Wikipedia talk:Notability/RFC:compromise

uselessness of the RfC

Looking over the course of the discussion, I see no consensus at all on anything. I see no single proposition which has general consent. I see enough diversity of opinions to indicate that perhaps the entire concept is useless altogether. What we want to have an article on, we find soemtreason for calling ntoable, but that doesn't advance things any. The question is what we want to have an article on. DGG (talk) 02:57, 18 October 2008 (UTC)

There's plenty of points that are agreed upon; spinouts have to show notability, there's a likelihood we can develop non-notable list allowances, and that we have SNGs that cannot override the GNG but can define more appropriate sources for certain areas and presumptions of sources for notability. --MASEM 03:49, 18 October 2008 (UTC)

It's not a ballot. We're not here to vote up individual propositions, but to generate discussion around a few concrete ideas instead of talking abstractly. As Masem has pointed out, there's a lot of agreement here. Randomran (talk) 08:33, 18 October 2008 (UTC)

I'm starting to get the impression that change based on consensus - where consensus is read as near-universal agreement - isn't going to happen on any guideline on Wikipedia. But I suppose the aim here is to look for strong, overriding, support instead. And while none of the proposals received this, there was enough to suggest that it is possible. I'm almost done analyzing A4, but what was common to most of the proposals was a "I support, however ..." and "I oppose, but ...". Perhaps if a proposal could be developed based on the "buts" and "howevers", it might gain something akin to consensus. :)- Bilby (talk) 16:39, 18 October 2008 (UTC)

I think you're onto it. Even between the "I wouldn't support this proposal as it is" kinds of comments, there is enough qualified support and qualified opposition to indicate that some of these proposals are close. Remember: a consensus isn't unanimity, nor is it what a majority of people want. It's what most people agree to and understand, even if it's not their first choice. Randomran (talk) 16:20, 18 October 2008 (UTC)

Canvasing

Just a heads up that I've spotted some canvasing here. Not so sure that it makes any difference.  :) HiDrNick! 20:06, 21 October 2008 (UTC)

• Slashdotter1: Please strike a blow AGAINST deletionism.
  • Slashdotter2: I prefer to think of it as Intelligent Editing.

This message approved by: VG ☎ 20:23, 21 October 2008 (UTC)

Closing

As it's been almost a week since Randomran's request to close the RFC, and no one's objected, I'll be going through and closing the discussions. It might take me a day or so to post the results, though, since I'm pretty busy (and sick) at the moment. Cheers!--Aervanath lives in the Orphanage 03:52, 23 October 2008 (UTC)

Thanks a lot for helping out and closing the discussion. Randomran (talk) 13:23, 23 October 2008 (UTC)

Results

Results table

Proposal	Supported	Opposed	Neutral	Support %	Result	Proposal was
A.1	61	130	18	29%	Failed	Every spin-out is notable
A.1.2	75	69	6	50%	No consensus	Spin-out articles are treated as sections of a larger work
A.2	82	59	2	57%	Majority, but no consensus	Every spin-out must prove notability
A.3	51	49	8	47%	No consensus	SNGs can define that some spin-outs are notable
A.4	51	35	6	55%	Majority, but no consensus	Lists may be exempted from the GNG
B.1	26	65	6	27%	Failed	Articles must meet the GNG and SNGs
B.2	66	17	3	76%	Passed	SNGs can outline sources that assert notability
B.3	23	31	19	31%	Failed	SNGs can define when sources probably exist
B.4	14	63	6	17%	Failed	SNGs are not needed
B.5	14	54	5	19%	Failed	SNGs override GNG
B.6	40	22	9	56%	Majority, but no consensus	SNG criteria support reasonable presumptions of notability
B.7	5	1	2	62%	Not enough input	SNGs (only) provide subject area interpretation of the GNG

My general interpretation of the results:

1. Consensus seems to be pretty clear on a few points:

1. WP:Subject specific Notability Guidelines are necessary
2. SNGs can never override Wikipedia:Notability
3. SNGs can describe what kinds of sources are useful for proving notability

1. Consensus can probably be achieved on:

1. What kind of notability criteria are needed for articles which are spun out from a larger article. There was little support for allowing any spin-out to be notable, but the opposite proposal, requiring that all spin-outs require independent notability, didn't achieve consensus either. It is likely that proposal A.2 could achieve consensus with some tweaks to the wording to address some of the objections given.
2. What kind of lists can be exempted from the GNG. This is also something that could probably achieve consensus with some careful rewording.
3. Whether SNGs can create a reasonable presumption of notability. There were 9 neutral !votes on this proposal (B.6), so if those could be convinced to move to support, this would probably reach consensus.

