This post originated from an RSS feed registered with Python Buzz
by Phillip Pearson.
Original Post: News Aggregator UI design
Feed Title: Second p0st
Feed URL: http://www.myelin.co.nz/post/rss.xml
Feed Description: Tech notes and web hackery from the guy that brought you bzero, Python Community Server, the Blogging Ecosystem and the Internet Topic Exchange
One of my "background" tasks at work is to write a news aggregator. I hacked a basic Bloglines-like aggregator together a few months back, and haven't touched it much since then. I had a good brainstorming / design session over a beer with Richard MacManus when I was in Wellington at the end of August, though, and I finally got around to adding in some of those ideas a few days ago.
We were talking about automatically ordering subscriptions based on some sort of "interest" metric. Most useful at the top, and so on. As a result I'm now tracking the following bits of my browsing behaviour:
- how many times I have clicked on each feed
- average number of new items in each feed that accumulate before I click on it
- average time between a new item appearing in a feed and me clicking on it
The last one seems most useful at the moment - it results in feeds which I read *often* ending up at the top. The only problem is tuning it so that a change in my behaviour is reflected relatively quickly in the aggregator display. Some people (e.g. Dave Winer, Marc Canter, Erik Benson, Dilbert) post often enough that they are already right up the top of my list after only a couple of days of tracking my clicks. Others (e.g. Paul Graham, Justin Hall) don't post that often, so I'm not likely to notice their new posts as quickly, and they'll never make it up to the top.
Perhaps I should make it possible to flag a feed so that it will jump up to the top when there are new posts. This would be nice for things like my feed of new comments, and for "interesting but low-traffic" people.
Previous Topic
