From: Sean Middleditch Date: 2002-02-23T13:36:12+09:00 Subject: Re: Ruby's Future I've only got two of the three, only one of which is still functional (other left off in the middle of extensive hacking, and doesn't come close to compiling, and I have no intention of attempting to fix or recent the changes). An older version of that *should* have been posted (a link, anyways) to this list. My last attempt I'll post on-line. It has a number of draw-backs, mostly from its use of recursive function calls to execute the byte code - this made it not only rather slow, but impossible to freeze the execution. It also doesn't have OO at all, nor a usable standard library. OO would be somewhat straight forward to add. Things like hashes and so on that aren't a language necessity (i.e. can be easily coded in the language itself) aren't part of the language - that is what made the language so small and fast. It's like RISC vs. CISC CPU's. ~,^ The URL for that (last updated in November) http://www.awesomeplay.com/personal/elanthis/scripts-0.0-20011108.tar.gz Looking thru it, you may see an idea or two taken from Ruby. ~,^ I'm starting up a *new* language (tentatively called mudparse, for use with AweMUD and other MUD engines - anyone interested in helping me get it right is free to contact me about it) that is designed *solely* for use in games and MUD's, which means I'm not going to try to build general programming features into it (which is what killed my last attempts). That freezing issue is actually the one that I've yet to actually complete. I keep trying to avoid memory usage, and trying to use short-cuts that make the freezing impossible. My next attempt will worry about functionality first, then optimizing *last*. Maybe I'll actually succeed then. ^,^ You might also take a look at some of the embeddable LISP interpreters out there. I found a couple (no, I don't recall their names) that were freezable/threadable from a controlling process, but like I said - I'm not forcing LISP on anyone. That is solely a science/geek language. ~,^ On Fri, 2002-02-22 at 21:01, Yukihiro Matsumoto wrote: > Hi, > > In message "Re: Ruby's Future" > on 02/02/23, Sean Middleditch writes: > > |I've implemented some 3 mini-languages in the last 4 months or so, each > |with a majority of the above features. Each of them easily beat Ruby in > |terms of speed, load time, resource usage, etc. > > Could you show me your mini-lang API if possible? They would help me. > > matz.