Very Quotable Lisp Bigotry
Over the weekend I listened to a recording of a talk by Paul Graham about his new dialect of Lisp, called ARC. The talk proved to be very quotable, although the truth can hurt sometimes.
“I read the first white paper about Java and Gosling actually says on the first page: This language is designed for like mainstream programmers right. Like y’know, like it’s some kinda euphemism.”
“If I’m writing a language for myself I can always hack it if it doesn’t do something that I want it to be able to do, right. Whereas if someone else is writing a language for me it better be totally reconfigurable ‘cos I can’t go and hack it necessarily to change it. It better be totally re-definable.”
”…You gotta be pretty humble designing this, you gotta design something that’s gonna be used by people who like intend to do things with it you never planned right ‘cos they have ideas you never even imagined right so you better make something like Lego that they can stick together in all kinds of ways and not get in their way…”
” the great thing about brevity is that unlike abstraction it can be measured.”
“I read the perl book to try and learn about this [brevity] I’m damned impressed with perl. I mean that guy is like Cesanne. He’s a total madman - violates all the rules and there’s definitely things you can learn from this.”
This is the funniest thing I’ve heard about perl since I read Erik Naggum’s tirade.
”…I think that paper of McCarthy’s in 1960 is just the most brilliant thing. I mean when you read that paper you realise he didn’t invent Lisp, he discovered it. I mean this is like Euclid’s elements here…”
There’s no wonder that mainstream development languages and tools are so big and weak: if “programming languages designed to empower smart people” really constitutes a niche market, it’s no surprise that Lisp, Smalltalk and Prolog aren’t well known in commercial circles.
The talk is available at http://technetcast.ddj.com/tnc_play_stream.html?stream_id=645