Robodini Q&A

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

Moderators: hgm, Rebel, chrisw

User avatar
Houdini
Posts: 1471
Joined: Tue Mar 16, 2010 12:00 am

Re: Robodini Q&A

Post by Houdini »

lkaufman wrote: Isn't this an EXACT description of the origin of IPPOLIT, except that it took longer than 3 months? If Richard does use info from Houdini RE in Critter (and I'm not saying that he does), it would be exactly the same as what the IPPOLIT author(s) did, from which Houdini started. I'm not saying I approve of this, but I think you are not in a position to complain about RE. If you want to be fairly safe from RE, don't start with someone else's open source code, write an original program.
You're the guy that has repeatedly pushed Richard to discuss the RE findings of Houdini for the sole purpose of improving your own engine. You are inciting Richard to commit illicit actions for your personal benefit.
You also seem very happy to justify your crime by pointing to an alleged crime committed by the unknown people that created IPPOLIT.
I have nothing to do with all this, and find your attitude simply disgusting.

As you seem to be so much in favor of RE, let me ask you the same question I asked Don higher in this thread.

Suppose someone RE's your next Komodo 6 and then:
- Discusses the results openly on this forum.
- Publishes the source code in a different language (say, Fortran or Pascal) so that it's an obviously non-competitive engine but still shows all the details of your evaluation and search.
- Integrates important search and evaluation elements in his own engine, and then distributes his engine source code to competitors.
- Provides help on the forums to other people that want to RE Komodo.
- Releases a free Komostrelka 1.1 that plays exactly like your commercial Komodo 6.

If you approve of all this, I will hire Jury Osipov to do the job.

Robert
mcostalba
Posts: 2684
Joined: Sat Jun 14, 2008 9:17 pm

Re: Robodini Q&A

Post by mcostalba »

I'd prefer still not comment on the RE of Houdini by Richard because the thing it is really complex and there are a tons of grey areas, and there is a lot of historical background that cannot be ignored.

But one thing on which I made up an opinion I am quite confident is that I'd would like to ask Richard to, now, remove the Robodini binary link because, given that is almost at the same strength of Houdini (although it is not SMP) it could affect Houdini sales and this, I believe, was not the intended target. Moreover the "bet" has been demonstrated won now, so there is no more point in keep it public.
kinderchocolate
Posts: 454
Joined: Mon Nov 01, 2010 6:55 am
Full name: Ted Wong

Re: Robodini Q&A

Post by kinderchocolate »

Don't overestimate Robodini. Peter Svidler won't stop buying Houdini just because there is another engine close to Houdini in strength. It won't affect the sales in any possible way. An engine can only be successfully if it is stable and reliable, Robodini doesn't have any of it.
Michel
Posts: 2272
Joined: Mon Sep 29, 2008 1:50 am

Re: Robodini Q&A

Post by Michel »

bob wrote: In the US as well. In that you simply can't patent software any longer. This changed at least 30 years ago. The only real protection here is copyright, which can extend to hardware since EEPROMS can be copied, which is copying a binary, which is copyrighted.
On what planet are you living? Software patents are very much alive.
The state of the art video codec H264 is protected by 1000s of patents.

http://www.mpegla.com/main/programs/avc ... c-att1.pdf

Just one obvious example.
User avatar
rvida
Posts: 481
Joined: Thu Apr 16, 2009 12:00 pm
Location: Slovakia, EU

Re: Robodini Q&A

Post by rvida »

mcostalba wrote: But one thing on which I made up an opinion I am quite confident is that I'd would like to ask Richard to, now, remove the Robodini binary link because, given that is almost at the same strength of Houdini (although it is not SMP) it could affect Houdini sales and this, I believe, was not the intended target. Moreover the "bet" has been demonstrated won now, so there is no more point in keep it public.
Good idea. I removed the original link. However, I'm afraid some people already created mirrors on file sharing sites which are out of my control :(
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: Robodini Q&A

Post by Dr.Wael Deeb »

mcostalba wrote:I'd prefer still not comment on the RE of Houdini by Richard because the thing it is really complex and there are a tons of grey areas, and there is a lot of historical background that cannot be ignored.

