Re: bug#7260: 24.0.50; DEL screwed up

[Stephen J. Turnbull]

David Kastrup writes:

 > > It has been addressed.  The only way this *can* be addressed is to
 > > make a decision one way or the other, and that's been done.
 > 
 > Uh no.  We are programmers.  Consequently, a _lot_ of work has been
 > invested in making transient-mark-mode less noisome, both before making
 > it the default, and in the aftermath.

Sure.  Most of those changes simply moved it closer to zmacs-regions,
so that now I personally find it quite usable. :-)

But I'm definitely not talking about delsel-mode which is currently
not on by default in XEmacs.  I'm talking about the current design of
t-m-m mode, which happens to fit my usage patterns well, and fits
others' rather poorly.  Lars and I cannot both have the default we
want.

 > So there certainly is more to address than whether to turn it on by
 > default, and that is being done.

Sure, but in fact a lot of people who use computers like (fsvo "like"
including "are used to from other systems") exactly the behavior that
Lars is objecting to, which (IIRC) according to the words he wrote was
C-x C-x activating the region, although he may have really been
concerned (at least partially) with the interaction with delsel-mode.

 > > FWIW transient-mark-mode (actually, zmacs-mode) ON has been the
 > > default in XEmacs since 1998 or so, and the complaints took about 12
 > > months to die out.  My impression at the time was that a majority of
 > > the complainers tried the new behavior and decided it was acceptable
 > > (may be biased, because I'm in that group), and a large minority
 > > decided to turn it off at initialization and stopped complaining.  New
 > > users never complain about it AFAICR.
 > 
 > I don't like the "implement and leave it" attitude.

It wasn't an implement it and leave it attitude.  In the "t-m-m by
default" controversy, Emacs people have made a lot of suggestions
(most of which I think are wrong-headed, but they're worth trying --
there's a lot I don't know yet about UI design, including some of the
big principles).  But in the case of XEmacs, the naysayers made no
constructive suggestions, just "oh God, turn it OFF!  Now!"  Had they
made suggestions, they would have been considered, I suppose.

 > Moving a controversial feature to a less controversial state causes
 > a lot of bile and bickering on the Emacs developer list.

Sometimes that's a waste of time.  IMO, t-m-m is one.  It's been tuned
by its users over a decade and through a rather heated controversy at
the time it was decided to make it default, and it's very similar to
zmacs-regions which has been similarly tuned over two or three
decades.  People who don't like it should probably just turn it off --
some things don't admit much improvement, at least not at reasonable
levels of developer effort.

 > It makes the list a less friendly place, and Emacs itself a more
 > friendly place.

I liked TECO Emacs's response to ESC ESC ESC, personally.  Made me
feel like one of the team, sorta like getting a wedgie in the locker
room....