time to fish or cut bait?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

time to fish or cut bait?

Post by bob » Thu Jun 24, 2010 4:50 pm

Graham said he was _told_ to remove the ip* links. The post from Steve shows this to be completely false, that the "legality determination" was left to the moderators. Graham clearly said "That is not for me to decide, I'll leave that to the experts."

So, what expert has said that ip* and friends are illegal? What did he "leave to the experts" (I assume he expected at least one to support his uninformed decision to not allow links?)??

Talk about contradiction and self-incrimination...

I believe the general feeling is that the ip* case is _very_ unclear. It is _extremely_ clear that it is not a pure copy of Rybka 3, bb's post certainly proves that beyond any reasonable doubt. What, if anything, was copied is unclear. For all we know at the present, the ip* program contains much of fruit. That would explain similarities to the Rybka series (for obvious reasons).

With no evidence supporting "clone" and some interesting information showing "differences" there is more than "reasonable doubt" here. In fact, a preponderance of the evidence also supports the legality of ip* since there is no evidence to the contrary at all, yet.

Nimzovik
Posts: 1831
Joined: Sat Jan 06, 2007 10:08 pm

Re: time to fish or cut bait?

Post by Nimzovik » Thu Jun 24, 2010 5:07 pm

I personally will never castigate anyone whom, rightly or wrongly, is following his conscience. This is what Mr. Graham was doing IMHO. Certainly many educated people asserted that Ipp*** was a clone when it first came out. Especially so since the names of the website of origin were so..... flaky. It certainly is not proven to 100% certainty not to contain pilfered Rybka code. When one speaks of a 'preponderance' of evidence one should support such assertions. It seems reasonable to me to say it is certainly an arguable point judging from all the past discussions. Please understand that I personally am not a programmer by any stretch of the imagination. I rely on those that seem knowledgeble. Even then I admit to being succeptible to deception. I suppose time will tell if the issue is legally pursued. However until then I guess it will all transmute to 'opinion' unfortunately.

Ryan Benitez
Posts: 719
Joined: Thu Mar 09, 2006 12:21 am
Location: Portland Oregon

Re: time to fish or cut bait?

Post by Ryan Benitez » Thu Jun 24, 2010 5:10 pm

Please be clear. Are you making any GPL violation claims against either engine? Ideas used from Fruit are not a violation unless code is taken outside of the GPL. I am sure the team of people that worked on Fruit 2.1 are proud to have impacted other engines in a positive way.

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: time to fish or cut bait?

Post by bob » Thu Jun 24, 2010 6:01 pm

Nimzovik wrote:I personally will never castigate anyone whom, rightly or wrongly, is following his conscience. This is what Mr. Graham was doing IMHO. Certainly many educated people asserted that Ipp*** was a clone when it first came out. Especially so since the names of the website of origin were so..... flaky. It certainly is not proven to 100% certainty not to contain pilfered Rybka code. When one speaks of a 'preponderance' of evidence one should support such assertions. It seems reasonable to me to say it is certainly an arguable point judging from all the past discussions. Please understand that I personally am not a programmer by any stretch of the imagination. I rely on those that seem knowledgeble. Even then I admit to being succeptible to deception. I suppose time will tell if the issue is legally pursued. However until then I guess it will all transmute to 'opinion' unfortunately.
First, only _one_ person asserted that it was a clone, because only _one_ person has the source code for the engine it was supposedly cloned from thru reverse engineering.

As for evidence supporting the "IP* is a clone" there is exactly zero from the author. Larry Kaufman said that a couple of the piece/square tables are _close_ to the values he gave Vas for Rybka 3. That is it.

As far as evidence supporting ip* as a "not a copy of rybka" engine, we have lots of people running test positions thru both engines and reporting a multitude of differences. DIfferent scores. Different depths. Different PVs Different times. We have the bb+ report that is incredibly detailed showing lots of differences between the two, compared thru disassembly of Rybka and comparing it to ip* and friends.

