[License-discuss] The per se license constructor

[Richard Fontana]

On Sun, Mar 17, 2019 at 6:02 PM VanL <van.lindberg at gmail.com> wrote:
>
>
> I think it gets back to the core purpose of the OSI: To be a steward for the OSD and to certify licenses as compliant with the OSD. There are many other good things the OSI *can* do, but that is the one thing it *must* do.
>
> So how does that get back to L-D and L-R? Well, the OSI board is not composed of lawyers. So L-D and L-R end up being systems for recommending (or not recommending) licenses for approval based upon various criteria.

That seems to imply that determining whether a license is compliant
with the OSD is task that is fundamentally within the competence of
lawyers in particular. I do not agree. The OSD (drafted by a nonlawyer
without access to legal advice, at least in its original form) is not
a conventional legal document -- it is not a court decision, statute,
regulation, treaty, constitution or the like. The OSD is a statement
of philosophy and policy aimed at nonlawyers.

> Those criteria seem to include:
>
> 1) Compliance with the OSD. This can have a more or less legal feel, but it seems necessary to identify whether particular licenses do or don't appear to comply with various parts of the OSD.

Okay, I understand the "more or less legal feel". It is true that
lawyers are generally trained (though, in general, not especially
well) to interpret legal texts in certain particular ways relevant to
the way the legal system works. And compliance of a license (itself a
legal text) with the OSD (again, not a legal text in any conventional
sense, but has some similarities to some kinds of legal texts) should
involve a kind of interpretation that resembles the sort of
interpretation at which, in theory (though often not especially in
practice) lawyers are supposed to be particularly competent. I would
argue that the work that most lawyers do bears little resemblance to
the activity of determining whether a license complies with the OSD.
On the other hand, a nonlawyer who is steeped in free software/open
source legal policy (and I know many people who fit this description)
might be much better qualified to interpret the OSD than the average
lawyer.

As I was recently saying elsewhere, lawyers often carry certain sorts
of biases and baggage (due to client interests, professional
specialization, mercenary aspirations, class and guild loyalties, or
otherwise) to open source-related work that can negatively affect how
they approach a task such as interpreting the OSD in relation to
software licenses.

> 2) The law. If the license is not legally sound, that is a good reason to recommend against it. Board members could vote to approve or deny based upon how persuasive those arguments are.

Perhaps. I don't think it is or should be a primary consideration, at
least normally. For example, you provided some arguments for why
SSPLv1 (I think it was v1) was not legally sound, and that was a
really useful contribution to the discussion, but I think those
arguments were ultimately less important than, for example, the
philosophical/policy question of whether SSPL was inappropriately
discriminatory in violation of OSD 5/6, or unreasonably burdened the
ability to use or distribute other software in violation of OSD 9.

> 3) Policy choices affecting open source usage or the open source community. Certain licenses may implicate policy issues that, on the whole, a board member could feel was detrimental to the community as a whole.

> 4) Community reaction. Licenses are community organizing tools, and having communities indicate a willingness to use a particular license is an strong indicator of quality or acceptability. Similarly, a strong negative reaction from community members could reasonably cause a board member to vote against a license.
>
> Thinking back on various discussions, I think that the most substantive discussions really ended up focused on one or more of these criteria. It also seems reasonable that criteria 1) and 2) are strongly legally inflected, but any member can have a well-reasoned opinion with regard to 3) or 4).

I would say 2) necessitates input from legal specialists, but not 1).

 Richard