:BURDEN are different sections. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)[reply]

Oppose

The wording is bad. "Eligibility" is the wrong term here; just because something is eligible doesn't mean it should be on Wikipedia. Anything may be on Wikipedia, the question is whether it should be, and the current wording does a much better job explaining that consensus etc determines what should or shouldn't be in an article on Wikipedia, as opposed to what may be or what is eligible.

Additionally, saddling this proposal to the discussion above is incredibly counter-productive, instead of waiting for its (non-)conclusion and starting fresh. Splitting off the discussion to a user talk page, not letting the original discussion come to a conclusion, tacking on a proposal-within-proposal... this all reeks of WP:GAMING. Bright☀ 04:28, 20 May 2017 (UTC)[reply]

It was not planned, just an opportunity that popped out of a 2 person discussion which I lurked into. North8000 (talk) 12:14, 20 May 2017 (UTC)[reply]

We don't need both a "Reject" and an "Oppose" section above, so I'm removing the reject section. I invite BrightR to bullet-point his oppose entry to keep the formatting consistent. Regards, TransporterMan (TALK) 16:14, 20 May 2017 (UTC)[reply]

• Oppose, but could support in a new RfC with a few stipulations:

1) This RfC closes and this new proposal is presented in a new RfC. The change needs wider support than is likely to happen in a new proposal within this very long RfC. Also, it’s been a long hard road for the major contributors to the discussion, and this can lead to a strong desire to reach any resolution, even a compromise that is worse than either of the two current versions – fresh eyes are needed for a sanity check.

2) The proposal includes a corresponding change to the lede. The February changes to the section header and the lede were the major problems with the changes (for me anyway).

3) Both the pre-February version and the new changes are included as options in the next RfC. Both have had considerable support in this RfC, and it wouldn’t be right to throw one or the other of them out of consideration.

I see this proposal as good groundwork for the next RfC, rather than something that can be considered as part of this RfC. Continuing the groundwork, I'd like to point out that there was a version of the lede that had widespread support among participants back in the February discussion. This reads: “However, being verifiable is not sufficient for including something.” (This is somewhat different than February change of “However, while verifiability is required for including something, it is not a reason for inclusion.”) For the section header, this could read “Verifiability is not sufficient for inclusion”. This might be a good option for the next RfC.--Wikimedes (talk) 18:53, 21 May 2017 (UTC)[reply]

That sounds like an excellent plan for achieving a no-consensus result. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)[reply]

• Oppose. The proposal has some merit, but lacks a critical element: a shared (common) understanding of what it means and implies. And it suffers by any connection with this current discussion. In that regard I concurr with Bright*'s assessment ("saddling this proposal to the discussion above is incredibly counter-productive ...") and Wikimedes' stipulation ("This RfC closes and this new proposal is presented in a new RfC.") This discussion has run for three months with nothing to show but how fundamentally conflicted we are on this policy; it is no basis for any kind of result.

It is an error to polish the language before agreeing on the policy and implied practices the language describes. We need to work on reconciling our different views on what should be included, the basis for including or deleting, etc. But prior to anything else we need to:

1) Close this RfC. Give it a decent burial to avoid spoiling any subsequent discussion.

2) Take a break. A rhetorical fire break as I suggested a month ago, so that when we come back to this topic (hopefully refreshed) we will be less likely to carry over the bad feelings and bad rhetoric that has poisoned this discussion.

Then we can (and should) proceed to deal with the questions of:

3) What should the underlying philosophy of inclusion be?

4) What kinds of situations or edits are we concerned about, and what kinds of edits should be allowed or disallowed?

5) What are the terms (language) by which we express the foregoing?

