Wikipedia talk:Wikipedia is being flooded

From Wikipedia, the free encyclopedia

Contents

[edit] Eventually...

There are only so many topics that have been covered in reliable sources. Eventually the "flood" should let up, with most/many of the new articles being about new topics, as opposed to the preexisting ones that we don't yet cover. Then perhaps the gnomes and copyeditors will be able to improve our content faster than it is being written. I don't think the rapid increase of articles is our biggest worry - the dismal state of the Vital Articles is a much bigger problem, in my opinion. Λυδαcιτγ 20:36, 1 July 2007 (UTC)

[edit] Legitimate problem

This is certainly a legitmate problem. Wikipedia is useful to many readers because its signal-to-noise ratio is very good relative to the internet at large. Having limitless numbers of (low quality) articles diminishes this attractive feature.

I think that excessively rapid growth in article numbers could be mitigated by placing some very mild limitations on article creation. Possibly:

  • limit article creation to logged in users
  • require a one sentence justification to accompany the creation of any new article.

ike9898 21:48, 1 July 2007 (UTC)

We already limit article creation to logged in users; hence Wikipedia:Articles for creation. Λυδαcιτγ 03:45, 2 July 2007 (UTC)
We already have a one sentence justification to acompany new articles. If you don't assert notability, it gets AfD's. or speedied. JetheroTalk 04:57, 9 July 2007 (UTC)

[edit] But are they read?

Counting the number of articles of each type is a flawed metric because it means that each article is considered equally important. The new articles will tend to have low page rank and rarely come up in searches, so they won't be read much. A better metric would be the number of page views that featured and good-quality articles get versus other kinds. Is this increasing or decreasing?

Skybrian 00:26, 2 July 2007 (UTC)

I'm not sure how relevant that is. Encyclopaedias aren't like other books, they're reference material. Most paper encyclopaedias will sit on a shelf and only be referred to every once in a while, with some of the more obscure or specialist topics not being looked at much at all. Does that mean those topics shouldn't be included? I don't think so. Waggers 10:49, 2 July 2007 (UTC)

Counting page views rather than articles gives us the probability that a person consulting Wikipedia will see a high-quality article. Isn't that what it's all about?

Skybrian 23:39, 8 July 2007 (UTC)

[edit] Various thoughts

As a highly active editor for the last six months, I have had time to consider the important issues raised by Giant Puffin in this essay.

I object now, as I did when I read Wikipedia:Wikipedia is failing, against measuring the success of Wikipedia by the number of good articles and featured articles. First of all, I have written one good article - endgame tablebase, which was unreferenced and disorganized when I started to rewrite it completely over several stages lasting almost a year long. It took me somewhere between fifteen and twenty hours to do it, and it was a labor of love, an opportunity for me to learn in depth about one of my favorite subjects, and to share that knowledge with whoever wishes to learn. However, I cannot focus my efforts consistently on most articles long-term. I have neither the time, nor the knowledge, nor the patience. I am also not particularly well-suited to the WikiProject style of collaborative editing.

The WP:1FA approach suggests that only high-level articles "count", and all other articles exist in order to provide a tiered framework for those core topics to sit at the summit of human knowledge. I don't think Wikipedia works that way. As a democratic project, Wikipedia invites twelve-year-olds to add a sentence of trivia about the latest Family Guy episode, and they do (and sometimes we have to revert them). The strength of Wikipedia is not in its depth, but in its breadth. There will never be a larger encyclopedia. It contains information about every active player in the National Football League, every small town in Wisconsin, and every character in the Pokemon series. Together, these three categories account for about five thousand articles, and only a handful of those (such as Tom Brady) are GAs and FAs. But really it cannot be any different. Who can write a Featured Article on Josh Miller or Jeff Feagles? As punters in the NFL, they are entitled to articles, but they will never be notable enough to rise to Featured Article caliber. The notion that any article can become featured is fiction. Some stubs will always remain stubs, and should always remain stubs.

The majority of my recent focus in article space has been on filtering the small, new articles to weed out what doesn't belong and send it to the appropriate deletion forum. Category:Uncategorized pages, which is populated by tagging from User:Alaibot, is perpetually backlogged with thousands of articles. Some of them are obvious speedy deletion candidates if anyone ever notices them. Others are excellent start-class articles that happen to need a category. I've seen both ends of the spectrum and everything in between. The point is that categorization - essentially a WikiGnome activity - will never lead any article on the path toward FA. It's just too far away, and if someone ever decides to adopt the article, they will categorize it themselves. I'm just trying to salvage Wikipedia's worst articles to make them passably readable and useful.

