The trouble with wikis
March 11th, 2010Discussions of context on the Web tend invariably toward Wikipedia. The site is a wondrous exemplar of the radical new ways in which information can be created and structured after we’ve cast off our legacy media forbears.
And yet.
What Wikipedia lacks in beauty, it makes up for in ugliness. The editing interface is hideously unfriendly. Discussion pages – while often fascinating – are a hot mess. There’s no easy way to satisfy the audience desire for both overview and update; if Wikipedia introduces you to a subject, it’s often a poor place to follow that subject as it develops. And the site emphasizes comprehensiveness over synthesis – you wouldn’t go to Wikipedia for a concise introduction to the challenges in U.S. immigration policy, you’d look to it for a thorough history of everything there is to know about the subject.
Elsewhere on the Web, we haven’t seen much evolution beyond the model of the wiki set forth by Wikipedia. Wiki markup everywhere tends to be unintuitive at best. Templating for wikis is a nightmare. And they haven’t yet developed the type of storytelling variety you see in blogs, which have now evolved tumblelogging, microblogging, photoblogging, videoblogging, audioblogging, and more. Few people are yet using wikis to create content that is strikingly visual and concise. It’s certainly possible to create a wiki that’s attractive, but it’s quite an undertaking.
The ideal wiki technology would make it possible for a community to create and maintain something like The Money Meltdown, and one day, perhaps even The Bold Italic. It would enable you to easily link episodic and comprehensive information; for example, an update to the wiki could be tied to a blog post explaining a new development. And the interface would be as friendly as that of WordPress for would-be wiki creators, editors, and designers.
This is where I would really love to see more open sharing of data and empirical evidence on the demographic break-down of how different users experience different interfaces. After consulting on the launch of Berkeley & the State Dept.’s Opinion Space project, I’m amazed at how much totally different feedback one gets on the interface from outside the Media Love Bubble. At HacksHackers Burt’s planning a follow-a-journalist type day; we probably need many more follow-reader type days too.
“we haven’t seen much evolution beyond the model of the wiki set forth by Wikipedia”
This is true, but a really good example of what’s possible is WhoRunsGov
I used to spend time complaining about things that are wrong with Wikipedia until I decided to spend that time trying to learn what is right about it.
As the creator of a wiki on the MediaWiki platform, I can (from the inside out), say that you are correct — and incorrect.
First off, when it comes to Wikipedia/MediaWiki, my opinion is a little like Churchill’s quote about democracy being the worst form of government except for all those other forms…
What Wikipedia (and I’m speaking more of it’s underlying CMS) lacks in a friendly, intuitive user interface (yes, its usability sucks and it’s ugly in the eyes of some beholders), it more than makes up as a tool for taxonomy and knowledge-management. It took me months of working with the platform in creating SmallBusiness.com to finally comprehend its beauty. (There are other hidden treasures related to scalability, distributed computing and the need to manage a global network with the fewest possible people, but those are not related to this particular thread.)
The Wikipedia/Mediawiki community understands the complaints you raise and if you dig, you’ll find some incredible extensions that allow the platform to be adapted to address usability issues (ironically, the last place you’ll see many of these improvements is Wikipedia itself, where the scale of the site makes it a complex challenge to “improve” things).
My main point, however, is this: In terms of this discussion — the future of context — Wikipedia is not broken. Is it perfect? No. Is it finished? No
But does it provide the best laboratory for understanding the way people who use the web reward contextual content?
Does it provide, perhaps, a lesson in the role of “beauty” (or the lack thereof) when it comes to contextual content?
Does it provide a roadmap to how contextual media is not about how easy it is to comment (a time-stamped content element) than to taxonomize (a word that probably doesn’t exist).