Only then will we have an adequate foundation and tools for working out the policy. ~ J. Johnson (JJ) (talk) 21:12, 21 May 2017 (UTC)[reply]

I wasn't going to comment on the procedural points, but the fact is that RFC's are open until they're closed. There's no semi-closed-waiting-for-formal-closing status. And until they are, in fact, closed they're open for further discussion and proposals. IMHO, the only reasonable conclusion on the current RFC prior to our new proposal was "no consensus". Whether you agree or disagree with that evaluation, our only intent and motive was to add another proposal which might actually result in a consensus. It's disappointing to see that folks are opposing it not on the basis of its merit or lack thereof but instead on the basis of some self-defined criteria about how RFC's are supposed to work (and, indeed, I would invite any closers to take into account the degree to which any opposes here are based on procedure/bureaucracy rather than merit/what's best for the encyclopedia). But this is a wiki and people can opine on whatever basis that they care to do so, including the color of the font in which I posted the proposal. Regards, TransporterMan (TALK) 22:22, 21 May 2017 (UTC)[reply]

I believe that there's a distinction drawn between "closed" (boxed or 'hatted', if you prefer; the point is that further comments are discouraged as pointless) and "had a formal summary statement posted". It is possible to "close" or stop a discussion without (or in advance of) a "statement" being posted. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)[reply]

I just want to comment on something J. Johnson says in his five point list above. He asks us to consider "What should the underlying philosophy of inclusion be?"... this goes way beyond the original intent of the section under discussion. The original "Verifiability is not a guarantee of inclusion" section (pre Feb.) was intended to be nothing more than a brief warning to editors, alerting them to the fact that there are other policies and guidelines that have an impact on inclusion/exclusion. It was never intended to be some sort of underlying philosophical statement (as if it were our WP:INCLUSION policy). And I would object to any attempt to turn it into one. Blueboar (talk) 13:59, 22 May 2017 (UTC)[reply]

Indeed. But some of the divisiveness in the this discussion arises from different takes on what should be included (such as everything verifiable having "a virtual guarantee of inclusion somewhere..."). Resolving matters such as that (at least as far as verifiability is concerned) could reduce conflict at a higher level. Likewise for reaching a mutual understanding of key terms and caveats.

That such philosophical considerations go "way beyond the original intent of the section under discussion" is hardly a credible objection, as the discussion here goes "way beyond" the original question ("Q: Was this a good idea?").

Note that the February change was not intended to be make any change in policy (or philosophy), it was intended as a clarification. What seems to have happened is that making a more specific statement revealed differences of interpretation of the underlying policy.

TM: "what's best for the encyclopedia" would be better organized discussions that don't plop all over the rhetorical landscape to little result. Which is why we have established procedures (such as RfCs) and rules (such as WP:V). To imply (falsely!) that anyone is insisting on procedure over merit is quite ironic, as Marshall's objection to the undeniable consensus reached in February was that it did not go through the bureaucratic procedure of an RfC.

What BrightR, Wikimedes, and I are for is starting fresh. What we oppose is not so much this "new proposal", as the context of basing it on the current RfC. Let this and any other proposals have their own RfC. But we need a real break, not just a speed bump in a crappy discussion that has burgeoned well beyond any chance of summarization. ~ J. Johnson (JJ) (talk) 21:35, 22 May 2017 (UTC)[reply]

Regarding the core items under debate, yours and Bright Round's ideas are the ones that I most agree with. But the topic of inclusion in general is determined by dozens of policies and guidelines plus editorial practices in general. I think that the more specific topic is that wp:ver is often misused on this topic. Some hints in the mis-summaries that were changed in February contributed to this. So IMHO the February change changed those from "does harm" to "really fixes the problem". IMO the new proposal went to the middle ground of just "does no harm"; I supported it as a compromise to settle it, seeing this as an intractable mess at the moment. Not pushing anything, just explaining my thoughts. Sincerely, North8000 (talk) 22:24, 22 May 2017 (UTC)[reply]

The underlying problem is that different editors here have different ideas – philosophies — of what should or should not be included, and how that is to be effected. Unless someone can find a way of how to actually accomodate different policies and practices of WP:V and inclusion, any "compromise" on the wording is only being ambiguous enough that each side thinks they have it their way, only to continue to butt heads in actual practice.

One way of dealing with this is to define the issue of WP:V so narrowly that it does not reach the matter of what should be included (etc.). (And this is the desire of some of us, to remove WP:V from inclusion/exclusion debates.) This requires careful and strict specification of the scope of the issue, which the OP here rejected.

Alternatively, we can't reconcile the essential differences – i.e., compromise — without dealing with where they come from. I am not saying we must deal with every policy and guideline that touches on inclusion. But as long as we are are not actually on the same page there will be conflict, so we have to get to the bottom of all the differences.

