Wikipedia talk:Requests for comment/Inclusion criteria for Lists

Timeline for RFC

Any plans to close this RFC once the discussion has run its course? We should also think about the next logical step. Shooterwalker (talk) 15:03, 24 August 2010 (UTC)

I think the "what next" is the more important issue. When we have that, then we can see whether the discussion has "run its course" and close. Blueboar (talk) 15:17, 24 August 2010 (UTC)

The idea I had in creating this is determine a language and a target for that language to be added to either existing p/g or a new one , whatever is best needed, to meet the consensus view of how lists are to be handled. When we have a reasonably good idea of what this language and target will be, we'll need to engage that target's talk page as well, such that we close this RFC and then add the language appropriately. We should be aware that if there's no obvious consensus, we need to then close this before it gets too far down the drain. --MASEM (t) 15:48, 24 August 2010 (UTC)

If I were uninvolved

If I were summarizing and closing this discussion as an uninvolved Admin (which I am not) I would do so as follows.

Summary of consensus on the following points: (This summary is generalized, the points are not intended to be proposed policy or guideline statements and there is an understanding that a great many scenarios and permutations were contained in the discussion that must be taken into account when actual policy is written. The summary points also include a brief assessment as to where policy change might be required to implement them.)

• Any notability burden put on a list should be placed on the List Topic (as described in the lead with discriminate inclusion criteria) not the List Title with notability determined by coverage of the topic in reliable sources. WP:GNG and [[WP:V--Mike Cline (talk) 20:09, 24 August 2010 (UTC)]].
  • State of consensus: ~75%+ of discussion participants support this general position
  • Policy implication: WP:SAL could be modified to strengthen and explain policy relative to this position.
• As a point of policy, lists are complementary to categories and the coverage of any given topic by both a list and category is permitted. Lists and categories on the same topic may indeed by duplicative and one or the other may not be appropriate for the encyclopedia, but duplicative lists and categories are dealt with on a case by case basis.)
  • State of consensus: ~90%+ of discussion participants support this general position
  • Policy implication: WP:CLN already states this. WP:ATA should highlight this policy
• List style articles may be titled List of …. or any other appropriate title that conveys the list topic in a concise manner and is consistent with WP:Article titles.
  • State of consensus: ~90%+ of discussion participants support this general position.
  • Policy implication: WP:SAL probably could be reworded to make List of … just another of many alternatives for list style articles.
• Purely (or mostly) navigational lists should not be separated from the article space (i.e. religated to the Project space).
  • State of consensus: ~90% of discussion participants support this general position.
  • No change in policy required. Policy implication: Some changes to WP:SAL may be necessary to clarify the allowance for the unique nature of purely navigational lists in the article space.

--Mike Cline (talk) 19:28, 24 August 2010 (UTC)— Preceding unsigned comment added by Mike Cline (talk • contribs) 19:01, 24 August 2010 (UTC)oops

Mike, I think that's a fair summary of the current state. I think that the "topic vs title" issue could be generalized to all articles (not merely lists). WP:N might benefit from a sentence like "Do not assume that the exact, quoted phrase used for the current article title is always the subject of the article."

Also, I'm not sure that 'separated from the article space' is the clearest way to express this point to people who haven't been following this (but I don't have a good suggestion offhand). WhatamIdoing (talk) 19:10, 24 August 2010 (UTC)

As far as the "separating" navigational lists from article space idea goes... you can omit that from the summary completely ... if you need to, consider it "Withdrawn". I raised it as an idea... and it is obvious no one liked it. No harm, no foul... and no need to even discuss it further.

More trouble than its worth to withdraw. It was a valid position that just didn't get any traction, no harm, no foul. No follow-on is required anyway.--Mike Cline (talk) 20:03, 24 August 2010 (UTC)

Something that I think still needs resolution is the distinction between topic notability and content notability... that notable items do not always mean a notable topic and vise versa. Blueboar (talk) 19:41, 24 August 2010 (UTC)

I think I know what you are saying, but I would contend two things: 1) notability in itself is a vague notion. Since we determine notability by coverage of something in reliable sources, its really sourcing that's important. 2) I think the collective notability of list entries around a discriminate inclusion criteria make the List Topic defacto notable. That thought is just a vague as notability as a concept is and probably very difficult to turn into a coherent, useable policy. But that said, its still all about sourcing, one way or the other. If an editor can't find reliable sources covering a list topic or the general locus of list entries, then the list probably is flawed in some way or the other. We need to write policy that allows for the maximum amount of lists, with good, well sourced content, that will make this encyclopedia a better product. The poor lists can be weeded out by diligent editors, but policy should encourage and support good lists.--Mike Cline (talk) 20:02, 24 August 2010 (UTC)

No Notability is for articles only, period. Content is explicitly exempt from N per WP:NNC. It still must meet V and BURDEN, but notability of article contents is a conversation that doesn't need to happen, ever, anywhere, period. Jclemens (talk) 15:38, 25 August 2010 (UTC)

I personally believe the "list of Xs" versus "Xs" debate is off topic, and too vague to really tell us anything. I'd almost agree with it in principle, and I'd almost agree about the title thing. But so what? "Virgin" (X) is notable, but we don't have "list of virgins". So it's never been a simple question of whether the overall topic is notable, but whether it's appropriate for a list. We run into the same problem for other weird lists. We have a list of tallest buildings, but it's inappropriate to have an article about "tallest building". And what's the topic for list of Sixth Feet Under deaths that makes it patently obvious the third-parties say the show is about death, and yet most reasonable Wikipedians wouldn't dare create a list of deaths? It's possible you've identified a principle where people agree, but what does it mean in practice? Shooterwalker (talk) 19:55, 24 August 2010 (UTC)

The above was an attempt to summarize the broad set of issues raised in this RFC and their current state of consensus, not to raise and/or summarize what is most likely the thousands of permutations of relationships between articles, lists, their titles and topics. For every rule, there will be a million exceptions, and for every exception, dozens of rules will be violated. We need to keep this at a high enough level of abstraction to allow reasoned decision making, unencumbered by every possible permutation of its implementation.--Mike Cline (talk) 20:09, 24 August 2010 (UTC)

I think - before we get too far off what should be on the RFC page - is that the notability of "List of X" vs X is important that we can justify such lists. This point is important to justify or nullify a large number of lists on WP. But that's half the picture. Now we have to consider the other aspect of selecting lists, that being in consideration of indiscriminate coverage and other metrics as outlined at WP:SAL. Your example of "List of Six Feet Under deaths" would be the type that would technically pass the notability barrier (Six Feet Under being notable) but not the indiscriminate consideration. Should we bring this up in the RFC? Or is SAL sufficient in addition to the points Mike brings up above, to account for most cases? --MASEM (t) 20:10, 24 August 2010 (UTC)

Yeah, sorry this is just rehashing the RFC on the talk page. I really don't think the principle that we verify "Xs" and not "list of Xs" helps us much, because I think it's just meaningless semantics. But it does help, in principle, to know that a list can resemble a category and there's nothing inherently wrong with that. It also helps to know that there's a consensus that not every category warrants a complementary list, and not every article warrants a complementary stand-alone list. The issue is when they do, and when they don't. Shooterwalker (talk) 20:38, 24 August 2010 (UTC)

It may seem like a meaningless semantic, but there are people involved in this discussion on the RFC that are insisting and aren't going to stay quiet until we've found a way that "every article topic is notable". Well, as Mike summarizes, we've got language to get to that point. Now we can talk more other practical issues. --MASEM (t) 20:55, 24 August 2010 (UTC)

• If anyone here hasn't already seen it, I'd like to re-suggest that you take a (long) look through User:Quiddity/Navigational pages RfC. It covers many of the same issues examined here, and presents all the proposed-solutions I've previously encountered. I've been following many of the arguments since late 2005, when a few of us who had helped with the Main Page redesign, overhauled Wikipedia:Category schemes (random 2005 diff), so more context is available, if wanted. All thoughts/questions welcome. Here, there, anywhere. :) -- Quiddity (talk) 20:24, 24 August 2010 (UTC)

I agree that Mike's summary is fair and accurate. However, I would suggest that some sort of policy implication is needed in regards to "navigational lists in articlespace". I've elaborated on that at the talkpage for the RfC in my userspace. HTH. -- Quiddity (talk) 20:24, 24 August 2010 (UTC)

Good catch, I tweaked my summary a bit.--Mike Cline (talk) 20:37, 24 August 2010 (UTC)

• I think this "uninvolved" summary is too one sided to be credible. It is foolish to believe that a tweak or two to WP:SAL is all that is needed to legitimise these proposals. The reality is that lists would have to be exempted from WP:N, WP:NOT would also have to amended to exempt lists as well. There is an element of wishful thinking that there are no alternatives to what is being proposed, and I think these need to be summarised as well. --Gavin Collins (talk|contribs) 22:54, 24 August 2010 (UTC)

• I don't see this summary as biased. Most of the other suggests (such that "List of X" has to be a notable topic) clearly don't have consensus, so they are rabbit holes we can avoid to arrive at the answer. As to whether WP:N or WP:NOT needs to be changed, we can figure that out based on the exact wording change at WP:SAL, if necessary. As it stands, I don't think we have to touch WP:N or WP:NOT to make this summary work. --MASEM (t) 22:57, 24 August 2010 (UTC)

• Gavin, as I've encouraged you in the past, your views are important but must be stated concisely in the context of the discussion. If you believe the summary above does not reflect the consensus on the issues in the RFC, then please provide us your summary of the consensus as you see it and the policy implications that consensus might drive. In this case we know what your position is and that's not what this thread about, it's about what we've determined from this RFC todate on the major issues. If you have a different summary than the one above please lay it out. We may have missed something. I know I did in my first pass.--Mike Cline (talk) 23:04, 24 August 2010 (UTC)

• To give you one example, my understanding is that is the term List Topic is still disputed, and therefore there is fundamental disagreement about where the burden put on a list to demonstrate notability lies. I think this issue has to be resolved first. --Gavin Collins (talk|contribs) 04:55, 25 August 2010 (UTC)

Using the following definitions as established at the beginning of the RFC for clarity here is a quick assessment of the various positions discussed in this RFC about this distinction (apologies upfront if I’ve misread or missed anyone’s position):

Positions

• Position A: List of X or List of X and Y where X or X and Y represents the list topic as described by the lead and inclusion criteria in the lead. Any notability burden is placed on the List Topic.
• Position B: List of X or List of X and Y where List of X and List of X and Y represent the list title. Any notability burden is place on the List Title.

Those generally or explicitly supporting Position A

• Cyclopia^talk
• Jclemens (talk)
• Cybercobra
• Polaron | Talk
• Davewild (talk) 18:28, 20 August 2010 (UTC)
•  Sandstein 
• Blueboar (talk)
• Sjakkalle (Check!)
• RJC Talk
• WhatamIdoing (talk
• postdlf (talk)
• MASEM (t)
• Orlady (talk)
• WF
• Alzarian16 (talk)
• Mike Cline (talk)
• ALR (talk)
• Slrubenstein | Talk
• ‒ Jaymax✍
• Stefan (Talk)
• Sandman888 (Talk)

Those generally or explicitly supporting Position B

• ThemFrom
• Zunaid
• Shooterwalker (talk)
• Gavin Collins (talk
• Active Banana (but even so, that would leave us with articles like List of fictional doctors)

Now, since the two positions are fundamentally opposed and in no way could an either/or policy be crafted that imposed both positions on an editor, it was my view of the discussion that the greater consensus was with Position A: The List Topic is X or X and Y not the List Title. We completely understand that you and a few others do not agree with that construct. There are two avenues we could take from this point on: 1) Proceed with ensuring List related policies reflect the consensus Position A. or 2) Discuss this simple (List Topic/List Title), but fundamentally opposed (it’s one or the other) construct for the next year using every conceivable example and eventually get back to where we are today. I personally don’t think the current state of consensus on this is going to change much. If you believe otherwise, please provide your summary of the current state of consensus for this RFC so that we can get on with editing.--Mike Cline (talk) 07:42, 25 August 2010 (UTC)

You place me in the correct group as far as where the burden of notability lies... However... I believe that a list's Title should reflect the list's Topic, and the Title should inform the reader what Topic is. The two concepts are closely intertwined. To me, the far more important distinction is the distinction between the Topic and the Content of the list.... the notability of the topic vs. the notability of the items listed. Blueboar (talk) 12:32, 25 August 2010 (UTC)

Indeed, and that's an editorial decision that takes into count our naming convention policy: WP:Article titles.--Mike Cline (talk) 12:39, 25 August 2010 (UTC)

I guess I'm still not sure what this gets us. Were people really insisting that "Delete: sorry, you have these perfectly good third-party sources talking about Green Bats of Japan, but we couldn't find a source about a List of Green Bats of Japan"? Shooterwalker (talk) 13:49, 25 August 2010 (UTC)

Absolutely, that's why its an issue.--Mike Cline (talk) 13:53, 25 August 2010 (UTC)

Wow, I had no idea. Shooterwalker (talk) 14:19, 25 August 2010 (UTC)

A more realistic example of how notability affects list articles is Wikipedia:Articles for deletion/Worst Britons, currently at AfD. Aside from POV concerns, there are also concerns that the list isn't notable in itself (that is, that this particular "list of worst Britains" hasn't received significant attention). ThemFromSpace 20:47, 25 August 2010 (UTC)

It's a bit inappropriate to presume which of those one might end up in, although the title should reflect the topic. My main interest is that there is evidence to support the assertions; a statement of the topic notability and a clear articulation of inclusion criteria.

ALR (talk) 12:41, 25 August 2010 (UTC)

Apologies for reading the tea leaves wrong on your position. I agree with your 2nd sentence as well and suspect that consensus on that point is near 100% except for those pesky navigational lists.--Mike Cline (talk) 12:53, 25 August 2010 (UTC)

I believe Position B is mistated. The difference of opinion is based on the the following:

• Position B: List of X or List of X and Y where the defintion of List of X and the definition of List of X and Y represents the List Topic as described by the lead and inclusion criteria in the lead, or by the list's title if a defintion is not explicitly described. Any notability burden is placed on the List Topic.

I think we are misunderstanding each other here, and this may be related to whether a list can be described as being or about X or whether described as being a particular set of X.

The confusion is probably related to the nature of definitions, which can be both descriptive ("This list is about...X") or stipulative ("This list contains...all known X"), but either way, the defintion needs to be the subject of at least one reliable source to justify inclusion. --Gavin Collins (talk|contribs) 14:29, 25 August 2010 (UTC)

Gavin, I'm not really following you here - do you refer to the definition of the list or definition of the topic or the definition of the criteria? ‒ Jaymax✍ 00:30, 1 September 2010 (UTC)

Mike, I think you are far too involved to even be speculating as to this RfC's outcome. The RfC has only lasted about a week, while they provisionally run for four weeks. Don't insert your own personal opinion in the form of a summary and let an uninvolved admin do this work in a few weeks. Stick to the discussion on the RfC for now. After the discussion period is finished (this is still listed on CENT so it will likely gain more participation) we can decide how to move forward. ThemFromSpace 20:37, 25 August 2010 (UTC)

I don't see any reason Mike can't state what he thinks is the outcome or where he thinks things are headed; it's not like he's trying to close the discussion. If I were the closing admin I'd see this "summary" as quite helpful, particularly because people are commenting where Mike has them wrong which clarifies their positions. I am clearer about Gavin's position from this than I might have been from merely reading the !votes and comments on the project page.--Doug.^{(talk • contribs)} 17:07, 5 September 2010 (UTC)

Should position A be written into WP:SAL? Sandman888 (talk) 09:40, 7 September 2010 (UTC)

suggested next step. work on WP:LISTOUTCOMES

It might be easier to work on a guideline if we were all working from the same evidence. I think a lot of us have different experiences with lists, and we can all think of a few AFDs that confirm our point of view, but might not have seen other AFDs that are trickier to explain. Some of the most useful discussions came about when we tried to explain why some types of lists are always kept, and some types of lists are always deleted. The most useless discussions were when we re-opened individual AFDs, and got bogged down in arguing about which way a borderline case should have gone.

I suggest that people create a page of WP:LISTOUTCOMES from recent AFDs (within the past few years). Maybe offer a deadline of 2 weeks or even a month just to gather and organize a pile of AFDs. We will see entire categories of lists that are red, and entire categories of lists that are blue, and then we will all be able to work from the same experience to craft a guideline. (Even if our points of view will remain different.) Shooterwalker (talk) 14:19, 25 August 2010 (UTC)

Not that its really necessary to do a big survey on lists to craft a guideline, I would suggest an alternative approach that would be much more positive. AfDs are required because editors created articles that were flawed in some aspect. We should not base our policies on experiences with flawed articles. Instead, I would survey the locus of all the really good lists--from navigational to significantly annotated content and build our policies based on the reasons they are good lists. Our policies need to encourage good lists and support the criteria that make them good. If that criteria is sound, then flawed lists can be easily dealt with by demanding editors.--Mike Cline (talk) 14:39, 25 August 2010 (UTC)

I would add this: we probably want to assert a few more things about lists beyond the notability issue that is close to resolution. (For example, "List of Y of X" where "Y of X" is not notable, but X is, as often the case in fiction coverage). We should try to come to a consensus on those additional points, but failing to do so, an OUTCOMES may be helpful to include as a backup. --MASEM (t) 15:05, 25 August 2010 (UTC)

I would support that approach. We can learn a lot by looking at Wikipedia:Featured lists in terms of what is appropriate. Ideally we would do both. Figure out the clear keeps, and the clear deletes, and build a guideline around those with room in the middle for interpretation. I think we should try to push this discussion as far as we can, of course. But I don't expect much more than an agreement in principle. Shooterwalker (talk) 15:08, 25 August 2010 (UTC)

Which leads to a practical question from a driveby editor: will the lists be deleted or not? Does it make any sense to edit any list now? Don't share your personal expectations, say a firm verdict, then stick to it. East of Borschov 15:45, 25 August 2010 (UTC)

As far as those yet uninvolved in this discussion, nothing on how present list handling has changed, only better clarified. We are seeking further clarification but it is not expected to change what current practice is. Edit away per all other policies and guidelines. --MASEM (t) 16:03, 25 August 2010 (UTC)

I wouldn't expect a moratorium on creating OR deleting lists. Whatever this guideline comes to be, expect it to describe how things usually go, but sometimes don't due to slight jumps in ideological editors one way or another. I do think we could achieve a lot by listing outcomes. It becomes more obvious what's actually going on, and gets us away from personal opinions. Shooterwalker (talk) 21:50, 25 August 2010 (UTC)

Clarification: what is a list topic?

Rather than launch into a long rebuttal of of Mike Cline's views in the Clarification section of theo RFC. I think we need to discuss and agree upon what is a list topic.

I think we are all agreed that this RFC is about the inclusion of list articles in Wikipedia, not the inclusion of individual items within a list. I think we are also agreed that the list topic is not the same as the list title, for they are distinct elements of an article.

However, some misunderstanding has crept into the discussion, and I quote Mike:

• List Title: Metadata that provides a unique identifier within WP and concisely conveys the subject of list. List titles are constructed typically as: List of X (List of Lakes in Montana), …ography of X (Discography of Jane Doe), or noun of X (Birds of Foobahland) and other variations.
• List Subject: (or List Topic): What are the list contents all about? As the above titles suggest, the lists are about Lakes in Montana, the recordings of Jane Doe, the Birds in Foobahland, etc.
• List Lead: Prose that establishes both the subject of the list and inclusion criteria for list entries.
• List inclusion criteria: Specific, verifiable criteria that establishes the boundaries of list entries. Inclusion criteria are directly related to and expand understanding of the List subject.

I think there is a fundamental misunderstanding in Mike's understanding of what is the list topic, which can be summarised as "the List of lakes in Montana is about the lakes of Montana". The burden of notability falls on the whether the lakes are notable, rather than the defintion of what the list contains.

My understanding (which could also be mistaken, in fairness) is that a list topic is the defintion or inclusion criteria for the list, e.g. "the List of lakes in Montana is about a defined set of lakes in Montana". The burden of notability falls on the list defintion, rather than the lakes themselves.

At this point, I have to ask, is this what Mike meant to say, or have I misunderstood? --Gavin Collins (talk|contribs) 09:47, 26 August 2010 (UTC)

I think this is purposely twisting Mike's language to fit your opinion. Further, based on Mike's summary, it would seem that most would also agree with his language of what a list topic is. You could try to argue it further (on the RFC page) but I don't think you're going to change anyone's minds about it. --MASEM (t) 12:38, 26 August 2010 (UTC)

Should this be moved to the main page? I think it needs broader discussion. Blueboar (talk) 12:17, 26 August 2010 (UTC)

As for my opinion... With these examples there are actually two topics... a general topic and a specific topic. The general topics are "lakes", "birds" and "Jane Doe" ... the specific topics are "lakes that are located in Montana", "birds that are found in Foobahland" and "songs that have been recorded by Jane Doe". We need to establish that both the general and the specific topics are notable. Now... I believe that if there is an existing "main article" about the general topic (where, presumably, we have established notability), there is no need to re-establish the notability of the general topic at the list article. If we can not establish that the general topic is notable, then it will be impossible to establish that the specific topic is notable (Lack-of-notability is inherited). However, even if the general topic is notable, we still need to establish that the specific topic is notable. If we can not establish this, then we have two options... we can either delete the list entirely, or we can subsume it into the article on the "main topic" Blueboar (talk) 12:58, 26 August 2010 (UTC)

In answer to Masem, no misrepresentation of Mike's position is intended, and already admitted that my own views may be found to be entirely mistaken, or that I have misunderstood what Mike's positon is.

I disagree with Blueboar that there is any requirement to find the general over-arching topic to be notable, for the simple reason is that a list can be notable in its own right, regardless of whether the general topic (say, "Lakes of Montana") even exists.

Where I would agree is that that lists are sub-topics of a general topic, e.g. the "List of lakes in Montana" is a sub-topic of the general topic "Lakes of Montana". However, where I am in disagreement with Mike is whether or not there is any burden to provide evidence of notability for the sub-topic, and if so, how or what form should that evidence should take. I think these are important questions that, with hindsight, we should have agree upon before the RFC started. --Gavin Collins (talk|contribs) 13:25, 26 August 2010 (UTC)

In earnest, I really don't understand the difference between the two positions. Shooterwalker (talk) 14:47, 26 August 2010 (UTC)

My understanding is that under Mike's framework, it is not necessary to provide evidence of notablity for the sub-topic if they part of a notable category. In the example given above, some of the lakes in Montana are notable, e.g. Flathead Lake, therefore the List of lakes in Montana is deemed to be notable by default. --Gavin Collins (talk|contribs) 15:06, 26 August 2010 (UTC)

A question for Mike... it sounds like you are saying that if all the items listed are notable, the topic is notable by default. Is this a correct representation of your view? Blueboar (talk) 15:10, 26 August 2010 (UTC)

I don't believe this is what Mike is aiming for; Mike can clear that up himself, but I want to intercede before I think we go too far down the wrong path. We are considering how to judge the inclusion-worthiness of "List of X" and specifically what aspect needs to be shown notable. I do not believe Mike is arguing (as he has spelled out) that "List of X" is included because each element that falls into "List of X" is notable while neither "List of X" or "X" are notable topics. Instead, I believe he is saying that when the "X" itself is a notable topic, "List of X" is inclusion-worthy regardless of the notability of the elements within the list. So in the example List of Lakes of Montana, to be included, the topic "Lakes of Montana" as a whole should be notable.

