Sunday marked the first time I was able to tell someone who was not a software person about my personal software project (in the most general way). The person in question was a singer in the amateur early music choir that I started singing with recently. Unlike me, she is a pro who sings with the Austin Opera. We had a gathering at a pub after our concert, and she, oddly enough, asked me if I had any other hobbies than singing. I replied that programming was both my job and my hobby. Then she asked what kind of software I write for fun.
I used to not be able to answer such questions, because "explore data computationally" or "tinkering with found data" are not very informative answers. But the project I've been working for the past year has enough specifics that I could summarize it as "trying to identify which visitors to my website are bots versus human visitors". It may even be understandable to non-technical people. So yay. She asked me if I was successful at it. I replied that the criteria for making such a determination are fuzzy. Looking at Apache server logs, you can tell that some log entries were caused by human visitors, by the patterns in the resources they requested. But many log entries look too ambiguous. And ambiguous criteria are hard to encode in a program. So my result was mixed, at best.
So that was it. The moral of the story, I finally have a project for which I actually understand what I'm trying to achieve and even describe it to someone else.