And all of that would be better done in a new RfC. ~ J. Johnson (JJ) (talk) 23:42, 23 May 2017 (UTC)[reply]

Comments: When a (the) break is over, or a new RFC is started, I would like to state that one word should not be negotiable as indicated in bold. As I read; "Only verifiable information may be in Wikipedia. Even if information is verifiable, this does not mean it must be included or kept. Whether information that is verifiable will be included, excluded, or removed is determined by other policies and guidelines and consensus. The onus to achieve consensus for inclusion is on those seeking to include disputed content.". It does not matter if an agreement is made between editors, or with some admin ruling, consensus is the actual determination of inclusion or exclusion. Exclusion can be the result regardless of the source. All the rhetoric about well sourced content belonging "somewhere" is irrelevant. In this such case, as the changing of policies or guidelines, according to current consensus that includes BRD, as soon as any changes were contested the content could have been reverted. But! that is another story. Otr500 (talk) 01:12, 28 May 2017 (UTC)[reply]

BRD is not policy, and it begins with warnings that it should not always been used. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)[reply]

I have added sources to verify appearances of actors/actresses, like David Pevsner and Herb Mitchell (actor). The issue of verifying an actor's appearance on any television show or film has been raised, especially in Wikipedia talk:WikiProject Film. Recently, the filmography of the late Erin Moran has been discussed, while it's been removed but then reinserted. Thoughts about the filmography issue? (Frankly, filmographies had not been discussed in this project talk page [WT:V]... until now.) --George Ho (talk) 20:20, 24 April 2017 (UTC)[reply]

Standard (uncontroversial) entries in the filmographies don't need to be sourced explicitly (strictly speaking each entry is primary on its own and usually there reference sites like IMDB are given in the article which can be used for verification as well). However controversial/disputed (uncredited) appearances may require (additional) explicit sourcing.--Kmhkmh (talk) 12:54, 2 May 2017 (UTC)[reply]

With regard to the linked examples. I'd say the explicit sourcing in Presner and Mitchel is ok, but not strictly required, i.e. optional and the temporary removal at Moran was inappropriate.--Kmhkmh (talk) 13:03, 2 May 2017 (UTC)[reply]

My feeling is that uncredited appearances should be explicitly sourced or tagged/removed. (Presumably) credited appearances don't require sourcing, as the credits themselves should suffice, but if an editor challenges the (presumably) credited appearance then it's probably best to provide a source rather than debate the need for one. If an editor is consistently making such challenges it may be a matter of editor conduct. DonIago (talk) 13:00, 2 May 2017 (UTC)[reply]

Source it, and use a good source, not IMDb that is partly user-generated, but something with a semblance of proper editorial fact-checking. Wikipedia is as good as its sources. Bright☀ 18:45, 2 May 2017 (UTC)[reply]

Sorry but that seems neither consensus nor current practice and on top of it it is often not feasible. "Good" sources as in high quality (scholarly) sources simply don"t exist for most actors/actresses. Aside from current practice and practicability aspects WP doesn't really require every uncontroversial simple factoid in an article to be sourced.---Kmhkmh (talk) 20:21, 2 May 2017 (UTC)[reply]

Just because you don't have to doesn't mean you shouldn't (damn that's a lot of negatives). I've fixed plenty of articles with "uncontroversial simple factoids" that were factoids in the original sense—not true but commonly held. These factoids are as good as their sources... Bright☀ 20:58, 2 May 2017 (UTC)[reply]

As I pointed out above as long as we're talking credited roles, then every listed movie in the filmography is already a primary source for its own entry. And from my perspective this is a case of "you can doesn't mean you should" rather than " you don't have doesn't you shouldn't". From my perspective such (redundant) overcitations are not good encyclopedic writing and hence not something we should recommend. I wouldn't keep any editor from sourcing every single entry in a filmography in his article if he insists on it, but we certainly should not create the impression here that such sourcing is a recommended practice (by consensus) and I understood your first posting as the latter.--Kmhkmh (talk) 21:37, 2 May 2017 (UTC)[reply]

These "sourced-to-themselves" entries have a tendency not to check out. Bright☀ 11:13, 8 May 2017 (UTC)[reply]

