[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: ZFS in Buster



[No need to Cc an extra copy, I've been a d-d subscriber since...
the 1990s?]

On 2019-06-08 13:00:02 -0400 (-0400), Sam Hartman wrote:
> Jeremy Stanley <fungi@yuggoth.org> writes:
> > Your earlier message also implied the motives behind
> > Conservancy's recommendations to be something other than a
> > desire to protect projects relying on free/libre open source
> > software licenses from making costly mistakes. Suggesting that
> > their interpretation of these licenses is driven by an ulterior
> > motive strikes me as a gross mischaracterization, particularly
> > in light of the ways in which they (individually and
> > collectively) have demonstrated a dedication to core values of
> > software freedom over the years.
> 
> I actually think that the SFC has a strong motive to defend
> copyleft. Certainly their leaders have taken strong pro-copyleft
> positions.
> 
> I don't think it is negative to them to describe them that way at
> all.

I'll concede, I may have been reading too much into Aron Xu's
accounting of the in-person debate, but it seemed like the
implication was that Conservancy was working in their own self
interest in this matter, and not simply out of a desire to protect
the (possibly tenuous, as you also indicate) legal grounds
underpinning the GPL and by extension those projects who have chosen
the GPL to represent their needs.

> I think there are times when the desire to establish a strong
> copyleft is in conflict with the desire to accurately articulate
> the legal risks associated with something Debian might do.
[...]

I wholeheartedly concur, which is why I contested the grounds for
the suggestion that Debian was coerced into its current position
regarding ZFS integration with the Linux kernel (I was not there for
the face-to-face discussions, so all I have to go on is the
established reputations of the parties involved).

> And the fact is we don't know what would happen in the courts in a
> lot of situations. I attended a session at Libreplanet this year
> where a lawyer talked about what actual rulings the courts have
> made about the GPL and copyleft. It's not clear. Cases like Oracle
> V Google and the German Vmware case both surprised us in various
> ways.

Yes, I see the VMware result as a sort of victory. While it's
unfortunate that the German courts effectively dismissed the appeal
on technicalities without weighing in on the issue at stake, at
least the vendor has agreed to cease future flagrant violation of
license terms by removing Linux components from their proprietary
software (even if this does nothing to address their past violation
the Linux contrbutors' legally-documented wishes with regard to
their copyright). But point taken, the choice to retroactively
underpin the free/open software movement (I have trouble seeing the
two in isolation from one another as some do) with copyright law is
akin to building a house on a foundation of sand.

> And there are cases where courts will listen to what happens in
> practice. If everyone or a lot of people are interpreting a
> license a particular way, that can potentially influence courts.
> Case law matters and case law is influenced by what companies
> actually do and how copyright holders and licensees actually use
> software.
> 
> > To be clear, I seriously doubt Conservancy (or more precisely,
> > the fine individuals within Conservancy involved in debating
> > this topic over the years) would have been "upset" if Debian
> > chose to act counter to their opinions. I'm quite sure they know
> > that they don't control the choices of the Debian community, and
> > are therefore not responsible for any additional risk that
> > Debian knowingly takes on itself or passes on to its users.
> 
> It's not the risk that Debian passes on to its users. It's what
> Debian says and what happens if that becomes part of an argument
> in a court case.
> 
> Also, even if it never comes up in court, Debian's actions could
> influence others. If Debian takes a strong position it makes it
> easier to argue that in order to get your software actually used,
> you need to interpret the GPL strictly. If Debian takes a weaker
> position then as a practical matter you can commercially succeed
> by releasing CDDL works that are combined with GPL works, at least
> until a court says you cannot.

While I respect that concern for case law and precedent may be at
the forefront of Conservancy's arguments and preferences in this
matter, I prefer to look at the underlying reasons for this. I
highly doubt they chose this stance for their own sake, but rather
for the good of software relying on the licenses with which they are
most closely affiliated. I've had plenty of discussions and been on
the sidelines for plenty more where these individuals left me with
no impression that they were motivated by anything other than the
proliferation and preservation of software freedom.

> I am quite certain that factors like this were considered by
> Debian. It was more than just the legal risks that were
> considered. I have not talked to the conservancy yet (although
> they did reach out to me when this issue came up again). I have
> read some of the correspondance between the DPL at the time and
> the FSF and I can assure you that choosing an interpretation of
> the GPL that defended copyleft was something the FSF cared about.
[...]

Of this I have no doubt, but to imply that Debian's choice was
anyone's besides its own does a discourtesy to the project as well
as to the entities implicated in purportedly forcing such a
decision. This was the only real thrust of my prior reply.
-- 
Jeremy Stanley

Attachment: signature.asc
Description: PGP signature


Reply to: