new packages freeze policy

[Scott Kitterman]

On Thursday 30 August 2007 15:51, Stefan Potyra wrote:
> Hi,
>
> it seems that there has been a misunderstanding of the new packages freeze
> (NPF) policy between archive admins and the MOTU team: While MOTU's assumed
> that any package in the new queue until new packages freeze will get
> processed, Tollef explained that the packages freeze for archive admins
> mean that new processing is merely stopped at NPF (apart from exceptions).
>
> In order to clear out this misunderstanding, I'd like to ask for opinions
> on what the ideal new package freeze policy should look like.
>
> From a MOTU point of view, the gist of NPF is to switch focus from
> reviewing packages to fixing bugs and thus having more time to get universe
> into shape for release.
>
> Hence I'm in favour of having everything processed that lands in the new
> queue until NPF.
>
> Additionally I can see these pros and cons for each model:
>
> * stop processing new after NPF
>   + relieves duties of archive admins from new processing at a distinct
> point in time.
>   + archive admins don't need to wonder wether a package is ok to process
>     if it sits in NEW after NPF
>   - as the delay of packages passing/getting rejected from the new queue
>     varies, people don't have certainty, if their package is looked at when
>     getting it finally uploaded.
>
> * process everything that hit new until NPF
>   + people aren't disappointed, in case they got a package uploaded in time
>     but it didn't get processed any more
>   + hence gives a reliable deadline
>   - opposite of pros of above
>
> If it should be a problem (or a burden) for archive admins to sort out
> wether a package is OK to process, e.g. motu-uvf could provide a list of
> packages in the new queue at the point of the freeze (as they're dealing
> with exceptions anyway).
>
> Of course the question right now is also what we'll be doing for Gutsy. I
> guess I should also state that some contributors worked hard to get their
> package into shape recently, and that deferring these packages in new right
> now would thus lead to at least some frustration.
>
> Cheers,
>     Stefan.

I have conspired with the other 40% of MOTU UVF that's awake now (3 of 5 
people) and so I think we have a quorum.  I'm going to file a blanke UVFe for 
all the NEW Universe packages that were uploaded before the official Freeze 
(0000 UTC 30 Aug - i.e. ~20 hours ago).  Once I have an ack from another 
motu-uvf, we'll subscribe the archive admins in the usual way.

People who had packages uploaded after that may file a UVFe request and 
motu-uvf will look at them and make a decision using the same highly 
scientific processes we use for regular UVFe processing.

I would ask the archive admins to process these packages through NEW on a best 
effort basis (knowing they all have other stuff to do).

For the future, I would suggest that the archive admins and the MC get 
together and set a New Package Freeze data early enough that the archive 
admins will commit to reviewing the packages before Hardy Herron's release 
(no guarantee stuff gets approved, of course).

Thanks,

Scott K