Author: Richard A. Fowell (fowell@netcom.com)
Date: 22:27:19 11/16/97
Go up one level in this thread
A "chess program - program" interface protocol/utility is certainly
something that would, to me, rank up there with such things as the
ICS protocol, PGN, and Auto232. What it needs, I think, in somewhat
backwards order, are:
A) People willing to use the final product
B) A set of requirements for the software to meet (A)
C) People willing to do the work of deriving (B) and writing the S/W.
Some thoughts on this matter:
A) Who would want such a program:
i) Programmers who want their software tested by the SSDF, but
don't run under DOS, hence can't use Auto232.
ii) Programmers who would like their software "server-compatible"
(like Crafty and Ferret(?) are).
iii) Programmers that want autotest capability in their software
(to play against other software)
iv) Users that want to run computer-computer tests
(like the SSDF, and reviewers like KK/me)
v) Users that want to use their software as a "robot" on servers
(a lot of people do that now, manually operated, and I've heard
a couple of them bemoaning the fact that they have to play
interface.)
B) Requirements: Basically, understand what is needed to satisfy the
needs of the people in Group A. Do a cost/performance sort on these,
to see how to meet the most needs for a given level of effort.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.