bug#7925: 23.2.91; report-emacs-bug doesn't like curly quotes

[Eli Zaretskii]

> From: Štěpán Němec <stepnem@gmail.com>
> Cc: 7925@debbugs.gnu.org,  cyd@stupidchicken.com,  rrt@sc3d.org
> Date: Sun, 30 Jan 2011 15:19:35 +0100
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> In any case, this reasoning is quite different from the one stated
> in your previous message

The previous message corrected an assumption that the request to use
English comes from the pure ASCII era.  This reasoning, by contrast,
tried to explain why asking for English does not guarantee 7-bit ASCII
text, even if the user strictly abides by that request.

> (although it doesn't make much more sense to me).

Would _any_ reasoning make sense to you in this matter?  If not, we
are having a religious argument, and then we better stop.

To make more sense out of it (hopefully), think about the various ways
a Lisp-level backtrace (or, indeed, any information from the Emacs
display) could be brought into the bug report (from the same Emacs
session, by copy-pasting via a windowing system, via a file, etc.),
and then try to imagine the set of possible encoding related mangling
that could cause it to be utterly illegible.

Anyway, it's time for me to bail out.  Someone asked a question why
report-emacs-bug does what it does.  I tried to answer, which
hopefully helped understand the reasons.  I don't have anything else
to say on this matter, and I really don't care if non-ASCII characters
will be allowed.  If a bug report uses non-trivial characters, its
submitter should consider the possibility that it will be read by a
smaller number of maintainers, in extreme situations not at all.