I seriously doubt that a high percentage of filmography entries are incorrect, but of course mistakes (or intentional "resume" inflation) are going to happen sometimes. There's also nothing to stop someone from adding to an article a secondary source about a film that doesn't actually verify what it was supposed to. The presence of a ref in an article just tells a reader where they can verify a statement in an article. It's still up to the reader to consult that source to see if the editor who added it was correct in relying on it. So primary sources such as films are no different in that. postdlf (talk) 12:30, 8 May 2017 (UTC)[reply]

Exactly, even having a source listed doesn't guarantee the information is actually from the source. I constantly flag and remove information that's {{failed verification}}. But it's all the easier not to provide a source than to provide a bogus source, and providing a source brings the reader one step closer to verifying the information. That's why if you can source something you should source it, even if it's something as "obvious" as every single movie an actor has ever been in... Bright☀ 14:35, 8 May 2017 (UTC)[reply]

I agree that only uncredited appearances need separate/secondary sources. Otherwise the inline list entry of the film or television episode is itself the source for the credited appearance. There's nothing magical about separately formatted footnotes. postdlf (talk) 21:49, 2 May 2017 (UTC)[reply]

You could use {{cite film}} or {{cite episode}}, but in the absence of a dispute, it's probably pointless: Joe Film appeared in Joe's Film.<ref>{{cite film|title=Joe's Film|at=Credits}}</ref> does not seem very helpful. WhatamIdoing (talk) 05:50, 3 May 2017 (UTC)[reply]

The filmography of Mary Tyler Moore, which has sourcing issues, was split into Mary Tyler Moore filmography and awards. I discussed this at Talk:Mary Tyler Moore#Mary Tyler Moore filmography and awards by saying that such splitting is subject to WP:CFORK, yet the editor said that there was clutter with tables. I have wondered whether the splitting was appropriate. I can't ping the editor yet, but I may do so later. George Ho (talk) 15:23, 3 May 2017 (UTC)[reply]

I very strongly doubt that the English Wikipedia has ever seen an actual POVFORK from WP:SPLITting off the long list of awards that a famous actor has won over the years, much less from splitting off a complete filmography.

Additionally, POVFORK (and other [i.e., the many acceptable] forms of content forking) isn't really a verifiability issue, and citing the sources for such a list does not determine whether the split was a good idea. If you actually suspect a POVFORK, then you need to be thinking about WP:POV problems, not about WP:V problems. WhatamIdoing (talk) 06:52, 5 May 2017 (UTC)[reply]

Just wondering... How would this discussion affect Featured Lists? "Lauren Bacall on screen and stage" and "Ethel Barrymore on stage, screen and radio" are Featured Lists, and the appearances are cited with secondary sources. More at Category:Actress filmographies and Category:Male actor filmographies, both at Category:Actor filmographies. BTW, I found Wikipedia:WikiProject Good lists created in December 2016, the same period where lists were discussed. The Good List criteria are not yet established, however. --George Ho (talk) 07:29, 14 May 2017 (UTC)[reply]

(Background.) When we want to verify information that we've derived from more than one page of a printed book we simply provide the page numbers. That might be a page range, but it might also be two or more pages that are not nearby each other.

Non-pdf e-books don't have pages. Most devices used by readers will break the book up into the equivalent of a page view, but when viewed as between different devices, an entirely different number of "pages" may be shown.

Still, so long as the information comes from one location in an e-book, it seems to me providing the URL that takes the person to that exact part of the source provides transparent verification, right? But what about when we have the equivalent of the need to cite two or more (discontinuous) "pages" from an e-book for a single citation; how do we provide a transparent citation that targets the two locations in it?

The only way I can see to do this is to provide two separate URLs in the cite (which AFAIK, cannot be done using our citation templates—can it?), or, really "clunkily", cite the same source twice in a row, with identical details but for the URLs targeting the two separate locations in it. Has this been discussed and is there any solutions or recommendations? Our need to cite e-books is only going to increase as time goes by.--Fuhghettaboutit (talk) 22:22, 26 April 2017 (UTC)[reply]

• I think best practice would be to include both links in the same reference, but how to go about doing that within our citation templates is an interesting question. @Headbomb: may be interested and is usually helpful with this kind of thing.—S Marshall T/C 23:30, 26 April 2017 (UTC)[reply]

Use {{rp}}.