Y'all play nice, now.--Aervanath lives in the Orphanage 06:03, 23 October 2008 (UTC)

This is a decent summary, but I think the reality is more nuanced than that. Reading the comments, there's not just support/oppose, but conditional support, and qualified opposition. Not to mention that more than 200 commenters weighed in on issue A in A.1, but less than 100 in A.4 -- we ought to read between the proposals to find Wikipedians' real position on issue A, and B for that matter. Nonetheless, thanks again for helping to close the discussion. I think the next step is selecting a detached third-party party who has the experience to read and close a lengthy discussion like this. I've been busy, but we can discuss this over the next week. I know this is moving slower than some people would like, but it's a difficult issue. Randomran (talk) 13:26, 23 October 2008 (UTC)

Sorry, I'd meant to put in a disclaimer to the effect that this needed to be analyzed in more depth before drawing any thorough conclusions. I don't have lots of time to really get in deep at the moment. I think it'll be a team effort, though. Cheers!--Aervanath lives in the Orphanage 14:12, 23 October 2008 (UTC)

No problem. It's useful just to have a brief by-the-numbers summary. I think it reveals where there's common ground, and where there isn't. Now we just have to understand what that common ground is, because it's never exactly the proposal that was put up. Randomran (talk) 22:29, 23 October 2008 (UTC)

My thoughts exactly.--Aervanath lives in the Orphanage 01:34, 24 October 2008 (UTC)

Ok, here's an attempted remix of A.1.2 and A.2;

Spin-outs must show some evidence of notability to merit a separate article.

Proposal: The notability guideline applies to every topic, but some notable topics are too broad to cover effectively in a single article. Sub-articles are permissible when one or more reliable sources, independent of the subject, provides significant coverage of the sub-topic, and the sub-topic is addressed at a general level in the (notable) parent article.

Rationale: We want comprehensive coverage, but there is no consensus to allow articles that rely exclusively on primary sources to support that coverage (which practice is already frowned upon by our no original research policy). A "one minimum" rule is less restrictive than WP:N, which prefers "multiple", and to some extent it allows the spinout to be treated as part of the main article, but is not carte blanche to create new articles with no secondary sourcing. Requiring the sub-topic to be addressed in the main article prevents any non-notable article from being eligible for spin-out status. A more lenient version of this might be adapted for lists. Fletcher (talk) 18:09, 25 October 2008 (UTC)

On the surface, that seems reasonable. I think we might encounter some resistance though, at least until we can agree on what the RFC itself means. It also means looking at the big picture, including a relaxed notability standard for lists, and relaxed sourcing requirements in SNGs. But I think this seems like a good general rule. I'm working on finding a neutral and detached third-party to analyze the RFC right now. Randomran (talk) 18:35, 25 October 2008 (UTC)

• I see one problem, in that a relaxation of the WP:N that Fletcher is proposing would allow the creation of content forks on topics that are not notable. Because WP:N is the only way to determine whether an aricle is a content fork or not by reference to the fact that reliable secondary sources have written about a particular topic, there is no other mechnanism to identify which sub-topic is a fork and which is not. The example I have cited before is the forks that have been created from The Terminator: Terminator (character concept) and Terminator (character) are both about the same topic, so how do you determine which is the content fork, and which is not if all sub-articles are exempt from WP:N? There may be an answer, but no one has come up with a working proposal as yet. The problem I see with analyising the RFC is to work out what are the feasible proposals that could work, not just what is the the most elegant or popular proposal. --Gavin Collins (talk) 11:41, 28 October 2008 (UTC)

• I see that as a separate issue: you can produce a content fork by utilizing [[WP:N] as it stands now. Some are legitimate, some aren't. I don't see Fletcher relaxing the requirement to the point of having no sources, one source might be enough (in the context of multiple sources for the broader topic). Again, not to say it's a complete proposal, but I wish people would try to build off of each other instead of shooting everything down. Maybe that can't happen until we have an independent party summarize the RFC (which is being undertaken right now, and will take a little while). Randomran (talk) 17:28, 28 October 2008 (UTC)

• Agreed; and I paired the relaxed sourcing requirement with the need for the main topic article to address the spinout topic. The theory is that the main article will summarize the most important elements of the topic, but not trivial stuff (or trivial stuff will tend to get pruned). For this to work, the main article should itself meet WP:N, so we know other sources concur with what we claim are the main elements of that topic. Fletcher (talk) 22:30, 29 October 2008 (UTC)

• In answer to Randomran, I think you meant to say that you can create an article about a notable topic that is a POV fork (by ignoring some sources and emphasising others to create a bias), but in my experience the a content fork is ususally an article about a topic with unproven notability; in both cases there is an over-arching topic of proven notability.
  In answer to both Randomran and Fletcher, it is not my intention to be one of those people that "shoots down" every proposal. However, I still see a fundamental problem: for every notable topic on Wikipedia, it is possible to create many many content forks that fail WP:N.
  For example, whilst Terminator (character concept) and Terminator (character) are potential forks, it is possible to create other (reasonable) forks using the same sources like, say Terminator (film series). The only way to be sure that one of these articles is not a content fork is to check for non-trivial real-world content obtained from reliable secondary sources that are genuinely about the specific itself. Following on from Randomran's earlier point about POV forks, you would also have to double-check that the content of the article is no biased in favour of one particular viewpoint (assuming there are many) in order to be sure the article is not a POV fork as well.
  Whilst I would not wish to discourage Fletcher from making suggestions, but there is a thorny problem which he needs to address in his proposals, namely articles that fail WP:N also tend to fail Wikipedia content policies as well. --Gavin Collins (talk) 11:02, 30 October 2008 (UTC)

