Slashdot Mirror
Wikipedia's Participation Problem
holy_calamity writes "More people use Wikipedia than ever but the number of people contributing to the project has declined by a third since 2007, and it still has significant gaps in its quality and coverage. MIT Technology Review reports on the troubled efforts to make the site more welcoming to newcomers, which Jimmy Wales says must succeed if Wikipedia is to address its failings."
Wikipedia was run by people that equated quantity with quality. It was routine to see someone heralded as authoritative because they had made tens of thousand or more edits. In reality the only thing that shows is that someone is obsessive compulsive, doesn't have a job or has a job where they don't have to work. The result was large numbers of articles that were complete and utter crap, a few that were well qualified and the constant question of was the last edit done by a PhD that's an expert in the field or a bored teenager?
It's long overdue for quantity to step to the wayside so that quality can step up to the plate. When wikipedia can stop ranking editors by quantity and start ranking editors by quality the entire site will gain credibility. The concept that just anyone can know what their talking and edit something accordingly leads to idiots that cite wikipedia over the CDC or a thousand other examples I can think of.
Wikipedia still suffers from tremendous a vocal minority on certain political subjects that are locked and to prevent any viewpoint other than the vocal minority that won the right to represent their view on the given subject. Wikipedia has made improvements, but it has a hell of a long way to go before it can be anything other than a starting point for the curious and gullible.
Pick almost almost any random article, something not too obscure. Look for some cumbersome or inelegant prose and clean it up. Don't even change anything factual, just make the article objectively clearer. This isn't even very hard to do, since many articles are written by technical-types who aren't very proficient at communicating. You see this sort of thing with engineers especially; the kind of people who resented having to take English classes.
Now wait about five minutes. Your edit will automatically be reverted by a bot squatting on the article. And after a few seconds you'll receive an automated message, usually beginning with an insincere and condescending, "Welcome to Wikipedia! I've automatically reverted your edit because...".
You can try to start an edit war, but the entrenched editors of most articles have more seniority than you, they're "experts", and it's really not worth the hassle just to make small changes. So you end up with a lot of articles which seem like they have been written by people with Aspergers, or a tenuous grasp of English. I can't speak to the editing climate in other languages.
I don't have a comprehensive solution to this problem, but it probably has something to do with getting rid of the automated bots which protect pages. That'd be a decent start.
If you build it, nerds will come. Soylentnews.org
Whilst I would NEVER quote it in an academic essay, as a source of information on non-controversial topics, (e.g. dates in history, who wrote what, basic chemistry and physics issues, all you ever wanted to know about British Railway stations past and present...) it's excellent. The sources that it quotes are the next step in serious research, with the best articles quoting online primary resources. A core question is 'were encyclopedias ever that much better?' They all come with their own agenda and biases. It's not perfect, but it's a useful resource, as well as providing the occasional giggle.
I converted a paragraph that looked like it had been run through Google translate a few times into actual English. It was reverted. The people that claim Wikipedia entries as their own are generally some of the dumbest people on the internet. The YouTube commenters are the ones in charge.
I asked Jimmy directly about this in a pretty even handed way when he did the Slashdot interview questions back in August. He responded:
" Things have mostly stabilized. It's still not a crisis, but I still consider it to be important. One of the most exciting developments is the visual editor, which I hope will bring in a whole new class of editors who were turned off by the complexities of wikitext."
More or less he dismissed the premise that there was a problem in the first place, and any issues that are left could be handled with a better editor UI. Now, I do think the Wikimedia editor needs work, but Jimmy is kidding himself. Maybe he'll get a new rush of editors when they release the new UI, but I'm not convinced they'll stay.
I have not tried to contribute to wikipedia yet (though I have thought about it, I have been unsure whether I want to try given the currently climate described), but it occurred to me: how does one become an editor?
I am wondering, if current editors are appointed and have permanent control and this is causing problems, what if Wikipedia switched to something akin to slashdot's moderation (and metamoderation) tool? Let random people vote on if they think the change was warranted. They don't need to be experts on the topic, just answer yes or no as to whether the change was significant and properly documented with references. If so, then vote ok, and overrule the mods that may be blocking it. Is that not possible?
I see a lot of complaints here that boil down to "Subject matter experts try to contribute knowledge, dorky editors revert them over some stupid policy I don't understand the purpose of." From there it just devolves into poo-flinging. Most of you are missing the major policy points that require those reversions, and the truly deep (and perhaps, unsolvable?) problems which are the reasons for those policies. The WP policy stuff is really doing the best it can do, and you're caught up in blaming the "lesser of two evils" results of the process.
The *real* problem for WP can be broken down like this: The WP guys really do genuinely want to build an archive of all human knowledge, freely available to all humans, while minimizing bias and falsehood. They want it to be crowdsourced, too. The obvious problem is that lots of the crowd will contribute false or biased things, and those things have to be filtered out somehow. The primary mechanism of filtration isn't "assign brilliant people in every field to fact-check submissions based on their own expertise", because frankly that would only lead to more bias and more problems. So instead of futilely trying to judge objective truth in that matter, they redefine the objective to a more-attainable version of the truth: we want what's commonly accepted as the truth, not what objectively really is the truth.
