[License-review] Approval: Server Side Public License, Version 1 (SSPL v1)

[Kyle Mitchell]

On 2018-10-18 15:10, Jim Jagielski wrote:
> > On Oct 17, 2018, at 9:44 PM, Eliot Horowitz <eliot at mongodb.com> wrote:
> >
> > We think it is time for an extension of copyleft.
> >
>
> If so, then this list and the OSI are NOT THE PLACE for
> this type of discussion, IMHO.
>
> The OSI is about determining whether a license abides by
> the OSD or not, not about determining the extent, value or
> viability of copyleft. As such, it should be abused in
> such a way as to "sneak-around" those to whom copyleft is
> of prime and vital importance in hopes of getting a
> license approved which "redefines" or "extends" copyleft.

As the last person to propose a radical copyleft license to
OSI, I agree with you, functionally.  Many knowledgeable
participants post on license-review, but license-review is
not the best or even a particularly good way of soliciting
or receiving their input.  Not for the one giving, and not
for the one receiving.

Neither is it a very good place to try and revise a license.
My compliance with the published request to propose early
and revise with feedback spawned complaints about revision
control that outlived discussion of the license.   And came
again to naught.

Substantively, I strongly disagree.  Copyleft is a fun name
for a category of license condition.  It belongs to any
particular organization no more than "public license" does.
Those who prefer particular implementations---and their
compromises---should play their hands, not claim house
privilege to make newcomers play their game at their odds.

As for OSI, judging by track record, no organization has
approved so many licenses expanding copyleft.  Whether we
call those "copyleft" or "reciprocal" terms is a matter of
terminology, or branding, not substance.  The substance is
making "free for free software only", which is how these
licenses are largely received and read until their
vulnerabilities become widely known, mean what it says,
despite changed legal and technical circumstances.  OSI has
arguably done more to maintain and legitimize radical
copyleft than any other organization.

Circa 2002, OSI was apparently eager for the job, and
specifically for corporations dual licensing software.
Those licenses remain on the books, though FSF rejected
several of them.  When I ask, I do not hear about any
appetite to recant the difference of opinion to mend the
rift.  After all, those licenses were essential to bringing
open source into corporations, starting with smaller ones
where allies enjoyed more discretion.  It was in part on the
basis of extending open source as a breeding ground for
business models accessible to upstart firms that the first
wave of firms came on board.  There is still a great
business use case, too.

This is meta, but it's also directly responsive to Eliot's
question about whether it's enough for _some_ valuable part
of the community to want a license, or whether every part of
the community has to sign off.  OSI's early welcome to
upstart dual licensors contributed to its success
proselytizing larger, incumbent firms.  Now incumbent
software companies are well represented through OSI, and
their views come through on this list.

The answer to Eliot's question that I've read in my own
experience here is that the list must reach consensus---of
whoever pays the time and attention toll to chime in---to
even put the license on the board's agenda.  Insofar as
large, established software companies must be part of that
consensus, many will not approve stronger copyleft licenses,
either for software freedom activists, or for their natural
allies in copyleft, dual licensing upstarts.  The purpose of
those licenses, for either part of their user base, is to
defend their users' efforts against incumbent software
companies, by the leverage of intellectual property.

Case in point, Mongo has drafted and applied a license with
specific, but unnamed, competitors in mind.  Those
competitors know who they are.  They're free to chime in
here, and to block consensus.  That time and attention toll
must pale in comparison to what they've spent reading into,
and through, AGPL.

-- 
Kyle Mitchell, attorney // Oakland // (510) 712 - 0933