Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / In retrospect (locked)
1 2 3 410 11 Previous Next  
Parent - - By bob (Gold) Date 2011-07-21 18:50
How deep does your ignorance of the law go?  How many times have you bought (say) a version of windows, or Intel's compiler, or some other commercial product, and then have to "accept" the license before you continue and are able to install the product?  No signed license for any version of windows, microsoft office, etc that I have, yet when it gets checked for validity by Microsoft, it always comes up "OK."

The rest of your post just shows ignorance of copyright law.  It is mine until I disclaim the copyright.  I have not.  End of story on that.  I simply, via the license agreement in the source, spell out what you can do with my code, and stay in compliance with copyright law, because I can either remain under copyright law but impose restrictions (which I did) or I can disclaim the copyright and make the code public domain, which I did not...

1.6.1 violated the license on at least two fronts.

*  Personal use includes any use you make of the program yourself, either by  *
*  playing games with it yourself, or allowing others to play it on your      *
*  machine,  and requires that if others use the program, it must be clearly  *
*  identified as "Crafty" to anyone playing it (on a chess server as one      *
*  example).  Personal use does not allow anyone to enter this into a chess   *
*  tournament where other program authors are invited to participate.  IE you *
*  can do your own local tournament, with Crafty + other programs, since this *
*  is for your personal enjoyment.  But you may not enter Crafty into an      *
*  event where it will be in competition with other programs/programmers      *
*  without permission as stated previously.            

(1) It can not be used in any event where other program participate unless it is clearly labeled "Crafty".  This would mean CCT-6 was a violation of the above.

(2) it can not be distributed without source being included.  It was sent to at least one computer chess rating list, which is _not_ "personal use".

I don't see where the "you encouraged..." comes from.  Seems like the license agreement is quite clear.  Play with it.  Modify it.  Learn from it.  But you can't use it to compete _anywhere_ unless it is identified as Crafty and you have my permission to do so...  Seems pretty simple to me.
Parent - - By AWRIST (****) Date 2011-07-21 21:50
Bob, after the magician David so and so retired, you are the best one in the States. Chris told you that you dont have a license agreement for Crafty which is true, then you begin to enter into magic mode and ask him how many times have you bought Windows and then signed the pop up "ok" and you as a computer scientist think that you had made a point but in fact you didnt. hehe

IMO we must find a general agreement for a minimum of humbleness or otherwise we will never understand the Levy wrongness against Vasik. Of course one had to repeat the basics of logic otherwise nobody could understand the details. In other words, you avoided the understanding question by simply calling out a voting. This is as if some coffeehousechessplayers would vote in a question for Kramnik against Anand. Of course such a basic vorting is rigged if the knowledge to understand the basics that are judged cant be understood by the voting majority. But this is no problem for spin doctors like Levy or Hyatt.

It's really beyond me why Vas has not yet called a legal court in a urgent question of protecting his rights and prevent further damages to his good name. Hopefully this will happen next week with the help of ChessBase that has good lawyers. It's time to tame the ICGA imposters and the ignorants of the secretariat. Make no mistake, these experts are no idiots in the underlying data. But they are completely helpless in the overall estimation of what is important and what is not. In other words this is a total confusion now. You dont even understand the difference between prosecution, experts and judges. You feel like being three in one. But in truth this could all be foreseen in advance. I have made that criticism more than once that unintentionally you naively destroy the old sport of computerchess because you just cant control the rules you wanted to put into the process. Or you violate the rights of the best like Vas and throw him out. But this isnt really what the community is wanting. The paying community, not the hypocrits among the other commercial players who are hiding in public. They all know that they would themselves never show their source codes. So by force the whole competition is dead. I reproached against you the warning that you would destroy computerchess as we all enjoyed it in the past. But you prefered not to listen because perhaps I'm just an outsider who cannot program. As if that had a negative impact on anything valuable in human life included the justice system and democracy.

