RE: Change `customize-save-variable' to work under "emacs -Q"?

[Drew Adams]

> > If you run "emacs -Q", and you have a piece of code that calls
> > `customize-save-variable', it will error out, saying that it won't
> > overwrite the conf.
> >
> > Would anybody mind if I changed that function to not error out, but
> > instead just do a `setq' on the variable in question, if 
> > we're running under -Q?
> 
> I think this is fine.

Ouch!  Maybe I'm misunderstanding, but a user in `emacs -Q' who does `M-x
customize-save-variable' and sees no error message will expect that it has been
_saved_, no?  Likewise any code that a user might run that calls
`customize-save-variable'.

How can we _silently_ convert saving to just setq?  The user will think that
something is saved when that is not the case.  Sounds like asking for trouble,
to me.  But maybe I'm missing something.

Why is this better than raising an error and letting the user then use `setq'
etc.?  Why hide the situation from the user?

I would think we would want to take advantage of the opportunity to remind the
user that s?he invoked `emacs -Q', so no init file, so no place to save the
option.

Also, why would we convert to just `setq' and not `customize-set-variable'?  The
latter DTRT wrt :set etc.