• Please understand that consensus defines policy and guidelines, not the other way around; the entire point of the RFC was to see if WP:N and other guidelines needed to be changed to reflect what editors expect of SNGs and spinout articles. The RFC shows that there's allowance for some non-notable spinouts that must be defined, which (at least as you argue) cannot exist under WP:N as it reads now, but since there's consensus to make the change, we can make the change. We'll still need to make sure that change itself meets consensus but we can point to the RFC as a basis for why that change was added. --MASEM 12:59, 30 October 2008 (UTC)

• I see that "Every spin-out must prove notability" is the consensus; so your assertion that there's allowance for some non-notable spinouts does not hold any weight. Even if it did, you would have change all Wikipedia content policies to accomodate this change, as articles on topics that fail WP:N are also likely to fail WP:NOT or WP:CFORK. If you propose changes to guidelines, you have to ensure they are workable, regardless of how popular the changes are, otherwise you will be building on sand. --Gavin Collins (talk) 13:56, 30 October 2008 (UTC)

• 57% is not consensus, Gavin. And more to the point, 43% opposition is strong evidence against consensus. Phil Sandifer (talk) 14:39, 30 October 2008 (UTC)
• I don't read consensus the same way you do; if A.2 won consensus, A.1.2 should have been far below 50%, as these two proposals contradict each other. Did someone miscount? From my (admittedly brief) reading of the arguments and !vote counts I'd say Wikipedians are fairly evenly split, with somewhat expressed preferences here and there but no overwhelming consensus. Hence the need to find a pragmatic compromise that both sides can live with. My proposed compromise seems workable, given that it excludes articles with zero RSS, and promotes better cross-communication between main articles and spinout topics. You haven't proven that anything other than an inflexible reading of WP:N will promote articles that contradict our policies. Fletcher (talk) 15:16, 30 October 2008 (UTC)

• In answer to Phil's earlier point, whether there is consensus or just a majority who hold the view that "Every spin-out must prove notability" is a matter of opinion, and although it does not invalidate Masem's view that "there's allowance for some non-notable spinouts", there is no evidence that the RFC supports this. I am happy to consider any proposal on their own merit, but I would be disclined to put one forward myself if I had to misrepresent the facts in order to justify itself. --Gavin Collins (talk) 15:27, 30 October 2008 (UTC)

• • I would point out that such articles routinely survive AfD, which seems to settle the issue of allowance straightforwardly. Phil Sandifer (talk) 15:33, 30 October 2008 (UTC)

I am somewhat unconvinced of this proposal - I think it preferable to lower the bar on type of sources rather than move the line from two to one - especially given that the threshold for inclusion is presumption of notability, not proven notability. It's often possible to find one good source and persuade people that another is likely to exist. But showing presumption for one source seems like a hazier issue. Phil Sandifer (talk) 15:19, 30 October 2008 (UTC)

Cautionary note on the results

Folks shouldn't get too hung up on these results. It's a useful shorthand, but a closer analysis could reveal something completely new, if not just offer a more nuanced perspective.

1. These results simply count it as a vote, instead of looking at the actual comments and discussion. The actual comments often reveal more "conditional support" or "conditional opposition", which would sway things one way or another.
2. These results treat each proposal as its own separate discussion. That may ignore the larger consensus at the issue-level: only 90 people weighed in on A.4, but 200+ people weighed in on A.1 and may have had comments about issue A at large that we should take seriously.
3. These results simply look for consensus on a proposal. They don't look for consensus elsewhere, such as on a guiding principle. For example, some of the opposers might agree with the supporters that we should relax the standard, just that a specific proposal goes too far.

That's why there is going to be an analysis from people who are further away from this dispute. Someone who actually reads every comment, across the individual proposals, and gives us some guiding principles moving forward. I know things are moving slow, but unfortunately this is the only way something this contentious can move forward on Wikipedia. Randomran (talk) 15:07, 30 October 2008 (UTC)

Well, and as I've suggested, there's not going to be a decisive result over this. The major piece of information coming out of this is that we're going to have to work to forge a compromise. Which, it should be noted, there are some good efforts towards. Mainly I think declaring victory for a given proposal is unhelpful at this juncture. Phil Sandifer (talk) 15:11, 30 October 2008 (UTC)

There's some stuff that has been clearly rejected, I think it's safe to say that. And some proposals have a lot of support -- but that indicates we need to modify or build upon those proposals before they can really reflect a consensus. But I'm not sure anyone really trusts each other to interpret this fairly, so let's not get into that amongst ourselves. We can begin working on a compromise based on what we've learned. And when the analysis is done, we'll have an even better idea of how to move forward. Randomran (talk) 15:55, 30 October 2008 (UTC)

As the editor who hastily mashed together those results, I have to agree with Randomran. All they are is a simple vote count. While it is useful for noting that some proposals clearly passed or failed, it is not useful for determining consensus in depth. So please don't read too much into them. Thanks!--Aervanath lives in the Orphanage 02:54, 31 October 2008 (UTC)