Bob, I dont blame you for something you are incapable to understand. It's a pity that you have such a power although you dont have a clue. Excuse me I dont want to give the impression that I would bully you. But in a way it's really a trivial thing.
Parent - By bob (Gold) Date 2011-07-21 22:33
There is going to be no agreement when one wants to argue against basic written law.  And copyright law is quite clear.  And I, as the copyright holder, can constrict that law in any way I choose since the code is mine.  To argue otherwise is ridiculous.  The GPL is a license agreement that has withstood challenges in a court of law.  And yet, when I download GCC, or gnuchessx, or whatever, I do _not_ sign anything.  There is an "implied acceptance" concept that is understood universally.  Why Chris doesn't understand that is not something I am going to lose sleep over...
Parent - - By Nelson Hernandez (Gold) Date 2011-07-21 23:10

> It's really beyond me why Vas has not yet called a legal court in a urgent question of protecting his rights and prevent further damages to his good name.

This is really unlikely unless these bozos push him further against the wall, and I think they are pretty much spent now.  You have to look at the big picture. 

Cluster: hasn't fully launched yet, let's wait and see on its financial success.  ICGA impact: none. 
Rybka 5: still on its way, maybe 5-10 months away.  Should be a highly competitive product.  ICGA impact: none. 
Side projects: we know nothing about these, but I will bet they represent a much bigger source of income than the rest.  ICGA impact: none.

Why would he upset this state of affairs by starting a money and energy-sucking suit?  You have to have a sense of proportion.  The ICGA is not a significant blow to Rybka at all, contrary to what many may think.
Parent - - By AWRIST (****) Date 2011-07-22 13:05
Thanks Nelson. I would agree with the bigger picture, but isnt it also a bit about Hans as the operator or Lukas? They are also discriminated. Also Jiri for the books. I would expect legal actions to get reality into talks or interviews. I believe every single word from Vas but then a case should be trivial. I'm not a lawyer but if Vas is correct a banning of his programs is crazy. Einstweilige Verfügungen or Unterlassungserklärungen, this is something lawyers do successfully. Perhaps someone could translate the two terms into English. Could someone give the address of a lawyer who's working for ChessBase?
Parent - - By Harvey Williamson (*****) Date 2011-07-22 13:35
You need to be clear it is Vas that is banned not Hans, not Jiri, not jeroen, not Lukas etc.....
Parent - - By AWRIST (****) Date 2011-07-22 13:55
Perhaps this could be important. Do you say that if they wanted to start with Rybka 4 or cluster, they could  like all others? Until now it didnt look this way IMO. Could you clarify the situation?
Parent - - By Harvey Williamson (*****) Date 2011-07-22 13:57
Vas is the programmer of Rybka and so rybka would not be allowed. however these guys are free to participate in other teams if they wish.
Parent - - By AWRIST (****) Date 2011-07-22 14:19
Aha, you think they could be without principles like yourself?
For Vas and also his collegues a direct action against their
discrimination should be undertaken. Since others argue with
other sports in general it should be evident that outsingled athletes
could well appeal at courts of the legal system in their country.

Again, you are  a proven cheater and although you were excused
you should not be a part in such internal examinations, you will
always like to cheat if in case of a danger that your expected outcome
should be hindered by certain obstacles..

Also in these talks here I think it's nonsense to expect from
you an outbalanced judgement, because you have the job of
showing alleged wrongs in Rybka and its author.

But I have something for you. You could become a legend in the history
of computerchess if you would publish the cheats in Hiarcs by Uniacke and
this way you would prove that those are correct who always said that
the commercial competitors would all take code from each other just to
minimize the advantages of a particular leader. Ok, you would lose your job
as operator but what you would win that were honesty and public admiration.
Parent - By Harvey Williamson (*****) Date 2011-07-22 14:21
Parent - - By Nelson Hernandez (Gold) Date 2011-07-22 14:17

> Einstweilige Verfügungen or Unterlassungserklärungen

Non-German speakers do not want to translate these words.  We merely wish to admire them.  The latter is 23 letters long and comes with an umlaut, what more can you ask for!
Parent - - By AWRIST (****) Date 2011-07-22 14:33
Let me make known the principal though. A normal court case might take years.
That really costs nerves and money. But in such a case here where a conflict means that someone
has to suffer already a damage without a chance that a won case could repait that later, then you
choose a as if direct step that a) puts you back into the race and b) forbids that the evil libel is repeated.

Einstweilig means you get a decision on a momentary basis to enter the race. And Unterlassung threatens them with a money punishment if certain evil things are repeated about you. Certainly this does exist in other countries too.

