Wednesday, June 23, 2010

Friday, June 18, 2010

Monday, June 14, 2010

Monday, June 7, 2010

Friday, June 4, 2010

Twitter services

I realized I haven't posted links to some Twitter services mentioned in some scientific articles. Probably they weren't worth to mention. By the way - I've found a very convinient content agreggator.


I'll map positive and negative words and then connect them with elections, cities and SouthAfricaan world championships.

I tried to perform some searches on Facebook, but there is a bug I cannot understand:

(same query but different date and some authorization exception)

Thursday, June 3, 2010

searching parsing doing

I'll use Facebook search (login needed)

It is as easy as fetching URL:

Problem emerge with parsing response:
"data": [
"id": "1522384619_124458707586819",
"from": {
"name": "Piyush Rana",
"id": "1522384619"
"picture": "",
"link": "",
"name": "Wall Photos",
"caption": "crazy vs hot graph for girls",
"icon": "",
"type": "link",
"created_time": "2010-06-03T07:45:09+0000",
"updated_time": "2010-06-03T07:45:09+0000"
"id": "612787180_124298310937744",
"from": {
"name": "Alan Rycroft",
"id": "612787180"
"message": "RT @RWWRSS: ReadWriteWeb: Twitter Exposes Intersections in the Social Graph",
"link": "",
"icon": "",
"attribution": "HootSuite",
"type": "link",
"created_time": "2010-06-03T04:32:03+0000",
"updated_time": "2010-06-03T04:32:03+0000"
"paging": {
"previous": "",
"next": ""

There are some java libraries and RestFB has what I've been looking for. Unfortunately there is one feature I miss very much - Operations without OAuth token (google groups).

I could:
1) parse response with JSON
2) parse response with ANTLR
3) change source code of the library
4) follow Desktop Application Authentication (seem to be the easiest task)