Re: inputting characters by hexadigit

[David De La Harpe Golden]

Stefan Monnier wrote:

> I'd drop the C-x 8 1 / 2, C-x 8 1 / 4, and C-x 8 3 / 4, replace them

Was going to suggest C-x 8 h - half C-x 8 q - quarter C-x 8 t - three
quarters  ... but then noticed iso-transl-set-language, some languages
use C-x 8 h...

But there's also the issue that unicode includes a number of other
common fractions, so it might be nice to allow for supporting 'em as C-x
8 1/3 for ⅓ and similar for ⅝ ⅗ etc. in future.

> with something else, and then use C-x 8 NNNN RET. 

Should allow for N to NNNNNN, not just NNNN  .

> Note that I'd also imagine the NNNN to use hex, so it would also conflict 
> with other
> existing bindings, 

Also N.B. conflicts change depending on M-x iso-transl-set-language   !

IMO might be better to just require a single character prefix, rather
than eating so much of C-x 8 space.

C-x 8 # 263a seems nice to me, though, my bad,
C-x 8 0 263a is perhaps easier to type for people who
aren't using british keyboards (# is an unshifted key for british/irish).

Is it a goal to eventually deprecate C-q entry?
Just wondering... C-q 0x263a  doesn't seem so bad...