Re:Sorry, Lisp. I tried.
by brlewis on Monday December 03, 2007 @09:56AM
(#85604)
Sounds like CL wasn't right for his particular requirements. He would have been better off with a Scheme implementation on the JVM -- Kawa or SISC.
[ Parent
| on HN ]
Maybe Lush? http://lush.sourceforge.net/lush-manual/f0288067.html
Rather than a Lisp that runs on the JVM, I'd be very interested in knowing a Lisp that somehow runs as Python bytecode. Does such a monster exist?
Re:Sorry, Lisp. I tried.
by danohuiginn on Monday December 03, 2007 @11:20AM
(#85624)
Googling turns up a few lisp-in-python implementations: <a href="http://www.biostat.wisc.edu/~annis/creations/PyLisp/">Pylisp</a>, <a href="http://www.xs4all.nl/~yduppen/site/psyche.html">psyche</a> and <a href="http://hkn.eecs.berkeley.edu/~dyoo/python/pyscheme/">pyscheme</a> (the last two being scheme, obviously). Haven't tried using them; maybe I will the next time I feel like frittering away a couple of hours.
Re:Sorry, Lisp. I tried.
by danohuiginn on Monday December 03, 2007 @11:20AM
(#85625)
clearly html links are the wrong way to go around here ;)
Thanks a lot! I had googled for that, but couldn't find anything interesting.
Lush is great for numerical prototyping, much better than MATLAB/Octave. Its matrix/tensor engine is one of the most awesome (as in being in awe) pieces of software I've ever come across. It is definitely mature and well-tested in that niche. It was created by leading neural network researchers.
Outside of its niche, it's pretty useless.
Re:Sorry, Lisp. I tried.
by viergroupie on Tuesday December 04, 2007 @04:11AM
(#85876)
I don't quite get the faith Lush seems to inspire in some of its users. It's terribly documented, has antiquated semantics (no closures), and isn't even type safe (you can segfault if you pass an object of the wrong type). I might consider using Lush when performance was absolutely critical, but otherwise I don't think it holds a candle to Matlab for prototyping.
Why? So you have access to the Python libraries from Lisp? That would probably be better accomplished by writing a layer that used the Python/C API to fbnd the functions into Lisp, rather than compiling the whole program to run on the (slow) python interpreter.
No, I just was thinking in some Lisp implementation that didn't require to install the JVM, but that could use the bytecode of a lighter language, like Python. Also because I suppose if a Lisp implementation is generating code for a specific platform, its documentation will refer to such platform, and I'd prefer to read about Python bytecode than about JVM.
But maybe I'm wrong and it's just useless prejudice of my part.
Re:Sorry, Lisp. I tried.
by dpapathanasiou on Monday December 03, 2007 @10:07AM
(#85608)
I was thinking the same thing -- there's also a Lisp for the JVM called Clojure (http://clojure.sourceforge.net/) which he might have found interesting.
I've been looking at that one, as well as Armed Bear Common Lisp. Shortly after writing that article I started reading up a lot on Erlang which I've come to like and now have some projects in it (notably wrotit.com, where the submitter probably found this article, since that's the domain on which this was submitted to reddit a few hours ago). I didn't discover ABCL or Closure until after I started using Erlang, so both are a bit of a back-burner educational thing at this point.