To those who have ever been in large Lisp programming projects:
I have two questions on lists used in large Lisp applications.
1. How long, do you think, become the lists? (1 2 3 4 ....)
How deep, do you think, become the lists? (((((....)))))
2. Were there frequent cases, you think, lists becomes both
long and deep? or ...
Thankyou for any guesses, feeling-from-experience, or whatever. -Kwang
·····@uicsrd.uiuc.edu
Center for Supercomputing R & D
Univ. of Illinois at Uarbana-Champaign
From: Brian Leverich
Subject: Re: long and deep?
Date:
Message-ID: <2075@randvax.UUCP>
In article <········@uicsrd.csrd.uiuc.edu> ·····@uicsrd.csrd.uiuc.edu writes:
>
> 1. How long, do you think, become the lists? (1 2 3 4 ....)
> How deep, do you think, become the lists? (((((....)))))
> 2. Were there frequent cases, you think, lists becomes both
> long and deep? or ...
>
For the knowledge-based simulation work we're doing, we typically have
thousands to tens of thousands of object attributes that are short shallow
lists. Then we have hundreds to thousands of attributes specifically
containing fact collections or event queues, with these guys being a few
hundred or more top-level elements and nesting typically no more than
three layers deep.
Incidentally, very few of our structures have pointers into their
interiors. KBSim would be a ripe field for the design of tuned garbage
collectors... Cheers. -B
--
"Simulate it in ROSS"
Brian Leverich | U.S. Snail: 1700 Main St.
ARPAnet: ········@rand-unix | Santa Monica, CA 90406
UUCP/usenet: decvax!randvax!leverich | Ma Bell: (213) 393-0411 X7769