Re: character syntax fixes needed

[Dave Love]

Richard Stallman <address@hidden> writes:

> As Stefan says, we choose the syntax for each character in any given
> mode that is most convenient for the users.

The issue is the default syntax for characters.  Obviously programming
language modes should use syntax that fits the language definition as
well as possible, and I've fixed a number of them for non-ASCII
identifiers.  We actually have tried to add syntax systematically in
characters.el on the basis of Unicode, like the balanced pairs I
referred to before.  It just needs more work.

It obviously isn't convenient for German, Swedish, Russian &c users to
end up with unbalanced parens when they use quotes in text just at the
whim of someone who doesn't know or care about the conventions of
their language.

> What Unicode says is at best a suggestion, perhaps worth
> considering.

As far as I know, it's authoritative, modulo occasional bugs and
obscurities.  What makes you think otherwise?  It's ridiculous to
propose considering each character and rejecting the Unicode
definitions in arbitrary special cases, if that's what you're
suggesting.  Who is in a position to do that, especially if
maintainers don't even know Western European usage?

Are you abandoning Unicode-based Emacs?