Occasionally I get interested in the science of social media. I think: wouldn’t it be interesting if I can analyse this in some grand way. Wouldn’t it be great if I could collect a load of tweets about something, crunch them and then make some world-changing conclusion.
Of course there’s Storify. I’ve not used it until today, but I’ve seen others produce interesting stories of events from them. I made my first one today. 2 minutes of signing up and clicking things created this (a story of posts about the recent London Google Teacher Academy).
What intrigued me is that during the same event, I noticed that ifttt.com broadcast a way of collecting tweets. There is now a recipe for collecting tweets with a certain hashtag and sending them all to a Google Spreadsheet. I’ve done this a few times with the #gtauk tweets and collected the tweets in three separate spreadsheets here:
Of course, the next challenge is to do something with all that information. This is where something like Storify comes in handy – it already has a way for publishing the posts in some interesting ways.
All I could thing of doing was making a Word Cloud of the tweets, which I did on my iPad (for the first spreadsheet) using an App called ‘Word Clouds‘.
For the second spreadsheet, I again took the tweets to word cloud, but this time used Wordle, which is slightly ironic because Wordle uses Java Applets and so doesn’t work on either of my chromebooks, nor my iPad. I increased the irony by posting the Wordle-generated images to the Google Teacher Academy Google+ Community.
I admit, that publishing this information is a word cloud is not the most interesting thing to do with these collected tweets – I’m still trying to think of a more useful or interesting way of crunching this data.
I have now finished this experiment by seeing how many #fail tweets are generated on Twitter in an hour. Here’s the Spreadsheet. I’m a bit disappointed really: there were only 74. I thought there would be more than that.