Bug#622371: transition: webkit

[Michael Gilbert <michael.s.gilbert@gmail.com>]

Kumar Appaiah wrote:

> Hi.
> 
> On Wed, Apr 13, 2011 at 09:05:06AM -0400, Michael Gilbert wrote:
> > Steve McIntyre wrote:
> > > Right. Are you deliberately trying to stall all development, or does
> > > it just appear that way?
> > 
> > I fail to see how proposing a more measured approach to testing
> > security for two very problematic packages qualifies as stalling all
> > development.
> > 
> > Yes, it delays newer versions of those packages and their dependencies
> > for a while, but I don't see how that is a real problem.  The transition
> > is still going to happen. Testing of those packages is still going to
> > happen. A year is a very long time.  I fail to see why two years must
> > be viewed as a requirement.
> > 
> > "Stall" is the wrong kind of framing since that connotes that some kind
> > of catastrophe (plane crash) is about to happen.
> 
> To the best of my understanding, testing security is provided on a
> best case basis. Given that testing is not as secure, why not just
> provide a lower priority to webkit from the testing security
> perspective while allowing the newer version to get more exposure
> before the next release, rather than introduce the transition at the
> end?

That is what will end up happening as soon as this transition starts.  I
will only have enough time to work on stable security updates.

My suggestion is not to move the transition to the end, but instead to
the middle.  I don't see why it has to be at the start.

Best wishes,
Mike