A proposal for separate English localization

[Toki]

On 10/17/2017 09:17 AM, Jan Iversen wrote:

> I have never understood why we make English a special case, and assume all developers are perfect English speakers.

Two reasons:
* Inertia;
* An unwillingness to create both English(US) and English(UK) as
independent l10n projects.

That every professionally run l10n project does «let the developer enter
a suggested string, which is then through tooling converted to an
identifier, and the suggestion is moved to pootle.» appears to be
irrelevant.

That every time something in the English(US) strings get changed, means
each L10N team has to spend hours figuring out what, if anything --- and
as oft as not it is nothing -- was changed, is more than ridiculous.
(IIRC, if the translators were paid at the usual and normal translation
rates, each time the English (US) strings changed, the cost would be
US$1,500,000.)

jonathon