That's my point. Value in Wikipedia can be measured in different ways. If I keep a bad article out of Wikipedia by alert patrolling of newpages and backlogs, that's half as good as creating a decent new article. Both actions enhance the integrity of the project, even though neither action will have the slightest bearing on the isolated corner of the project known to us as Featured Article Candidacies. I agree that these backlogs keep getting longer and longer, and it is impossible to turn back the clock, but there does not need to be a specific response to it, other than to spend less time commenting on Wikipedia essays and more time fixing up articles. (Haha.) With that, I wish everyone good luck in the never-ending endeavor. Shalom Hello 01:04, 2 July 2007 (UTC)

[edit] Chiefs and Indians

I don't agree with the underlying logic of this essay (too many stubs being created relative to GA/FA promotions) because that assumes that stub/start/B are bad and that FA/GA is the end goal. A better barometer would be how many articles are tagged needing attention (and don't get me starting on how flawed the tagging system is...), how many bad articles are there relative to total articles? My opinion? Too many chiefs, not enough indians. The best editors turn into admins who get bogged down chasing down vandalism, involving themselves in Incidents. Too many editors investing themselves in fighting over policy and procedures, burning out and leaving. Too many intelligent people fighting over their preferred format or version. --maclean 01:38, 2 July 2007 (UTC)

Many of the stub and start articles are bad. If they're not already tagged with various templates, they should be. Im in full support of people making new articles, but only if they are of a useful quality. Creating several stub articles is completely useless, especially if they become forgotten and untouched - • The Giant Puffin • 08:47, 2 July 2007 (UTC)
I certainly agree with the point about getting bogged down in policy etc. We say that Wikipedia is not a bureaucracy, Wikipedia is not a discussion forum and Wikipedia is not a battleground yet the reality is somewhat different. Waggers 10:41, 2 July 2007 (UTC)
Very true. I was a fairly prolific article-writer before I became an admin in April; since then I've only done substantial work on one new article, History of Nebraska, which I started yesterday. Nor do I do as much maintenance or anti-vandalism as I used to. I find myself getting bogged down in the wikipolitics, as, sadly, do most admins. Waltontalk 15:45, 2 July 2007 (UTC)
"The true solution is to gain more members who want to truly improve Wikipedia, not vandalise it or join and forget about it". I agree with this opinion. And what puts me off most from getting involved more isn't the vandalism, it's the experienced "know-it-alls" who know nothing of the topic but use their superior editing skills to enforce their opinions and revert perfectly sensible and well-cosidered edits on technical grounds rather than offering a helping hand. There are quite a few like that...why can't they beat up on the real villains rather than the honest learners?

[edit] various thoughts (me too)

Interesting essay yet unlikely to lead to much change. Shalom makes a number of important points and the situation is indeed not as bleak as the essay suggests. I'd like to add a few thoughts of my own.

