Right now I’m in a state of wondering as to quite where my interests lie. While I still read about and am interested in web standards, the push towards them and the inventive things that they allow you to do, writing about them is no longer the fire that it once was. I still find Linux and open-source in general interesting, but I’ve been so busy recently I haven’t really had the time to mess with my system and in the process learn some interesting snippet that’s lead me on to other interesting prospects.
I’m now starting to think about interesting ideas for my university project next year. The project is three or four months long; and it is pretty much all that you do for those months. Therefore I have to come up with something interesting. If I can’t find something that will interest me for three months, I will not do well. This is a fact. I’m terribly bad at doing things that I find fundamentally unexciting and that are not urgent. This would mean that I would do nothing for two and a half months and then have to hack out some code that would suck in a couple of weeks before the hand-in date.
On the other hand, if I find a subject that I can really get my teeth into, I will produce something great. Of that I have no doubt. It’s just finding that project that is interesting and I actually have the ability to do. At the moment I find things like Zeroconf pretty interesting; the idea of a computer you just plug in and have it up and working seems rather good to me. Say a media server: the box is really just a big media store that uses Zeroconf to advertise its services on the network. Other clients, such as a hi-fi, can then be plugged into the network, seek out the media servers on the network and present this to the user.
The general idea could be extended. For example a backup server could advertise this to the network. Any client that wants to back up simply looks for a backup service and backs up. This could be an interesting project as it brings several issues with it. One such issue is that if you have multiple backup servers on the network — and clients just pick one at random that is free — then which server holds a given backup? Perhaps the servers could maintain an index of backups that they share with other backup servers. If each held the index then servers could be plugged and unplugged at will and an admin would be able to query any for the location of a given backup. There are other items that I haven’t really thought about, but to implement a reliable scheme would certainly be interesting.
However, my attentions will probably have shifted by the time projects have to be done. Still, even if I didn’t do it, maybe someone else will have a go. I for one would like the media server idea to see some hard light of day!