On Thu, 2005-10-13 at 11:04 -0300, Hernan Tylim wrote:
> Hi Göran,
> 
> [...]
> > Now, the idea was not to post about this until Ken and I had learned
> > more :), but I couldn't help it. What do you guys think?
> 
> I have  a question. One that I have since a very long time, but I always 
> feared to pop up one of these kind of threads.
> 
> What happens with the packages that nobody wants to steward ? I ask 
> because I think that nobody will take care of something that don't use, 
> or directly need.
> 
> So. If there will be packages with no stewards. What will happen with 
> the fixes and enh related to them ?

I want to be optimistic but I agree with you that it is unlikely that
all the classes will ever be covered by Steward teams and they certainly
will not all be covered from the beginning.  For that reason I do not
advocate completely dropping the current system.  I would prefer to slot
in the new system into the existing system.

Under the current system we effectively have 3 layers:

1. Harvesters

2. Reviewers (volunteers from the public, coming and going at will)

3. Developing public at large

My intention was to start the new system simply by slotting in the new
Steward teams in at level two:

1.  Harvesters

2.  Stewards and Reviewers

3.  Developing public at large

For an issue related to a class handled by a Steward team then the
stewards would take the report, do what's necessary, and forward the
results on to the Harvesters.  For an issue related to a class not
handled by any Steward team the current system would still hold and the
community would be responsible for a fix and giving reviews.  In this
way there is no collapse as we transition from one system to the next
and if (as is quite possible) a complete transition is not possible then
we maintain a system that works at least as well as what we have today.

Now you didn't address this issue, but since it has come up, I don't
want to ignore it:  Some Steward teams from day one will have the
confidence and experience to be able to produce quality fixes and they
should be able to submit updates directly and not require time from the
Harvesters.  I would update my diagram but this gets a bit beyond my
ASCII art capability (and patience).  Over time as other Steward teams
develop they would then be given direct access to the update mechanism.
Perhaps in time most if not all teams will have direct access.

But the need for Harvesters will not end there.  Back to your original
issue it is unlikely that there will be a Steward team for every issue
and we need a fallback.

Ken
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20051013/db5ab4e7/attachment.pgp