So there is no _clear_ decision here. But there is more showing that at the very least, it is nowhere near just a "reverse-engineered Rybka 3". It is much more than that. Was that its origin? Who knows. There are lots of arguments to make here. Perhaps it was copied from fruit, which would certainly give it similarities to Rybka. Perhaps it was reverse-engineered and then modified significantly.

Seems the only practical thing to do is to error on the side of caution and assume it is OK until proven otherwise. Rather than to assume it is a derivative until that is proven false. The author _could_ step forward with significant proof of that if he wanted to. But he hasn't. One should also consider _that_ as a form of evidence, as in the old expression "the silence is deafening".

Rybka is certainly questionable in terms of legality, based on the Fruit comparison. I've not said "let's declare Rybka illegal and no longer allow it to participate." Maybe it has been modified enough so that no original parts of Fruit remain. Or maybe not. With no concrete evidence, we have erred on the side of being conservative and letting it compete. Seems the same courtesy should be extended to the ip* family, although certainly we should follow past practice and only allow _one_ of them in any event, not the whole inbred family.

Chan Rasjid
Posts: 567
Joined: Thu Mar 09, 2006 3:47 pm
Location: Singapore

Re: time to fish or cut bait?

Post by Chan Rasjid » Thu Jun 24, 2010 6:07 pm

The sponsor is unequivocal that only illegal softwares and links to such be disallowed - nothing mentioned about clones, derivatives, disassembly, etc...

So far illegal software means an illegal "application" - NEVER SOURCE CODES TO BE DOWNLOADED AND COMPILED BY END USERS TO AN ILLEGAL APPLICATION - non-programmers might have a different way of viewing source codes.

Illegal source codes giving rise to an illegal application means the sources of a software was "stolen" and leaked to the public domain. Vasik never ever said anyone stole the source codes of Rybka 3 or that the Ippolit source codes are the same as that of Rybka 3.

So the condition that the Ippolit source codes being illegal can never arise.

Rasjid
Don't believe when you're told "There's no free lunch!" There is Linux.

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: time to fish or cut bait?

Post by bob » Thu Jun 24, 2010 6:08 pm

Ryan Benitez wrote:Please be clear. Are you making any GPL violation claims against either engine? Ideas used from Fruit are not a violation unless code is taken outside of the GPL. I am sure the team of people that worked on Fruit 2.1 are proud to have impacted other engines in a positive way.
I am saying, unequivocally, that code from fruit was copied and used in Rybka 1. Whether or not this code remains in current versions is unknown, but highly probable. ip* is unknown. I have not personally compared it to fruit, but that might be an interesting comparison. Although I am not sure it would be convincing since if you copy from the original, or a copy of the original, you could end up with the same thing and it would be pretty tough to decide whether the thing is a copy of the original or the copy.

Nimzovik
Posts: 1831
Joined: Sat Jan 06, 2007 10:08 pm

Re: time to fish or cut bait?

Post by Nimzovik » Thu Jun 24, 2010 7:53 pm

bob wrote:
Nimzovik wrote:I personally will never castigate anyone whom, rightly or wrongly, is following his conscience. This is what Mr. Graham was doing IMHO. Certainly many educated people asserted that Ipp*** was a clone when it first came out. Especially so since the names of the website of origin were so..... flaky. It certainly is not proven to 100% certainty not to contain pilfered Rybka code. When one speaks of a 'preponderance' of evidence one should support such assertions. It seems reasonable to me to say it is certainly an arguable point judging from all the past discussions. Please understand that I personally am not a programmer by any stretch of the imagination. I rely on those that seem knowledgeble. Even then I admit to being succeptible to deception. I suppose time will tell if the issue is legally pursued. However until then I guess it will all transmute to 'opinion' unfortunately.
First, only _one_ person asserted that it was a clone, because only _one_ person has the source code for the engine it was supposedly cloned from thru reverse engineering.

As for evidence supporting the "IP* is a clone" there is exactly zero from the author. Larry Kaufman said that a couple of the piece/square tables are _close_ to the values he gave Vas for Rybka 3. That is it.

