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

Fri Aug 29 07:16:58 UTC 2014

This got lost inside the moderation queue. - RF

On Sun, Jul 13, 2014 at 08:29:18AM -0700, Jim Wright wrote:
> All,
> 
> Please see below some proposed revisions to the UPL in response to community feedback.  The comments I made to the JCP Executive Committee are somewhat specific to the use of the license in the JCP (though as I've said, we hope to use it elsewhere as well), and I don't want to move the JCP discussion to this list obviously, but the changes are of general applicability and I am forwarding it as-is just for the sake of transparency.  
> 
> After some thought, I believe that whether future versions of the Software or Larger Works are to be licensed is better left to the larger works designation, since this will permit either a license grant covering only the work as of the date of the contribution (therefore rendering the license closer to the way most of us interpret the MIT license in the absence of a larger works file, only with well defined patent grants), or a broader grant, depending on the desired application.
> 
>  Regards,
>   Jim
> 
> 
> Begin forwarded message:
> 
> > From: Jim Wright <jim.wright at oracle.com>
> > Subject: Proposal for EC
> > Date: July 10, 2014 3:39:49 PM PDT
> > To: Patrick Curran <patrick.curran at oracle.com>
> > 
> > Patrick,
> > 
> > Please transmit the message below to the EC on my behalf.
> > 
> >  Best,
> >   Jim
> > 
> > ----------
> > 
> > All,
> > 
> > I have listened to the questions and comments of the various parties in our recent meetings, and here is a proposal for use of a revised UPL and some background on my thinking as per our discussion.  The revised UPL reads as follows:
> > 
> > The Universal Permissive License (UPL)
> > 
> > Copyright (c) <year> <copyright holders>
> > 
> > Subject to the condition set forth below, permission is hereby granted, free of charge and under any and all patent and copyright rights owned or freely licensable by each licensor hereunder, whether an original author or another licensor, to any person obtaining a copy of this software, associated documentation and/or data (collectively the "Software"), to deal in both
> > 
> > (a) the Software, and
> > (b) any piece of software and/or hardware listed in the LARGER_WORKS.TXT file if one is included with the Software (each a “Larger Work” to which the Software is contributed by such licensors),
> > 
> > without restriction, including without limitation the rights to make, use, sell, offer for sale, import, export, have made, have sold, copy, create derivative works of (provided that this does not license additional patent claims beyond those covering the unmodified Software and Larger Works), display, perform, distribute, and sublicense the Software and the Larger Work(s) on either these or other terms.
> > 
> > This license is subject to the following condition:
> > 
> > The above copyright notice and either this complete permission notice or at a minimum a reference to the UPL shall be included in all copies or substantial portions of the Software.
> > 
> > THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
> > 
> > I am also attaching a redline against the prior version.  This would be used with a larger works file containing something like following: The Reference Implementation of JSR-XXX Including Maintenance Releases as defined In the Java Community Process Program.
> > 
> > There are three revisions to the license plus the larger works proposal here:
> > 
> >  - One, it removes the reference to future versions of the Software and Larger Works from the base license text and pushes this, for the purpose of the JCP, to the larger works file, and limits the scope of those future versions in that larger works file to the final RI and Maintenance Releases.  I deemed this ok for use in various other contexts as well, as you can recover all or some future versions of whatever you are seeking the license for by referring to it in the larger works file where appropriate.
> > 
> >  - Two, it clarifies that arbitrary derivative works are not necessarily licensed for additional patent claims that might be infringed by changes (most folks did not interpret the license this way but because a couple asked about it, we clarified).
> > 
> >  - Three, I added a minor language tweak adapted from a suggestion by Simon Phipps re: the possible absence of a larger works file (probably not relevant here but included for completeness).
> > 
> > This revision now permits the license grants from contributors to cover up through the final version of each RI, while *not* extending the grant so far as to major new releases which change the functionality substantially from the original JSR, which is a question that has been raised by HP and perhaps questioned by others.  JSPA signatories would still retain their defensive termination rights both to the extent not actively contributing to an RI, and with respect to independent implementations not based on the RI code.
> > 
> > Finally, regarding the questions on the Apache license, while this was addressed when this subject first came up many months ago, as well as on the OSI list at least in part, I will address it briefly again here.  The Apache license is widely regarded to be incompatible with the GPLv2.  The reasons why it is incompatible are not up for debate here, or really relevant in my view.  The consequence of this incompatibility is that software licensed under the Apache license may not be freely commingled with software licensed under the GPLv2, with or without the Classpath Exception added, and each addition of an Apache licensed component therefore creates confusion and additional risk for all involved.  People cut and paste code, there is debate about subclassing creating derivative works, etc., and the problems created by license incompatibility are something we need to avoid, and easily can in this case - the UPL provides both Oracle and other consumers with all the rights they need, is compatible with both GPL and commercial licensing, and is also flexible enough to be compatible with other licensing regimes down the line if that ever happens.
> > 
> > HP has asked about the possibility of modifying the Apache license for use for this purpose instead, but this is then not a standard licensing regime and at that point no different than modifying the MIT license as I did here other than to grant less rights to Oracle and to the rest of the community in the output.  Likewise, I have been asked repeatedly about defensive termination clauses, but this is not an effective defense against trolls, and has the obvious purpose of attempting to preserve patent claims against Reference Implementations or the platform.  Subjecting both Oracle and everyone else to the risk of losing the ability to ship RIs, needing to indemnify commercial licensees, etc. in a dispute between the parties sitting at this table seems ill advised at best and likely only to incentivize the wrong kind of behavior.  Clear and broad rights in the RIs are a more effective way to foster collaboration between all parties here, and as I said, Oracle would be contributing code under this license too.
> > 
> > In closing, I should note that I and my colleagues have spent quite a bit of time trying to figure a way to satisfy as many people as possible without giving away the farm.  One thing we discussed was simply having JSPA members & affiliate members grant licenses to Oracle with scope at the boundaries of the JSPA grant, and non-members able to contribute under the UPL.  I'm not sure any of the open source forges/foundations would want participation in this bifurcated manner though, even if the rule was that JSPA members are dual licensing under whatever terms we agreed on and an open source license like Apache, and non-members under UPL and Apache.  If anyone wants to think further about this and suggest something other than just another open source license with narrower grants, or bat it around a bit, I'd be very happy to do that.
> > 
> >  Regards,
> >   Jim
> > 
> > 
> > 
> 

> _______________________________________________
> License-review mailing list
> License-review at opensource.org
> http://projects.opensource.org/cgi-bin/mailman/listinfo/license-review