The sun is a star.^[1]: sun The moon is not a star.^[1]: moon

Although I would not use URLs at all, and simply give the name of the chapter in the {{cite}} template to avoid all of this.

The sun is a star.^{[1]: chapter 1} The moon is not a star.^{[1]: chapter 8}

Bright☀ 08:50, 27 April 2017 (UTC)[reply]

Aaaarrrgghhf! No, no, not {{rp}}; it's ugly! it's too horrendous to be tolerated!!

Folks, look, the seeming difficulty here is more conceptual than real. And is readily handled. But there needs to be clarification of some concepts.

First: URLs are not citations. You can't just point somewhere and say: that's it. Proper citations (here, as well as well-established practice elsewhere, print and non-print) should give enough bibliographic detail to describe and identify the source – including such extremely important parameters as who wrote it, and when – along with any info to help locate it. A reader/editor should not have to follow a url, and then poke around a bit, to determine whether the author is a world famous expert, or an unknown clown.

Second: all of that info – the "full bibliographic details" – applies to the source (or "work") as a whole, and need be, and indeed, should be, cited only once per source, per article. And that constitutes the full citation.

Now your question really has two parts: 1) how to cite a source (any kind) more than once in an article (i.e., "re-use" it)? And: 2) how to cite a specific location in an e-book?

To cite a source more than once, many editors use "named refs", but that is quite unsatisfactory, and forces use of such crudities as "rp". A much better practice is the use of short citations (short cites), such as "Smith, 2001", that refer back to the full citation. These can be done as straight text, or with various templates, such as harvnb. And you can have as many as you need, where you need them, and each one can have a specific "in-source" attribution (page number, or whatever).

As to the second question, I think you're there already: specific URLs. Just do something like "Smith, 2001, <specific url>", with whatever information you think would be helpful. Not hard at all. ~ J. Johnson (JJ) (talk) 20:35, 27 April 2017 (UTC)[reply]

Short citations are way worse than {{rp}}, and are rarely used on Wikipedia. Bright☀ 13:41, 28 April 2017 (UTC)[reply]

How are short cites "way worse"? They are pretty much standard in much academic publishing, losing out in some scientific and medical journals only where for reasons of space they use a bastardized form of citation. On the other hand, {rp} is peculiar to WP, unfamiliar, and looks ugly as sin where it intrudes into the text (being exceedingly non-transparent).

"Rarely used" is debatable. Over 26,000 articles use {harvnb} (which is only one way doing short cites), while use of {rp} is just short of 24,000. That most editors put their citations in "named-refs" (which then forces use of {rp}) is hardly an informed choice: most editors, when they start using citations, look around, and, like puppies, do it the way they see "everyone else" doing it. And having, with great difficulty, learned how to do citation one way, they don't want to get off that horse. ~ J. Johnson (JJ) (talk) 22:50, 28 April 2017 (UTC)[reply]

There are over 5 million articles on the English Wikipedia, 0.5% is pretty rare. But let's not argue; I encourage you to add or fix citations on articles with short citation style. Compare <ref name="author-year" />{{Rp|pages}} with <ref>{{Harvnb|author1|author2|year|pages}}</ref>. Look at all the redundant info, and if you add or change the name of one of the authors, or if there's no author name, or if you change another minor detail in the ref, the short reference breaks, so maintenance is needlessly complex. With named references, you change the reference once and nothing breaks. And of course, the reference section itself is much more concise with long refs than short refs.

In bullet points:

• No duplication in the reference section (short refs require two separate ref sections)
• Only one click required to get to the complete reference (short refs require two)
• Direct links to and from a reference and each of its usages (short refs cannot be linked back)
• Updating requires editing only a single place (short refs require updating all places where the ref occurs)

Bright☀ 19:02, 2 May 2017 (UTC)[reply]

• Let's not wander off topic... the question was not about citation styles.. but how to precisely cite e-books (which don't have page numbers), which is problematic no matter what style one uses. Per SAYWHEREYOUGOTIT... if you got it from a kindle or some similar device, that is what you should cite. There is no URL to cite. So the best you can do is cite chapter ( and perhaps paragraph number. Blueboar (talk) 23:38, 28 April 2017 (UTC)[reply]

It's not entirely off-topic as, given some form of identifying or linking to specific in-source locations, there is the consideration of how it should be displayed. Where Fuhghettaboutit mentions URLs I am assuming he has in mind some kind of source accessible on the Internet.