Unfortunately I'm not a lawyer.

I have another question to the topic. Say a private organizer would allow Rybka but then urges them to show their source code. Ok that is one side of it, but is such a wanting believable, could a single person control the source code at all? Or is it just another variation of saying to f**off. Because who of the commercial players would open their source code for some private tournaments?
Parent - By Nelson Hernandez (Gold) Date 2011-07-22 18:03
NOBODY has a tournament big enough or important enough to compel commercial authors to release their source code.  Because if you are smart enough to understand the code, then you are a security risk for any author!  Computer chess teaches us again and again and again: trust no one!
Parent - - By Moz (****) Date 2011-07-24 16:51

> The latter is 23 letters long and comes with an umlaut, what more can you ask for!

Well, so much for that famous German efficiency we've all heard so much about. It certainly doesn't apply to the language!
Parent - - By Nelson Hernandez (Gold) Date 2011-07-24 17:54
What a thing to say!  They can sometimes put more meaning into one of their composite words than we can put into a sentence.  I'm not an expert in this subject, obviously, but I do remember the impressions that were left after two years of high school German almost 40 years ago.  I fondly remember the uproar that took place when we translated "Hans is looking for you" (Hans sucht dich).  It took a couple of minutes for the class to settle down and continue the lesson.
Parent - By Moz (****) Date 2011-07-24 19:26
Parent - By RFK (Gold) Date 2011-07-24 19:35
It is the German weltanschauung !
Parent - By Chess_Rambo (***) Date 2011-07-30 06:05
We use one word where you have to use many. :twisted:

Unterlassungserklärung == cease-and-desist declaration
Parent - - By turbojuice1122 (Gold) Date 2011-07-18 18:34
Let's not take the unreleased Rybka 1.6.1 and use its transgressions in place of arguments concerning publicly released versions of Rybka.  This isn't the first time that I've noticed this being done.
Parent - - By Watchman (***) Date 2011-07-18 20:28

>This isn't the first time that I've noticed this being done.

And this isn't the first time I have noticed you trying to wiggle out of an argument you've lost turbo.

"Show me some copied code!"

"Look here... should be easy enough even for you to see."

"Foul!  I cry foul!  You should have understood I wanted to see copied code released on a 'large scale' and not on a 'small scale'!"

Parent - - By turbojuice1122 (Gold) Date 2011-07-18 21:51
Actually, I mixed things up somewhat here.  Nonetheless, obviously Bob's statement was an exaggeration, which is what my original reply was getting at (as opposed to the post to which you replied, which is where I mixed things up between posts).  I know there is a copied code in pre-1.0, unreleased Rybka versions, though obviously if it was 100%, as Bob stated, the playing level would have been far higher.  Thus, the statement about taking it with a grain of salt.  My reply concerning 1.0 and later was thinking about a different post, and I later realized that I had replied to the wrong reply.
Parent - By bob (Gold) Date 2011-07-19 02:04
I can not guess why it played so poorly.  I suppose there -were- changes somewhere.  But I can tell you this, whatever they were, they were not obvious, and not in very many places.  Beyond that I can't guess.  Every piece of code examined was identical to standard Crafty source.  Does that mean there were no changes?  No.  But with so many samples taken, with zero "hits" that found original stuff, it suggests that there were not many, at best.
Parent - - By bob (Gold) Date 2011-07-19 02:09
Why?  Because you don't like the topic, since it shows blatant copying, and then blatant lying when Vas claimed that _all_ versions of Rybka, even the early ones, were 100% original?  It does create some difficulties for you it seems...  So let's not talk about that...
Parent - By turbojuice1122 (Gold) Date 2011-07-19 16:16
As I noted to Rob, my reply didn't really make much sense because I had confused different posts with each other.
Parent - - By bob (Gold) Date 2011-07-21 22:34
So if you rob from Peter, it is ok, but if you rob from Paul, it is not???


Let's just ignore that major transgression and move on to others.  Which we will also ignore, until there are no more to deal with???
Parent - - By turbojuice1122 (Gold) Date 2011-07-25 13:44
All I'm saying is that the pre-released Rybka versions cannot be used as evidence of ICGA violations.
Parent - - By bob (Gold) Date 2011-07-25 17:44
(1) any version that copies code is in violation of copyright.  Clearly 1.6.1 fits that and I do not hear a single person arguing otherwise.   Instead, since that is irrefutable, it becomes "OK, it was copied, but it was an early version..."  or something equally "forgiving."

