[frio]

That's actually something I thought on for a while, but never got the time to implement. Sort of an anti-Reddit; you add your RSS feeds, and then like or dislike various things you read. The software, using various ranking metrics, then begins to recognise the content in the RSS feeds, and displays what you consider more interesting more prominently (so it acts as the ranking algorithm without social input).

I don't think something like that would be hard to implement. Ranking metrics could be simple at first: whether a site is "preferred", whether content is agreed with or disagreed with (the software should display both prominently, otherwise it ends up as a circlejerk), and whether the content is interesting or disinteresting.

However, the big problem I saw with such a setup was that it de-socializes the web; while I get a custom news feed, I don't get the quality comments I generally find on aggregators like HN or Reddit (so long as it's a decent subreddit anyway). So, the next iteration could use a distributed database backend (like CouchDB) to allow comments to be shared cross-node; people share a small web presence on their site, and you can choose to follow their comments on articles, like some sort of distributed Twitter.

The self-defeating part of the exercise that got me was that at that point, you've basically made discovery difficult. Which you'd address with a central hub. At which point, why not just use HN/Reddit anyway?