Strelka again

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

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
Rolf
Posts: 6081
Joined: Fri Mar 10, 2006 10:14 pm
Location: Munster, Nuremberg, Princeton

Re: Strelka again

Post by Rolf » Fri May 11, 2007 7:17 pm

Uri Blass wrote:
Sergei Markoff wrote:I have sent messages to Bryan Hoffmann and Dann Corbit offering them to be the experts. Yury Osipov give me a permission to do with his sources everything what I want. He is really depressed with this story. I hope he will not stop his work after such a "greeting" from the community.
In this case I believe him
Tell him that he should understand that the suspect was based on history of new strong programs that were discovered as clones and it is not something personal against him.

The choice of the name also gave the impression that it may be a joke about rybka.

Uri

Uri, BTW thanks for your in general decent contributions in this witch-hunt. But still I want to state that here with this description I must strongly disagree. Before I explain my standpoint I want to say that technically I still dont know what is going on in the debate. But the aspect of decency is my interest.

You can speak for yourself when you say that it wasnt against the author, personally, but because of earlier cases with clones in the end.

Uri, with all respect I must insist that even such a history of bad experience does not justify that exorcists with their provoking speech mistreat Sergei who figurated as sort of replacement for the real author. (personal attack deleted)
It's not trivial for me to define what is the true scandal because I live also under that cloud of habits and routines but if one dares to make one's own thoughts then it's easier.

Instead of viewing the many authors, their engines and the different strengths, as an art, we assure that we are caring for a climate that appeals to low instincts of misbelief, treason and malicious gossip. And then comes a real icon of computerchess and claims that this is because we were not living in Heaven but in the full dirtyness of our World. Fool of me who always believed that computerchess was a branch of computer sciences; now I'm taught that it's allegedly a section of Scotland Yard.

I urge the community to be friendly towards their new members and to finally convict someone only AFTER all possible doubts have been refutated, normally trivial things in a democratic society.
-Popper and Lakatos are good but I'm stuck on Leibowitz

Alexander Schmidt
Posts: 1098
Joined: Thu May 10, 2007 12:49 pm

Re: Strelka again

Post by Alexander Schmidt » Fri May 11, 2007 7:29 pm

Hello Sergei,

It's a shame, that Yuri is forced to show his sources.

As I wrote in another post: I wouldn't do that, he showed his sources to a trustable person, that is more than he has to do.

I would understand if he don't want to share his work with us in the future after all. And I can also imagine other programmers with such jewels on the hard disk thinking twice before they release something.

The people only search for similaries, not for differences...

Disapointed,
Alex :(

User avatar
Graham Banks
Posts: 34518
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: Strelka again

Post by Graham Banks » Fri May 11, 2007 7:35 pm

Rolf wrote: I urge the community to be friendly towards their new members and to finally convict someone only AFTER all possible doubts have been refutated, normally trivial things in a democratic society.
Wise words.

User avatar
Thomas Mayer
Posts: 383
Joined: Thu Mar 09, 2006 5:45 pm
Location: Nellmersbach, Germany

Re: Strelka again

Post by Thomas Mayer » Fri May 11, 2007 7:41 pm

Hallo Alex,
Alexander Schmidt wrote:The people only search for similaries, not for differences...
have you looked at the positions I gave ? These are simply the first few given positions in the CCC archives about Rybka's abnormalities. When I start I really didn't know what I would find... but you may look yourself...

