[Rd] Blue Sky Statistics - GPL Violation?

[Andrew Simons]

Thanks Simon,

Yes, what stands out with BSS is the degree of integration with R.

"By contrast, pipes, sockets and command-line arguments are communication
mechanisms normally used between two separate programs. So when they are
used for communication, the modules normally are separate programs. But if
the semantics of the communication are intimate enough, exchanging complex
internal data structures, that too could be a basis to consider the two
parts as combined into a larger program."

... and them not providing any source code.

It seems to me to be a violation of the GPL, but i appreciate that is a
"courts decide" question.

There's nothing I can do about it, not being a copyright holder of any R
code, but I post here in case someone here wants to take a look. There's
pros and cons to pursuing GPL violations -- with one pro being that
everyone benefits from the release of additional GPL code.

Kind regards


On Wed, 29 Jun 2022 at 09:11, Simon Urbanek <simon.urbanek using r-project.org>
wrote:

> Andrew,
>
> I'm not a lawyer and this does not represent a legal advice nor do does
> this represent a view of any entity nor organization. The content of this
> e-mail is entirely speculative.
>
> I have never heard of BSS before your post, but based on a quick peek they
> do appear to distribute Python and R binaries in their app. Although most
> people avoid doing that to prevent legal challenge, GPL generally allows it
> as long as they publish the corresponding sources (R, all GPL packages
> etc.) along with the binary release and it is just a distribution
> convenience (mere aggregation) and not a tighter coupling. In plain speak,
> anyone should be able to replicate the open-source part of the distribution
> with the sources they release. Any front-end typically requires a back-end
> that links to R, that piece also has to be included in the published source
> as it is covered by GPL. I didn't check what they publish, but as long as
> it covers the above, I would expect that to cover the GPL requirements.
>
> It's generally not hard to de-couple the open-source back-ends from a
> commercial front end. Once IPC/pipes are involved the coupling is typically
> considered loose enough to not make the GPL viral (though, it's not
> entirely clear-cut may depend on the protocol etc.). GPL only guarantees
> that the company has to release the sources for the back-end part which
> embeds R, but not any front-ends that talk to it. There are actually quite
> good, stable open-source back-end solutions, so most companies avoid the
> issue by simply using those since then they don't need to worry about being
> responsible (that's why most big players let the user install R), but it
> seems BSS didn't go that route.
>
> Again, just my speculative interpretation of information available to me,
> not a legal advice. If you are interested, you should check on the sources
> they release with the binary (or provide if you ask) as that is the key
> aspect.
>
> Cheers,
> Simon
>
>
>
> > On Jun 28, 2022, at 12:06 PM, Andrew Simons <simos1990000 using gmail.com>
> wrote:
> >
> > Hello,
> >
> > I notice that v10 of BlueSky Statistics is quite tightly integrated with
> R,
> > but is not released under a GPL compatible license.
> >
> > Do people think this represents a violation of the GPL?
> >
> > Kind regards,
> >
> > Andrew
> >
> >       [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-devel using r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-devel
> >
>
>

	[[alternative HTML version deleted]]