  • Wikipedia's strength is indeed breadth and even more importantly a convenient interface to navigate the whole mess. So having lots and lots of articles, even short, stubby, will-never-be-anything-close-to-GA articles is a good thing. But in my mind the crucial misconception of the essay is that this breadth hinders the progression of quality articles. What would happen if, say, we were to declare a moratorium on new pages or started to agressively delete all of the hopeless stubs? Truth is: not much. Creating a stub takes 30 seconds so even assuming that all that time would go into editing more important articles, the gain would be marginal. There would be less time spent deleting a new article that says "John Doe is gay" and more time spent deleting the sentence "John Doe is gay" from the article particle physics.
  • I regularly patrol new pages and CAT:CSD to weed out the junk and I'd be very interested in a statistical study of the rate of creation vs the rate of deletion. My rough estimate is that at least a third if not a half of newly created pages get deleted within, say, a month (excluding things like redirects or pages created as subarticles of pages that have become too long) and I suspect that this proportion is increasing. This is not so surprising since we are indeed coming closer to a shortage of valuable new topics.
  • In many ways, Wikipedia is best viewed as a heap of projects which just happen to share a common url. There's the Wikipedia on physics, the Wikipedia on geography, the Wikipedia on Pokemon, the Wikipedia on jazz and they are maintained independently by groups of people with little if any overlap. Each of these communities polices itself, organizes itself, sets priorities for itself and all are making progress. What does this have to do with anything? Well I think the people building the featured articles should be viewed in many ways as one of these communities. Yes, I also wish we had more gnomes, more competent copyeditors like Tony, more people with specific expertise and so on. But the fact is that we're not going to take a 12-year old and force him to become the main force behind the rise of History of Scotland to FA status. What we can maybe dream about is that some of the 12-year olds that create pages for every new Beyoncé single will mature into excellent editors. Wikipedia has to face the fact that its pool of quality editors is limited but that it can still benefit tremendously from a huge pool of people who are pushing the project forward through smaller contributions.
  • Let's pause for a moment and imagine what we would like Wikipedia to ultimately be. We'd like it to have all the breadth it currently has and all the depth that traditional encyclopedias have. We'd like all articles to have FA-status or at least GA-quality for the less important topics. We'd have featured articles on LeBron James, Lebesgue integration, go (board game), frogs, Novosibirsk but also featured articles on Bimbo Coles, conull sets, parcheesi, Hylobius abietis, Tatarsk. Currently, most core topics have quite good articles, important topics have decent-on-average articles, not-so-important topics have low-quality start articles and most fringe topics have stub articles. Everyday , more articles move up this ladder: more GAs take the complex next step towards FA, more articles of secondary importance come closer to GA and the pool of fringe articles gets new stubs. That's how we should expect it to be. Pascal.Tesson 02:04, 2 July 2007 (UTC)
Well said, especially your first point. Λυδαcιτγ 03:54, 2 July 2007 (UTC)
Yes, I also agree with most of these comments. Although I concur with the basic premise of this essay - that we need to stop expanding Wikipedia and focus our efforts on cleaning up what's already here - I don't think there's any concrete way to achieve that. As Pascal said, a moratorium on new articles would not affect our other problems, such as the massive rate of vandalism; such a moratorium would do more harm than good. And I have also noted the fact that our pool of expert editors is limited. Having substantially rewritten Politics and taken it to GA status, I know that it will not make FA in the short term - because I was out of my depth, not having true academic expertise in the subject, and there simply wasn't anyone else bothering to work on it. However, in the long term, we can hope that with Wikipedia's growth, core topics will not be neglected. Waltontalk 15:44, 2 July 2007 (UTC)

[edit] Stiffling Information More Harmful

If even .01% of the Wikipedia articles are GA or better that is justification enough for Wikipedia. It is an open source form of information. Contributors and editors will often have divergent vierws over what is relevant. For example, an article on "Microsoft" might be newsworth to an American, while to a person in Nepal such an article might be worthless and an article on the agricultural value of "ox dung" might be all important. I think we are foolish in the Wikipedia community if we think we can create somethign that is all things to all persons. I think we should let those post who wish to and only place mininal limits on them...such as are required by the rights of others (privacy, libel, slander, maliciously false information). We should believe that people accessing Wikipedia will use it as a first stop to find something on everything, but not that they will use it as the ultimate credible source. We want tem to look at other encyclopedias, others articles in newspapers and journals (which are also often false), and whatever other sources of information they can find. When they do this then our goal of open source information is achieved. Finall, therefore, I would like to suggest that perhaps we consider not restricting information on Wikipedia as this article suggest, but leaving the flood gates wide open as the more information there is out there the better we serve the international community. --Robert 07:19, 2 July 2007 (UTC)

[edit] Rate of GA/FA articles being listed vs Rate of articles created

There is no way that the rate of GA/FA being listed will be faster than articles being created simply because GA/FA needs to go through a process before it can be listed while articles can be created at the click of a button. This situation is almost unavoidable especially when Wikipedia is getting popular. However, with reference to the chart on the page, you can see that it will soon reach a plateau and not growing at an exponential rate. Another thing I want to point out is that some articles have reached GA status already, but haven't been nominated by someone yet, so the actual # of GA is higher than the statistics that show us.

