There’s always a nice thing to do during insomnia: close all the open ideas you’ve got in mind!
So I wanted to have something working about yesterday Twitter Timezone idea, here is the result after using Twitter4R and Sparklines gems:
The sparkline is only created from my friends last 20 twitts. With so few samples, we can’t really see any ‘timezone‘ effect, but some social behaviors appear:
- Rush hours: some people twitts mostly during the evening;
- Exclusives: others have only one or two big peaks, it would mean most of their 20 twitts were written in the same period of the day;
- Flat twitting: twitts seems equally reparted all along the day.
This experience might be extend in the next days, opening it more to public test, and it’s still looking for your suggestions to improve by itself.
One of the interesting part with Twitter is its worldwide use. On my Twitter, I usually get European twitts at the end of the afternoon, and South-American twitts during the night.
Comparing my twitting habits with those of my friends, each one of us could have a twitting timezone, that we’ll be surely different even if you live in the same ‘real’ timezone.
Here is a screenshot of what could be the result:
It could be easily done using the API, unfortunatly, Twitter reduce the API use to one request per minute, so it’ll take one hour to build such a timezone with all your 60 friends… but I’m pretty sure someone will have fun to do something like this soon.