[haiku-development] Re: Request to work on newer buildtools integration (GCC/Binutils)

  • From: Ingo Weinhold <ingo_weinhold@xxxxxx>
  • To: haiku-development@xxxxxxxxxxxxx
  • Date: Tue, 03 Nov 2009 10:39:01 +0100

On 2009-11-03 at 06:56:04 [+0100], David McPaul <dlmcpaul@xxxxxxxxx> wrote:
> 2009/11/3 Ryan Leavengood <leavengood@xxxxxxxxx>:
> 
> > On the note of commit access voting, what is the process for handling
> > negative votes? Is one enough to stop the process, or is it down to a
> > high majority (maybe 80% positive or something)?

For votes on any proposals a simple majority is required (i.e. the sum of 
all cast +1, +/-0, and -1 should be > 0). I don't think we've ever 
discussed a different procedure for votes on commit access. We could raise 
the hurdle for commit access, but I don't see a convincing reason.

> Niels has made some good points that really apply to anyone who wants
> to work on Haiku but I think that if someone wants to write code for
> haiku and a current developer thinks it is worth granting them access
> then I say go for it.

Well, I don't really agree with this. There should always be a phase for 
would-be committers to prove that they are "worthy" to gain commit access. 
That means they should have provided enough patches to allow any developer 
to check that this is the case.

I totally agree with Niels that what we've seen from Joe so far doesn't 
really justify commit access yet, but I think we should make an exception 
in this case, since updating the buildtools cannot reasonably be done 
without commit access (you have to import into the vendor branch, merge 
into the trunk, resolve conflicts, apply additional changes) and none of 
the developers who have worked one the build tools so far (Oliver, Michael, 
and myself) aren't really motivated to do that. So it's either no updated 
tools or giving Joe a chance.

CU, Ingo

Other related posts: