Lisp

[Gnu/Vince]

I just read the article that this Yahoo! dude wrote. He's like "Lisp is the most powerful", "We wrote everything SO much faster", etc. I just wanted to have input from guys who know Lisp, is that true or is he being just a little-bit over-zealous?

[jemfinch]

It's probably true, but it doesn't necessarily make me want to know lisp.

Jeremy

[Ben Briggs]

wmHardRock:

Could you please post the link? I'd be interested in reading it.

Thanks

[Gnu/Vince]

http://www.paulgraham.com/lib/paulgraham/sec.txt ==> ASCII Version
http://www.paulgraham.com/lib/paulgraham/sec.pdf ==> PDF Version

[BrianDrozd]

It depends on the application. I doubt that they tested it on math-intesive applications or operating systems. But certainly parsers are easier to write in lisp and generally do run faster; that's what the language was designed to do, after all.

[kmj]

I have can't argue with anything he says, but then again, I don't know enough to argue. :)

I do think lisp is cool, though :)
...course, I think alot of languages are cool.

"The Day of the Jackal" was really cool, btw. :D

[jemfinch]

Originally posted by BrianDrozd:
It depends on the application. I doubt that they tested it on math-intesive applications or operating systems.


CMUCL, a Free Software implementation of Common Lisp (CL) has been shown to be faster than C on several numeric benchmarks (sorry, no link.)

And operating systems have already been written in Lisp. Have you heard of a Lisp Machine? Everything on that box is written in Lisp, kernel and all.


But certainly parsers are easier to write in lisp and generally do run faster; that's what the language was designed to do, after all.

Lisp is definitely a general purpose language. If it can be said that it was designed for any one purpose (which it wasn't,) then that purpose would be AI, since Lisp is one of the few languages that considers data and code the same thing (and that's much more pervasive than simply having first-order function objects.)

Jeremy

[kmj]

I do have one problem with that article... the author basically says, "lisp is better because lisp has macros"... (obviously we're not talking about c-macros here)... but he doesn't try to explain macros at all. He just says (paraphrasing)"well, let me tell you that 20-25% of our program was written in macros, which means that 20-25% of our program was doing things that you can't easily do in normal programming." ...that isn't exactly logical, by any strict sense of the word, he never even makes an effort to show that macros are powerful...