WBEC Ridderkerk new results.

[Leo Dijksman]

Geschrieben von: / Posted by: Leo Dijksman at 15 February 2004 17:05:51:

1st Division:
Round 3:
Ktulu 4.2 11½0 2.5/4
WildCat 3.0 00½1 1.5/4
------------------------------------
King Of Kings 2.52 ½011 2.5/4
Comet B.68 ½100 1.5/4
------------------------------------
SOS 99.11.03 ½01 1.5/3 = 1 game to go!
Delfi 4.4 ½10 1.5/3
------------------------------------

3rd Division:
Round 4:
Butcher 1.42c 1010 2.0/4
NullMover 0.24d 0101 2.0/4
------------------------------------
Hagrid 0.7.56 1½0½ 2.0/4
Dorky 3.48 0½1½ 2.0/4
------------------------------------
TRACE 1.25 01½1 2.5/4
CyberPagno 2.01 10½0 1.5/4
------------------------------------

5th Division:
Round 4:
MrChess 2.1 - Grizzly 1.40.1b 1-0 53
Grizzly 1.40.1b - MrChess 2.1 0-1 66
Round 5:
Cilian 4.13 - Chiron 0.38 =-= 55
Chiron 0.38 - Cilian 4.13 1-0 65
Simontacchi 1.8a - BigLion 2.23i 0-1 47
BigLion 2.23i - Simontacchi 1.8a 1-0 51
EnginMax 5.11c - NoonianChess 3.6e 1-0 69
NoonianChess 3.6e - EnginMax 5.11c 0-1 42
Qalat 0.19b - Neurosis 1.7b 0-1 58
Neurosis 1.7b - Qalat 0.19b =-= 128
Ax 0.8 - NagaSkaki 2.41 =-= 90
NagaSkaki 2.41 - Ax 0.8 1-0 75
Faile 1.4 - SmallPotato 0.6.1 1-0 67
----------------------------------------------------------

6th Division:
Round 1, 2 and 3:
Cefap 0.7.2 - Fafis 0.85 0-1 47
Fafis 0.85 - Zotron 2003 R7.5 0-1 61
Fafis 0.85 - Adam 1.6 1-0 95
Round 4:
ChessRikus 1.4.60 - KKFChess 2.6.1 =-= 128
Enigma 1.1.3 - Deuterium 04.01.04.1 0-1 128
Rainman 0.7.5 - Robin 0.9.86 0-1 49
PolarChess 1.3 - SdBC 0.4.13.0 1-0 26
Smash 0.8d - Replicant 1.2e 0-1 35
TSCP 1.81 - LarsenVB 0.05.01 1-0 31
Embracer 1.12 - ApiChess 1.29 1-0 33
Beaches 1.52 - Booot 2.4 0-1 49
Aice 0.55 - Parrot 031231 1-0 75
Golem 0.4 - BSC 2.8 0-1 42
Belzebub 0.64 - Fafis 0.85 1-0 73
----------------------------------------------------------
After a little conversation with Robert Hyatt and Daniel Shawul I have
decided to take DanChess out and replaced it by Fafis 0.85!

Crosstables and pgn on the WBEC homepage.
Enginelist, latest updates/new engines:
15/02/2004:
Updated: PostModernist 1009
Leo.


WBEC Ridderkerk homepage.

[Uri Blass]

Geschrieben von: / Posted by: Uri Blass at 15 February 2004 17:34:40:
Als Antwort auf: / In reply to: WBEC Ridderkerk new results. geschrieben von: / posted by: Leo Dijksman at 15 February 2004 17:05:51:

1st Division:
Round 3:
Ktulu 4.2 11½0 2.5/4
WildCat 3.0 00½1 1.5/4
------------------------------------
King Of Kings 2.52 ½011 2.5/4
Comet B.68 ½100 1.5/4
------------------------------------
SOS 99.11.03 ½01 1.5/3 = 1 game to go!
Delfi 4.4 ½10 1.5/3
------------------------------------

