My two cents on Rybka's disqualification

[bob]

I think it is very likely that Vas has done something wrong. My big problem is with the process.

1. Vas was singled out. Where is the reverse engineering of Shredder, Hiarcs, Junior, etc.? Why has Vas been singled out for this scrutiny and the other commercial vendors have not?

Because a formal complaint was filed with the ICGA. This came from the strelka release that showed how close rybka and fruit were... Others will go under the microscope over time. Loop is already being discussed.

2. The process to find similarity is someone's recent invention. Has it been tested on other similar systems (meaning a test of similarity between TSCP and Brutus is absurd, a test between Junior and Shredder and other high end programs that are binary only against high-end open source programs makes sense)

Do you mean Mark's evaluation comparison? If you read his stuff completely, he _did_ compare a large matrix of programs to see who was similar to who. And Rybka/Fruit was the only one that reached a level of similarity that one had to believe in a rare 7.5SD event to be able to conclude the two programs were independently developed with so much similarity.

If the experiment is run without any controls, then what have we really shown?
To my way of thinking, it would also be a good idea to perform the following exercise:
A. Take a strong open source program and copy it.
B. Make a bunch of changes and form a binary.
C. Study the program and use the algorithms in a new program.
D. Compare the results of the tool for case B and C
I doubt if anybody wants to put the work in to do this, but it seems a good way to model the problem and test the results of analysis.
3. The prosecution was formed by asking "Hey, anybody who wants to throw stones at Vas, gather over here in the barn."
Imagine if a jury were formed in that manner.

Juries are pretty much formed in that manner. Random sample of the population. The ICGA "jury" was not out to crucify Vas, they were out to answer a complaint. The prosecutor does pretty well go around looking for people that can impeach the defendant's credibility, provide evidence to support a guilty verdict, etc. Again, we were _not_ the "finders of fact" (the jury). We were "the presenters of evidence".

4. Look into your own source code at the following:
A. PVS search
B. Null move reductions
C. LMR reductions
Now, do these look a whole lot like those found in some other programs? If so, why are you not prosecuting yourself? If it is OK, then where is the line to draw in the sand that says "This amount of similarity is OK, but this amount is cheating"?

Because you are still stuck on ideas. Which is different from _identical source code_. Extremely _different_. The idea argument is bogus, because it has not been mentioned once in the report we wrote, nor in the ICGAs verdict. This was about copying source code. nothing more. Nothing less.

We are not talking about "similarity". We are talking about copied source code.

[bob]

He was repeatedly asked to join the Wiki discussion.

Now you are speaking like committee was a neutral party and Vas was friendly asked to just asked to join the committee.

If the committee was neutral why then in the final report there is nothing which speaks for him, but it's is easy to find sarcastic remarks.

If the committee was acting a prosecutor, what was the point to ask the defendant to help prosecutor.

The committee was playing both sides, and that was really unfair.

Vas was given 30 days, _after_ the report was finished, to respond to any/all points raised. He did nothing. That is somehow "unfair"?

The "panel" was very much like an open court, where all sides meet in a single room. One person presents evidence, the other side then attacks the evidence, or the credibility of the witness, or whatever. And this back-and-forth continues until both sides are satisfied they have "made their points". Then on to the next witness or piece of evidence, and this repeats. It is as fair as it can be. If a defendant clams up and refuses to make any statements, that does not make the proceeding unfair. It marks the defendant as not very smart.

Normally a defendant either has his own lawyer, or gets one appointed by the court to ensure a fair trial. Why didn't the ICGA appoint one or more panel members to this role?

Edit: even Milosevic got a court-appointed lawyer when he didn't recognize the jurisdiction of the Hague tribunal.

That was Vas' responsibility. The ICGA is not a court, with a budget, and money to pay people to defend him. This is more like a "small claims court" in the US where you don't need nor get a court-appointed attorney. You present your own evidence, the other side does the same, and the judge renders a verdict.

[zamar]

In this case, Joona is simply right.

Oh and.

If you want crafty to win anything again you need to pla·gia·rize some of the ideas from Ippolit (Rybka 3, not Fruit anything). Other chess engine authors know this well. Or, you could use your reverse engineering skills to decipher Rybka 4.1 and use those ideas instead. The big cluster fast testing of eval terms that you use a lot, where did you get that idea from, btw?

Lastly, do you have any knowledge of who the people behind the pseudonyms of Head Comrade Yacov and Roberto Pescatore are?

Thank you.

Bob is the pioneer of the computer chess and a person I highly respect. Stop that utter nonsense!

[garybelton]

Does that answer your question???

