[ESS] Key bindings for some features {was 'Adjust R option width'}

[Friedrich.Leisch at tuwien.ac.at]

>>>>> On Fri, 20 Jan 2006 17:30:16 +0100,
>>>>> Martin Maechler (MM) wrote:

  FrL> Umm, the emacs manual says

  FrL> As a user, you can redefine any key; but it is usually best to
  FrL> stick to key sequences that consist of `C-c' followed by a letter.
  FrL> These keys are "reserved for users," so they won't conflict with
  FrL> any properly designed Emacs extension.

  >  (I have know the above for more than 10 years)


I would have thought so, hence was surprised to see a proposal
violating the rule

  FrL> so ESS should use a different binding (same for your "\C-cf").

  > I disagree. I had spent quite a bit of  thought on these many
  > years ago. My points on this have been

  > - ESS already uses so many other key bindings, it's not easy to
  >   find another  C-c C-<letter> one which also should be mnemonic
  >   to some extent {and I didn't want to use
  >   more complicated keys; or things like [F1] etc}.

Give me a break, ESS has/needs by far fewer bindings then many other
major modes I use on a regular basis, e.g., auctex certainly has (and
will always need) much more!

  > - ESS does not define these bindings 
  >   but provides a function that the user has to call {maybe in
  >   her/his init files} which activates the key bindings.

Now that is IMHO cheating the rule ;-)

As you may imagine I have a full set of private C-c<letter> bindings
(grown over more than a decade of using emacs), and I guess the same
is true for most other long term emacs users. 

Now if you provide a function providing hopefully useful bindings that
overlap with mine I simply couldn't use it ... hence ESS might as well
not ship it. Of course I could look at the function creating the
bindings, adjust it to be compatible with mine, etc., and for ESS I
would probably do it, because I use it a lot.

But imagine every emacs package would use your approach/attitude: Are
you seriously suggesting that each user should figure out his private
bindings for each package he uses in order to efficiently use all the
features? That's a nightmare come true. 

Just my 2c.

Fritz

-- 
-------------------------------------------------------------------
                        Friedrich Leisch 
Institut für Statistik                     Tel: (+43 1) 58801 10715
Technische Universität Wien                Fax: (+43 1) 58801 10798
Wiedner Hauptstraße 8-10/1071
A-1040 Wien, Austria             http://www.ci.tuwien.ac.at/~leisch