From: Reijer Grimbergen ETL GO JP> Date: 23 jan 1996 Subject: Summary on game notation? (summary, but still rather long) Since I started the issue on Shogi game notation and have thus far kept quiet and let the discussion run its course, it may be a good idea for me to try and summarize it. As could be expected, the discussion has sometimes digressed to the related Japanese-Western culture issue. The differences between Japanese and Western culture can be discussed at great length, as can be the need to resolve them or not and if so, how. It is my personal opninion that shogi-l is not the platform to discuss these issues, and newsgroups like soc.culture.japan are a better forum. Again, that is my personal opinion and anyone is free to take these issues to Shogi-L. I will just not get involved in these discussions. As for game notation, it seems that there are two major decisions involved: * First, whether to use Kanji or not. Most people seem to agree that using Kanji is not very convenient. It is rather difficult to publish magazines with kanji without using special Japanese (or Chinese) software. Also, most western people will not use Kanji to write their games, so there would always be the need for translating actual game scores (kifu) to kanji. Of course, if kanji would be convenient for everybody the discussion we have had would be irrelevant and we would all just use the Japanese notation system (even though there are some variants here as well). * Second, should an international notation system stay as close as possible to the original Japanese system or be a chess-type system? Again, this can be broken down into several parts: 1) How to describe the pieces? Again, kanji seems to be inconvenient. However, someone mentioned the use of graphic symbols in most international chess magazines. Although shogi will not be as international a game as chess for quite some time, this might imply that kanji will return to game notation in the future. If not Kanji, what letters to use to describe the pieces? Again, most people seem to agree that the Japanese names for the pieces are inconvenient, since there are four pieces starting with K (Kaku, Kyo, Kei and Kin). Using two letters for a piece is not desirable. I have typed in many games in the past and I can tell from experience that extra typing work should be avoided at all costs. Furthermore, in most non-Japanese magazines and most of the non-Japanese or bilangual books the English names are already used for the pieces. Of course, some details could still be discussed, like using D (for Dragon) instead of +R and H (for Horse) instead of +B. Again, this would save typing, but there is no convenient way to use a single symbol for +N or +L. I think these details are not very important and might even be included as an option into a description of an international notation system. 2) How to decribe the squares? I think that this is indeed a complicated issue and the discussion on Shogi-L seems to reflect that. It is my personal opinion that Fairbarn (and before him maybe Leggett, even though I have never seen his book) made a serious mistake by using a chess type notation for Shogi games. Translating Shogi games to and from Japanese notation to what I shall refer to as "Fairbarn notation" is almost impossible without using a Shogi board. The system I use for notating my games is a direct westernization of the Japanese notation. For example 1.76P 34P 2.66P 84P instead of 1.P7f P3d 2.P6f P8d. Only if multiple pieces can go to one square I use the starting square as an extra instead of the kanji used in Japanese notation. For example 58G49 instead of G4i-5h or 58Gm (with m for "migi"= right). However, I would not recommend to the Renmei to use such a system. Again, Fairbarn-notation is used in most non-Japanese publications and by almost all non-Japanese Shogi players. Therefore it has already become a standard and it is hard to change that, even if there are better systems available. Most experts agree that VHS was the worst of the three video systems introduced in the seventies, but everybody knows what has happened to V2000 and Betamax. Of course even in Fairbarn notation there is room for discussion. For example, the "-" between piece and square seems to have disappeared. P-7f as in Fairbarn is now just P7f. Also, should we use G4i-5h, G4i5h or G45h? Again, I think these are minor details and there is no point in discussing them yet. Conclusion: 1.P7f P3d 2.P6f P8d 3.S7h P5d 4.R6h G6a-5b may not be the best way to describe a game, but it is not a bad way to describe a game and is already some kind of standard. Reijer -- Dr. Reijer Grimbergen Electrotechnical Laboratory Palcious Tsukuba 302 1-1-4 Umezono 1-24-8 Ninomiya Tsukuba-shi, Ibaraki-ken Tsukuba-shi, Ibaraki-ken 305 JAPAN 305 JAPAN E-mail: grimberg etl go jp Tel: 0298-59-1606 WWW: http://www.etl.go.jp:8080/etl/suiron/~grimberg Tel: +81-(0)298-54-5080 extension: 67431 Fax: +81-(0)298-58-5918