Re: Thoughts on GPL's Appropriate Legal Notices? or the CPAL?
On Thu, Dec 15, 2011 at 08:20:59PM +0000, MJ Ray wrote:
> Can you remember what arguments support the idea that "Powered by
> SugarCRM" is an author attribution?
>
> Especially in cases like this, I think the man on the Clapham omnibus
> would probably assume it referred to the running software, rather than
> the non-author licensor - it's not a Matrix-style or BlackMirror-style
> human-powered world. This seems deeply counter-intuitive to me and
> it's not in the FAQ.
I don't have a good response to this, other than to say that I
remember having the same thought and convincing myself to my own
satisfaction that I was being too strict. 4.5 years later I think you
make a reasonable argument.
I am not sure that the "Powered by" aspect of the issue was considered
in great detail by the FSF apart from my own thinking about it. I
recall that RMS was convinced that an "indicator-of-origin logo" was
logically equivalent to an author attribution.
> > SugarCRM relicensed to GPLv3 shortly after the FSF's final publication
> > of GPLv3. The additional licensing terms concerning the logo and so
> > forth that were in place at the time met with the FSF's approval.
>
> This raises two questions that I can't follow now:
>
> 1. Anyone know where is that approval documented?
No, but it's no secret; I've spoken about it publicly before, and I
would suppose SugarCRM publicly pointed it out (I seem to remember
criticism arising at the time from some users of the 'community'
edition of SugarCRM). One could contact the FSF for some confirmation
I suppose.
> 2. What were the terms in place around 29 June 2007 compared to now?
I don't have it at hand but I imagine what was in place for the
initial GPLv3 version is still publicly available. I don't know if the
terms have changed and I haven't bothered to check, have had no need
to check (I don't think there's been any interest in attempting to
package SugarCRM in Fedora, which I imagine might represent a daunting
technical packaging challenge if not a legal one), and haven't wanted
to check (since it would bother me if there's been restrictiveness
creep).
The released language of the 2007 SugarCRM terms reflected some
editing by me (acting on behalf of the FSF which was my client). Their
original language, written with the near-final draft of GPLv3 in view,
was overreaching. I can't recall all the issues but one I remember
quite clearly was the issue of the basic nature of the Appropriate
Legal Notices requirement (which I remember well had, in its GPLv2
version, long been thoughtfully criticized by some on
debian-legal). The original SugarCRM language assumed that they could
legitimately require preservation of the "Appropriate Legal Notices"
in the form in which it was received by the licensee. That is not what
GPLv3 (or the counterpart provision in GPLv2) says; rather, a licensee
can be required to perpetuate the legal notices, but it could be done
in an entirely different way from the way chosen by the upstream
licensor. So, yes, the FSF believed that a logo-preservation
requirement might be a "reasonable author attribution" in some cases,
but one could choose to display the logo in a different way from that
chosen by upstream. SugarCRM accepted this.
> > As an example of how I've applied this in practice, working with the
> > Fedora Project, we succeeded in getting Zarafa to revise its AGPLv3
> > additional terms to scale them back to approximately what SugarCRM had
> > been doing.
>
> I thought Zarafa relied on a trademark.
> I don't know the Fedora
> Project well enough: where might I find that work described? (So far
> I found https://bugzilla.redhat.com/show_bug.cgi?id=498194 but it's
> not specific about this aspect.)
We had some non-public discussion with Zarafa's lawyers, so this isn't
fully documented. SFLC got involved in that discussion too, I believe
representing the FSF (this was after I'd joined Red Hat of course). I
don't remember the details (indeed there's a chance I'm
mischaracterizing the specifics when I say scaled back to the SugarCRM
approach), but I'll try to dig them up and publish what I can
somewhere.
Thanks, and sorry if this is mostly or entirely off-topic.
- RF
Reply to: