At 2018-12-17T21:36:28+0100, Ingo Schwarze wrote: > My comment was specifically about a character for which Branden > said that Unicode support is so recent (U+1E9E LATIN CAPITAL LETTER > SHARP S)
I did say this.
> that the operating systems of some users may not have support for it
> yet
This, I did not say, but it can be reasonably inferred.
> and may continue to show it as a small letter even after applying
> towupper(3).
No. I neither said nor suggested this, because I did not assume that
the classical .tr request could reasonably be made to expose an
interface to to*{lower,upper}.
I was assuming we'd have to kludge around the lack of such an interface,
and I didn't know how we were going to do that. Tadziu and Werner
almost simultaneously proposed using .char to define a character escape
to provide an uppercase "SS".
> Even assuming that fear is well-founded, it's a fringe concern.
This, I can concede. All the German I have, I acquired prior to Reform
der deutschen Rechtschreibung von 1996 (and it has decayed aggressively
since), so I feel particularly ill-suited to judge issues of
orthography.
I will happily refrain from inflicting my poor schoolhouse German on
people if we can get all the darned misuses of "actual" out of our man
pages. ("Actual" is not a synonym for "current".)
Bernd in particular had a predilection for the construction, "the actual
version is here: $URL". I was always baffled by that, until I found out
it was actually (natch) quite a well-known barbarism[1].
I think it would be better to extend groff to expose the underlying
locale-aware C case-transformation functions, and _not_ try maintaining
our own mappings. Nevertheless I value the recovery of distinct
lettercases in man page titles and section names sufficiently to
undertake that work if it's the price that must be paid.
Regards,
Branden
[1] https://en.wikipedia.org/wiki/False_friend#Shared_etymology
signature.asc
Description: PGP signature
