22 Jun 2000 ahosey   » (Journeyer)

I went to hear Bill Joy's keynote yesterday. It was very interesting to hear someone of his stature and experience say some things very similar to things I've been thinking. Like it's time to move to a high level language as the de facto standard if we want more reliable powerful software. (At least, higher than C. I like C a lot, but I think it's time for the general populace to move on, leave C in its appropriate niche, much like what has happened with assembler.)

There were some sour grapes in the keynote, like things about Microsoft and the failed Java standards track. True or not, I think those things would have been better left out, because now a lot of people will remember the speech and just remember "There's Bill Joy, being a bitter kook again" and his points about computer technology and its advancement, which were very valid, will tend to be forgotten.

Josh and I were talking about making higher level features more common in programming practice. (This was before the keynote, too.) We think one could take error exceptions, for example, and move them into the OS. So you could have throw() and catch() system calls, for example. This would make it easier for language designers to implement these features, and you could even have them available for free to application coders from C. The idea is to take higher level programming features and move them into lower levels of the system, make them easier for programmers to have available, and hopefully make them more ubiqutious in programming practice. You can't do this with everything (Josh convinced me you can't do a generalized garbage collector in the kernel without substantial performance loss) but you can do it with some. There are some techniques already moving in this direction (like kernel-space threads instead of user-space threads are a perfect example) but no one has tried error exceptions that I know of.

Latest blog entries     Older blog entries

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!