Greets, Thomas

User avatar
smirobth
Posts: 2307
Joined: Wed Mar 08, 2006 7:41 pm
Location: Brownsville Texas USA

Re: Strelka again

Post by smirobth » Fri May 11, 2007 8:21 pm

Sergei Markoff wrote:I have sent messages to Bryan Hoffmann and Dann Corbit offering them to be the experts. Yury Osipov give me a permission to do with his sources everything what I want. He is really depressed with this story. I hope he will not stop his work after such a "greeting" from the community.
Hi Sergei,
I have seen many of the unkind things that have been said and certainly can understand his feeling depressed. However another way to look at it is that people think his program is so very strong, something highly unusual for a new program, that many assumed it "must" be a clone. In a way it is an odd sort of compliment regarding his impressive achievement, something for which he could feel very proud. I certainly hope he doesn't stop his work after making such a great start.
- Robin Smith

Alexander Schmidt
Posts: 1098
Joined: Thu May 10, 2007 12:49 pm

Re: Strelka again

Post by Alexander Schmidt » Fri May 11, 2007 8:42 pm

Thomas Mayer wrote:have you looked at the positions I gave ?
Hello Thomas,

I didn't follow the discussions from the beginning, I saw Eduard Nemeths positions, most of them had to do with underpromotion. You cant take 4 positions with underpromotions as 4 hints. It's 1 hint. Strelka and Rybka Beta don`t know underpromotion. Thats all. Also the other position was not convincing. ( http://f50.parsimony.net/forum202220/messages/23616.htm )

Maybe there where a huge ammount of such positions, then I didn't see all of them. But after my experiences I don't count much on positions. I try to look deeper: movegen, evaluation, knowledge, protocol handling, strings if possible.

I have my own special positions. In my positions I found no similaries, that was enough for me.

Thomas, it took a few minutes to find out that there are different things like the movegenerator. At this point the people said it is pure Rybka with a kind of Adapter in a new exe. I expect from experts that they look on different sides of an engine before claiming it's a clone.

OK, probably I am wrong, the experts didn't say its a clone, but they made the community believe it is a clone.

Best,
Alex

NKOTB

Re: Strelka again

Post by NKOTB » Fri May 11, 2007 8:47 pm

smirobth wrote:
Sergei Markoff wrote:I have sent messages to Bryan Hoffmann and Dann Corbit offering them to be the experts. Yury Osipov give me a permission to do with his sources everything what I want. He is really depressed with this story. I hope he will not stop his work after such a "greeting" from the community.
Hi Sergei,
I have seen many of the unkind things that have been said and certainly can understand his feeling depressed. However another way to look at it is that people think his program is so very strong, something highly unusual for a new program, that many assumed it "must" be a clone. In a way it is an odd sort of compliment regarding his impressive achievement, something for which he could feel very proud. I certainly hope he doesn't stop his work after making such a great start.
I don't agree with personal attacks either but I understand your comment. My guess is every author here would take it as a compliment if their engines were comparable to Rybka. It will not be too long before we have final words on this.
.
On another note, if I release a program there will be no debate if it is a Rybka clone or any clone for that matter. The only debate will be if it is a three or four digit ELO. :lol:

Zlaire

Re: Strelka again

Post by Zlaire » Fri May 11, 2007 8:56 pm

Sergei Markoff wrote:He is really depressed with this story. I hope he will not stop his work after such a "greeting" from the community.
Sorry to say it but I find things like this rather silly. If people come with claims, just prove them wrong and get on with it, just like happened here. I'm very happy Strelka seems to turn out to be a genuine engine, and I'm sure everyone else here is.

It shouldn't be hard to understand why people are suspicious about a new strong engine that comes out of 'nowhere'.

Making a big thing about how evil everyone is just seems ridiculous.

I for one hope people keep checking new engines, as long as you're in the right you have nothing to 'fear'.

ernst
Posts: 336
Joined: Thu Mar 09, 2006 5:00 pm

Re: Strelka again

Post by ernst » Fri May 11, 2007 9:02 pm

It takes two source codes to compare.
This post may either be cause or result of misunderstanding.

cooldalek

Re: Strelka again

Post by cooldalek » Fri May 11, 2007 9:04 pm

Basically it shows a very sad side of a lot of people here

A lot of people actually WANTED strelka to be a clone so they can go on worshipping Rynka

I feel for the authorl. He may have been planning on going commercial and his reputation has been damaged

Post Reply