Re: face-remapping patch

[David Kastrup]

Richard M Stallman <address@hidden> writes:

> I think we should continue with local variable bindings (of various
> sorts) rather than change to specifiers.  The fact that these bindings
> apply to the same variables that ordinary code can look at
> is a great convenience.

As I said: specifiers have the advantage of being able to
define/differentiate behavior even in the case where the locus is not
established yet.  We have a similar thing with face specifications.  We
never had this for image specifications, I think (where one can specify
in XEmacs what to do for B&W, monochrome and more complex displays).

If I get the terminology more or less right, a specifier is then
instantiated on a particular device.  That's where our localities come
into play.

I doubt that we have enough different and pressing applications for
"uninstantiated specifiers" that a generic approach is really needed.

And I don't think that conflicts in our current locality schemes will
magically disappear by switching to different data structures where we
basically tell people "sort out conflicts yourself if you understand
them".  Choice is no substitute for ease of use.

-- 
David Kastrup