The Longest Now

On Wikinews : Africa vs. Endor
Saturday October 30th 2004, 6:46 am
Filed under: %a la mod

Just yesterday, Rebecca made a few good suggestions about how a prospective Wikinews project could proceed without annoying any traditional journalists or alienating their audiences. The idea of tackling new topics for investigative journalism is a particularly interesting one. I think, however, that most original journalism will start out addressing local news that traditional media haven’t deigned to pick up (and this may have a broader audience than traditionally expected… but then when’s the last time that any journalist, traditional or not, asked you what you wanted to read?).

But when she talks in terms of choosing angles on stories, she misses one of the strengths of this particular wiki tradition. What wikinews should be able to do better than any other news source, is mention and contextualize all of the major angles on a story (including, perhaps, a novel angle not covered by other media); expose the aspects of a news report that are hotly contested among its various authors; and expose the revision process involved in newsmaking.

She gets in a dig about Middle-earth having better coverage in the encyclopedia than most of Africa, referring to Ethan Z‘s comment last month that the article on the Congo Civil War was shorter than that on Tolkien’s Middle-earth. I feel bad that the source of these claims is right here in my backyard, so let me try to set matters right. The initial distribution of content on Wikipedia was spotty, influenced by the interests of the initial contributors. When you have a blank canvas, you have to pick somewhere to start. Since then, the shared goal of a neutral, comprehensive encyclopedia has guided how coverage has broadened.
Many contributors to Wikipedia are not contributing in their area of expertise, but instead researching new things as they contribute articles where the encyclopedia needs them most (see for instance the recent new-article contest focusing on filling article requests).

Yes, Wikipedia has fantastic, perhaps unequalled coverage of Middle-earth — currently there are almost 900 related articles, half of which are short descriptions of the hundreds of characters and places that compose that most detailed of fantasy worlds. And yes, Wikipedia’s coverage of Africa pales in comparison to its coverage of other continents.

Nevertheless, there are 3000 articles directly related to Africa, including, for instance, Economy of Africa and Congo Free State. There are no articles of such depth or quality about hobbits and elves — despite Tolkien’s talent, he could hardly compete with the exotic detail of real life.

[If you’re curious about the numbers, I just spent half an hour reviewing these topics via the beta categorization system.]

4 Comments so far
Leave a comment

I’m sorry my way of framing the issue of systemic biases in peer produced media got under your skin. I’ve offered the observation – not mine, by the way, but Xed Mac’s – as a pithy way of making a more complex point: the beauty and brilliance of wikipedia is a function of the fact that people write what they want to write about. The downside of this strength is that the group work is stronger in areas that are of interest to the writers than in areas where they lack expertise.

I don’t find it especially reassuring that wikipedia authors are branching out into topics they’re not naturally inclined to cover. While I appreciate the sentiment behind projects that try to get wikipedia authors working on novel subjects, I’d much rather see a concerted effort to change the peer base, and recruit more people who would naturally be inclined to write articles on African economics. While I’m sure I could research and assemble a competent article on Ents, I would be at a disadvantage, writing on a field where I’m inexperienced, and, historically speaking, uninterested – I worry that authors taking on subjects they know little about – except that they’re uncovered – may be as lost as I would be writing about Tolkien.

Defending wikipedia’s structure by demonstrating that there are three times as many articles on a real place where 800 million people live than on a compelling fictional universe seems to demonstrate my point as much as yours. (In making the statement in public, as I did in a speech last week, I’ve used the phrase “More information on Middle Earth than on Central Africa”, a comparison that may still turn out to be true.)

My point is not that there’s an optimal Africa/Tolkein ratio for an online encyclopedia, but that a peer produced system will come into being differently than a top-down encyclopedia. This is not a bad thing – it makes Wikipedia tremendously useful for some purposes – but it means that there are ways in which a peer-produced encyclopedia will come up short in comparison to a traditional encyclopedia. What percent of the total articles on Wikipedia cover African issues, versus a traditional encyclopedia?

Holding wikipedia and a traditional encyclopedia side by side, I suspect that wikipedia has better and deeper coversage of certain areas (technology, some aspects of politics, some types of media) than the traditional encyclopedia, and weaker coverage on some economic, political and historical topics. (I’m looking for good ways to do this experiment, and would be grateful for your thoughts on it.) Even if that proves to be the case – that’s okay. That’s how peer production works. Once we’ve identified the problem, we can work to change the peer group working on wikipedia and broaden coverage on certain topics. Or we can decide that wikipedia may have a different coverage profile than a traditional encyclopedia.

What concerns me is that wikipedia is increasingly held up as a _replacement_ for traditional encyclopedic media, and I don’t think it’s there yet. Just because wikipedia, in terms of raw numbers, has more articles than a traditional encyclopedia, it does not mean that coverage in all areas is as broad or deep.

Comment by Ethan Zuckerman 10.31.04 @ 7:36 am

Thanks for sharing your reactions, Ethan. You haven’t gotten under my skin yourself. I was pleased to see your original post on the subject, and am excited by the attention it has gotten — if I am touchy about exaggerations of this class of comparison, it is precisely because I want people to focus on how the pool of contributors can be and is being expanded. Imperfect coverage should not be construed as a reason to discount what information exists. And I want the metric for coverage to be updated as quickly as Wikipedia itself is.

There is slightly more information in Wikipedia on Middle Earth than on Central Africa, though that will not be the case for much longer — in the former case, most everything that could be written about the subject has been. And one could say more strongly – there should be no single country with less information about its history and inhabitants, than exists about Middle Earth. I’ll let you know when Central Africa is no longer a small enough region for comparison.

I share your concerns about Wikipedia being used as a replacement for traditional sources for research. One of the misfortunes of our society is how quickly something easier to use can completely replace a predecessor which was still superior in many ways. In this particular case, I am sure that we can eventually have uniformly broad and deep coverage in a free encyclopedia.

Comment by sj 11.02.04 @ 3:31 pm

Endor is still winning.

Comment by Blue 05.15.10 @ 3:35 pm

In some ways, modern pop culture continues to grow and attract new editors more than writing and research about Africa. Pop culture is right there, in the face of Internet junkies, and it’s fun to think and right about — until we get an equal number of readers and editors who are from Africa, this balance won’t shift.

But as for Endor in particular, African content is growing steadily across all topic areas, whereas there’s a finite amount that can ever be written about Endor, and we’ve covered most of it.

Comment by metasj 05.17.10 @ 4:55 pm

Leave a comment
Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Bad Behavior has blocked 440 access attempts in the last 7 days.