Notice there is no notation symbol for stalemate

[UCLARodent]

I would like to propose that we use the percentage sign to denote stalemate.

23.Kg8 h7+
24.Kh8 Kh6%

[Crume]

Out of curiosity, why? Stalemate would be obvious from playing through the game?

[wbport]

@Crume

Out of curiosity, why? Stalemate would be obvious from playing through the game?

It wouldn’t be necessary to play through the game.

BTW, I think the colors were reversed in that example. Perhaps something like 22. … Kg8 23. h7+ Kh8 24. Kh6%

[Crume]

@wbport

@Crume

Out of curiosity, why? Stalemate would be obvious from playing through the game?

It wouldn’t be necessary to play through the game.

If all you want to know is the end result (win, lose, or draw), what added value is provided by knowing the draw was a stalemate? What added value is knowing the draw was an agreement, 3-repetition or 50-move draw?

I don’t learn, or take any enjoyment, from the result. That comes from going through the game score(s). YMMV.

[GeneM]

@UCLARodent

I would like to propose that we use the percentage sign to denote stalemate.

23.Kg8 h7+
24.Kh8 Kh6%

I guess most players today rely on the absence of an ending ‘#’ to indicate “Resigns”? Obviously “1-0” does not distinguish.

Most painfully missing from chess notation, as provided by ChessBase Mega and TWIC .PGN, is clock times.
All those electronic DGT boards and clocks used in major tournaments, and of course in WCChamp title matches, are capturing clock data. But .PGN and subsequent publications of the games is simply losing and wasting the clock info.
The human side of the battle is revealed in greater depth by clock information.

[ChessSpawn]

@UCLARodent

I would like to propose that we use the percentage sign to denote stalemate.

23.Kg8 h7+
24.Kh8 Kh6%

Good idea! And %%% could be draw by repetition.

[wbport]

@Crume

If all you want to know is the end result (win, lose, or draw), what added value is provided by knowing the draw was a stalemate? What added value is knowing the draw was an agreement, 3-repetition or 50-move draw?

Someone might think, “Hey, that game ended in a stalemate! I’m going to play through it to see how that happened.”

If someone doesn’t have the time or interest in going through several hundred games, a little more information might be helpful.

[Smythe_Dakota]

Well, let’s see. A pound # sign is sometimes used for checkmate, and a stalemate is half a checkmate. The pound # sign is composed of four straight lines, so the stalemate symbol should be composed of two. How about “=”? No, that symbol is already taken (at least if the game is annotated). How about “+”? Nope, that means check. That leaves “||”.

@ChessSpawn

… And %%% could be draw by repetition.

And how about %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% for the 50-move rule?

Bill Smythe

[GeneM]

@Smythe Dakota

@ChessSpawn

… And %%% could be draw by repetition.

And how about %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% for the 50-move rule?

Ha! Reminds me of Luke Surl’s chess start position cartoons in the 2010/Dec Chess Life (note to Crume: on page 8.).

[ChessSpawn]

@GeneM

@Smythe Dakota

@ChessSpawn

… And %%% could be draw by repetition.

And how about %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% for the 50-move rule?

Ha! Reminds me of Luke Surl’s chess start position cartoons in the 2010/Dec Chess Life (note to Crume: on page 8.).

Pretty clear, eh?!

[JWhist]

i wouldn’t recommend % since pgn uses that to distinguish clock times

[LaughingVulcan]

It depends on which variation of the PGN standard is followed (1994 or 2001,) but either way “%” is involved with command sequences to be ignored by PGN parsers, but is usable for any other kind of command or specialized parsing (which includes clock times as one of the few enunciated commands in the 2001 standard.)

@http://www.drpribut.com/sports/final.txt

3.2.3 Commands

Command names start immediately after the [%.
They consist of one or more alphanumeric characters
They are terminated by the first space character encountered.

We are not proposing any fixed canonical list of commands. The idea is to describe a generalised command syntax, not to impose a particular set of
commands.

We are proposing a set of commands to deal with time handling. “clk” in the example above is the proposed command name for recording move times. See section
4 for the detailed specification of this and other time related commands.

So it can be used for other things besides clock command and control. A truly compliant PGN parser should probably ignore any “%” it can’t parse - but there is also a recommendation that move displays should bypass the display of command strings… so it’s hard to say.

What is wrong with using “st” for stalemate in a written scoresheet and appending the comment {stalemate} on the last move in a PGN file, before the terminator?

[Crume]

@LaughingVulcan

What is wrong with using “st” for stalemate in a written scoresheet and appending the comment {stalemate} on the last move in a PGN file, before the terminator?

Because it’s a draw like any other draw. Any master level game will be an agreed draw before ever getting to stalemate. No reason for an additional symbol or notation other than “1/2-1/2”.

