Re: Debian doesn't have to be slower than time.

[Michael Neuffer <neuffer@sciobyte.de>]

Quoting Colin Watson (cjwatson@debian.org):
> On Sun, Feb 17, 2002 at 12:41:44PM +0100, Michael Banck wrote:
> > On Sun, Feb 17, 2002 at 10:23:20AM +0100, Michael Neuffer wrote:
> > > Most of the times however it is simply a question of getting all
> > > affected packages rebuild. And this is exactly where we have most 
> > > of our problems. 
> > 
> > There are auto-builders. There's an auto-builder for i386, too. What
> > would be the difference between a build on s390 and a rebuilt on i386?
> > (and all other arches)
> > 
> > If it's really only a rebuild thats needed, I guess we could trigger
> > that now, no?
> 
> I often say that getting the packages rebuilt is the bulk of the work,
> but I don't really mean that the mechanical process itself of rebuilding
> is what takes the time. The bulk of the work lies in making the change
> to the package maintenance scripts, testing the change, discovering that
> the package only actually built properly on the maintainer's system in
> the first place, fixing that, making sure that the fixes don't break
> anything else unforeseen, and so on. It usually doesn't take long for a
> single package, but it's all human effort that has to be multiplied up
> several hundred times.
> 
> An autobuilder can't help significantly with much of this, unless we
> stop testing packages and declare that we no longer care about quality,
> which I don't think is the case.


There are two things:

1. Packages should not only build on the maintainers machine
   but they should be able to compile in a "standard" environment
   to begin with.
   We can ensure this by forbidding binary uploads.
   Source only uploads have to pass the "autobuilder test" to
   be added to the incoming queue.  
 
2. The output of the autobuilder should be send to the maintainer.

With these two things we make sure  that we indeed DO care about 
quality, even more then we do now.

Junichi Uekawa tried to rebuild the packages and already started 
filing bugs. With this automated he wouldn't have to do this menial,
tedious and and often unthankful task manually.

Cheers
   Mike

-- 
People often think of research as a form of development -- that it's 
about doing exactly what you planned, doing it on time, and doing it 
with resources that you said you'd use.  But if you're going to do 
that, you have to know what you are doing, and if you know what you 
are doing, it isn't really research."
             --Dave Liddle, The New Yorker, Feb. 23/Mar.2, 1998, p84