I do agree that the rate of GA/FA promotion is far slower than others. If we want to put the promotion rate back to scale, we need to simplify the processes. OhanaUnitedTalk page 09:27, 2 July 2007 (UTC)

There is also a backlog at the GA review. Go figure. JoeSmack Talk 15:56, 2 July 2007 (UTC)

Relative growth rates for Wikipedia and for GA/FA has actually been improving steadily (probably for the main reason that new article creation has slowed considerably over the last 6-9 months). Certainly the promotion processes are not completely optimal, but there are not any easy fixes except for increased participation. Without that, streamlining will simply mean less attention to problems in articles. However, the quality discussion is going to change dramatically over the next year or so, when stable versions permit us to more accurately measure the worthwhile articles that are not GA/FA.--ragesoss 13:44, 3 July 2007 (UTC)

The main problem is that there're not enough reviewers for GA/FA/peer review (and similar) OhanaUnitedTalk page 14:46, 8 July 2007 (UTC)
Also, not enough good reviewers. Awadewit | talk 14:50, 8 July 2007 (UTC)
Is there anything that can change this? It's possible to ask people to change their welcome note to new editors and mention about what they can do (such as peer review, XFD, RFA, portals, WikiProjects) in addition to telling them the 5 pillars of Wikipedia. OhanaUnitedTalk page 06:25, 9 July 2007 (UTC)

[edit] To obsess about the author's use of the FA/GA metric is to miss the point

The author has a point

"Good article" growth rates do not perfectly reflect how many articles on wikipedia are useful. We know that. The author knows that. He's just using G.A. rates to illustrate in simple terms the growing disparity between quantity of articles on wikipedia, and quality of articles on wikipedia. Those who obsess about his use of "Good Articles" as a measurement are missing the point - they are focusing too much on the medium he is using, rather than the message. The message is this: wikipedia is drowning in articles. Yes, many of them may be interesting, accurate, or potentially useful. But the problem is that this are not sustainable.


The problem with too many articles

There are many things to describe on the world. Too many things to continuously describe accurately on wikipedia. This is the same reason we have notability guidelines - many subjects are so small in irrelevant that no more than a few people on earth care about them.

The problem is that its difficult to determine what subjects perceived as relevant today, such as digimon, will have their influence persist into the future. If you think of every film or movie ever made, or will be made, do you think a movie like The_Girl_from_Monday will be considered important 50 years from now? The upkeep required to constantly chronicle pop culture and create an impromptu database of as many historic, geographic, and scientific nuances as possible would be much better spent solidifying and improving the information we have about the slightly more-likely-to-be-relevant topics.

But what about breadth?

I agree with the statement that the strength of wikipedia lies in its breadth - that is, I am glad that wikipedia covers so many topics. But similarly, even though I agree the sustenance of humanity lies in access to water, I would not like humanity buried in water.

We need to make sure we do not bite off more than we can chew and up with an entirely disreputable mass of disorganized information. The more well-written the average article looks, the better. If we can channel energy used in making new articles into improving the articles that exist, I believe we should.

-Monk of the highest order(t) 21:34, 4 July 2007 (UTC)

