vi editor enhancement request

[Doug Gwyn]

In article <15948 at bloom-beacon.MIT.EDU>, jik at athena.mit.edu (Jonathan I. Kamens) writes:
> 1. Why is emacs more universal than vi, if different versions of emacs
>    have different command sets, ...

(1) EMACS runs on DECsystem-10s, and lots of other non-UNIX systems.
    There are flavors of EMACS for personal computers, even the Apple II.

(2) The basic default key bindings are practically the same across all
versions of EMACS.  ^F goes forward a character, ^N goes down a line, etc.
One can cutomize his own use of many flavors of EMACS via .emacsrc-like
files, and there is even a "vi" emulation available for some EMACS.
(The converse is impossible => EMACS is more universal in a third sense.)

> 2. Why is emacs more universal than vi, if emacs is NOT shipped by
>    default with most Unix systems, while vi IS shipped by default with
>    most (if not all -- anybody here work on a system that doesn't have
>    vi :-) Unix systems?

Serious UNIX houses generally provide some local flavor of EMACS in
addition to whatever is shipped by system vendors, and of course you
can get GNU EMACS for free.  We've had at least five flavors of EMACS
here, one of which we selected years ago to officially support lab-wide.

> 3. Why do you have different versions of emacs on your VAX, Ultrix and
>    Sun machines is GNU emacs will compile for all of them?

Why not?  Is it a problem, or what?

> >It is difficult to change editors but in this case it is worth it. IMHO.
>   Now *this* I agree with :-)

Me, too, except I would go further and say that the TYPICAL user would
be better served by an editor like "sam".  I reserve EMACS for special
circumstances, and "vi" only when there is no better choice.

The main thing "better" about EMACS over "vi" is that EMACS is modeless.
Experience (backed by some of the usual silly academic "studies") has
shown that modeless interactive editing is overall more effective than moded.