News for hackers. Stuff that matters.
Re:Language Adoption and Lisps by Hexstream on Wednesday April 02, 2008 @10:59PM (#153593)
"I don’t mention Common Lisp here because, again I stress personally, I hope we stop encouraging it’s active use."

Keep hoping.

"I consider [Common Lisp] [...] over-engineered for almost any other scenario."

Pretty broad claim, in what way is CL over-engineered and for what particular scenarios? And how does that justify discouraging its use?

Sounds like someone is looking for excuses not to get busy learning a language in depth... Seriously, if you don't feel like learning a language, fine. But don't criticize that which you only understand the basics of.

Re:Language Adoption and Lisps by redline6561 on Wednesday April 02, 2008 @11:07PM (#153599)
Hi. I acknowledge that this wasn't well defended. Honestly, I didn't know many people even read my site and I certainly didn't post the article expecting it to make it's way to Y-Combinator. It was a personal draft that I was going to touch up tomorrow but I have a 5am flight so I'm going to sleep for now. Suffice it so say I plan to learn Common Lisp and tried to stress in the first paragraph or so that these are personal inexperienced opinions. I'll try to give it a fair treatment later on when I'm able. Call this an early impression.
Re:Language Adoption and Lisps by jcl on Thursday April 03, 2008 @12:34PM (#153957)
I was curious about that comment as well... I really like the cleanliness of Scheme, but I also find myself attracted to Lisp's ability to selectively optimize a program, for example by declaring the type of variables. To my knowledge this ability is not included in any Scheme implementation. Does such a Scheme actually exist?
Re:Language Adoption and Lisps by redline6561 on Thursday April 03, 2008 @12:43PM (#153973)
Apologies for the site being down at the moment. I'd reset the server but I'm in Chicago. I know that PLT Scheme recently had a system for optional types implemented and there was a paper posted on Lambda-the-ultimate.org. I think it's in SVN. http://lambda-the-ultimate.org/node/2622

Apparently, Bigloo has supported this for a while: http://www-sop.inria.fr/mimosa/fp/Bigloo/doc/bigloo-22.html

One should also consider that there are implementations like Gambit-C and Chicken that compile to type-annotated languages, not that it's necessarily what you had in mind. Just food for thought.