3rd Division:
Round 4:
Butcher 1.42c 1010 2.0/4
NullMover 0.24d 0101 2.0/4
------------------------------------
Hagrid 0.7.56 1½0½ 2.0/4
Dorky 3.48 0½1½ 2.0/4
------------------------------------
TRACE 1.25 01½1 2.5/4
CyberPagno 2.01 10½0 1.5/4
------------------------------------

5th Division:
Round 4:
MrChess 2.1 - Grizzly 1.40.1b 1-0 53
Grizzly 1.40.1b - MrChess 2.1 0-1 66
Round 5:
Cilian 4.13 - Chiron 0.38 =-= 55
Chiron 0.38 - Cilian 4.13 1-0 65
Simontacchi 1.8a - BigLion 2.23i 0-1 47
BigLion 2.23i - Simontacchi 1.8a 1-0 51
EnginMax 5.11c - NoonianChess 3.6e 1-0 69
NoonianChess 3.6e - EnginMax 5.11c 0-1 42
Qalat 0.19b - Neurosis 1.7b 0-1 58
Neurosis 1.7b - Qalat 0.19b =-= 128
Ax 0.8 - NagaSkaki 2.41 =-= 90
NagaSkaki 2.41 - Ax 0.8 1-0 75
Faile 1.4 - SmallPotato 0.6.1 1-0 67
----------------------------------------------------------

6th Division:
Round 1, 2 and 3:
Cefap 0.7.2 - Fafis 0.85 0-1 47
Fafis 0.85 - Zotron 2003 R7.5 0-1 61
Fafis 0.85 - Adam 1.6 1-0 95
Round 4:
ChessRikus 1.4.60 - KKFChess 2.6.1 =-= 128
Enigma 1.1.3 - Deuterium 04.01.04.1 0-1 128
Rainman 0.7.5 - Robin 0.9.86 0-1 49
PolarChess 1.3 - SdBC 0.4.13.0 1-0 26
Smash 0.8d - Replicant 1.2e 0-1 35
TSCP 1.81 - LarsenVB 0.05.01 1-0 31
Embracer 1.12 - ApiChess 1.29 1-0 33
Beaches 1.52 - Booot 2.4 0-1 49
Aice 0.55 - Parrot 031231 1-0 75
Golem 0.4 - BSC 2.8 0-1 42
Belzebub 0.64 - Fafis 0.85 1-0 73
----------------------------------------------------------
After a little conversation with Robert Hyatt and Daniel Shawul I have
decided to take DanChess out and replaced it by Fafis 0.85!

Crosstables and pgn on the WBEC homepage.
Enginelist, latest updates/new engines:
15/02/2004:
Updated: PostModernist 1009
Leo.

Can you give more details about the reason for replacing DanChess?
What parts of Danchess are identical to Crafty?
Uri

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 18:01:45:
Als Antwort auf: / In reply to: Re: WBEC Ridderkerk new results. geschrieben von: / posted by: Uri Blass at 15 February 2004 17:34:40:

1st Division:
Round 3:
Ktulu 4.2 11½0 2.5/4
WildCat 3.0 00½1 1.5/4
------------------------------------
King Of Kings 2.52 ½011 2.5/4
Comet B.68 ½100 1.5/4
------------------------------------
SOS 99.11.03 ½01 1.5/3 = 1 game to go!
Delfi 4.4 ½10 1.5/3
------------------------------------

3rd Division:
Round 4:
Butcher 1.42c 1010 2.0/4
NullMover 0.24d 0101 2.0/4
------------------------------------
Hagrid 0.7.56 1½0½ 2.0/4
Dorky 3.48 0½1½ 2.0/4
------------------------------------
TRACE 1.25 01½1 2.5/4
CyberPagno 2.01 10½0 1.5/4
------------------------------------

5th Division:
Round 4:
MrChess 2.1 - Grizzly 1.40.1b 1-0 53
Grizzly 1.40.1b - MrChess 2.1 0-1 66
Round 5:
Cilian 4.13 - Chiron 0.38 =-= 55
Chiron 0.38 - Cilian 4.13 1-0 65
Simontacchi 1.8a - BigLion 2.23i 0-1 47
BigLion 2.23i - Simontacchi 1.8a 1-0 51
EnginMax 5.11c - NoonianChess 3.6e 1-0 69
NoonianChess 3.6e - EnginMax 5.11c 0-1 42
Qalat 0.19b - Neurosis 1.7b 0-1 58
Neurosis 1.7b - Qalat 0.19b =-= 128
Ax 0.8 - NagaSkaki 2.41 =-= 90
NagaSkaki 2.41 - Ax 0.8 1-0 75
Faile 1.4 - SmallPotato 0.6.1 1-0 67
----------------------------------------------------------

