Re: Coding system sniffing.
David Kastrup writes:
Whether or not it makes sense in the light of a sensible design: if
it
is reasonably reliable, there is a chance that people will adapt the
few cases where it misses to Emacs' heuristics, and if XEmacs had
different heuristics, this would be sort of a pain in the neck.
Sure. However, Mule history is rife with cases where Mule stubbornly
insists that it knows better than the user, and at least in XEmacs, as
you know very well, recovery from mistaken judgments about coding is
frustrating at best, and goes downhill rapidly from there.
By "design", I don't mean heuristics. I mean implementation details.
Before we start doing fancy stuff, we need to have a straightforward
way for the user to get a specific coding system, even if it's tedious.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta