Can we talk?

I recently discovered that Ray Ozzie is experimenting with blogging. In case you haven’t heard of him, he was the creator and developer of Lotus Notes. He founded Groove networks in 1997 to take groupware in a new, more secure, and decentralized direction. Because of his years of experience, he’s got terrific insights into how users behave in collaborative environments, particularly with regard to security.

He predicts: “If we continue on the current trajectory, e-mail will become the place where you receive stuff from people you don’t know, and Groove and other collaborative environments will be where you work with people you know.” This is already true for me. I already do a great deal of productive work interaction through instant messaging and IRC. E-mail has become a tool for archiving information, exchanging less-pressing thoughts, and spam.

The architecture of our collaborative environment matters a great deal in our productivity and the quality of the conversation, he says. Blogs improve the signal-to-noise ratio by creating distributed conversation threads that naturally omit the spammers and flamers because nobody links to them. If you have a blog, you can participate in the conversation. The conversation can be guided as blogs link between each other. Civilized public discourse can return: blogs allow everyone to have the power of their own press.

Speaking of blog architecture, a number of folks are working on the BlogMD Initiative. The name made me think of medical blogs, but in actuality they are talking about ways to improve the metadata (MD, you see) exposed about blogs. There are other similiar projects: BlogChalking wants bloggers to add geographic and demographic information to their blogs. They’re off to a good start with thousands of people adding blogchalk meta tags, but the data isn’t completely reliable due to formatting issues (Some people use a postal abbreviation, such as TX, instead of the full state name, Texas). BlogMD seems to be focused on data that is typically available on blog web pages, such as last update time and URL. Having a way to access this consistently and programmatically for all blogs would be helpful. And I’m sure there’s other metadata that would be useful.

I’ve got to run to vespers at church, but I also want to mention there is a privacy concern with some metadata, particularly the demographic data available when blogchalking. Many times we read stories in the newspaper and have no idea about the demographics of the author or editor. This hasn’t particularly harmed newspapers. How much metadata should be available?