3 ms·
I see what you're saying, and broadly, I agree with you. I think we have different definitions of article creep. The things you talk about are real problems. T
by mdisraeli 16y ago
I see what you're saying, and broadly, I agree with you.
I think we have different definitions of article creep. The things you talk about are real problems. The problem for me is that when I have encountered the term "article creep", it has mostly been with respect to articles about episodes from highly-notable television series. Really, we have two different types of article creep:
Firstly, we have the pure vanity cruft, new articles created without any references which simply are impossible to verify, or have been created by sources too close to the origin. These genuinely do polute wikipedia, since they typically do not attract new editors, and rarely get linked to (both internal and external to wikipedia).
Secondly, we have genuine quality writing that is well-researched, clearly verifiable, useful to a wide audience, and can be well-linked. This second category would attract editors, maintains itself, and is generally useful. Yet, for various reasons these can get accused of ruining wikipedia.
Part of the issue is that most wikipedia editors have been untrained, and seeminly have not even had and understood something akin to GCSE history. A significant number joined to promote their point of view, rather than to develop a quality resource. The easiest solution to the problem, given this, is as you support. The best, however, is a different matter.
- tptacek 16y agoShouldn't I be able to counter this argument by asking you to cite a topic that Wikipedia covers poorly as a result of these policies?
- mdisraeli 16y agoPractically any notable serial format fictional work, for a start. The subject as a whole gets reasonable coverage, but the coverage of the material within, arguably the source of the notability, has typically been removed wholesale from wikipedia. The inclusion of this material could be said to be a matter of taste, however I have yet to see anyone counter it's inclusion with arguments which cannot be disproven. A large number of topics about anything not angloamerican. I've seen before now the members of the national sports teams for middle eastern countries put up for deletion as "not notable". Anything, absolutely anything, that is best verified via physical books and records. These inevitably end up declared as not verified, or non notable (since the internet doesn't know much about it). Given it's very late over here, I'm not going to go and look up the rest of the list. But as a counter counter-argument, can you cite a topic that wikipedia's coverage is significantly enhanced as a result of these policies?
- tptacek 16y agoYes. Companies. There are hundreds and hundreds of tiny technology companies created every year, and Wikipedia only covers the ones that have managed to get written about somewhere else. Wikipedia's business coverage would be a "who's who" mess of press releases without these policies. Another one: industry jargon. When I was on AfD patrol, I was constantly having to knock back random three letter acronyms people had invented to promote or position their companies. If you find industry jargon in WP, there is at least a decent chance that three people in the world have spontaneously used the term before. Another one: open source software. Obviously it would be bad if Wikipedia was Freshmeat, but I'll give you an even more concrete example: I helped maintain a couple pages that summarized a bunch of different software solutions (such as "Comparison of DNS Servers"). Those pages were dumping grounds for everyone's 0.01-alpha releases, each of which usually came with added columns on the page's table that was a "yes" for that NN piece of code and a "no" for everything else. Note that in the latter two cases, NN articles are creating worse problems than simply littering the WP and abusing it for company promotion; it's also tainting WP's coverage of topics, confusing lay readers. It is definitely a WP problem that "tracking down reliable sources" devolves to "search NEWS.GOOGLE.COM". On the other hand, the authors of pages can cite books in AfDs, and after doing almost 100 AfDs, I never saw such a citation get knocked down.