Re: RFS: windowlab (updated package)
Hello,
fredag den 9 april 2010 klockan 21:06 skrev Christoph Egger detta:
> On Fri, Apr 09, 2010 at 05:10:28PM +0200, Mats Erik Andersson wrote:
> > Dear mentors,
> >
> > I am looking for a sponsor for the new version 1.40-1
> > of my package "windowlab".
> >
> > The upload would fix the bug 575793.
> > The upstream author has resolved an otherwise blocking call in X.org.
>
> You're still having that problematic licensed bit of code. If you
> think it's no problem, could you please document that in
> debian/copyright?
>
This code is now incorporated by the upstream author of Windowlab,
Nick Gravgaard, nothing is my insertion any more.
Gravgaard stated in a response to the Ubuntu bug report that he
altered Kilgard's code. I searched through the new licensing and
file heads, but no mention of this new origin. The only pointer is
at the end of 'events.c'.
My impression this time, was that since Gravgaard accepts this
altered code, of expressed origin with Kilgard, he judges it to be
acceptable and compatible with the AEWM clauses.
The blocking bug that this code is intended to remedy is a common
trait of all unpatched descendants of Aewm, ever since glibc-2.7
or a recent X.org, of some reason unknown to me, began stumbling
on signals.
If you take these words of mine into regard, and your greater
experience with licenses, do you concede a suitable comment in
'debian/copyright' to cover the back for Debian GNU/Linux?
Clearly Gravgaard is not so concerned as we are.
In an almost palpable sense, I judge the publication of the bug
resolution to be a matter of some urgency, taking priority over
finalizing the exact phrasing of licenses, just to provide the
few users that do exist, a fully functional package. (I wonder if
I will have to pay hard time for this loophole phrasing?)
The Ubuntu team is presently triaging windowlab-1.39 and
windowlab-1.38 for this bug, and for Debian I would like
to use a further update to the next packaging 1.40-2, to
conscientiously think our copyright notice through again.
The upstream author seems to have lost his hesitance, so
this is now a matter for Debian to resolve on its own.
> +/* Copyright (c) Mark J. Kilgard, 1994, 1995, 1996. */
> +
> +/* This program is freely distributable without licensing fees
> + * and is provided without guarantee or warrantee expressed or
> + * implied. This program is -not- in the public domain. */
>
....
> You've now replaced
>
> build:
> dh build
>
> with
>
> build:
> dh @
>
> Any reason you don't do a general
>
> %:
> dh @
>
> that would fit all targets?
>
Yes. Performing that wildcard replacement for all targets, would
erase my efforts to set a priority for the window manager ranking,
thus setting it to an improperly low value instead of the chosen 40.
On the other hand, I am still reluctant to implement a debhelper
override for the install target, since that would mean using a
version of debhelper so recent that backporting to, or building
inside Lenny would be a mess.
MfG
Mats Erik Andersson, fil. dr
Abbonerar på: debian-mentors, debian-devel-games, debian-perl, debian-ipv6
Reply to: