On Wednesday, 26 September 2018 at 20:43:47 UTC, Walter Bright
wrote:
On 9/26/2018 5:46 AM, Steven Schveighoffer wrote:
This is a non-starter. We can't break people's code,
especially for trivial reasons like 'you shouldn't code that
way because others don't like it'. I'm pretty sure Walter
would be against removing Unicode support for identifiers.
We're not going to remove it, because there's not much to gain
from it.
But expanding it seems of vanishingly little value. Note that
each thing that gets added to D adds weight to it, and it needs
to pull its weight. Nothing is free.
I don't see a scenario where someone would be learning D and
not know English. Non-English D instructional material is
nearly non-existent. dlang.org is all in English. Don't most
languages have a Romanji-like representation?
It's not that they don't know English. It's that non-English
speakers can process words and sentences in non-English much more
efficiently than in English. Knowing a language is not binary.
Here's an example from this years spring semester and NTNU
(norwegian uni):
http://folk.ntnu.no/frh/grprog/eksempel/eks_20.cpp
... That's the basic programming course. Whether the professor
would use that I guess would depend on ratio of
English/non-English speakers. But it's there nonetheless.
Of course Norway is a bad example because the English level here
is, arguably, higher than many English countries :p But it's a
great example because even if you're great at English, still
sometimes people are more comfortable/confident/efficient/ in
their own native language.
Some tech meetups from different countries try and do things in
English and mostly it works. But it's been seen consistently with
non-English audiences that presentations given in English result
in silence whereas if it's in their native language you have
actual engagement.
I fail to understand how supporting a version of unicode from
(not sure when it was released) 3 billion decades ago should just
be left as is and also cannot be removed when there's someone
who's willing to update it.
C/C++ have made efforts in the past to support non-ASCII coding
- digraphs, trigraphs, and alternate keywords. They've all
failed miserably. The only people who seem to know those
features even exist are language lawyers.
This is not relevant. Trigraphs and digraphs did indeed fail
miserably but they do not represent any non-ascii characters. The
existential reasons for those abominations were different.
Anyway, on a related note: D itself (not identifiers, but std)
also supports unicode 6 or something. That's from 2010. That's a
decade ago. We're at unicode 11 now. And I've already had someone
tell me (while trying to get them to use D) - "hold on it
supports unicode from a decade ago? Nah I'm not touching it". Not
that it's the same as supporting identifiers in code, but still
the reaction is relevant.
Cheers,
- Ali
