Google Knol: What is it? Where is it going?

It's early morning and the buzz on Google's Knol is already building fast. I'm not going to rehash what other's are saying, you can go read them for yourself.

Google calls a "knol" a unit of knowledge (this from the people who misspelled "googol"). Google says the goal is to "find a way to help people share their knowledge", and Google Knol is the place where they can do that; as authors, contributors and commenters. Everyone has jumped on this and said it's a Wikipedia competitor, and maybe in the long run that is true, but that ignores an important distinction; Knol is focused on highlighting authors. Google calls this the "key idea", and I think they are absolutely right.

Wikipedia leverages the wisdom of the crowd to build collaborative articles. It relies on multiple authors, many eyes, consensus and majority rule to get accuracy. In a lot of cases that works well. However, it suffers from all the usual problems of a democratic system. Backroom deals can skew the results. Controversial subjects can require special protection, which gives more control to the editors. And majority rule can stifle new ideas or legitimate criticism. Again, those in control of the overall system can exercise a great deal of power that isn't especially visible to the outside world. And of course, sometimes the things which "everybody knows" aren't always correct.

If Wikipedia is a communist democracy (and I mean that in a completely positive sense, you can't truly have the former without the latter), then Google Knol is a meritocracy. The key is something that has been talked about in social networking circles for several years. Knol depends on reputation. The author of the article is prominent. You see everything else they have written. You see what their peers think of them (and who their peers are). You see what commenters have said about them. Knol is blogging with a focus, and attempt to move beyond general topic pundits and bring in the specialists. The author of the article is a known identity which can be tied to other articles in the past and future. (Note that I don't say "person". It could be a group, and of course we don't necessarily have to know the physical identity. The key behind combining reputation and anonymity is the concept of a long term identity. In the ideal system, nobody knows that you're a dog, but they know that you're the same dog.) Knol attempts to ensure accuracy by assuming that a persistent identity (e.g. your Google account) will encourage you to try and maintain a good reputation. Your reputation in turn depends on how much support you can garner from your peers, contributors and commenters.

The usual problems with online reputation systems apply here of course. Online identities can be discarded when they become tarnished. In some cases that's a feature—there are certainly aspects of my past I'd love to discard that easily—but if identities are easy to come by it weakens the power of reputation. That is countered however, by the fact that it takes time to build a reputation, and discarded articles don't drive traffic. More worrisome is the degree to which people can jazz the system by creating multiple identities that work together to build a buzz and the appearance of consensus. But then, Wikipedia has the same weakness. Even real world systems are susceptible to fake groundswells.

In general, I think the idea has a lot of merit, and it's likely to result in a lot of in-depth and well organized articles (the current Knol screen shot shows a very professional looking page—much nicer than your typical wiki). The big question is whether it will gain the breadth that Wikipedia has, and how it will evolve over time? Who maintains articles when the author loses interest (or dies)? People can make contributions and comments, but they aren't directly editing (or will it allow edits, but with publication under control of the author?). I keep coming back to the first Wikipedia edit my daughter made. She was writing an article on the Oregon Trial (a mock tourism brochure, actually) and in the course of her research she discovered the Wikipedia had the length of the trail wrong—so she fixed it. How easy (and immediate) would that process be using Knol? And what happens when over time there are thirty different articles on the same subject? Have we just recreated the web? (Well, at least we know it won't do away with the need for Google's search engine :-).

Like most Google projects, Knol is starting out on an invitation basis, although in this case I suspect invitations will be a bit harder to get than usual. The initial focus will probably be more on quality than quantity. I think the idea of a reputation-based system, and the appeal of an author-centric system, will make it successful, but I don't see it replacing the Wikipedia. If anything, I think merging the two concepts would make more sense. Combining both authored and crowd-sourced systems into a single repository. It seems unlikely that Wikipedia would do anything so drastically different, and starting a "new" Wikipedia would be hard for anyone to do, so unfortunately it's not likely to happen. I guess we'll all have to get used to searching two locations and sending our edits to two different sites.