TalkChess.com

I apologize in advance for the length of my post.

I want to thank everybody who has contributed to all the GPL discussion in the last weeks. i think the dialogue is important for all, I found them enlightening…i've learned a lot.

What I learned, above all, is that there exists a wide disparity in exactly what is meant by the GPL, and that there are many many diverse ideas, views, perspectives, and interpretations.

Some people interpret it in a rather strict sense, others much more loosely. I can’t say where i (personally) stand...i need to think a lot more about it and carefully consider the many viewpoints.

Fabien Letouzey doesn't seem to be concerned at all, and seems to have an
extremely liberal viewpoint, as Grant kindly pointed out in this thread: http://64.68.157.89/forum/viewtopic.php?t=23361

and as Fabien indicates here after seeing the Stelka 2.0 source code:

“No worries as far as I am concerned.
Ideas are not a legal property.
The code was rewritten so it's OK with me.
Tournament organizers might think differently.”

It appears here he recognizes fruit 2.1 in Stelka (“The code was rewritten”), but from what I see, the code is not re-written at all!:
http://64.68.157.89/forum/viewtopic.php ... ka+++fruit

Maybe he recognizes things I didn’t even notice, as I was searching for exact or near-exact matches!

Jury Osipov also has a very liberal viewpoint apparently, he emailed Christophe and wrote:
- Obviously Vas has borrowed many things from Fruit.
- UCI parser does not count.
- Borrowing is a normal practice in chess programming.

Chris W., Uri Blass, and many others make a distinction between certain parts of the program:
the “UCI/user interface/parameter passing/non AI parts …” and “engine parts”.

i.e. certain parts of the code are critical, or more important, than others…and scrutiny should be restricted to these areas

Most of Uri’s posts appear to be a sincere attempt to obtain some ‘clarity’ about what is ok and what is not.

Bob H., Christophe, Zach, and many others clearly take a more rigorous view, asserting than any ‘block’ or chunk of any code is relevant…i.e. chunks of identical/equivalent source code is very concerning, and lots of them almost certainly constitute a violation.

Imagine the young promising programming student who decides, yeah, I’d like to try my hand at a chess engine. He looks around and sees this great wealth of GPL/Open-source chess engine code. But clarity of proper usage is missing, guidelines are absent or difficult to discern, there is but one document for him– the GPL license (over which there is so much disagreement).

It’s really important that the chess engine community establish some clear-cut guidelines. It’s especially important for new, young and upcoming programmers. Make it easy for them work within ‘the framework’ of what is acceptable.

One thing is for sure…i’m more confused than ever.

Norm

kranium wrote:One thing is for sure…i’m more confused than ever.

So all my attempts to clarify have been for naught, eh?

Seriously, I learned a great deal about copyright, GPL, and that there are several common preconceived ideas about it all.

Ok, I know where I stand:
it should be strictly interpreted and strictly enforced.

Unfortunately, lawsuits are risky and expensive, and the FSF is unlikey to go diving in head 1st concerning every alleged breach. In addition the manpower and resources required alone would be prohibitive.

So unfortunatelty, there is only the court of public opinion left, and the chess engine community is left to police itself...

clear rules and guidelines are glaringly absent and desparately needed!

kranium wrote:Ok, I know where I stand:
it should be strictly interpreted and strictly enforced.

Unfortunately, lawsuits are risky and expensive, and the FSF is unlikey to go diving in head 1st concerning every alleged breach. In addition the manpower and resources required alone would be prohibitive.

So unfortunatelty, there is only the court of public opinion left, and the chess engine community is left to police itself...

clear rules and guidelines are glaringly absent and desparately needed!

Just continue as normal. Use ideas, read books, contribute to forums, study source, study other engines, write your own code. That's what programmers do.

There is no problem in the way engine writers work. The problem has been the wild and venomous attacks carried out with dubious motive and even by those guilty of exactly the same thing as they accuse only a few weeks ago.

The chess engine development process is just fine. It produces 3000 ELO plus programs, ideas are widely distributed and discussed, there's lots of learning and lots of scope for more.

chrisw wrote:... The problem has been the wild and venomous attacks carried out with dubious motive and even by those guilty of exactly the same thing as they accuse only a few weeks ago.

Please expand on this. Which venomous attacks? Which dubious motives? Who is also guilty of exactly the same thing? Who is accusing?

I haven't seen any of it. Or if I have, it's been buried sufficiently that I missed it.

chrisw wrote:

kranium wrote:Ok, I know where I stand:
it should be strictly interpreted and strictly enforced.

Unfortunately, lawsuits are risky and expensive, and the FSF is unlikey to go diving in head 1st concerning every alleged breach. In addition the manpower and resources required alone would be prohibitive.

So unfortunatelty, there is only the court of public opinion left, and the chess engine community is left to police itself...

clear rules and guidelines are glaringly absent and desparately needed!

Just continue as normal. Use ideas, read books, contribute to forums, study source, study other engines, write your own code. That's what programmers do.

There is no problem in the way engine writers work. The problem has been the wild and venomous attacks carried out with dubious motive and even by those guilty of exactly the same thing as they accuse only a few weeks ago.

The chess engine development process is just fine. It produces 3000 ELO plus programs, ideas are widely distributed and discussed, there's lots of learning and lots of scope for more.

this thread is simply asking for clear GPL guidelines, it was not meant to be a part of the flame wars. that's why i created a new topic.
i think my original post is quite reasonable and respectful of everyones opinions.

bnemias wrote:

chrisw wrote:... The problem has been the wild and venomous attacks carried out with dubious motive and even by those guilty of exactly the same thing as they accuse only a few weeks ago.

Please expand on this. Which venomous attacks? Which dubious motives? Who is also guilty of exactly the same thing? Who is accusing?

I haven't seen any of it. Or if I have, it's been buried sufficiently that I missed it.

Bob, he' referring to me. search the forum for xyclops if you want to catch up.

Chris-

Do you disagree that guidelines would be a benefit?

As far as I can see, it can't hurt. Everyone would know exactly what is acceptable, and not. The previous GPL discussions, which have been substantial, during the last weeks could have been avoided. I don't see how providing this information can be anything but a good thing.

and of course my motive is influenced by my experience, but it's far from dubious...you'd feel the same i'm sure. a GPL should apply to everyone.