6th Division:
Round 1, 2 and 3:
Cefap 0.7.2 - Fafis 0.85 0-1 47
Fafis 0.85 - Zotron 2003 R7.5 0-1 61
Fafis 0.85 - Adam 1.6 1-0 95
Round 4:
ChessRikus 1.4.60 - KKFChess 2.6.1 =-= 128
Enigma 1.1.3 - Deuterium 04.01.04.1 0-1 128
Rainman 0.7.5 - Robin 0.9.86 0-1 49
PolarChess 1.3 - SdBC 0.4.13.0 1-0 26
Smash 0.8d - Replicant 1.2e 0-1 35
TSCP 1.81 - LarsenVB 0.05.01 1-0 31
Embracer 1.12 - ApiChess 1.29 1-0 33
Beaches 1.52 - Booot 2.4 0-1 49
Aice 0.55 - Parrot 031231 1-0 75
Golem 0.4 - BSC 2.8 0-1 42
Belzebub 0.64 - Fafis 0.85 1-0 73
----------------------------------------------------------
After a little conversation with Robert Hyatt and Daniel Shawul I have
decided to take DanChess out and replaced it by Fafis 0.85!

Crosstables and pgn on the WBEC homepage.
Enginelist, latest updates/new engines:
15/02/2004:
Updated: PostModernist 1009
Leo.

Can you give more details about the reason for replacing DanChess?
What parts of Danchess are identical to Crafty?

No parts of it are identical. Some parts are very similar. It is clear that DanChess has used crafty algorithms. Dr Hyatt and I have exchanged perhaps 20 emails on the subject. We disagree about both the spirit and the extent of whether or not DanChess is a crafty clone. He says it is, and I say it is not.
But it does not matter, because DanChess is undergoing a complete rewrite.
There is one routine which I must agree is extremely similar. It seems to be a major bone of contention for Dr. Hyatt.
This is from DanChess:
int BOARD::SEE(int from,int to)
{
BITBOARD attacks,tmp;
RooksQueens =Pieces[white][rook] | Pieces[black][rook] | Pieces[white][queen] | Pieces[black][queen];
BishopsQueens=Pieces[white][bishop] | Pieces[black][bishop] | Pieces[white][queen] | Pieces[black][queen];

int attacked_piece;
int sq,dir;
int col,nc=1;
int swap_list[32];
attacks = AttackTo(white,to) | AttackTo(black,to);
attacked_piece=PieceV[piece[to]];
col=opponent(Player);
swap_list[0] = attacked_piece;
attacked_piece = PieceV[piece[from]];
Remove(attacks,from);
dir=direction[to][from];
if(dir) attacks = RevealedAttacks(attacks,from,dir);