(2) that version did violate tournament rules for CCT6 which used ICGA rules.

(3) during the ICGA investigation Vas explicitly stated that _all_ versions of Rybka, even the early ones, were 100% original.  That is a statement we certainly can check.  And it is false...
Parent - - By turbojuice1122 (Gold) Date 2011-07-27 13:40
I think that I once held a basement tournament using ICGA rules.  Ooops...I guess I accidentally included two versions of Rybka...I guess that's an issue??  As for "even the early ones, were 100% original", I think that Vas was clearly talking about 1.0 Beta and other versions in the Rybka 1 series.  It would have been silly to discuss this for Rybka 1.6.1 because that version was never released.  I think that you and everyone else understand this, too.
Parent - - By bob (Gold) Date 2011-07-27 16:26
No.  But if you modify my source and give it to a 3rd party and it participates in _his_ basement tournament, you violated my license agreement.  BTW, I don't think anyone considers CCT as a "basement tournament".  Those have been overall the _largest_ tournaments that have been held, one having something like 50+ participants.  And it was a tournament where the _authors_ wanted to use the ICGA rules, for the same reason we want to use them in ICGA events.  And also, CCT is an event for the authors.  Sound familiar???

You can try to allow him to escape, but his comments are quite incriminating.  He explicitly talked about "all versions being original" with the early versions being mtd(f), which 1.6.1 was certainly _not_  Whether 1.0.0 or whatever was mtd(f) I don't know...

And, for the record, 1.6.1 was released to at least one rating list group.  Which is _explicitly_ against my license.  And rating lists generally do not accept clones of others without permission of the original author and a statement showing that the program is a clone or close derivative.  Many have been _removed_ from rating lists for failing to do that.  From Le Petite, to voyager, to el chinito, and others that don't immediately come to mind.
Parent - By turbojuice1122 (Gold) Date 2011-07-28 15:00
Yes, this isn't "allowed", but on the other hand, there's really not anything that can be done about it if it's done--and it has NOTHING to do with the ICGA.  Using ICGA rules doesn't make it an ICGA event.  Obviously, version 1.6.1 would not have been included in the "all versions are original" because this wasn't really "Rybka as we know it", and it wasn't released publicly or commercially.  It just so happens that it has the same name as the engine that won the World Championship from 2007-2010 and dominated the rating lists for a 5-year period.
Parent - - By Ugh (*****) Date 2011-07-28 16:06
What is the provenance of this 1.6.1 version? You say it was given to ONE person, a third party, who entered it into the basement tournament?

So, we have, according to your assumptions, 1.6.1  leaves Vas's hand in 2004?

Then it plays in the tournament rather badly. Not like Crafty at all which presumably would have, or did, do better?

And then this version sits in a cupboard presumably until 2010? when it falls into your hands?

1. There's a tampering problem. How do you know the version 1.6.1 you have is the same as the one entered? How do you know somebody didn't recompile a mangled Crafty and deliver it to you, directly or indirectly as Rybka 1.6.1?

2. There's an initial provenance problem. How do you know the 1.6.1 actually used in the tournament came from Vas?

Any court would ask you to PROVE provenance. Can you?
Parent - - By Nick (*****) Date 2011-07-28 16:18
Trotsky, just so you know that I believe Bob Hyatt is prone to hearing things from others and not checking them out himself, he can believe that a rumour is gospel truth, if it agrees with his internal view of the world.


I point out in this post that he basically told an untruth about me (I was quite shocked) and he admitted he was wrong in the follow up.  He (like everyone else) has the capability to make stuff up to suit personal outlook - I think he made an honest mistake (for which I forgive him) but it does show a tendency.
Parent - - By bob (Gold) Date 2011-07-28 20:57
Why didn't you post the next reply from me?  "Then I was wrong and mis-remembered..."

I suppose that didn't fit with your implication, however???

There was a _lot_ of discussion during that tournament.  And afterward.  About the cluster stuff.  I stated what I recalled, and then said that I must have made a mistake in the attribution.

