[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Joey Hess is out?

Andrew McGlashan <andrew.mcglashan@affinityvision.com.au> wrote:

> I would rather it be done right [release...], than be done by an
> otherwise /not/ necessary arbitrary date for a freeze.

I'd like to remind you that not having a pre-set freeze date didn't go
so well in the past. Getting Sarge out for example took ages, because
there was no communicated date set by the release team and the
contributing DDs were in the dark as what to do until which date. Then
the freeze came and lasted for over 1 year which made everybody angry
and in the end, Sarge release with very outdated packages, because of
the uncoordinated freeze peroid.

You _have_ to set a freeze date and you have to set it really really
early so that every DD, DM and contributor knows what is comming when
and to plan the packaging schedule for that date.

You don't want to be in the middle of planning the upload of a new
upstream release which results in a library transition or the need of
binNMUs for dependant packages when, *plop*, suddenly the release team
drops the freeze on you.

Without an early set freeze date you will never get every package
maintainer (of over 38.000 packages) coordinated on a single date to
stop developing the next version and to just fix bugs.

No early set freeze date may have worked in the days of Woody with just
8.000 packages but in the current state, this will never work.


Sigmentation fault. Core dumped.

Reply to: