the silver bullet

Posted by: stak
Tags:
Posted on: 2006-06-13 15:03:15

i came across this article on OSnews today, and it seemed kind of interesting. the author claims that brooks' "no silver bullet" argument only applies to algorithmic programming, and proposes that we write programs in a different manner. from what i understood (and i could be totally wrong here), what he's proposing sounds a lot like VHDL, where outputs are a function of the inputs with no side effects, and everything operates concurrently (for the most part, anyway).

it sounded kind of interesting, and i wanted to procrastinate studying for my networks midterm tomorrow, so i hacked together a little framework to try it out, along with a sample "echo" program. the app echoes stdin to stdout (but only if stdin changes - if you type the same line twice, it doesn't re-echo) and ends when you type "exit". basically, the program is going to be a bunch of variables encapsulated in their own class, each with an update() method. it can register dependencies on other variables, so that when those variables get updated, the dependent variable also gets it's update() method called.

it seems kind of interesting, although i have my doubts as to whether you could really write large applications with this model easily. of course, i'm writing it in java, which doesn't help. a new syntax/language would probably simplify things a lot more, but i don't really feel like writing a parser for it. i'll probably try to write some more complicated apps later, possibly something with a GUI to make things interesting..

Name:
Comment:	Allowed expansions in comments/replies: [i]italic[/i], [u]underline[/u], [b]bold[/b], [code]code[/code], [sub]subscript[/sub], [sup]superscript[/sup], [url=http://some.url]linked text[/url]
Human verification:	=

[ Add a new comment ]