Wikiscape

For me, 2007 was the year of the wiki. I find myself looking up the oddest words on Wikipedia; did you know wiki is hawaiian for “fast”? It does and being able to look up such things all in one place made the web seem smaller. Really though, it just became useful.

Google wants in on it; Knol (ironically, Wikipedia has the best info on it) is a sort of content hub that allows authors to create articles on any topic. Users can edit pages, though only when given permission from the initial author. It’s a good idea: compete with Wikipedia for the top search spot, but show Google Ads on the side. Google is fast becoming an eco-system; a virtual mega-mall that fills our every need so we don’t ever have to leave the safe zone.

On closer inspection, Knol looks more like ex-wikipedian Larry Sangers doomed Citizendium project. Citizendium is a Wikipedia fork that aims to rid the world of those ever-clever “you can’t trust Wikipedia because anyone can edit it” comments. The idea is that all its articles are to be “approved” by someone in-the-know. At the time of this writing, Citizendium sports a whopping 46 approved articles.

Of this triad, I see both Wikipedia and Knol keeping their buoyancy. Citizendium is a club of nerds, whereas Knol gives an incentive for authors to work up a reputation in the community. It remains to be seen whether Knol can compete with Wikipedia on the quality of content. My prediction: unlikely, but no matter: Knol will have a different angle, and that’s fine. Citizendium, on the other hand, is like doing homework without the goal of a diploma in the end.

The problem isn’t redundancy. Having started a thriving danish wiki on comics, I’ve learned that the most common question from the community is: why not just use Wikipedia? It’s a good question that stems from the idea that Wikipedia is the be-all end-all wiki. The question is usually settled by comparing similar articles on different wikis; go on, read about Star Wars on Wikipedia, Wiktionary and Wookieepedia. You’ll see three different articles that have each their use.

Wikis are here to stay, even idiot wikis like Conservapedia (I’ll defend to the death your right to say it, and so on). Wikis are likely to change the world, in fact, and I’m not referring to comic-book databases alone. I’m referring to the fact that knowledge does to ignorance and injustice what water does to fire. Puts it out. Righteously.

There’s plenty of room for improvement in the wikisphere. Currently, articles are written in plain text and special wiki syntax and inter-wiki links are manually created. Most of this should be automated, and editing should be rich text, at least optionally. The whole reading, editing and discussing content experience should be overhauled to fit the true meaning of the word wiki: fast. Coming in 2008? I’d like that.

4 thoughts on “Wikiscape”

  1. Larry Sanger says:

    I’m afraid your post suffers from a lack of proper perspective.

    First, the Citizendium — started with a budget of $0 just over one year ago — has over 4,600 articles under development and hundreds of people contributing each month. At our one year mark, even including a six-month private pilot project, we had slightly more words than Wikipedia did after its first year (five million). We have accelerated our growth rate significantly and are poised to accelerate even more in 2008, for a whole variety of reasons. Of course, the rate at which we approve articles is something we intend to work on.

    As to Knol’s chances and its comparison to us — well, I’ve discussed that at length in various posts for the Citizendium blog. And there was a very interesting discussion of that question recently in Ars Technica.

  2. Joen says:

    Hello Larry, and thanks so much for taking time to respond here.

    While my language in this article is colorful and my notes on Citizendium are unjustly short, I’d like to point out that I do respect your work. I know that you’ve helped start Wikipedia and I know that your intentions with CZ are noble, so here’s my respect for that.

    I am aware that you’ve discussed these things at length before, I also read the column on Ars Technica you mention. So I’ll try my best to keep my comments short, but I think you deserve some elaboration.

    There are a number of reasons why I currently do not believe in the Citizendium project. First and most importantly, I believe CZs biggest problem to be its staunch head-to-head competition with Wikipedia. CZ uses the same basic Monobook skin that Wikipedia uses, this makes it very hard for the layman user to see CZs distinguishing qualities. Since most of the content is also directly overlapping (sometimes even a GNU copy) and in both cases encyclopedic, it’s hard to see why one should use CZ over Wikipedia when the latter has far more, and generally well-written content.

    Secondly, both Google and Wikipedia continuously enjoy huge amounts of positive press, while CZ has gotten off to a really bad start. Citizendium, I think, is currently labeled as “those who would usurp Wikipedia”. Had things been handled differently, one might compare Wikipedia and Citizendium the way one would compare two traditional print-based encyclopedias: on quality of content, not survivability.

    That’s not to say that the Citizendium can’t be successful in its own right. Running a wiki myself, I know a tad more about wikis than the regular Wikipedia critic. I know that while gathering a wiki-community is an uphill battle, it is also a one-way battle. (Nearly) every edit and every contribution adds to the collective information pool driving the site in the right way. Content will show up on Google, and with each search, the best article will win. So from that point of view, CZ will only be doomed the day you stop paying the server bills.

    To thrive, however, Citizendium needs to be positioned differently than it is now. My approach to this is from a usability and design perspective, which matters more than one might think. Citizendium needs to be an alternative to Wikipedia. It needs a well designed logo and a great design that isn’t Monobook. It needs to stand out with an air of knowledge that will appeal to teachers and professors.

    As for Knol, I think it’ll succeed simply because it’s Google. They’ll design it well, they’ll support it by integrating into their services; Gmail, Calendar and whatnot. Incidentally, I think Knol will also be positioned as an alternative to Wikipedia, but this is really all moot since we haven’t seen much yet. It could very well turn out that Knol is simply Google Page creator, rebranded and with inter-wiki linking.

  3. David Gerard says:

    Currently, the problem with rich text editing in MediaWiki (which is the software Wikipedia and the other wikis you name run on) is that there is no defined syntax for wikitext! Really – it’s defined as “whatever the parser does” (processing the wikitext into HTML with a series of regular expressions). This is, obviously, way less than ideal. There’s work going on into reverse-engineering a proper grammar for it, so that third-party implementations are at least possible. This will allow high-quality WYSIWIG editing, non-MediaWiki wikitext converters and all manner of nice things. The status of the current effort is “promising vapourware”. See http://www.mediawiki.org/wiki/Markup_spec and the current effort, http://www.mediawiki.org/wiki/Markup_spec/ANTLR . There’s also wikitext-l, a mailing list for working out the weird corner cases in wikitext … and so far wikitext appears to be made of corner cases. Every quirk you can think of is actually used in practice. Gah.

Comments are closed.