User:Jorge Stolfi/DoW/Notability

UNSORTED DRAFT - NOT READABLE YET

The Wikipedia notability guideline (WP:N) is a page in the "Wikipedia:" namespace, supplemented by about a dozen sub-pages, that defines when a topic can be considered "notable". It is meant to flesh out the notability rule the principle that only "notable" topics can have have Wikipedia articles. By and large, the guideline defines a topic as "notable" if it has got plenty of references in "independent" sources.

Belief in the notability rule and its guidelines are the defining characteristic of the so-called deletionists. The rule as been used as the sole excuse to delete thousands of articles on popular subjects were deleted because they allegedly failed to meet that. These range from articles on pop-culture topics like characters from computer games, to local-interest topics like small enterprises, small-town politicians, and elementary schools, but also minor novels, songs and movies, as well as many biographies of scientists, politicians and athletes who were not considered to be "sufficiently famous."

The notabilty guideline page claims to be a consensus, and the notability rule is stated by deletionists as if it were a core Wikipedia policy. In fact neither claim is true. Like many other "policies", they are not "official" in any sense, and are anything but consensual.

Cost of the rule
Needless to say, the author of a deleted article is hardly ever pleased to see his work — which often cost him hours to write — to be bluntly labeled as "worthless" and summarily deleted, in such a way that he cannot even see the comments in the talk page.

"Notability rules are needed to justify deletion of non-notable articles"
The reasoning used to defent the notability guideline, in particular, is pretty circular: "We need to have a notability requirement, because otherwise Wikipedia would get thousands of articles on non-notable subjects, and that is bad because WIkipedia articles must be on notable topics.

The requirement for "notability" is not an axiom. The question is not whether mailmen and high school teachers are notable or not, or whether evidence of their notabilty is likely to come up soon. The question is whether we should worry about their notability at all — and, most importantly, why.

"Notability is required by WP basic principles"
meeting the notability guideline does not mean meeting a V, NOR, NPOV, and vice-versa.

"Notability is the same as verifiability"
Notability is often confused with verifiability. "The notability guideline is not something separate from the verifiability policy, rather, it is a guideline for implementing it." and 2. How do we write about it properly in the absence of significant coverage?

But there is big a logical gap between "verifiability" and "notability". The "verfiability" requirement follows directly from the fundamental goal of Wikipedia (to be a repository of useful, and hence true, factual information); is a practical necessity; and is a real consensus among all regular editors, whether or not they read Wikipedia project pages. The same can be said of several other requirements on article contents, such as "no libel", "no copyrighted material", "not a resume/directory/textbook/manual/blog/database/"etc.

The "notability" requirement, on the other hand, is none of those three things.

I really don see the connection. Indeed, most of the disputed examples mentioned in earlier threads are articles that are judged to be "non-notable" according to the present guidelines, but which contain information easily verifiable by any editor (e.g. "Steven O'Clock has been professor of drama at Trifling College since 1971"). Most editors apparently agree that "verifiability" is necessary (and I do), but obviously there are many who do not think that "notability" is necessary (and I don't). And there seem to be many more who agree that notability is needed but do not like the current guidelines and/or the current AfD procedures.

As for logical necessity, there is nothing in Wikipedia's goal that rules out having an article on every human being who has ever lived, every street and condo in the world, every species living os extinct, every song that has ever been broadcast or sold, and every book that has been printed — as long as the article's contents meets the other well-established and consensual guidelines.

"A good encyclopedia does not cover non-notable topics"
The notability rule and its guidelines are often justified by claiming that a "traditional encyclopedia would not have an article" on those subjects. Indeed, but that is because paper is very expensive, and its authors must be paid for each word that they write. But "Wikipedia is not a paper encyclopedia" and our writers work for free. These are the key advantages of Wikipedia over traditional electronic encyclopedias, which allowed it to become much larger and much more interesting than any of them. Indeed, I doubt that one would find articles on Dichlorotetrakis(dimethyl sulfoxide)ruthenium(II), Zuncich Hill, or Mexico (Bob Moore song) in a traditional encyclopedia. Should we try to fix those "flaws", or be proud of them?

"Notability is necessary to preserve Wikipedia's prestige"
Notability has been proposed as a way to defend Wikipedia's reputation as an academic resource (more to the point, we are not considered a serious encyclopedia by most academic institutions) and opening the doors to all sorts of sundry and pointless topics is probably the best way to divest us of any shred of credibility we still have.


