NB – Please note that I wrote most of this post about a week ago but didn’t feel like it was finished for some reason. By rights the date on this ought to be 18th of June.
I’ve found an ajax polling system that takes one closed question, adds it to a database and serves back the results for the poll immediately. Exactly what I need. I had some trouble getting it to work but the basic problems I didn’t figure out myself were solved on this useful ubuntu forum discussion.
I’ve got the basics working (I think). Now I need to fit it in to my framework and make it look right. One thing you’ll notice with this is the fact that the graph is very basic. Aren’t there lots of more sophisticated graphs out there? Yes. There are. I’ve been reading data visualisation blogs for a while now though and have come to the conclusion that you could do a phD in this stuff and it wouldn’t necessarily work that well. I’m excited by data visualisation but it still feels like it’s at a really early stage in its development.
This is graph about MP’s salaries around the world is a good example of this. After a minute or so I start getting stuck on the user interface (some of the labels stick and then repeat and get in the way), it’s all bit ‘mystery meat’ (you don’t know what you’re getting when you mouse over something), I can’t be bothered to tally up the figures in the slide out boxes with what they mean and I generally don’t feel any wiser about politicians’ wealth. A shame. There’s bound to be a fascinating story in here somewhere but I can’t see it. In fact I’m not sure how many really massively successful data visualisation projects I’ve seen.
My interest with the questionnaire is in quite vaguely defined parts of the human mind. It’s about the thought process of answering the questions probably more than the data that comes out of that. I do think there’s a need for agency in seeing results and getting a sense that people are adding to something. Also, people just like being nosy and want to see other people’s answers. Graphs are important. I’m just reluctant to make that much out of them when they’re not at the heart of what I’m trying to do.
The data visualisations I’ve seen which I’ve enjoyed have all made me feel a bit like a kid. They make me feel excited that I can turn this button and then this happens and that there are all of these different shapes and things. That’s great. The problem is, it brings it to the level of a toy. That’s also great, because it’s a really cool toy with some very useful and interesting potential applications. I want to play with it in the future, preferably when I know lots more about php and Processing. For this project, though, I don’t want the questionnaire to be in competition with a toy. I want it to have some level of give and take about it, so people feel that they can put something in to it. I just don’t want that give and take to be via a toy.
You can follow any responses to this entry through the RSS 2.0 feed.