As far as evidence supporting ip* as a "not a copy of rybka" engine, we have lots of people running test positions thru both engines and reporting a multitude of differences. DIfferent scores. Different depths. Different PVs Different times. We have the bb+ report that is incredibly detailed showing lots of differences between the two, compared thru disassembly of Rybka and comparing it to ip* and friends.

So there is no _clear_ decision here. But there is more showing that at the very least, it is nowhere near just a "reverse-engineered Rybka 3". It is much more than that. Was that its origin? Who knows. There are lots of arguments to make here. Perhaps it was copied from fruit, which would certainly give it similarities to Rybka. Perhaps it was reverse-engineered and then modified significantly.

Seems the only practical thing to do is to error on the side of caution and assume it is OK until proven otherwise. Rather than to assume it is a derivative until that is proven false. The author _could_ step forward with significant proof of that if he wanted to. But he hasn't. One should also consider _that_ as a form of evidence, as in the old expression "the silence is deafening".


Rybka is certainly questionable in terms of legality, based on the Fruit comparison. I've not said "let's declare Rybka illegal and no longer allow it to participate." Maybe it has been modified enough so that no original parts of Fruit remain. Or maybe not. With no concrete evidence, we have erred on the side of being conservative and letting it compete. Seems the same courtesy should be extended to the ip* family, although certainly we should follow past practice and only allow _one_ of them in any event, not the whole inbred family.
Bold letttering mine Certainly one could suggest that the different node counts etc. could be the result of additional code. I am not trying to detract from the perhaps original code added. It is indeed possible that many of the IPP** family have morphed almost beyond recognition. No debate there. In terms of errring on the side of caution it was not unreasonable to temporarily ban the IPP*** engines for the sake of integrity on the part of Mr. Graham. The issue was handled relatively well considering the nature of the beast. Should Ipp*** be banned currently? I am not qualified to state.

Dann Corbit
Posts: 11237
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: time to fish or cut bait?

Post by Dann Corbit » Thu Jun 24, 2010 8:20 pm

bob wrote:
Ryan Benitez wrote:Please be clear. Are you making any GPL violation claims against either engine? Ideas used from Fruit are not a violation unless code is taken outside of the GPL. I am sure the team of people that worked on Fruit 2.1 are proud to have impacted other engines in a positive way.
I am saying, unequivocally, that code from fruit was copied and used in Rybka 1. Whether or not this code remains in current versions is unknown, but highly probable. ip* is unknown. I have not personally compared it to fruit, but that might be an interesting comparison. Although I am not sure it would be convincing since if you copy from the original, or a copy of the original, you could end up with the same thing and it would be pretty tough to decide whether the thing is a copy of the original or the copy.
Because the souce code for Rybka 1 is not availble, the only thing that is available is Zach's reverse engineering job.

Since Rybka 1 is bitboard and not array based like Fruit, it is literally impossible to simply cut and paste things like eval and search. The only things (therefore) which can be cut and paste operations are simple and trivial utility routines. Whether this has happend or not is again open to debate.

Here is what is literally and undeniably clear:
Vas carefully studied Fruit and then either copied code or wrote his own version. If he copied then he committed a crime. If he wrote his own version then he may have done somthing that chess programmers do not like but which would nonetheless be legal.

I think it is strange to say you know which of the two possibilities actually occurred.

And (let us suppose) that a few utility routines were copied and used directly. Since he did give credit to Fruit in his beta documentation, would this usage be within the category of "fair use"? Again, I do not think it is possible to say.

It seems possible to me that your dislike of the action (CLEARLY he has used Fruit ideas) may be coloring your notion that you can plainly see the path he has taken to get there.

User avatar
Graham Banks
Posts: 34200
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: time to fish or cut bait?

Post by Graham Banks » Thu Jun 24, 2010 8:22 pm

bob wrote:Graham said he was _told_ to remove the ip* links. The post from Steve shows this to be completely false, that the "legality determination" was left to the moderators. Graham clearly said "That is not for me to decide, I'll leave that to the experts."

