Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty question about root move order?

Author: Pham Minh Tri

Date: 21:53:20 03/06/01

Go up one level in this thread


On March 06, 2001 at 18:31:58, Robert Hyatt wrote:

>On March 06, 2001 at 13:27:06, Severi Salminen wrote:
>
>>>What you say is true, but Severi was not referring (I think) to reinvent the
>>>wheel but more to "cut and paste" code and then modify it.
>>>In science you repeat many times a recipe, but you do not use the same
>>>reagents that the original author.
>>>I partially agree with him. That is why all the code in my program has been
>>>written by me. However, I did not reinvent the wheel, I listened to this
>>>group (when I started was r.g.c.c.), I read the little things that are available
>>>on the net and then I implemented them with _my_ code. Was this the most
>>>efficient way? I have no idea... maybe not, but it fulfilled my goals.
>>
>>Yes, this is exactly the point. The reason why I pointed this out was that a few
>>months ago someone tried to get credit with a "modified" TSCP. He had only (as
>>far as I know) altered a few parameters, like piece values, changed the name and
>>started the hype. That is what I do not like at all, and without any sources
>>available this would not have happened. Not that it offended me somehow but I
>>feel bad when someone takes credit from others' work. But this is not about
>>ideas and techniques but about pure code which is copied or only slightly
>>modified.
>
>
>This has happened several times in the case of Crafty as well.  IE Le Petite
>was one well-known copy that claimed to be a brand-new program.  There is
>another one I discovered a month or so ago but the name escapes me right now,
>but someone here was beta testing it and noticed it looked a lot like crafty
>when they fired it up in a dos window.
>
>Of course, this nonsense also happens in science, and is the reason for
>existing patent law.  And there are lots of patent infringement cases going
>on all the time...
>
>
>
>
>>
>>But to be more precise: I don't like the fact that there are _complete_ sources
>>available. It would be better to have partial examples (like the structure of
>>normal search()) available, or a Computer Chess FAQ. This way new programmers
>>would know where to start but they would still use their own brains to actually
>>make things work. This all is about copyright.
>
>I agree to a point.  But for many people, they don't have the time to write a
>whole program.  But they do have ideas to try, and in the case of crafty, a
>few good ideas from others have made it into the program.  Ideas that might not
>have been possible had the person first had to write a complete program before
>he could test the idea...
>
>
>
>>
>>
>>>So, it is good to copy few things...
>>
>>I must admit that I also have peeked to Crafty sources when I programmed SEE. I
>>really had no idea where to start and got plenty of ideas there.
>>
>>>One little point that I disagree with Severi is that "cutting and paste"
>>>does not hurt creativity in this field, because most people that do that
>>>do not have the time to do it in any other way. So, they wouldn't be doing
>>>any chess programming at all.
>>
>>That is partially true, partially not. If there had not been any sources
>>available, some of the people who would have "cutted and pasted" would have
>>instead started from scratch and maybe inented something new. But on the other
>>hand now when we _have_ sources available we have more programmers trying things
>>and get maybe new ideas that way.
>>
>>Severi

Remember that many people and I are learners, not inventors like many other
people here. Actually, I do not see the diffirence between the learning from
textbooks or the learning from open source codes. I think one of the main
purposes of that codes is for training. "Cut and paste" is also a method of
learning. Some people here want and can invent something new, but some others
will be happy if they could understand only some pieces of your codes. Do not
blame with their happiness or make any over expectation about new inventions
(likely someone is requiring all people must be as smart as him to enter this
field).

I do not believe that we could write **all** our code without learning /
modifying from other codes. Is your QuickSort (or AlphaBeta) algorithm totally
diffirent from all others? Of cource not. And you should have the same answers
with many parts of your codes. Certainly, if you are expert or work for longer
time, your proportion of own codes would be much bigger than the new one's.

And note that in science, people could "cut and paste" any thing with enough
referrent and acknowledgment.

I do not agree with somebody here about cutting off all free source codes. This
would make benefit for few people who have commercial software but slow down the
progress of science and reduce the number of potential people, who could
contribute to it. (If there is not any open source codes like Crafty and others,
the popular of this club would be drop dramatically). I like the current balance
of commercial and free software and more people are getting benefit.

In my case, I make an easy plan. I will learn from you all and every sources
with critical mind, do all copyrights (no worries, isnt it?), try some my ideas,
try all in some virgin areas (like chinese chess), first for curious and fun,
then release (freely) them if they are useful. Hope that works :-)



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.