:: Re: [DNG] new dbus packages for ASC…
Page principale
Supprimer ce message
Répondre à ce message
Auteur: Antony Stone
Date:  
À: dng
Sujet: Re: [DNG] new dbus packages for ASCII in experimental -- PLEASE HELP TESTING
On Thursday 15 June 2017 11:18:29 KatolaZ wrote:

> On Thu, Jun 15, 2017 at 12:02:02PM +0200, Svante Signell wrote:
> >
> > gconf2 is not installed but gconf-service, gconf2-common and libgconf-2-4
> > are.
> >
> > Are they really needed? Trying to remove any of them causes
> > emacs24/emacs25 to be removed. (And that is of course not allowed :( )
> >
> > Maybe emacs packages for X should be rebuilt??
>
> Hi Svante,
>
> I was considering that possibility, but didn't really want to go down
> that path unless it is strictly needed, since we have far more
> important priorities.
>
> In most of the cases those dependencies do not come from upstream, and
> have been instead introduced by Debian packagers, who have made a lot
> of effort to entangle as tightly as possible hundreds of packges with
> GNOME stuff.


How does this work out on a Debian system when you select a different DE, then?

> That's really annoying, to be honest, also considering that the desktop
> share of GNU/Linux is below 1% of the total market, and that the Debian
> share of that small percentage might be less than 5% overall, taking into
> account Mint, Ubuntu, and the rest.


Well, surely that's an argument for saying "why bother with Linux on the
desktop?" or "why bother with Debian on the desktop?" (neither of which I
agree with, by the way) rather than "why does Debian focus so strongly on
Gnome?"

Surely the important measurement is "what percentage of Debian users want to
have Gnome as their desktop?"

> So we are talking of a relatively useless chain of dependencies introduced
> to make happier a relatively minuscle minority of users.


It only affects people who have already decided to use (a) GNU/Linux, and (b)
Debian (and then (c) want to have choice about their desktop environment), so
surely either:

- the number of people affected by this decision is so small that there's no
point in bothering about it either way

or

- what's important is the percentage of people who've already made decisions
(a) and (b), who are then affected by choice (c).

Again, I think the important question is "what percentage of Debian users want
Gnome, and what percentage want something else?" That determines what degree
of effort is worth putting into integrating, or separating, Gnome and the rest
of Debian.

> And all this comes from a "universal" operating system....


I thought that just meant it runs on lots of different hardware, not that it
aims to please all of the people all of the time?



Antony.

--
"Remember: the S in IoT stands for Security."

- Jan-Piet Mens

                                                   Please reply to the list;
                                                         please *don't* CC me.