[License-review] Request for Approval of Universal Permissive License (UPL)

[Henrik Ingo]

On Sun, Apr 13, 2014 at 8:55 PM, Jim Wright <jim.wright at oracle.com> wrote:
> The idea here is that the original authors of the Software are not generally going to be contributing to everything, but more often just to one thing or a set of things defined at the beginning - as I mentioned in my response to Karl, my first intended application is the Java Community Process.  If the authors all agree, they can edit the file later, but they must all agree before they could do that just like they would need to agree to otherwise alter the license grant (e.g., by just editing the text of the license).  In general, I don't realistically see these files getting edited much, but rather set to a scope at the beginning of a project, for the purpose of allowing the contribution to something that was planned for in advance.  It's doable, but any time you want to change the scope of a license, if you have multiple contributors, this can be kind of a pain and the license does not solve that problem.
>

Your own words here should have been a strong clue for you that this
license fails basic assumptions of open source, like elaborated in
more detail by Christopher. The whole point of open source is that you
do share your code with (potentially) everything, and a recipient can
use it for anything without asking for permissions from all original
authors. (At least under the same license, sometimes even under a
different license.)

Note also that everyone who commented failed to understand the clause
about the LARGER_WORKS.txt. The reason seems clear now: what you're
trying to articulate simply didn't make sense in an open source
license.

You seem to argue in other responses that for copyright purposes this
is still a valid open source license, it is just the patent grant that
is limited to some specific scope. In that case however the
justification for approving this license is undermined from a license
proliferation point of view: This license doesn't add anything new
when compared to existing permissive licenses.

Especially I should note that many people will argue that MIT and BSD
licenses do include an implied patent license due to words such as
"use" or "sell" (...while some argue the opposite, there isn't a
consensus.) Hence your proposed license is not a better alternative,
as you're *explicitly restricting* patent rights to a given scope, not
really granting them.

In addition of course use of this license with an open source
copyright license but restrictive patent grant would create a lot of
confusion in the open source community and should therefore probably
be recommended against.

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

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