[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: computer-go: Is the Go Modem Protocol used in actual tournaments ?
From: Christian Nentwich <c.nentwich@xxxxxxxxxxxxxxxxx>
> kinds of connections many times, and rs-232 is much easier :) Have you ever
> set up an ethernet network for machines that didn't have built-in cards?
I can't help the comment that your world-view (that setting up an
ethernet network is difficult) shows that you have been using the wrong
operating system for too long :)
But then so have most go programmers, I suppose..
Christian
Christian,
I have the same world-view that you do, but I don't think it's a big
issue here.
The issue is standards. Unless we have a really good reason to change
them, we shouldn't because it causes disruption, confusion and
frustration to everyone. We need GMP as a low level interface because
people can be assured that things will always work. That's what I
want, because I don't want to rewrite my own interface everytime the
latest and greatest gizmo comes along.
If something is horribly wrong with GMP, I would have a different point
of view. But I don't see this. Do people feel this is the case?
By the way, I didn't have any problem using the gmp code to interface
my program. And there was some example code that was easy to follow.
I don't think there is any good arguments that it's too hard to do, I
can't imagine designing anything much easier than this.
It would be nice if there was a nice clear document to describe how to
use gmp.c. This would help for those not willing to grovel through
the code and the example. But this isn't a failing of the protocol,
just the documentation. The xboard protocol for chess has a well
written explanation of how to write interface code, such a document
could be modelled after that one.
I'm not endorsing GMP, I'm just saying it's there, it works, it's in
wide use, and you had better give me a really good reason to waste
time on redoing my interface.
Don