Jeremy Hylton : weblog : 2003-10-28

Learning About Weblog Services

Tuesday, October 28, 2003, 10:53 p.m.

A bunch of notes about services for reading weblogs. I want to learn more about these things.

Roland Tanglao's weblog has notes about the aggregators discussion at BloggerCon.

Technorati and Feedster were mentioned as having "contextualization and global view." Sounds like they are some kind of search engine. I need to find out what these services are. I visited the Technorati site once, but left without figuring out what it was for. The design was far from obvious.

It should be straightforward to hook an aggregator up to a search engine like ZCTextIndex. Bummer. It looks like no one ever packaged it for standalone use.

It should also be possible to have more structured queries. I took a quick look at XQuery and XPath, but the syntax turned me off. Erik Meijer has a better idea: provide XML support integrated directly with your favorite programming language.

If I wanted to read a lot of blogs, it seems obvious to hook the feed up to Spambayes so that I can focus more on the things I am interested in. I like reading Fredrik Lundh, but I don't expect I'll ever learn Swedish.

Another interesting idea is to use feeds to drive Plucker. If you've only got whatever will fit on a Palm or PDA, you need to be a little selective. (The little aggregator I'm running currently produces 476KB of HTML, and that's only a dozen or so feeds.)

It might be nice to have a second-order aggregator. The first-order aggregate is just the feeds that you are intentionally reading. The second-order aggregate is everything that they are reading. Either scrape their reading lists from the weblogs or get a narrower view based on what they actually link to in their weblogs.

Nearly forgot this one. Need to find out about Atom.