This post originated from an RSS feed registered with Java Buzz
by Fred Grott.
Original Post: Why Twitter Track Will Never Work
Feed Title: ShareMe Technologies LLC-The Mobile Future
Feed URL: http://www.jroller.com/shareme/feed/entries/rss
Feed Description: A Weblog about Java programming and digital convergence on mobile devices in such areas as P2P and collaborative technology.
Its kind of simple why Twitter Track was not working. Imagine if you will for a moment a database. You have regular queries, and than
you have dynamic queries getting composed cached and maybe even operated on via stored procedures. That is what is at Twitter's core.
You as a person are node level one. You are finite set of nodes. Now, pretend for a moment that you have 20,000 followers, which is level two of nodes and is closer to becoming not finite but infinite. Now send track through these two level of nodes and guess what it suddenly become sat least to the database as infinite, databases are not good at solving infinite problems people.
The infrastructures a that are good at scaling to close to infinity are networks not databases. Specifically federated networks. Could Twitter off load enough through routing 3rd party apps through Jabber Gateway to bring it back? Not really as it would be the same database loading levels.
Sorry folks I wsh I had good news about Twitter Track function coming back, I do not.
Previous Topic
