+1 to Marvin's "I hope that most projects won't bother" although there
needs to be something a little more than a blank piece of paper.

The best approach, IMHO, is to simply make it official that the project
adopts the same byelaws as project x, y or z. Pick an established project
that has a minimal set of stable byelaws and go on from there. Some
projects like to refer to the original project pages, others make a local
copy. Both approaches have their advantages.

The important thing to note is that, in my experience there is very little
that changes from one project to the next given that we try to minimize the
number of rules we wrap a project in. Probably the only item I see regular
disagreement between projects is how much merit is needed for committership
and then PMC membership but even this should not be codified into the
byelaws since projects tend to change the height of the barrier over time.
All we need is a way to decide if someone is a committer/PMC member.

Ross


Ross Gardler (@rgardler)
Senior Technology Evangelist
Microsoft Open Technologies, Inc.
A subsidiary of Microsoft Corporation





On 1 October 2013 08:13, Marvin Humphrey <mar...@rectangular.com> wrote:

> On Tue, Oct 1, 2013 at 7:50 AM, Jordan Zimmerman
> <jor...@jordanzimmerman.com> wrote:
> > Wow. I missed that. I'll work omit for Curator. I'd like to see the grad
> doc
> > updated to mention this.
>
> I hope that most projects won't bother.  We need rules because we need a
> framework to resolve disagreements, but bylaws are really hard to get right
> and every little flaw is an opportunity to waste time arguing over
> legislative
> minutia.  Projects which inherit as many well-exercised rules as possible
> from
> the foundation rather than creating their own rules ex nihilo get to spend
> less time debugging legalese.
>
> Marvin Humphrey
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
> For additional commands, e-mail: general-h...@incubator.apache.org
>
>

Reply via email to