September 4th, 2006

Filed under:
Books, Technology
9 Comments »

Article Length in Wikipedia

You know, I don’t spend much time in the Wikipedia community. I probably make a small (often trivial) edit to an article about once a month (here’s what I’ve done, though I sometimes forget to log in when making edits). On the other hand, I visit the site on a daily basis.

I went there today curious about the etymology of ‘hollaback girl’, subject of Pink’s Gwen Stefani’s popular 2005 single. I found a fairly decent explanation:

For the remainder of the evening, Stefani and Williams incorporated the inspiration into the lyrics that would eventually become “Hollaback Girl”. The two of them decided that Stefani did not have to have an answer for her intentions and that the choices she made were based on what she felt was wrong or right.

Here’s what’s interesting: the article runs to about 3000 words. The average Wikipedia article has about 400 words. For a random comparison, the article about the fish that brought about Steve Irwin’s untimely demise (man, Australia has so much stuff that can kill you) has 1800 words.

In traditional encyclopedias (and most reference books), longer articles imply greater importance. The article on, say, Canada is longer than the article on Campbell River (and the article on the town is presumably longer than the article on the river). Article length, then, is an important piece of metadata for the reader.

Page scarcity isn’t an issue online–we have as much ‘ink’ as we want. So, a fleeting pop single gets more words than a fish that’s been around for millions of years.

I don’t necessarily think we should retain the dead-tree paradigm of ‘more ink = more important’. We should, however, think about presenting other metadata to denote notableness or importance. If I don’t know, how can I determine whether Ernest Hemingway is more notable than, say, some regionally famous author? We could show the number of page views, the amount of discussion and so forth, but I’m afraid those metrics would equate timeliness and controversy with importance. Any suggestions?

I did find these guidelines on Wikipedia for article length, but that’s more of a technical discussion than a philosophical one.

By the way, if you use Wikipedia as much as I do, I encourage you to make a donation. I just did, and now I feel smug and guilt-free. Alternately, you can always buy a shirt, mug or thong.

Comments: 9 Responses so far

Lisa
September 4th, 2006
2:38 pm

Not Pink — Gwen Stefani

[Reply]

Travis
September 4th, 2006
3:07 pm

They don’t sell thongs. (Too bad we can’t edit your blog post.)

[Reply]

darren
September 4th, 2006
3:19 pm

Lisa: Thanks for that. Can’t keep my hardboiled songstresses straight.

Travis: I glanced at one store, and saw that they don’t do the CafePress thong. In truth, I just thought it was a little funny to imagine buying a Wikipedia-branded thong. So, I left it in, hoping that one of the other stores did.

Which raises the question: why don’t they? It doesn’t cost them anymore to list them?

As for editing my entries, I wouldn’t mind, actually, except that the site lacks the critical mass necessary to self-moderate the edits.

[Reply]

jd
September 4th, 2006
4:28 pm

i recently discovered that the ThunderCats (children’s show from the 90s) entry is about twenty times longer than the one on Picasso’s blue period. and, it has better pictures.

i wonder if there really is a need for a system that tells people the relative importance of the topics being covered, though. usually, it’s the sort of thing that can be picked up reading the entry. i guess one extra level of “metainformation” would be to organize all the entries into a giant tree-graph, where relative positioning would provide some clues.

[Reply]

Air
September 4th, 2006
4:43 pm

Call me a perv, but I just spent five minutes searching for a wikipedia thong, and only then read the comment (above) that there are no thongs. Now how sad is that! And it’s very sad about Steve Irwin, but I’m sure he would have wanted it that way had he been given the choice.

[Reply]

Ian King
September 4th, 2006
9:11 pm

Article size is also one of my big issues with Wikipedia. Subjects active after 1994 tend to become ridiculously long. The trouble is that the dominant Wikipedia attitude is to err towards including every last bit of information because there’s no real size limit, and articles are overrun with fan trivia, speculation, and laundry lists as a result.

The argument that space constraints don’t exist in electronic media is shallow. Increasing the size of any work, not just a wiki article, may be cheap, but the audience’s time is not. Frittering their time away by adding trivia of marginal utility — often obscuring important points and rendering the article useless in the process — is disrespectful. What is ten minutes of *your* time worth?

Ultimately, I don’t think that metadata — especially that drawn from unskilled users — is a substitute for good editing. You mentioned two misleading measures. Rating or discussing a subject’s relative importance also won’t mean much unless it’s weighted to register the opinions of actual experts in a field, and Wikipedia is pretty hostile toward any expertise that doesn’t involve PHP or JavaScript.

In the few well-edited Wikipedia articles I’ve read, the lead paragraph and article size are actually good indicators; such an article won’t include oodles of irrelevancies and the lead functions as an abstract. Written and edited judiciously, that first paragraph is enough to cue the reader to read on or move on.

[Reply]

Andy K
September 6th, 2006
12:45 am

The way I see it, Wikipedia is not an encyclopedia. It is what a bunch of people write about what they know. And if the goal is record knowledge, why would you leave anything out? The minute details are often the ones that make interesting associations to other articles. Plus, if you didn’t want every single person adding their two bits, why make it a wiki in the first place?

Consider the art on the Beatles’ Sgt. Pepper’s album. I have seen the album cover, I know it’s famous, and I know it contains lots of images of various people, but I haven’t ever been bothered to wonder who they are and what they represent. Without ever having looked it up, I bet there is a Wikipedia entry explaining it all. Useless trivia, entertaining links or insight into musical genius and social commentary, it’s all the same in the end.

For sure, good writing and good editing can make the entries easier to read and the audience quicker to find the info they want. But I feel the role of Wikipedia is to amass as much knowedge as possible, the long tail of every topic, if you will, so you can see where all those tails cross. That’s something that has never existed before and is good fun to surf. Good writing is a bonus.

Sure enough:
http://en.wikipedia.org/wiki/List_of_images_on_the_cover_of_Sgt._Pepper%27s_Lonely_Hearts_Club_Band

and more interesting: http://en.wikipedia.org/wiki/Sgt_Pepper%27s

As for metadata (or you know, a rating on a scale of 1..10) to denote importance, wouldn’t it all be subjective? Hollaback, as unlistenable as I think it is, was obviously very important in many ways to many people. We don’t really know how influential Picasso was and to how many people (though maybe there is a historical perspective on this–so many articles fail to put historical subjects into context). But now that wikipedia is recording the present, in 100 years people will know what impact Hollaback had. Actually, I find that wikipedia is becoming a repository and explanation of pop culture–a good yet embarrassing thing for us Gen X and Y’ers who are getting disconnected.

[Reply]

Pseudo_Naufana
October 7th, 2006
5:27 pm

There used to be a Wikipedia Thong, I bought one for a girlfriend a couple of years ago.

[Reply]

Wikigroaning, An Wikipedia Article Length Redux | DarrenBarefoot.com
June 5th, 2007
2:09 pm

[...] Geoff linked to my post about the peculiar article length trends in Wikipedia, and also referenced a Something Awful article entitled “The Art of Wikigroaning“. What’s Wikigroaning? For example, the article called “Knight“. Then, find a somehow similar article that is longer, but at the same time, useless to a very large fraction of the population. In this case, we’ll go with “Jedi Knight“. Open both of the links and compare the lengths of the two articles. Compare not only that, but how well concepts are explored, and the greater professionalism with which the longer article was likely created. [...]

Add a comment