[GeneM]

@LaughingVulcan

It depends on which variation of the PGN standard is followed (1994 or 2001,) but either way “%” is involved with command sequences to be ignored by PGN parsers, but is usable for any other kind of command or specialized parsing (which includes clock times as one of the few enunciated commands in the 2001 standard.)

A general problem with the .PGN specification is that it fails to require a NewLine at the front (and back) of each non-move element, such as in front and back of each {comment}.
This failure makes many .PGN files humanly implausible to read.

.PGN should satisfy the needs of both machine and human. It easily could have done so.

[LaughingVulcan]

Hmm… I thought in the PGN standard that CRs or CR/LF were completely ignored in parsing; only reserved characters mattered. In the 1994 standard, it is noted that local convention may require different forms of newline characters, and that newlines may be considered a “white space” for the purposes of tag separation (along with spaces.) The only limitation I can read is that a string line must consist of 255 characters or less, though conventions of the day might restrict that to 80.

The 1994 standard also allows the use of a semicolon (“;”) to make a comment to end-of-line without the use of the brackets.

The proposed draft 2001 standard from Dr. Pribut mentions new lines not at all.

(All of that depends on that I’m parsing the standards text correctly.)

That is, you could have a PGN file like this:

[Event "Sample"]
[Site "?"]
[Date "2010.12.17"]
[Round "?"]
[White "Goodpawn, Garry"]
[Black "Badbishop, Boris"]
[Result "*"]
[WhiteElo "3500"]
[BlackElo "100"]
[PlyCount "5"]

1. e4 {White looks good here.} e5 {Perhaps Black should have tried the Siclian.} (1... c5) 2. Nf3 {Darren loves open games!} (2. Bc4 {The Bishop's Opening seems to be the coming thing, yet again.}) 2... Nc6 {I have you now, Obi-Wan!} 3. Bb5 *

[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "1-0"]
[PlyCount "4"]

1. f4 e5 2. g4 Qh4# {Although a checkmate one could simply comment "st" were it a stalemate.} 1-0

Or this…

[Event "Sample"]
[Site "?"]
[Date "2010.12.17"]
[Round "?"]
[White "Goodpawn, Garry"]
[Black "Badbishop, Boris"]
[Result "*"]
[WhiteElo "3500"]
[BlackElo "100"]
[PlyCount "5"]

1. e4 
{White looks good here.}
 e5 
{Perhaps Black should have tried the Siclian.} 
(1... c5) 
2. Nf3 
{Darren loves open games!} 
(2. Bc4 
{The Bishop's Opening seems to be the coming thing, yet again.})
2... Nc6 
{I have you now, Obi-Wan!}
3. Bb5 *

[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "1-0"]
[PlyCount "4"]

1. f4 e5 
2. g4 Qh4# 
{Although a checkmate one could simply comment "st" were it a stalemate.} 1-0

Or even this…

[Event "Sample"] [Site "?"] [Date "2010.12.17"] [Round "?"] [White "Goodpawn, Garry"] [Black "Badbishop, Boris"] [Result "*"] [WhiteElo "3500"] [BlackElo "100"] [PlyCount "5"] 
1. e4 {White looks good here.} e5 {Perhaps Black should have tried the Siclian.} (1... c5) 2. Nf3 {Darren loves open games!} (2. Bc4 {The Bishop's Opening seems to be the coming thing, yet again.}) 2... Nc6 {I have you now, Obi-Wan!} 3. Bb5 *

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "?"] [Black "?"] [Result "1-0"] [PlyCount "4"] 1. f4 e5 2. g4 Qh4# {Although a checkmate one could simply comment "st" were it a stalemate.} 1-0

And they should all parse the same.

Of those three, with the first game ChessBase only gives fits when I paste the third version back into a new game file. With the second game, in the second variation ChessBase wants to make the ???ed game tags a pre-game comment but posts the moves and score correctly.

But if I’m right, the problem is with those who write display interpreters, perhaps parsers, and file-writing from interpreters, as opposed to the standard itself.

Though I could be way off. I’ve read the standard a few times but have never had to program with the PGN standard (aside from some aborted attempts to write parsers.)

[adp]

Chess notation is woefully inadequate with respect to outcome, even, sometimes, when you do play through the game. Ever see 1-0 or 0-1 with position approximately even? Don’t you ever wonder whether the loser resigned or ran out of time? Also, 1/2-1/2 is redundant. You only need 1/2 to indicate a draw.

[KeithAmmann]

Not necessarily. 1/2-0 or 0-1/2 could represent some form of forfeit, couldn’t it?

[tmagchesspgh]

I would opt for a simple symbol like ~ to represent stalemate. The symbol is on the same line of the keyboard as #. I do not see the ~ symbol used in either of the ECO/Informant or NIC code systems.