I note that this should not be taken to invalid cases where "List of X" is actually a notable topic. I will also note that we should also discuss at some point another possible means for list inclusion implied by Gavin/Blueboar in which each element of a list (navigational or otherwise) is notable but neither "List of X" or "X" is a notable topic on its own. There probably needs to be clarity to make sure it is a non-indiscriminate grouping and doesn't imply correlation without causation. But I believe that is a completely separate case from what Mike was discussing in the first point. --MASEM (t) 15:26, 26 August 2010 (UTC)

OK... Masem (and only Masem, please)... Do you see a difference between saying that the topic is "Lakes of Montana" and saying the topic is "lakes that are located in Montana"? Blueboar (talk) 15:52, 26 August 2010 (UTC)

Out of context, there's no difference, they refer to the same things; but in context of Wikipedia, they are different. "Lakes of Montana" gives me the impression that the collection of lakes in Montana is the topic as a whole. The latter seems to suggest that we have a lakes and these are are put together simply because they share one equivalent feature: they all are in Montana. The former is preferred because, as it suggests a collective group with specific examples being listed below, then likely with a bit of research and the like, the geological history of the general grouping lakes, their ecology, impact on tourism/economy/environmental aspects, and other concepts can be discussed in the context of the grouping of Lakes of Montana. Of course, this is a blind assertion, it does really depend on the context of the list description text and what lakes are listed to make the proper judgement. --MASEM (t) 16:47, 26 August 2010 (UTC)

List of lakes in Montana breaks down this way:

• List title: List of lakes in Montana
• List topic: Lakes in Montana
• List inclusion criteria: Named (as named in USGS database) lakes in Montana by county (supported by reliable sources)
• List entries: List of lakes in county X (Similar inclusion criteria for each county list, except individual list entries are lake names. (each named lake in each county list is supported by a reliable source)

So where does the notability burden fall and how is it satisfied? Named lakes as a geographic feature (regardless of where they are) are considered notable Wikipedia:Notability (geography), and Montana as a state is notable. Therefore any list of notable things with a common characteristic (ie. named lakes in this case) in the state of Montana should satisfy the notability of the List Topic. The article is not about some List of … (I know of no source that has commented on the collective set of lakes in Montana as a set), the article is about lakes (their name, location and other data). The topic is modified by the state Montana and by Montana counties. The reason for that is merely article size. A list of 200 entries is about max for WP on the article size front. In Montana the number of named lakes exceed 2000, so a single list is not possible and the by county approach allows each county list to be linked to a related county article. From a purpose standpoint, these lists are both navigational, developmental, and informational. There is a distinction between the literal words of a List Title and the literal words of the List Topic. Notability burden should be on the topic not the title.--Mike Cline (talk) 15:55, 26 August 2010 (UTC)

AH... Mike brings up a very important point that we have not discussed before... the inherent notability of certain broad topic categories... in this case, the fact that geographic places (including both lakes and states) are considered notable by default. This does muddy the waters somewhat. Blueboar (talk) 16:42, 26 August 2010 (UTC)

No, because they are notable because generally they have been covered (documented) by a reliable source (in this case the USGS). They aren't notable because they just exist, they are notable because they are named and data has been accumulated on them by a reliable source.--Mike Cline (talk) 16:51, 26 August 2010 (UTC)

There must be hundreds of if not thousands of lakes in Montana that fit that criteria, but I would dispute their notability based on map data alone, but that is a seperate issue. I think what Mike is infering from the inclusion criteria he has described above is a variant on WP:NOTINHERITED based on category: if the category is notable, then the members of that category inherit that notability. --Gavin Collins (talk|contribs) 17:01, 26 August 2010 (UTC)

I always thought the "inherent geographical notability" only applied to populated places. Alzarian16 (talk) 17:02, 26 August 2010 (UTC)

The notability of individual lakes is a completely separate discussion - maybe necessary to have but not here. For purposes of this discussion, I strongly recommend we start from the working principle that every USGS-listed lake in Montana is notable, and worry about the list of these as the issue. --MASEM (t) 17:04, 26 August 2010 (UTC)

I thought the purpose of this little sojurn was to clarify the distinction between list topic and list title and where the notability burden (if any) should fall. Based on the discussion, that's been accomplished.--Mike Cline (talk) 17:09, 26 August 2010 (UTC)

I think the problem is that you have identified the topic of a list to be the category it is is a member of, rather than the selection criteria for the list itself, i.e, a list is sub-set of a notable category, rather than a list is a set with its own in its own right. An analogy would be Matryoshka doll: just because the doll is notable, that does not mean one its components (the baby doll) is notable. --Gavin Collins (talk|contribs) 17:31, 26 August 2010 (UTC)

Really. What's the difference between establishing the notability of "lakes in Montana" and establishing the notability of "list of lakes in Montana"? Either way, aren't we just looking for multiple, reliable third party sources that cover multiple lakes in Montana (as a group or as a series) in direct detail? I just don't get the difference.Shooterwalker (talk) 20:55, 26 August 2010 (UTC)

It is an important distinction since in past debates, editors opposing a particular list have stated to the effect: No reliable source has defined what a List of lakes in Montana looks like or contains so it must be a madeup topic or OR. The same type statement cannot be said of Lakes in Montana.--Mike Cline (talk) 21:02, 26 August 2010 (UTC)

Let me try to explain the difference. The "Forbes' Index of the 100 richest people" (I probably have the title wrong... but I think you know what I mean) is an example of where a list is itself a notable topic. It is a well known published list. We might present that topic in paragraph form, or as a list... but either way, the topic of the Forbes Index is notable. As far as I am aware, there is no comparable published and notable "list of lakes in Montana". So the list itself is not a notable topic. However, the topic of "lakes that are in Montana" may well be considered notable enough for inclusion in Wikipedia, and one way to present such a topic is through a list. Blueboar (talk) 21:21, 26 August 2010 (UTC)

In the case of the Forbes List you could have two different articles (but they would require slightly different names to distinquish them): Forbes List of The 400 Richest Americans could be an article that explains the structure, compliation and history of the Forbes list. That article could include an embedded list that listed the current 400 list entries. If that embedded list was too large for the main article it could be broken out as a standalone list (copyvio considerations asided) entitled Current list of Forbes 400 Richest Americans. In the first case, the article is a normal article and the List itself must be notable WP:SAL wouldn't apply. In the second case, the list topic is the membership of the Forbes 400 list (clear inclusion criteria) and WP:SAL would apply.--Mike Cline (talk) 22:46, 26 August 2010 (UTC)

Wow, okay, that's really enlightening. Then yes, I think it's ridiculous to demand that every list be a notable list in of itself, like the billboard 100. For most Wikipedia list articles, it would be sufficient to find a source that even for a couple good paragraphs covers a group or set or category of things. In other words, you verify the notability of the set/group/category of things, rather than verifying the notability of someone's special top 10. Is that what we've been arguing about all along? Shooterwalker (talk) 01:53, 27 August 2010 (UTC)

SW - Excellent encapsulation of the issue. Indeed that distinction between List Title and List Topic has been a sticking point. Have you ever thought about a job writing policy for WP?--Mike Cline (talk) 07:29, 27 August 2010 (UTC)

Thanks! Do they give out jobs for policy writers? Shooterwalker (talk) 14:34, 27 August 2010 (UTC)

Notability has nothing to do with the title, only the topic. Why is it ridiculous to demand that every list topic be a notable list if we ask the same thing of every article topic? The scenario that Shooterwalker describes is based on WP:INHERITED, and although it sounds simple, it is actually difficult to justify, because this boils down to the idea that coverage in one article relates to another list. I think the coverage of the Forbes 400 is rather poor, and the coverage of List of members of the Forbes 400 (2008) is not great either, but a better example would be Sunday Times Rich List and Sunday Times Rich List 2008 indicates that the 2008 list (as well as all the other years) is a distinct and notable topic in its own right. The evidence of notability is weak, but the reader benefits from the context this provides.

Going back to Mike's example, and I think this case list sums up the position regarding the relationship between the general over arching topic ("Lakes in Montana") and the list topic ("List of lakes in Montana"):

Case 1: "Lakes in Montana" is not notable, nor has a list of lakes ever been published, in which case it could never have been notable;

Case 2: "Lakes in Montana" is notable, but a list of lakes has never been published, in which case it could never have been notable either;

Case 3: "Lakes in Montana" is notable, and although a list of lakes in Montana has been published, there is no evidence that he list is notable;

Case 4: "Lakes in Montana" is notable, and the list of lakes in Montana has not only been published, but it is also notable;

Case 5: "Lakes in Montana" is not notable, but the list of lakes in Montana has been published, and it is also notable.

I think "List of lakes in Montana" is a case 1 example at this time. There is evidence to suggest that Flathead Lake is notable, but that is a separate topic in its own right, even though it is a member of the same category. The key point is the notability means just that: each list has either been "noted" in accordance with WP:N or they have not. Notability is not deemed to be inherited, nor is it transferred just because it is a member of a particular category, or sub-category of multiple categories. An example of each case might be as follows:

Case 1: Masonic building, List of Masonic buildings

Case 2: Isle of Wight, List of people from the Isle of Wight

Case 3: Heroes (TV series), List of characters in Heroes

Case 4: Sunday Times Rich List, Sunday Times Rich List 2010

Case 5: Nixon's Enemies, Nixon's Enemies List

I think we have to move away from the idea that notability is collectively shared between topics in a category, because lists can be members of multiple categories. The list topic is not the list title, but it is not the list category either. --Gavin Collins (talk|contribs) 09:04, 27 August 2010 (UTC)

This is tricky stuff and we're getting back into the area where people clearly disagree. But at least we've established that isn't not enough to verify individual list members (too cold), but that you don't need to WP:verify notability of a specific written list (too hot). We're somewhere in agreement that you verify notability of a set of things (just right), just that there's a lot of ambiguity in that which you've pointed out. I actually think this would be an excellent point for a guideline to start out. As vague as it is, it would bring just a little more sanity to list discussions where there are people who want to keep everything or delete everything. Shooterwalker (talk) 14:34, 27 August 2010 (UTC)

The problem is, if a list does not have any evidence that it is notable, then there is no verifiable evidence that is encyclopaedic in any way. Baring in mind that lists are little more than bare bone articles, one would have to question why an encyclopaedia needs so many of them if a few sentences of commentary will do a better job. Lists for lists sake is not the way forward, and the List of Masonic buildings: since it has not been published outside of Wikipedia, one can only guess what rationale there is for creating in Wikipedia. Maybe the creators of such a list are seeking academic recognition for their work? Who knows! At least with notability, you know where you stand. --Gavin Collins (talk|contribs) 00:05, 7 September 2010 (UTC)

The problem and magnitude of the problem

One of the pitfalls with examining isolated examples of data in an effort to propose new solutions to a perceived problem is that the isolated example may or may not demonstrate the problem in a way that makes a compelling case for one solution or another. In the case of this List RFC, what is the problem we are attempting to resolve my proposing some policy change? As I see it, one of the perceived problems is that some lists exist and/or are created that are indiscriminate, non-notable or both. Now no policy wording in WP is going to prevent the creation of indiscriminate, non-notable lists. If policy could prevent the creation of articles that don’t stand-up to our policies we would not need a deletion mechanism. On the other hand, policy should be written to encourage discriminate, notable lists while allowing for removal of indiscriminate, non-notable lists. One of the problems with changing policy is assessing the impact those changes might make the on class of articles it is intended to regulate. If we concede that WP contains some number of indiscriminate, non-notable lists is a reasonable statement of a perceived problem, then I believe asking the question What is the magnitude of the problem? is a reasonable request, before new policy is formulated. If there are two potholes in a 1000 miles of highway, you don’t need to hire 100 inspectors to search a 1000 miles of highway for potholes every day, you just need to fix the two potholes. If there are 1000s of potholes, the inspectors might be a reasonable solution.

So, to that end, I would suggest that the editors who hold the minority position in the discussion—that the burden of notability and discrimination falls on List of X and not X (a position that would require a written guideline change to enforce)--compile a list of WP standalone lists currently existing that would fail in their opinion the new notability burden and be removed from WP. To ensure the real magnitude of the problem is understood, this enumeration ought to represent at least 10% of our current lists and a broad range of subject matter. That seems that it would be a reasonable sample to judge the impact of a policy change of this nature and eliminate the vagaries of using isolated examples.--Mike Cline (talk) 11:18, 27 August 2010 (UTC)

• I think it's dangerous to ask people to compile "bad lists" that should be deleted, just as much as it's dangerous to compile "bad deletions" that should have been kept. Talking about specific examples will just rehash what goes on at AFD. Talking about what "should have" been done will almost always lead to arguments.
• A more productive idea would start from the premise that the community is always right, and that the AFD process usually represents consensus with the exception of 10% of outliers. If we compiled 100+ AFDs and/or lists, we'd quickly see there are some lists that are almost always kept, and others that are almost always deleted. Then we could write a guideline that describes current practice. But we'd have to agree that current practice is basically fine, if only as a starting point. Shooterwalker (talk) 14:43, 27 August 2010 (UTC)

SW - Although I don't disagree with what you've said above, I do think you misconstrued the purpose of the request. A minority number of editors are proposing that we change the burden of notability for list topics to List of X instead of our current practice on X. If this change was adopted it would have a significant impact on the current lists in WP today. Reviewing AfD results would not give us any sense of that impact because most lists at AfD were either kept or deleted by evaluating the notability of X, not List of X. The list that is being requested above is not a list of Bad Lists as they exist under current guidelines, but instead of list of lists that would become Bad Lists if the focus of notability changed from X to List of X. That is the only way the impact of this radical change can be assessed.

I will agree that evaluating AfD for the various reasons that lists under our current policies were either kept or deleted would help in crafting better list related policy wording that would encourage good lists and possibly deter bad lists. Unfortunately the radical List title vs List Topic proposal by Gavin and others hasn't been part of the AfD decision process (even though its been agrued in AfD debates many times).--Mike Cline (talk) 20:46, 27 August 2010 (UTC)

Its not a minority, Mike. Our existing guidelines on notability are clearly focused on topic, rather than category they belong to as you are suggesting. Most lists are kept or deleted in expectation that the topic is notable: there is no other metric for inclusion. Please drop the idea that list title has anything to do with notability, I think we are agreed it is not the case. --Gavin Collins (talk|contribs) 00:31, 28 August 2010 (UTC)

People have the funniest way of agreeing. I think everyone understands that the list title isn't the issue. So let's focus on the next step. It's a combination of extracting whatever few abstract principles we can from this discussion, and moving onto some kind of WP:OUTCOMES for list to ground our next discussion. That way we won't argue about what common practice is. It will be in front of our face. Shooterwalker (talk) 15:19, 28 August 2010 (UTC)

SW - Unfortunately there is no agreement with Gavin's minority position that the List topic equates unequivocally to List of X not X. Whether we call it the list title or topic is irrelevant. One only has to follow Gavin's positions on the RFC page as well as on other policy pages to know that what Gavin means is: If the literal List of X isn't notable in its own right, it doesn't matter if X is in anyway notable or not. It is not only a gross misinterpretation of WP:NOTINHERITED but it is a completely unreasonable and untenable position given the current state of lists on WP. His is clearly a minority position if one reviews the locus of comments on the RFC page. Indeed we all would like to move on beyond this pothole.--Mike Cline (talk) 15:34, 28 August 2010 (UTC)

That seems like an odd position to take (and you're probably right that it's a minority position that shouldn't stop us from finding an overall consensus). But let me ask Gavin himself. Gavin, do you believe that every list in Wikipedia needs to be a real world list that has gotten wide attention in third party sources, like New York Times Medical School Rankings, The Billboard 100, or Nixon's Enemies List? Or is it enough that all Wikipedia lists focus on a class or a set that has been covered in third-party sources (as a class or a set, not just as individual separate entries)? Shooterwalker (talk) 15:47, 28 August 2010 (UTC)

Don't ask me for is it of no matter what I think. What do our policies and guidelines say on this issue, Shooterwalker? --Gavin Collins (talk|contribs) 20:29, 29 August 2010 (UTC)

You're entitled to your opinion, and I guess that includes the right to withhold it. I guess we can summarize consensus on this issue, basd on what a few of us have read from this discussion There's a consensus that if multiple third-party sources discuss a set or group of things in direct detail, then that set or group is appropriate to be covered as a list on Wikipedia (assuming it meets other policies like WP:NOT). Shooterwalker (talk) 20:56, 29 August 2010 (UTC)

What you are describing is a notable over arching category, not a notable topic. Neither Wikipedia's content policies nor the notability guideline mention categories at all. The only mention of categories is in WP:NOTINHERITED, so there is no evidence that this view is the consensus by any strech of the imagination. For a topic to be notable, it has to have been "noted" in accordance with WP:N. Notability cannot be infered from membership of a category, as there is no verifable evidence to support such an inference. --Gavin Collins (talk|contribs) 08:15, 30 August 2010 (UTC)

I always thought that the topic of a list is a group of things as an overall group (and not the individual members). Shooterwalker (talk) 01:04, 31 August 2010 (UTC)

Yes... what I am suggesting is that the way to identify and evaluate this topic is to identify and evaluate what is common to the over all group and not the individual members of the group. Blueboar (talk) 16:06, 31 August 2010 (UTC)

That is correct: a list is a topic in its own right because it is a defined group of things, and is similar to a mathematical set in htis regard. A category, on the other hand, is not defined in terms of elements, but is defined in terms of characteristics. --Gavin Collins (talk|contribs) 15:56, 31 August 2010 (UTC)

This distinction, while perhaps logically valid, is arcane and I don't think helpful - the people to whom any policy or guidelines apply are not going to draw the distinction. Furthermore, there is significant overlap, depending how the set is defined. When we are talking about list articles, there is (usually) no assumption that the list is 'complete' - what defines a list here is typically not the set of entries that make it up at any point, but the basis upon which entries are included - I think ‒ Jaymax✍ 01:29, 1 September 2010 (UTC)

I have to disagree because the difference between mainspace pages and navigational pages is clearly understood: just becasue a topic is a member of a Wikipedia category, that is not a valid rationale for inclusion as a standalone article or list. Categories are simply navigational aids: they were never intended to provide evidence that a topic is notable because it shares a common attribute with another topic. The argument that notability can be inherited through categorisation has been rebutted at WP:NOTINHERITED. --Gavin Collins (talk|contribs) 10:10, 1 September 2010 (UTC)

Throwing thought out there

This is a subject in which I am very interested - I've only just come back after an extended wikibreak, so it's going to take me some time to catch up. However, here are some random thoughts.

1) I am very much in the List of X must not require that the list or act of listing in and of itself must be notable. As a thought experiment, an article "List of lists of richest people" would be good for the encyclopaedia, (as would an article "rich lists") REGARDLESS of whether such a list of lists has achieved notability independently. This inherently means that the compilation of such a list will involve Original Research - in the same way that compiling an Article must involve OR. Of course, neither should ever present OR, content must be sourced.

2) I firmly believe that the significant bulk of rules/policy/guidelines for lists should apply equally to in-article lists and stand-alone lists. There may be need for some criteria that determines whether a list best sits in-article or stand-alone; but one of those criteria would have to be size - that simple fact (meh, my opinion) leads to an obvious conclusion - A good in-article list should be moveable to being a stand-alone list-article fork, just like any sub-topic that gets too big is.

3) Lists should be of the form "List of Discriminator X" - I'm not talking about titles per se here - but that the criteria for list inclusion is inherently part of the question. Let's take the three 'levels' here: we essentially have potential article 'topics' of LDX, DX, and X. It seems to me that Gavin et al feel LDX should be notable of itself, Mike et al feel it is sufficient for X to be notable. But the question of D is very important, and largely missing from the debate above. "List of Michigan Lakes over 1000 sq m" - So assume 'lakes of Michigan' is notable (in the sense of a non-listing article). A list of all of them is both pointless and unhelpful. Consensus seems to be that it is not required that the act of listing lakes in Michigan be itself notable. But, with the BEST INTERESTS OF WP in mind, I would argue that the discriminator 'over 1000 sq m' should be permissible to be arbitrary, and decided by consensus by the list article editors. Because no other option (that I can see) makes for a better encyclopaedia.

4) Really restating the above - you CAN'T IGNORE the inclusion criteria for list content in this discussion - because that goes straight to scope, and so to notability questions.

5) A useful discriminator in many cases is "List of notable X" - where each entry has it's own article, but "List of significant X" where the list-article editors develop the objective criteria for significance through consensus should be equally acceptable. Some would call that OR, to which I'd say nonsense - because no OR is being presented in the list beyond the regular (non-list) article process of deciding what is in or out through consensus - but by dint of it being a structured list, it is possible to achieve a documented consensus that doesn't need re-consideration for each new factoid someone wants to add.

Phew - done for now. ‒ Jaymax✍ 01:16, 1 September 2010 (UTC)

I would say the above is an excellent characterization of the way things are and ought to be. I have said many times in this RFC and in this essay Creating a better list that inclusion criteria for any list must be discriminate. Your points are right on!--Mike Cline (talk) 01:47, 1 September 2010 (UTC)

I see where Jaymax is coming from, I think what he is proposing is essentially an exemption for lists from WP:N. No editor has been brave enough to propose such an exemption in a clear and explicit fashion, and I think this is about as good an explanation as to why such an exemption is considered to be justifiable, and provide a rationale for unpublished lists and list without notability to be included in Wikipedia. If put your proposal for an exemption in this RFC, whilst at the same time explaining why that it represents a deptarture from Wikipedia's existing policies and guidelines in an explcit and honest way, then other editors will be able to comment on it in an honest way too. In contrast, the proposals of Masem, Blueboar and Mike Cline are not explictly claiming exemptions, and their proposals do not explain why an exemption is considered to be necessary. --Gavin Collins (talk|contribs) 10:34, 1 September 2010 (UTC)

I disagree. I cannot read anything in Jaymax's proposal which would exempt lists from WP:N. Most people in the RFC are of the opinion that the topic that the list covers needs to be notable and that this topic is subject to the same notability criteria as anything else. The idea that the list itself must be notable is in the distinct minority. An article titled "List of X" is not about a list of X, it is a list about X. Sjakkalle (Check!) 11:19, 1 September 2010 (UTC)

We have been through this already. Looking back at Case 5 (above), How can "List of X", be about X, if X does not exist as a topic? how come Nixon's Enemies is not a notable topic, yet Nixon's Enemies List is notable? I don't think anyone has seriously thought this through, and I am mystified as to how this idea that the list topic is subject to the same notability criteria as anything else, yet the list topic does not have to be notable - this seems to me to be a blatant contradiction. How can this contradiction be explained? If someone explain using the example of List of Masonic buildings as to why this list is deemed to be is notable if it has never been published nor commented on, bearing in mind that a the topic Masonic building has not either? --Gavin Collins (talk|contribs) 12:44, 1 September 2010 (UTC)

Apples and Kumquats - Neither Nixon's Enemies List nor Master list of Nixon's political opponents are list articles, they are just regular articles with the word List in their titles. Neither of these articles are governed by WP:SAL. And even though they contain embedded lists, they are indeed articles about real-world lists and not List articles. There is a distinct, clear and unequivocal distinction.--Mike Cline (talk) 12:50, 1 September 2010 (UTC)

I am not following Mike. If the topic of an article is a list...then it is not a list article...even if it has the word List in the title...and because the topic of the article is a real-world list...it is an article about a list...but not a list article. ???? --Gavin Collins (talk|contribs) 13:22, 1 September 2010 (UTC)

I would agree that Nixon's Enemies List is an article that happens to be about a list... but I think Master list of Nixon's political opponents is clearly a list article (ie an article that organizes its material in list format). It may or may not be a stand-alone list... but it is a list. Blueboar (talk) 13:48, 1 September 2010 (UTC)

I think you need to recognise that Nixon's Enemies List is a bona fide list article. Sure, it contains more coverage than your average list (List of Heroes characters being another example of a list plus extended coverage), but it is still a list article about a list topic. Look at the article yourself, and you will see that it is more or less comprised of three parts: the list defintion (the list was compiled in a memo to Nixon and subsequently published as part of the Senate hearings relating to his impeachment), the list itself, and commentary, criticism and analysis from reliable sources about the list (how and why it was compiled, its significance etc). It is no different from any other list, other than its list topic is the subject of richer variety of sources than your average barebone list. --Gavin Collins (talk|contribs) 14:15, 1 September 2010 (UTC)

Nixon's Enemies List is an article about a list, where the list itself takes up less than half the article. There is even a suggestion on the page to transwiki the list-section over to Wikisource, and if that happens there won't be a list there anymore.

As a side issue, I find your tone: "I think you need to recognise that..." as rather arrogant. Sjakkalle (Check!) 14:28, 1 September 2010 (UTC)

Just because an article contains a list does not make it a list article. Is Periodic table a list article because it has the list of elements within it? Absolutely not. Same with Nixon's Enemies List - it happens to contain a list that is part of a notable topic. --MASEM (t) 14:30, 1 September 2010 (UTC)

Just because an article contains a list does not make it a list article, but if it is about a list then it does. The Periodic table is indeed a list article, and I put it to you that like Nixon's Enemies List it contians three parts: a defintion (its structure), the list or table itself, and commentary about the table. The fact that the article may be shorter or longer than the List of Heroes characters does not negate this analysis. If the topic of an article is a list, then it is a list. I should add that list is simple form of table , so don't tell be me that it is not a list article for this reason. --Gavin Collins (talk|contribs) 14:51, 1 September 2010 (UTC)

Gavin, I believe your interpretation is far out of line from what consensus has been historically on list articles. (The question has been discussed at length between the Featured Article and Featured List areas in the last several years). Even if the list takes up the bulk of the article, it can still be considered an article and not a list (see Characters of Final Fantasy VIII). Neither Nixon's list or the Periodic Table would fall into the definition of list used by our featured content. (I will note we do have alternate versions of the Periodic Table, including Periodic table (standard) and Periodic table (large version) which are considered lists.) --MASEM (t) 15:08, 1 September 2010 (UTC)

It is not far out of line, how could it be? We know that a list is a type of article (both WP:LISTS and WP:SAL say so), so the distinction between a list article and an ordinary article is an artificial one - they are all mainspace pages. The only difference between a list topic and an article topic is that...one is about a list, and the other is not. This is common sense, not out of line. --Gavin Collins (talk|contribs) 15:20, 1 September 2010 (UTC)

The distinction between an article and a list is not what it is about, which you are implying, it is about it's structure. That's a huge difference, and why one needs to look more than just the title to understand that, say, "Nixon's Enemies List" is formatted as an "article" article that happens to be about a list. --MASEM (t) 15:30, 1 September 2010 (UTC)

I consider "list articles" to be the articles where most of the content is organized in list form, rather than in paragraphs (prose form). Periodic table is about a classification of the elements, and consists mostly of prose and paragraphs about how it was devised and discussion of the various components in the table. Naturally, the table is in the article (as it is in any good encyclopedia), but it makes up only a small part of the article, so I would not consider the article as a list. At best, an article containing a list. Sjakkalle (Check!) 15:32, 1 September 2010 (UTC)

Format is of no significance. The focus of an article is its topic. The focus of a list article is the list topic. --Gavin Collins (talk|contribs) 15:37, 1 September 2010 (UTC)

Gavin - An interesting interpretation that might be better understood if we applied your logic to a variety of articles that are clearly about lists. Here’s a two part question. Part 1 is a very simple question that requires a simple YES/NO answer. Any elaboration for part I isn’t required. Part 2 need only be answered if the answers to Part I aren’t consistently Yes or No.

