The Internet is a real information juggernaut, and a testament to just how big it’s becoming is the number of aggregators that are popping up all over the place. RSS feeds and keywords have served us well, however news aggregators have started to use other methods of organisation.
We’ve discovered
Tiinker, a site coded by two graduates of the University of New South Wales’ School of Computer Science and Engineering. It does away with traditional methods and instead provides a newspaper style web page of content suggested by its algorithm based on user habits.
The interesting power of Tiinker is that unlike other sites, the system does not assume that everyone’s internet usage habits are identical. Instead of relying on voting systems, Tiinker’s artificial intelligence algorithm textually analyses all the articles made available to the reader. It still injects random articles for the sake of variety, but most of its results are based on reader feedback and behaviour. Tiinker delivers these results into a custom page of articles with similar and related content from the avalanche of stories it indexes each day.
We spoke to the developers about the Tiinker code, and as it turns out it’s highly modular and easy to add new features to the site. One of the things on their to-do list is to add the ability to flag interesting articles for friends who use Tiinker. Developer and co-founder Oleg Sushkov said: “There is so much information on the internet, that you need all the help you can get sifting through it, and your friends are the perfect candidates since you are likely to share common interests.”
Tiinker has recently completed beta testing and is now
available to the public.