So, what expert has said that ip* and friends are illegal? What did he "leave to the experts" (I assume he expected at least one to support his uninformed decision to not allow links?)??

Talk about contradiction and self-incrimination...
I have tried to answer this everywhere that you've posted it, so here goes again:

If you could see the contents of the mods forum (which I can't quote because it's private), you would be the one owing me an apology for what you've just said.
I'll point out again that the other mods interpreted what we were told regarding disallowing links the same way as I did.
gbanksnz at gmail.com

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: time to fish or cut bait?

Post by bob » Thu Jun 24, 2010 8:44 pm

Nimzovik wrote:
bob wrote:
Nimzovik wrote:I personally will never castigate anyone whom, rightly or wrongly, is following his conscience. This is what Mr. Graham was doing IMHO. Certainly many educated people asserted that Ipp*** was a clone when it first came out. Especially so since the names of the website of origin were so..... flaky. It certainly is not proven to 100% certainty not to contain pilfered Rybka code. When one speaks of a 'preponderance' of evidence one should support such assertions. It seems reasonable to me to say it is certainly an arguable point judging from all the past discussions. Please understand that I personally am not a programmer by any stretch of the imagination. I rely on those that seem knowledgeble. Even then I admit to being succeptible to deception. I suppose time will tell if the issue is legally pursued. However until then I guess it will all transmute to 'opinion' unfortunately.
First, only _one_ person asserted that it was a clone, because only _one_ person has the source code for the engine it was supposedly cloned from thru reverse engineering.

As for evidence supporting the "IP* is a clone" there is exactly zero from the author. Larry Kaufman said that a couple of the piece/square tables are _close_ to the values he gave Vas for Rybka 3. That is it.

As far as evidence supporting ip* as a "not a copy of rybka" engine, we have lots of people running test positions thru both engines and reporting a multitude of differences. DIfferent scores. Different depths. Different PVs Different times. We have the bb+ report that is incredibly detailed showing lots of differences between the two, compared thru disassembly of Rybka and comparing it to ip* and friends.

So there is no _clear_ decision here. But there is more showing that at the very least, it is nowhere near just a "reverse-engineered Rybka 3". It is much more than that. Was that its origin? Who knows. There are lots of arguments to make here. Perhaps it was copied from fruit, which would certainly give it similarities to Rybka. Perhaps it was reverse-engineered and then modified significantly.

Seems the only practical thing to do is to error on the side of caution and assume it is OK until proven otherwise. Rather than to assume it is a derivative until that is proven false. The author _could_ step forward with significant proof of that if he wanted to. But he hasn't. One should also consider _that_ as a form of evidence, as in the old expression "the silence is deafening".


Rybka is certainly questionable in terms of legality, based on the Fruit comparison. I've not said "let's declare Rybka illegal and no longer allow it to participate." Maybe it has been modified enough so that no original parts of Fruit remain. Or maybe not. With no concrete evidence, we have erred on the side of being conservative and letting it compete. Seems the same courtesy should be extended to the ip* family, although certainly we should follow past practice and only allow _one_ of them in any event, not the whole inbred family.
Bold letttering mine Certainly one could suggest that the different node counts etc. could be the result of additional code. I am not trying to detract from the perhaps original code added. It is indeed possible that many of the IPP** family have morphed almost beyond recognition. No debate there. In terms of errring on the side of caution it was not unreasonable to temporarily ban the IPP*** engines for the sake of integrity on the part of Mr. Graham. The issue was handled relatively well considering the nature of the beast. Should Ipp*** be banned currently? I am not qualified to state.
The problem is the time-frame. This surfaced near the beginning of the _last_ moderator's term (Steve, Dann and myself). We initially prevented links, just as you suggested. But after 3+ months, with _zero_ evidence to justify this, we removed the ban on the links. Graham _instantly_ started removing/deleting/moving threads about ip* within an hour of being elected.

Reinstating the link ban was unjustified. Keeping it going is also unjustified. How long do you wait for proof?

Post Reply