As to not having page numbers: wow. That's a step backwards, page numbering being one of the principal improvements made in the development of books as we now know them. Without page numbering table of contents are just minimal outlines, and indexing is meaningless. If e-books have no such equivalent (perhaps paragraph numbering?), then the answer to the question presented is: No, one cannot "cite ... one part of an e-book", transparently or otherwise. ~ J. Johnson (JJ) (talk) 21:26, 29 April 2017 (UTC)[reply]

Most e-readers do not display page or paragraph numbers... the best you can do is manually count paragraphs from the start of the chapter. So to form a precise citation you might have to create something like:

• "Author, Book Title, Reliable publishing co., 2016, (Kindle edition) Chapter 3, Paragraph 527."

It might not be a standard citation format, but it would at least allow others to find the information in the book and verify that it supports what we say in our article. Blueboar (talk) 22:16, 29 April 2017 (UTC)[reply]

There is no problem citing paragraphs, but there could be problems in counting them. (E.g., is your comment just above one paragraph? Two? Or three?) Perhaps we could start a campaign to persuade publishers of such content that if they don't include paragraph numbers (like some journals do) we won't allow them to be cited. ~ J. Johnson (JJ) (talk) 23:35, 29 April 2017 (UTC)[reply]

You can also provide short quotations, using |quote= if you happen to be using citation templates. Most e-readers permit searching, so this is an easy way to find the correct part of the source. You can also use the |at= to record free-form locations (e.g., |at=dust jacket or |at=Near "The Moon, like a flower...". WhatamIdoing (talk) 06:01, 3 May 2017 (UTC)[reply]

Opinions are needed on the following: Wikipedia talk:Citation overkill#Citations. A permalink for it is here. Flyer22 Reborn (talk) 06:20, 9 May 2017 (UTC)[reply]

@TransporterMan: You said "let's at least not misstate the policy". I agree with the sentiment, but adding that "information that is not cited... may be removed" does not match the policy. WP:BURDEN applies to information that is not verifiable, but for verifiable information, it doesn't require that the reference is in the page, only that it is known. Also if the information is not challenged nor likely to be challenged, there's no need for an inline citation. Diego (talk) 14:34, 16 May 2017 (UTC)[reply]

Not correct. As established here at this talk page again and again, uncited information may merely be removed. If there is any additional requirement on that it is only that the removing party state a "concern" that it may not be verifiable (and even that is doubtful as a requirement due to the "should" in that sentence). There is no burden on the removing editor to establish, or even attempt to establish, that the information is, in fact, unverifiable. Best regards, TransporterMan (TALK) 16:10, 16 May 2017 (UTC) Oh, and PS, the "likely to be challenged" standard only applies to the introduction of new material. The standard for removal of material after it has been added is in BURDEN and says, "Any material lacking a reliable source directly supporting it may be removed and should not be restored without an inline citation to a reliable source." — TransporterMan (TALK) 16:22, 16 May 2017 (UTC)[reply]

You forget that WP:PRESERVE is also policy, forbidding the exclusion of content that belongs in the article. The idea that any content without a direct citation may be removed for any reason, at any time, has been defended at this talk page, but it has not been encoded as such in WP:BURDEN, and is contrary to the editing policy. The "any material may be removed" that you quote has lots of qualifications and caveats in this and other policies. Diego (talk) 18:39, 16 May 2017 (UTC)[reply]

And the next question is: Does the material "belong in the article" (or not)? Nine times out of ten, that's really what is underlying most debates about a removal (and opposition to the removal)... a disagreement as to whether the unsourced material should be in the article (or not). Blueboar (talk) 19:16, 16 May 2017 (UTC)[reply]

Right, but BURDEN may only be invoked under a genuine belief that the content can't be supported by reliable sources. Deleting content merely as a way to force others to source it, without oneself checking whether it's easy to verify, has been deemed disruptive by previous consensus. Diego (talk) 21:48, 16 May 2017 (UTC)[reply]