while (attacks)
{
if ((tmp = Pieces[col][pawn] & attacks)) sq=FirstOne(tmp);
else if ((tmp = Pieces[col][knight] & attacks)) sq=FirstOne(tmp);
else if ((tmp = Pieces[col][bishop] & attacks)) sq=FirstOne(tmp);
else if ((tmp = Pieces[col][rook] & attacks)) sq=FirstOne(tmp);
else if ((tmp = Pieces[col][queen] & attacks)) sq=FirstOne(tmp);
else if ((tmp = Pieces[col][king] & attacks)) sq=FirstOne(tmp);
else break;
swap_list[nc] = -swap_list[nc-1] + attacked_piece;
attacked_piece = PieceV[piece[sq]];
Remove(attacks,sq);
dir=direction[to][sq];
if(dir) attacks = RevealedAttacks(attacks,sq,dir);
nc++;
col=opponent(col);
}
while (--nc)
if (swap_list[nc] > -swap_list[nc-1]) swap_list[nc-1]=-swap_list[nc];
return swap_list[0];
}
This is from Crafty:
/*
*******************************************************************************
* *
* Swap() is used to analyze capture moves to see whether or not they appear *
* to be profitable. the basic algorithm is extremely fast since it uses the*
* bitmaps to determine which squares are attacking the [target] square. *
* *
* the algorithm is quite simple. using the attack bitmaps, we enumerate all*
* the pieces that are attacking [target] for either side. then we simply *
* use the lowest piece (value) for the correct side to capture on [target]. *
* we continually "flip" sides taking the lowest piece each time. *
* *
* as a piece is used, if it is a sliding piece (pawn, bishop, rook or queen)*
* we "peek" behind it to see if it is attacked by a sliding piece in the *
* direction away from the piece being captured. if so, and that sliding *
* piece moves in this direction, then it is added to the list of attackers *
* since its attack has been "uncovered" by moving the capturing piece. *
* *
*******************************************************************************
*/
int Swap(TREE * RESTRICT tree, int source, int target, int wtm)
{
register BITBOARD attacks;
register int attacked_piece;
register int square,
direction;
register int color,
nc = 1;
int swap_list[32];
/*
************************************************************
* *
* determine which squares attack for each side. *
* initialize by placing the piece on first in *
* the list as it is being captured to start things off. *
* *
************************************************************
*/
attacks = AttacksTo(tree, target);
attacked_piece = p_values[PcOnSq(target) + 7];
/*
************************************************************
* *
* the first piece to capture on is the piece *
* standing on . *
* *
************************************************************
*/
color = Flip(wtm);
swap_list[0] = attacked_piece;
attacked_piece = p_values[PcOnSq(source) + 7];
Clear(source, attacks);
direction = directions[target][source];
if (direction)
attacks = SwapXray(tree, attacks, source, direction);
/*
************************************************************
* *
* now pick out the least valuable piece for the correct *
* side that is bearing on . as we find one, we *
* call SwapXray() to add the piece behind this piece *
* that is indirectly bearing on (if any). *
* *
************************************************************
*/
while (attacks) {
if (color) {
if (WhitePawns & attacks)
square = FirstOne(WhitePawns & attacks);
else if (WhiteKnights & attacks)
square = FirstOne(WhiteKnights & attacks);
else if (WhiteBishops & attacks)
square = FirstOne(WhiteBishops & attacks);
else if (WhiteRooks & attacks)
square = FirstOne(WhiteRooks & attacks);
else if (WhiteQueens & attacks)
square = FirstOne(WhiteQueens & attacks);
else if (WhiteKing & attacks)
square = WhiteKingSQ;
else
break;
} else {
if (BlackPawns & attacks)
square = FirstOne(BlackPawns & attacks);
else if (BlackKnights & attacks)
square = FirstOne(BlackKnights & attacks);
else if (BlackBishops & attacks)
square = FirstOne(BlackBishops & attacks);
else if (BlackRooks & attacks)
square = FirstOne(BlackRooks & attacks);
else if (BlackQueens & attacks)
square = FirstOne(BlackQueens & attacks);
else if (BlackKing & attacks)
square = BlackKingSQ;
else
break;
}
/*
**********************************************
* *
* located the least valuable piece bearing on *
* . remove it from the list and then *
* find out if a sliding piece behind it attacks *
* through this piece. *
* *
**********************************************
*/
swap_list[nc] = -swap_list[nc - 1] + attacked_piece;
attacked_piece = p_values[PcOnSq(square) + 7];
Clear(square, attacks);
direction = directions[target][square];
if (direction)
attacks = SwapXray(tree, attacks, square, direction);
nc++;
color = Flip(color);
}
/*
************************************************************
* *
* starting at the end of the sequence of values, use a *
* "minimax" like procedure to decide where the captures *
* will stop. *
* *
************************************************************
*/
while (--nc)
if (swap_list[nc] > -swap_list[nc - 1])
swap_list[nc - 1] = -swap_list[nc];
return (swap_list[0]);
}
As you can see, the algorithm is identical. However the implementation is different. Dr. Hyatt calls it a cut and paste. I call it an implementation of an algorithm. You cannot copyright an algorithm. You have to have a patent for that.
About 30% of the DanChess code is similar to the crafty code. DanChess plays 300 Elo weaker than crafty.
So that is the story. I think it should serve as a severe warning about using ideas from crafty. Basically, I now think it is a very bad idea to read the crafty code.



