Re: Customize Rogue

[Luc Teirlinck]

Juanma Barranquero wrote:
   
   I think our difference about custom is philosophical. To me, customize
   should be *only* an easy-to-use interface to setting Emacs variables,
   modes, etc.

As opposed to what?

   In that light, getting a message that says "Value changed
   outside of customize buffer" is only meaningful if meant to indicate a
   posible problem (for example, that I did set a variable instead of
   running the corresponding minor-mode function, as you point out
   elsewhere).

It is a complete mystery to me how you can see all kinds of ominous
hidden meanings in a straightforward statement like:

"State: this option has been changed outside the customize buffer."

This is supposed to tell you the current customization status.  If you
changed it outside the customize buffer, then why would you want it to
give you false information?  That is what it is: information, not a
moral judgment.

   As it stands now, if I
   set up a variable in my .emacs and then do a M-x customize-variable I
   get a vague feeling that I'm somehow "doing something wrong"... :)

If you intend to use the customize buffer to change the value (as most
people do when they use M-x customize-variable), then you are doing
something wrong, and should be alerted to this.  If you set an option
with set-activate in your .emacs and then change your mind, you have
to do that in your .emacs, using the customization buffer will have no
effect.  (I believe Per already pointed this out to you before.)

   instead, it has
   all to do with the fact that I want the "right" to make customize
   believe that I did M-x customize, even if I didn't.

I believe you want customize to respect your right not to use it.  How
can it possibly do that if you fool it into believing that you do use it?

   I'm not so sure. I'd say (but I've not checked it) that there are
   special :set functions that are lambdas and not functions, because
   they're intended to make data self-consistent or whatever.

Yes, but this is exactly the problem set-activate is trying to
address.

   > However, set-activate will relieve you of the problem of
   > finding it (fixing the doc-string to mention the function would also
   > be a good idea).

   I like it (with another name). But I would also make it set 'saved-value
   or whatever.

Why?  To avoid the above "State: " line?  As pointed out, there are
situations where the user needs that information.

    Still, I'd like customize to consider an existing custom-setq more
    prioritary than the stuff it usually writes.

You can easily achieve that effect by putting the custom-set-variables
and custom-set-faces forms early in your .emacs, before any
set-activate forms.  I personally would prefer not to have any
conflicting customizations at all, but that is your choice.

Sincerely,

Luc.