[MPEG-OTSPEC] Some ideas for an alternate feature variations mechanism

[Skef Iterum]

A few weeks ago I sent an email about three concerns Adobe hoped to 
discuss at this week's meeting. One of those was the problem of too many 
regions when mixing features, outlined in 
https://github.com/MPEGGroup/OpenFontFormat/issues/53 .

As Behdad notes there, the associated proposal is a minor adjustment 
that arguably doesn't address more basic problems with the existing 
feature variations specification. He and I have subsequently been 
sketching out a larger change in the form of an alternative mechanism 
that could be added to the existing one. Yesterday I made a very rough 
draft of how that might work in terms of tables and records, linked at 
https://github.com/MPEGGroup/OpenFontFormat/issues/53#issuecomment-1678593050 


It's far too early to call this new sketch a proposal. However, I 
encourage anyone who will be attending the upcoming meeting and is 
interested in this corner of the specification to take a quick look at 
the discussion in the issue and the "new_substitution.pdf" document. 
This is the sort of thing where if the basic idea looks promising, a 
smaller even-more-ad-hoc group could discuss it on GitHub or elsewhere 
and perhaps come up with something more substantial for a subsequent ISO 
meeting.

Skef