my ftp site {remove http:// unless you like error messages}

[Matthias Gemuh]

Geschrieben von: / Posted by: Matthias Gemuh at 15 February 2004 18:54:05:
Als Antwort auf: / In reply to: Re: WBEC Ridderkerk new results. geschrieben von: / posted by: Dann Corbit at 15 February 2004 18:01:45:

Dr. Hyatt calls it a cut and paste.
I call it an implementation of an algorithm.

Some people call it search and replace.
This can be done without even understanding the algorithms.

I think the fairest way to write an engine is to start from scratch or
skeletons as on Bruce Moreland's page.
Perfectly acceptable is also to start with Tscp or weaker (see Trace).
To start with a "search and replace" of Pepito, Amy, etc is not OK.
BTW, I think you have revealed more of Danchess source code than
pleases the author, who generally ignores clone questions.

/Matthias.


BigLion + Taktix

[Uri Blass]

Geschrieben von: / Posted by: Uri Blass at 15 February 2004 19:04:23:
Als Antwort auf: / In reply to: Re: WBEC Ridderkerk new results. geschrieben von: / posted by: Matthias Gemuh at 15 February 2004 18:54:05:

Dr. Hyatt calls it a cut and paste.
I call it an implementation of an algorithm.

Some people call it search and replace.
This can be done without even understanding the algorithms.

I think the fairest way to write an engine is to start from scratch or
skeletons as on Bruce Moreland's page.
Perfectly acceptable is also to start with Tscp or weaker (see Trace).
To start with a "search and replace" of Pepito, Amy, etc is not OK.
BTW, I think you have revealed more of Danchess source code than
pleases the author, who generally ignores clone questions.

/Matthias.

Based on the posted code it seems to me that the programmer understood the algorithm.
I may be wrong but I also did not get the impression that he started with crafty and changed it.

I do not think that the author has problems with Dann's post.
Uri

[Matthias Gemuh]

Geschrieben von: / Posted by: Matthias Gemuh at 15 February 2004 19:30:45:
Als Antwort auf: / In reply to: Re: WBEC Ridderkerk new results. geschrieben von: / posted by: Uri Blass at 15 February 2004 19:04:23:



Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

/Matthias.


BigLion + Taktix

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 19:32:24:
Als Antwort auf: / In reply to: Re: WBEC Ridderkerk new results. geschrieben von: / posted by: Uri Blass at 15 February 2004 19:04:23:

Dr. Hyatt calls it a cut and paste.
I call it an implementation of an algorithm.

Some people call it search and replace.
This can be done without even understanding the algorithms.

I think the fairest way to write an engine is to start from scratch or
skeletons as on Bruce Moreland's page.
Perfectly acceptable is also to start with Tscp or weaker (see Trace).
To start with a "search and replace" of Pepito, Amy, etc is not OK.
BTW, I think you have revealed more of Danchess source code than
pleases the author, who generally ignores clone questions.

/Matthias.

Based on the posted code it seems to me that the programmer understood the algorithm.
I may be wrong but I also did not get the impression that he started with crafty and changed it.
I do not think that the author has problems with Dann's post.

That is what Dan did in my view.
That is no more acceptable than using crafty without permission. In fact, Tom is also very definite about the use of his code.
Or with any other program.
Your understanding is the same as mine. He cleary had to make many adaptations. For instance, his data structres are not the same.
I will admit that it is a grey area.
Matthias has a very good point. I should have asked his permission.



my ftp site {remove http:// unless you like error messages}

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 19:35:05:
Als Antwort auf: / In reply to: Dann as judge is too liberal. geschrieben von: / posted by: Matthias Gemuh at 15 February 2004 19:30:45:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

That is no more acceptable. Not at all. Why should a program of lower strength lose all of it's rights. In fact, I think your assertion is ridiculous.
If you use the crafty algorithms then you have done nothing wrong. You cannot copyright an algorithm. You can only patent it.
That means it is not illegal or immoral to use someone else's idea if you use the algorithm and not the code.



my ftp site {remove http:// unless you like error messages}

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 19:44:11:
Als Antwort auf: / In reply to: Re: WBEC Ridderkerk new results. geschrieben von: / posted by: Dann Corbit at 15 February 2004 19:32:24:

Dr. Hyatt calls it a cut and paste.
I call it an implementation of an algorithm.

