Blogging what I know so far

I’ve worked out that I was doing stuff wrong with Github and in fact I can pull 5000 requests an hour from their API, so I’m reviving that code I wrote and doing some cool processing of the streams on there.

As part of this, I’m documenting what I know so far and in an abstract form have the following posts (due to pop in a few weeks)

Mon Mar 18 2013 11:00:00 GMT+0000 (GMT):
Fri Mar 15 2013 11:00:00 GMT+0000 (GMT):
Thu Mar 14 2013 11:00:00 GMT+0000 (GMT):
Wed Mar 13 2013 11:00:00 GMT+0000 (GMT):
Tue Mar 12 2013 11:00:00 GMT+0000 (GMT):

Proof-reading would be met with much happiness, I'll be writing a few more shortly on a less abstract topic (Github streams) and doing some cool demo off the top of it.

Wed Mar 20 2013 11:00:00 GMT+0000 (GMT):—emitting-commits-as-their-own-events.html

Thu Mar 21 2013 11:00:00 GMT+0000 (GMT): [](
And another

And another, this time doing a basic temporal query—temporal-queries,-who-doesnt-trust-their-hardware.html

If anybody has any ideas of interesting queries to run, do let me know - I intend on making an infograph type of thing out of this—crossing-the-streams-to-gain-real-insights.html

Crossing the streams

Note: I’ll actually run the scripts for these up on my server for the next couple of weeks so the graphs more accurately represent reality

Temporal averages

Next up, some basic sentiment analysis and then I'll build a super cool hacker-news friendly "sentiment analysis over time" page

Basic sentiment analysis of different languages—sentiment-analysis-of-github-commits.html