Day 40 - Tweets analyser, Git analyser idea

I started reading Emma (Austen) today! The comforts of a story being linearly narrated are manifold, I missed that! (Both my previous books were confused narrations which switched here and there and totally left me in reading-epilepsy, I liked that too, it’s good to be back nonetheless) This story is also written in late-18th century English with the whole flourish and the Keira Knightley and Rosamund Pike-esque characters. Emma Woodhouse’ introduction was solid, Mr. Knightley just entered the scene!

(That applies to Tuesdays this semester)

Meanwhile, fun fact, most of my previous posts have been really late, about 30 minutes before midnight or hours after midnight. I love Git for this extent of statistics reporting that’s built into it. For the longest time I have wanted to build some kind of analyser or a gem of some sort which will use a ruby wrapper for the Git binary and show some kind of “cool” statistics about projects. Specifically, I wanted to use it to find out how “active” an open source project was: how many new people contribute each week, how many contributors return and contribute regularly, what kind of times do they work in between. I think the last metric is especially important in finding out if a project is something that the contributors do for work or pursue as personal projects.

This could also be done for Github users on their public repositories to see what type of code they are writing and what time they are writing it in. All of this is great talk, but

Talk is cheap

So, this still remains in the “ideation” phase and never made it into the real world. I found this repository through this article: x0rz/tweets_analyzer. Given a username, it tries to find some stuff about the user, specifically it uses the timezone, the times that the user often tweets in, the users that are most interacted with and retweeted: this probably shows people that the user has turned on notifications for (?) because it’s next to impossible to continuously retweet someone unless you are pinged each time they tweet something. Twitter is an incredibly hard platform to just stay focused on one thing, if you follow a few 100 people and then you scroll down for about 2 minutes, your feed updates with 4-5 tweets and the blue dot next to the “HOME” icon is hard to ignore and just keep browsing. This is how FOMO works, yes? (Note to myself: I should add FOMO to the awesome-social-science list that I maintain)

POST #40 is OVER