Voting: repotag for EPEL

[Thorsten Leemhuis]

On 05.04.2007 11:04, Axel Thimm wrote:
> On Thu, Apr 05, 2007 at 10:19:29AM +0200, Thorsten Leemhuis wrote:
>> I don't care what they decide, but in this case I think it's [FPC's]
>> business [...]
>> People [of the EPEL committee] that vote should know what they vote
>> about to make the decision, and that's not the case here afaics.
> You need to decide, do you make it FPC's call or not? You can't have
> it both ways, e.g. put the FPC through making a decision and then
> turning it down.

Sure.

> And the FPC is not a political organ, EPEL is (in this context). Don't
> pass on the Old Maid to the FPC. The decision on whether or not to
> introduce repotags is ours, and that's all what the voting proposal is
> about.

I disagree. The decision about to introduce repotags is IMHO one that
needs to be done by EPEL and FPC together, as the decisions, if repotags
are worth the trouble IMHO depends on how the technical solution looks
like and what kind of trouble it creates for the package maintainers.

That just fair to everyone (including the FPC) as we can't simply dump
work on the FPC like this.

> Consider the EPEL/FPC relation here like management and
> engineering. The manager decides, the engineer turns the decision into
> bits.

So "Manager=EPEL" and "Engineering=FPC"? Then I strongly disagree. The
Fedora Board is the only manager in Fedora-land afaics. FESCo is some
kind of manager for SIGs/Groups below it (like EPEL and FPC), but also
does a lot of engineering.

EPEL and FPC further are on the same hierarchy in Fedora-land, and they
can ask each other kindly to do something, but they can't say "you have
to do foo now because we want it like that".

I other words: FPC is no dumping ground for realizing something EPEL wants.

> If you like to decide on implementation details within EPEL the FPC
> will be grateful to only get a ready EPEL-releated guideline to vote
> on.

I don't want do decide on implementation details. But the decision how I
vote if repotags are worth the trouble is influenced by the rough
concept how to realize them. And that's why I'd like to see that solved
first before the vote.

> [...]

Axel, the best thing at this point of time afaics is not a binding vote
with the question "Should EPEL carry a repotag". A more simple
questioning "Should we investigate further if we want to use a repotag
and how it could be realize it if we want to" would be much better.

My option on this whole issue, in case it matters:

"Should EPEL carry a repotag" -> I don't want to make that decisions
now, before I know if the gain is worth the pain (if there is pain).

"Should we investigate further if we want to use a repotag and how to
realize it if we want to" -> yes

CU
thl