One part I didn't catch when first replying to this thread was Ben's
focus on input methods.
I'm curious about that - could someone give more details?
- Is there any consistency among the input methods in how mnemonics are
framed? That is, how do you tell the IME that you're starting a
mnemonic? Do you switch into "mnemonic mode" and type several of them,
or do you enter them as <leadin> <mnemonic> <leadin> <mnemonic>?
- How do the input methods deal with finding the end of a mnemonic?
While a lot of character mnemonics of 1345 are 2-characer, there are a
lot that are not , and there doesn't seem to be a consistent character
set for "intermediate" and "ending" characters ("16" is "VULGAR FRACTION
ONE SIXTH", while "16." is "NUMBER SIXTEEN FULL STOP").
I'm curious to learn about whether the IMEs (EMACS and SCIM was
mentioned) actually use 1345 or some adapted subset of it - if the
latter, documenting the adapted subset that has proved useful in
practice might be a worthwhile exercise.
Ietf mailing list