Re: I'll be a son of a bitch.
On Wed, Apr 13, 2005 at 09:22:24PM -0700, Steve Langasek wrote:
> On Wed, Apr 13, 2005 at 06:44:33PM -0600, Ean Schuessler wrote:
> > I didn't post that status nor are we actively monitoring it. Someone from
> > Alpha needs to get proactive and run the ball if they care about that
> > machine.
>
> The problem there, however, is that there aren't actually any Alpha porters
> today. Alpha is port*ed*, past tense, with very little ongoing work; the
> people listed at www.debian.org/intro/organization are not currently
> involved (at least one of them is no longer active in the project), and
> those of us who take care of the alpha-specific code bits on an ongoing
> basis, like the installer, kernel, and bootloader, have no formal status as
> porters. We also have no authority over build daemons.
Well, I don't think it's quite that dire.
First, AFAIK, there is no such thing as a "formal status as porters."
Porters generally seem to be people that are active on the list for a
specific architecture and help to make things run on that arch. I
greatly appreciate the efforts you and others put in to keep the
installer, kernel, and bootloader going.
One nice fact about Alpha is that just about everything else Just Works
(TM). Alpha is an ancient port in terms of both the Linux and Debian
timeline. It was our first 64-bit port, and I think our second non-i386
port. There isn't a lot that refuses to build on Alpha these days, and
there also aren't many packages that have arch-specific bugs on Alpha
these days either. Those that do probably manifest themselves on other
64-bit platforms, too.
But it's misleading to assume that there are no other people that help
out with Alpha matters. As an example, I'm not terribly active on the
Alpha list, but I do have an Alpha, and have recently worked on obscure
exim4 bugs, helped out a bit with the GHC port, and found a working
ctx+util-vserver combination for Alpha and shared these observations
with the Debian maintainer.
> So it's all well and good to say that "someone from Alpha" needs to get
> proactive, but AFAICT, that's an organizational null pointer; and I think
> this is in fact part of why our ports have been so hard to corral for sarge,
> because "porters" only exist for new ports, and we have no other process for
> people to assume responsibility for the overall health of a port.
I think there are people, but they are less centralized than you would
prefer. There are always arch-specific bits, such as bootloaders. But
past that, there's little more to maintaining an active port other than:
1) Finding problems, reporting bugs, submitting patches
2) Answering questions on the mailing lists from developers that need
help solving an arch-specific bug
3) Making sure we have a working buildd and debian developer machine
And 3 is as much a DSA problem as anything.
-- John
Reply to: