I am finally starting to see the mesh of innovative web sites that are allowing data to slosh around freely and expose people to more conversation. I must admit that I'm a little late getting onto the RSS bandwagon, but this is still a vital part of the equation not only for humans but for machines. There are a number of efforts to allow you to use the same profile on different sites, allowing you to connect with friends you may not have even known used these other sites. APIs allow third-party applications and other sites to send and receive messages and multimedia from not just computers, but also mobile devices. And these services will begin to truly become efficient as add the ability to push data (XMPP) in addition to pulling data (HTTP, RSS).
As many have pointed out, the noise level becomes a problem with so much text sloshing around. Duplicates are pretty easy to filter. The solution here is the same as personalized news. You rate items based on their relevance to you, and some algorithm figures out similarities between items you rated highly. There are several factors that could potentially be important to this kind of algorithm: keywords, length, author or originating source, ratings by users with similar tastes. That last one is enormously powerful. In fact, Netflix is running a contest to see who can do it the best.
I don't know if it will ever be possible to have completely synchronized discussions. But there are various hacks for now, such as tracking keywords on Twitter.
No comments:
Post a Comment