But one thing on which I made up an opinion I am quite confident is that I'd would like to ask Richard to, now, remove the Robodini binary link because, given that is almost at the same strength of Houdini (although it is not SMP) it could affect Houdini sales and this, I believe, was not the intended target. Moreover the "bet" has been demonstrated won now, so there is no more point in keep it public.
Try to limit a virus from replication once it entered the blood stream of a host....

Good luck with that regards,
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
lkaufman
Posts: 5960
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA

Re: Robodini Q&A

Post by lkaufman »

Houdini wrote:
lkaufman wrote: Isn't this an EXACT description of the origin of IPPOLIT, except that it took longer than 3 months? If Richard does use info from Houdini RE in Critter (and I'm not saying that he does), it would be exactly the same as what the IPPOLIT author(s) did, from which Houdini started. I'm not saying I approve of this, but I think you are not in a position to complain about RE. If you want to be fairly safe from RE, don't start with someone else's open source code, write an original program.
You're the guy that has repeatedly pushed Richard to discuss the RE findings of Houdini for the sole purpose of improving your own engine. You are inciting Richard to commit illicit actions for your personal benefit.
You also seem very happy to justify your crime by pointing to an alleged crime committed by the unknown people that created IPPOLIT.
I have nothing to do with all this, and find your attitude simply disgusting.

As you seem to be so much in favor of RE, let me ask you the same question I asked Don higher in this thread.

Suppose someone RE's your next Komodo 6 and then:
- Discusses the results openly on this forum.
- Publishes the source code in a different language (say, Fortran or Pascal) so that it's an obviously non-competitive engine but still shows all the details of your evaluation and search.
- Integrates important search and evaluation elements in his own engine, and then distributes his engine source code to competitors.
- Provides help on the forums to other people that want to RE Komodo.
- Releases a free Komostrelka 1.1 that plays exactly like your commercial Komodo 6.

If you approve of all this, I will hire Jury Osipov to do the job.

Robert
Regarding RE an engine and then writing your own based loosely on what you learned, as IPPO did to Rybka and as you allege that Richard did to Houdini, I don't know if it is illegal or not, there seems to be a dispute about that. If it is illegal, it is pretty much unenforceable, unless the similarities are so strong as to remove any doubt of its origin. So it is my position that it is up to the original author to make it reasonably difficult for someone to do this, mainly by not using substantial code that is already open source. It's kind of like saying that if you leave hundred dollar bills out on your lawn overnight, don't be surprised if someone steals them. In other words, don't make it so easy.
Regarding discussing results, once someone does the RE and used the knowledge himself, making the ideas public can't be criticized. If you consider the first action a crime like theft, then it's as if the thief chose to share his loot with the world. If it's not a crime, then certainly no harm done by sharing.
Releasing actual code from RE or releasing a virtual clone of an engine is much more clearly wrong and illegal than just getting and/or sharing ideas from RE. So I don't approve of Robodini or Strelka, or Houdini if it was based on a non-public domain derivative of IPPO as opposed to the original IPPO as has been alleged here(I don't have any special knowledge of this point except what I read in this forum).
So yes, I would not be happy if someone did all of the above to Komodo 6, though some of them would not bother me much. If someone does this to Houdini because they feel that Houdini forfeits the right to protection because it copied protected code, that is a reasonable position morally (I don't know about legally, I'm not a lawyer). In that case it comes down to whether Houdini did or did not use protected code.
We don't worry about this problem for Komodo because it is totally original, so it would be a lot of work for someone to RE it fully. Maybe when we are clearly superior to Houdini someone will do it, but not until then, I suppose. If you don't want people doing this to Houdini, it is in your power to make it extremely difficult. Rewrite the code from scratch, duplicating the function, as it is alleged that Richard Vida did with Critter more or less. If he could do it, you should be able to as well if you are as good as he is. That would also make your new engine immune from any potential lawsuit by Norm etc for using protected code.
Finally I would add that we have received very little benefit (maybe five elo or so in total) from purported Houdini ideas exposed in this forum. It seems that most of them only work in Ippo-derived engines. I am quite curious about how Houdini 3 gained 60 elo over Houdini 2, but I doubt that the answer would be very relevant to Komodo. Maybe part of it came from Komodo ideas someone found by RE, I wouldn't know unless someone reveals the entire list of changes from H2 to H3.
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: Robodini Q&A