Some people call it search and replace.
This can be done without even understanding the algorithms.

I think the fairest way to write an engine is to start from scratch or
skeletons as on Bruce Moreland's page.
Perfectly acceptable is also to start with Tscp or weaker (see Trace).

That is what Dan did in my view.
That is no more acceptable than using crafty without permission. In fact, Tom is also very definite about the use of his code.

This is from the TSCP readme file:
" LEGAL STUFF
According to copyright law, you are not allowed to distribute copies of TSCP
or anything that's derived from TSCP without my authorization.
Version 1.4 of TSCP is the first version to include copyright notices, but
previous versions are also protected under law. If you are distributing an
earlier version of TSCP or a derivative work without my authorization, you are
acting illegally.
For more information about copyrights, visit this web page:
http://lcweb.loc.gov/copyright/"
[snip]



my ftp site {remove http:// unless you like error messages}

[Uri Blass]

Geschrieben von: / Posted by: Uri Blass at 15 February 2004 19:48:01:
Als Antwort auf: / In reply to: Dann as judge is too liberal. geschrieben von: / posted by: Matthias Gemuh at 15 February 2004 19:30:45:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

/Matthias.

I understood from Dann's post that 70% of Danchess code is not similiar to crafty.
Uri

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 19:57:04:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Uri Blass at 15 February 2004 19:48:01:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

/Matthias.

I understood from Dann's post that 70% of Danchess code is not similiar to crafty.

Many parts are completely different. All of this is neither here nor there. He is doing a complete rewrite to 0x88. He has some ideas that are fresh and interesting and found nowhere else.
Amazingly, he already has a working version. It's just as strong as the old one.
Daniel Shawul is a very smart person. Expect big things from him. He does not have a lot of experience. But that will gather like a snowball down a hill.



my ftp site {remove http:// unless you like error messages}

[Matthias Gemuh]

Geschrieben von: / Posted by: Matthias Gemuh at 15 February 2004 20:11:22:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Dann Corbit at 15 February 2004 19:35:05:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

That is no more acceptable. Not at all. Why should a program of lower strength lose all of it's rights. In fact, I think your assertion is ridiculous.
If you use the crafty algorithms then you have done nothing wrong. You cannot copyright an algorithm. You can only patent it.
That means it is not illegal or immoral to use someone else's idea if you use the algorithm and not the code.

We disagree on these issues like day and night.
Weak programs generally lack ideas worth protecting.
Starting with such is OK because all key routines eventually get replaced,
and final strength depends then on author's work.
If I start by using modified key routines of a strong program,
then that is wrong.
/Matthias.



BigLion + Taktix

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 20:14:47:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Matthias Gemuh at 15 February 2004 20:11:22:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

That is no more acceptable. Not at all. Why should a program of lower strength lose all of it's rights. In fact, I think your assertion is ridiculous.
If you use the crafty algorithms then you have done nothing wrong. You cannot copyright an algorithm. You can only patent it.
That means it is not illegal or immoral to use someone else's idea if you use the algorithm and not the code.

We disagree on these issues like day and night.
Weak programs generally lack ideas worth protecting.
Starting with such is OK because all key routines eventually get replaced,
and final strength depends then on author's work.
If I start by using modified key routines of a strong program,
then that is wrong.

They have the same legal protection as the strongest ideas.
So it is okay to steal for a little while, because you will eventually replace the things you took?
I agree that we disagree totally.



my ftp site {remove http:// unless you like error messages}

[Marcus Prewarski]

Geschrieben von: / Posted by: Marcus Prewarski at 15 February 2004 20:28:15:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Matthias Gemuh at 15 February 2004 20:11:22:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

That is no more acceptable. Not at all. Why should a program of lower strength lose all of it's rights. In fact, I think your assertion is ridiculous.
If you use the crafty algorithms then you have done nothing wrong. You cannot copyright an algorithm. You can only patent it.
That means it is not illegal or immoral to use someone else's idea if you use the algorithm and not the code.

