tuesday, 20 january 2009

posted at 07:36

I haven't written any code for ages and ages. I'm having a lot of troubled getting motivated, and although I've had a few idea I haven't really come across anything sufficiently challenging or world-changing to really get excited about it. Some ideas I've had that haven't got anywhere, in no particular order:

  • A group chat client (IRC, Jabber MUC, etc) that allows images, videos, etc from the web to be embedded directly in it so that when we share links I don't have to click everything. I started prototyping this in Perl using WebKitGtk, but it doesn't have API that allows a page to be added to incrementally, and I don't really want to get back into hacking on WebKit right now.

  • A version of Settlers of Catan with a really sweet AJAX/DHTML/Web 2.0/buzzwordish interface. Currently the options for playing Settlers on a computer appear to on your local machine with Pioneers or on the web through Java applets. I got as far as thinking about the AI and poking at the Pioneers code a bit before realising that the only interesting bit is too hard for me which means borrowing the AI code and then writing the web interface, which is totally boring.

  • A browser-based sprite engine. THe idea is you'd throw in a bunch of images, specify coordinates and any filters/transformations you might want, and it takes care of their display. Again though, its hacking browser stuff which I'm pretty much over at this point in my career.

  • A web-based forum that doesn't suck. I'm quite over every bit of web software requiring a database and having insanely complicated interfaces. I had a thought of some kind of cross between Digg, the Joel On Software forums (design notes) and an image board. Basically no mad hierarchy of forums and posts, no threads, no userpic/sig/meta clutter on every post, and no login requirement. Internally its very much like blosxom, just a simple one-file script that you can drop into your cgi-bin and it just goes, and that you can extend in any way you like with plugins. The whole thing would be backed by git and just be sweetness. I made a good bit of progress on this but at the end of the day its basically a repeat of the code I've just written to get this blog updated, and its for web forums which I hate and avoid as much as I can anyway.

  • A database diff tool for Lotus Notes databases. This is kind of interesting, and it would help a lot at work and dodge the need for us to buy any more stupidly expensive licenses for a commercial product, but the Notes API really really sucks and I haven't been able to get a working development environment working on Debian yet, which means I'm going to have to use a Redhat VM if I want to make this work. That amount of effort for what is really part of my job in my spare time is just too much.

So there you go. A few reasonably good ideas, but no motivation. As a result I've been killing time playing games, but that's starting to wear a bit thin. I'm not sure what to do right now, because I really feel like I want to write something. Motivation came easily when I was working on jabberd2 - we were changing the world! It was easy for AROS because I'd never worked on an so many different facets of an operating system before and everywhere I turned there was something new to be done. None of the above so far fit into one of those categories. Every existing project I've considered so far doesn't seem to be either groundbreaking enough or have enough for me to learn to make it worth the effort. I'm not sure what to do now.

Its likely I'm growing old and curmudgeonly. I hope I can get interested soon, if for no other reason than my bus trips are getting boring.