 * Wikipedia will never be a credible source, will never meet any academic standards, and will never get anywhere near the editorial, stylistic and typographical quality of a traditional encyclopedia. It lacks the means to get there, and therefore it should not aim for that goal.  Any editor who is working under that assumtion is certain to be bitterly disappointed.

As for Wikipedia's credibility before the general public, what damages it is incorrect information — especially on important topics, or in articles that, by their polished writing and typesetting, may induce readers to believe that they are reading reliable, reviewed sources. My only idea on how to alleviate this problem is to put up a disclaimer in small but bold red letters, e.g. " All information in this site was contributed by volunteer unsupervised editors and is not to be trusted without independent checking. ". It may not be good marketing, but it is just basic honesty and humility.

Having an article on a non-notable garage band does no harm to our credibility. Having an article on a non-notable professor, lawyer, or small-town alderman (provided that its contents complies with all other WP requirements) is not only harmless in general, but may be actually useful, sometimes for hundreds or thousands of readers. A million such articles will probably be useful to ten million readers.

"Deleting non-notable articles improves Wikipedia"
As for practicality, first, the supposed advantages of removing articles on "non-notable" subjects are more than offset by the cost (in editor's time and morale) of the ensuing disputes, which are entirely based on a bunch of *arbitrary* and *subjective* notability guidelines.

Until 2005 or so the number of articles in WP was doubling every year; if growth at that rate could be resumed, it would reach 3 billion articles in 10 years, and 30 billion in another 3 years — which would be more than enough for the universal coverage of the aforementioned areas.

There are more than 1,600,000 articles still marked as "stub", and probably another 1,000,000 non-stubs that need urgent attention. Even if we settled for the modest goal of bringing all those articles to "Start" quality (forget "Good Article"!) within five years, while weeding out the usual amount of vandalism, we need to recruit something like 10,000 new commited editors. We need much more than that if we also want to fill the most obvious gaps in coverage. Unfortunately the number of such editors has been shrinking, not growing, since 2006.

A first-time editor is unlikely to be "hooked" until he has created an article on a subject that is dear to him. Coincidentally, since 2006, anonymous users have been blocked from creating new articles. In retrospect, that may go a long way towards explaining the "2006 catastrophe" that we see in the statistics: why would a reader bother to register, if not to watch and/or get credit for edits to "his" article — which he created as an anonymous user? Why would he bother to register before creating his first article, if he has no idea of what that experience is like? Even if the new editor gets past that hurdle, his first new article will probably fail to meet the "notabilty" criteria, and will be promptly and rudely challenged for deletion. That will remove most of his motivation to contribute to WP. On the other hand, if we could get that editor hooked, he would probably contribute another 100 good articles and tens of thousands of cleanup edits over the next few years (and perhaps even donate some money!). The loss of all those contributions must be counted as a very real but invisible cost of deleting a single "non-notable" article, no matter how obvious it "non-notability" is. In retrospect, the measures implemented since 2006 to curb non-notable articles may have been like a farmer killing all the bees because they bothered the cherry pickers.

In summary, Wikipedia desperately needs more editors to clean it up, and a lot more to make it grow again; but the only way to get those people is by becoming MUCH more tolerant about "non-notable" or "too specialized" new articles.

As a minimum, there should be a waiting period of at least one year between the creation of a new article and any proposal to delete it that is based solely on the "non-notabilty" of its subject. That is probably enough for the article's creator to become a regular; by that time, having an article deleted will hopefully be no more disagreable than having a bad tooth removed.

Deltionists say that it is insane to put up with worthless articles on blatantly non-notable garage bands, corner stores, PTA presidents and high school fads, even for a limited period.

Wikipedia's fame and value are not due to it being any of those things. What brings readers here is that Wikipedia (1) is extremely accessible, (2) is very likely to give some useful and usually mostly correct information on the most unlikely topics one can think of, and (3) is chock-full of useful links, internal and external. Thus, to increase its value to the public, we must strive to expand its coverage and close all its red-links and "black-links" (names and terms that should be wikilinked, but aren't). And this, in particular, means creating an article — even if a single-line long — for every person whose name is mentioned in Wikipedia, notable or not

For example, some deletionists would consider inappropriateto have an article on a guy who has ben running for a town supervisor in every election for sixty years straight can be considered a public figure, and has been covered in local papers or maybe even a locally-produced book about the town.

