(Re)defining the News Aggregator
Reading Dave's in-progress What is a News Aggregator?:
A news aggregator is a piece of software that periodically reads a set of news sources, in one of several XML-based formats, finds the new bits, and displays them in reverse-chronological order on a single page...and the rest leads from there. First thing: I think this is an unfortunately limited definition. It specifies the particular interface by which Radio presents information. And, as a comment on one of my first experments in tweaking AmphetaDesk's interface attests (no author link):
I've been wanting to use an aggregator since I first heard of them (RU7) but have always felt that because poor information architecture/presentation they tended to make tracking a large number of sites harder rather than easier.I didn't expect my first attempt to be called "Brilliant! Lovely! Perfect!", but I did think that I was on to something. Simply displaying the new bits in reverse-chronological order is too limiting - it's only one attempted solution at the problem of aggregation. My solution isn't an ultimate one either.
Aggregators desperately need to grow toward more flexibility and scan-ability. A few things I'd really like to see addressed:
- De-emphasis of seen & older items from sites, but not complete hiding. Context between entries on weblogs is important.
- Optional grouping of items from the same or similar weblogs. Context between entries, and between blogs is important.
- Emphasis of newer items, tracking the time line and signalling attention to changes. Radio does this, but mostly to the exclusion of other concerns.
- Preventing sites with few updates from getting lost in a wash of frequently updating sites. Some of the best sites may update once every few days with something worth reading, but simple reverse-chronological order pushes the quiet sites out in the maelstrom.
So, while Dave's working on defining the News Aggregator, I think it's a good time to redefine it a bit while he's at it.