Check out Symmetric Chess, our featured variant for March, 2024.


[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Comments/Ratings for a Single Item

Earlier Reverse Order Later
Heterodox chess piece Unicode proposal[Subject Thread] [Add Response]
Garth Wallace wrote on Tue, Jul 5, 2016 08:36 PM UTC:

I'm currently working with the WFCC on a proposal to encode symbols used for chess variants and fairy chess problems in Unicode so that they can be used in plain text. Currently it consists of:

  • The 3 cardinal rotations of the orthodox piece symbols, in white and black
  • The 4 intermediate rotations of the knight symbol, in white and black
  • Neutral (half white, half black) symbols, upright and in the above rotations
  • The equihopper symbol and rotated equihopper symbol, in white, black, and neutral
  • A florette symbol (used in some places for the Rose) in white, black, and neutral
  • The knighted compounds (amazon, archbishop, marshall) in white and black
  • A joker/jester symbol in white and black
  • Shatranj fers and alfil/elephant symbols in white and black

The last three are probably most applicable to chess variants rather than fairy chess problems. I was hoping to be able to propose a wider set of variant pieces (I'd love to be able to justify a cannon and gryphon, for instance) but ultimately I haven't been able to find any examples of their use in running text.

The goal here is to allow these symbols to be used for things like figurine notation without having to rely on inline images or font-switching. Support for these symbols would come "for free" in any software that supports the appropriate version of Unicode (though displaying them would still require the user to have a font that contains them).

If anybody is interested, I can supply a PDF of the proposal, though it's still a draft (the exact ordering of characters is still in flux). I'd appreciate any input or feedback from the chess variants community.

(This is a followup to this old thread, posting as a new thread because the old one's title was a bit misleading)


🕸Fergus Duniho wrote on Wed, Jul 6, 2016 02:58 AM UTC:

This seems interesting. Please post a link to your PDF.

The last three are probably most applicable to chess variants rather than fairy chess problems. I was hoping to be able to propose a wider set of variant pieces (I'd love to be able to justify a cannon and gryphon, for instance) but ultimately I haven't been able to find any examples of their use in running text.

What is running text, and what is the significance of not appearing in running text? A cannon should be included if only because of its use in Chinese Chess.


Garth Wallace wrote on Wed, Jul 6, 2016 05:14 AM UTC:

Here's the link to the PDF on Google Drive. I hope that works. That's the original draft I submitted to WFCC. Since then the neutral compounds have been dropped (the WFCC reps didn't think there was a need, and I see their point) and the number of columns has been reduced to 6.

By "in running text", I mean in-line with written text, like in figurine notation, as opposed to appearing in an accompanying graphic like a diagram. This is important because Unicode is intended, first and foremost, for encoding text. The Unicode Consortium is unlikely to accept a symbol for encoding that doesn't have a good case for appearing in a text-like context (according to their criteria for encoding symbols, "symbols...whose identity must be able to be automatically interpreted and processed in ways that are similar to processes on text" or having "uses and usage patterns analogous to the notational systems used for writing"). Unicode has generally held that chess diagrams are inherently two-dimensional and therefore not "analogous to writing", even if fonts and typesetting have traditionally been used for them. To some extent I'm relying on the principle that, if a given symbol is used in chess diagrams, that symbol is used to refer to the same piece in figurine notation (this is essentially the definition of figurine notation). I'm not sure how far I can take that argument, though.

The problem with the cannon, I believe, is that xiangqi players in general don't even use a symbol for it, since xiangqi pieces are not sculpted figures but round tokens with the names of the pieces written on them. They just use the Chinese characters for cannon and catapult (ç‚® and ç ²), and those already exist in Unicode.


Garth Wallace wrote on Thu, Jul 21, 2016 09:30 PM UTC:

Have you had a chance to take a look at the proposal?


Garth Wallace wrote on Wed, Aug 10, 2016 01:42 AM UTC:

Here is an updated draft, incorporating a revised piece arrangement and some (hopefully) clarified language: https://drive.google.com/open?id=0B_uaELFCyUCcVU4wZGl6dnNlNkE


Garth Wallace wrote on Wed, Aug 10, 2016 01:43 AM UTC:

Er, here's the link


Garth Wallace wrote on Tue, Aug 30, 2016 10:14 PM UTC:

The WFCC has given its official thumbs up for the Unicode proposal project. Could anybody representing Chessvariants.com get in touch? It would be great if we could get all of the stakeholders involved so we can make sure everybody is well served by the final draft.


🕸Fergus Duniho wrote on Wed, Aug 31, 2016 01:03 AM UTC:

For it to be useful here, it would have to be very thorough and comprehensive, and I don't get the impression that it will be. So, I expect I will just continue to use image files to represent pieces, and because of this, I don't consider myself a stakeholder in this.


Garth Wallace wrote on Wed, Aug 31, 2016 06:26 AM UTC:

Well, that's partly why I asked. The current draft is limited by my research skills. Information on the symbols used by fairy chess composers was relatively easy to come by, but I had a much harder time finding info on symbols used by variant creators, so I only included the knighted compounds since I knew they were used in several relatively popular variants, and the shatranj pieces because I had an unambiguous example of their use in notation in a published document. That doesn't mean a case can't be made for other pieces, it just means that I don't have enough data to make a case. So I'd like to involve people with more expertise on that front.

