Re: set-background-color

[David Kastrup]

"Eli Zaretskii" <eliz@is.elta.co.il> writes:

> > From: David Kastrup <David.Kastrup@t-online.de>
> > Newsgroups: gnu.emacs.bug
> > Date: 17 Feb 2002 14:49:46 +0100
> > 
> > This is another case where the behavior to a user appears quite
> > unacceptable, as valid color names are rejected: it is perfectly
> > understandable that they are not completed, but they are not even
> > accepted when you press return.
> 
> They are not valid colors for the display being used.

I have a true-color X display capable of displaying the colors such as
"#c0c0c0" that M-x set-background-color will not accept unless called
non-interactively, in which case it has no problems in setting the
background appropriately.

> But to the user, the colors are still treated as invalid, because they
> don't exist as far as the terminal is concerned.

If a true color X display considers simple grey value colors as
non-existent, this is a problem with the implementation, I dare say.

> 
> > That there may be technical reasons underlying the unfortunate
> > current behavior, may be an explanation for the current behavior,
> > but not an excuse.
> 
> I don't think I need an excuse: the code in tty-colors.el is designed
> and written based on years of experience, and works well enough to not
> deserve being called ``unfortunate'', even though it's version 1.0.

Who is talking about tty-colors.el?  It is not even loaded here when
I call set-background-color.

> 
> > I find it counterproductive if core developers scorn criticism
> > because they do not feel able or willing to cope with the raised
> > issues.
> > 
> > I find it perfectly understandable if they say current priorities do
> > not permit them any work on that.  But if they suggest that they
> > consider an obviously dissatisfactory behavior correct, they
> > discourage other people from trying to fix it.  Because those other
> > people must then assume that their fixes will never get accepted.
> 
> Frankly, I don't understand this outbreak of negative emotions.

Where do you see "negative emotions"?  I simply say that I do not find
the current behavior suitable for the user, and do not see how the
internals causing it would be relevant for the sake of discussing the
appropriateness of the user experience.

> I don't think I deserve that for patiently explaining the reasons
> for the current behavior and details of the internal design.  If
> you, or someone else, have more questions, please ask, and I will
> try to answer them as best as I can.

The question is why
M-x set-background-color
is not supposed to accept "#c0c0c0" while
M-: (set-background-color "#c0c0c0") RET
works just as intended on a true color capable X display.

> But I believe I do have a right to have an opinion, and to express
> it without being pounced upon.

So do I.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum
Email: David.Kastrup@t-online.de