I wonder if some of this could be solved by breaking the project into
sub-projects.
This would allow the "super-committers" to focus on those projects where
the risk of incorrect code being submitted is high or where the impact
of coding missteps would have the biggest impact on the overall
stability and reliability of the entire suite.
The projects where code errors are more easily detected or have only
local impact could be tested and reviewed by people with different skills.
It would also encourage a clear delineation between projects which would
reduce the scope of the impact of code errors.
It would also attract more reviewers and contributors since the scope of
one's involvement would be clearer and would allow people to get active
in areas that interest them.
At the moment it seems that the project is too big and too complex so
that the current committers have too many areas to watch and are unsure
about how to attract and vet new people.
It might also make the process of fiddling with the core much cleaner.
If a committer on a sub-project needs to have a change to the core, they
would have to request this and there would be a discussion about the
impact on the core and other dependent projects.
It appears that this might be missing in the current setup and some of
the "trust" issues have come about from experience where changes made to
support a particular business case have had impacts that the contributor
did not anticipate.
Ron
On 21/10/2014 9:27 PM, Pierre Smits wrote:
Scott,
I thank you for your patience and eloquence to explain your viewpoint as a
PMC Member regarding the subject to every participant in this community.
For sure, the willingness of every participant in this community to discuss
contentious or controversial issues with an open mind (and with the best
interest of the project at heart) is something that will have the consensus
of all within this community.
We understand that you are expressing only your viewpoint and concerns as
only one member of the PMC.
I wonder: is this low commit:review ratio you're talking of supported by
any kind of numbers? And is this complaint/concern you're expressing not
the result of the code of conduct for committers, or lack thereof? Why is
this now - while we are discussing how to get more committers - a reason
for concern? And is this a concern of all PMC Members?
We have had very little complaints about such code commits up to now. And
if there were any, these issues were resolved quite fast. Isn't it so that
committers review code patches by contributors? And that it is part of the
responsibilities of committers. But we also know that committers review
committed bugfixes by other committers seldomly. But we trust committers to
do the right thing when committing changes, don't we?
Are you now saying that the PMC is regarding this as something to be
concerned about? And that all within this community should be concerned
about this? That current committers don't apply due diligence when it comes
to committing changes? And that we must have some kind of super-committer
policing the committers? With as a result of not having enough
super-committers, the entire PMC feels that we must accept that not more
eligible contributors are invited to be a committer?
I wonder, given that you say that you don't speak for any of the other PMC
Members - except Jacopo, can each of the other PMC members share her or his
viewpoint on this?
The controversy regarding commits I can think of (at the top of my head) is
that the PMC allows major extensions (improvements) to be committed without
prior review.
The other controversy I can think of is that, while you are trying to
explain at great lengths how cautionary the PMC is with respect to inviting
new committers (and new PMC members), a contributor with only 178 postings
in the user ml, 114 in the dev ml, and about 11 patches submitted and 2
publications in a period of 6 years makes it to become both committer and
PMC member within the last 3 months of those 6 years..
Though welcome the addition is, this contradicts anything of the concerns
you expressed.
Again, I understand and accept that you are expressing only your viewpoint.
So again, I invite every other PMC Member to share her or his viewpoint as
well. So that the entire community can learn how this issue, this
controversy is regarded in the entire PMC.
As always, and discussing with an open mind,
Best regards,
Pierre Smits
*ORRTIZ.COM <http://www.orrtiz.com>*
Services & Solutions for Cloud-
Based Manufacturing, Professional
Services and Retail & Trade
http://www.orrtiz.com
--
Ron Wheeler
President
Artifact Software Inc
email: [email protected]
skype: ronaldmwheeler
phone: 866-970-2435, ext 102