We disagree on these issues like day and night.
Weak programs generally lack ideas worth protecting.
Starting with such is OK because all key routines eventually get replaced,
and final strength depends then on author's work.
If I start by using modified key routines of a strong program,
then that is wrong.
/Matthias.

I have to agree with Dan on this one, I don't see why it is okay to copy code from a weak engine. For one thing who is to say what is weak and what is not. Also if the ideas are so easy why copy. Maybe I only care about this because my engine is weak, but I find it a little irritating when a new engine appears that is really a TSCP clone with a few modifications and someone claims it as their own. I don't spend a lot of time working on my engine but it wasn't until version 0.8 of my engine until I could say it was really stronger than TSCP.
-Marcus

[Roger Brown]

Geschrieben von: / Posted by: Roger Brown at 15 February 2004 20:47:02:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Matthias Gemuh at 15 February 2004 20:11:22:

Weak programs generally lack ideas worth protecting.
Starting with such is OK because all key routines eventually get replaced,
and final strength depends then on author's work.
If I start by using modified key routines of a strong program,
then that is wrong.
/Matthias.

Matthias,
I am simply assuming that you are being misunderstood here and that what you are saying is not what you mean to say.

It cannot be that you are saying that because my engine is not capable of beating a patzer it is alright to take critical ideas from it but if it is Cunning 2000 with an Elo of 3500 then leave it alone?
The intellectual property of a patzer is to be respected as much that of a GM strength world beater.
Going down this road is dangerous as the next question then becomes - who judges what is weak?
Let us tread carefully here and respect all ideas, weak and strong.
Later.

[Matthias Gemuh]

Geschrieben von: / Posted by: Matthias Gemuh at 15 February 2004 20:48:54:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Marcus Prewarski at 15 February 2004 20:28:15:

We disagree on these issues like day and night.
Weak programs generally lack ideas worth protecting.
Starting with such is OK because all key routines eventually get replaced,
and final strength depends then on author's work.
If I start by using modified key routines of a strong program,
then that is wrong.
/Matthias.

I have to agree with Dan on this one, I don't see why it is okay to copy code from a weak engine. For one thing who is to say what is weak and what is not. Also if the ideas are so easy why copy. Maybe I only care about this because my engine is weak, but I find it a little irritating when a new engine appears that is really a TSCP clone with a few modifications and someone claims it as their own. I don't spend a lot of time working on my engine but it wasn't until version 0.8 of my engine until I could say it was really stronger than TSCP.
-Marcus

Starting with a weak engine is just a compromise for people who want
to see some action of their own routines really fast by temporally
using borrowed helper routines.
I started BigLion from scratch and for a long time his 4 sparring
partners were Tscp, Monik, LarsenVB and Oswald.
After more than 2 or 3 years, BigLion is still struggling to get basic
algorithms right.
So my posting does not mean to encourage any form of "copy and paste".
/Matthias



BigLion + Taktix

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 21:02:52:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Matthias Gemuh at 15 February 2004 20:48:54:

We disagree on these issues like day and night.
Weak programs generally lack ideas worth protecting.
Starting with such is OK because all key routines eventually get replaced,
and final strength depends then on author's work.
If I start by using modified key routines of a strong program,
then that is wrong.
/Matthias.

