Why UTF-8/16 character encodings?

monarch_dodra monarchdodra at gmail.com
Thu May 30 03:40:47 PDT 2013


On Thursday, 30 May 2013 at 10:13:46 UTC, Dicebot wrote:
> On Thursday, 30 May 2013 at 09:36:43 UTC, monarch_dodra wrote:
>> What about Chinese? Russian? Japanese? It is doable, but I can 
>> tell you for a fact that they very much don't like reading it 
>> that way.
>>
>> You know, having done programming in Japan, I know that a lot 
>> of devs simply don't care for english, and they'd really enjoy 
>> just being able to code in Japanese. I can't speak for the 
>> other countries, but I'm sure that large but not spread out 
>> countries like China would also just *love* to be able to code 
>> in 100% Madarin (I'd say they wouldn't care much for English 
>> either).
>
> What about poor guys from other country that will support that 
> project after? English is a de-facto standard language for 
> programming for a good reason.

Well... defacto: "in practice but not necessarily ordained by 
law".

Besides, even in english, there are use cases for unicode. Such 
as math (Greek symbols).

And even if you are coding in english, that don't mean you can't 
be working on a region specific project, that requires the 
identifiers to have region-specific names (AKA, German banking 
reference).

Finally, english does have a few (albeit rare) words that can't 
be expressed with ASCII. For example: Möbius. Sure, you can write 
it "Mobius", but why settle for wrong, when you can have right?

--------

I'm saying that even if I agree that code should be in English 
(which I don't completly agree with), it's still not a strong 
argument against unicode in identifiers. In this day and age, it 
seems as arbitrary to me as requiring lines to not exceed 80 
chars. That kind of shit belongs in a coding standard.


More information about the Digitalmars-d mailing list