In "Re: Diaspora, politics, and MI", Chris G Demetriou
<Chris_G_Demetriou@ux2.sp.cs.cmu.edu> wrote: 
> > Jukka writes:
> 
> > >Now that I'm talking about these things I'm receiving personal flame mail
> > >from the Ones Who Know Better.  Very incouraging.  Someone says "he has
> > 
> > I don't know who the "Ones Who Know Better" are, but certainly nobody should
> > be sending you flamage.
> Despite the thoughts of a few people, flaming the NetBSD developers is
> not the way to encourage their cooperation or get your source code
> into the tree.

I think the problem evidenced by both parties here is lack of understanding
of how each person "fits" into NetBSD.  NetBSD is a community effort...
when somebody in the "community" writes a driver, suggests a bug fix, etc
they want to see it integrated so that the rest of the NetBSD users can
benefit from it.  It is the responsiblity of the NetBSD core group to make
sure that the drivers/bugfixes/etc fit the general paradigm of NetBSD, and
to attempt to keep the quality of the code in the NetBSD tree as high as
possible.

People writing drivers/etc should NOT expect that their changes will go in
immediately, but they SHOULD expect that they will go in sometime (or else
somebody from the core group will have a discussion with them about what
needs to be fixed in order to get the code into the tree).

The NetBSD core group, should review each piece of code that gets added to
the
tree, but should try not to take on more than they need.  If the person
submitting the driver/fix/etc says it works on XYZ systems that they have
tested it on, and that they've been using it for XYZ months w/o incident,
take that into account....don't think you have to go over the source with
a fine tooth comb.  Give the person submitting the code some credit...they
have just as much interest in having a stable, professional system as you
do.

It really pisses me off to see bickering and splintering like this.  There
are a boatload of driver enhancements/bugfixes/code cleanups that have
been submitted which have NOT been integrated.  In the interest of harmony
amongst everyone involved with NetBSD, there should be some relaxing of
restrictions here....the core group needs to start trusting more in the
user
who submitted the change, so that changes get integrated quicker, and
people
who submit the changes need to accept that it may take a while before the
port maintainer has a chance to look over the code, and get it moved into
the tree. The current system feels very much like a commercial entity...
the core group pretends to "own" the code, and user submit bug fixes which
the core group will look into "when it gets some time"...this is wrong. 
The
core group is no more important than the users submitting changes...they
just
have a different "job".

There is a happy medium out there, which won't overload the core group
with lots of code review, and yet will still maintain a high degree of
quality.  We just have to find it....and quit all this silly bickering.

-- 
   //////////////////////////////////////////////////////////////////////
  //  Scott Ellis   //   sellis@gi.com   //   sellis@rohan.sdsu.edu   //
 //////////////////////////////////////////////////////////////////////
// WARNING: This signature warps  time and space in its vicinity    //