I have to agree with Dan on this one, I don't see why it is okay to copy code from a weak engine. For one thing who is to say what is weak and what is not. Also if the ideas are so easy why copy. Maybe I only care about this because my engine is weak, but I find it a little irritating when a new engine appears that is really a TSCP clone with a few modifications and someone claims it as their own. I don't spend a lot of time working on my engine but it wasn't until version 0.8 of my engine until I could say it was really stronger than TSCP.
-Marcus

Starting with a weak engine is just a compromise for people who want
to see some action of their own routines really fast by temporally
using borrowed helper routines.
I started BigLion from scratch and for a long time his 4 sparring
partners were Tscp, Monik, LarsenVB and Oswald.
After more than 2 or 3 years, BigLion is still struggling to get basic
algorithms right.
So my posting does not mean to encourage any form of "copy and paste".
/Matthias

I think if you use someone else's algorithm as a benchmark reference and do not actually use it in your released code, than you have committed no wrongdoing.
It appears that I simply misunderstood your intended use.
Of course, I am not a lawyer, so my opions have very little value in this arena.
The subject of legal software protection is something that I have studied extensively, however.


my ftp site {remove http:// unless you like error messages}

[Matthias Gemuh]

Geschrieben von: / Posted by: Matthias Gemuh at 15 February 2004 21:31:05:
Als Antwort auf: / In reply to: I beg your pardon? geschrieben von: / posted by: Roger Brown at 15 February 2004 20:47:02:

Weak programs generally lack ideas worth protecting.
Starting with such is OK because all key routines eventually get replaced,
and final strength depends then on author's work.
If I start by using modified key routines of a strong program,
then that is wrong.
/Matthias.

Matthias,
I am simply assuming that you are being misunderstood here and that what you are saying is not what you mean to say.

It cannot be that you are saying that because my engine is not capable of beating a patzer it is alright to take critical ideas from it but if it is Cunning 2000 with an Elo of 3500 then leave it alone?
The intellectual property of a patzer is to be respected as much that of a GM strength world beater.
Going down this road is dangerous as the next question then becomes - who judges what is weak?
Let us tread carefully here and respect all ideas, weak and strong.
Later.

Starting with a weak engine is just a compromise for people who want
to see some action of their own routines really fast by temporally
using borrowed helper routines.

Hi Roger,
The key words are "compromise" and "temporally" .
I like to quote Trace as beautifully acceptable example.
I can bet that Trace was "clean" of Tscp pretty fast.
/Matthias


BigLion + Taktix

[Sune Fischer]

Geschrieben von: / Posted by: Sune Fischer at 15 February 2004 21:37:22:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Dann Corbit at 15 February 2004 19:57:04:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

/Matthias.

I understood from Dann's post that 70% of Danchess code is not similiar to crafty.

Many parts are completely different. All of this is neither here nor there. He is doing a complete rewrite to 0x88. He has some ideas that are fresh and interesting and found nowhere else.
Amazingly, he already has a working version. It's just as strong as the old one.
Daniel Shawul is a very smart person. Expect big things from him. He does not have a lot of experience. But that will gather like a snowball down a hill.

Sounds like he fits the classic profile of a chess programmer, full of fresh ideas and lots of energy
The question is if these on-the-paper ideas actually work.
Only time will tell.
The are already many smart people in chess programming, the question is
not if he is 'smart' but if he is 'smarter'
-S.

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 15 February 2004 21:40:46:
Als Antwort auf: / In reply to: Re: Dann as judge is too liberal. geschrieben von: / posted by: Sune Fischer at 15 February 2004 21:37:22:

Anyway, I think Dann's judgement is extremely liberal.
Dr. Hyatt certainly knows why he speaks of cut an paste.
I wouldn't go that far, but it sure looks like search and replace.
Would be OK if starting point for that is Nero.
I can clone Crafty and end up with only 10% total code size by
using similar data structures and mimmicking Crafty's 10 key routines.

/Matthias.

I understood from Dann's post that 70% of Danchess code is not similiar to crafty.

Many parts are completely different. All of this is neither here nor there. He is doing a complete rewrite to 0x88. He has some ideas that are fresh and interesting and found nowhere else.
Amazingly, he already has a working version. It's just as strong as the old one.
Daniel Shawul is a very smart person. Expect big things from him. He does not have a lot of experience. But that will gather like a snowball down a hill.

Sounds like he fits the classic profile of a chess programmer, full of fresh ideas and lots of energy
The question is if these on-the-paper ideas actually work.
Only time will tell.
The are already many smart people in chess programming, the question is
not if he is 'smart' but if he is 'smarter'

Probably determination is even more important than that. Some of the smartest chess programmers simply got tired of it.



my ftp site {remove http:// unless you like error messages}