On Fri, 19 Jun 2020 at 01:26, Corentin Jabot <corentinjabot_at_[hidden]> wrote:
>
>
>
> On Fri, 19 Jun 2020 at 00:13, Ville Voutilainen via SG16 <sg16_at_[hidden]> wrote:
>>
>> On Thu, 18 Jun 2020 at 21:55, Jens Maurer via SG16
>> <sg16_at_[hidden]> wrote:
>> >
>> > So, it seems we would increase consensus in EWG if we
>> > added emojis to the valid identifier characters.
>>
>> While my views on this matter should be well-known by now, I have a
>> question about this.
>> What is the rationale for emojis in identifiers?
>
>
> I am actually not sure that there is a strong one, but as P1949 will remove support for the few emojis that happen to be supported,
> there is some push back.
> The fact that C++ allows some emojis seems rather accidental as it only allows a subset of them (in a way that is problematic in terms of diversity).
>
> I could give you some reasons like, maybe it appeals to young people learning the language or it's fun in slide code / twitter poll.
> But the truth is, in languages that do support them ( Swift , raku), it seems barely used beyond a novelty effect, and it would have a non-negligible cost
> both for us and implementers.
>
> Either way, the status quo is not satisfactory, support needs to be fully added or fully removed.
>
> I hope we understand that making sure people can write code in their native language is a different matter than allowing emojis.

That's all fine by me.

I'm confused to the hilt by this:

"So, it seems we would increase consensus in EWG if we
added emojis to the valid identifier characters."

The paper I read didn't seem to go into that direction. That quoted
bit (which I copy-pasted, it's not a drunken
transformation) seems like it's a completely new direction.