User:Masem/i

Anyone that has spent a small amount of time on the various Village Pumps knows that there is a long embattled issue over notability, mostly related to works of fiction but includes others fields as well. This has led to the "inclusionists" vs "deletionists" monikers that have been reported in third-party sources. Part of this is simply due to Wikipedia's growing pains: back when Wikipedia started, people wanted to created articles about anything and everything, but as we have started to mature and becoming a serious source of information, shucking off the initial flaws that some critics had about the model, we are considering how Wikipedia needs to be defines to still allow growth but to maintain its new image. However, part is this is also how many of our policies and guidelines have organically grown through the years, with changes of name, merging of policies, and the like. It is a hodgepodge, and though they are not unwieldy, they are a formidable barrier for newer editors to become accustomed to as well as to try to reaffirm what Wikipedia's purpose is.

One of the primary guidelines at issue is notability, which simply states that for a topic to be considered notable as to merit a stand-alone article, it should have "significant coverage in secondary sources". While it doesn't seem to do this at first glance, the result of this statement is actually trying to do two things: be an inclusion guideline (as to the inclusion of a topic by its sourcing), and be a content guideline (as to when such topics merit full articles). This fact is emphasized by the need to have the section Notability of article content that states that once the article itself is notable, the content itself does not need to be. For many fields, Wikipedia's definition of notability is trivially met - history, wars and military, and so forth. But other fields, fiction primarily but including living persons, musicians, athletes, geographic topics, and the like, all have to deviate from this basis of "significant coverage in secondary sources" by the need of "sub-notability guidelines" (such as WP:BK and WP:MUSIC). These sub-guidelines are not bad, per se, but as most define cases that a topic should be included as long as they meet a specific criteria. Most of these ultimately can meet the general notability guideline as sources come in, but because of the nature of what the authors of these guidelines believe Wikipedia should include instead of going after what can be sourced at the present, these guidelines have to exist. This is very very awkward, and reflects on the organic growth of the guidelines in the first place.

I propose changes that will take time to work through and much discussion, but are aimed to be as invisible as possible to the average user of Wikipedia. The changes are meant to deconstruct existing policy and guideline and rebuild them in a manner that is more reflective of what Wikipedia is today instead of the hodgepodge of policies that have been come out over time. Specifically, the idea is to remove the concept of notability, replaced by an inclusion guideline and a guideline about article worthiness. These are two very separate concepts and should be developed separately as I explain below.

Inclusion guidelines

While the mantra Wikipeida is not paper is sometimes thrown around haphazardly to explain some actions, it is clear here that we have no bounds and thus we should aim to include as many topics as possible in the work. This factor should only be limited by two guiding pillars of Wikipedia: we should not be an indiscriminate collection of information and we should only include topics that can be verified by sourcing. We are not going to include every one of the six billion people on the planet even though they can all be verified, as this is indiscriminate, but we are going to include those that have contributed in some way to society, whether in politics, entertainment, or sciences. We should be including every flora and fauna species, every major human settlement recorded by governments, every athlete that's played at the professional level of their sport, every major character of any television show, and so forth. All these can be backed by sources - maybe not third-party or secondary sources, but they can be confirmed to the level that is needed to be verifiable. There are limits, of course, and these also need to be defined.

Thus, the major change here is to specifically create inclusion guidelines. By necessity, these are going to have to be presented in a hierarchy by field, and possibly even more detailed to individual WikiProjects. Such guidelines should define whitelists - what topics should be includes regardless of any other factors - and (ideally smaller) blacklists - topics that should not be included; this may leave grey areas within the respective fields, but that's fine - we should aim to default on inclusion until shown otherwise that inclusion is indiscriminate. There will also likely be a couple overarching inclusion guidelines, the most likely one being "notability" of the topic - the significant coverage of a topic by secondary sources. Even topics on the various blacklists that meet these overarching inclusion guidelines are appropriate to be included, and should be considered the exceptional cases.

Key, however, is that topic inclusion has nothing to do with a topic meriting an article in any way. Inclusion means that it is discussed somewhere on Wikipedia. It may be a sentence, it may be five-part article. The key is to use cheap redirects and disambiguation pages to help readers to locate these included topics easily. We are not paper, we have plenty of space to provide one of the most comprehensive indexes of human information via the search bar on the left hand side, and much more through Wikipedia's own search features, but at the same time, the quality and maintenance of Wikipedia must be a concern. With a current estimate of 2.5 million pages and only about 200,000 "active" editors (at least 1 edit in the last 30 days), this presumes that an editor is actively responsible for 10 to 15 articles by themselves. While this seems to be reasonable, this is not how Wikipedia editors are distributed. Adding more pages without increasing the number of editors places more burden on maintaining the work from vandals and the like.