Is that such a bad thing?  I didn't hear things from "others" for this particular topic.  I heard or read it myself.  I just apparently got the person making the statement's name wrong.  What more is to be said?
Parent - - By Nick (*****) Date 2011-07-28 21:06

> Why didn't you post the next reply from me?  "Then I was wrong and mis-remembered..."

If you re-read my post you'll see that I did mention that you admitted you were wrong in the follow-up.

> I suppose that didn't fit with your implication, however???

It's very clear that you don't check your facts, and the above CCC link is just one example of that propensity.  Did you check Zach's disassembly of Rybka to ensure that it matches the code he put in the report?  Did you check the output from the "Rybka 1.6.1" you have to ensure that it could play the moves it allegedly did in CCT6?
Parent - - By bob (Gold) Date 2011-07-28 21:58
The FACT was correct.  The detail about who provided the information was not the key point.  We were having a discussion about the cluster rybka's search approach which was not very good at the time.  I did not even cite your name, if you recall...  So if that minor point is enough to say "I don't check my facts" then I suppose I was guilty as charged.  However, the FACT being discussed, split at root, was 100% accurate, which was the key point of the discussion.  Sort of like getting the day of the week wrong, but getting the event that happened laid out correctly, when the day was not that important...

oh well...

a mistake...
Parent - - By Nick (*****) Date 2011-07-28 22:01

> a mistake

Yes, it doesn't matter and like I said, I forgive you.

The main points ...

Did you check Zach's disassembly of Rybka to ensure that it matches the code he put in the report? (because It doesn't) and did you check the output from the "Rybka 1.6.1" you have to ensure that it could play the moves it allegedly did in CCT6?  (I am sure that Nelson has the PGNs, he has everything).
Parent - By Watchman (***) Date 2011-07-28 22:13

>Yes, it doesn't matter and like I said, I forgive you.


You should be asking him to forgive you acting like such a bonehead and bringing something like that up...

"I'm going to give you a tasty morsel of 'public gossip' with which we can further discredit Bob."

Parent - - By bob (Gold) Date 2011-07-28 22:15
No I did not check to match 1.6.1's moves to CCT6.  I don't see the point.  The point is that 1.6.1 contains a ton of crafty code.  That I did check.  And found other copied code that Mark/Zach had not paid any attention to, they had already found so much there...

I do not know what you mean about "checking assembly language" in the context of the PSTs that are being discussed.  There is no assembly language there...
Parent - - By Nick (*****) Date 2011-07-28 22:22

> No I did not check to match 1.6.1's moves to CCT6.  I don't see the point.

The point is that we don't know if this 1.6.1 that you have is the one that actually played in CCT6, which seems to be the cause of a lot of the fuss.   For due dilligence this has to be checked.

> in the context of the PSTs that are being discussed.  There is no assembly language there...

Zach produced bogus Rybka code, in C, supposedly from a disassembly of Rybka ..

This "static const int KingRankOpening = 0;" and the subsequent use of it in a multiplication to produce a King PST value.  My question is: do you believe that this code is in Rybka? (The setting zero followed by multiplication of it).
Parent - - By bob (Gold) Date 2011-07-29 02:07
It is in fruit.  What makes you think it is not in Rybka?  Did you overlook that?
Parent - - By Nick (*****) Date 2011-07-29 07:04

> It is in fruit.

That is incorrect, it is not in Fruit.  Line 48 of pst.cpp in Fruit 2.1 shows:

static const int KingRankOpening = 10;

10 is not 0.

I'll ask you again, do you believe that "static const int KingRankOpening = 0;" and the subsequent multiplication of it to obtain King PST values, as shown in the bogus Rybka code in the Zach report, is actually in Rybka?

You've skipped over responding to the checking of the output of the Rybka 1.6.1 binary that you have to see if it matches the actual moves that Rybka played in CCT6, so I will assume that you did this that and there was a high degree of "moves don't match".
Parent - - By bob (Gold) Date 2011-07-29 16:33
You are losing me.  If you go down to the "kings" section of Zach's report, you see this:

static const int KingFileOpening = 10;
static const int KingRankOpening = 10;

Which is exactly what is in my copy of fruit 2.1 at lines 47 and 48...
Parent - - By Nick (*****) Date 2011-07-29 16:38

> You are losing me

I will go slow.