Part I – The following five articles are clearly about Lists of some sort. Are these articles in your opinion, stand-alone lists (as you suggest above) and subject to the content guideline: WP:SAL – Yes/No?

• The Clements Checklist of Birds of the World
• Fortune 500
• Nikkei 225
• The 500 Greatest Albums of All Time
• Who's Who in Australia

Part II –If the answers to any of the five above are inconsistent (ie. some yes and some no), what are the distinguishing characteristics that make one or more of the articles a list article (subject to WP:SAL) and the others not subject to WP:SAL? This distinction, if it exists, will be important in crafting future guidelines relative to lists vs articles.--Mike Cline (talk) 18:03, 1 September 2010 (UTC)

Mike, I think you are setting me up for a is a trick question here. Not all of the articles are about lists per se, so a simple yes or no is not appropriate.

• The Clements Checklist of Birds of the World is an article about a book;
• Fortune 500 is about the Fortune 5000 lists in general, i.e. it is an article about a category of list that is published every year (see Case 4 above);
• Nikkei 225 is a bit of a mess (see below);
• The 500 Greatest Albums of All Time is a list article;
• Who's Who in Australia is an article about a book.

The article on the Nikkei 225 for all intents and purposes is a list article, but I am not sure it should be. My understanding is that, like all stock market indices, the membership of the list is reviewed regularly, and members may be dropped suddenly if they no longer meet the inclusion criteria. This article therefore has a problem: it is tying to be about a category of list (lets call it "Nikkei 225" in general, focused on the significance, history and development of this list category) whilst at the same time listing the components of the list for the year 2009/10 (lets call it "List of Nekkei 225 companies 2009/10"), so it is also contains a snapshot of the index at a particular point in time. The list for 2009/10 is embedded in the article, but I don't think it should be. The list should be split out if there is evidence that the 2009/10 list is notable, or dropped if it is not. Note that the contenst to the list should be listed in the Category:Nikkei 225, but for some reason that has not happened.

I am not sure what you are trying to prove, if anything, but in each case a reliable source is need to tell us what the article is about (i.e. what is the topic) and a further sources are needed to provide evidence of notability in each case. --Gavin Collins (talk|contribs) 18:44, 1 September 2010 (UTC)

No, reliable sources are not needed to tell us what any article is about. Editors choose what the (proposed) article is to be about. There are zero independent reliable sources that say "Wikipedia should have an article on this subject." There are zero independent reliable sources that say "Gavin thought he should be writing about the Heroninos Archive, but actually, I'm going to insist that the subject of the page he created be Predatory birds in fictional universes!"

Editors, not sources, are the ones who say "(I believe that) Wikipedia should have an article on this subject." It happens that if sources don't exist for the subject, then the community will (usually) refuse to keep the article -- but this does not change the fact that it is the editor who takes the action of selecting the subject.

Put another way: Sources are passive objects. Editors are not. Choosing a subject is an action, not a fact. WhatamIdoing (talk) 19:04, 1 September 2010 (UTC)

Part II of the question remains somewhat unanswered (notability of these lists is not part of the question). These are all articles about lists, but lists that take different forms. For:

• The Clements Checklist of Birds of the World
• Who's Who in Australia

You claim they are merely articles about books and not list articles, therefore (I presume) WP:SAL doesn’t apply.

For: The 500 Greatest Albums of All Time an article about a magazine story, you claim that the article is a stand-alone list article and (I presume) WP:SAL applies.

For: Nikkei 225 an index of the top 225 stocks in the Toyko Stock Exchange and Fortune 500 a specific list published by Fortune magazine you waffle a bit. Are they standalone lists or are they articles?

So you gave two No, one Yes and two maybes. What distinguishes The 500 Greatest Albums of All Time as a standalone list article from The Clements Checklist of Birds of the World and Who's Who in Australia for the purposes of applying WP:SAL. The only difference between these articles that I can discern is that two of the lists are in book format and one is in magazine format. None of the three actually include the subject list in the articles. The articles just discuss the contents of the list in general. Two of them included embedded lists of various highlights of the overall list. So I guess the burning question is when is an article about a List not a List article and not subject to WP:SAL?--Mike Cline (talk) 19:11, 1 September 2010 (UTC)

Gavin, if you believe that format is irrelevant, then you've got a completely different definition of "list article" from everyone else. WP:LIST says: "List articles are encyclopedia pages consisting of a lead section followed by a list (which may or may not be divided by headings)." I don't think you'll find anyone else who believes that Periodic table is a "list article". If Periodic table were to obtain Featured status, it would be as a Featured Article, not as a Featured List.

Format is key to determining whether an article is a "list article". -- Quiddity (talk) 02:44, 2 September 2010 (UTC)

Forgive me for being so ponderous, but there is a lot a mud being thrown at me, whilst at the same time there is no acknowledgement that what I am trying to explain to you about lists is either embedded in existing policies and guidelines, in third party sources (e.g. Set theory) or is just plain common sense. If we are to work towards a shared understanding of this issue, we need to acknowledgement that there is merit in each editors viewpoint.

1. In answer to WhatamIdoing, of course editors have the discretion to create or delete articles, but there is no denying that a reliable source is needed to tell us what the article is about, for going back to WP:BURDEN which says "f no reliable third-party sources can be found on a topic, Wikipedia should not have an article on it." I think you will agree that this applies to lists as well.
2. In answer to Mike, you are wrong about The Clements Checklist of Birds of the World and Who's Who in Australia: these are definitely articles about books. The fact that the books might contain lists suggests that they could be used a source for compiling lists, but the absence of a list in the article means that we can say they are definitely not list articles. Even though the books themselves contain lists or are about lists, these articles are about the books as publications. They are no different in this regard from, say, Encyclopaedia Britannica, whose content is effectively organised into a list or series of lists, but the subject of the article is the publication.
3. In answer to Quiddity, to some extent format matters in the sense that a list is always going to be a tabulated column of elements, even if it is only one rather than multiple columns - all of the this is explained in WP:LIST and WP:SAL. The article on the Periodic table is very definitely about a list (arguably the most important list in history), but whether you want to categorise it a list or a non-list makes no difference to how we treat its content and its topic: they are both governed by Wikipedia's content policies and inclusion guidelines. Format is not relevant in the sense that, even if the prose exceeds the list by a multiple of say 100x, the focus of all that prose is still the list itself, its definition, history, development, and the commentary about its significance. To some extent, the Periodic table is knockout proof that list topics should not be treated any differently from topics that are not formatted in list form, and is also proof that the topic of a list article is the list, not how its elements are categorised.

Stepping back from the detail a moment, I can see that there is a reluctance to recognise that a list is just one way to organise content around a topic, but that does not mean that the subject matter (the topic) or the content should be treated differently from any other article. Barebone lists contain very little in the way of content that can be used to define their subject matter, and some editors have seen this lack of content as reason to treat lists as if they are not bound by content policy or notability guidelines. If we can agree to dismiss this line of thinking, then I think we will find there is a way forward to reconcile our views. --Gavin Collins (talk|contribs) 07:13, 2 September 2010 (UTC)

Wow - I have not read all the above, I will later, but Gavin, you said: "I think you need to recognise that Nixon's Enemies List is a bona fide list article" - This is the CORE of this endless debate it seems to me. That article is not (to my and other's minds) a list article. So I posit a working definition: A list article is an article in which the primary content is a list compiled by wikipedia editors. To avoid a possible argument about OR, I mean this in the same sense that a non-list article is similarly compiled by wikipedia editors. ‒ Jaymax✍ 08:26, 2 September 2010 (UTC)

I see where you are coming from, but I suspect your working definiton does not apply in every case. For argument sake, lets set the scene and say a list has four components:

1. a title;
2. a defintion;
3. the list itself;
4. additional coverage, such as commentary, criticism and/or analysis.

In the scenario which I think Jaymax has in mind when he wrote his working definiton relates to a barebone list, in which the defintion and commentary are minor components of the article. In this scenario, the list makes up the bulk of the article's content. I think this is what Jaymax had in mind when he says that a list is an article in which the "primary" content is a list.

However, conisder this scenario: as the depth of coverage increases, then:

• the volume of content contained in the commentary section may, over time, exceed that volume of content contained in the original list;
• as the article develops, the article changes from being a barebone list to something more substantial;
• despite that fact that the article is now "primarily" made up of commentary, the subject matter has not changed: its subject matter is still the list.

These are characteristics of lists that make it to FA status; they contain more commentary than a barebone list, but they are still list articles. A good example of this scenario List of Smithsonian museums, where the volume of commentary exceeds the volume of the list itself.

So my response to Jaymax is to consider the followsing: it makes no difference whether the list is the "primary" content; the fact that other components of the list article are less than than or greatly exceeed the size of the list makes no difference to whether or not an article is a list article. I think Jaymax would recognise that commentary is important, and regardless of whether a list is the subject of a small volume of commentary, or even large volume of commentary (such as the Periodic table), that does not mean that its nature has changed in any way - the article is still about the list, and the topic of that list is provided by its defintion. --Gavin Collins (talk|contribs) 09:30, 2 September 2010 (UTC)

(edit conflict)

Okay - for clarity, when I say compiled I do not in any sense mean replicating a list that is sourced (or potentially, a subset thereof - but I'd have to think about that more.

It seems from the debate above that existing definitions of list in guidelines etc may not distinguish between lists that are compiled by editors, and those that are extracted from a singular master source. This (to my mind, and if true) is daft - there is blatant qualitative distinction, and it's hard to see how how any guideline could serve the encyclopaedia well without recognising the distinction. ‒ Jaymax✍ 09:47, 2 September 2010 (UTC)

I do see the distinction and I think this is the difficulty lies when it comes to reaching a shared view about list inclusion criteria given these two conflicting perspectives:

1. I would argue that the qualative disctinction can be negative, i.e. lists that are compliled by editors without reference to a master source are effectively original (primary) research. Of course this cannont be proved, and herein lies the locus of the dispute, as whether or not a list has been madeup is a matter of opinion, not fact, and there are quite a variety of opinions on this issue.
2. The alternative viewpoint, as I understand it, is more positive: that list compiled by editors is useful, e.g. homemade lists can be useful for navigation.

In some ways I can undertand why navigational list are useful (see the more about this in the detauled discussion at WT:NOT#Linkfarm), but my reservations about primary (original) research lead me to disagree with this premise. To avoid WP:NOT#OR, I think there should be clear distinction between navigation pages (WP:CATEGORY) and mainspace content pages, in which case navigational lists have no rationale for inclusion. What ever view you have, I think you will see that this issue is yet another bone of contention, because a list can have a dual role (being both content and a navigational aid) and which takes priority is, again, a matter of opinon, not fact. --Gavin Collins (talk|contribs) 10:16, 2 September 2010 (UTC)

(edit conflict)

Gavin, I said in my bit at the top of this section that I thought guidelines for lists should (mostly) apply regardless of whether it is a list article or an embedded list. But a large part of this debate surrounds notability - what has the right to be an article. In the case of the Smithsonian Museums article, I would comfortably say that "Smithsonian Museums" would be at least an equally appropriate title, and I don't see it as a list article per se, because the primary content is not the list - it may have been once. The case of a list article gaining extra content until it is no longer a list article, is somewhat comparable to the case where a list in an article becomes so expansive that it is forked off into it's own list article. ‒ Jaymax✍ 10:21, 2 September 2010 (UTC)

I would disagree with you, becasue the scope of the topic "Smithsonian Museums" is already covered by the article Smithsonian Institution, which is the umbrella organisation (or over arching category) which the museums fall under. In my view, the list is a notable list topic in its own right, and as you can see from Smithsonian Institution#Smithsonian museums, the list is treated as sub-topic as you suggest. --Gavin Collins (talk|contribs) 10:36, 2 September 2010 (UTC)

(edit conflict) the third - I'm gonna quit for a bit now.

Gavin, replying to your subsequent comment, I'm going to paraphrase you somewhat liberally to hopefully make a point clear:

articles that are compliled by editors without reference to a single master source are effectively original (primary) research. Of course this cannont be proved, and herein lies the locus of the dispute, as whether or not an article has been madeup is a matter of opinion, not fact, and there are quite a variety of opinions on this issue.

The alternative viewpoint, as I understand it, is more positive: that articles compiled by editors are useful, e.g. homemade articles can be useful for an encyclopaedia.

I am convinced that the correct analogy (broadly) is to compare a singular list item is to a single fact (or para or aspect) in an article, and that it doesn't make sense to view the entire list in the context you present. (again, I'm focussed on editor compiled lists here; and in no way limiting that just to navigation lists) ‒ Jaymax✍ 10:44, 2 September 2010 (UTC)

The problem with this analogy is that it conflicts with generally accepted Set theory, which I believe you would have been taught when you first entered secondary school. If a list can be likened to a mathematical set, then a list is an topic in its own right, i.e. it is more than just the sum of its singular list items. Like sets, the subject of list is determined by its defintion that determines which items are included in it. --Gavin Collins (talk|contribs) 11:15, 2 September 2010 (UTC)

FWIW - I majored in pure math, and while you're right in essence, the analogy you try to draw from that is fallacious. Article content is likewise a set. If we titled all our articles 'Article about X' it would be apparent that the distinction you are trying to draw does not really exist. We don't so title non-list articles, because to do so would be linguistically dumb. ‒ Jaymax✍ 12:15, 2 September 2010 (UTC)

I suppose a mathemation could extend set theory to non-list articles too, except it would not be as clear what the equivalent of a singular list item in a non-list article. The analogy is good in the sense that it serves to identify the defintion of list to be the list topic. Looking at all the other ways a list topic can be defined, they seem to imply the defintion does not need to be sourced in anyway, which raises my suspicion as to their validity. I don't think there is any other way to define how a list topic is compiled, or to identify what makes two simialr lists different, I think the analogy that a list is a simple example of a set is good for now. --Gavin Collins (talk|contribs) 12:36, 2 September 2010 (UTC)

The continued references to mathematical Set theory are not useful in this discussion. As interesting as it might be as an analogy in a list related discussion, it is absolutely irrelevant to any change the WP list related guidelines. Our current practice is based on the consensus that list style articles are indeed encyclopedic and those guidelines have resulted in 1000s of good lists that meet those guidelines and remain in the encyclopedia by consensus of the community. None of that consensus nor our guidelines relative to lists has anything do with Set theory. I trust we don't go a path of policy development based on mathematical or any other theoretical construct. This is an encyclopedia. Our guidelines need to encourage and support the creation and improvement of good encyclopedic content regardless of the format it takes.--Mike Cline (talk) 20:30, 2 September 2010 (UTC)

I totally agree - it's of academic interest to a math geek, but is not useful. All I can say in response to Gavin's "definition does not need to be sourced in anyway, which raises my suspicion as to their validity." is that for any article, the scope of the article is not sourced, only the topic. An article is "Collection of info about X", where X is the topic. All a list article is, is a constriction of 'info about X' to where the 'info' is of subtype 'instance of'. ‒ Jaymax✍ 22:25, 2 September 2010 (UTC)

I agree entirely, becasue this restriction is what makes a list topic distinctive in its own right. I think set theory is a useful analogy in this regard, wheras what is a "good" or "bad" list is just a matter of (unsububstantiated) opinion. I think it is convenient to presume that list topics do not have a definition, yet such a definition is required by WP:Source list. This is not a new concept by any means; it is just one that some editors would like to brush it under the carpet because they would rather not have to go to the effort of finding a reliable source to provide evidence of notability.

Standing back from our discussions for a moment, you would think that list are the centre of the earth when it comes to encyclopaedic articles. But I think their importance is highly over rated, in the sense that a few sentences from a reliable secondary source is as good as or much better than a list that provides no context to the reader. If list topics are so useful or valuable as has been claimed, then I don't understand why there is such strong resistance to sourcing article topics. --Gavin Collins (talk|contribs) 23:56, 2 September 2010 (UTC)

Gavin, I re-read WP:Source list. Firstly, your comment wanders from the need for a definition, to reliable source, to notability. It is not clear to me what exactly you are saying. I cannot see where the MoS section you linked comes anywhere close to stating that a definition of the list is required other than "discussed on the talk page in order to attain consensus". It states list items must be sourced (just like all content). Indeed, the title at that link is listed items.

As for the importance of lists - it is only, perhaps, because you are determined that the definition be RS (and perhaps N) [despite, I would say, both consensus otherwise, and strong logical argument that doing so would cause lists to be treated differently to regular articles] - that this has become so major.

I do think list guidelines etc are due some tweaking to make things clearer and easier - but I am of the opinion that it is what you are striving for that would require a significant change to existing policy. ‒ Jaymax✍ 00:11, 3 September 2010 (UTC)

I think your reading of the guideline is too restrictive. If a list has to contain a definition, then that is list content as well, and that too has to be sourced. --Gavin Collins (talk|contribs) 05:48, 3 September 2010 (UTC)

Um, no - list definition != list content, emphatically so when the section heading is list items. The guideline section you linked is quite clear, and does not come anywhere near saying what you are suggesting. ‒ Jaymax✍ 06:12, 3 September 2010 (UTC)

Let me quote it to you, just to be clear: Lists, whether they are embedded lists or stand-alone lists, are encyclopedic content as are paragraphs and articles, and they are equally subject to Wikipedia's content policies such as Verifiability, No original research, Neutral point of view, and others. --Gavin Collins (talk|contribs) 06:33, 3 September 2010 (UTC)

Yah, I read it. It's talking about list items. Taking stuff out of context doesn't help you demonstrate that your argument holds up - rather the reverse. ‒ Jaymax✍ 06:49, 3 September 2010 (UTC)

Moving this here from project page 'cos doesn't belong there, and doesn't deserve a new section or anything...

I just went through Wikipedia:NOTINHERITED a few times, because I see Gavin relying on it often. I want to point out that, as worded, it doesn't really apply to this debate at all. It deals with whether notability of X falls through to AN INSTANCE of X, or ACROSS distinct instances of X, whereas what we are discussing here is whether notability of X applies to a list (complete, or of discriminated examples) of instances of X, which is quite different. ‒ Jaymax✍ 07:30, 4 September 2010 (UTC)

To be fair, there are some lists which both are notable as lists, and where the list makes up such a significant amount of the article that it could be considered a list article, List of air carriers banned in the European Union is an example. However, these are the exception rather than the rule. On the other hand, an article can be about a list without containing the list, and that would not be a list article. An example is one of my articles: Register for Governmental Approval of Financial Obligations where the frequent updates made to the registry made me prefer to just link to the external (official) website for it. Sjakkalle (Check!) 13:34, 3 September 2010 (UTC)

Exactly. Wikipedia is not simply a mirror site for lists. There has to be some evidence of notability. --Gavin Collins (talk|contribs) 23:55, 6 September 2010 (UTC)

Side discussion

I think a fair question to ask at this point is if there is anyone else that agrees with Gavin's position that may be reading this? Right now, Gavin's view seems like an extreme minority position, and if we are spending all this time to try to satisfy one view when consensus is somewhere far away, something is very wrong here. If there are more people that are interpreting list articles in the same way Gavin is, then yes, we need to work on that more so that we can find that consensus. However, I am just not seeing that being the case right now. --MASEM (t) 13:09, 2 September 2010 (UTC)

That is not a fair question, it is simply an Ad hominem circumstantial attack and is out of order. The subject of this discussion is Jaymax's proposal, but if you want to launch personal attacks against me, please direct them to me at my talk page. Consider this a warning. --Gavin Collins (talk|contribs) 13:21, 2 September 2010 (UTC)

It is a completely fair question: if you're the solitary standout against Jaymax's proposal (which seems everyone else is cool with to build consensus on), and you're demanding we bring the consensus to you, that's improper consensus-building and tedentious editing, and we need to move on so we can all get back to editing the actual work. But if there are others that agree with your position, but having spoken up yet, then we need their input to know that we need to find a way to reconcile these positions. --MASEM (t) 13:31, 2 September 2010 (UTC)

Consider this a second warning Masem. The discussion is not about me in fairness. We can discuss this point at Wikipedia:Wikiquette alerts if you continue to attack me personally. Please take this to my talk page or cease as a matter of courtesy. --Gavin Collins (talk|contribs) 13:37, 2 September 2010 (UTC)

I think everyone needs to calm down. Masem said IF you're the solitary standout, you would be in a minority position. If you're not, you're not. So let's see if anyone speaks up. Consensus is not the same as unanimity and sometimes we have to move on despite one person's viewpoint. Shooterwalker (talk) 15:13, 2 September 2010 (UTC)

I have to ask Masem to clarify what he means by "Gavin's position"?... Gavin has taken positions on a number of issues discussed on this page, some of which I agree with, and some of which I disagree with. Blueboar (talk) 22:27, 5 September 2010 (UTC)

Before Gavin added the header, this section was part of the previous one (edit: Jaymax's proposed question to Gavin) w/o a header specifically more on what WP considers a "list" article. That's the only position that I specifically address there. --MASEM (t) 00:56, 6 September 2010 (UTC)

Examples of implementation

I'm not aware of the practical implications of policy discussions, but this is something that's been discussed before. On the AfD of this list, there where general agreement that the topic was non-notable, but the list nevertheless should be kept. Should that then be deleted according to the new consensus? Sandman888 (talk) 08:23, 4 September 2010 (UTC)

I'm not sure how I would have voted before this debate, but after thinking about lists a lot of late, I would have def voted delete, because the topic X is not notable. The content rightly belongs in the List of Nobel laureates by university affiliation article - and the correct article being poorly formatted should never be an excuse to retain. Thanks for presenting this case - it raised some interesting thoughts. ‒ Jaymax✍ 10:08, 4 September 2010 (UTC)

List length and WP:SIZE

Cut'n'paste from discussion at, Wikipedia talk:Notability.

Intriguingly (well to me, anyway) WP:SIZE and WP:SPLIT is the one area where I can see some sense in real special handling for lists - Take for example List of Nobel laureates. This list would be improved (better for the encyclopaedia) with the addition of a short summary and perhaps even a pic for each laureate. However, that could well push it over a reasonable size limit. The notability of the topic is clear, the value of the list is clear, the encyclopaedic improvement from additional content and context for each entry is clear - but perhaps the list would then require splitting into different articles on an arbitrary basis (by year? by discipline? by surname?). It would be a wrong to force notable, lengthy and useful lists to be less encyclopaedic due to mechanical size constraints. ‒ Jaymax✍ 18:06, 4 September 2010 (UTC)

The point being that if List of Nobel laureates by university affiliation was formatted per List of Nobel laureates affiliated with Princeton University it would almost certainly go well over size, even though it would be better for the encyclopaedia - which would force arbitrary breaks (affiliations A-F, G-O, P-Z or whatever). Now compare that with an in-article list (non-list-article), that is a good, useful, encyclopaedic list, that grows to the point where it pushes the article over length. It makes sense to fork out the list into a new article, for purely technical reasons. So while the technical (and usability) constraints of WP:SIZE add value to a 'prose-based' article - forcing (eg) careful consideration of which significant sub-topic is the most notable so it can be spun out, or (via consensus) where there might be unnecessary bulk of little encyclopaedic value in an article; the same cannot readily be applied to lists. If the consensus has determined what makes for good encyclopaedic content (items, and item data eg: narrative, photos) in a list, and the list gets to big - then what?