Information structure

Once we have determined what topics are to be included, now it is time to decide how to present them. As noted, the goal is not to make an article for each included topic but instead to determine how to organize included topics into a comprehensive encyclopedia. Thus, what is needed is a way to determine when should a topic deserve its own article, and this will be a function of several factors, not including size suggestions and limitations, reliable sourcing and the likelihood of the topic to be discussed in a manner that doesn't introduce original research or biased editing, in addition to other content policy. A topic to be included that can only be verified from one source likely cannot be expanded significantly to grow beyond a stub. Similarly, a topic that is only sourced to primary material must be carefully written to avoid bias and conjecture on the topic. Given wide inclusion guidelines, more often than not, included topics will not have sufficient content or sourcing to merit their own article. This does not mean we don't include these topics - just that we have to figure out a larger structure for including such topics. This is where non-standard articles such as lists and tables can help to cover numerous topics at one go that otherwise would not merit their own articles. This arrangement also does not prevent future expansion of a topic in a list or table if new information is available later.

Thus, the second part of improving our policies and guidelines is to work out what are our ideal qualities that articles should possess (in light that no article is ever considered perfect). This will inherently be very flexible - what an "encyclopedic" article is to one editor will be very different compared to another editor. That said, we can set some merits such as size, quality and quantity of references, and general content of such articles. We also must make absolutely clear that if an article on a topic fails these, deletion is never appropriate if the topic has already been determined worthy of inclusion, and instead seek where to retain that information. This may mean that while some editors may write dissertation-length articles on topics, a merge may reduced this to one or two sentences. How topics should be organized into articles will be partially set in overall policy and guideline, and partially on the various WikiProjects.

How do to this

There is no easy way to do approach this solution. The way that makes the most sense with the least disruption is to develop these policies/guidelines in some project space, improving them and seeking consensus, with the clear understanding that once they are accepted they will replace existing guidelines. For example, as noted, WP:Notability will no longer have the same impact and either this will be replaced with a new version, or demoted from a guideline in place of a notability guideline for the inclusion aspects. This is a Wiki-wide impact change and thus will need to have a wide consensus before implementing it.

As to the specifics, the first step would be to work at morphing the current subnotability guidelines into inclusion guidelines for specific fields, and encourage Wikiprojects to create there own inclusion guidelines. The second step would be to work on the general definition of when a topic should get an article, identifying that it may be necessary to seek more detailed advice from Wikiprojects or elsewhere for specific fields.

However, there is also a need for people to work cooperatively to meet this goal. There is a present mindset that if there's no article on a topic we don't cover this topic (to the point where some call the suggestion of merging the same as deletion.) The fact that topic may not merit articles should not be seen as a way of diminishing whatever importance a topic has to an editor, but to help instill good editing practices across all editors. We want Wikipedia to be a quality product. We are slowly slowly overcoming the media's reluctance to accept us due to issues with verifiability, we should now start considering how to improve the overall presentation of material to make it much more useful resource.

How this impacts fiction

By and large, the issues of notability and sourcing and topics having their own pages mostly affect how fiction is approached. Our current methods for handling fiction are, for the most part, poor - not necessarily due to depth or requirements of sourcing, but simply because fiction is the elephant in the room - compared to any other topic, it necessitates a different approach. However, at the same time, we don't want to create methods for fiction to be covered to ad infinitum details.

Wikipedia is not paper, but imagine that mantra if Wikipedia was truly a printed version in a book with infinite pages. A article on a TV show would likely contain: an overall description of the show (including a description of the universe should it need it), a list of the major and minor characters, a detailed list of episodes (with plot summary), development information, reception, and legacy of the show. This would all be one single article in the infinite printed work, but there's no reason that the index could allow cross referencing of individual characters and episodes within the context of that show. The only technical limitation that we can't do that on Wikipedia electronically is our size guidelines. Likely, for a long running show like Star Trek, the size of such an article would quickly exceed 100k (if not a few megabytes), a value impractical for the work. Instead, we use summary style to split the article into logical segments; this may include a separate page about the characters, a separate page about the episodes, with certain characters and episodes having even their own page due to volume of information. Our "index" would use redirects and disambiguation pages to cross-reference all of these. There would be effectively no difference between the "printed" version and this electronic version beyond the use of article boundaries to separate the parts of the overall topic.

What needs to be done, in light of rejigging all of our guidelines for this, is to make sure that, as long as we are sticking to article size restraints, that the summary style approach of writing about a topic is fully supported. It is important to realize that "articles" and "topics" are two separate concepts. "Articles" are artificial bins that we sort information on a topic in. If a topic can completely fit within one bin, then great. If not, then we let the topic fill all additional bins as needed, while adhering to encyclopedic and editorial standards. This is not to imply that for a work of fiction, absolutely every character, every episode, every object in the fictional universe gets a bin - most of these will fail to have significant information to write a good encyclopedic article. Some bins may contain several sub-topics of a topic, aka the various lists of characters and episodes that we have about. Again, these would all be part of our infinite paper encyclopedia, so should not be excluded from the electric version.