Modified Hosking Notation

 If you've had any interactions with me over the years, you'll probably think this post is hypocritical.  I've been pretty adamantly against adding new forms of shogi notation, because every time it's changed, it splinters English players' ability to read a little bit more.  That being said, what I'm detailing here is not a new notation format, but rather an update to formatting an existing one, so I hope you'll forgive me.

The English shogi community on the internet has pretty much settled on using some form of Hosking notation - named for and created by British shogi champion Tony Hosking - in day to day communication.  When I was a new player, it was primarily Hodges notation that was used, so I think this is a good direction to give us parity with Japanese shogi materials.  Kitao-Kawasaki Notation - created by ladies professional Madoka Kitao and shogi promoter Hidetchi - attempts to bridge the gap even farther by replacing the pieces in the notation with the kanji for the pieces.  I like this notation system.  It's perfectly reasonable, even if it does kind of feel like the kind of chess notation with pictures of the pieces instead of letters.

Kitao-Kawasaki notation, in my opinion, has two big downsides:

1. It's less legible.  I understand that players will have to learn those kanji anyway to properly read shogi books, but for people who don't have any Japanese knowledge, recognizing kanji as a pictograph on a board and recognizing it in text are actually two different skills.  This isn't terribly hard to overcome, but it is a barrier nonetheless.

2. It's much more difficult to type.  Needing to constantly switch my IME back and forth to type 歩-76 歩-34 歩-26 歩-44 銀-46 飛-42 takes considerably more keystrokes than P76 P34 P26 P44 S46 R42.  Even worse, the amount of keystrokes increases if your IME refuses to acknowledge the shogi reading of the kanji on your first go.  Typing 飛 as hisya and then backspacing the 車 so it doesn't try to write 日 or 火 makes it more of a hassle than typing a single R.

These downsides have rendered Kitao-Kawasaki notation basically completely unused outside of books published specifically by Nekomado. The speed of writing notation with Latin letters is a clear benefit, which is why you'll see it used on discord, reddit, and between anglophone players on shogi websites.  

That's not to say Nekomado's notation is the only one with downsides.  There are a couple of big ones with Hosking Notation, too:

1. As written in Shogi Foundation books, Hosking notation is a nightmare to read for people with reading disabilities.  The lack of any tangible separators for moves, and the format of just dumping all the moves within a paragraph makes it extremely difficult for someone with dyslexia, for example, to parse.   

 2. It uses the Chess style of counting moves.  Hosking counts one move as being a pair of moves by sente and gote, which feels a bit antithetical to the goal of bringing shogi notation closer to the Japanese.  Nowadays, pretty much all shogi players have abandoned the concept of ordering moves in pairs.

It's for all these reasons that I'd like to introduce the Modified Hosking Notation, which essentially aims to merge the best parts of Kitao-Kawasaki Notation with the best parts of Hosking Notation, unifying them into one notation system.  This will be the system I personally use to write shogi materials, and I would be pleased if other people decided to use it as well.

As with the previous example, in Modified Hosking, a fourth file rook opening would begin ☗P-76 ☖P-34 ☗P-26 ☖P-44 ☗S-46 ☖R-42.  While I would never expect players in a discord server to take the time to use the shogi piece symbols before moves, it's my personal belief that they're incredible useful for making moves more readable.  Move numbers could be placed before the player symbols if your commentary values being able to call back to specific moves.

The other major thing that Hodges and Kitao-Kawasaki do that I think is important is having move type identifiers for every move.  - for move, x for capture, and * for drop (we reluctantly are forced to accept Hosking's ' as a substitute for drops because of reddit's and discord's markup language).  The move type indicator not only makes the moves more legible, but reflects how we describe the moves with words.  P-76 is often read as "Pawn to Seven Six," and Sx55 as "Silver takes Five Five." 

There are two topics that I believe will come down to each individual writer's tastes in notation.

The first is whether to omit a square upon a recapture - that is, when a piece captures on the same square that the opponent moved to last turn.  In practice this leads to writing a rook pawn trade as ☗P-24 ☖Px ☗Rx ☖P*23 ☗R-28.  My personal opinion is that not only is this an elegant way to translate ☖同歩, it actually helps with move sequence recognition, because you already know where the last move took place.  It even reflects how I've heard players say the moves out loud: "Pawn to 24, Pawn takes, Rook takes."

The other topic is how to render the promoted pieces.  The Dragon, Horse, and Tokin (promoted pawn) all have unique names that are mirrored in English.  As a result, many players - myself included - write these as D, H, and T within notation.  So this leads to situations such as ☗Bx11+ and then ☗Hx21.  The alternative is using the promoted symbol for them, such as +R, +B, and +P.  I see no real problem with this, because it does create less ambiguity as to which piece it was before promoting.  However, my personal opinion is that we should, again, have notation be a reflection of how we speak.  For that reason, I intend to continue using D, H, and T, while the other pieces are +S, +N, and +L simply because we call those pieces Promoted Silver, Promoted Knight, and Promoted Lance anyway.

This is my recommendation for how to write shogi books and articles in English.  It should be a given that when discussing the game in casual communication, you can and probably should take formatting shortcuts.  As always, I also love to hear others' opinions, because I think it's very important that community decisions are done by the community and not by a single figurehead.

p.s. - As an aside, it's very important to note that ☗ and ☖ should be observed as a "filled in shogi piece" and a "hollow shogi piece" instead of "black" and "white."  I've seen concern about these symbols causing confusion in dark themes, because the "colors" swapped.  Japanese materials continue to use these as is even in dark themes.  Remember, shogi pieces don't actually have colors!