Some observations on Wikipedia as a source

Old college threads.
Locked
User avatar
Skepticism and Animal Feed
Auron
Posts: 3238
Joined: Sat Oct 30, 2004 11:47 pm
Location: Arlington, VA

Some observations on Wikipedia as a source

Post by Skepticism and Animal Feed »

I love Wikipedia. I have used it as a source for almost every quizbowl question I have ever written. I hear that other people sometimes use Wikipedia as a quizbowl source too, or may be thinking about doing so. This post is my advice to them.

One thing you often hear is that Wikipedia is full of lies and mistakes. I think this is overplayed. There are mistakes, and there is vandalism, but most of the information on Wikipedia is correct. I'd go so far as to say that 99% of information on Wikipedia is correct, and that this rate is probably improving.

But even though Wikipedia is mostly factually correct, it does have serious problems, and these problems are systematic, endemic to Wikipedia itself rather than to any individual article, subject matter, or field of knowledge. These problems emerge largely not from people being malicious, but from people rationally responding to Wikipedia's incentive structure. I want to talk about some of these serious problems that I've noticed, because if you write questions out of Wikipedia and do not correct for these problems, your questions will suck.

And even if I’m wrong about accuracy, and Wikipedia IS a den of lies like Matt Weiner and other critics say, then you should still read this post, because double-checking clues from Wikipedia for factual accuracy alone isn’t sufficient to correct the problems I’m about to discuss.

I. Wikipedia articles blather on about irrelevant things

You cannot assume that just because something is mentioned in a Wikipedia article that it has any shred of importance, even if it is discussed at length.

For example, let's look at Wikipedia's article for Medieval Times (the dinner and tournament chain): http://en.wikipedia.org/wiki/Medieval_Times

Roughly half of the article is about a 1997 incident in which Medieval Times was forced to file bankruptcy because it had to pay millions on back taxes. For somebody who wants to understand Medieval Times, this is completely irrelevant: the bankruptcy did not affect Medieval Times's operation and did not shut down any dinner shows, and further it occurred almost 15 years ago. In a hypothetical tossup on Medieval Times, it should be at most a lead-in, and probably not even mentioned at all.

Wikipedia articles blather on about irrelevant things because there is no penalty for article writers who do this: as long as the information isn't incorrect, isn't POV, etc., nobody will delete it or discipline the author. This makes Wikipedia different from almost every other resource you could possibly use for a question. Contrast that with, say, a print source, where going on for too long will cost the publisher real money, meaning that your irrelevant things will be cut out. What does Wikipedia care if a few bits of its enormous server space will be taken out by a three paragraph digression?

This problem is even worse for events that occurred AFTER Wikipedia was founded. When a major event unfolds, Wikipedia people constantly edit the article. Pretty much every news story around the world that mentions the event gets incorporated in.

Read about something like Russia’s invasion of Georgia, or the civil war in Libya, and it becomes basically a play-by-play. Facts are put into the article as they come up, before any determination can be made as to their relevance, and are rarely deleted after they turn out to actually be irrelevant. There are entire sections of articles like these called “international reaction” (or something similar) that tells you what the foreign ministers of countries like Portugal and Hungary said. Now I love Portugal and Hungary as much as anyone else, but let’s face it, the opinions of their foreign ministers have been irrelevant for centuries.

Another great example is articles about professional athletes. I can’t tell you how many articles for MLB players I’ve seen that contain a sentence like “on July 7, 2006, he hit a game winning three-run home run against the Padres.” You know that, on July 7, some guy saw that home run, was like “whoa this could be notable”, and edited that in. Guess what, that home run turned out to be non-notable and didn’t ultimately affect the standings in a relevant way. But it’s still there: it’s not wrong so it probably won’t be deleted.

II. Wikipedia articles completely leave out important things

Let’s go back to the Medieval Times example. If you've ever attended a Medieval Times dinner show (and I have on multiple occasions), you will know that they feature not only jousting, but also feats of equestrian riding and falconry. You will also know that the various knights involved in the joust have different personalities and affiliations: one represents the church, one is evil, one is good, etc. You will also know that they represent different regions of Spain, and that each section of the crowd is assigned a color and told to cheer for the knight of that color. All of these are important things to know about Medieval Times, and none of them appear in the Wikipedia article about Medieval Times.

Just as there is no penalty for talking about irrelevant things on Wikipedia, there doesn’t seem to be any sort of penalty for leaving out important things. Again, I don’t suspect that this is malicious: I don’t think there are people who are ideologically opposed to falconry who are trying to cover up the fact that Medieval Time features falconry. The person who wrote the article might have had incomplete knowledge: he might have never been to a Medieval Times and honestly didn’t know that there was falconry. He might have known about it, but not had a good citation. Wikipedia is increasingly fanatical about requiring citations, and he might have figured that leaving out the part about falconry was better than getting into an edit war with somebody over whether or not there is actually falconry. Note that the Medieval Times article cites an obscure feminist film criticism textbook to prove that Medieval Times is mentioned in the movie “The Cable Guy”. Or the guy who wrote an article might have just been a bankruptcy law hobbyist who made an article about Medieval Times because he wanted to discuss the bankruptcy case. I think this last explanation, when generalized, explains a lot of things: people write Wikipedia articles about things that interest them and leave out things that don’t interest them, probably hoping that people who are interested in those things will fill in the gaps. Well, often they don’t.

