Spider friends pages for a list of named twitter accounts. e.g.
[login to view URL]
[login to view URL]
...
[login to view URL]
The spider will output a simple Pajek .net format file
where vertices (named accounts) are stored as
<id> "<label>"
and edges are stored as:
<from_id> <to_id> <weight>
(weight is a count of the number of links between vertices -- in this instance weight should ALWAYS be 1)
Here is an example format:
*vertices <# of vertices>
1 "http://twitter.com/briansolis"
2 "http://twitter.com/another"
3 "http://twitter.com/another"
*edges
1 2 1
1 3 1
2 3 1
May not need MySQL back end -- depends on memory handling.
There will be two other closely related projects that follow this one. A satisfactory result will lead to automatic selection for the next project(s)
Spider can run from the command line (I use Mac OS X by preference, and PREFER cross-platform tools)