FWIW, my first programming language was РАПИРА, here is its Hello-World:

ПРОЦ СТАРТ();
    ВЫВОД: "ЗДРАВСТВУЙ, МИР!";
КНЦ;

I don't recall the keyboard I used back then to have any Latin characters.

The Soviet knockout of Space Shuttle automatically landed in 1988 while controlled by the software written in what we would now call a domain-specific language ПРОЛ2, essentially Prolog with Russian keywords. They would unlikely go into all that trouble if they didn't think it would improve their productivity.

-----Original Message-----
From: Ville Voutilainen <ville.voutilainen_at_gmail.com>
Sent: Tuesday, July 30, 2019 11:35 AM
To: Lev Minkovsky <lminkovsky_at_[hidden]>
Cc: JeanHeyd Meneide <phdofthehouse_at_gmail.com>; keld_at_keldix.com; unicode_at_isocpp.open-std.org <unicode_at_open-std.org>
Subject: Re: [SG16-Unicode] Ideas for the future

On Tue, 30 Jul 2019 at 18:30, Lev Minkovsky <lminkovsky_at_outlook.com> wrote:
>
> Alas, speakers of RTL languages would probably find it more convenient to use standard English C++.
>
>
>
> I never mean the C++ to be translatable into all 200+ world languages
> that have writing. Each and every such translation will indeed be a
> major undertaking. The language keywords will actually be an easy
> part. More difficult will be to translate the library. I suspect the
> only realistic solution will be to create a set of “national” headers
> that will map onto the established definitions. For example, the
> стдвв.г header I used in the example could include the following
> definition

The last time I ran into localized programming languages were certain spreadsheet automation languages, and the users of those that I talked to very strongly asked "please don't allow anyone to do this ever again".