Of course, if a list "list of X" is a valid article if the topic X is notable (per consensus I think), then the issue goes away - but there is scope for us to explicitly recognise that in WP:SIZE (in currently acknowledges "Two exceptions are lists and articles summarizing certain fields." but that is not quite the same as acknowledging that "prose facts about X" and "list of instances of X" might well belong in separate articles for size reasons, or that "list of instances of X" might itself need multiple articles to do the encyclopaedia justice. ‒ Jaymax✍ 07:07, 6 September 2010 (UTC)

{note to MediaWiki devs - dynamically user sortable and groupable giant tables when you can; thanks} ‒ Jaymax✍ 07:07, 6 September 2010 (UTC)

True... on the other hand, splitting off a list to conform to WP:SIZE can sometimes result in a list that has been from its context (often resulting, intentionally or unintentionally in a POV fork). Sometimes it is better to ignore WP:SIZE. Blueboar (talk) 13:27, 6 September 2010 (UTC)

One solution that I know has been done is what we did at 2010 in downloadable songs for the Rock Band series (and yes, this has been through an AFD wringer too, to demonstrate the appropriateness of the list); note there are 3 other similar lists for 2007-2009. There is also Complete list of downloadable songs for the Rock Band series, which includes all 4 lists through transclusion and clearly is far beyond SIZE (and generally warned indirectly to the reader that it will be large to avoid "OMG ITS HUGE" comments. We chose a format that captured the basic details for each song (possible a problem with the Princeton Nobel alum list), and then set a default sort and list break criteria on the same metric, that being the year of release. In the Nobel case, I would certainly say that default-sort by year and splitting by year by quarter centuries would allow for a format like the Princeton version to be used.

But another comment there: I see at least three different ways the Nobels are presented. A short list that just gives years and names across all 5 awards (List of Nobel laureates); a version similar to the Princeton one but specific for each award (List of Nobel laureates in Chemistry) but then you have all these other weird cross-categorizations, like List of Nobel laureates by university affiliation, List of Nobel laureates by country List of Nobel laureates by country per capita, etc. At some point this is excessive list making just because the list can be made, as most of these can be evaluated if there was a master list with all the details on it ala the Princton format. (it would also eliminate the need for the Princeton list as a separate article as long as sortable tables were used). You would still have a very large list and would need sublists and transclusion to make it work effectively for some readers, but its very possible to do it. --MASEM (t) 13:46, 6 September 2010 (UTC)

I like the tenor of Masem's point above in this regard. Indeed, complex subjects with lots of relevant elements do lend themselves to all sorts of lists that slice and dice the subject. Purely from a practical, usefulness measure, lists about Nobel laureates make perfect sense when organized by field, by university, and by county. If I were a high school or college student researching something about the Nobel Prize, these breakdowns might be very useful. The last list mentioned--List of Nobel laureates by country per capita begins to stretch on the basis of practicality (for the most part, because the prime discriminator--population is so volatile.) Indeed all the elements in these lists could be combined in a sortable list that would allow the reader to selected the relevant relationship elements they need. Yes, today WP:SIZE is still a constraint. What I really like about Masem's comments is this. He's not talking about these lists as good or bad, but instead is talking about ways that we can improve WP content. All the content in these lists is good content, we just need to find more creative ways to encourage the compilation of this good content into better articles and lists. What isn't required, and should be discouraged, are guideline and policy efforts that discourage that creativity and inclusion of good content for the sake of preventing (something that rarely happens anyway) the occasional flawed list or poor content. Our policies should encourage good content and good articles, while maintaining the minimum necessary mechanisms necessary to cleanup the marginal stuff.--Mike Cline (talk) 14:15, 6 September 2010 (UTC)

I see WP:SIZE as connected to the notability guideline and the guideline on summary style. Wikipedia is not paper, but there's only so many kilobytes and articles you can have before you screw up the noise-to-signal ratio. You have to prioritize. And if you spin something out or divide an article into two, you have to prove that it's notable in order to justify having more coverage. Otherwise you're adding to the noise instead of the signal. There's no absolute limit to the number of articles or lists we can have, but there is a limit in terms of quality. Shooterwalker (talk) 16:27, 6 September 2010 (UTC)

The way I see WP:N, WP:SS, and WP:SIZE is to imagine that WP is capable of being an infinitely large book allowing for articles to grow to as many pages as needed to be comprehensive. It is still trying to be a reference book, so signal-to-noise is very important; thus we would still only have topics that are notable, and a notable topic cannot be filled with minutae like trivial pop culture references. What is "comprehensive" does need to be a consensus-based discussion as otherwise you make walled gardens (as we had in the past with fiction, and likely how we have now with many topics on mathematics). But the point is this - as long as a list or table would be considered part of a comprehensive article of any size in a printed reference work, there is zero reason why it should be exclude from Wikipedia the electronic version. Furthermore, it is necessary to understand the electronic version will have things that can't be done easily in print; thus the need to understand where SIZE, SS, and Notability need to overlap and interact. Same is true with large lists; within a printed version it's easy to cross-categorize the large list, but not the same w/in the electronic version. So we have to come to some means to implement such lists to make them appropriate reference sources without diluting the information. So a question to always ask would be "how would this be done in print" or even "would this be done in print"?, leading to question why we have a list of Nobel Laureates by capita - I cannot see that being part of a Nobel article in a printed work. --MASEM (t) 16:42, 6 September 2010 (UTC)

I don't have much to add that I didn't already say. But I also see a relationship between WP:SIZE and WP:IINFO. A discriminate list would probably fit within 100kb. Many (but not all) lists that are more than that are likely something inherently unmanageable like list of virgins. (Although even list of Japanese virgins would be an issue.) Shooterwalker (talk) 22:53, 6 September 2010 (UTC)

I'm certainly not trying to discount our avoidance of being indiscriminate. Would you expect "List of virgins" to appear in any printed reference work? That's an easy first pass to say if we're just including a list just because we can. (that is why, in part, notability does come into play in this evaluation - if there's no aspect of the list that is notable - the list itself, its definition, its main topic, or any of its entries) why are we including it? --MASEM (t) 23:11, 6 September 2010 (UTC)

I don't understand why you are ignoring WP:AVOIDSPLIT. You just can't dump a load of non-notable stuff in a list. Instead it should be summarised. After all Wikipedia is not a list of everything that existed or exists. --Gavin Collins (talk|contribs) 23:52, 6 September 2010 (UTC)

Please tell me exactly what language of this: Editors are cautioned to not immediately split articles if the new article would meet neither the general notability criterion nor the specific notability criteria for their topic. Instead, editors are encouraged to work on further developing the main article first, locating coverage that applies both to the main topic and the subtopic. Through this process, it may become evident that subtopics or groups of subtopics can demonstrate their own notability and help justifying splitting off into their own article. If information can be trimmed, merged, or removed, these steps should be undertaken first before the new article is created. bars the creation of non-notable list articles that would otherwise be included in a larger article if SIZE was not a problem. "AVOIDSPLIT" does not say "Never split", it says "avoid splitting without trying other options". --MASEM (t) 00:13, 7 September 2010 (UTC)

Ah, a guideline that explicitly puts the burden on the splitter to improve the main article would be welcomed. The lists of "list of nobel laureates X" has become something of a joke, but any of the sub-list are surprisingly resistant to deletion. Should that be added/made more clear at NOTSPLIT? Sandman888 (talk) 09:30, 7 September 2010 (UTC)

Summarising the discussion

Is it about time to close this discussion? I think an uninvolved admin should be able to look through the main page and draw out some principles, even if there are outstanding issues that have no consensus. Then we can discuss a next step. Shooterwalker (talk) 01:10, 8 September 2010 (UTC)

I think that the most contributors to the RFC believe that the inclusion of lists is a matter of editorial discretion & judgement:

• Based on the premise that editors should decide which lists are useful to the reader;
• Editors know when comprehensive coverage of a general topic at detailed level is needed;
• Readers need navigational lists, list summaries (outlines) and lists of lists;
• Lists with a clear focus and a defined scope are needed by the reader.

Although these views are entirely reasonable, it is difficult to reconcile the idea that editors should have greater discretion over creating lists articles than non-list articles, given the prohibition on original (primary) research. --Gavin Collins (talk|contribs) 14:52, 8 September 2010 (UTC)

Given that notability on Wikipedia itself is a form of original research (we don't seek sources that assert the statement that "X is notable", but instead that there are sources that describe X in-depth, and the original research coming from when we consider there to be enough sources to claim that), it is difficult to see the problem here. Editor must engage in a minimal amount of original research in assembling articles and content, and that's the application being used here. --MASEM (t) 15:11, 8 September 2010 (UTC)

I understand what you are saying Masem... but I have to disagree with how you say it... establishing notability is a form of research, yes ... but it is not Original research as Wikipedia uses that phrase. Blueboar (talk) 17:50, 8 September 2010 (UTC)

But neither is saying "these elements form a list that is part of a notable topic". That is, there is WP:OR that applies strictly to content to prevent people from inserting their own questionable expertize, and then there is "original research" that is necessary for collecting sources and grouping information into cohesive topics and information; all of our policies and guidelines for building the encyclopedia are this type of acceptable original research. The point is this: what Gavin is trying to assert above - that it is "original research" on assembling a list article where "List of X" itself is not a notable topic - doesn't jive with how WP:OR is used normally. --MASEM (t) 21:06, 8 September 2010 (UTC)

"Original research" refers to facts and statements which are unsupported by sourcing, not the way we choose to write articles. Apart from direct quotations, everything on Wikipedia is original whether they are prose articles or lists; that is a requirement to avoid running afoul of copyright law. These original articles should be well-founded in sourcing, so that the article's content can be verified. An original article based on research is not the same as original research. Sjakkalle (Check!) 11:58, 9 September 2010 (UTC)

That's probably what I was trying to say, more succinctly. --MASEM (t) 12:53, 9 September 2010 (UTC)

Thanks for acknowledging that consensus is somewhere looser than your preferred position. I know it's looser than mine too. But I don't think that people are arguing to make it totally discretionary. There's a lot of consensus to avoid totally original topics for lists like "list of fictional characters with the name john". I really do think we should get an independent admin to close the discussion rather than arguing about what the consensus is. Shooterwalker (talk) 19:48, 8 September 2010 (UTC)

I think a more correct statement is that the consensus in the RFC is looser than is permitted by Wikipedia core content policies. For this reason, I don't believe that list inclusion based on what editors think are useful to the reader is a feasible proposition within the context of Wikipedia's existing framework of policies and guidelines.

The problem as I see it is that if a list has not been published, then it is not verfiable. Going back to what Blueboar said earlier, the issue being discussed here is the inclusion of list articles in Wikipedia, not the inclusion of individual items within a list. It is difficult to understand how any list that has not been published by a reliable source is in anyway compatible with core content policy. The idea that editors' personal ideas for list topics, editors' interpretations as to what the reader wants, or editorial opinions as to what is useful do not have a place in Wikipedia.

Regardless of whether Masem thinks an list is or is not original research, it is impossible to prove it is not original research if it has not been published. Herein lies a difficult problem. --Gavin Collins (talk|contribs) 11:20, 9 September 2010 (UTC)

As long as there are reliable sources to show that elements of a list meet its definition (either per element or the list overall), it is far from the type of Original Research that we disallow. As Sjakkalle points out, the choice of a list's definition is not the same type of Original Research as unsupported claims of list membership; and in fact the type of "original research" that we need to engage in in the first place to summarize sources for the work. Of course, the definitions could be poor themselves, involving inappropriate cross-categorizations, overly indiscriminate and broad membership, or overly trivial and excessively narrow membership, but those are problems that other policies/guidelines (specifically WP:NOT) deal with, and not itself an issue with being Original Research. --MASEM (t) 12:53, 9 September 2010 (UTC)

Pure OR like List of Presidents with facial hair or List of fictional characters who can manipulate fire would still get deleted. There's a middle ground that you're not seeing that appears to have a consensus. But once again, it's best to leave this to an independent admin. Shooterwalker (talk) 15:03, 9 September 2010 (UTC)

If you're certain that's "Pure OR", then you might like to take List of United states presidents with facial hair during their tenure to AFD. But you might want to look at sources, like the sources named on the page and others before you incorrectly assert that none exist. WhatamIdoing (talk) 21:06, 9 September 2010 (UTC)

Wow. I'll just leave that one alone. But my overall principle still stands. You can't just make up a set of things with your own criteria. If a third-party has talked about a group of things that all meet some criteria, that's different. Shooterwalker (talk) 19:55, 11 September 2010 (UTC)

SW – things are a bit slow in this discussion, so here is a question re WP:Madeup. Take notable subject X. There is a robust article on X in WP. There are a great many books written about various aspects of X. There are way too many books to include them all in a further reading section of the article on X. So if an editor created a standalone list article entitled Bibliography of X with clear inclusion criteria that list entries must be books about the subject of X would you consider the list as a WP:Madeup topic if either of these conditions existed:

• No reliable source has ever published or commented on a Bibliography of X as a standalone reference (not a bibliography at the end of a reference).
• Reliable sources on the subject of X or elements of X contain Bibliographies listing books relative to the subject of the source, but not necessarily all inclusive of X.

To rephrase the question another way. When would a Bibliography of X be considered a madeup topic when it contained a list of books clearly related to the otherwise notable subject of X? --Mike Cline (talk) 20:17, 11 September 2010 (UTC)

• This is a tough issue and really getting too specific to be useful at this point. I think this is really more of a question for WP:NOT. The closest thing I can see for that is something like list of songs in Rock Band. It's not a list I'm crazy about. But it's well done and there's enough community practice around it that I'm willing to let it work itself out. I'd feel the same way about creating bibliography articles on Wikipedia -- every guideline tells me it's a bad idea, but if the community practice is a good one, then maybe you can make some sense of it. ... at this point I would prefer to focus on high level issues. Like, do lists have to be notable, and what kinds of stuff would a third-party have to cover. In the long run, we could start to work out specific exceptions or clarifications. But only in the long run. Shooterwalker (talk) 20:25, 11 September 2010 (UTC)
  • • Re: the Rock Band list, a few traits pop out to me as relevant to why it's appropriate for Wikipedia. First, Rock Band is indisputably a notable game. We would not suffer a list of songs featured in a non-notable video game. I don't give a whit about the subject matter (I don't play video games at all), but dislike is obviously not relevant. Second, Rock Band is a game that is about playing music; therefore the songs that are played are unquestionably relevant, and integral, to the notable article topic. Third, there is a finite number of songs featured in Rock Band; the list you link to is only for the 58 songs featured on the game disc itself. Fourth, a supermajority of the songs are themselves notable (I presume by the bluelinks), and all but a few are by notable bands. Fifth, whether a song was included on the game disc is verifiable and NPOV, so there is no controversy over a song's inclusion in the list. Not all of those points are necessary in my view for this list to be valid (and some additional arguments for its validity were made in its AFD and a related AFD), but the list's inclusion on Wikipedia should be unquestionable given all of those points together. postdlf (talk) 21:52, 11 September 2010 (UTC)
      • Postdlf's statement above - well-stated - is a good reason to make sure we distance ourselves from WP:IDONTLIKEIT-type arguments. We are looking to build soemthing more than a basic encyclopedia, personal dislike or preference for a topic should not come into play as long as the end approach is something that can be held consistent across all topic fields. Hence why the language that seems to be emerging from this RFC seems completely appropriate for most situations. --MASEM (t) 21:57, 11 September 2010 (UTC)
  • You said above [bibliographies]]...every guideline tells me it's a bad idea. I interpret that to mean in a broad sense that even of a list topic is notable, there is clear and discriminate inclusion criteria, and list entries can be verified against that criteria, there are still some broad list topics that are unsuitable (and would always be unsuitable) for WP. Where would we spell out that policy? Where would we say A list of people associated with a notable topic is OK, but a list of books about the same topic is not. A list of events associated with a notable topic is OK but a list of organizations related to that topic is not? I think is extremely unproductive to try and legislate against one kind of list and allow others. If the list topic is notable, inclusion criteria is clear and discriminate and entries are verifiable against that discriminate criteria, then it shouldn't matter what the list is about--events, people, books, music, organizations, et. al. and our guidelines should be clear on that.--Mike Cline (talk) 20:50, 11 September 2010 (UTC)
    • The only other criteria I would add to Mike's clear points is, "is this list now stand-alone due to SIZE/summary style issues?" to the equation. If its a small list that can fit in a small article, regardless of notability or other facts, it probably should be included in the larger article. --MASEM (t) 20:54, 11 September 2010 (UTC)
    • A list of books about X should be uncontroversial if it only includes books that are notable and/or by notable authors, and if it's verifiable and NPOV that X is the subject matter. List of books about Israel should qualify (though appropriate for sublist organization). List of books supporting the Israeli occupation of the West Bank probably not. postdlf (talk) 21:18, 11 September 2010 (UTC)
      • I agree with you 100% because the list topic: Supporting the Israeli Occupation of the West Bank is POV and probably not notable as a topic in its own right. Whereas a Bibliography of the Israeli Occupation if the West Bank is fine because Israeli Occupation of the West Bank is a notable topic.--Mike Cline (talk) 21:35, 11 September 2010 (UTC)
        • I imagine there probably is significant RS coverage of support for the West Bank occupation as a (meta)topic; notability isn't the question there. I would imagine such a list would be more of a problem because of the intractability of determining whether a given book supports or doesn't support something, when given as a binary statement of fact. Such a list would probably become bogged down in POV editing wars regarding the argued position of one author or another. But the subject of a book, regardless of any opinions therein, should be much easier to determine in most instance, and we'd have library classifications at a minimum to guide us. postdlf (talk) 21:52, 11 September 2010 (UTC)

Bibliography of X is original research if it has not been published before. If it has not been published by a reliable source, it is just not reliable. --Gavin Collins (talk|contribs) 21:14, 11 September 2010 (UTC)

How would your opined principle apply to a "further reading" section within an article? Or to an editor's decision that X, Y and Z sources all pertain to article A, and information from those three sources should be used to expand A, and not article B? postdlf (talk) 21:20, 11 September 2010 (UTC)

Gavin - Please explain if it has not been published before in the context of bibliographies. You've repeated this published before statement a great many times in this discussion. Most scholarly works on subject X contain comprehensive bibliographies. Do those Bibliographies of X meet your been published before standard? Or, does your standard require that someone has published a complete work, entitled: Bibliography of X and clearly defined that bibliography as a list of books about X? An either/or answer is sufficient. --Mike Cline (talk) 21:45, 11 September 2010 (UTC)

Like I said, it's not going to be a productive discussion to talk about exceptions where we create lists of things that have NOT been covered by third-parties. We need to clarify the overall rule first, and let the exceptions fall into place over time. Why did we get started on this when we're still trying to figure out some basic stuff? Shooterwalker (talk) 22:20, 11 September 2010 (UTC)

SW - I don't think we are talking about exceptions here. The basic debate is and has always been: does the notability burden fall on List of X or List of X. The above discussion is just a rehash of that in the context of bibliographies. There are four key elements of a standalone list:

1. List Title (conveys the contents of the list within the constraints of our naming conventions)
2. List topic (bears the burden of notability if such a burden is required by the nature of the list (navigational only list exempted))
3. List Inclusion criteria in lead - establishes the criteria to verify for list entries and further establishes the list topic. (No one is contending as I can tell that inclusion criteria should not be discriminate.)
4. List entries - individual pieces of content that must be verifiable against the lead's inclusion criteria.

The only real issue as I see it, is what is the list topic? List of X or List of X. The majority as I see it believes it is X. The only question that remains is what specific changes need to be made (if any) to WP:SAL and/or WP:N to codify that position.--Mike Cline (talk) 22:34, 11 September 2010 (UTC)

We've probably talked about that to death too. At this point it's a matter of finding someone who will close the discussion. We can open a sequel to this discussion if we want to develop it further. But I'm comfortable codifying whatever we got here, even if it's high level and wishy washy. Further changes can always happen a few months down the line through the WP:BRD process. Shooterwalker (talk) 22:42, 11 September 2010 (UTC)

Concur and request made: Here--Mike Cline (talk) 22:53, 11 September 2010 (UTC)

Thanks a lot for trying to move the process forward. I backed you up. Shooterwalker (talk) 23:44, 11 September 2010 (UTC)

In answer to Postdlf, a "further reading" section within an article can be added because Notability guidelines do not directly limit article content. But putting such a list in a separate stand alone article means that it has to meet the notability guideline.

In the real world, an academic might put together a list of works by an author or composer, e.g. Köchel catalogue. But if a Wikipedian does the same, that is original (primary) research. The point of making this distinction is two fold: the list topic (or the idea for the list) needs to be attributable so someone other than a Wikipeidian, and the content of the list should be taken from a reliable source. If a list does not meet WP:BURDEN, then there is no rationale for inclusion, and its content is not reliable.

This why many madeup lists like List of Masonic buildings are so problematical. The editors who created it can't agree on what the inclusion criteria are (should it include Masonic hospitals?), as the idea for the list (the list topic) does not come from a reliable source. Nor can they agree on which items should go in the list because its content does not come from a reliable source; instead the editors add items based on categorisation, which is a purely subjective metric for inclusion.--Gavin Collins (talk|contribs) 09:45, 12 September 2010 (UTC)

Neither my question, nor your original comment above, had anything to do with notability. This is only about what constitutes original research and what is "reliable" as you put it, and you haven't addressed my questions other than to restate your opinion.

Lets say I find three separate books at the library, all of which are biographies of Joe Schmoe that would satisfy all reliable source requirements. No single source has ever compiled information from those three books, or ever even listed them together. I then use those three biographies to A) write a single article about Joe Schmoe, and list those three books in a "references" or "bibliography" section within that article; or B) discover that there is already a Joe Schmoe article, and merely add those three books to the "further reading" section rather than use them to expand the article; or C) include the books in a Bibliography of Joe Schmoe or List of books about Joe Schmoe. Please explain why you think A and B are reliable acts and not original research, but C is unreliable and original research. Because I don't understand at all why A, B, and C would involve different kinds of editorial judgment; all involve determining that the books are about the same Joe Schmoe and no more. postdlf (talk) 12:44, 12 September 2010 (UTC)

Postdlf knows that A and B are usually added to existing article topics (space allowing) in accordance with WP:NNC. Such lists are subject to content policy such as WP:V, so even if they are included in an article, such lists cannot be made up. If we look at a real list article, such Ernest Hemingway bibliography, we can see for ourselves the problems associated with original research: it contains works by Hemingway himself and (unbelievably) film and television adaptations of his works by other authors. The only reliably sourced list (within the list) is the "Posthumous works", is are taken from a reliable source. The underlying presumption is source of the list is verfiable and notable, but alas that is not always the case in practise.

Can Postdlf give an example of list he has himself compiled? --Gavin Collins (talk|contribs) 17:26, 12 September 2010 (UTC)

Notability is a red herring in this case; the question is, from an original research standpoint, why A, B, and C are different in your mind? The way you are describing what you being is inappropriate original research is what is used in the construction of every wikipedia article as we are creating summaries of topics that have never been published before - which of course is an impossible position to try to compromise with. --MASEM (t) 17:32, 12 September 2010 (UTC)

Notability is never a red herring; while WP:NNC does not directly limit article content of notable topics, there are other content policies that do. The reliability, size and context provided by the coverage in a list is important in relation to the principle of due weight and other content policies. If the content of a list is not significant, there is no good reason for inclusion within an article. A good example is the List of Heroes characters: its content is not worthy of inclusion within the the Heroes (TV series), because it is primarly original research that goes into needless detail; why Masem believes such a poor list should be included in Wikipedia is highly questionable.

For standalone lists, notability is highly important. I would suggest to you that Ernest Hemingway bibliography could be a notable list topic, but it has not been fully sourced at this time. In its current state, it is based on original research of sub-standard quality and needs to be improved. This is why notability is so important for lists: content derived from the creative efforts of Wikipedia editors is not a good as content derived from reliable sources, even though we must assume good faith in our editors. --Gavin Collins (talk|contribs) 08:25, 13 September 2010 (UTC)

Gavin, I've said it before: I agree that List of Heroes characters is presently a very poorly-written list because it is endlessly detailed. But there is a huge difference between a poorly written list and an improper list for Wikipedia. The former can be fixed; the latter cannot. We should not be letting lists that have writing and other content problems be part of the discussion about the inclusion of lists.

The question that is still being asked of you that you are side-stepping is this: is there any difference, considering only the original research standpoint, of a bibliography assembled from various sources, placed within the context of an article and the same, placed as a standalone list? Your statements suggest "yes", which is completely counter-intuitive to how we use "original research" (not Original Research) as goo editors and researchers to assemble sources in the most comprehensive manner even if we introduce overarching aspects that were not original spelled out in the sources - standard editing practice. --MASEM (t) 12:55, 13 September 2010 (UTC)

That was not what I was saying; I was refering to where WP:NNC is relevant. Simply put, if a list is original research, then it should not be placed within an article, nor should it be the subject of its own standalone list article - I think we are all agree on this. Likewise, if a list is not taken from a reliable source, then it should not be given the same weight as coverage from a reliable source in accordance with WP:WEIGHT.

Where I think where there may be a misunderstanding is to what extent is original research is tolerated as proxy for a reliable source. Going back to the list Ernest Hemingway#Selected list of works, we know this list is unreliable (it has not been sourced and is incomplete), but it is tolerated with the context of this article as a proxy for a list that could be sourced. However, if it was a rubbish list ("List of Hemingway books featuring bullfights"), then it would not be tolerated at all.