That's absolutely false and has been refuted here again and again. Point to that consensus if you can. Moreover, PRESERVE is expressly conditioned upon V being satisfied before PRESERVE kicks in: "as long as any of the facts or ideas added to an article would belong in the 'finished' article, they should be retained if they meet the three article content retention policies: Neutral point of view (which does not mean no point of view), Verifiability and No original research" and "Several of our core policies discuss situations when it might be more appropriate to remove information from an article rather than preserve it. Wikipedia:Verifiability discusses handling unsourced and contentious material" (emphasis added). (And that begs the question of whether PRESERVE, though designated a policy, really requires anything at all.) Best regards, TransporterMan (TALK) 22:48, 16 May 2017 (UTC)[reply]

Once upon a time, the policy used to say "However, this policy should not be used to cause disruption by removing material for which reliable sources could easily be found — except in the case of contentious material about living persons, which must be removed immediately." I still feel that this should be true to some extent. But, just so we're clear: One should not go around deleting relevant material that meets the other guidelines if sources can easily be found, should one? Sławomir Biały (talk) 23:31, 16 May 2017 (UTC)[reply]

That's already provided for in the policy: The burden for seeking out sources, whether easy or difficult to find, should be on the editor who initially adds the material or who restores it after it has been deleted, that is, the persons asserting that the material should be in the article. The burden should not be on the person who removes unsourced material. (Just as it should not be on a person who challenges unsourced material without removing it.) Per the policy, the only burden on the remover or challenger should be (if any, see above) that they have and state a concern that the material is unverifiable. Best regards, TransporterMan (TALK) 04:05, 17 May 2017 (UTC)[reply]

Thanks, Blueboar. Your new version is much closer to the wording at WP:CHALLENGE. Diego (talk) 14:52, 16 May 2017 (UTC)[reply]

I think that the February 2017 wording was better, but am ready to compromise to really settle it. But not open to creating a shifted starting point for future discussions. So I'm not sure whether to revert or not. North8000 (talk) 15:38, 16 May 2017 (UTC)[reply]

We could rather revert to the pre-february 2017 version as a starting point. It was a long-standing stable version, after all. Diego (talk) 15:51, 16 May 2017 (UTC)[reply]

According to the RfC there does not seem to be a consensus to change from pre-february 2017 version, so why have the changes not been reverted back to that version? -- PBS (talk) 15:59, 16 May 2017 (UTC)[reply]

FYI... No problems on my end with a revert to the pre Feb version. Blueboar (talk) 16:09, 16 May 2017 (UTC)[reply]

I'm also fine with the pre-Feb version, but I'm not sure that there's any real consensus for anything in the foregoing discussion, at least not a consensus of the quality needed to change this core policy. Best regards, TransporterMan (TALK) 16:11, 16 May 2017 (UTC) And PS, I've reverted to the post-Feb version as the last at least semi-stable version. That does not mean that I support that version since I actually support the pre-Feb version, but if we're going to continue to draft on this it ought to be done here on the talk page, not in the policy. Best regards, TransporterMan (TALK) 16:40, 16 May 2017 (UTC) And PPS since there's a continued discussion about closing, above, I think I need to make my position clear about my preferences. My preferences are, in order from most preferred to less preferred, (a) The February text with the pre-Feb heading (I think I'm the only one asserting that combination), (b) The pre-Feb text and heading, (c) The Feb text and heading, and (d) any of the post-Feb substitutes which have so far been proposed (here or by editing the article). I find any of (a) through (c) acceptable in the order of preference just stated, but I do not find any of the (d) versions acceptable and object to the adoption of any of them. Best regards, TransporterMan (TALK) 04:14, 17 May 2017 (UTC)[reply]

The Feb version was arrived at in a 27,000+ word discussion that ran for 23 days (Feb 7th - March 2nd) with no objections at the end of the process. And it merely summarized the actual policy. I'm ready for compromise be we need to recognize that. North8000 (talk) 18:38, 16 May 2017 (UTC)[reply]

