What should Robert do? (read post first before you vote)

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

Moderators: hgm, Rebel, chrisw

What should Robert do ?

Leave things as they are, I don't care
24
26%
Give credit to Norman and Milos for their initial work
20
22%
Compensate Norman and Milos financially
6
7%
I only want Robert to admit the Robbolito origin
41
45%
 
Total votes: 91

User avatar
Rebel
Posts: 6991
Joined: Thu Aug 18, 2011 12:04 pm

What should Robert do? (read post first before you vote)

Post by Rebel »

Folks,

I hate to see Robert dragged through the mud every time and this poll is meant to suggest some constructive solutions and to express your feelings as a guide for him what to do. But first allow me to put things in a historic nutshell before you cast your vote.

1. When Ippolit appeared in 2009 it immediately was claimed by Vas as Rybka. But Vas never provided the proof for that. Sven asked Vas direct questions about his Ippolit claim but Vas refused to back-up his accusation.

As I see it, accusations without providing evidence are worthless and when given the chance by the initiative of Sven (June 2010) Vas failed again what he should have done in the first place when he made the Ippolit claim 8 months earlier.

By his refusal to do so the claim lost its right, perhaps even on a juridical base. So Ippolit is freeware. Everybody is free to start from it. Julien as a lawyer might want to shed his light on this.

2. Ippolit became Robbolito, released as freeware also. But Robbolito was Italian and thus an obstacle for non Italian speaking people to start from it. Here Norman and Milos did a huge translation job and made the free Robbolito source code understandable in English. Then Norman released this english version under GPL.

3. Robert without a shred of doubt started from this English version. It's as crystal clear as the pre-Rybka's were rooted in Crafty. I have not heard one programmer saying otherwise. There is full consensus among programmers on both cases which is quite unique I must say :lol:

4. There is absolutely no kind of wrong doing if Robert would have started from the Italian freeware Robbolito or the raw Ippolito, fact is he did not and started from the GPL protected English Robbolito which only protects the work Norman and Milos have done to Robbolito.

And herein lies the origin of the constant scolding of his name being a cloner in endless loops.

I have vote myself for option-2.

In the hope the scolding might stop one day.
mar
Posts: 2554
Joined: Fri Nov 26, 2010 2:00 pm
Location: Czech Republic
Full name: Martin Sedlak

Re: What should Robert do? (read post first before you vote)

Post by mar »

Quoting Norman from another thread:
kranium wrote: our work included:
porting from raw unstable ippolit linux release to windows
much debugging: resolution of unhandled exception errors, etc;
the addition of pondering and ponderhit
optimization of the alignment of hash structures
different and improved bit scan functions
implementation of mersenne rand functions
new faster version of popcnt functions
cache prefetching
smooth scaling of null move
improved/corrected time management
better UCI GUI compatibility
CPU load
upper/lower bound printing
move on ponderhit
etc.
and of course, the translation from very difficult to read code to into common and easily understandable programming terms

the development history is extremely well documented, and
all this source code is freely avaialble...anyone can examine it to verify these facts
So 99% of these changes are worth absolutely nothing in terms of strength.
Mostly cleanup.
What has value is probably really only the translation, if they understood what was going on because the original "author(s)" had absolutely no clue what some parts of the code were supposed to do.
What's funny is that Houdart cries because someone REd his engine, claiming it's illegal,
but the ippo* family is exactly that, so sort of double standard.
If RE is illegal then ippo* is too. But if it's public domain then it's ok?
Funny really :)
kranium
Posts: 2129
Joined: Thu May 29, 2008 10:43 am

Re: What should Robert do? (read post first before you vote)

Post by kranium »

Ed,
i'm not interested in being compensated, and am not seeking individual recognition...
i did it for the love of chess and programming

i do hope that he retracts this statement:
Houdini wrote:Houdini does NOT contain any Ippolit code.
Robert
and give full acknowledgement to the real Robbolito authors Yakov Petrovich Golyadkin, Igor Igorovich Igoronov, Roberto Pescatore

PS - his web page acknlowledgement for 'ideas' has changed and now includes Robbolito:
"Without many ideas from the excellent open source chess engines Ippolit/Robbolito, Stockfish and Crafty, Houdini would not nearly be as strong as it is now."

so there may be hope after all...
jd1
Posts: 269
Joined: Wed Oct 24, 2012 2:07 am

Re: What should Robert do? (read post first before you vote)

Post by jd1 »

kranium wrote:Ed,
i'm not interested in being compensated, and am not seeking individual recognition...
i did it for the love of chess and programming

i do hope that he retracts this statement:
Houdini wrote:Houdini does NOT contain any Ippolit code.
Robert
and give full acknowledgement to the real Robbolito authors Yakov Petrovich Golyadkin, Igor Igorovich Igoronov, Roberto Pescatore

PS - his web page acknlowledgement for 'ideas' has changed and now includes Robbolito:
"Without many ideas from the excellent open source chess engines Ippolit/Robbolito, Stockfish and Crafty, Houdini would not nearly be as strong as it is now."

so there may be hope after all...
Very good! And thanks to Ed too. I vote for Option 2.