From the persective of WP:NOT#OR, Ernest Hemingway bibliography is more problematical. The author worked as a journalist, so a lot more of his material was published than this list suggests. I put it to you that amateur lists like this might be good by school homeowrk standards, but I think they are barely tolerable elsewhere because they are not taken from a reliable source such as this.

Once again I put it to you that if you believe lists do not have to verfiable or notable, then you should put forward a proposal to have them exempted from content policy. If you continue to advocate that made up lists comply with content policy, then we will never reach a shared understanding of what the incluison criteria for lists are.--Gavin Collins (talk|contribs) 13:53, 13 September 2010 (UTC)

That arguement is baseless. There is no change in how content should be treated as "original research" between being embedded in a larger article or in its own standalone article. If it is "bad" Original Research that is used to push excessive POV and synthesis, it shouldn't be tolerated period, embedded, standalone, or otherwise. If it is "accepted" original research used to group common aspects of a topic in the act of summarizing and writing about them, then it shouldn't matter how its presented, the only issue becoming a question is if the stand-alone version really needs to be stand-alone (which has nothing to do with original research at all). --MASEM (t) 14:06, 13 September 2010 (UTC)

List Topics should demonstrate notability, ie. Ernest Hemingway should be notable. List inclusion criteria should be discriminate: ie. Books or major works by Ernest Hemingway. List entries need to be verifiable against inclusion criteria. List entries are not burdened with individual notability unless that is explicit in the inclusion criteria. There is no blanket application of notability to content, only topics. Content has to be verifiable, not notable. Gavin's assertion of OR in the context of bibliopgraphies is absolutely inconsistent with encyclopedic practice!--Mike Cline (talk) 14:20, 13 September 2010 (UTC)

Gavin, you're good at repeating yourself but bad at advancing your arguments or actually responding to the points people have made and the questions they have raised. I hope it's at least clear to you that your interpretations of these policies and guidelines is a minority one, though you try to phrase them as abstract fact. You also keep sliding between complaining about a hypothetical list not satisfying GNG and being OR, clearly not the same things, and this among other things has made it impossible to keep a discussion with you focused. This is really becoming tendentious on your part. You're not convincing anyone and you're not saying anything new, despite the various ways we have tried to address your comments and explain our own. So I think continuing this with you is a waste of time; I urge everyone else to move on as well. postdlf (talk) 15:02, 13 September 2010 (UTC)

I think it is intuative that a list is a source of coverage just like any other source, and pretending that it is OK to make up a source of coverage in the absence of a published source is not going to work. Even if Postdlf wants to ignore this distiction as the minority view, he still has a problem with how to distinguish between lists that are original research, and those that are not. This is an important distinction, for my guess is that every single list that has ever been deleted was original research.

For all intents and purposes, a list is no different from any other source of coverage, whereas Masem and Mike's views (which are an exemption from WP:NOT#OR in all but name) is that "List of Hemingway books featuring bullfights" would qualify for inclusion (as bullfighting is a notable theme in his works), regardless of weight or notability, or even if the list had been published before. --Gavin Collins (talk|contribs) 15:12, 13 September 2010 (UTC)

I think the problem here is that we are dealing with two related, but not identical concepts. WP:N and WP:NOR can and do overlap... but they are not quite the same. Thus, we can have lists that fail WP:N but pass WP:NOR... lists that fail WP:NOR but pass WP:N... and lists that fail both. All three scenarios are problematic, and lists that fall into all three scenarios should be considered as possible deletion candidates... but the reasons for deletion are distinct in each case. Blueboar (talk) 15:29, 13 September 2010 (UTC)

I can't imagine a list that would fail WP:NOR but pass WP:N - there is an inherent conflict in this statement, since a list can only be notable if it verfied, and that excludes any possibility of original research. Can Blueboar provide some examples to illustrate his point? --Gavin Collins (talk|contribs) 16:24, 13 September 2010 (UTC)

Sure... I think a List of people currently involved in a secret conspiracy against the government would be a very notable topic for a list (we could easily establish that the topic is notable, as the concept of someone being involved in "secret conspiracy against the government" is discussed in lots and lots of sources)... However inclusion of any particular person would be completely unverifiable (how can you verify something that is secret). I would definitely give a Delete !vote if such an article were created and came up at AfD... but my reasoning for deletion would rest upon WP:V and WP:NOR, and not upon WP:N.  :::::Blueboar (talk) 14:55, 15 September 2010 (UTC)

I think I detect a hint of irony here, in which case my answer would be "how could a List of people currently involved in a secret conspiracy against the government be presumed to be notable if it is a secret?". I would suggest that Blueboar has been watching Three Days of the Condor a little too frequently of late.

If he is being serious, I think Blueboar is mistaken. The presumption that such as list is notable, in the absence of verifiable evidence, is simply not proven. In the absence of evidence, there is no way to prove it is not original research.

I think you will agree that most editors are willing to tolerate original reasearch if they think a list is plausible, interesting or useful. However, this is problematical from the perspective of content policy. --Gavin Collins (talk|contribs) 15:52, 15 September 2010 (UTC)

OK... I'll give a few more examples...

1) List of time travelers... the topic of time travel, and of people who travel in time is notable. It is a key genre in fiction, and noted scientists have discussed the topic (noting for example, that a time traveler could create a "grandfather paradox" etc.). However, we could not populate such a list due to WP:V and WP:OR concerns.

Or if you want an existing article....

2) List of wars between democracies. The topic of "war between democracies" is a notable topic... it relates directly to the Democratic peace theory. However, as we have discussed at NORN, because sources disagree as to what constitutes a war, and what constitutes a democracy... populating the list requires OR synthesis. The notability of the topic can be established under WP:N, but the list itself violates NOR. Blueboar (talk) 16:21, 15 September 2010 (UTC)

I have no idea if these are published lists in their own right, but even if they are [1], then they need to be verified to avoid plagiarism, as well as notable if they are to used in standalone articles. I disagree with you that synthesis is allowable to draw up these lists, for there really needs to be a published source that has undertaken that step already. List of Masonic buildings is a classic example by your own admission: the contributors can't agree on its scope because publishing such a list is what reliable sources do, not Wikipedians. We can't create new topics out of our own ideas for the simple reason that if a published source has already done so, we would be stealing their ideas without realising it.--Gavin Collins (talk|contribs) 21:17, 15 September 2010 (UTC)

I am not saying that synthesis is allowable to draw up these lists. In fact, I am saying exactly the opposite... that synthesis is the reason why these articles should not be allowed... even though the topic is notable. With the List of Masonic buildings, on the other hand, we have two problems... a problem with synthesis, and a problem establishing that the topic is notable. They are both problems, but they are distinct issues. Blueboar (talk) 21:51, 15 September 2010 (UTC)

Summarizing (again)

May I suggest that we re-boot the summarization. Rehashing where we disagree is not going to move us forward. Let's try to identify the things we do agree on. Blueboar (talk) 13:48, 12 September 2010 (UTC)

• I agree with that. :) Shooterwalker (talk) 16:38, 12 September 2010 (UTC)
• I agree as well. So we are agreeing to agree. But what are those points of agreement?
  • I agree that List topics (List of X) should be notable. I agree that list inclusion criteria should be discriminate in the context of the list topic. I agree that list entries should be verifiable against the inclusion criteria. I agree that the List Title should convey (within the contraints of our naming conventions) the idea that the list is an enumeration of logical things relative to the list topic X.--Mike Cline (talk) 14:59, 13 September 2010 (UTC)

I agree... these are the points where we agree. :>) Actually, I think this is a very good summary of where all (or at least most of) the various proposals overlap. I could easily see something like this going into WP:N. It is short, to the point. Blueboar (talk) 15:12, 13 September 2010 (UTC)

I think it's more complicated than that. I think the only simple statement we could make regarding lists and notability is that "standalone lists must pertain to notable topics." From that, we can perhaps generally describe some kinds of lists that may be valid:

1. Lists of X that index notable topics (i.e., articles and proper redlinks) by a basic definition X of what they are (lists of people, of companies, of films, etc.), often organized into sublists.
2. Lists of X of/by Y that index notable topics X in relation to another notable topic Y (e.g., lists of people by city, of buildings by city) or by a shared fact Y (e.g., lists of films by year of release, lists of politicians who were impeached) or by a variable fact (e.g., lists of rivers by length, lists of countries by population) where those shared/variable facts are encyclopedic and relevant to the subject.
3. Lists of X of Y that are split off from notable topic Y primarily for size reasons, even where most or all of the Xs may not themselves be notable topics (e.g., lists of episodes of a television series, lists of buildings by architect).

I think we have a clear consensus, both here and as reflected in existing practice throughout Wikipedia, that a list may satisfy notability requirements in any one of these three manners. Have I left any other kind of list out? Note that I've purposefully omitted things like Nixon's Enemies List, where a historical list is itself an article topic. postdlf (talk) 18:59, 13 September 2010 (UTC).

Nicely captured. List topics must in some way demonstrate notability. Articles about Lists are articles not Lists. Lists are a special style of article that enumerate things about notable topics such that we need some additional guidelines WP:SAL around their construction.--Mike Cline (talk) 19:09, 13 September 2010 (UTC)

The idea that "List of X" is notable because X is notable is based on a notability fallacy, and I think it is travesty to misrepresent the contributions to the RFC in this way.

Similarly, the idea that a list based wholly on original research because they are spinoffs is not going to stand up to peer revew. I can't see that there is consensus to ignore content policy at all, and pretending that it does not in the absence of sourcing is not going to work.

What is missing from this proposal is recognition that a list (like any other source) must be verifiable, and if it is to be included in Wikipedia as a stand alone article, then a list (like any other topic) be the subject of reliable sources to demonstrate notability. Lists without notability fail WP:NOT#DIRECTORY. No amount of wishful thinking is going to change this. The idea that list is notable if it has not been noted in accordance with WP:N is nonsense.

The achillies heal of this proposal is that is does not show how it is possible to distinguish between lists that are original research, and those that are not. If there is no verifiable way to do this, then I would suggest that we try another proposal: one that regognises the nature of compiling homemade lists. --Gavin Collins (talk|contribs) 10:01, 14 September 2010 (UTC)

And again Gavin, the point you seem to miss(?), is that ALL article body-content is compiled through Original Research - I have yet to see you make the argument as to why lists should be treated differently to all other articles, just because the primary article content has additional structure. To be fair, I recognise that you perceive this somehow differently - but I have not seen you clearly address this aspect. ‒ Jaymax✍ 11:20, 14 September 2010 (UTC)

Articles are not compiled by original research. The are compiled by citing sources that are address the topic directly and in detail. This is what distinguishes original research/synthesis from the compilation and summarisation of reliable sources within articles. Be clear in your mind about this. --Gavin Collins (talk|contribs) 11:39, 14 September 2010 (UTC)

Gavin, I am utterly clear in my head about this, and I've done plenty of hard thinking about it. And to my mind you are obviously, demonstrably, objectively, wrong - there is no way that I can see to argue that the COMPILATION of the content of an article does not constitute original research, but the compilation of the content of a list does. A compiled list needs to cite sources for content in EXACTLY the same way a compiled article does. The distinction you are trying to draw does not exist. I almost beg you to think about this for a while so that you can become clear in your mind about the objective truth of it. ‒ Jaymax✍ 12:13, 14 September 2010 (UTC)

I think it entirely possible that I may be mistaken. But that still does not answer my earlier question, how is possible to distinguish between lists that are original research, and those that are not? If you can explain the mechanism that would identify List of Masonic buildings or "List of Hemingway books featuring bullfights" as being or not being original research, then maybe I can change my way of thinking. --Gavin Collins (talk|contribs) 12:41, 14 September 2010 (UTC)

This is why it is necessary to make a distinction between upper-case Original Research, the bad type, that talks about putting novel claims and improper synthesis into content of WP articles - and of course the type we have to discourage completely - and lower-case original research, the type that takes a source, recognizes what topic(s) that source applies to, and placing summarizing content that is sourced into the appropriate section of the relevant article(s) - the type of original research fundamental to building the work. There is no difference between if lower-case original research is being applied to a list or an article; it's readily acceptable as long as it's not pushing a specific POV or other content policy problem, such as actually introducing upper-case Original Research.

The question Gavin's trying to ask here, that is, about specific list definitions, is not really about Original Research vs original research, but instead simply if the list definition is one that is appropriate for Wikipedia. Certain, a pre-published, notable list is fine, but it's clear from this RFC that WP is more accepting of lists that don't have a pre-published definition - that is, we accept lists that are created based on lower-case original research . Part of this discussion has revolved around the aspect of a notable topic that the list is based on, which certainly seems to be part of the rationale for acceptance based on general AFD consensus. But the other part seems to involve if the list definition has been selected to avoid being too broad or too narrow. The reason we likely would not have a list on Hemmingway works featuring bullfights is that it is too tight a classification considering the rest of his works. Similarly we would not have a complete list of all Masonic buildings since that is potentially infinitely long; instead a restricted list to notable or other similar trait (like nationally-historic) does narrow the criteria to an acceptable level. Are there any easily-derived rules from these to determine if the scope of the list is appropriate? The only one I think we could write is that if the number of members of a given qualification are very limited, we should not further restrict that list; if the list membership is very large or potentially infinite, further restrictions on inclusion are needed. The point here: this aspect cannot be confused with original research or Original Research; it is possible to have list definitions that do fail Original Research, generally due to POV-pushing, but that is only an aftereffect of our allowance for editors to define lists based on notable topics using lower-case original research. --MASEM (t) 13:35, 14 September 2010 (UTC)

Gavin, That’s a rather simple task that involves the editorial discretion we give all our contributors. If in fact someone created a List of Ernest Hemingway works featuring bullfights there are three hypothetical scenarios under which such a list might exist.

• Scenario 1: The total number of EH works exceeds the feasible number of entries for a single list. An editor decides to break up the works into some logical topical groupings—war, bullfighting, fishing and hunting. One of those grouping is bullfighting because a large % of the works contain that theme. Then any work that is included in the list must be verifiable against that theme—easily done via book reviews, etc. The editorial discretion here is the topical (vice alphabetical, genera or other organizational construct) organization of the list. It’s the same type of editorial discretion we use when creating a variety of section headings for any article.
• Scenario 2: A reliable source has written about EH’s prolific use of bullfighting in his works, thus making the EH-Bullfighting connection notable. Again the inclusion criteria is clear and individual entries should be verifiable against the inclusion criteria. The editorial discretion here is to create a list to highlight the works of the notable EH-bullfighting connection.

Both Scenario 1 and 2 are legitimate rationales for a list of this type and these lists should not be deleted.

• Scenario 3: An editor just decides to create this list without any relationship to the total number of EH works or other lists that might organize his works in a different way. There are no sources that say the EH-bullfighting connection is notable. Individual entries could still be verified against inclusion criteria but because the topic is not notable and the topic is not the result of editorial decisions about how to organize a large number of works into logical, multiple lists, it is likely that this scenario would result in a deleted list.

None of these involve OR, 1 and 2 are good lists, 3 is a bad list. --Mike Cline (talk) 13:43, 14 September 2010 (UTC)

I refer everyone again to my comments re: Gavin in the previous section. I detest making discussions personal, but this really is a question of one person who just refuses to get it arguing with everyone else, and it's long ago stopped being constructive. Given that this section was supposed to be about summarizing the points on which there is consensus, Gavin's holdout is really irrelevant and disruptive to bring up here. Which is why I believe it is a waste of time for us to continue to try to change his mind.

I will make a couple further comments, however: my summary of the ways in which lists could sufficiently pertain to notable topics obviously (to all but one) did not address other requirements such as verifiability because it didn't seem an open question that the inclusion criteria of a list, and the inclusion of every entry on that list, must be verifiable, NPOV, and not violate WP:OR by creating a novel synthesis that the sources do not support. A list compiled from multiple sources can of course be sufficiently verified by those sources if it is clear they are using the same definition, such that a list of X that includes A, B, and C as X can be cited to three sources, each for one entry, that all define them as X by the same standard of what X is. That a prose article on Reginald Schmoe could be converted into a List of events in the life of Reginald Schmoe should illustrate that compiling an article and compiling a list from multiple sources do not pose different verifiability problems. And this again is where Gavin disagrees with everyone else, so I really don't care to again hear his opinions repeated on this matter because I'm done listening to him or trying to convince him.

So...back to business... Can anyone point out a valid kind of consensus-supported list that I omitted from my examples, or offer any other further thoughts on notability regarding those forms of lists?

Mike: I think your scenario 3 goes to what I termed in my #2 as the issue of whether the shared fact the list is based on is "encyclopedic and relevant to the subject." Listing works by a theme that no source has commented on as significant, and of which there is no sign that anyone studies in these works, would be trivial and unencyclopedic, even though verifiable. Extreme forms would be List of X of Y and of Z, the proverbial "unencyclopedic cross-categorization" too often trotted out in AFDs (often applied against simple categorizations), where a subject is indexed by two completely unrelated facts, even though true and verifiable: a list of [notable] people who worked for Walmart and were convicted of sex crimes (well, I would hope there is no relation there). Such a list would also be arguably POV, and violate SYNTH by implying a conclusion not supported by any source, namely that there's a relationship or causation there. This is not a new principle, though it's regularly misapplied. postdlf (talk) 14:16, 14 September 2010 (UTC)

We are once again getting sidetracked into where we disagree instead of focusing on where we agree. Please re-focus. Blueboar (talk) 15:03, 14 September 2010 (UTC)

In order to reach agreement, I think there is going to have to be acceptance that what is being described is inclusion based on meaures of subjective importance, which is what is being suggested, but slightly different ways. Just because Elephants and Flying are notable, that does not justify a "List of flying elephants", I think we are all agreed. But in essence, the List of Masonic buildings is in exactly the same boat, because it is the architectural equivalent of Dumbo:

I heard the Wailing Walls, I've seen a Church that falls,

I seen a a city wink its eye

But I be done seen 'bout ev'rything

When I see a elephant fly.

I think the only consensus that can be drawn from this RFC is that most editors have high tolerance levels when it comes to unpublished lists, provided that they appear to be plausible. I can't provide a defintion of what is or is not plausible, but you have each in turn given some good indicators that a list might withstand AFD if it is plausible: "restricted" (Masem), "logical", "thematic", "feasible", or "connected" (Mike Cline), "encyclopedic", "relevant" or "related" (Postdlf). All of these are measures of subjective importance, without a doubt.

Similarly, I think Blueboar is also of this persuasion. The idea that notability is "transfered" is more or less the same as saying that a list is plausible if there is another article about a closely related topic makes it seem plausible.

However, I must stress that they are not measures of notability, even though they are good faith meaures of the value of a particular list. Going back to the proposal from Themfromspace, I think it is clear that list that are subject to peer review (Category:Featured lists) are notable, in the sense that they have been "noted", rather inheriting notability from another article, or the members of the list.

If we can agree to set aside the idea that homemade lists are notable, then we can agree on a way to accomodate them within Wikipedia. The best way to do that is to treat homemade lists in a similar way to disambiguation pages: treat them as pages that are not articles, but aids to understanding a category of topics in barebone form. If there is a way of creating a "homemade list" template, then perhaps they have a future.--Gavin Collins (talk|contribs) 15:42, 14 September 2010 (UTC)

If by "homemade list" you simply mean compiled from and verified by multiple sources in such a way as to only draw conclusions supported by those sources, rather than a list that has already been published as a list by a reliable source, that's not a distinction supported by consensus. I'll use my last bit of patience to again state that a "list is notable as a list" is not a coherent, consensus-supported standard, so your criticism based on that standard of comments that are applying different standards are irrelevant here where we are summarizing what does have consensus.

The clear consensus is that 1) lists are viewed as a format of indexing and/or presenting information about notable article subjects, and so satisfy notability requirements through their relation to notable topics, not as lists themselves, such that List of events in the life of Reginald Schmoe would satisfy WP:N by Reginald Schmoe being notable (just as Reginald Schmoe does not need to be notable as an encyclopedia article); 2) that such a list would be verifiable and not OR if Reginald Schmoe was verifiable as a single topic and each event could be cited to a reliable source, and that these are the exact same standards and editorial judgment we'd use for writing a prose article on Schmoe's life; and 3) that once it has already been determined that "Reginald Schmoe" is a notable article subject, whether it made sense to present facts about Schmoe as a list rather than only in prose form in Reginald Schmoe would not involve another notability analysis specific to the list, but rather a different kind of editorial judgment, as would whether such a list should be further divided into sublists. Does anyone other than Gavin disagree that we have agreement on these points, or disagree that these points reflect community practice? postdlf (talk) 16:49, 14 September 2010 (UTC)

Perhaps Postdlf has missed the point that a list compiled from and verified by multiple sources has to be published by a third party before it can be included in Wikipedia. From the standpoint of WP:RS, a list is no different a source from say, an academic paper or novel: any conclusions that can be drawn from a lists (i.e. how its members have been selected) must come from a published source, not madeup.

Perhaps Postdlf missed the point I made earlier about Ernest Hemingway bibliography, that most of it is not taken from a reliable source, and cannot be trusted. The list includes adaptations of his work written by other authors, so it is not a bibliography of his work per se.

However, the list does contain a sub-list which is the subject of reliable sources (his Posthumous works) which can be verified. The key point to note is that both the defintion and the content of homemade lists cannot be checked against a reliable source. No matter how good editorial judgment is when it comes to putting items into lists, it cannot be verified unless it is backed up by a reliable published source.

Where I think we need to agree is how to treat homemade lists that have not been published. If we can agree that they should be treated differently from content pages, then we can remove the conflict between them and content policy which is very clear about the prohibition against original research. --Gavin Collins (talk|contribs) 08:45, 15 September 2010 (UTC)

So you're saying we should only have a list article where the entire list has been published in another source. Wouldn't that be a copyright violation? We don't allow articles that are copied entirely from another source, so why lists? Alzarian16 (talk) 09:49, 15 September 2010 (UTC)

It depends, like any other source, on what the circumstances and terms of use are. For instance, the New General Catalogue was first published back in 1880's and therefore the original version is not subject to copyright. More recent revisions to the the original are, e.g. The Historically Corrected New General Catalogue, and so reproducing them in full would be a copyright infringement. I am no expert in the field of copyright, but I expect they are no different from any other source in this regard. Since most lists are only a small part of a particular publication, using published lists is unlikely to be a copyright infringement in my view.

Where copyright violation is more likely to be a problem is where a homemade list does not cite its source. Citing the source of a list is needed to avoid plagiarism - see Feist v. Rural. --Gavin Collins (talk|contribs) 11:08, 15 September 2010 (UTC)

Does anyone other than Gavin disagree that we have agreement on the points I raised above, or disagree that those points reflect community practice? postdlf (talk) 12:46, 15 September 2010 (UTC)

• I think we are agreed it is based on a notability fallacy. --Gavin Collins (talk|contribs) 12:57, 15 September 2010 (UTC)
  • Anyone other than Gavin? postdlf (talk) 13:49, 15 September 2010 (UTC)

• I agree that List topics (List of X) should be notable. I agree that list inclusion criteria should be discriminate in the context of the list topic. I agree that list entries should be verifiable against the inclusion criteria. I agree that the List Title should convey (within the contraints of our naming conventions) the idea that the list is an enumeration of logical things relative to the list topic X. I agree that even purely navigational lists and lists of logical cross categorizations need to demonstrate some encyclopedic notability/suitability. We don't need more bad lists, we need more good lists. I hope there is consensus on that last point. Let's go build them--Mike Cline (talk) 15:59, 15 September 2010 (UTC)
• I agree with "1) lists are viewed as a format of indexing and/or presenting information about notable article subjects," but I do not agree with "and so satisfy notability requirements through their relation to notable topics, not as lists themselves, such that List of events in the life of Reginald Schmoe would satisfy WP:N by Reginald Schmoe." We need third party sources that have identified that the parameters of the list/list inclusion criteria are themselves something that have been shown to be worthy of looking atedited to add/encyclopedic. Active Banana ( ^bananaphone 16:51, 15 September 2010 (UTC)

Well stated. this is part of what I was trying to get at with my "Option C" on the main page. Blueboar (talk) 17:01, 15 September 2010 (UTC)

I might agree with your second sentence depending on how you think we establish that the list parameters are "worthy". Based on this RFC and my observations of overall Wikipedia practice, that's not necessarily a notability issue, but another kind of issue. Active Banana (if that is your real name), what do you think that analysis involves? postdlf (talk) 17:02, 15 September 2010 (UTC)

I realize this term is just as vague as worthy, but encyclopedic is what we are shooting for!--Mike Cline (talk) 17:20, 15 September 2010 (UTC)

by "worhty"/"encyclopedic" - essentially the same criteria that we use for any article: WP:N parameters that third party sources have discussed in a more than trivial manner. Active Banana ( ^bananaphone 18:15, 15 September 2010 (UTC)

Encyclopedic does not = notable; notability is an analysis for whether a subject merits an article, while encyclopedic may refer more broadly to when a fact merits inclusion in an article or how those facts should be organized. On that distinction, I agree with "encyclopedic," because I think the kind of editorial judgment used in constructing an article around a notable subject is really no different in kind from the judgment regarding what kind of lists we should construct listing notable subjects or elements thereof. But many valid ways of listing or indexing haven't necessarily been "discussed", though perhaps widely utilized in general kind, and there may be permutations of such indexes that we may simply be the first ones to compile even though the parameters are completely verifiable and encyclopedic. And that's one of the main issues this RFC has been about.

