> I estimate it would only
> take a couple days to get all of KDE 3.3 ready for release, besides
> being built.
Well, speak for yourself. :)
I had assumed that 3.2 was going to ship with sarge (given how close the
upstream 3.3 release was to the freeze, and given that the 3.3 upload to
unstable came completely unannounced). As a result, I had prioritised
my time to look at sarge bugfix issues, with a plan to work on my KDE
3.3 modules after the freeze.
There is a fair amount of work involved in polishing the packaging for
these modules (these being kde{addons,artwork,edu,sdk,toys,webdev}).
I'd estimate probably two days full-time work, which would be fine were
it not for the fact that I am being paid to do a full-time job that does
not involve KDE packaging. Assuming I were to do this in my usual free
time, I'd say it'll take me about a week.
(I should note here that I quite deliberately worked through my sarge
TODO list over the last several months, so that no last-minute rush
would be necessary. My last uploads that weren't closing newly found
bugs were in July.)
Anyway, after this week of work there is NEW processing to worry about.
Then it hits unstable, and only then will people really start to test it.
I'm sure people will step up to NMU these major upstream releases for me
since I can't afford to drop everything and work on this suddenly-urgent
3.3 packaging task that has come unannounced. However, given the help
that people have offered in the past with KDE packaging, I expect this
will probably consist of someone making sure the *.install files are
correct, dropping in some new control entries and saying she'll be right
mate. Even in the days where I took far longer to produce unofficial
3.2.0 packages than the other maintainers, I never saw anyone offer to
do a thorough license audit of my packages, or update the multitude of
kdesdk manpages, or convert all the icons to decent-looking XPMs for
users of other window managers. Amongst other things.
There's also the fact that 3rd-party apps have broken with KDE upgrades
in the past. There will be virtually no time at all for testing this
once NEW processing is done and all of KDE 3.3 has worked its way into
sid.
At any rate, I don't see how you can claim that finalising KDE 3.3 is
going to be faster than finalising 3.2. The security patches are
trivial to apply (update from BRANCH and/or apply the diffs that are
explicitly available on ftp.kde.org for 3.2.3), and as for the kmail
bugs -- have you checked with upstream as to whether the patches are as
complex as you say, or is this just speculation?
Anyway, I must stop writing, I have to get back to frantic last-minute
KDE 3.3 packaging in the event that we decide to push forwards instead
of back.
b.
