[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#645656: network-manager in Gnome

On Fri, 2011-11-04 at 20:55 +0100, Tollef Fog Heen wrote:
> So since gnome-shell actually needs gnome-bluetooth, the dependency
> should be demoted to a Recommends?
Well but then it would be enough for gnome-shell to depend on it.

And one should perhaps try to, whether it's easy to patch it, that it
does not longer depend on it, cause this is really stupid.

And yes I know Stefan's argument, that we're just packagers... but we
change/improve so many things... why not here, if upstream makes stupid

> I believe the Gnome packaging team would be happy to accept more members
> if somebody wants to work on this and keep maintaining it.
You shouln't take my comments as criticism or offense on the gnome
packagers... they do a great job and I guess most people appreciate
Gut given that we know that they're quite competent and given that (in
the view of many people) upstream has gone a bit mad... it's not
unreasonable to ask them, to give the users their freedom back.

> I suggest you file a bug about it so it can be fixed.  (I didn't find
> one filed by you at least, but I didn't look very hard.)
No I haven't yet... I suffer from so many small and nasty problems since
gnome3, that I even don't know where to start.
And most of the "bugs" are upstream anyway and I guess we all know their
stance on "more features/functionality".

I mean we all know the heavy criticism that takes place upstream, about
GNOME shell, etc.
It's surely not useful to repeat this here.
No matter who's right, one thing is certain IMHO... all the "old" ways
(like having panels and so)... or e.g. not having NM, but managing this
e.g. via ifupdown alone... were proven workflows for years or decades.
Forcing users to switch because upstream believes it know better, is not
just wrong, but also arrogant.

On Fri, 2011-11-04 at 21:38 +0100, Josselin Mouette wrote:
> Because doing so is like ensuring there will be some systems that
> work correctly, wasting our time on bug reports that can be fixed just
> by installing the missing recommends.
> Worse, it would let metapackages migrate to testing without the
> appropriate dependencies.

I don't see why.
E.g. I installed my gnome-core without gnome-packagekit... and it works
just fine... and I'm also quite sure that there's no need for many other
things (epiphany,.. etc. etc.) I've configured to be never used, and
gnome never somehow started automagically.

> Are you running the latest kernel? There have been a lot of fixes in
> drivers for PA, and it’s now working on almost all systems (of course
> there are still unlucky ones).

> The same holds for NM. Most criticism has been inherited from version
> 0.6 which was full of design flaws. Upstream and us made this a
> dependency now because we believe it will be easier to fix the
> bugs than to support different underlying subsystems.

Well but then you admit,... that it was not added because it's really
needed, aren't you?!
And btw: I use NM now too on my notebooks, I reported things to the BTS
which for me still don't work,... but nevertheless, I don't like people
being more or less "forced" to use it.

And what starts as deps in meta-packages easily out-grows to real hard
deps from other packages.

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply to: