[License-discuss] dual licensing and the Open Source Definition

Tue Dec 17 19:34:58 UTC 2013

On 12/14/2013 02:21 PM, Engel Nyst wrote:
> Hello,
>
> On 12/12/2013 11:46 PM, zgilboa at culturestrings.org wrote:
>> As per the Open Source Definition, commercial use of Open Source
>> software must be permitted, yet "the license shall not require a
>> royalty or other fee for such sale."
>>
>> One interesting side-effect of the above is that software can be
>> released under a strong copyleft license, for instance the GPL, and
>> yet be accompanied by the option to "buy one's way out of the
>> license," thereby releasing the buyer from any and all obligation to
>> make the modified source available to the public.
>
> I'm not sure there's causality between the two. Commercial use must be
> permitted (not denied, as in "non-commercial use"), and, there are other
> business models than those based on copyright restrictions (fee per 
> copy).
>
> Commercial services around the software, customizations at user's place,
> are a few other possibilities. They are alternatives to 
> fee-per-copy-otherwise-restricted traditional models.
> This is a major point, in my opinion, that should be understood.
>
> However, indeed this isn't happening in the scenarios you're referring
> to: proprietary-GPL dual licensing or open core business models are a
> weird case that has emerged in practice. It's quite common place today;
> I say weird not because it's uncommon (it's not) but for several other
> reasons, among which one similar to yours if I understand you correctly:
> this particular dual licensing doesn't seem to serve the goals intended
> by the license (not by OSI). Whether you consider these goals focused to
> assure software freedom/openness or to promote alternative business 
> models.
>

Indeed, the purpose of my question was to find out whether a single 
license could guarantee that the source remains available in all 
scenarios, and also ensure that using the library would be free for all 
end-users, yet entail a (one-time) charge for vendors of 
commercial/proprietary software.  I think there are several advantages 
(both practical and technical) to using a single license over multiple 
licenses, but at the same time consider such advantages to be secondary 
to offering at least one license that is OSI-approved.

>> 1) to what extent does the GPL meet the OSI promise regarding the
>> source of Open Source Software remaining open? After all, if vendors
>> can take GPL'ed software and buy their way out of the license so that
>> binaries, with or without changes, can be distributed without
>> restriction and without a corresponding source, then something is
>> probably not working the way it was originally intended.
>
> It's an interesting perspective you seem to have, that GPL may be "at
> fault" on this point. I'm not sure I disagree, depends what exactly 
> you mean and where you're looking for improvements; but again, it has 
> nothing to do (AFAICT) with OSD/OSI. Rather, I can see why among all 
> licenses, GPL and AGPL are "favorites" in this game.
>
> Copyright system is such that a license like GPL, which requires a
> strong leveling of the playing field for contributors (nobody can use
> copyright to restrict works/derivatives), is exposed /more/ in a sense
> to copyright privilege when a single owner adds up for themselves
> separate rights from all contributors. When one can what no one else can
> (as opposed here to permissive licenses, where everyone can restrict
> derivatives), that one has more power.
>
> That's a philosophical matter, though, and it doesn't mean at all that
> the GPL by itself doesn't work to keep software open. GPL and many 
> other licenses make the software open source.
>
>> Then again, it seems to me that the possibility to regulate one-time
>> charges for commercial use from _within_ a license should be much
>> preferred over a de facto option to bypass the license altogether.
>
> See above on business models. (and please see below for another option)
>
> On 12/13/2013 10:02 PM, zgilboa at culturestrings.org wrote:
>> The framework in which I am interested would allow a distinction
>> between derivative work in general, and changes to a specific library
>> in particular.  In other words, I'm looking for a license that would
>> guarantee source availability of all patches that were applied to an
>> open source library, but not necessarily of the code that uses the
>> interfaces which that library exports.
>
> This goal (and only it) sounds like you may want to take a look at MPL
> or LGPL. With MPL it's simple, files that contain code of the original 
> work are covered by copyleft. Outside files (even within a "fork" of
> the library/framework, though, note), are not bound to any condition.
> Depending how you interpret derivative works (in copyright law, and add
> the intention of LGPL licensors), LGPL copyleft seems to cover the
> library/framework, but not the software using its API. The actual scope
> is disputed though.
> Also, there are others with approximately this scope.
>
> On 12/13/2013 09:46 PM, zgilboa at culturestrings.org wrote:
>> Here, too, my issue is with the idea that an effective open source
>> license needs to be "upgraded" in order for software to become
>> profitable for its copyright holder, or convenient for the vendors
>> of commercial software that would like to use it. I am also not sure
>> that a license that allows the modified source of a library to no
>> longer be available is an "upgrade" to begin with, but that is of
>> course subject to interpretation.
>>
>
> I sympathize with the sentiment. In any case, I don't think another 
> license that is proprietary is an improvement.
>
>> 2) Consider the case of an individual entrepreneur who created a
>> software library, and who would like to require vendors of
>> commercial products that _depend_ on that library to pay a _one-time
>> fee_, but otherwise be permitted to use the library or distribute it
>> in any way they see fit without additional charges, and provided that
>> the original source code, along with all changes that were applied to
>> it, remain available to the public. Would that author be able to
>> release his/her library under an OSI-approved license? Having gone
>> through the various licenses on the site, I was unable to identify a
>> single license that adequately meets this scenario.
>
> Licenses that require a fee per copy from commercial entities are not 
> compatible with the Open Source Definition.
>
> You may be interested in another experiment done with licensing to
> address something similar enough (AFAIK): in some licenses, for a
> limited time, like one year, proprietary derivatives are allowed, then
> when the time elapses, their source is supposed to become automatically
> open, under a copyleft license. You may want to look at this Transitive
> Grace Period license[1] for an example (actually the best example I 
> know of, others are proprietary or intended to; this isn't).
> Please note that the license itself is not OSI-approved (the software 
> using it is dual-licensed with GPL AFAIK, so there is still the GPL 
> path for downstream).
>
>
> [1] https://github.com/zooko/tahoe-lafs/blob/master/COPYING.TGPPL.rst

Thank you for making so many excellent points.  The model based on a 
transitive grace period is very interesting, even though it still 
requires dual-licensing for full compatibility with other open source 
projects and libraries.  One way or the other, then, it might have to be 
a dual-licensing model that offers the GPL in conjunction with the 
license of choice...