incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ross Gardler <>
Subject Re: Radical revamp
Date Wed, 18 Aug 2010 14:23:54 GMT
On 18/08/2010 14:25, Carl Trieloff wrote:
> On 08/17/2010 07:48 PM, Ross Gardler wrote:
>> On 17/08/2010 17:35, Carl Trieloff wrote:
>>> To this question, (what will make this model succeed or fail) I find
>>> myself not coming
>>> to defensible answers... I would love to see thoughts of others on this
>>> question.
>>> The thread implies it comes down to the 3+ members on the project.
>> It's not that simple for me.
>> A single member who either knows all they need to know or (more
>> likely) knows when they need to ask for guidance from their peers is
>> all that is needed if (and only if) the IPMC trusts that member to do
>> the right thing.
>> However, there are a number of reasons why a single mentor does not
>> work out including some fundamental requirements for binding votes on
>> certain activities.
>> For me success or failure comes down to the quality of the mentoring
>> and the willingness of the project committers to learn how to apply
>> the Apache Way to their community.
> Ross, I buy that, however what guidelines would IPMC use to allow a
> podling to use this model? Does this mean we have 'lesser' and 'greater'
> members or mentors?

It is nothing to do with "greater" or "lesser". It's to do with 
respecting different people with different approaches.

 > Or is it a podling asking to use the model and then the IPMC deciding 
 > based on knowing who is on the podling PPMC to allow it, or if it is 
 > open to any podling, how do we as the IPMC understand that the
 > project is ready to graduate?

Different mentors are better at helping different types of podlings. 
Some people are detail heavy (good for those with little understanding 
of the Apache Way), others are a bit slapdash* (good for those already 
most of the way there).

I think Greg suggested that it should be the board who decide if a 
project should go straight to TLP with mentor oversight (the radical 
revamp proposal). I'd hope the opinion of the IPMC would have the 
opportunity to influence this decision.

As I said elsewhere I'd love for the IPMC to focus more on training and 
supporting mentors than on ticking boxes for podlings. Knowing which 
mentors are the appropriate type is as important as knowing which 
projects are the appropriate type for this model.

As for how do we know when it is ready to graduate, this is still 
something that has not been addressed, and I don't feel able to address 
it now. I suspect it can't be addressed until we have tried it a couple 
of times.

My gut response is that if a project is trusted to go straight to TLP 
with mentors then we should also trust the mentors and project 
committers to know when it is time to graduate (there are a few 
requirements that should be present e.g. an audited release put out 
there using the correct infra@ processes).


* so as noone takes offense let me say I'm thinking of my own approach 
when I say "slapdash" ;-)

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message