I suspect we'd have far fewer lists and less of these issues to worry about if we had a database of article information that was searchable by cross-referenced facts. I've been thinking about List of federal judges appointed by George H. W. Bush, for example. I don't know that anyone has "discussed" this list as a whole, and that shouldn't matter, because obviously it's verifiable, NPOV, completely encyclopedic, and well grounded in notable subjects. It was compiled from the Biographical Directory of Federal Judges, which is searchable, so that the Directory does not itself maintain a list, but rather is comprehensive and searchable by appointing president allowing that such a list could be compiled. The same result would be reached (with much greater effort, but no less verifiability) if we had 192 separate reliable sources for each of the judges each saying they were appointed by Bush (and a source confirming the number of 192 was complete, of course). Similarly, I've compiled lists of all Supreme Court opinions authored by each justice by term for quite a few years now; these lists are probably nowhere else published in this form because online databases like LEXIS or WestLaw can generate the same information with an easy search of their comprehensive databases, without the need to maintain it as a distinct document. postdlf (talk) 19:04, 15 September 2010 (UTC)

Hmmm... Given our discussions, I actually have to wonder about that article... I suppose grouping Judges by the President who appointed them is a conceptually valid way to organize and break up a list of all Federal judges (other, potentially less POV, ways to organize the same topic would would be to list them by decade of appointment, or by home state)... but if that is the justification, I do have to wonder why we don't have a List of federal judges appointed by Warren G. Harding, a List of federal judges appointed by Lyndon B. Johnson, a List of federal judges appointed by Franklin D. Roosevelt, or lists of federal Judges appointed by most of the other Presidents? If the point of the list is to say that there is something notable about judges who were appointed by that particular President, then article needs to do a better job of establishing what that something notable actually is. But I digress. Blueboar (talk) 19:29, 15 September 2010 (UTC)

All other presidents should have such lists because all (but one) other presidents appointed federal judges, and (as the bluelinks you apparently unwittingly presented indicate) they all do, except for those who appointed fewer than ten judges (see note at top of Category:Lists of United States judicial appointments by president). And we also have lists of federal judges by the court to which they were appointed, by length of service... That's one of the great features of lists: they are not mutually exclusive. postdlf (talk) 19:45, 15 September 2010 (UTC)

When I am wrong, I am wrong... and I admit it. My excuse... While I was typing the above, I did a show preview to see if there were articles for Judges appointed by other the Presidents, and I got red links... I realize now that I checked them without the middle initials, then neglected to check another Show preview when I added the initials, to see if the links were now blue... (d'oh). My apologies for the confusion. Given that there are other lists for judges appointed by president... I will retract that concern... as I said before, I think that grouping US Federal judges by the President who appointed them is a valid and logical method of breaking up the topic of all US federal judges (since List of US Federal Judges would be overly long). Creating sub-lists grouping this topic by appointer is no different than grouping by alphabet, or by decade. Blueboar (talk) 22:15, 15 September 2010 (UTC)

Indeed Encyclopedic <> Notable but what it does mean: Synonyms for encyclopedic: all-embracing, all-in [chiefly British], all-inclusive, broad-gauge (or broad-gauged), compendious, complete, comprehensive, cover-all, cyclopedic, embracive, exhaustive, full, global, inclusive, in-depth, omnibus, panoramic, thorough, universal is instructive. We just need to find a simple way to capture this in practice. --Mike Cline (talk) 19:16, 15 September 2010 (UTC)

• Not much to add. But sign me up for option C. I don't think that X's notability makes any list in an around X notable, like "List of X's that Barack Obama has met", or "List of X's with red hair", or "List of businesses that have endorsed X". The X IS the topic of the list. You need to verify notability of "businesses that have endorsed X" as a topic/set/group, in the sense that some third-party source has talked about this group of things. Shooterwalker (talk) 22:00, 15 September 2010 (UTC)
  • Shooterwalker, I think you're making the same mistake a lot of people are in this RFC: trying to use notability criteria as the sole gateway for preventing or deleting bad lists. Notability criteria is not the only hurdle a list must pass. Saying that notability criteria can be satisfied by certain kinds of lists is not to say that every possible list of that kind is also NPOV, V, or encyclopedic any more than every possible fact regarding a notable subject is encyclopedic and should be added to that subject's article. We also wouldn't add a list of all people Barack Obama happened to meet to Barack Obama (even if article size was not a concern) for reasons other than the "non-notability" of meeting Barack Obama; notability is simply a nonsensical way to approach that kind of editorial judgment, and might even give the wrong answer given the vast news coverage typically given to everything a president does every day. Such a list would be better criticized as indiscriminate, and probably unverifiable because different sources probably wouldn't agree on what it meant to "meet" someone (shake their hand? speak to them? be present in a group in front of them? in the same room?). So again, notability criteria is not the only way, nor always the best way, to criticize bad lists.

    You're also not accurately critiquing the general list forms I and others have tried to describe, because none of your examples of bad lists are simple "lists of X"; they're "lists of X that Y" (people who have met Barack Obama) or "X by Y" (people by hair color). postdlf (talk) 23:24, 15 September 2010 (UTC)

    • Sorry, wasn't trying to complicate things. I was referring to the example "List of events in the life of Reginald Schmoe" where simply proving that Reginald Shmoe is notable would be enough to justify this list. I don't think that fits the general "list of Xs" form, which prompts a whole other end of the debate. You're right that we probably agree on more general "List of Xs" lists, it's just that the "List of Xs in Y" or "List of Xs that Y" that are trickier. Shooterwalker (talk) 02:25, 16 September 2010 (UTC)
      • We'd normally call such a list a timeline or a chronology (see Timeline of the Presidency of Barack Obama (2009) or Vincent van Gogh chronology for example). And I merely meant that proving that Schmoe is notable would be enough to get the notability beast off of such a hypothetical list, I did not mean that there were no other requirements for keeping the list in addition to a prose article. It's a telling hypothetical in my view because it simply doesn't make sense to ask whether a timeline of a notable individual's life is notable. A timeline is just a different presentation of the kind of biographical and/or historical information you'd have in a standard prose article. So notability analysis simply does not meaningfully answer the question of whether such a timeline list should be created/kept once you've already decided that the subject being timelined is notable; a different and further analysis is called for. And the same is true in evaluating the relationship of X to Y in those trickier lists; it's a different analysis than notability.

        List of presidential trips made by Barack Obama or List of state visits made by Queen Elizabeth II might be an interesting comparison to your "people who have met X" hypothetical. postdlf (talk) 05:21, 16 September 2010 (UTC)

Two points:

(1) Active Banana, above you say "essentially the same criteria that we use for any article" and then reference WP:N. But N is a criteria for Articles, NOT article Content. The content of a list, just like the content of any other article, should indeed be subject to the same criteria; but WP:N is specifically EXCLUDED from content issues. Rather, other policies and guidelines, notably RS, NOT etc, apply, and are implemented in the first case per consensus on the article talk page. It makes no sense using existing policy to apply WP:N to list article content.

(2) Reginald Schmoe is perhaps not an ideal example, for the following reason: I would argue that if there is an article The life of Reginald Schmoe which is agreed notable, then a list of events in the life of Reginald Schmoe would be appropriate, because it's quite reasonable to regard a life as a set of events. However, a biographical article about a person, is ipso facto an article about the life of that person - It's a stretch to argue that a person is a completely distinct topic from that persons life, hence if RS is notable, so is Life of RS, and if Life of RS is notable, then so is an article that happens to document that life in list form, rather than in prose.

‒ Jaymax✍ 06:20, 16 September 2010 (UTC)

Just to clarify, I am NOT talking about the specific content of the list article. I am talking about one or more people having published something non-trivial covering "a group of items fiting certain specified parameters" which we then use as the list criteria for the content of the list article. Active Banana ( ^bananaphone 19:05, 16 September 2010 (UTC)

Expanding a bit - if we took the example List of Relatives of Reginald Schmoe, then the topic of the list is Relatives of Reginald Schmoe, and as such would have to meet notability criteria in it's own right. I note Shooterwalker above interpreting my 'B' option as implying that Relative and Reginald Schmoe being notable would be sufficient under the B option - that is absolutely not the case, and I don't think anyone is arguing for that degree of freedom. ‒ Jaymax✍ 06:27, 16 September 2010 (UTC)

It's another list of X of Y example. The relationship of X to Y must be encyclopedic

Jaymax is making the mistake of assuming that a list is notable if it associated with a notable topic, but as stated before, that is a notability fallacy. His justification for creating homemade lists based on original (primary) research is pure fabrication, and does not accord with exisiting practice.

There are many ways to present biographical information: as a conventional article, as a list (a timeline), as a obituary (as is often done in newspapers), or as an essay (e.g Reginald Schmoe, his role in history") . Even if all these article forms can be compiled using the same sources of coverage, this is not done in practice, as the sources are usually rolled up into a biographical article, which straddles both thematic as well as chronologoical coverage.

However, if a timeline of an individual's life is presented in a standalone article, it would have to be have been published as a single source or work, and commentend on by other reliable sources to justify its inclusion in Wikipedia as a standalone list. I have never seen this in practice, and I concur with Active Banana's views, as this is the existing consensus.

If you look at the way that bibliographical coverage is presented in Wikipedia, you will see that there are probably no timeline's about individual lives. There are no timelines associated with the lives of, say, Queen Victoria, Albert Einstein, Joseph Stalin or Mao Zedong in Wikipedia.

In short, there is no justification for concocting madeup lists of this type. --Gavin Collins (talk|contribs) 09:54, 16 September 2010 (UTC)

Gavin, you are very good at stating things repetitively, it does not make what you state any closer to being true. You can keep referencing WP:NOTINHERITED as often as you like; it would be better if you read it, understood it, and realised is says nothing along the lines you keep trying to invoke. But hey, I said that before; you remain intransigent and generally ignorant of what the project space pages you continually refer to actually say.

But my main point is this, any intelligent, rational reading of my point above, would show that I am in no way proposing 'concoctions' of the type I describe, but pointing out why, as a straw-man test-case, the example being discussed is less than ideal. That was all.

Finally, I am no longer going to engage you in debate. I have had enough of what, I am forced to conclude, must be quite deliberate twisting and ignoring of others arguments. I have had enough of your endless repetition of points that have been thoroughly de-constructed by others previously. I have had enough of you being so arrogantly certain you are right about everything. Winning arguments by sticking your fingers in your ears, which seems to be your strategy, works for 5 year olds, it should not work here. Clearly, in a limited way, I can no longer AGF with regards to you, and so recuse myself accordingly. ‒ Jaymax✍ 11:27, 16 September 2010 (UTC)

I can understand your impatience, but you have to acknowledge that there is merit in this point: timeline lists are seldom used, if at all, for biography; simply saying I am intransigent or ignorant does not refute this argument. I suggest you drop the ad hominem attacks, and ask yourself why this is so. Actual practice indicates that list a "List of events in the life of Reginald Schmoe" would be not be appropriate topic for a list. Whether this is due to a lack of notablity, or some other reason, I can't say for sure, but there is a complete absence of evidence to support your claim that madeup lists are considered appropriate for biography.

What I think yourself, Mike Cline and Blueboar now have to try and work out is why there is such a glaring exception to your idea that "List of X" is notable if "X" is notable. It seems to me that "List of events in the life of Reginald Schmoe" and List of Masonic buildings share exactly the same characteristics: they are tolerated, but they are not compliant with content policy as they have not been published in the real world. --Gavin Collins (talk|contribs) 11:53, 16 September 2010 (UTC)

Gavin, you know from the WQA you put against me that discussing your talk behavior is not an ad hominem attack. You are not coming to the consensus that everyone else has and persisting in that, and either you are going to need to realize that and come to meet us there, or consensus will move on and will ignore anything else you have to say. The case here for lists is pretty clear - we have a good idea of what needs to be changed and added to other guidelines and as a result we need to define the language to go into them per consensus to complete the process. You may think this is a completely wrong step and against policy, but consensus create policy, not the other way around; policy is changed if it is found lacking. Everyone else has acknowledged this and cannot sit around waiting for you to do the same. --MASEM (t) 12:34, 16 September 2010 (UTC)

summarizing again - ye olde subheader

Lets stay on topic. The simple point I have made is that timeline lists are seldom (if ever) used as source of coverage for biographical topics. What is the reason for this, and what to lesson can be drawn from this point? --Gavin Collins (talk|contribs) 12:50, 16 September 2010 (UTC)

1. In reply to your iterations of that idea from above ("... you will see that there are probably no timeline's about individual lives ... timeline lists are seldom used, if at all, for biography ..."): Answer: See Category:Personal timelines (at least 2 of which are Featured Lists). That category is as small as it is, because {{Include timeline}} and mw:Extension:EasyTimeline are bloody complicated, and plain text/table lists are not "all they could be" (eg Graphical timeline of the universe) and hence are less inspiring to write.

   And of course, ideally much of this could be done by extracting information with metadata/microformats – e.g. We should eventually be able to enter the url for List of ships attacked by Somali pirates into a site/program, and it will use our metadata to autogenerate a graphical timeline of events – but that's a whole different can of worms...

2. A lot of your arguments seem to result from a different understanding of WP:OR than the other participants in these threads. Now, I'm not even close to being well-read in this area of our rules, but as I understand it, WP:OR is intended to prevent "original synthesis" of information (drawing conclusions), and is not cautioning against "original compilation" of information (putting information from 2 sources into 1 article). Possibly I have that horribly wrong (and I need more coffee still), but it's what I've been thinking for a while whilst watching this.
3. We keep turning back to the abstract mumblings, whereas we should now be concentrating on: What exactly needs to be changed at WP:SAL (and elsewhere) in order to allow the existence of all the things that most of us agree should exist? whateverthehellwecallthem (timelines, glossaries, indexes, outlines, topic bibliographies, lists of lists, and other lists-that-are-originally-compiled-here-in-a-neutral/verifiable/consensus-way). Gavin keeps pointing out that the rules don't allow for some of these, as they currently are, and if we get specifically exact he is correct. So, we simply need to amend the rules, as we've often done when they don't match up to reality. Descriptive not prescriptive (usually) and all that.

HTH. I'll give the addendums once I've been corrected...! (and ccofffeee....) -- Quiddity (talk) 18:47, 16 September 2010 (UTC)

To get back to the discussion, Jaymax, your comments here state that you didn't think the Reginald Schmoe timeline was a good hypothetical, but your last sentence in that comment makes exactly the point I was trying to make through that hypothetical (maybe you meant not a good example for someone else's point?). It isn't an easy proxy for all kinds of lists, but it does well illustrate that there is not a firm line between the editorial judgments involved in compiling a prose article and compiling a list, and that the latter is not more difficult to verify or filter down to the encyclopedic than the former, and this was my point.

Even if a list purports to be a complete listing of its items ("this is a list of all species of the bird genus Foo"), or starts intersecting items by different facts that may or may not be sensible to intersect ("list of U.S. Congressmen who own three cars"), the way in which those claims or intersections are verified, or determined to be encyclopedic, still does not differ in a list from how such statements in a prose article would be verified. An article on genus Foo would need to verify a statement therein that there were exactly 50 species in the genus (and what those species were) no less than a standalone list of those species would. And an article on an individual congressman or the U.S. Congress as a whole would require verification, and a showing that it was encyclopedic and not random trivia, for a statement that congressman Schmoe was one of fifteen who owned three cars. postdlf (talk) 18:24, 16 September 2010 (UTC)

Re Schmoe, given that I agree with the point you were making, I think it's correct that I was referring to someone else's use of your example. Re Foo, I'm a bit concerned that a list article that lists 50 species and discriminates to "all" members of Foo, thus allowing a reader to infer that there are exactly 50 (known) species, is the same as making the explicit statement that there are 50 (known) species. My approach has always been to assume a list, as with any other article, is incomplete. Furthermore, what if you have a source saying there are 50 species, and then one is sequenced and found to belong elsewhere. You have to take it off the list, but you might not have a source saying that there are only 49 species for some time. ‒ Jaymax✍ 00:41, 17 September 2010 (UTC)

We are starting to blend different things here. One is establishing that the topic of the list is notable... the second is establishing a criteria for inclusion in the list that is not based on OR... the third is establishing that the items listed are verifiable within that criteria. These are all different issues, even though they are all resolved the same way... by reference to reliable sources. Blueboar (talk) 18:42, 16 September 2010 (UTC)

I was trying to separate those things!  ; ) postdlf (talk) 18:46, 16 September 2010 (UTC)

WP:SYN does talk about the need for wikipedia editors to not combine content from Source One with content from Source Two in a way that presents something that is not in either Source One or Source Two. but that is different than saying all content in an article must be from a single source, which is actually generally advised against. Active Banana ( ^bananaphone 18:59, 16 September 2010 (UTC)

IMO, "Establishing a criteria for inclusion the list that is not base on OR" is the definition of "discriminate". This is one of the problems with "list of virgins" -- you're going to have to do a lot of digging and arguing to figure out what belongs in there or not. There are some people who try to get around this by saying "virgin only counts if you're over 18 and still haven't had sex". But this only compounds the "original research" problem. You're either making up your own definition of "virgin", or you're changing the list topic to something even less encyclopedic ("List of people over the age of 18 who have not had sex"). In summary, I think this is showing that "notable" and "discriminate" are two requirements that are different, but complementary. Shooterwalker (talk) 21:07, 16 September 2010 (UTC)

I agree for the most part on that assessment (I would argue that "notable" is clearly a part of being discriminate, but to simplify the language, this approach works too). However, one can also be too discriminate to the point of being trivial, creating cross-categorizations, or otherwise just creating a list for list's sake, so there's another "bound" here on top of notable and discriminate. --MASEM (t) 21:57, 16 September 2010 (UTC)

I used to think notability could do it all. But yeah, I think some lists can be notable in that it gets people to talk about it, but inlusion in the list is indiscriminate, vague, and full of point of view and bias. So I DO think "discriminate" and "notable" are two different things. I used to think "unencyclopedic cross categorization" was covered under "notable" too. That is, if there's third parties who have talked about "drug dealers who live in gothic architecture", then it would go from a ridiculous intersection to something truly of importance to human knowledge. But I might be wrong. What do you think? Shooterwalker (talk) 00:11, 17 September 2010 (UTC)

For the sake of semantic simplicity, I think it's more helpful to define "discriminate" as the stuff that relates to list items, and thus distinct from list topic "notability". I think that "encyclopaedic cross categorisation" is a notability issue, but there is a difference between "drug dealers who live in gothic architecture" and "drug dealers who are notable", likewise "rivers in montana with waterfalls" and "rivers in montana over 3000mi". In both cases the first is cross-categorisation, and the second is discrimination. I've been trying to work out how to express that distinction as a test; and it's not easy. What is clear, is that in the discriminating cases, it would be fine to title the articles "drug dealers" and "rivers of Montana", but that would not make sense for the cross-categorisation cases. ‒ Jaymax✍ 00:30, 17 September 2010 (UTC)

So may I suggest we have three points to define an appropriate list (Ignoring if the list itself is a notable topic like Nixon's enemies or AFI 100 Movies)

1. Notable - the list should about a notable topic.
2. Discrimiante - the list should neither be too broad nor too narrow
3. Stand-alone per Summary Style - the list is too large to fit into the larger topic that it is about.

(the last has seemed to be implicitly implied in the discussion here) --MASEM (t) 01:34, 17 September 2010 (UTC)

1. Notable - 'about' is too broad - 'instances of' ?
2. Discriminate - ... objective criteria. Something about sourcing for every entry must show entry meets criteria? ‒ Jaymax✍ 01:52, 17 September 2010 (UTC)

Well, "discriminate" the way I'm thinking about it, it's not about sourcing though a good point is that the list criteria should be something that an element can be readily sourced to through reliable sources. eg, List of virgins is impossible to proof from the start. Instead, I'm thinking more about the membership size of a given list definition; is it finite, relatively unchanging, etc. etc. --MASEM (t) 01:55, 17 September 2010 (UTC)

1) If those points will allow all the test cases (plus those mentioned above) to exist to everyone's satisfaction, then it should be sufficient. If not, then it needs to be expanded to allow for them. (Or alternatively, one of the other proposed solutions needs to be adopted for the page-types that do not fit within those 3 points).

2) For discussion purposes, redlink examples are sometimes useful, but good bluelinks (that get argued over) might be better. E.g. for sprawling multi-page lists, the first item in Wikipedia:Featured lists#Culture and society (of which some components are Featured, and some are not) is a good example; and for split-out-for-size-sub-lists, something like the items in Category:Lists of things named after mathematicians (Euler and Einstein are the best examples there) might be useful.

HTH (does it? my comments above received no reply? I'll be quiet(er) if I'm getting core details wrong...! :) -- Quiddity (talk) 02:08, 17 September 2010 (UTC)

I can see where Jaymax is coming from, but I still maintain that his arguments are fundamentally flawed. This is a bone of contention, so I preface my comments by saying that I may be wrong ("Think it possible that you are mistaken"). I will try and work through my thinking with a new example that clearly illustrates my views, so I would be grateful if you will be patient with me.

Intuitively, I can tell that the idea that a list is notable because it is about a notable topic is a notability fallacy: the idea that any number of lists can be notable because another article topic is notable is not accepted in Wikipedia. However, it goes deeper than that: what is being implied is that a list is verifiable because another article topic is verifiable. Although this is being denied, this iswhat is being proposed: an exemption for lists from WP:N and WP:V.

At the core of this argument is original research. The reason why many editors are putting forward the idea that "a list is notable because it is about a notable topic" is that it gets around the problem of having to provide attribution for lists. If a list has not been published, it has been madeup. A lot of lists in Wikipedia are not verifiable at all, in the sense that neither the definition nor the the content is veritable - these are pure original research. Other lists are sourced from lots of different sources, but they are synthesis. At this point, this is where many editors will disagree with me, so let me provide an simplistic example:

List of items that is not synthesis

1. This
2. list
3. is
4. not
5. synthesis

N.B. This list has been taken from a single reliable published source.

List of items that is a synthesis

1. This
2. list
3. is
4. a
5. synthesis

N.B. This list has been taken from multiple reliable published sources.

What has been suggested to me is that "List of items that is a synthesis" is not a synthesis, on the grounds that it is simply a summary of sources about a related topic, and this is a reasonable argument, I will concede. However, I think this is a category mistake, because there is no evidence to suggest that it is not synthesis.

To illustrate this, let me read back the second list in prose form:

This list is a synthesis.

The point I am making here is that in prose form, we would reject this statement as synthesis straight away. If the sentence contained a strong expression of opinion, we would delete it without thinking twice: it is clearly synthesis.

The source of disagreement is that we don't consider homemade lists to be synthesis is because the list members appear to be separate and verifiable "factoid"s in their own right, and so synthetic blurring does not seem affect them. However, this ignores the fact that the list as a whole is made up from a rag bag of sources whose relationship is inferred, but not proven. It seems to me that at the core of our disagreement is the disputed issue of whether a list is a topic in its own right (a set) or whether it is merely the sum of its associated parts.

I think we are agreed that the border between assembling coverage about a topic from reliable sources and synthesis is a matter of subjective judgement, and it is not easy to determine when the cut-off point has been reached. Having said that, to demonstrate that a list is not original research or synthesis, I would argue that to justify the inclusion of a list in Wikipedia, it should have been published already, because (to paraphrase WP:SYNTH) a reliable source would be needed to demonstrate that the members of a list belong in that list.

The above example is very simplistic, so it might be too trite for out purposes. But looking at large lists such as List of Heroes characters, I would say that this list is not only a synthesis of sources, but at its core it is basically an essay topic, whose subject matter is not defined in any way. Masem has argued that this list's problems are purely content based, but I think it goes deeper than that: it is basically an unverifiable, unpublished topic in its own right, like the example above.