Jerry
kranium
Posts: 2129
Joined: Thu May 29, 2008 10:43 am

Re: What should Robert do? (read post first before you vote)

Post by kranium »

mar wrote: So 99% of these changes are worth absolutely nothing in terms of strength.
Mostly cleanup.
Martin-
it's hard not to disagree with this...
the many code and speed optimizations, pondering, etc. improved strength measurably

the abundant 'cleanup', error resolution, bug fixes, etc. were all critical...
i.e. any engine that is stable is much stronger than one that crashes continuously...
:wink:

it was only after all this debugging and optimization that it could be tested and it's full potential recognized...
for a short time it was #1
...until the very 1st Houdini was released (just slightly stronger)

any one using this code as base for a new engine would have several months of thankless/tedious work already done for him
mar wrote: What's funny is that Houdart cries because someone REd his engine, claiming it's illegal,
but the ippo* family is exactly that, so sort of double standard.
If RE is illegal then ippo* is too. But if it's public domain then it's ok?
Funny really :)
good point...
:lol:
Last edited by kranium on Fri Feb 01, 2013 10:20 am, edited 1 time in total.
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: What should Robert do? (read post first before you vote)

Post by zullil »

kranium wrote:Ed,

i do hope that he retracts this statement:
Houdini wrote:Houdini does NOT contain any Ippolit code.
Robert
and give full acknowledgement to the real Robbolito authors Yakov Petrovich Golyadkin, Igor Igorovich Igoronov, Roberto Pescatore
In what sense are they "authors?" In what sense are they "real?" :D

I have no doubt about the origin of Houdini. What still puzzles me is the origin of Robbolito. What else could it be but a reverse-engineering of Rybka? Anything that you'd care to address?
mar
Posts: 2554
Joined: Fri Nov 26, 2010 2:00 pm
Location: Czech Republic
Full name: Martin Sedlak

Re: What should Robert do? (read post first before you vote)

Post by mar »

kranium wrote:the 'cleanup', error resolution, bug fixes, etc. were all critical...
i.e. any engine that is stable is much stronger than one that crashes continuously...
:wink:
I wonder why it was so buggy in the first place. There's something fishy about that :wink:
kranium
Posts: 2129
Joined: Thu May 29, 2008 10:43 am

Re: What should Robert do? (read post first before you vote)

Post by kranium »

zullil wrote:
kranium wrote:Ed,

i do hope that he retracts this statement:
Houdini wrote:Houdini does NOT contain any Ippolit code.
Robert
and give full acknowledgement to the real Robbolito authors Yakov Petrovich Golyadkin, Igor Igorovich Igoronov, Roberto Pescatore
In what sense are they "authors?" In what sense are they "real?" :D

I have no doubt about the origin of Houdini. What still puzzles me is the origin of Robbolito. What else could it be but a reverse-engineering of Rybka? Anything that you'd care to address?
well, thats a good question Louis,
i can only say this:
they are (of course!) real people, and have been quite active communicating info and responding to questions on the ippolit.wikispaces.com site
in addition, there exists a whole team of volunteer 'decembrists' (programmers as well as testers) that were/are active developing source code for the effort

it's unfortunate they are using pseudonyms, many have excluded the engines from testing, as if the use of pseudonyms is evidence of some sort of wrongdoing

1st:
Roberto Pescatore is a very common name...
probably 100s or 1000s in Italy alone?
2nd:
pseudonyms fit perfectly with the whole communist style 'workers' revolution
i.e. the Decembrists have numbers...

of course anything may be true...doesn't much matter it's past history and probably not worth discussing
Last edited by kranium on Fri Feb 01, 2013 10:48 am, edited 2 times in total.
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: What should Robert do? (read post first before you vote)

Post by zullil »

kranium wrote: of course anyhting may be true...doesn't much matter it's past history
No, only the truth can be true. And it matters.
Lavir
Posts: 263
Joined: Sun Oct 28, 2012 11:45 am

Re: What should Robert do? (read post first before you vote)

Post by Lavir »

mar wrote: What's funny is that Houdart cries because someone REd his engine, claiming it's illegal,
but the ippo* family is exactly that, so sort of double standard.
If RE is illegal then ippo* is too. But if it's public domain then it's ok?
Funny really :)
OMG.

The two things have NOTHING in common.

1. It has not been proven that Ippo is RE of Rybka. Vas never did care to release proof and there are many contrasting opinions on the matter anyway. So the RE (and in which form has been done) is "cloudy" at best.

2. Since Vas (the direct interested party) never did commit proof explicitly not in action nor in words (saying in an e-mail that you are sure it is RE of rybka doesn't constitute proof of anything) Ippolit has become ipso-facto freeware and public domain after nothing has been done about it and it has been used in fact by EVERY top engine author (funny how people like you keep forgetting this and quote only Houdart using the source).

3. In this case instead there is 100% proof of Robodini being a RE of Houdini 3 (as it is obvious, isn't it?) and Houdart has all the rights not to be happy about it as Vas could be if he wanted to (if he was 100% sure as in this case that Ippo was RE of his work); but if you do nothing or you don't care there is a prescription in law - do you know about it? After the prescription the proprietary becomes another (and in this case the "another" is public).

So, how can you compare the two and say that Robert is using double standards is beyond me. Matter of fact is: Ippolit has become public domain (if you like it or not makes no difference) for what happened, Robodini will NEVER become such.