Notability would have to be derived from something like recognition within the profession; writing for a national mail carrier publication, being covered by one as a subject (directly, probably not in passing), or receiving some kind of recognition by award, etc. The greyer area is when there's somebody who does public work that is not local and/or they have received some degree of nonlocal coverage, and perhaps some degree of recognition within their field, and is the type of person a traditional encyclopedia would have an article about, and might have included if they had unlimited space.

A traditional encyclopedia would not have an article on a literal garage band, corner store, PTX president, or high school fad. It may on a band with recordings, national radio play and reviews, or a national fad, etc.

Even if one were to accept the notability rule per se, there is the much fuzzier question of defining what exactly is "notable enough". Some deletionists would not consider a local band "notable" even if it has published almums, got coverage in papers, played in radios, or even become a national fad for a short while.

Notability and AfD are not going to give Wikipedia a better academic reputation, I think overall only scholarly contributors or scholarly editorial oversight could do that. Serious students might use WP to get a general idea regarding something and then go to the cited sources and read and cite them directly but not cite Wikipedia, or at most note that they learned of a source through WP. Professors do not have so much of a problem with that and may even encourage it on occasion, if students have a better ability to assess the quality of sources than many WP editors do. On a more limited level, individual articles might be capable of being cited, but the author citing it is going to have to (1) cite a particular revision of a page and (2) explain in their paper why they believe citing Wikipedia had merit in this case. Articles on topics that are not just the type of thing in a traditional encyclopedia but which actually have entries in them, but which are mediocre, poor, or bad be it due to poor writing, poor understanding, poor sources or vandalism are more of a problem to academic credibility than having articles on things of lesser notability.

Notability as encapsulated by the general notability guide is a good rule of thumb for most subjects, as if there aren't multiple reliable sources writing indepth about a topic then 1. Why should an encyclopedia write about it if nobody else has?

As for why Wikipedia's editors and articles are not expanding as fast as they used to, look at Google Trends: The searches for Wikipedia plateaued at the end of 2006 and have been stable for three years. If our readership isn't growing, and our readership is where we get our editors from, why would you expect the number of editors to be rising?


 * It is hard to believe that the Google plot reflects a sudden leveling off of Wikipedia readership. I have been told that the "2006 catastrophe" was the result of a permanent chang in Wikipedia's public image in the aftermath of the Steigenberg incident. That does not seem to fit.  One would expect a sudden jump (up or down) in activity and then a slow recovery, complete or partial, as people forgot the incident. But we don't see that; edits actually kept increasing for a while after the incident, and the new article rate did not show any immediate effect.  But within a year all trends changed permanently to a new steady regime.

My guess is that what we see in that "access" plot is actualy edits by regular editors. An average pure-reader reads one Wikipedia article a week, perhaps less. An average regular editor, on the other hand, probably makes 100 page fetches a day, probably more. So, even if there are 100 times more readers than editors, still 90% of the traffic will be due to editors. If these are slowly shrinking (as all other plots seem to indicate), the steady number of accesses may mean that the number of pure-readers is still expanding. Perhaps this information can be obtained from the HTTP server's logs? Is is feasible to identify which accesses came from editors?

If this guideline were to go away, something very similar would take its place. "Notability" as a standard is the natural consequence of our basic content principles. The neutral point of view is a fundamental, and non-negotiable, principle of the project. It requires that we present a topic as it is evidenced in the body of reliable sources. If there is no coverage in reliable sources, we cannot possibly meet that principle. Verifiability states that if there is no outside coverage of a topic, we do not cover it in Wikipedia. We are not permitted to engage in original research to complete an article about a topic, thus the need sources that report on and interpret a topic for us is reinforced. If all we can draw from independent reliable sources is a mere assertion of existence or a similarly short amount of information, it not appropriate to have the article on Wikipedia as we are explicitly not a directory.

If we can only create a complete article by heavy reliance on primary, dependent, and/or self-published sources, it is not appropriate to have the article on Wikipedia as it is explicitly not a vehicle for promotion. (In respective order, the relevant policies are Neutral point of view, Verifiability, No original research, and What Wikipedia is not.) In the absence of a formal notability policy, we would still effectively have the requirement. The most basic content principles of Wikipedia demand that sufficient sources be available to meet their requirements. If you boil it down, notability is really nothing more than an expression of that demand.