At the core of my views is that there is a burden to demonstrate that a list topic as a whole has been published to demonstrate it is not original research. Whether or not the members of a list are verifiable or notable, or are members of notable category has nothing to do with he list itself.--Gavin Collins (talk|contribs) 09:46, 17 September 2010 (UTC)

No, what you are saying about original research and synthesis is not following established measures and what's even required for fundamental encyclopdia building. Assembling a number of sources and putting them together to make a statement is "synthesis" by definition, but not necessarily synthesis outlined at WP:SYNTH to push a specific point-of-view, which is the type of synthesis we disallow per NOR. Every WP article (non-list) is doing the type of synthesis you are arguing we cannot do for some types of lists. This is a necessary measure of synthesis for proper encyclopedia building from a number of sources. Just because in your example of list-to-prose there would be a source next to every word does not discount this idea; if they are all taken from discrete sources then each source needs to be cited. It looks poor but that's maybe why the original editors used a list to avoid that. This continued claim that list articles cannot be assembled from an agglomeration of sources is bogus when you put it next to standard article creation processes. This is not to discount the fact that one can still create a list definition/topic that where, despite all the entries being validated in meeting it, is used to create truly original research to push a POV. "List of rich widows who's husbands died in unexplained manners" for example would beg the question if the widows off'd their husbands themselves, even though one would be able to likely source news stories for each case on there. That's an issue of topic selection though, not original research in assembling the list. --MASEM (t) 13:08, 17 September 2010 (UTC)

I think that argument is reasonable, but its not complete. There needs to be evidence that an amalgam of sources is not synthesis, and your argument that whether a list looks "plausible" does not do that. Taking your example, if a "List of rich widows who's husbands died in unexplained manners" sounds implausible, then a "List of rich widows" does seem to be entirely feasible and plausible by the same measure. However, there is no evidence to disprove that either list has been made up or plagiarised from a published list - plausible or implausible, they are both in the same boat, they are both madeup. There is some evidence to suggest I am right regarding this example, as the equivalent List of richest people is a published list, not a homemade list.

It is not a bogus to say a reliable source is be needed to demonstrate that the members of a list belong in that list, i.e. the list has been published. Simply claiming that a list is not synthesis is not enough, and this principle is core policy. Plausibility is not a recognised as a valid inclusion criteria, although it might provide a good explanation as to to why homemade lists are tolerated.

I recognise that I could be entirely mistaken. But this is the same view put forward by ThemfromSpace, but we have approached this issue from entirely different connections, and independently come to the same conclusion. --Gavin Collins (talk|contribs) 17:42, 17 September 2010 (UTC)

Here is how I think it goes...

Acceptable List: List of Xs that Y

Lede: This is a list of notable Xs that have attribute Y. Xs that have attribute Y have been discussed in many sources <ref to a few sources that establishes that the topic is notable.> To qualify for this list, an X must be notable (as demonstrated by having an existing Wikipedia article, or through citation to a reliable sources) and must include sources that demonstrate that the item is an X with attribute Y.

• Item 1 <ref to show item 1 meets inclusion criteria>
• Item 2 <ref to show item 2 meets inclusion criteria>
• Item 3 <ref to show item 3 meets inclusion criteria>
• Etc.

Flawed List: List of As that B

lede: This is a list of A's that B. A's are notable. Items listed here are all notable As and have attribute B

• Item 1
• Item 2 <ref that shows Item 2 is a notable A, but does not mention B>
• Item 3 <ref that relates Item 3 to A><ref that relates Item 3 to B>
• Item 4 <ref that mentions that Item 4 is an A that has attribute B in passing>
• etc.

In the first example, we properly establish through sources that the topic is notable... We then we set up a non OR criteria for inclusion (based on those sources, and thus not OR), and finally each item is sourced to show that it meets that inclusion criteria.

In the second example, we are relying on the notability of A, without properly establishing that B is note worthy attribute for an A to have (thus attempting to inherit reliability). The inclusion criteria is flawed as no source establishes that combining A and B is not OR. The items are flawed as they are all improperly sourced, even in connection to the flawed criteria. Blueboar (talk) 18:21, 17 September 2010 (UTC)

Either I am not explaining myself, or maybe the point that I have made that notability is not inherited is not accepted. How can you demonstrate that items on your list are notable, or that they should be included in a particular list? I think these are two issues that need to be agreed on or refuted: that being a member of a Wikipedia category (even a notable one) is not a justification for the inclusion of a list, as this is self-referencing, nor does is notability of the list itself inherited from a category or an association with a notable topic or group of notable topics. The notability of one topic is not evidence that another topic is notable.

How can you tell if a list is flawed if there is no original list source to compare it with? Going back to the WP:OR issue, there has to be direct and detailed evidence that a list exists, it cannot be inferred from an amalgam of sources. --Gavin Collins (talk|contribs) 01:08, 19 September 2010 (UTC)

We are approaching this from the wrong perspective

For lists, a unique, enumerative style of WP article, we need to be answering the following questions.

1. What are the distinct characteristics of a Good List, ie a list worthy of WP
2. What are the distinct characteristics of a Bad List, ie a list unworthy of WP

If we can agree on these characteristics, we can surely agreed on the policies/guidelines necessary to ensure our lists meet the desired characteristics of good list. --Mike Cline (talk) 03:16, 17 September 2010 (UTC)

It's worth noting that (a) the list is rarely the entirety of the list article, and (b) those characteristics may well also apply to lists which are not the primary article content.

I would kick off Good List characteristics with my thoughts as follows (very much worded in my style, I don't recommend this language):

• The list is a collection of instances of a class, and that class should be a notable topic (meets WP:N)
• The list has well defined criteria, which can be objectively applied, which specify the subset(s) of the class which the list includes. Criteria should be such that the list is encyclopaedic (particularly with regards WP:NOT)
• Each list entry must be (WP:V) sourced to demonstrate it is an instance of the class, and that it meets the criteria.
• Lists, other than strictly navigation lists, should go beyond merely identifying the instances, and should do so in a consistent manner for all list entries (as far as possible), and additional list item information must also be sourced.
• Lists should form part of an article about the topic generally, even if the list is the substantive part, unless size restrictions make this impossible. ("List of" should be avoided in titles, unless there is already a prose article dealing with the topic.)
• (something about) oversize lists broken out into multiple articles, and lists repeated under different sorting/grouping in different articles due to mechanical constraints...

Only the first item is specific to stand-alone lists, and even then is a good rule-of-thumb for in article lists. ‒ Jaymax✍ 04:25, 17 September 2010 (UTC)

• I'm not too far away from Jaymax. I think the requirements are threefold:
  1. Notable: The class of things that the list covers is referenced to third-party sources.
  2. Discriminate Class: Deciding what does or does not belong in the list is obvious, and is unlikely to be challenged in most cases.
  3. Summary Size: The list is neither too large nor too small.
• To turn it on its head, this would be a bad list:
  1. Not notable: Someone fails to find third-party sources, or can only find third-party sources for a different class. (e.g.: sources cover "Drug addicts" and "self-help gurus", but not "Drug addicts who became self-help gurus".)
  2. Indiscriminate Class: Something where it's impossible to draw the line, or the line is drawn arbitrarily. Like "movies that take place in nature".
  3. Summary Size: like "list of virgins" at one end, or "list of non-protestant U.S. presidents" at the other.
• ... plus whatever is in WP:NOT. Like, no price lists and so on. Shooterwalker (talk) 06:49, 17 September 2010 (UTC)

I think you are missing a key point here. A reliable source would be needed to demonstrate that the members of a list belong in that list, i.e. it is a published list. The only dsicriminate class of lists are those that have been published. Everything else is simply an indiscriminate list of stuff. --Gavin Collins (talk|contribs) 09:51, 17 September 2010 (UTC)

Gavin, could you provide us two examples: One example would be an article that demonstrates your ideal list characteristic as stated above, ie published list and the other would be an article that violates that characteristic. Ideally these examples will be ones that have not been discussed in this RFC todate. Ideally these examples will represent quality lists differing only in the published list characteristic and not burdened by other flaws. Ideally these examples will speak for themselves and not require alot of dissection from us or interpretation from you to make the point. Could you do that for us.--Mike Cline (talk) 14:48, 17 September 2010 (UTC)

To add to Mike Cline's question for Gavin... does it matter if the third party sources cover the list of things as a list, or can they cover it as prose? (In my opinion it doesn't matter, so long as it's covered. The decision to switch between list-style and prose-style is an editorial decision, not an original idea.) Shooterwalker (talk) 15:51, 17 September 2010 (UTC)

Here are two examples you should be familiar with:

1. A verifable, notable list: List of Schindlerjuden
2. An unverifiable, homemade list: List of people from the Isle of Wight

For some reason, I believe the second one qualifies for inclusion under your proposal. --Gavin Collins (talk|contribs) 19:34, 18 September 2010 (UTC)

- Thanks Gavin. Indeed the List of Schindlerjuden -- a published list would set a very high inclusion bar for any list style article. It is interesting to note that if the source of this list was not in the public domain, this would be a classic copyvio. In other words, absent the Public Domain nature of the information, the editors who created this list just copied someone elses work. And you are correct Gavin, that current list policy (not my proposal) allows lists such as List of people from the Isle of Wight-- a discriminate compilation of notable individuals with a significant association with a notable topic. Since propably every individual on the list could be legitimately (and verifiably) incorporated into prose in the Isle of Wight article, it seems reasonable that a SAL compiles them and presents that otherwise legitimate and verifiable content into a more useful form (for size, readability and navigation reasons). Why should that content be valid in one context and not another? There is nothing unverfiable, POV or OR in the article, thus it meets our standards. And if not one word of our policies changed about lists from today on, it would still meet our standards.--Mike Cline (talk) 19:53, 18 September 2010 (UTC)

Gavin, if the List of people from the Isle of Wight were retitled "index of Wikipedia articles about people from the Isle of Wight" (all else remaining the same), would you still adjudge it "unverifiable", or "homemade" in violation of WP:OR? Please answer yes or no, and then explain why that would or wouldn't change your judgment. postdlf (talk) 20:07, 18 September 2010 (UTC)

The issue of copyright is helpful in relation to List of Schindlerjuden, in the sense that providing a source for a list demonstrates that is not a copyright violation, or that its terms of use have been honoured. I have discussed this issue of copyright already in relation to the New General Catalogue, and a list is no different from any other source in this regard - you have to cite your source, and any restrictions attached to it. If the "List of Schindlerjuden" is the subject of any copyright restriction (such as owners of the source document, Yad Vashem, have to be named as such), then citing the source and stating the restriction is the right thing to do. We could try and get around this restriction by making up a homemade list by reconstructing it from different sources to avoid having to cite the list directly (e.g. by using the book or the film as a source), but essentially to to do this would be plagiarism. I think this is a knock-out argument as to why a list must be verfiable as a whole, not just in part: to demonstrate that a list not been plagiarised from an original source. Homemade lists are always at risk of having been plagiarised from a previously published source.

With regard to List of people from the Isle of Wight it is not reasonable to simply assert that current list policy allows such lists, because to suggest that a list can be included simply because the individuals on the list have a significant association with a notable topic is based on a notability fallacy, and there is not policy or guideline that justifies this approach. I have debated with Masem about WP:SAL, and it is my view that the approach it adopts is intellectual rubbish. Just because individuals on the list may be the subject of their own article, that does not mean the list is notable. The knockout arguement against such approach is that, while many of the individuals might have their own article in Wikipedia, this may not be evidence that they are notable, nor does it validate the list inclusion criteria. Using Wikipedia as a justification for inclusion is an self-referencing, and whether it was labled as an index does not change that fact. --Gavin Collins (talk|contribs) 00:59, 19 September 2010 (UTC)

No one else in this discussion agrees with your interpretation of WP:NOTINHERITED, the supposed basis for your repeated claims that everyone else is committing a "notability fallacy." Those who have commented on it have expressly rejected your interpretation as incorrect, and argued that it does not apply to this issue. Can you provide a diff from this discussion that shows otherwise? If not, please stop continuing to repeat your interpretation as if it were objective fact, without advancing your argument or responding to counterarguments. That is just tendentious and disruptive.

The same goes for your "list is [not] notable" complaints. A clear consensus of participants have rejected that mode of analysis, and I know I have repeatedly pointed out to you that I am not claiming a list is or isn't notable because I don't consider that a useful approach to a list. Can you provide a diff from this discussion that shows otherwise? You just repeat yourself as if you haven't read anything anyone else has written, or you haven't understood it. Can you show me you understand why I and others have a problem with you doing this, and why we feel it isn't constructive? postdlf (talk) 14:45, 19 September 2010 (UTC)

In fairness to you, I freely admit that my interpretation may be mistaken. However, given the clear and unambivalent way in which your assert that the notability of "List of X" is derived from topic X, I could not describe this scenario in any other way other than a list inherits notability from another article topic. Regardless of what you think the consensus is (lets say you are right), then I still think the problem of assuming notability jumps from an article to a list is that this idea has been debunked by WP:NOTINHERITED as being a fallacy. I see this as just another variant on Masem's pet proposal, namely that spinoff articles inherit notability from their "parent", and that proposal did not achieve consensus, but his arguments have always tried to get around WP:NOTINHERITED in various convoluted ways. His is a plausible argument, but its Achilles heal is based on the idea that notability can be infered based on personal opinion, rather than presumed based on verifable evidence of notability. The key to verifiable evidence is that it is based on significant coverage that addresses a topic directly and in detail. I see no direct or detailed evidence that "List of X" is notable, even if there is direct and detailed evidence that topic X is notable in your scenario. --Gavin Collins (talk|contribs) 18:12, 19 September 2010 (UTC)

You are completely wrong on my account, Gavin. 1) Notability only determines when a topic should be given its own article 2) Notability does not limit coverage of a topic 3) Coverage of a topic may have to span several articles due to SIZE and summary style (and in some cases, this is where some lists are placed). Ergo, spinouts and supporting articles have a notable topic, that of the parent one, not a new one that is inherited as you keep insisting (though it is certainly possible that they are a new topic in-of themselves). In most cases of these lists, that's the same rationale we're establishing here and why there should be no different between a list that would be part of a larger article and one that is standalone. Also it should be reminded that NOTINHERITED is an essay and thus has no consensus behind it though I again, my stance requires no use of inherited notability. --MASEM (t) 18:20, 19 September 2010 (UTC)

Masem, I agree with your point... but with a caveat... if you spin out material from a parent article in the wrong way, I think you can end up creating a new topic. Thus, some spinouts are fine, while others are not. It depends on what was spun out, and how it was spun out. Blueboar (talk) 18:24, 19 September 2010 (UTC)

Oh sure, there's a lot of finer details; SPINOUTs suggests several as triming and other steps before creating one, and obviously some spunout content could be problematic (I would not spin out the plot of a movie from a movie article if it is long, for example). However, the point here is that not all immediately-obviously-notable spinout articles/lists are bad as Gavin tries to insist by the NOTINHERITED argument. If the spinout contents would have been something in the article without a SIZE limit in the first place, but not necessary to the general understanding of the topic, it's likely appropriate as a spinout. --MASEM (t) 18:29, 19 September 2010 (UTC)

Masem always ignores WP:AVOIDSPLIT when he invokes WP:SIZE, which is why I characterise his views as convoluted. Coverage of a topic may have to span several articles, as he says, but in order to split out a sub-topic into its own standalone article, there needs to be evidence that the sub-topic is notable in its own right. Without such evidence, his views are based on WP:NOTINHERITED in all but name. --Gavin Collins (talk|contribs) 18:35, 19 September 2010 (UTC)

AVOIDSPLIT does not disallow spinout articles that don't have a specifically individual notable topic as you think it does. It says avoid them by doing other things first (trimming, etc.), but outlines the spinout as the last resort. This is consist with the use of spinouts across WP. But even more so, per this RFC, it suggests that if a spinout is along the lines of "X of Y" where Y is a notable topic, and X of Y would be something normally covered as part of Y, there's good reason to support the spinout being notable under Y itself. --MASEM (t) 18:58, 19 September 2010 (UTC)

Gavin, you're continuing to do the exact same thing: claiming someone else's view violates an essay, WP:NOTINHERITED, without elaboration and according to an interpretation of that essay that no one else here agrees with, an essay that isn't binding policy or even a guideline to begin with. What do you think that accomplishes? I've asked you to put up or shut up: show us that someone other than you agrees with your interpretation (for starters), and figure out a way to explain your opinion in a way that actually advances your argument rather than just repeat your opinion. You have not. In my view, apart from the issue of whether it's desirable to treat an essay as a binding law, your interpretation completely ignores the actual language of the essay: "[N]otability of a parent entity or topic (of a parent-child "tree") does not always imply the notability of the subordinate entities. That is not to say that this is always the case...or that the subordinate topic cannot be mentioned in the encyclopedia whatsoever. Often, a separate article is created for formatting and display purposes; however, this does not imply an "inherited notability" per se, but is often accepted in the context of ease of formatting and navigation, such as with books and albums." (boldface emphasis added). So you're claiming that people are committing a fallacy by violating an essay, an essay that does not even set forth anywhere near an absolute prohibition on justifying a standalone lists or articles by their relationships to notable topics. So I'm going to ask you again to stop merely repeating an opinion without providing different or additional support for it. Regardless of whether you are "mistaken" or not, how you are going about this is only disruptive and tendentious. So unless you have something new to say, I don't see you having a further role to play here. postdlf (talk) 21:59, 19 September 2010 (UTC)

Personally, I think you are both trying to form a rule where no rule can exist. Sometimes a sub-article does have the same topic as a parent, and notability of the parent does logically carry over to the sub-article. But at other times this is not true. The spin-out is actually a new topic. You are both correct (and incorrect) because both types of spin outs exist...

I am going to suggest a possible solution... what if we require that notability be either established or re-established in every article (regardless of whether it is a parent or a spin out)... if we do this, I think we eliminate the problem. If the topic of a spin out is legitimately the same as the parent article, re-establishment should be easy... the work has already been done in the parent. a few sentences added to to the lede of the spin out (essentially repeating what is stated in the lede of the parent article) should be all that is needed. The sources used to establish notability in the parent article will establish the notability of the sub-article. If these sources don't establish the notability of the sub-topic, then editors know that there is a problem... we need to either re-work the spin out so it does match the sources, or freshly establish notability for the new topic. Blueboar (talk) 19:43, 19 September 2010 (UTC)

I agree with your starting point that there's no absolute either way to prohibit or permit a split-off, but I'm not sure I follow on your solution of re-establishing notability. Could you give an example of an existing subtopic split-off (or one that failed AFD) that would require a new notability analysis? postdlf (talk) 22:03, 19 September 2010 (UTC)

I think Blueboar idea of "sub-article" is a little confused. Either he means a sub-topic that is sufficiently notable to justify its own standalone article or list, or he means something else that does not exist. The idea that "The sources used to establish notability in the parent article will establish the notability of the sub-article" is, in my book at least, the perfect definition of inherited notability, and that idea has been debunked as a notability fallacy. How notability can be "restablished", I have no idea, but perhaps Blueboar can provide an an example.

Getting back on topic, I disagree with Shooterwalker's approach because it is not based on direct evidence that a list is notable:

1. Notable: A list inherits notability from a class of notable topics that the list covers or is in some way "related".
2. Discriminate Class: Deciding what does or does not belong in the list is based on subjective importance.
3. Summary Size: The list size is also determined by subjective importance.

If I can pick holes in this approach, then other editors can too. --Gavin Collins (talk|contribs) 23:24, 19 September 2010 (UTC)

None of these are based on established practices of the cited essays. --MASEM (t) 02:04, 20 September 2010 (UTC)

...and I think that's really all that needs to be said about that. No need to waste any more time. postdlf (talk) 03:58, 20 September 2010 (UTC)

Back to Blueboar, to rewind a bit so we can resume the discussion... As I said (to Blueboar), I agree with your starting point that there's no absolute either way to prohibit or permit a split-off, but I'm not sure I follow on your solution of re-establishing notability. Could you (Blueboar) give an example of an existing subtopic split-off (or one that failed AFD) that would require a new notability analysis? postdlf (talk) 03:58, 20 September 2010 (UTC)

List minimums

Is it reasonable to think that establishing a [generally] recommended minimum number of entries for a SAL? Many new lists get created with 2 or 3 entries, and it is unclear as the probability that additional entries could be forthcoming. If the WP:SAL read something like this: Standalone lists should generally have at least X discrete entries. Lists with fewer entries are generally better included in the related article as an embedded list. I don’t know what a reasonable minimum would be, but having one might prevent some really flawed SALs from being created.--Mike Cline (talk) 14:58, 17 September 2010 (UTC)

• It's a good question. The best way to answer might be to go to actual practices and find some short lists. Shooterwalker (talk) 15:53, 17 September 2010 (UTC)

My hope (call it an agenda if you like) - and this is something I have only come to after becoming involved in this RfC - is to dilute the distinction between list articles and prose articles. List of Smithsonian Museums came up at some point, and what struck me was two-fold. (1) The prose content of the article was significant (and interesting), and (2) the article would be better titled just Smithsonian Museums because that would make it a good home for further development of an article on that notable topic, where "List of" might prevent same.

I have no real problem with a stub article, that includes a list of three instances, if there is not an existing topic article - providing it's not title "List of..." and includes a descriptive lead. I can't see the sense in a SAL of 20 items, if there's a perfectly good article for the topic where the list would properly belong. List formatted content is just content at the end of the day; To me is there should be reasonable justification for separating listed instances from prose facts, and in the absence of such justification, they should be contained in a singular article (even if it's a stub which has started out incorporating a short list).

So to answer Mike's Qn, I don't think there is value in an arbitrary minimum per se, but rather some broader reasons to avoid creating articles as SAL in the first place, which would cover the issue. ‒ Jaymax✍ 01:28, 18 September 2010 (UTC)

I don't think that an arbitrary limit is a good idea. List of cemeteries in San Francisco would be very short (two?), but it would be a reasonable enough way to present information about the history (they removed nearly every grave in the city after the earthquake) and the very few cemeteries that remain. With a section on history, and a couple of paragraphs about each remaining location, it could be a perfectly acceptable article. WhatamIdoing (talk) 00:15, 20 September 2010 (UTC)

Mini-RFC: where do we put this stuff?

We have a consensus on a few things. (Most people agree on a few principles.) We're getting to the point where we might want to close this discussion. The way I see it is we have a few options:

1. Add a section to WP:N about lists.
2. Spin WP:SALAT out into a new guideline.
3. Move the WP:SALAT section to the top of WP:SAL and add something to the lead.
4. Something else...
5. Tweek List naming conventions in WP:SAL --Mike Cline (talk) 16:34, 17 September 2010 (UTC)

I'd pick WP:N. But what does everyone else think? Shooterwalker (talk) 16:31, 17 September 2010 (UTC)

I'm for 1+3+5+more. WP:N needs something SHORT that says list articles should list instances of a notable topic. SALAT needs rework and elevation. But SAL has become bloated with a lot of stuff that is list-generic rather than stand-alone specific, and that (to my mind) helps make it seem a bigger deal than it perhaps needs to be. Also, a lot of tangential but intertwined stuff (is that possible in 3D) relating to list content issues has been discussed, and that should be addressed somewhere also; but it's not relevant to N or SAL, except to emphasise that list content is a separate matter. (or am I making it all too hard?) ‒ Jaymax✍ 01:46, 18 September 2010 (UTC)

I would be in favor of a short paragraph in WP:N (I do think something is needed there)... pointing to a re-worked SAL for further explanation and examples, etc. Blueboar (talk) 02:06, 18 September 2010 (UTC)

So it looks like there's an emerging consensus for adding/modifying both WP:N and WP:SAL? No spin-outs... Shooterwalker (talk) 03:16, 18 September 2010 (UTC)

Thinking atype: WP:Lists would be the place to talk about what makes for good lists generally (anything not article-specific). WP:SAL for notability issues, oversize issues, and SAL naming etc. ‒ Jaymax✍ 05:11, 18 September 2010 (UTC)

Agree, think 1) is needed, but only something very short, then a bit more in 3). --Stefan ^talk 08:15, 18 September 2010 (UTC)

case study

I waited until an AFD on this list closed because I didn't want to engage in canvassing.

• List of Star Control races

This embodies more or less what I feel is MINIMALLY necessary for a list. Maybe not 70k worth of details. This list is in rough shape. But it meets the bare standards of inclusion.

1. Notable: There are several third-party sources that talk about this class of things (races in the Star Control series). Not just Star Control in general.
2. Discriminate: The bar for inclusion is obvious and unlikely to be challenged in most cases. Any alien race in the game is put in the list. (--Gavin Collins (talk|contribs) 15:39, 20 September 2010 (UTC)There's a small chance that someone might say "do we really include information on the unplayable races that are mentioned in the game", but that wouldn't invalidate the idea that most of the members are obvious and involve no original research to identify.)
3. Summary style: this class of things isn't so huge (list of alien races in fiction) or so small (list of Star Control 2 versions) that it would be inappropriate for an encyclopedic article.

If you look at the AFD on this list, there were some counter arguments.

1. Notability: Some people really wanted third-party sources for the individual list members, which is inconsistent with WP:NNC, and (I think) inconsistent with consensus here.
2. Discriminate: Some people thought the word "races" was confusing. But to me it's not a crippling issue. The title could be "aliens" or "characters" and the list would still be basically good.
3. Summary style: Some people thought there was too much detail. I'd agree with that. But it's not a reason to get rid of the list. It's a reason to clean it up.

