Re: i18n/gettext?

[Richard Stallman]

    There were a few areas demanding special care.  One is that, given
    the structure of LISP itself, the textual domain may have to switch
    quickly and often while the code is executing, so provisions should be
    constructed to ensure every string belongs to its proper domain.

We could perhaps put a text property on the string's contents to say
what domain it is in.

      Another,
    quite related to the first, is whether documentation strings need to be
    translated or not (I think they should).

I think we should support translation of documentation strings.
I think that should not be particularly hard.

    who strongly objected that we mark strings
    to be translated, so departing from how we do for most other packages.

If that was me (I don't know if it was), I don't remember having
thought about the issue.  Perhaps that means it was someone else.

There is a clear and large advantage in not marking the strings
to be translated: marking them would be a gigantic job and would
change most of the Lisp files.  Avoiding that gigantic job
is a tremendous advantage.

So the question is, why is it hard to arrange not to mark them?
If that is a harder job than marking all the strings, we may
as well mark all the strings.  But since that is such a big job,
it is worth making a considerable effort to avoid it.