Proposal to add standardized variation sequences for chess notation

Michael Everson everson at evertype.com
Tue Apr 4 19:02:32 CDT 2017


On 4 Apr 2017, at 18:54, Richard Wordingham <richard.wordingham at ntlworld.com> wrote:
> 
> On Tue, 4 Apr 2017 01:30:05 +0100
> Michael Everson <everson at evertype.com> wrote:
> 
>>> I'm trying to work out whether we need a variation sequence for "chesspiece in a sentence”.  
>> 
>> Of course! Haven’t you ever seen chess problem texts? Check out the Fairy Chess proposal for encoding additional characters. Plenty of examples there. 
> 
> Your examples did not have to contend with the possibility of fonts that only support the variants for drawing chessboards.

Um, what? 

Why would anyone make a font that supports the variants for drawing chessboards (which require the encoded characters 2654..265F) not put in glyphs for those? 

FontLab is the program I use to add OpenType features to my fonts, and if I try to add a sequence like 2654 + FE00 and the font doesn’t have a 2654, if flags it as an error and insists that the character appear in the font. OK, someone could be perverse and not add glyphs to those code positions, but…  

But nobody making a chess font with actual support for chess would do that. 

So this is another red herring. As far as I can see, your worries are groundless, and nothing has suggested that there’s something wrong with the proposal. 

Also, having implemented it in three or four different fonts now, I find that it works. It does the job, and it’s easy to use to edit.

>> Sorry, I meant “Of course **not**!” that is, chesspiece in a sentence is extremely common, and should be the default (not stylized) form. We can’t repurpose that to be “chesspiece on a white square” because it hasn’t been previously and changing that would affect the layout of existing data.
> 
> But would not your proposal make it legitimate for a font to supply only chess pieces on dark backgrounds for the chess piece characters?

What does “legitimate” mean? 

Nothing prevents someone from drawing the 16 Myanmar base characters with rings at the ends of their glyphs even though now VS are being recommended for that presentation. Is it legitimate to do that? Of course it is. It’s legitimate to make Myanmar fonts with square glyphs rather than circular ones. 

This proposal provides a stable encoding model for drawing chessboards simply, with fonts. Currently there are other fonts which do this, but they do not share encodings, and so sharing chessboard data is dependent on whether you have set up your board in the same font encoding that somebody else is using. Otherwise it doesn’t work, and your text is corrupt and you have to re-key various elements in order to use the glyphs of the other font. This problem is described in detail at the beginning of the proposal. It is the same problem we had with ISO/IECE 8859-1, -2, -3. -4 etc before we had the UCS. So: we have unstable non-Unicode encodings for chessboards now, this proposal provides stable Unicode encodings. 

This can only benefit the community of users of chess fonts. Anybody who isn’t setting chessboards is unaffected, just as I am unaffected by variation selectors used for glyph variation in mathematical fonts. (I might add the slashed zero glyph to Everson Mono, though.)

This proposal does this while leaving the base characters alone so they can be used as chesspieces in text (as they have been since Unicode 1.1) and by adding a mechanism to construct the glyphs necessary for presenting chessboard data.

This proposal uses a mechanism which has already been used for dozens of regular characters and 310 times for some popular pictographs. No new characters need to be added. Just a list of items in a text file. 

Can you identify an actual problem? 

Michael Everson


More information about the Unicode mailing list