Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [technology-pmc] Formalizing some practices


On Thu, Feb 26, 2009 at 5:42 PM, Wayne Beaton <wayne@xxxxxxxxxxx> wrote:
I think that Eric's point is that if you make it so that you have to work to a number, than there'll be gaming. I feel that any attempt at quantifying will ultimately fail.

Agreed.
 


In the end, there has to be a demonstration of real merit. Participation in bugs, dev list, newsgroups, are all such demonstrations. Maybe it's enough to state that the candidate must demonstrate real merit and give suggestions on how that can be shown.

I tend to avoid using negatives, but we might consider explicitly stating that it doesn't matter if you've been hired to do the job; you still have to prove to the community that you're worthy of the appointment.

This is already covered in [1]:

There must not be any hint of "we (company W) hired person X to work on project Y thus person X should elected a committer".

..so repeating it should do no harm.

[1] http://wiki.eclipse.org/Development_Resources/HOWTO/Nominating_and_Electing_a_New_Committer
 


Wayne

Abel Muiño wrote:
Maybe a sum of several numbers would make a best criteria?

I usually feel that community members who collaborate on the dev list to add
greater value as committers than members that "just fix bugs", because they
are involved in the project communication and have a vision on its future
(of course "bug fixers" are still very valuable).

Also contributors of big chunks of code (something that needs IP review and
adds new functionality) look like better future committers even if the only
interact with a single bug report. Besides, they tend to be active on the
dev lists too (at least, for researching the info they need for the
contribution).

So let's say a committer canditate must score 10 points to be elegible.
- Questions on dev list: 1 point (maybe with a cap... questions show
interest, but a committer who only asks is not really a good candidate).
- Answers or suggestions (dev list or user newsgroup): 2 points
- Bug patches or test cases: 2 points
- Big contributions (require IP review): 5 points

(numbers are more or less random).

On Thu, Feb 26, 2009 at 4:53 PM, Eric Rizzo <
erizzo-eclipse@xxxxxxxxxxxxxxxxxx> wrote:

 
On 2/25/2009 11:23 AM, Wayne Beaton wrote:

   
===Electing Committers===
The Eclipse Development process allows a project to set its own criteria
for electing new committers. The Technology PMC requires that all
committer nominations include specific rationalization of the merit of
the individual. At a minimum, we expect that all committer nominations
include a list of at least three bugs (including ids) that the candidate
has worked on.

     
Like the others, I'm concerned about putting a specific number of bugs as a
requirement. Seems to me that will inevitably lead to attempts to game that
system. I would suggest simply omitting the "at least three" part from the
last sentence.

Eric

_______________________________________________
technology-pmc mailing list
technology-pmc@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/technology-pmc

   



 ------------------------------------------------------------------------


_______________________________________________
technology-pmc mailing list
technology-pmc@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/technology-pmc
 
_______________________________________________
technology-pmc mailing list
technology-pmc@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/technology-pmc



--
Abel Muiño - http://ramblingabout.wordpress.com/

Back to the top