[License-review] Submission of the Upstream Compatibility License v1.0 (UCL-1.0) for approval

Henrik Ingo henrik.ingo at avoinelama.fi
Tue Nov 29 10:13:43 UTC 2016

On Tue, Nov 29, 2016 at 1:43 AM, Richard Fontana <fontana at opensource.org> wrote:
> My general sense of the license-review discussion of the UCL, and this
> is confirmed for me by re-reading it now, is that opinion was mostly
> negative, with concerns being expressed about the asymmetrical nature
> of the license. This matches my personal view of the license. Thus I
> would recommend against approval.

For the record then: I didn't participate in this discussion,
precisely because I had nothing negative to say. OTOH I haven't
invested enough time to strongly endorse it either.

But on a general level I don't think that asymmetric licenses are
negative as a category, nor against the OSD clause against
discrimination. I have worked most of my career for companies that
accomplished the same goal by using *GPL as the OSI approved license,
and requiring a separate CLA for receiving upstream contributions
(from outsiders, which tend to be a limited set of contributors). The
difference in my opinion is that while these models can be rightfully
criticized, and also I have done so, there's no question that the GPL
is clearly compatible with the OSD. I think it is important for OSI to
acknowledge that OSD requirements are narrowly focused on the license,
and do not include:

 - requirement for a vibrant diverse open source community to actually
exist or grow
 - requirement for the software to be of any particular level of quality
 - requirement for the software to be cheap or free
 - etc...

And of course there clearly was never a requirement for the software
to only be available under a single license, or the same license to

The new contribution of this license seems to be that it tries to
capture within a single document both the open source license and the
upstream CLA (and I repeat, this is based on a superficial reading and
more to make a general point). IMO this is equivalent to the
established practice of using GPL and a separate CLA and cannot in
itself be categorically in violation of the OSD.

As for what to do with this submission: While I think it would be
wrong and dishonest for OSI to rule that this is categorically against
the OSD, if there's a strong view that this is nevertheless a negative
thing, maybe some of the following arguments are more suitable?

- We don't believe this license is useful, or that it is an
improvement over existing practice of having a separate OSI approved
license and separate CLA, and due to license proliferation concerns
choose to not certify it.

- While the license is OSD compliant from a narrow point of view, the
modern OSI is now only certifying licenses against some higher
standard, in order to promote good licenses that encourage commonly
agreed best practices. (This would be a major policy decision, but the
OSI is free to make such.)

- It could be argued that having a separate CLA, that contributors
must acknowledge by the act of signing it, is actually a preferable
process for asymmetric upstream contributions. It could very well be
undesirable to endorse a practice where a random contributor has given
away more rights than he himself gets by the mere act of sending a PR
to a github repo. IMO the established practice that you, for example,
agree to the GPL by way of contributing to a code base that is GPL, is
well justified because what you agree to give away is the same that
you received - and hypothetically understood to receive - by simply
using/copying the software. Even if you might not actually understand
the GPL, at least there's some fairness and balance that justifies
this process. OTOH it might be undesirable to create a situation where
developers contribute to some repository without understanding what
they are giving away, or even without understanding that there's an
asymmetric license in use. (For me personally this would be the
strongest, or only, argument against approval.)

- Any other technicality or detail prevents accepting the license.

To repeat: I didn't invest enough time to strongly endorse approval
either, this post is merely a reaction to the discussion about the
asymmetric nature being objectionable.


henrik.ingo at avoinelama.fi
+358-40-5697354        skype: henrik.ingo            irc: hingo

My LinkedIn profile: http://fi.linkedin.com/pub/henrik-ingo/3/232/8a7

More information about the License-review mailing list