I don't think so. BB (or BB+) said (in open forum) that fast tc cluster testing was Rajlich's (only?) contribution to comp chess eval testing ..
are you saying that you invented it instead?? Could be that BB was wrong of course.

Not a clue,

Thanks for putting that in print.

As this is the 2 cents thread, what we see here is simply:

http://en.wikipedia.org/wiki/Kangaroo_court

It's not surprising as this community is full of quarrelsome dissidence.

If Vasik were to sue you (for your online slanderous comments) and the ICGA then I am not sure of the outcome.
If the Courts awarded in his favor then the US Taxpaxer (UAB funders) would no doubt have to pick up the tab anyway, so your personal finances are probably safe.

Very disappointing situation nonetheless.

[bob]

The "panel" was very much like an open court, where all sides meet in a single room. One person presents evidence, the other side then attacks the evidence, or the credibility of the witness, or whatever. And this back-and-forth continues until both sides are satisfied they have "made their points". Then on to the next witness or piece of evidence, and this repeats. It is as fair as it can be. If a defendant clams up and refuses to make any statements, that does not make the proceeding unfair. It marks the defendant as not very smart.

There are 20 silent persons in the room and 10 persons furiously prosecuting you. After discussion, these 30 people will produce a report for final judgement. What's the point of even entering the room? For everything you say, there will be many persons trying to prove it wrong and very few (if any) trying to prove it right. Really "open wild-east court".

In a court proceeding (and I have served on 4 juries over my lifetime) every bit of testimony is taken apart, criticized, impeached when possible, to make things as favorable for one side as possible. It is up to the other side to tear that evidence apart and convince the jury it does not support a verdict of guilty.

BTW, you probably meant "wild-west", not "east". Vas could have invited any of his supporters to join the panel, just as a defendant can invite as many lawyers and witnesses as he deems necessary. When one chooses zero (0) one's chances are certainly hurt.

[Harvey Williamson]

I don't think so. BB (or BB+) said (in open forum) that fast tc cluster testing was Rajlich's (only?) contribution to comp chess eval testing ..
are you saying that you invented it instead?? Could be that BB was wrong of course.

Did BB(mark) say vas invented it or contributed to it?

[garybelton]

Bob is the pioneer of the computer chess

Agreed.

and a person I highly respect.

That used to be the case on my part.

Stop that utter nonsense!

What nonsense? We are seeing very un-Christian bevhavior here, my apologies if that does not mean anything to you.

[zamar]

What nonsense? We are seeing very un-Christian bevhavior here, my apologies if that does not mean anything to you.

It makes me sad, but pointless personal attacks are definetily not solution to that problem.

[garybelton]

http://www.open-chess.org/viewtopic.php ... fast#p7024

From BB+ talking about Rajlich ...

"As I've said before, I think modern testing methods (at ultra-fast time controls) is one of his major innovations. "

Sounds like Bob needs to rewrite that post for BB+, as clearly Bob invented it. OMG did BB+ made a mistake??!!

Congratulations to you and Hiarcs on your 2008 WC title.

yls.

[bob]

Does that answer your question???

I don't think so. BB (or BB+) said (in open forum) that fast tc cluster testing was Rajlich's (only?) contribution to comp chess eval testing ..
are you saying that you invented it instead?? Could be that BB was wrong of course.

I am not aware of Vas doing "fast cluster testing". I am aware of his doing very fast (hyper-bullet is the term Kaufman used) testing to play enough games. I believe they used a 4 or 8 cpu box to play 4 or 8 games at a time. Beyond that, I am not sure. You can certainly look back through CCC to see when I started my cluster testing. It was on an obsolete cluster that had 128 dual-cpu nodes and is well over 5 years old. I could probably find out when we started. I just checked and found some February 2005 dated files there. I can't say that was the beginning, it is possible that files were copied around by me and copies get the creation date set to the date they are copied, not the date of the copied file.

Rats, just found some 2004 dates and a bunch of earlier 2005 dates. So that is getting pretty close to "the beginning" I would think...

Not a clue,

Thanks for putting that in print.

As this is the 2 cents thread, what we see here is simply:

http://en.wikipedia.org/wiki/Kangaroo_court

The "court" tried Vas in absentia _solely_ because he chose to not participate. That is his fault, not ours. One can't compel one to defend himself if he refuses.

It's not surprising as this community is full of quarrelsome dissidence.

If Vasik were to sue you (for your online slanderous comments) and the ICGA then I am not sure of the outcome.

I am. One can't "slander" by "telling the truth."

If the Courts awarded in his favor then the US Taxpaxer (UAB funders) would no doubt have to pick up the tab anyway, so your personal finances are probably safe.

Very disappointing situation nonetheless.