III. Wikipedia sucks at giving you the big picture

The combination of I and II means that Wikipedia articles are often insufficient to give you a good "big picture" understanding of something you're unfamiliar with.

By big picture understanding, I mean things such as

(a) is this thing important at all
(b) WHY is this thing important
(c) what are the most important things about this thing

Wikipedia can teach you a lot of clues about a thing, but it can be very unhelpful when it comes to figuring out which of those clues matter.

If you lack a big picture understanding of your question subject, odds are your question will suck. You won’t know which clues to use: you may use clues that are obvious to everyone or clues that nobody has ever heard of. You will also neglect to put in clues that would be useful and helpful. And even if you miraculously use good clues, you’ll probably have a poor idea of how to put them in order, muddying the Westbrookian battlefield.

Also, not having a big picture understanding hurts your own improvement as a player. If you don’t have Jonathan Magin’s proverbial map of the subject and can’t connect it to other things (or even connect its various pieces internally) you won’t remember clues about it as well. So using Wikipedia alone as a source hurts yourself as much as it hurts others.

IV. Wikipedia makes up names for things

Wikipedia articles need names. Wikipedia literally could not function if its articles did not have names. Sometimes, a notable thing or event doesn't have a widely accepted name, or it doesn't have a name yet because it is in progress or recent. Wikipedia often responds to this conundrum by just making up a name.

This is especially true for events. Before you mention an event name you found on Wikipedia (or, god forbid, underline and bond an event name you found on Wikipedia), you should google that name and see if it shows up in non-Wikipedia sources. And remember, all of the mirrors of Wikipedia are NOT non-Wikipedia sources. You’ll actually have to click on all of your results and make sure they are not just copy/pasted versions of the same Wikipedia article you are fact-checking.
Bruce
Harvard '10 / UChicago '07 / Roycemore School '04
ACF Member emeritus
My guide to using Wikipedia as a question source
User avatar
Skepticism and Animal Feed
Auron
Posts: 3238
Joined: Sat Oct 30, 2004 11:47 pm
Location: Arlington, VA

Re: Some observations on Wikipedia as a source

Post by Skepticism and Animal Feed »

Wikipedia articles also have a tendency to over-rely on direct quotations from sources of dubious importance.

See the article on the Hipster subculture, for instance:
http://en.wikipedia.org/wiki/Hipster_(c ... ubculture)

Nowhere does this article actually define Hipster. Instead, it offers an impressively long series of quotes about hipsters, some of which purport to define the term, most of which are simply strings of quasi-academic terms. Nowhere is it indicated why we should care about these people being quoted and nowhere is there an attempt at a synthesis.

Somebody who isn't already familiar with hipsters will simply leave this article confused.
Bruce
Harvard '10 / UChicago '07 / Roycemore School '04
ACF Member emeritus
My guide to using Wikipedia as a question source
User avatar
Habitat_Against_Humanity
Rikku
Posts: 476
Joined: Sun Jan 21, 2007 8:51 pm
Location: Syracuse, NY

Re: Some observations on Wikipedia as a source

Post by Habitat_Against_Humanity »

Before we go any further, I think this needs to be seen if it hasn't already: http://en.wikipedia.org/wiki/User:Jnc/Cold_War_Classic
Rachel
UChicago 09
User avatar
AKKOLADE
Sin
Posts: 15783
Joined: Thu Apr 24, 2003 8:08 am

Re: Some observations on Wikipedia as a source

Post by AKKOLADE »

Didn't Bruce write that?
Fred Morlan
University of Kentucky CoP, 2017
International Quiz Bowl Tournaments, CEO, co-owner
former PACE member, president, etc.
former hsqbrank manager, former NAQT writer & subject editor, former hsqb Administrator/Chief Administrator
Edmund
Wakka
Posts: 176
Joined: Sun Feb 07, 2010 8:25 pm

Re: Some observations on Wikipedia as a source

Post by Edmund »

Thanks, Bruce, for this. The next time I need to explain why someone is not to write a question straight out of Wikipedia, even though it's probably factually accurate, I will direct them to this post.

I don't think that Medieval Times has a footing in Europe, but I have been here. It was my friend's birthday - he got dressed as a wench and much fun was had by all.
Edmund Dickinson
UK Quizbowl
University of Oxford '11
User avatar
Habitat_Against_Humanity
Rikku
Posts: 476
Joined: Sun Jan 21, 2007 8:51 pm
Location: Syracuse, NY

Re: Some observations on Wikipedia as a source

Post by Habitat_Against_Humanity »

Fred wrote:Didn't Bruce write that?
Yes indeed, though I was hoping some people hadn't seen it yet.
Rachel
UChicago 09
User avatar
Skepticism and Animal Feed
Auron
Posts: 3238
Joined: Sat Oct 30, 2004 11:47 pm
Location: Arlington, VA

Re: Some observations on Wikipedia as a source

Post by Skepticism and Animal Feed »

You may also be interested in the sequel, which was deleted even faster

http://en.wikipedia.org/wiki/User:XXxJe ... es_Delight

The first and only time I was able to cite Petey Pablo's PhD dissertation.
Bruce
Harvard '10 / UChicago '07 / Roycemore School '04
ACF Member emeritus
My guide to using Wikipedia as a question source
Locked