[Fedora-packaging] Re: Kernel Module Packaging Standard Teleconference

[Thorsten Leemhuis]

Ville Skyttä schrieb:
> On Wed, 2006-08-16 at 12:43 +0200, Axel Thimm wrote:

> and that the
> scope of the discussion needs to be limited to whether uname-r gets
> moved to the packages' names and its direct unavoidable consequences --
> the only real technical design issue.

I agree that this is the only "only real technical design issue" of the 
current standard. But I'd don't think that we hastily should change this 
detail. Two weeks should be enough to look at this precise problem 
closer and work out and test the solutions (one solution of course is 
uname-r, but I *currently* think it's not the best one -- especially not 
for RHEL, where the uname -r might be confusing). I roughly another way 
how this might be fixed in
https://www.redhat.com/archives/fedora-packaging/2006-August/msg00086.html
already.

>  Everything else in the "kmdl"
> proposal is more or less cosmetics and implementation details, and
> adopting it would mean throwing away quite a bit of 

s/a bit/a lot of/ IMHO -- it were probably around 15 to 20 IRC meetings 
IIRC (probably even more) and a lot of traffic on the mailinglists where 
FESCo (including jeremy and spot) discussed each and every detail in 
depth. We even evaluated most of the stuff that's made different in the 
kmdl scheme during that phase. We for example looked at a debuginfo 
solution similar to the one in Axels kmdls and chose not to use it; we 
looked at a one srpm approach for both userland and kernel-module and 
decided to split. That are -- as scop outlined above -- implementation 
details that are already used in some packages in Extras, under Review 
for Extras and another repo that's using the same scheme. We IMHO 
shouldn't change them without a good reason.

> work (reinventing
> the wheel from the POV of the current scheme and its adopters) from
> several parties for questionable gain.

Otherwise I'm strongly in agreement with that paragraph.

[...]

CU
thl