If we can channel energy used in making new articles into improving the articles that exist, I believe we should. Well, yes of course we should but as I said earlier this is most likely not possible. The energy that is put in creating a digimon stub cannot be magically turned into adding precise sources for photoelectric effect. The reason why FA and GA are being created at snail pace because it's just a long and tedious process to build top-notch articles not because too many people are busy with digimon. In fact, experience shows that having too many people working at the same time on FA articles is almost counter-productive and the final polishing of these articles is often due to a handful of experts. Pascal.Tesson 23:32, 4 July 2007 (UTC)
Maybe a better standard would be "B" articles or higher, which is actually the level at which a lot of articles in Encyclopedia Britannica and other encyclopedias would be assessed. There are 42,871 articles at that level or higher, or about 4% of our assessed articles and about 2% of our total articles. John Carter 17:30, 5 July 2007 (UTC)
I agree. I think it would also help if the standards for 'B' articles were clearer, both for writers and assessors. I've noticed a substantial variation across projects in the all-important difference between 'Start' and 'B'; some projects seem wary of giving anything above 'Start' to any but the most complete articles.
Perhaps it's also time for more granularity at the lower end of the assessment spectrum? If 'Stub' were used only for the most stubby of articles then a drive to improve all stubs wouldn't seem quite so out of reach. Espresso Addict 18:10, 5 July 2007 (UTC)
It would definitely help if we had something between Start and B. I only rate articles B if they are basically complete, and basically sourced. There are lots of articles that are more-or-less complete in coverage, but with only partial sourcing; there's a big difference between "not a stub" (i.e., Start), and "a full article, but only about half sourced" (i.e., still Start).--ragesoss 03:10, 7 July 2007 (UTC)
Indeed. Also I'd like some extra stub levels: there's a big difference between a "one sentence, no sources" stub and a "two or three paragraphs, partially sourced" article, which is still usually going at stub. To get back towards the point of the essay, I believe extra levels at these points would be a big incentive for improving the quality of the overall encyclopedia from the bottom up (rather than the top down, as most current FA/GA-type drives do). Espresso Addict 03:49, 7 July 2007 (UTC)
I would worry that too many gradations would lead to fights over the ratings rather than active contributions to the articles. Although the gradations are broad, I think that they have to be since they apply to so many different types of articles. I actually think that it is good that an article remains at "start" before it is well-sourced. I like the idea that it is only "beginning" since its research has only begun. Awadewit | talk 03:18, 7 July 2007 (UTC)
It just seems to have led in practice to such a coarse filter that (1) there's little incentive for the editor(s) involved to improve the article; and (2) it's difficult if not impossible to make any meaningful statements about the quality of the encyclopedia, despite the various ongoing rating drives. Espresso Addict 03:36, 7 July 2007 (UTC)
They are not driven to improve the encyclopedia simply for the sake of improving the encyclopedia? I would think that there are very few people here that are driven by ratings alone. The time and effort it takes to write a good article is not proportionate to the "B", "GA" or "FA" rating. And the FA process is anything but encouraging - it is positively discouraging sometimes. Awadewit | talk 04:13, 7 July 2007 (UTC)
Of course there are other incentives than ratings to improve articles, although the flourishing barnstar culture tends to suggest that brownie point earning would be helpful. My point was that the rating system as currently managed (at least on those projects I've observed) does more to put people off improving articles than it does to incentivise them. The number of times one sees comments along the lines of "this is a perfectly good article, it's simply insulting to label it 'Start'"... Espresso Addict 05:17, 7 July 2007 (UTC)
I think that the barnstars are slightly different - they are personal rewards for contributors. I thought that the article ratings were supposed to help people understand where wikipedia stood at large (so that essays such as this could be written). Editors know their own articles best (I would like to think)- sometimes I think they should rate them themselves (up to a point). On the other hand, I think what a nightmare - everyone will rate their articles "B" (I think of my freshmen - "I deserve an "A" because I tried"). Awadewit | talk 05:55, 7 July 2007 (UTC)
However much article ratings should be for the benefit of the whole project, editors do take them very, very personally. Confession -- I used to rate my own articles. It was rather hard to be completely objective and I did tend to the slightly generous side (though I never awarded anything above than Start). I think that if there were more levels and more precise criteria (probably on a project-by-project basis), self-appraisal would be less of an issue, and it might just reduce all those article rating backlogs. Espresso Addict 06:32, 7 July 2007 (UTC)

I didn't realize that people took it so seriously. I certainly don't. I took that whole ("This is not a negative grade. There are no negative grades in Wikipedia" line pretty seriously, I guess.) Also, I rate my own articles all of the time. If I had to wait for other people to rate them, it would never happen. I write on topics that wikipedia seems to consider pretty obscure (usually I edit alone). Since I tend to be a perfectionist, I tend to be pretty harsh on myself. What I do take seriously is the FAC process, but that is an entirely separate discussion, I think. Awadewit | talk 06:50, 7 July 2007 (UTC)

I'm not quite sure why editors take the rating so seriously. They do realize how arbitrary it is, don't they? Since I'm a graduate student in English, I know that if you get a bunch of professors or grad students in a room and ask them to grade a series of essays, they will not all give them the same grades (sad, but true). There are just too many different factors affecting grading. Since those are people trained to assess writing (supposedly), what can you expect from those not? Have you ever seen the breakdowns for the grading rubrics on something like the GRE essay section? They are no more specific than what we have. Being that specific only paralyzes graders. Also, it takes GRE or SAT graders so long to grade because they have to come to a consensus about the grade of your essay (2 or 3 graders have to agree within a point or something like that) - perhaps there should be something like that here? Three people have to agree on the rating? Ah, what a nightmare. Awadewit | talk 06:50, 7 July 2007 (UTC)

