I finally handed in my thesis on Wednesday (25th May), which was a great relief. It’s quite strange sitting here typing this and not feeling like I should really be working on the thesis. In the end, I was pretty happy with it — the time spent paid off and the final monstrosity was quite readable for a fifty-page document on machine learning and RSS feeds. A big thanks to all the people who helped me during the thesis.

I was bored by work so I took some photos of my guitar instead:

For those who are interested, I’m trying out using Backpack, 37Signals’ new toy, to manage Bluefunk and Feedfunk’s progress in a fairly ad hoc way. You can follow goings on at Bluefunk’s and Feedfunk’s pages respectively.

I watch people using computers, and I have to say that I agree with what Blake Ross has to say. In my mind, what many developers need to think about is making things work. As opposed to making MSN Messenger look a bit crazy.

I think that we need to stop sticking in shining bright new things just because we can and work out whether they are pretty useless, really, and concentrate instead on making what’s there already work flawlessly. Office suites come to mind here. Why not make them not crash, rather than making pretty Document Recovery dialogs? (Or in OpenOffice’s case, rather confusing ones). Thesis writing has concentrated my mind on this one. I should’ve just used Lyx, I know…

Feedfunk is nearly functionally complete — or at least as complete as I want it to be for my project. I’m soon going to start work on bits of Backpack page, and watch as I cross bugs and tasks out.

I am also progressing through my thesis — up to about 35 pages at the last count — which is coming on nicely. I have a problem now because I’ve finished most of the things I consider interesting to write. This means that coding bits on Feedfunk is far more tempting than it should be, given that I still have plenty of thesis to do. I’ll do my best to resist the temptation.