 * It seems that there is a persistent misundertanding here. What you describe is "verifiability", not "notability".
 * "Jimmy Hoffa's dying words" are notable but not verifiable. An article on this topic would be necessarily empty, to comply with the verifiability requirement; and therefore would be speedily and consensually deleted.
 * On the other hand, the existence, line of business, and street address of the "JJ & F Market, Palo Alto, California" are easily verifiable, but hardly notable (at least by the current guidelines). I still cannot see what harm there might come to Wikipedia from allowing the creation of a one-line article with this topic and title, with just that information.
 * Said another way:
 * Verifiability is about the article's contents. Any unverifiable parts of an article may be challenged and/or moved to the talk page and/or deleted and/or discussed on the talk page, etc., as per established guidelines customs, and according to the views of individual editors.
 * Notability is about the article's topic. Articles on non-notable topics should be tolerated because (as argued above) they are likely to bring more good than harm, and because of the incurable subjectivity and arbitrariness of the "notable" concept. The contents of any such article should be evaluated and handled with the same procedures and criteria that apply to other articles (including verifiability, non-libel, non-private, non-promo, neutrality, etc.). That article may be deleted, or turned into a redirect, only if it becomes (definitely and incurably) devoid of acceptable contents.
 * Some of the comments in favor of the notability requirement seem to say: "We need a notability requirement, because otherwise wikipedia will be flooded with thousands of articles about non-notable topics, and that is very bad because it violates the notability requirement." Or, "we need a definition of notability and a notability requirement, because otherwise we could not justify the deletion of articles whose only fault is the non-notability of their topic." Obviously one cannot make a convincing case for "notability is necessary" if one starts from the premise that notability is necessary.


 * Please review my statement again. My comments clearly encompass far more than simple verifiability, such as the need for appropriate sources to avoid entries that fit into what Wikipedia is not and for sufficient sources to be able to ascertain the neutral point of view. "Notability" is a term that unfortunately leads to misunderstandings, much like "neutral point of view". The latter is often misinterpreted as requiring some engineered neutrality or fair balance, but instead refers to the position of editors who should neutrally report what reliable sources state about a topic. Similarly, notability is not about some subjective measure of importance or noteworthiness, but rather that reliable sources have made substantive note of a topic. Notability is not the tautology that you present, but rather the natural consequence of our basic content policies as I noted above. Without substantive coverage in multiple independent sources, it is impossible to meet the requirements of those fundamental content principles. You can call that standard whatever you like (I'd prefer something like "Wikipedia:Sufficient sources" as it would avoid many misunderstandings), but it is the underlying principle of Wikipedia's notability guideline.


 * I agree with Jorge Stolfi's comments. It's never been proven to me how it harms WP to have a well-written and referenced (if possible) stub on a non-notable (whatever that might mean) topic. In addition, too much angst is caused by not considering "notable" on a sliding scale (all too often "notable" is treated as black and white). Apologies in advance for the inevitable international outrage, but I simply can't bring myself to find the parish of Saint George Gingerland in Saint Kitts and Nevis notable. I'm sorry, but there—I've said it. Oh, I know that they have "goats, which wander and graze freely", that "it faces into the trade winds", and that "some of the other plantation buildings, including the old windmills, have been turned into guest suites", but I really struggle to find it notable. Should we have an article on the parish of Saint George Gingerland? Of course we should. I'm sure that the "total population of circa 2,500" of the parish (not to mention the "pigs, cows and horses [which] are fenced or tethered") find it very notable. WP is able to document a great adventure, and is a wonderful present for all mankind (now and in the future)—let's miss as little as possible of it.


 *  (Straying into the wilderness of verifiability) I'm now concerned that there are no references (in the article in my example above) proving that those goats graze freely, nor that those pigs are tethered. Should I delete that information? Of course if we do decide to delete that, then I would estimate that over 75% of WP has to be deleted with it (in order to apply consistency, and so as not to have the 2,500 people of Gingerland take it personally). To complete the process, after deleting everything that isn't verifiable or notable, we can print WP and sell it with a title "Almost the Encyclopædia Britannica".


