[bug#45692] [PATCH v4 3/3] gnu: Add ZFS service type.

[Maxime Devos <maximedevos@telenet.be>]

[-- Attachment #1: Type: text/plain, Size: 2942 bytes --]

zimoun schreef op ma 06-09-2021 om 10:08 [+0200]:
> Hi Maxime.

Hi zimoun,

> 
> On Thu, 02 Sep 2021 at 22:57, Maxime Devos <maximedevos@telenet.be> wrote:
> 
> > See also <https://sfconservancy.org/blog/2016/feb/25/zfs-and-linux/>;.
> 
> Reading the Software Freedom Conservancy analysis, from my
> understanding, the issue is about distributing the resulting *binary*
> Linux+ZFS.

Previously I thought that ‘Is The Analysis Different With Source-Only Distribution?’
somehow did not apply here, though I cannot remember the reasoning well (maybe
my reasoning was bogus?).

However, SFC did note in that section:

‘Nevertheless, there may be arguments for contributory and/or indirect copyright
infringement in many jurisdictions. We present no specific analysis ourselves on
the efficacy of a contributory infringement claim regarding source-only distributions
of ZFS and Linux. However, in our GPL litigation experience, we have noticed that
judges are savvy at sniffing out attempts to circumvent legal requirements, and they
are skeptical about attempts to exploit loopholes. Furthermore, we cannot predict
Oracle's view — given its past willingness to enforce copyleft licenses, and Oracle's
recent attempts to adjudicate the limits of copyright in Court. Downstream users should
consider carefully before engaging in even source-only distribution.’

I'm completely unfamiliar with the notion of ‘contributory copyring infringement’
or ‘indirect copyright infringement’.  ‘Savvy judges skeptical at attempts to
exploit loopholes’ seems plausible to me in my inexpert opinion.

>   Other said, the distribution of the zfs.ko is an issue, not
> provide a way to build it locally.  Well, it does not appear to be a GPL
> violation, IIUC the SFC analysis.

I neglected that the terms of the GPL that come into play depend on whether
one is only distributing source code (*) or also binaries, and whether one is
distributing _modified_ source code.

I don't quite see a GPL violation anymore if we only distribute unmodified
source code.  However, what about freedom (1) and (3) (freedom to [...] and
change the program in source form and (3) distribute modified versions)?
I was going to write something here about why that would not be (legally)
possible because of the CDDL and GPL, but I forgot why it wouldn't be
possible.  Maybe it is actually possible?

(*) raid5atemyhomework noted that guix does _not_ distribute source code,
it only points to source code locations.  I don't quite agree.  From my
point of view, on whose server the source code is hosted is merely a
technicality, guix is just ‘out-sourcing’ the source code distribution.
Besides, ci.guix.gnu.org keeps a copy of the source code, and
(guix download) will try to download from ci.guix.gnu.org.

> Thanks for taking the time to carefully review all the aspects.

Greetings,
Maxime

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 260 bytes --]