Whee. I was catching up at
http://sigfpe.blogspot.com/ [0] when I realized that I'm finally in some sense starting to be able to read haskell code. Yay! My usual way of learning a language is to read a lot of related concepts and eventually try to sink into a zen of the language; but haskell has felt particularly difficult to get into, because its background is very abstract. A language that draws upon category theory in introductory tutorials is a language that takes some thought.
Now I should try programming some stuff in haskell, to go from a read-only language to a read-write one. One of the downsides, though, is that I don't program enough for pleasure: I'm a lazy person easily contented. Russ may
write his own blog system but I just use livejournal. My immediate custom needs are usually satisfied with simple shell/perl/python scripts.
I really do need to program more, and push stuff out more aggressively. The head maintainer of
color-theme seems to have dropped out of sight again, so I should roll up some pending commits, and see what it takes to get people to start using the new version. Yet another minor patch I had sent in to xemacs seems to have dropped down the memory hole, and the hashtable patches that I wrote 2(!) years ago are never going to go in at this rate.
A large part of the xemacs problem is my own fault, because I don't like having to bug people to get a patch in ... but it's left me feeling a bit disillusioned, particularly when
rants like Steve Yegge's latest are going around...
[0] By the way, the entry on negative probabilities shows an interesting perspective on quantum mechanics.