Post by Dr.Wael Deeb »

rvida wrote:
mcostalba wrote: But one thing on which I made up an opinion I am quite confident is that I'd would like to ask Richard to, now, remove the Robodini binary link because, given that is almost at the same strength of Houdini (although it is not SMP) it could affect Houdini sales and this, I believe, was not the intended target. Moreover the "bet" has been demonstrated won now, so there is no more point in keep it public.
Good idea. I removed the original link. However, I'm afraid some people already created mirrors on file sharing sites which are out of my control :(
I sense a high dose of regret in your thread Richard.....
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
User avatar
Ajedrecista
Posts: 1969
Joined: Wed Jul 13, 2011 9:04 pm
Location: Madrid, Spain.

Re: Robodini Q&A.

Post by Ajedrecista »

Hello:
rvida wrote:
mcostalba wrote: But one thing on which I made up an opinion I am quite confident is that I'd would like to ask Richard to, now, remove the Robodini binary link because, given that is almost at the same strength of Houdini (although it is not SMP) it could affect Houdini sales and this, I believe, was not the intended target. Moreover the "bet" has been demonstrated won now, so there is no more point in keep it public.
Good idea. I removed the original link. However, I'm afraid some people already created mirrors on file sharing sites which are out of my control :(
Good point by Marco and good gesture by you, Richard. In fact I have seen mirrors of Robodini 1.1 and even some executables of hacked Houdini 3 copies that play at full strength IIRC. Anyway, Robert did things better this time: Houdini 2 was hacked in two days more less while the hack of Houdini 3 took around a month.

------------------------

http://www.chess2u.com/t6799p15-robodini-engine#40803

I read in Chess2U Forum that Critter should be in beta testing phase but the truth is that it is slightly delayed. Can you give more details about your improvements, please? Thanks in advance.

Regards from Spain.

Ajedrecista.
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: Robodini Q&A.

Post by Dr.Wael Deeb »

Ajedrecista wrote:Hello:
rvida wrote:
mcostalba wrote: But one thing on which I made up an opinion I am quite confident is that I'd would like to ask Richard to, now, remove the Robodini binary link because, given that is almost at the same strength of Houdini (although it is not SMP) it could affect Houdini sales and this, I believe, was not the intended target. Moreover the "bet" has been demonstrated won now, so there is no more point in keep it public.
Good idea. I removed the original link. However, I'm afraid some people already created mirrors on file sharing sites which are out of my control :(
Good point by Marco and good gesture by you, Richard... but I have seen mirrors of Robodini 1.1 and even some executables of hacked Houdini 3 copies that play at full strength IIRC. Anyway, Robert did things better this time: Houdini 2 was hacked in two days more less while the hack of Houdini 3 took around a month.

------------------------

http://www.chess2u.com/t6799p15-robodini-engine#40803

I read in Chess2U Forum that Critter should be in beta testing phase but the truth is that it is slightly delayed. Can you give more details about your improvements, please? Thanks in advance.

Regards from Spain.

Ajedrecista.
I am afraid that any future update of Criitter will be marked with a black flag....

Critter will never be the same regards,
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….