Yesterday I wrote and published my first open source software, the awfully named jspsi-go-example. It is an experiment module for PsiTurk. PsiTurk makes it easy to run psychology experiments on the web using Amazon Mechanical Turk, where over half a million people do little tasks for a spot of coin. Because I don’t live where my university is, and because my project has a limited budget (as far as payout funding goes for projects with this much programming and set up) I plan on using MTurk this module as the basis of my dissertation experiment. So my little experiment module should come in handy. I even made an icon.
Josh’s tutorial has some good starter code. It includes the ability to determine whether or not an answer is correct given some computed criteria, but does not pass that information back to the PsiTurk server or its database. So I wrote that bit.
Then I figured I’d open source it so others wouldn’t have to do the same thing I did. I had only used GitHub as a place to download things from, except when I was writing my masters thesis in LaTeX, where I did use a private repository as a backup system. But I’d never published code. That was very easy to learn, thanks to their great documentation and client-side GUI interface. I didn’t fork Josh’s project, which was probably a faux pas, but oh well; he starred my tweet about this so everything’s okay.
(Technically it’s not really my first; I contributed a kernel patch to Linux in the early 2000s that fixed a strange bug for some weird hardware I had. I had no idea what I was doing but it worked.)