Yes, I often tend to write on the obscure, so self-assessment does make a lot of sense. As to taking the rating seriously, if one expends a lot of energy on an article, and the only feedback one gets is someone driving by and rating it 'Stub', then it can be rather upsetting, particularly if one isn't used to receiving criticism. I do wish more people wrote explanatory comments when they rated, though I've seen those come over badly, too. I'm a Brit, so I don't know precisely what you mean by GRE or SAT, but I don't think there's any hope of having multiple people rate articles; the backlog is just too overwhelming.
I would hate for the assessment to be (for me, for example) "why are you using this random person's book instead of SparkNotes"? Meaning, why are you using actual scholarship instead of a commercially-produced cheating tool? I can see it now. I agree that multiple people can't rate, but that is really the only way to achieve some sort of consistency (GRE and SAT - standardized tests for graduate school and college in the US - sorry). Awadewit | talk 08:01, 7 July 2007 (UTC)
One would hope that reviewers would stick to areas that they knew just a teensy bit about... though ignorance never seems to stop certain contributors. Espresso Addict 00:07, 9 July 2007 (UTC)
PS I'm enjoying this discussion, but I'm just off for the weekend and won't have any internet access, so apologies for both present terseness and future muteness till Monday. Espresso Addict 07:32, 7 July 2007 (UTC)
How horrifying. No internet access. I can't stand it even in the car. What happens when I want to know the entire list of dictators assassinated by a foreign power while driving down the interstate? Awadewit | talk 08:01, 7 July 2007 (UTC)
Out of curiosity. Has anyone ever tried to figure out at what level the articles in, say, Britannica would be assessed at? If that were done, then we'd have a much clearer picture of our standard of quality vs. Britannica's standard. John Carter 15:59, 8 July 2007 (UTC)
I've not seen a recent edition of Britannica, but as I recall the 1950s edition had entries of around 1-4 paragraphs on most subjects, with no pictures and usually no references (although all articles are signed), ie somewhere in the decent stub to start class, if you ignore the lack of references -- one reason why I find the "Wikipedia is failing, almost all our articles are stubs/starts" argument unconvincing. Espresso Addict 20:30, 8 July 2007 (UTC)

[edit] It's not the stubs

This is a pretty good essay - I agree with much of it, but I just thought I'd make one slight comment. Stubs and start-class articles aren't bad. Even Britannica has short articles. Some subjects are notable but just don't have the wealth of sources out there to become FA-quality. It's the poorly written, unwikified, unsourced articles that are the problem, and it makes no difference whether these are one sentence or 100 kilobytes. (Although the former is much easier to clean up.)

On a slightly different note, I'm thinking of doing another week for Gnome Week. I'm just not sure which one. I'm thinking something in October or December might work (elves maybe, or Halloween things - it's a stretch, but that's not the point). Crystallina 06:34, 5 July 2007 (UTC)

I'd agree with this comment. Most paper encyclopedias are largely composed of articles of no more than two or three paragraphs. I'm worried that the drive to expand articles to feature length can sometimes lead to unnecessary bloating. Concision is often beneficial, particularly where it involves the exclusion of repetition and irrelevant detail. The problem isn't length, it's quality. Espresso Addict 15:23, 5 July 2007 (UTC)
I thought all of this was taken into consideration within the rating system. A single paragraph on Plato, for example, would be insufficient, while a single paragraph on one of the nineteenth-century state legislators of South Carolina may be all we can muster or, indeed, need to muster. Also, in case anyone forgot, GA was apparently originally started to rate articles that could not make it to FA. Note that the very first sentence of the GAC page is: "Wikipedia:Good articles is a list of articles considered to be of good quality but which are unlikely to be suitable featured article candidates." I don't think it is being used that way (certainly, I have not been encouraged to use it that way), which accounts for some of the backlog. Perhaps it should return to that mission and perhaps there should be a special group of GA reviewers dedicated to "stubs"; a great "stub" can be submitted there under special guidelines. Awadewit | talk 03:07, 7 July 2007 (UTC)
This is certainly not the way the rating system is being used at the moment, at least in those areas of which I'm aware. Perhaps it's because much rating is being done in a hurry, but there's a tendency, I think, to rate by rote length (plus a few other simple criteria), rather than any consideration of how far the article has got towards its ideal length. It wouldn't matter so much if it didn't lead directly to "the sky is falling" analyses such as this one. Espresso Addict 03:27, 7 July 2007 (UTC)
Perhaps it should? Awadewit | talk 05:55, 7 July 2007 (UTC)
What, lead to sky-is-falling analyses? I fear they're as well harkened to as the climate change reports landing on Mr Bush's desk. Espresso Addict 07:12, 7 July 2007 (UTC)
The problem with not making one's pronoun referents clear. Although grammatically the "it" referred to the "analyses", I really meant it to refer to the "rating system". I meant to say that "it", the rating system, should be used the way I described. Sorry. Appalling on my part. That's why I get for writing so late at night. Awadewit | talk 08:04, 7 July 2007 (UTC)

