The Wikipedia Debate

Getting away from politics for a moment (maybe), one of the most interesting debates going on at the moment is about the Wikipedia, the open-source encyclopedia-like reference site that anyone can edit. I've been interested in projects like Wikipedia in theory for a while -- in part because I'm interested in whether such open systems can be useful for the intellectual work of politics -- but recently have been amazed by how vastly useful it is. For example, to take a kind of narrow interest of mine, twice in the last few weeks I've needed (or wanted) to know who was chair of a particular congressional committee at a particular time in recent history. That's difficult information to find; some of the committees have histories on their own web sites, but most don't. But it's all there on Wikipedia.

But there is certainly unevenness in Wikipedia, as some topics -- especially technology -- are covered in staggering detail and others less so, so it's hard to know whether a comparable bit of info in another field would be found. On the other hand, it only takes a few seconds to find out. If you find it, is it accurate? I'm reasonably confident about such basic fact information as who chaired the House Ways and Means Committee in 1967. But there's a lot more to an encyclopedia than such raw facts, and there are questions about the Wikipedia's reliability when it comes to historically contested issues, such as the year of Alexander Hamilton's birth, or those subject to interpretation, which is probably most of human knowledge.

The debate is summed up in this article in Wired News, but the source of most of it is on the Many-to-Many blog, a group blog devoted to social software (and often way over my head -- what are "folksonomies," anyway?), particularly the arguments of Clay Shirky and danah boyd. I tend to lean toward Clay's position, which is basically that the Wikipedia is an amazing process and initiative, not directly comparable to the Encyclopedia Britannica but with some great strengths (correcting errors, covering recent events such as the Tsunami, being free and accessible, and covering some topics more thoroughly). I love the underlying "wiki" concept, which is to make it easy to fix errors rather than making it impossible to commit the errors. In an earlier test of Wikipedia, errors were deliberately added to several articles to see how quickly they were corrected; most were fixed within hours.

On the other hand, I'm sympathetic to the other side of the argument, which is that a factually unreliable encyclopedia is pretty useless. (As I told my daughter a few weeks when she was waving a magic wand around a little too wildly, "A broken magic wand is just a stick.") And there are plenty of unsupported assertions, dull writing, and misplaced priorities in Wikipedia articles. But I'm not really an encyclopedia reader anyway; just the ability to use Wikipedia to check an unfamiliar name or find a weird fact instantly is amazing. As Clay Shirky says, it is more of a competitor to or supplement to Google than to the Brittanica encyclopedia.

Much of the debate has been about whether there should be some special privilege for "experts" in contributing to the Wikipedia, and whether the resistance to that option shows a cultish anti-elitism. Expertise is oddly distributed, and many people with obscure interests are as credible as experts on the issue as tenured, overcommitted academics. I'm surprised, though, that Wikipedia doesn't have any sort of system for vetting the reputation of individual contributors, in the way that Amazon reviews do, or that blogs generally do. If I could see something that said that the original entry on a topic was written by so-and-so, who is either (1) a credentialed expert in the field or (2) author of 15 other good Wikipedia entries or (3) highly rated for her corrections or something like that, it would help. And then perhaps to see the same for other contributors to/editors of the page, ranked by the magnitude of their contributions.

I'd also be interested in knowing if a Wikipedia author represents a strongly held view about a contested topic. I don't object to that, and in fact like reading something that I know might be a little opinionated if I can basically situate the opinion in relation to other views. Probably the greatest encyclopedia in history was the 1911 Britannica, with longer, signed entries by many of the leading minds of the day. Their views were idiosyncratic and I'm sure there were "errors" even relative to what was thought at the time, but at least a reader knows where those errors are coming from. On the other hand, that 1911 encyclopedia was not what we now consider an encyclopedia, just as Diderot's was not, and perhaps Wikipedia is not either.

Comments

[Dan]

As a teacher, I must say that Wikipedia is causing a whole host of new problems. In fact, it has gotten so bad that I give out their Disclaimer at the beginning of each semester, which is a wonderful explanation of the problems of verifiability. That said, it also provides a nice moment to discuss legitimacy and perspective, as well as the privilege afforded to so-called experts. I like it as it is. The problem isn't with Wikipedia - that will always be problematic (and, hence, exciting). The problem is teaching people about how to use the internet IN GENERAL.

[joe o]

I am impressed with the accuracy of the Wikipedia and with pretty much all of the info on the Internet you find with high google ranks. The Wikipedia has a lot of information you can't find in conventional encyclopedias. It is so quick, you can look stuff up just for the hell of it.

The critique of Wikipedia by the makers of conventional encyclopidias is pretty shameful. Its like reading a critique of automobiles from a blacksmith. You really have to hope no one is stupid enough to follow their advice.

[Grant]

Folksonomy

[Steve]

Mark, it's entirely possible to see who made any given change to a page; you can also see any person's change history. These are not the most user-friendly of pages, but the information is there.

[Aaron]

I believe a study of Wikipedia was done where, rather than serious errors, changes were made in the details of various posts. These went undiscovered as I remember it.