 * I also agree with Jorgs and you make a great point about the sliding scale. To a small extent our policies already say different topics require varying levels of sourcing, e.g. WP:V has a section titled Exceptional claims require exceptional sources. For what we could loosely call high impact topics,  I agree with Vassyana that notability follows naturally from our basic content policies - e.g.  with an article about recent original research that makes scientific claims, you need coverage in reliable independent sources otherwise we cant judge whether the article is about some random pet theory or even a hoax. Likewise with a sensitive political topic you need multiple quality secondary sources or its very hard to achieve NPOV. On the other hand, there seems to be much less reason to demand quality independent coverage for articles about say characters from popular novels or computer games. Providing we AGF that folk care about accuracy and fairness and that these articles are less likely to attract POV warriors, we can expect NPOV will generally be achieved just from the primary sources and perhaps fan sites, which should also be sufficient to Verify any dubious seeming claims. Granted its much easier to stray into non compliance with policies when you lack good secondary sources, but that really doesnt seem to offset the benefits you and Jorg talk about that follow from allowing articles that millions of people care about. PS - to proof the goats graze freely you can use the village grounds itself as the primary source - we do have an essay   linked to from Reliable_sources   that says we can verify from plain site as long as viewing is accessible to the public.


 * Vassayana's comments make sense; except that I still think that the concept he calls "notability" is what is generally understood by "verifiability".


 * As for the "notability rule" being a logical necessity: a basic one-line article on the "JJ & F Market, Palo Alto" would satisfy every fundamental Wikipedia requirement about content (verifiability, neutral point of view, encyclopedic tone, appropriate level of detail, etc. etc.), just as well as a (presumably much longer) article on "le Moulin Rouge, Paris".  The only difference between the two is the degree of "notability" of the respective topics. Hopefully this example shows that the notability requirement does not follow from the other requirements; to make it hold, it must be asserted separately. So the question still is, why do we need that rule?  Until someone can show that deleting such a "JJ" article would make  Wikipedia concretely better,  I will assume that it is basically a matter of taste: namely, some editors want to get rid of those articles, just because they do not fit their notion of what is a "respectable" or "nice" encyclopedia; while other editors have different tastes.  If that is so, then no wonder this talk page has got that long...


 * Im reading Vassyana's point as being the core policies depend on notability – perhaps in a similar way to the security of a lock depending on secondary qualities like strength and complexity of the locking mechanism. My intpretation is thats correct, but only for an important but small sub set of the articles folk like to create. Otherwise I agree with pretty all you say.


 * I don't think that Vassyana is saying that the core Wikipedia policies depend indirectly on some underlying notion of notability. Rather, I think the argument is that Notability is implied by the core policies. And I agree that Jorge has shown that there is no such implication. I think that the vast majority of non-notable articles also have some other problem such as lack of verifiability. The question is, assuming that an article includes 100% verifiable information, how do we decide whether or not the article belongs in Wikipedia?
 * Personally, I probably take a middle-of-the-road view: on the one hand, I don't think Wikipedia needs an article on every retail store verifiably listed in a phone book. On the other hand, I'm a bit uncomfortable because I've seen this guideline misused by people who don't understand it, and who can blame them? The guideline is difficult to understand as a result of years of creep that have given us 10+ sub-guidelines, when it ought to be a very simple and flexible principle.


 * I agree with this statement, in that, while meeting WP:N usually means you're meeting V, NOR, NPOV, etc. at the same time, WP:N should not be considered as a guideline that is meant to support those policies. WP:N should be considered as a means to judge non-indiscriminate topics. The GNG provides one clear guidance - when other sources talk about the topic in depth - but it is by no means the only way we decide what is and isn't indiscriminate. I think WP:N isn't wrong, but it is misused, because too many people see it as the only means of measuring "indiscriminate", where there certainly are several, unstated others (based on AFD results) Now, I will still point out that we don't let the core policies like V, NOR, NPOV be violated, but it is still careful to note that meeting the "not indiscriminate" aspect of WP's missions and meeting these other policies are two different tasks that happen to overlap nicely at the GNG


 * If you grab a handful of sand and keep squeezing, eventually the sand will trickle out (and the remainder will no longer look like sand). The more people try to define "notability", the more sub-guidelines will be necessary. Of course it gives all the wiki-lawyers something to do.