The reason for this distinction is it can be enforced easily: all contributions of knowledge have to be backed up by external, 3rd party, editorially-verifiable sources. Then at least the version of truth that passes WP muster can be said to have passed societal muster in general before it arrived on WP, which diverts a large part of the truthiness problem. This leads to a pair of lesser problems, both of which deserve attention, but are very difficult to solve:
1) WP doesn't accept original work from subject matter experts. Even if you *are* the world's leading authority on Quantum Chromodynamics, it's not good enough for you to create an account in your name and start adding random facts from your head to an article about QCD Coloring. Even you, the SME, needs to actually referenced a published book or peer-reviewed journal article for each factoid you add to an article. Obviously, this pisses off SMEs that know what they're talking about; it's annoying to be required to find what is probably an objectively less-qualified source than yourself to back up your claims. Unfortunately, it's the only way to prevent false SMEs: people with an inflated view of their expertise and/or a clear fringe bias. It's also the only way that a committee of non-SME editors can validate the process.
2) Perhaps worse is the problem of self-referential loops with the 3rd-party sourcing. A number of issues come together to create the problem, and a typical example goes like this: A well-meaning person edits an article on Palm Trees in Florida, and adds some hearsay non-sense they heard from their neighbor about a new type of pest imported from Cuba that's attacking the trees and how they might all be gone within 10 years due to this pest. Because very few editors or bots are actively watching the Palm Trees in Florida article, this bullshit goes undetected for a while. Let's say two weeks later, someone gets around to reverting the edit for lack of a verifiable source. However, in the intervening two weeks, a well-meaning reporter for a local news station in Florida happens on the article, sees this shocking fact about Palm Trees dying to pestilence, and writes a local new story about it.
She doesn't cite Wikipedia because, well, that would seem unprofessional. So when the original submitter sees the reversion, the submitter goes googling for evidence to back up the claims and get un-reverted. She stumbles on the local news story and brings it back to the edit war as a verifiable source. The editors pretty much have to accept it, and a new and totally invalid factoid has erroneously become a part of human knowledge.
The problems here are man
I would like to echo this, if only to provide some balance to all the complaints. Not to deny that unwarranted (or at least poorly explained) reversions and deletions occur, but for a lot of these claims, upon closer examination, it turns out the edit had problems and violated one of Wikipedia's policies (most often because it's not the kind of information Wikipedia includes, with lack of sources the second-most-common reason). For some of the Slashdotters complaining, I would be astonished if their edits did in fact adhere to Wikipedia policy and style, just given the way they post here.
I've written at least a half-dozen Wikipedia articles or major article sections on a few different topics (bios of various authors, among others). None were summarily deleted or reverted; as far as I can tell (I don't keep close watch on them), they all still exist, and most still incorporate significant amounts of my contributions. They all seem to be in better shape now than when I first wrote them.
Here's what I did: First, I registered an account. Whenever I was thinking of making major changes/additions to an existing article (in some cases it was non-existent or a stub), I first posted on the talk page, expressing my interest in the topic, diplomatically describing what I thought could be improved, outlining what I was planning to do, and asking for comments and suggestions. This rarely got any replies, but ensured I wouldn't be stepping too hard on any toes.
In parallel with this, I figured out what the best available sources on the topic were: online material, sure (more for orientation than for incorporating), but preferably published books. If I didn't have them already I got hold of them, either through the library or second-hand (or Google Books if not in copyright). In some cases, articles in academic journals were more useful, and I was able to access these through a university library. I read or at least skimmed them.
Then I wrote the contribution, using (not copying verbatim) the information from the references and providing citations to them. I tried to write in a straightforward but encyclopedic style, to format it to be consistent with the rest of the article, and to follow the letter and the spirit of Wikipedia rules. I submitted my edits, and sometimes made another comment on the talk page, inviting comments or explaining any deletions I'd made.
I'm not claiming this is a guaranteed way to get your text on Wikipedia, but in my experience it worked. While I've also had some negative Wikipedia experiences (with smaller edits to other pages, where I didn't go through such an elaborate procedure), overall I find that as long as you understand the basic Wikipedia principles (what kind of stuff belongs and doesn't belong), can separate yourself from the work enough that your edits aren't blatantly about your own ego or biases, have solid sources for what you write, and can string a coherent sentence together, people will usually take you seriously.
There are many disgruntled attempted-Wikipedia-editors out there, some with more legitimate grievances than others. Sometimes it seems like most people who have tried to contribute got shot down. But of course, the most embittered ones are the ones most likely to go on about it: the real distribution could be quite different. Or maybe it's simply that most people aren't capable of making good, suitable contributions to an open encyclopedia. Would that be particularly surprising?