[praktike]

Mark, you gotta check this out:

http://www.govtrack.us/

It's right up your alley.

[none]

The intersting part will come as Wikipaedia levels off in total content. The current model has focused on quantity and ensuring there is something on each topic. There is a long entry on Alexander the Great, great thing about this system is that the quality will now increase as that is what is left as a Work.

[Slowking Man]

An article version validation system is currently in the works. Think Slashdot, but without the trolling and flaming. Essentially, it will allow revisions of articles to be rated. A link to the last "stable" version will be prominently featured, allowing users to more easily circumvent vandalism, questionable edits, and the like. The delay is mainly technical; the Wikimedia Foundation is severely lacking developers to write the Mediawiki software and maintain the servers.

[Matt Brown]

There are quite a few pages on Wikipedia that have been automatically created by programs working from data sources; one set of which is on places in the US from official census data.

That's the kind of page that experiment that changed subtle data altered. The thing is, the quality of error-checking differs per article. If there are few or no current contributors interested in a certain article, its accuracy can't be nearly as good as a popular page - where many people will oversee changes.

[Fred Bauder]

On Wikipedia an editor who "represents a strongly held view about a contested topic" hides behind an assertion that it is they who are following the Wikipedia:Neutral point of view policy and that those who oppose them are violating it. In some instances there are small groups who work together. This is one of the more refractory aspects of Wikipedia policy. A candid statement of bias would constitute a violation of policy. In that respect Wikipedia is rather like a University political science department.

[trolls on patroll]

Wikipedia is just a user interface, a portal. The "neutral point of view" is more or less a syntax convention, how one writes, not what one writes about. It just means that whatever bias is presented by the particular people involved, it must be presented AS IF it were objective - "A says B about C" and so forth. But the vast majority of sentences obviously are statements of "what is" not "who said it", and must be, so it is highly selective which statements are challenged or disputed. The particular body of people at Wikipedia's portal doing this just do not matter as much as their systemic biases, that is: they are mostly bored Internet users! and tend heavily to young males with typing skills in the US and UK. So what they believe is what will be presented as if true.

What is important is the common licensing under GFDL of all the material permits anyone to start a new portal and correct the most grievous errors and biases even despite systemic opposition. The Wikinfo - http://wikinfo.org - has helped this happen in some areas, though often the best of the articles end up copied "back" into Wikipedia.

It is also important that portals be troll-friendly and not permit "sysop vandalism": the revert or delete of articles based on "who wrote them" rather than on factual inaccuracy and so on. No one can tell any other editors' intent, and so the only difference between a "sysop" and a "troll" is who has got there first and acquired the power to "block" the other's IP or "delete" the other's pages.

Sysops as a class do not acknowledge that there is even such a thing as sysop vandalism. This is the main problem WIkipedia has had in its history.

We, trolls, do acknowledge that some trolls such as SOLLOG's followers at WikipediaSucks.COM do in fact commit vandalism. This makes us better than the sysops, as a class. As if our choice to work only in equal power relationships did not already.

[Dan Keshet]

This is to Dan and any other teachers whose students use Wikipedia. Instead of sending them to our disclaimers, you might like to send them to Researching with Wikipedia, a guide we wrote that includes some of the weaknesses of Wikipedia, but also our strengths, and how to use Wikipedia to delve deeper into a topic, find more primary sources, et cetera.

If you have more that you would like to add to that page or experiences you have of students who have used Wikipedia for research, you can edit it or leave a message on it's talk page.

Most Wikipedians understand that Wikipedia is not intended as a primary research tool. However, Wikipedia can be very good at giving researchers unfamiliar with a subject an overview of the terrain that will help them research further.

[some guy]

FWIW,

The other thing Wikipedia is vulnerable to is people with extremist agendas turning everything into a "hotly contested topic" producing articles that basically say "Opinions About Shape of Earth Differ."

I visited the Wikipedia page devoted to the 2002-03 carnage in Gujarat, India, on two seperate occasions. The first time it was a reasonable summary of the accounts that had appeared in mainstream international news accounts. The second time someone had gone through and inserted the BJP talking-point objections to nearly every sentence of that summary.

I'd hate to see what could happen to an entry, on say, evolution. Or dinosaurs, for that matter.

[Warchild]

"intellectual work of politics" I'm not certain if that is intended to be a joke. I took it as one.
I like Wikipedia, the insertions and the deletions and the fanatical devotion of some of it's editors. It produces a nice place to inspire your own thought and perceptions while still providing the bare bones of "facts" enough facts to pursue your own ideas.
Wiki's and blogs, I think, are the future of the internet. A place for communication, not preaching, but communication as in dialogue instead of monologue

[dfg]

I am always surprised by the extent of Wikipedia's coverage, often finding a short, readable article on some topic of interest. It is clear that contributors not only add to the content of the articles, but also correct grammar, spelling and readability. Of course, as has been mentioned, if facts are really important, the researcher should always be looking at more than one source anyway. Especially useful are the imbedded links to other related articles as well as links to other web sites. Really, a nice contribution to knowledge.

[sfdsd]

fdgdf