There are also open questions about several pieces. For example, should a camel piece be considered unified with the existing DROMEDARY or BACTRIAN CAMEL characters? Since some games call a vao an "archer" or "bowman", should an archer and a diagonal cannon be considered variations on the same character? These aren't questions I can answer on my own, they would depend on consensus.

But I never expected you to replace images on the website or in Game Courier with Unicode characters, particularly in board diagrams. That's not really what this is about. This is so that people can use figurine notation for chess variants in plain text contexts like email, and more conveniently in styled text contexts like Word documents.


🕸Fergus Duniho wrote on Wed, Aug 31, 2016 04:45 PM UTC:

Since some games call a vao an "archer" or "bowman", should an archer and a diagonal cannon be considered variations on the same character?

The use of arrow for vao is an English translation I made of the Chinese character I used to represent the piece in Yang Qi. But the name of arrow may be used for other pieces in other games. The archer and bow pieces I made for the vao were just ideas that I made images for without using them.

In general, we may class piece images into two main categories. One category is for images of particular pieces with particular powers of movement, and the other category is for images suitable for any piece using a particular name. As an example, the following two images could be used for David Paulowich's Unicorn piece, which moves as a Bishop or a Nightrider. The first image is specific for those powers of movement, while the second is suitable for any piece called a Unicorn, such as the 3D Unicorn in Raumschach, the Unicorn in Grande Acedrex, or the knight-camel compound in Musketeer Chess.

These two categories should be kept separate, and having a piece in each category that would fit the same piece should be okay. The first category would consist of images for some basic piece types and their compounds. The second category would include images illustrative of particular names commonly used in variants, these typically being names of animals, both real and mythological, weapons, military ranks, royal titles, and occupations.

The use of upside-down and sideways pieces for fairy pieces is an artifact of having a limited selection of piece images, and as long as distinct images can be provided for different pieces, they are no longer needed.


Garth Wallace wrote on Fri, Sep 9, 2016 07:33 AM UTC:

I think making the case for a lot of these would come down to showing that they are used in games with significant player bases. For example, I know that cannon symbols are used in Shako, but how many people play that? How people regularly play Tamerlane Chess or Grant Acedrex?

Thanks for reminding me of the unicorn in Raumschach. AIUI that's the most popular form of 3D chess, so that could be a point in the unicorn's favor. On the other hand, I don't know of any fonts that include it.

I wouldn't say the rotated pieces are no longer needed. They're certainly needed for chess problems, where they are a well-established notational practice. They were originally used due to the hassle and expense of cutting new type, but now they're traditional for problemists (and convenient when dealing with an essentially unlimited variety of possible pieces). I understand the variant community generally prefers more directly illustrative symbols though.


Garth Wallace wrote on Tue, Sep 13, 2016 08:44 PM UTC:

Related news: a new proposal for xiangqi and janggi pieces (PDF) was just submitted to Unicode yesterday (not by me). It supersedes earlier proposals submitted by the governments of China and Korea, which had gotten some pushback and seem to have been abandoned by their proposers. This new proposal looks like it's focused more on emoji display.


Garth Wallace wrote on Fri, Nov 18, 2016 11:27 PM UTC:

Unicode has responded to the proposal. Their response is section 13 of this document. To sum up: they were receptive to the idea of encoding additional chess piece symbols, but skeptical of the particular set that was proposed, and requested a revised proposal. They thought that several of the pieces did not have enough evidence of use in text in the examples provided.

Unfortunately, that includes all of the variantist pieces: the fers and elephant, the joker, and the knighted compounds. They specifically say that the shatranj pieces should be taken out of the proposal to be considered separately. And they made it clear that they will not accept examples of use in diagrams as implicit evidence of a need in plain text, so only samples of figurine notation will be acceptable.

So I've been trying to pull together more evidence, with the help of the WFCC. So far, I've found examples of the knight-bishop and knight-rook in figurine notation, but when it comes to the amazon and joker I'm coming up short.


Garth Wallace wrote on Mon, Sep 17, 2018 05:57 PM UTC:

An update on this: the proposal was ultimately accepted. The final proposal mostly covers symbols from the fairy chess problemist tradition (e.g. rotated pieces), but does also include the knight-rook, knight-bishop, and knight-queen compound symbols. The characters are currently planned to be added to Unicode 12.0, which IIRC is scheduled to be published in March 2019, in the new "Chess Symbols" block (U+1FA00–U+1FA6F) that was added in Unicode 11.0 and currently includes symbols for Xiangqi.


Garth Wallace wrote on Tue, Nov 6, 2018 05:58 PM UTC:

The beta review period for Unicode 12.0 has just begun. This contains the heterodox chess piece symbols. The code point assignments are now fixed; the beta period is to allow developers (and font designers) to get a head start on anything that depends on the newly assigned characters before the Standard is formally released in March.


15 comments displayed

Earlier Reverse Order Later

Permalink to the exact comments currently displayed.