[MPEG-OTSPEC] Many to many substitution and localization

[Simon Cozens]

On 30/09/2024 11:14, William_J_G Overington via mpeg-otspec wrote:
> What needs to be done, by me or by others, for the suggestion in this 
> thread to become formally considered for inclusion in the standard please?

This is a really good question. My personal opinion:

1) Ensure that any suggestion is written in as broad terms as possible, 
allowing for general purpose rather than specific purpose use.

2) Gather evidence of broad consensus that the new feature is desirable. 
One person thinking that something is a good idea does not make it so.

3) Ensure that the font format is the most appropriate place for the 
standard. Can/should it be done by a layout engine? A rendering engine? 
A higher-level protocol? If so, it does not belong in the font format. 
If you think it should belong in the font format even though it can be 
done in other part of the text stack, demonstrate evidence.

4) Ensure that you're using the terms of the specification precisely. 
For example, there is no such thing as a "liga table" in the OFF 
specification, so it's hard for other people to follow what you mean 
when you say you want to "adapt a copy" of it. To do this, you may need 
to read, and possibly even understand, the specification.

5) Develop, or commission, a working implementation so that the feature 
can be demonstrated and evaluated.

6) Write the suggestion in a way that is suitable to be inserted into 
the existing specification.

7) Once all the above is done, gather evidence of broad consensus of 
approval.

Good examples of how to do the above seven steps are:
https://github.com/harfbuzz/boring-expansion-spec
and https://github.com/googlefonts/colr-gradients-spec

Bluntly, you are currently zero for seven.

Simon