[Bug 226377] Merge Review: rpm

[Max Spevack]

On Fri, 24 Aug 2007, Ralf Corsepius wrote:

> In second place, seeing rpm5 in Fedora (which could be read as Fedora 
> supporting rpm5) is threatening the rpm.org project.
>
> In third place, setting up rpm5 not to conflict with "Fedora's" 
> nominal rpm is a true technical challenge.

I would like to address these two points.

The RPM 5 situation has been a very complicated one, as there are 
several different elements involved.

ELEMENT #1 -- The Fedora Board (and therefore me) have said publicly 
that rpm.org is Fedora's upstream, and that Fedora's commitment to 
rpm.org is not something that is being questioned.

Most recently, this was discussed publicly here -- 
http://lwn.net/Articles/237700/ -- at the very bottom of the article. I 
stand by what I said.  Fedora is committed to rpm.org as its upstream.

ELEMENT #2 -- The Fedora Project needs to be open-minded and inclusive 
to new ideas, and alternative implementations of things.  What I have 
suggested to Robert in the past was quite simple:  If he has a package, 
and if it can merit inclusion into Fedora on its TECHNICAL MERITS alone, 
then why would I do anything to stop it?

In my opinion, there is a major difference between "allowing a package 
to be in the Fedora repository" and "using a specific package as the 
building block for the entire distribution". The first (allowing 
packages into the repositories) is a technical-level decision. The 
second (which packages are the core pieces of the Fedora distro) are 
both technical and political.

In my opinion, at this point rpm5 is clearly in the first category, and 
not the second.

I'm not in the business of telling people what they can or can not 
submit for inclusion into the Fedora repositories.  That is what our 
already-existing review processes are for.

I am in the business of (along with other folks) determining what 
comprises the core components of the Fedora distribution.  And in that 
regard, the answer is the code produced by the upstream that is RPM.org

--Max