In Zach shows the following bogus Rybka code:

static const int KingRankOpening = 0;

for (sq = 0; sq < 64; sq++) {
P(piece,sq,Opening) +=
KingRank[square_rank(sq)] *

For the third time, do you believe that this code is actually in Rybka?  I am trying to ascertain what you understand.
Parent - - By bob (Gold) Date 2011-07-29 16:58
Now I see your confusion.

(1) you realize that the code on the right is imaginary?  It is the code on the left, with the weights modified, so that you get the same PST values that Rybka _actually_ uses.  None of that code on the right is real.  It is what is necessary to produce the rybka PST values.  And it matches exactly with the code on the left, EXCEPTING the weights, which are way different.

(2) the weights on the left should _never_ match the weights on the right, because of the different material values used in the two programs.  And those weights are different (left to right) in every example, for that reason. 

That's about as clear as it can be.  If you can use the same _executable_ code and produce the fruit PST with one set of weights and the Rybka PST by just changing those 2 values, the two PSTs are nowhere near independent...
Parent - - By Nick (*****) Date 2011-07-29 17:02

> you realize that the code on the right is imaginary?

This is the key phrase where you at last admit the bogus Rybka code is made up, congratulations on being honest at last.

The ICGA placed this document on chessvibes for all chessvibes readers to read, mainly chess players and non-chess programming people like us on this forum.

With help from posts by other chess programmers I have found that the PST Rybka "code" in this Zach Wegner document (the very first code people see compared between Fruit and Rybka) is totally bogus and very misleading to anyone just glancing at the document. Folks (like me and other chessvibes readers) read the two columns of PST code, "Fruit" and "Rybka", see that they match and think maybe Vas was guilty of taking code from Fruit. But there is *no code* like this in the Rybka executable *at all* and Zach made it up.  It was not obtained using disassembly.  Each of the Rybka PST code tables in that document should be EMPTY of code rather than show Fruit code in Rybka as they do now.

I encourage anyone with connections to chessvibes to ask them to link to this post.
Parent - - By bob (Gold) Date 2011-07-29 18:14
You act like this is new information.  It is not.  In the top of Zach's PST description you find this:

Piece square tables are a very simple technique used for basic evaluation. For every piece type and square, PSTs
have a value for that piece being on that square. Fruit uses a clear and simple but effective way of calculating the
tables. Looking at Rybka's PSTs, we will see that they are calculated using these exact same constants except with
different weights. Also, note that here too that the PST values are hardcoded into the Rybka executable file, they are
not calculated at startup like Fruit's. The code shown here is simply the functional equivalent; it calculates the Rybka

Isn't that pretty clear?  At least to a programmer that understands the idea of "statically initialized"  rather than "calculated at startup".  I don't see any sense of being dishonest here at all.  One is interested in the data values, which are what defines program behavior.  Trying to show that they are "related" takes some sort of approach since you can't compare numbers and code against each other...
Parent - - By Nick (*****) Date 2011-07-29 18:31

>  I don't see any sense of being dishonest here at all.

Putting imaginary code that looks like Fruit under the heading of Rybka is dishonest.  But, in doing it with a caveat that many chessvibes users would not read or understand, it becomes honest to you. 

It's the same kind of honesty as:

"No one on the panel thought him innocent" .. without mentioning that only 14 of 34 voted.

This is binary 1/0 thinking, not like the reality of the real world.  Sorry, not buying it.
Parent - By bob (Gold) Date 2011-07-29 21:07
Buy what you want.  If you can't read the report _carefully_.  And I have _repeatedly_ used the word _CAREFULLY_ here, then it is hardly my fault if you get confused.  It is not for beginners, as I mentioned...
Parent - - By RFK (Gold) Date 2011-07-28 22:23
Hey Vas did pretty damn good using your code considering he placed 53rd -he could have been last. That would have been really insulting to your code fetishizing ego.
Parent - - By bob (Gold) Date 2011-07-29 02:05
So if he copies code and improves it significantly, it is OK.  If he copies code and breaks it so that it performs significantly worse, that is also ok?
Up Topic Rybka Support & Discussion / Rybka Discussion / In retrospect (locked)
1 2 3 410 11 Previous Next  

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill