To this end we analysed the RSS feeds of the named news aggregators and compared

1. timeliness – how often is the feed updated with new stories (interpretation: the more often, the better)
2. continuity – how well are topics pursued within a medium, that is do stories pick up on topics previously reported or is the news agenda very fragmented and disconnected (interpretation: less fragmentation is better)

As you might have read rather too often on this blog: the results are yet preliminary but hey, this is an academic work in progress. What we can infer from the results so far is:

• The traditional paradigms still outperform all competitors when considering “World News” (ie. international affairs, etc). They update very frequently (completely new content every 24hrs) and have a broad range of topics on their coherent agenda.
• The distributed paradigms do not yet work for news not concerned with technology but for this focus they perform very well. For example Digg receives about 100 submission per hour in the technology section alone but the community applies very hard filters to these submission so that only 1% eventually make it in the popular technology category (that translates to roughly one new 1 article per hour).
• However, what becomes clear from the analysis is that the news agenda of Digg and Slashdot is rather fragmented and in addition also very narrow ie. most stories come up only once and the few topics they really focus on are covered again and again and again (think Google)
• All aggregators, even the distributed ones such as Digg and Slashdot show very clear update cycles with low or none activity at night and very little during weekends.

From these findings so far it seems likely that distributed paradigms can compete with traditional modes of aggregation provided they have a sufficient user base interested in relevant topics. So far, however, relying only on Digg and Slashdot will give you mostly unconnected technology stories (but a lot of them!) and next-to-nothing about what else is going on in the world.

There are still a number of issues remaining including our selection of news aggregators and to optimize the continuity measure but for those of you who are already interested, you can download the slides.

PS: The project did arise out of a greater research project on the Performance of distributed problem-solving networks that the Oxford Internet Institute did in collaboration with the McKinsey Technology Initiative. You can find the relevant reports here.