Dave Winer writes about how OPML could be used as the base for creating the next generation of directories. Forst, he explains the rationale:
We accept that there are two main directories (Yahoo and DMOZ), but why do we accept that?
Suppose someone said there were two home pages for the Web. How good could they be?
What if they said that in order to write for one of these home pages you either had to be an employee of a relatively small company, or be part of an open source project. What if the subject you’re an expert in already has an editor in DMOZ? What if you already have a job you like and don’t want to go work for Yahoo? And how many people can Yahoo afford to employ to work on their directory? Wouldn’t it make more sense to open the process up, as the Web is open, and let the cream rise to the top?
First, we need to break the monopolies, to open up the idea of web directories to competition, so there can be as many sub-directories for a category as there are people who have an interest and have expertise.
It’s all just a bootstrap away. People have to start creating and maintaining sub-directories, and then link to each other through their directories, and away we go. This is something that is entirely in the user’s hands. The technology is already fully invented.