Stable Release Updates types proposal (was Re: Fedora Board Meeting Recap 2010-03-11)

[Kevin Kofler]

Rahul Sundaram wrote:
> It is interconnected in my argument and doesn't make sense to debate in
> parts.  If you avoid breaking ABI stability, you can avoid unnecessary
> churn and one of the benefits ( think resource cost - infrastructure,
> mirrors etc)  of that is users with low bandwidth systems being able to
> take advantage of Fedora more.  While you can always brush off any
> suggestion with a position of "take it or leave it", it is importance to
> recognize that there is room for improvement.    If we didn't care about
> people with low bandwidth systems, we wouldn't be having yum-presto and
> LZMA compressed RPMS  So claiming that users with such systems should
> just go away doesn't fit into the development efforts already made to
> accommodate such users.

The point of those technologies is to reduce bandwidth consumption WITHOUT 
destroying what Fedora is about. They're technical solutions to technical 
problems. They have no drawbacks. (Well, yum-presto can actually make 
updates take longer on slow CPUs, but it can be disabled, so this is not an 
issue.) The changes you and some others are advocating, on the other hand, 
turn Fedora into something entirely different, and throw away some of our 
unique advantages.

        Kevin Kofler