Wikipedia. . .
Well, everybody else is having a say. . .
My stance on Wikipedia is the common one: It's a good place to look for some things, but always bearing in mind it's got many innaccuracies as well.
But one claim that pro-Wikipedia people make I do want to debunk: The claim that Wikipedia is the best system because it uses peer review, an ancient, well-established system.
However, this is not the case: Peer-review isn't used on Wikipedia.
Seriously.
When information is made available for peer review, people look at it, and point out mistakes & omissions. The mistakes are corrected, ommissions filled in, and it gets sent out for review again.
Eventually, with this process, you (ideally) get an article that has had all mistakes found and all blanks filled in. This is a useful article. It's useful, because it's a one-way process: Once a mistake is removed, it doesn't get put back in. Once missing information has been added, it doesn't get removed.
With Wikipedia, the process looses the one-way process that makes peer review work. An article is put up. Somebody sees a problem and changes it. Somebody else doesn't like the change, and reverts the article back to the original.
Because Wikipedia uses a dynamic, two-way process, it never acheives authoritive status: Mistakes can always be, and often are, put back into a corrected article. For instance, one person's experience:
I corrected some howling, stupid, this-will-get-you-a-fail-in-first-year-engineerin
The collaborative approach works great for things like the Linux kernel: If somebody makes a change, (a) it gets examined by experts before being accepted, and (b) if the change is inappropriate, the kernel breaks. A simple acid test: It works or it doesn't. Just writing isn't enough, it has to be written right
If somebody changes Wikipedia, it gets no expert examination, and it doesn't matter if the change is truth or fiction: You don't have to write accurately, you just have to write.
So whilst I agree that peer review is a good system, that's not relevant, because Wikipedia doesn't use it. Mistakes get introduced as well as eliminated using the Wikipedia approach. If it did use peer-review, if it used a one-way review system done by experts, Wikipedia would become authoritive. So long as it uses the current, anybody can change anything system, it never will. It will even drive away the experts so vital to peer review: Why would, for instance, an expert on evolution waste time writing useful documentation if it keeps getting erased by creationists? And vice versa, of course. . .
Maybe somebody should fork Wikipedia, and create "Wikipeerdia", a wiki where changes are only accepted after they get properly reviewed. Might be interesting. . .
3 Comments:
I disagree here. It's an evolutionary process. Something that's slow. But there are some really great articles on Wikipedia. And the key is "matured". The "matured" articles have gone through many such "two-way" processes as you point out and ultimately the information is refined further and further till it reaches a point of saturation.
Unfortunately it's slow and not always perceptible. Newer articles will take ages before somebody seriously does some work on it. And in a way, it's to be expected because it's a voluntary effort. Nobody's getting paid to review wikipedia entries. But there are experts who quietly watch the changes to certain articles and correct them.
The biggest problem seems to be that the expertise seems to be too thinly spread out because wikipedia is a general encyclopaedia and not a topical one. Therefore you will get some poorer articles here and there. But my general experience is that there are some authoritative articles on almost every subject you can think of and to my mind that is the purpose of Wikipedia - to provide basic knowledge on most topics and then point out to authoritative sources for further information. It's amazing that Wikipedia has reached the level it has, considering that it's maintained mostly by thousands of volunteers all over the world.
The review process is definitely "democratic". And like democracies it does take time to filter out the bad and accept the good. It's a slow process. But that's the price of freedom... Otherwise there's no need for a wiki at all. It might as well be static HTML pages maintained by a small group of individuals.
"With Wikipedia, the process looses the one-way process"
sorry for being anal, but "loses", not looses.
I agree. the example of creation vs evolution demonstrates that it could go back and forth forever. It shouldn't be possible to remove anything, just add your correction. eg the creationist could add to the evolutionist but neither could delete so the reader can make the choice on what to believe. As long as things can be deleted why would anyone go to the trouble of adding things.
I believe wiki will die because of this.
Post a Comment
<< Home