[ATrpms-devel] Re: Changing foo-kmdl- to kernel-module-foo-? (was: Blackmailing ...)

[Axel Thimm]

Hi,

well the shouting is not stopping ;)

Please let me know your opinion on this. Renaming the kernel module
rpms will be painful, so please back-up the suggested change or reject
it, so we get an opinion :)

I am holding back rebuilding kernel modules for the latest released
ATrpms kernels, as this is still an open issue.

Shout! :)

Thanks!

On Thu, May 06, 2004 at 11:05:59AM +0200, Axel Thimm wrote:
> Hi,
> 
> some of you may be aware of the thread below. To summarize Dag
> explicitely broke compatibility with ATrpms in order to enforce his
> kernel naming and versioning semantics. This is not nice.
> 
> Anyway, independent of blackmailing attempts, how do you feel about
> changing the foo-kmdl- string to kernel-module-foo-? There are
> drawbacks like
> 
> o ATrpms userbase has gotten accustomed to the kmdl naming
> o kernel-module-foo-<kernel version/release>-<foo version/release>
>   is non-intuitive and doesn't sort next to foo in guis like synaptic.
> 
> The pros are, that this scheme is supported by Panu Matilainen, the
> Red Hat/Fedora apt guru (cced). Supported means that kernel upgrades
> automatically perform kernel module upgrades.
> 
> The kernel-module-foo-<kernel version/release>-<foo version/release>
> is currently used by fedora.us and announced to be used by
> PlanetCCRMA. There are only one or two such packages existing (most
> notably alsa), but counting packages is futile.
> 
> I would had preferred to convince the other parties of getting foo to
> the front of the naming scheme, but I think it is quite hopeless. The
> discussion about this is sabotaged again and again, so I am not going
> to try again :(
> 
> We don't have any voting regularisations, but I'd like to make this
> decision dependent on you in the old Athen's democratic style
> (listening to the volume of the voters' shouting).
> 
> Start shouting ;)
> 
> On Thu, May 06, 2004 at 10:39:00AM +0200, Axel Thimm wrote:
> > On Wed, May 05, 2004 at 04:39:10AM +0200, Dag Wieers wrote:
> > > On Wed, 5 May 2004, Axel Thimm wrote:
> > > > Please fix the breakage and we can try to discuss the kernel module
> > > > policies in peace and without enforcement methods? Don't do that for
> > > > the sake of friendship, do that for the sake of the community.
> > > 
> > > I'll make it very simple, if Axel adheres to the kernel-module-naming                                                    
> > > standard and you can convince Matthias about the kernel-module-versioning,                                               
> > > I'll implement the kernel versioning too.  
> > 
> > *Sigh*, which brings us back to where we were at the beginning.
> > 
> > You are breaking things for setting up requirements to fix them
> > again. That's exactly what the Subject is about, you try to enforce
> > your decisions upon others by brute force.
> > 
> > And why does Matthias have to get involved into this cat fight? He is
> > doing very well ignoring this thread I believe, which should be
> > probably what we should be doing also. You either analyse and
> > accept the issues with the versioning you propose yourself or not.
> > 
> > So let me wear my Anti-Terrorism suit and say "Blackmailing won't get
> > you further".
> > 
> > For what it is worth I am very open to changing the strings of kmdl to
> > kernel-module, as you already know:
> > 
> > http://lists.atrpms.net/pipermail/repo-coord/2004-March/000083.html:
> > > For getting a common versioning scheme I'd be willing to give up
> > > "kmdl", that's not a real problem.
> > 
> > But until now you were bundling it with the flawed versioning you use,
> > which is technically not acceptable.
> > 
> > So to make it even simpler, I [*] will "adhear" to a common naming of
> > kernel-modules independent of any of your actions, and I discard any
> > blackmailing by principle.
> > 
> > At any rate, please stop breaking repository compatibility
> > _unconditionally_. Get back to the community.
> > 
> > [*] I will delegate the decision to atrpms-devel, atrpms is not "Axel"
> > anymore.
-- 
Axel.Thimm at ATrpms.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.atrpms.net/pipermail/atrpms-devel/attachments/20040511/f39209a8/attachment.bin