[maemo-community] How Social News ranks news items

[From: Henri Bergius]

Hi, all

We had a chat with Dave a couple of days ago about how the Social News
algorithm works for ranking the most relevant posts from the Maemo
community.

As you know, we're aggregating many blogs to Planet Maemo, but also
have some other aggregations, like maemo.nokia.com news.

The Social News system looks at local votes (thumbs up / down) as a
factor to determine important items. But unlike many think, it is not
the only factor used. There are several, with different weighting:

* Comments the item has on maemo.org (available only for local news items)
* Diggs
* delicious bookmarks
* Links on other blogs (via Technorati)

Like account or app Karma, the social news algorithm is also easy to
tweak to include other factors, so the weighting system is constantly
open for discussion.

Some relevant blog posts:
http://bergie.iki.fi/blog/calculating_news_item_relevance/
http://bergie.iki.fi/blog/maemo_social_news_launched/
http://www.inventive.fi/sosiaalista-suunnittelua-maemoorg-news/ (in Finnish)

Source code:
http://trac.midgard-project.org/browser/branches/ragnaroek/midcom/org.maemo.socialnews

-- 
Henri Bergius
Nemein - Web Craftsmanship
http://nemein.com