I think that we are adding a layer to the core difference of opinion that unnecessarily moves it from challenging to hopelessly mired. The core difference of of opinion, is, beyond / stipulating meeting verifiability being a requirement for inclusion, to what extent is or isn't it a force or reason for inclusion? The extra layer that is miring it down is that the discussed wording wades into a bunch of other areas and procedures that are not even in dispute. For example , rules for removing material. North8000 (talk) 23:54, 16 May 2017 (UTC)[reply]

• Comment -- I strongly disagree with the statement that WP:BURDEN applies to information that is not verifiable. This is very vague -- who determines whether information is "verifiable"? Similar to TransporterMan, I don't believe that consensus for such an interpretation exists. K.e.coffman (talk) 02:18, 17 May 2017 (UTC)[reply]
  • Perhaps it will make more sense like this:
    1. If you personally know that a given piece of information is verifiable, then you personally should not remove it solely on the grounds that nobody has yet typed a citation after that information.
       You could add sources, you could tag it, you could move it to another article, you could copyedit it, you could do any of the other suggestions at WP:PRESERVE, you could even remove it on the grounds that it's WP:TRIVIA or WP:POVish or WP:SYNTH, but you personally should not remove information that you know is verifiable with a claim that you think it is not possible to verify it, when you have actual, positive knowledge that it could be verified. Put more simply, if you cannot comply with BURDEN's request to "please state your concern that it may not be possible to find a published reliable source for the content" without telling lies, then you should not remove that material under BURDEN.
    2. However: if you personally do not already know that the material is verifiable, then, yes: You may remove that material. There are no other requirements: if it is uncited, and if you personally do not already know that the material is verifiable, then you are entitled to blank it. I would only add to this that you please remember that it is both uncollegial and anti-policy to take actions that you personally believe will have a (net) negative effect on an article.
  • And: Remember that every editor knows different things. So while I couldn't ethically claim BURDEN for, say, the recent news about genetic treatment for sickle cell disease, most editors probably could. "I can't claim BURDEN" is different from "Nobody can claim BURDEN". WhatamIdoing (talk) 16:37, 17 May 2017 (UTC)[reply]
    • Re: "If you personally know..." -- How would one know what's in another editor's head? The current interpretation is cleaner, IMO. K.e.coffman (talk) 18:08, 17 May 2017 (UTC)[reply]
      • You don't need to know what's in another editor's head. You only need to know what's in your own head. I would be acting wrongly if I were to remove uncited material that I personally know to be verifiable, and then claim that I am doubtful that it is possible to verify this. The policy does not authorize me to tell deliberate lies about the potential verifiability of a statement, even if I feel certain that I won't get caught. Another editor, looking at the same sentence, but without (for example) the benefit of having read the same news article yesterday, could invoke BURDEN quite honestly for the same material. As Diego succinctly put it, "BURDEN may only be invoked under a genuine belief that the content can't be supported by reliable sources". Your genuine belief in the material's non-verifiability may be genuinely and honestly mistaken, and your genuine belief is allowed to be based on nothing more than a wild guess, but you must not blank uncited material under BURDEN when you have actual, positive knowledge that the material is verifiable. So the policy won't support me blanking uncited material about the genetic treatment of sickle cell that was reported this week, but (perhaps) you could; on the other hand, I could lazily challenge the verifiability of any number of uncited sentences in WWII-related articles, and you, by virtue of your greater knowledge about WWII, would be restricted from making the same claims. WhatamIdoing (talk) 05:33, 19 May 2017 (UTC)[reply]

One of the most difficult things for editors who are knowledgeable about a subject to accept is that their knowledge can be challenged by editors who are not knowledgeable. Yet this is true... unsourced information can be challenged (and removed) on the mere belief that it is unverifiable. It is up to those who know the information is verifiable to provide citations when citations are requested. Blueboar (talk) 13:13, 21 May 2017 (UTC)[reply]

I think that a main point of WhatamIdoing's post is that you should not do a wp:ver-based removal unless you have a sincere question about its verifiability. I wish that that point were more prominently enshrined in policy. North8000 (talk) 13:38, 21 May 2017 (UTC)[reply]

Yes... but but she goes on to say that a sincere question may be based on a good faith belief of unverifiability. That is an important point for those with knowledge of verifiability to remember. Others don't necessarily know what you know, so it is up to you to supply the sources when sources are requested. Blueboar (talk) 14:01, 21 May 2017 (UTC)[reply]