Just thought some people might be interested as it helps focus the discussion. Shooterwalker (talk) 05:47, 19 September 2010 (UTC)

Here's the problem that makes it difficult. It's first off a list of fiction, and thus new issues dealing with WP:NOT#PLOT and WP:WAF come into play. It's also a video game series, and we strongly warn against WP:GAMEGUIDE type content. In terms of determining a list, the former issues likely come into play with Discriminate - an atypical list for fiction (that of all alien races in a series) is going to be scrutinized more if it's not written well out-of-universe. The latter issue on GAMEGUIDE points also towards the discrimination aspect, in that generally on the VG side we don't consider such lists appropriate. In other words, this list fails primarily due to being indiscriminate; it's a notable topic and too large for that topic, but covers a lot more outside of what we'd expect for an encyclopedia. But I'd put it more that cleanup can be done before outright deletion. --MASEM (t) 06:01, 19 September 2010 (UTC)

Its more than just a content issue. Nominally, this article takes the form of a list, but in reality it is an essay. Like all essays it contravenes WP:NOT#OR because its subject matter (Star Control races) is not the subject of any significant coverage that has been published by a reliable source. It is in exactly the same boat as List of Masonic buildings: it is an orphan list, without any real world parent. --Gavin Collins (talk|contribs) 08:56, 19 September 2010 (UTC)

Gavin, I don't think anyone but you believes the Masonic buildings list is a helpful example to keep bringing up, because the problems particular to that list are not illustrative of the more general problems you try to use it for. Several people have said that. Can you show me a diff from this discussion that shows otherwise?

You seem to be under the impression that anything not notable is therefore necessarily violative of WP:OR. I don't think anyone else in this discussion believes that. Can you show me a diff from this discussion that shows otherwise, or correct me in my interpretation of your opinion? That something doesn't satisfy WP:GNG does not in and of itself prove that it violates WP:OR. A statement of fact could be verifiable to a single source, not multiple, and be the subject of minimal coverage not significant, or be sourced to a single primary source, and still be verifiable and not WP:OR. The WP:GNG test is to determine whether something merits inclusion. It can function as a good safeguard for preventing WP:OR, but it is not the only way to prevent WP:OR. It seems that you disagree with this, but you can't articulate why and you can't show any support for that opinion. And the comments in this RFC have shown that there is a strong consensus in favor of the interpretation that I just described. Can you show me a diff from this discussion that shows otherwise?

Given that many of us have expressed that we think you are just repeating your solitary opinions, I think it's fair to ask for some showing that what you keeping claiming are "points" that other people are ignoring are actually accepted by someone other than you, because we read you the first time, disagreed, and moved on. You have not. And if you can't, then you need to figure out another way to participate, or to stop participating here. postdlf (talk) 13:53, 19 September 2010 (UTC)

First off, this list needs to be understood as a subtopic of a notable topic though arguably not notable in its own right. It's clear that if the video game series was not notable, this list would not be appropriate.

I think a necessary part of any video game article (or series of articles about a game or game series) is a description of the characters that players can play. Further, the "history and reception" section of this list provides quotes from game reviews showing that the diversity of aliens in the series was one of its big selling points; the aliens were "noted" even if not individually.

That the list is written primarily in in-universe style is its biggest flaw. I would expect a description of a video game character to tell me what I can do with it when I play it and how that differs from using other characters, but this list mostly treats them as just elements in a narrative without grounding it in game play (see my complaints about this in an old-timey AFD for another video game character). But this is a matter for cleanup, not deletion. That these aliens are part of the game, have particular names, and do particular things in the game is verifiable. Whether it's necessary to go into non-player character aliens is a matter for further editorial judgment on what that contributes to understanding of the topic. It's possible that once the list is rewritten (possibly changed into List of player characters in Star Control) that it won't be too substantial to be included in Star Control, the article on the game series as a whole, or broken into sections for each individual game article (all of which currently just link to this list), though I suspect much of the reason for keeping it as a separate list is that each successive game reuses player characters introduced in the previous ones. And that might be one of the best reasons for keeping a standalone list: when its entries pertain to multiple articles, such that there is no one single merge target.

Anyway, that's my analysis for this kind of list. Tolerate it notwithstanding its obvious flaws because of its relationship to a notable topic, clean it up, and then figure out whether to keep it standalone or merge it. postdlf (talk) 14:33, 19 September 2010 (UTC)

I think you may have missed a key point: Star Control races is not a notable topic. I don't see how you can say a "List of Star Control races" is notable if its imediate parent topic is not notable. Or maybe I am mistaken, perhaps you are implying that the List is notable, because both it and Star Control races are notable because Star Control is a "related" article. Either way, an infered "relationship" is not evidence of notability. Whether "List of Star Control races" is the child, grandchild or third cousin twice removed, it still needs to demonstrate that it is a notable list in its own right, as notability can't be inherited. --Gavin Collins (talk|contribs) 23:08, 19 September 2010 (UTC)

But there are third-party sources that cover Star Control races in direct detail, and even go so far as to describe them as a notable reason for the game's critical and commercial success. It's a textbook case where the WP:GNG has been satisfied. Shooterwalker (talk) 23:37, 19 September 2010 (UTC)

I think that is pure speculation. If Star Control races is a notable topic in its own right, then I think it would have been split into its own article some time ago. I think the reality is that Star Control is a badly sourced article, and splitting it into two badly sourced articles would be a good example of WP:AVOIDSPLIT. For that reason, the existence of Adding a List of Star Control races" does not improve things either.

"List of Star Control races" is probably a good example of how lists are being used to circumvent WP:AVOIDSPLIT and WP:NOT. Most editors are quite tolerent of lists, because they are assumed to be low on content, and not "serious" mainspace pages. But I think it is about time we put an end to dumping topics that are not notable into lists in the hope that the conflict with content policy and notability guidelines will go away. --Gavin Collins (talk|contribs) 08:51, 20 September 2010 (UTC)

As another example... List of Kings of Arnor. Yes, the Lord of the Rings is a notable work of fiction... but does that mean everything associated with LOTR is notable? Even if we take an in-world perspective, the Kingdom of Arnor (and the genealogy of its Kings) is at best relegated (by the author himself) to being "background material"... stuck in an appendix and not part of the main story line. Is this really a notable topic? Well... strictly applying WP:N, I would say no. There are no reliable sources that are independent of the subject that discuss the topic. On the other hand, there are a lot of LOTR fans who will say it is a notable topic. If we were to send this to AfD I seriously doubt it would be deleted. Blueboar (talk) 12:56, 20 September 2010 (UTC)

There are certain fictional series that are going to be hard to make people balance lists appropriately; this is but one of those. We have to remember that for a lot of these, this was all before the Great Fiction Purge circa 2005 (this list being 2004). While I would not disagree that a cadre of editors will vote "keep" at AFD for it, I would also think that more reasonable editors would recognize it as a problem in its present state.

My suggest is that fiction-based lists are a problem wholly unto themselves with the issues of WAF and NOT#PLOT beyond the acceptance of the list itself. They are areas where most shouldn't have been spunout in the first place but were created when Wikipedia was younger and naive and notability never existed. It is an area that requires more unique cleanup, and that should only be worked through once we've gotten rules set for other things. --MASEM (t) 13:09, 20 September 2010 (UTC)

Dumping non-notable sub-topics into lists in the hope that they don't get deleted is a poor mans solution to a lack of notability, and creating a new ruleset to allow this, as Masem is implying, is not the way forward. What is need is clear guidance in WP:N that this is not a workable solution, but alas, this is not the case. Instead foonote 7, which is effectively an exemption for lists that gives rise to the idea that articles like List of Heroes characters are allowable, but they are not. --Gavin Collins (talk|contribs) 14:34, 20 September 2010 (UTC)

No, actually, using lists to contain non-notable elements that would be otherwise covered as part of a notable topic is established practice, originally outlined in Wikipedia:Deletion policy/Minor characters (ca 2005) and eventually moved to the footnote as you indicated. As it has been there for several years, it would suggest there's consensus for that.

If WP was more a traditional encyclopedia, I can understand getting rid of such lists and sticking to a stricter version of notability (one possibly engrained in policy). But we're more than just a traditional encyclopedia; there is nothing that limits what can be included per wide consensus save for the few edicts the Foundation passes down (such as issues of biographies of living persons). Consensus has decided that notability is a good measure for a topic to have a page on WP, but it has also decided that notability should not limit coverage and that a single topic can span several pages. In the cases of lists, if the same list would be acceptable within the main topic (as would be the case with Heroes here), a standalone list is probably going to be ok, give or take other concerns ("does it even need splitting?" "has trimming been tried?" etc.), because consensus has determined that a list of characters (embedded or not) is part of the coverage for a long-running TV show - that's part of its encyclopedic coverage. The problem with lists based on fiction is that it is very easy to poorly write them, and I will completely agree with you that List of Heroes characters is a poorly-written article for fiction. Same with the Star Control races. Poor writing, however, is never a reason to delete, unless attempts have been made to clean up without success. Both of these mentioned lists can be cleaned up to bring it to an out-of-universe approach to the races or characters. It has been determined by consensus both lists - embedded or not - should be part of the coverage of their respective topics, and since both main parent articles have SIZE issues, they are both prime as stand-alone lists that fit with the concepts developed in this RFC. Again, this is why Wikipedia is unique - the content is determined by consensus reflected into policy and guidelines, and is a living work, meaning that today, we may want to emphasizes grouped character lists over individual articles, while tomorrow consensus may turn and decide we need separate articles for every character regardless of sources. Obviously, its not going to change that fast, nor do I think we'll ever return in that direction in the near future, but this is just to hilight that we have to follow how the trends of consensus work, and cannot introduce language that is against consensus. --MASEM (t) 15:17, 20 September 2010 (UTC)

The consensus view that Masem describes does not exist, in the sense that WP:SIZE is not an accepted criteria for inclusion for lists and articles. Even if we ignore our disagreement about whether notability applies to lists or not, Masem is still ignoring WP:UNDUE. It is the general consensus in Wikipedia that articles and lists do not go into needless detailed exposition of every single detial about a topic. Rather, coverage given to topics (and sub-topics) is regulated by the weight afforded to them by reliable sources.

For example, the coverage given to the non-notable characters in the List of Heroes characters outstrips the coverage given to the television series by a factor of 5:1 or more (33,000 words vs. 6,500 words). If lists are not source from reliable sources, but madeup instead, how do we now where the appropriate balance is between comprehensive coverage ends and needless detail begins? --Gavin Collins (talk|contribs) 15:39, 20 September 2010 (UTC)

It's been general practice to use third-party sources to verify notability overall. Any detail about the subject can then be verified with a mix of all kinds of sources, and there is no rule against using primary sources (only that they are used "with care" and to avoid delving into "personal experience"). Going into too much detail is usually tempered by manuals of style, including WP:SUMMARY, and an overall consensus to make good editorial decisions on a case by case basis. Shooterwalker (talk) 16:52, 20 September 2010 (UTC)

While in principle, Gavin, you are correct that articles and lists should "not go into needless detailed exposition," there's no objective answer to what constitutes needless detail, certainly not an answer that would apply across all subject matter. That the list of Heroes characters can be trimmed is probably not controversial, but that wouldn't change the fact that there are too many characters to list in the main article on the show, and for those that don't merit independent articles there is no other centralized place to describe them. Because of the very type of show Heroes is (was?), it juggled a lot of characters all at once and killed off and introduced other characters throughout its run. It isn't "made up" that they are characters in Heroes; that is verifiable in each and every instance. And thanks to the internets not being paper, there are an increasing number of independent, reliable sources that review each and every episode of most network television series (even going back to many long-since canceled ones), providing not only detailed plot summaries but also production information and analysis of character development and continuity.[2] So it simply isn't productive to fret over such character lists existing, rather than about how to improve them. The sources are out there, and we don't delete what can be fixed. postdlf (talk) 17:07, 20 September 2010 (UTC)

There is an objective answer to what constitutes needless detail, and that is every topic or sub-topic that is not notable. The biggest problem I see with the idea that "List of X" is notable because "X" is notable is that there is a huge quantity of primary source material that can be brought into Wikipedia if this approach is adopted. For fiction there are List of characters, list of episodes and lists of fictional elements, fictional locations, timelines, not to mention the "Universe of X" type lists that used to be of fashion. Masem's mouth must be watering at this point, because he is very much in favour of extensive coverage of fictional topics, even non-notable ones, but consider this:

If company X is notable, then is "List of company X's products" also notable? Is a "List of company X's stores" notable? Is a "List of company X's factories notable? Marketing people will tell you that every company director wants to see a picture of his factory put in their marketing material, but why not have it embedded in a list on Wikipedia instead? Where I come from, this proposal would viewed as a public relations manager's wet dream - inclusion criteria that provide inherited notability that can be used to circumvent WP:NOT#SPAM, in the same way this approach is being used circumvent other parts of WP:NOT.

In some ways, this is already happening. I don't know what the justification is for List of Nokia products, but I know there are lots of of homemade lists like List of IBM products or List of acquisitions by Oracle that show there lots of scope for these sorts of lists. Perhaps I have a bright future as a hired photographer of factories and offices to put in these homemade lists - there will be lots of work for sure. Perhaps I will be able to fit in with the "consensus" after all. --Gavin Collins (talk|contribs) 21:06, 20 September 2010 (UTC)

Whether the "list" is notable is no more a coherent question than whether a prose article is notable, for reasons I've stated elsewhere and don't feel like repeating. Whether the subtopic being listed is independently notable is at best a proxy for the real questions, and is too poorly tailored to those issues to really help with any of those lists. Certainly IBM's products have been the subject of significant coverage in multiple reliable sources, which doesn't tell us what the best format is for presenting that information or how much detail is appropriate. The same is true of acquisitions by Oracle, a notable company that has apparently purchased a lot of notable companies, judging from the list's bluelinks, and though I see that list is currently sourced entirely to Oracle press releases, I can't imagine that those acquisitions went unreported by business papers.

The real question is what do these lists accomplish in this encyclopedia? And that's a question of the relationship of the list to what it lists, i.e., in what way the list pertains to notable topics, and I view that relationship as similar if not identical to the relationship of an article topic to the information and subheaders in the article. What understanding about IBM, or its products if independently notable, does that list of its products further? What understanding of Oracle, or the companies acquired, does a list of its acquisitions further? Or what would it accomplish to have a list of a company's factories? (answer: possibly quite a lot).

You can think of that as a "notinherited" problem if you like, though you are apparently unique in that approach, so it's hardly a persuasive or relevant rebuttal to anything. Even assuming you are interpreting it correctly (which would require ignoring its actual language), I can't say I'm particularly troubled by the thought that Wikipedia has lots of lists that are contrary to an essay that says certain deletion arguments regarding a guideline are generally unconvincing. postdlf (talk) 21:40, 20 September 2010 (UTC)

(ec)Why does it have to be notable? The only place that notability is used is to determine stand-alone topics. Beyond that, we include what consensus wants, generally limited through avoidance of indiscriminate information as outlined through WP:NOT. We are paperless, there is no reason not to include information that can be verified and summarized in an encyclopedic manner that would not normally be covered by a typical work.

And please stop mischaracterizing what I want. I want to see appropriate reins on fiction coverage, no "each character needs an article", but I will insist that there part of accepted coverage of a work of fiction includes a list of brief character and episode descriptions, whether part of the main article or as a separate article; beyond that, there needs to be strong justification (as the case for Star Control races). --MASEM (t) 21:50, 20 September 2010 (UTC)

Whether the "list" is notable is no more a coherent question than whether a prose article is notable is a common misconception that ignores the fact that only lists from reliable sources are notable. Homemade lists are not reliable, they are madeup. --Gavin Collins (talk|contribs) 06:47, 21 September 2010 (UTC)

Policies have consequences-Help us decide

Gavin, whatever language our core policies and our content guidelines use, that language and its interpretation has consequences in relation to existing content and new content. Ideally, any change in policy would retroactively apply to existing content and new content should comply with the new policy. Given your position that a List must have been previously published as a list to establish notability and inclusion in the encyclopedia how would you see such a significant policy change playing out in regards to existing content. Clearly from your examples above, your position—a published list--as a content guideline is not followed by the great majority of lists currently in WP. If, indeed, the community adopted your proposed content guideline for list—a published list as the norm for list inclusion, how would you see the community purging (or adjusting) existing content that did not meet the new content guideline? Since the new inclusion criteria is pretty black and white, ie. if the list has never been previously published, it shouldn’t be in WP, would your preferred method be to identify all the lists that meet the new criteria, and do a blanket deletion of those that don’t? Or would you prefer to take each individual list to AfD, even though we know upfront that ~95%+ of all current lists wouldn’t meet the new inclusion criteria?

Gavin, these are not idle questions. Understanding the consequences of any change in policy and how those consequences might play out is an important part of policy decision making. Most of the proposals re lists in this RFC would result in policy/guidelines that would help us identify and improve flawed list around the fringes. Your proposal might have major consequences. Help us decide what the consequences might be if the community decided to adopt your proposal.--Mike Cline (talk) 15:41, 19 September 2010 (UTC)

I think it would have important repercussions. But one thing is sure: Wikipedia would be much improved if lists that have been made up were replaced by lists that are sourced directly. Wikipedia should not be used as a platform for original research, as the compilation of lists that are not taken directly from reliable sources will not be taken seriously by readers. --Gavin Collins (talk|contribs) 22:55, 19 September 2010 (UTC)

We understand your motivation to improve WP with this proposal, but you failed to answer the question. How would you recommend we deal with the consequences? It is evident, that no matter how hard we tried, we would not find sources that support a previously published list construct that you propose. So, in light the the fact that most current lists would fail this inclusion critieria, how would you recommended they be dealt with if such a proposal was adopted.--Mike Cline (talk) 23:18, 19 September 2010 (UTC)

I too would like to read a more relevant answer from you, to Mike's questions in paragraph 1. Bear in mind that there are at least 56,000 List-class articles currently. -- Quiddity (talk) 03:47, 20 September 2010 (UTC)

I don't know if what Mike says is true or not. My only guess is that lists that are "navigational" (i.e. "List of X" where X is a category) are likely to be converted to categories, so that the boundry between "content-space" (aka mainspace) and "category-space" (aka categories) becomes clearer. Self-serving Wikiprojects such as Wikipedia:Outlines will be phased out, for a present they are little more editorial walled gardens that are based on madeup inclusion criteria that provide spurious justification for linkfarms. There are already moves along these lines in evidence: see User:Karanacs/Outline RfC draft, for instance. Beyond that, I have not idea. --Gavin Collins (talk|contribs) 08:38, 20 September 2010 (UTC)

Then it seems really a bad idea that has unknown consequences across thousands of existing articles, yes? Even the current task of trying to repair copyright infringement from one editor across ~10,000 articles by blanking them is drawing criticism (see Wikipedia:Administrators' noticeboard/Incidents/CCI)), do you think consensus would accept the removal of 56,000? What about lists that are already considered featured content? If anything, this alone tells me where consensus sits for these. --MASEM (t) 12:50, 20 September 2010 (UTC)

I think there lies the problem. If our policies and guidelines ignore or skirt around the issue of madeup lists, then when someone calls "time please", and suggests this practice ends, then there is going to be a lot of objectors. To some extent, I see this is happening already, and I am supprised that I have not been already burnt at the stake for sticking my neck out over this issue. But the longer we allow policies and guidelines to remain silent on how they apply to lists, the more difficult it is going to be to reach any consensus about how to deal with madeup lists.

As I see it, our content polices currently contain lots of exemptions for lists that have no intellectual basis along the lines of "WP:NOT#DIR or WP:NOT#INFO prohibits certain types of list... but that does not apply if the list links to another article". Likewise I see de facto exemptions from notability guidelines such as WP:BK along the lines of "Capsule reviews are not evidence of notability...but if they are put list of TV episodes that is alright". If anything, it is dangerous not to treat lists in the same way as articles, because it gives rise to the idea that arbitary exemptions are normal, not abberations.

Whether we choose to take a stand, ignore or skirt around this issue will have consequences, some good, some bad. --Gavin Collins (talk|contribs) 13:45, 20 September 2010 (UTC)

Gavin, why is it you think that a "madeup" navigational list of FOO that you do not believe can be verified to actually consist of members of the group FOO, could nevertheless be acceptable as Category:FOOs? I know you don't believe categories need to be notable in the same way you believe lists do, but article inclusion in categories still must satisfy WP:V, WP:OR, and WP:NPOV. And you have claimed that a list that isn't notable as a list (i.e., that list has been itself published elsewhere) is therefore unverifiable. How is it that adding Category:FOOs to an article X is more acceptable than adding a link to X to a list of FOO? postdlf (talk) 14:25, 20 September 2010 (UTC)

The simple answer is that using a link to a Wikipedia article as a basis for list inclusion is self-referencing, the worst example being "List of list" type articles. The conflict with existing policy is obvious to me: if WP:BURDEN says that "If no reliable third-party sources can be found on a topic, Wikipedia should not have an article on it", then the inclusion of lists or articles are sourced from entirely from links to other Wikipedia articles contradicts this.

By contrast, self-referencing is the raison d'etre of categories. It is their job to reference Wikipedia articles and lists, and categories can even reference each other,but...in a useful way. --Gavin Collins (talk|contribs) 15:00, 20 September 2010 (UTC)

No, that's not correct. A list is not "sourced" by the wikilink of an entry any more than a category is "sourced" by the category tag. A category tag, though it cannot be sourced directly, still needs to be supported (or supportable at the very least) by a sourced statement within the tagged article. See Wikipedia:CAT#Categorizing pages ("It should be clear from verifiable information in the article why it was placed in each of its categories."). So the only difference I can see between verifying inclusion in Category:People from the Isle of Wight or List of people from the Isle of Wight is where it is sourced: inclusion in the category can only be sourced within the included article, while the list inclusion can be sourced directly within the list. But the underlying statement of fact, that "this is a person from the Isle of Wight", is the same in either format. If the list is unverifiable, as you believe, then how is the category also not unverifiable? postdlf (talk) 15:36, 20 September 2010 (UTC)

Its not verifable because there are no reliable sources to say that the list exists in the real world. This is certainly true of lists that are just links to Wikipedia articles. If navigational lists don't exist in the real world, then why are they exempt from WP:BURDEN? Your answer would be "List of X is verfiable" because X is verifiable. But what is certainly missing from these lists is verifiable definition that is not self-referencing. --Gavin Collins (talk|contribs) 21:19, 20 September 2010 (UTC)

Gavin, that's completely nonresponsive to my comments and questions. Did you have difficulty understanding what I wrote? Please read it again and try again to actually answer my questions. postdlf (talk) 21:43, 20 September 2010 (UTC)

In fairness, I did respond but you can ignore my point if you wish. Articles and lists sourced from Wikipedia is not what this project is about. A link is not the same as a citation, and it is certainly not evidence that a list has been taken from a reliable source. --Gavin Collins (talk|contribs) 06:50, 21 September 2010 (UTC)

Gavin, please don't waste any more time; if you're not going to actually read what others have written, and actually participate in a discussion, then don't post anything. "A link is not the same as a citation" is the same as I already said, "a list is not 'sourced' by the wikilink of an entry," so it's really mindboggling that you can pretend that is somehow a rebuttal.

The only way in which your comment was a "response" was that it was posted after what I had written. All you did was restate something you had previously said with no development, and without making any comment about my post at all, which was about comparing how lists are sourced with how categories are sourced. Your "response" did not even mention categories. And you did not answer my very clear question to you: "If the list is unverifiable, as you believe, then how is the category also not unverifiable?" postdlf (talk) 07:09, 21 September 2010 (UTC)

Then, in fairness to Postdlf, perhaps I have not explained myself clearly. This is quite difficult, as WP:CATEGORY does not really explain why categories and lists are different, and perhaps this lies at the heart of why we are in disagreement. The only definition I can see as to what defines a category within the context of Wikipedia is that "a category is a page that exists in category space". The precise inclusion criteria for categories is not defined, but I think Postdlf is correct when he says that inclusion is based on evidence "sourced within the included article". In the absence of any clear guidance, this is how I understand categories and lists to be different:

1. Every article topic should be a member of one or more categories, which is an "essential" or "defining" feature of an article topic that it shares with other, similar, article topics. Categories are therefore "built up" from articles with common features, i.e. there is a sort of many-to-many relationship between categories and their members. Categories sit outside mainspace, so membership does not have to verifiable in every case (e.g. stubs with little content).
2. By contrast, list membership is built up from the top downwards, i.e. a list has a unique definition (regardles of whether defintiion is implicit or explicit), such that there is a sort of unique one-to-many relationship between a list and its members. Editors cannot decide up what goes into a list, as a list's defintion (or the idea behind a list) must come from a reliable source to demonstrate it is not original research (or that that the idea for the list has not been plagiarised).

The idea I am trying to describe is that a category "emerges" from the "sources within the included articles", but a list is defined by one source who created it or who came up with the idea behind it. A category does not need a defintion to exist, but if that defintion for a list is not sourced, then it fails WP:BURDEN.

I think, therefore, that the disagreement between us originates from the idea that lists can be compiled in the same way as categories, i.e. from the bottom upwards. --Gavin Collins (talk|contribs) 11:48, 21 September 2010 (UTC)