I went down to Holyoke and had a fit: ie had my new suit fitted. The seamstress had an eastern european accent and was rather fierce: "Don't look down! Hands out of the pockets!" It was over in a matter of moments. I have to go down once more on Friday to pick it up. The gala is on Saturday.
For the past few months, I've been disappointed that I can't make myself take more pictures. So I decided to buy a new, small point-and-shoot camera. I got a Cannon SD780 IS. I've only taken a few pictures, but I'm pleased so far. It's a really small camera and, like all the Canons I've had, it takes really good pictures.
I spent much of Thanksgiving day working on a program to generate "dot" files from the raw data from the virus simulation. "Dot" is the language that graphviz uses to build branching diagrams. It's incredibly simple: you basically just define which nodes link to which, and graphviz does the rest. You can specify lots of other stuff and, in particular, you can align nodes into ranks. my program takes any one infection code, creates a time-line with the dates, indicates which nodes go with which date, and then looks up and defines the relationships. The hardest part was reminding myself how to work with arrays in PHP. I knew what I wanted, but I couldn't figure out how to insert the values into the array. I got it wrong about four times before I finally figured out how to do it: it was simple, of course: too simple. I was looking for a function, but you didn't need to use one: you could just use simple assignment.
The patterns in the data are interesting. It's a bit disappointing how little data we actually got: most people who became "infected" did nothing. But that, in itself, is data. I've learned a lot about how to make the simulation work and, next time, I think I'll be able to make it work better.
- Steven D. Brewer's blog
- Log in to post comments