[edit] GA/FA stats

Wikipedia:Featured_article_statistics and Wikipedia:Good_article_statistics in a See Also section might be good for this essay. The essay is focused on number of Wikipedia articles and not nescesarily GA/FA ratios though - anyone think it should or shouldn't be added? JoeSmack Talk 20:31, 7 July 2007 (UTC)

[edit] Assumptions of unlimited growth and other comments

The article seems to imply that new (and uninformative) stubs will continue to be added indefinately at a high or increasing rate (similar to the world's population, of which a portion are wikipedians, so that's fitting) and that meaningful edits and improvements to the stubs will continue to be done at a steady but lower rate, as with a limited resource, which is fitting if you feel the number of 'experienced' and committed wikipedians has plateau'd. Given those assumptions, the logic that we'll never catch up and are destinated to have a resource of decreasing 'average' quality makes sense.

However, the 'total' quality of the collected knowledge of humankind as represented by wikipedia will continue to increase, regardless of the 'gap' described above. So I really don't see the sky-is-falling problem but rather a this-is-a-bigger-project-than-we-anticipated-challenge. People interested in only 'quality' will only look at the FA/GA sub-pedia, and they will have more and more articles in that category each month. (Now, if the 'demoted' FA/GA rate was greater than the promotion rate, I would be worried).

The concern with the disequilibrium between new/bad articles and vandalism versus new good articles is in part an unlimited growth argument, with limited resources, and can probably be related in some clever way to Malthus and his theory on human suffering, but I will leave that ideas as a stub and let someone cleverer than I to make the connection. The point is, however, as Λυδαcιτγ seems to have also indicated above on Canada Day, that the system might, like population often is, be more like a biological s-curve than an exponential curve, in that there are in fact limits in many of the domains of knowlege that need to be encoded in an encyclopedia. There are a finite number of countries, cities, people, events, species, body parts, medical conditions, soccer tournaments, video game characters, musical bands, or politicians that people care about. That number is huge, and it is much bigger than 1.8 million, but it's limited. The gap between what's encoded and what people think ought to be is still huge, making for fertile ground for very rapid expansion of so-called 'poor' articles. But once there is a stub for every hamlet in Australia entered diligently (and importantly, without some trigger happy AfD mavrick deleting them faster than they can get made because they are not in their idea of notable or encyclopeadic), the folks interested in Australian Geography may start to improve the hamlet articles rather than add more stubs. As Pascal.Tesson seems to mention above, I think that's how and why some of the domain wikiProjects were born--essentially, organically as domains matured and realized they had a 'poor' page for almost every topic they needed to cover but wanted greater quality on their 'vital' topics.

So absolutely, do whatever you are 'inspired' to do in order to 'inspire' others to improve the quality of articles, and the tech guys will continue to make better tools to help us keep track of what is yet to be done or needs to be fixed, and the 'process' folks will hopefully continue to simplify the policies and tags and templates required to manage pages -- but encoding all of human knowledge is probably a mind-boggling several decade process at best, and that's just to get caught up on an inventory of what we think we've already learned in the past ~5,000 years. So as long as, net, people are contributing constructively, then even a sloppy placeholder isn't out of place.

So my focus would be on things that seem to already be core to the wikipedia philosophy: a) enticing vandalous editors into contributors, b) curbing the negative efforts of some disgruntled or malicious editors (or admins), c) removing the (evident) barriers to encouraging new editors to feel like this is the place they should (learn to properly) document their (neutral) view of the world rather than having to set up 100's of independent 'reference' style wiki sites because they've been discouraged from editing here by the 'old-school' or 'know-it-all' crowd, d) improve technologies and 'processes' for 'rating' but also 'auditing' and 'curating' articles so that mistakes, unsourced info and other problem areas can more easily be identified, highlighted, fixed -- but don't worry about the fact that there are more topics than authors at this time. Do you really *ever* need a feature article quality article for every pokemon character? No. Does that mean we should not have a pokemon stub for each character, if that what people (and future rocket scientists) want to take time documeting? No. Take a deep breath. Let it go. Pick a different success metric. JetheroTalk 05:16, 9 July 2007 (UTC)