[License-review] Approval: BSD + Patent License

Tue Mar 14 21:11:11 UTC 2017

In light of Luis’s comment, which is valid and spotted a drafting flaw, I’ve slightly revised the draft (in the attached, and directly below) to put a clearer condition of revocability on the patent license.  I’ve used language more consistent with BSD’s language (using the term “condition”) although have not used “breach” as that’s not terminology used in BSD.
I’ve also removed the square bracketed portion, as per the discussion below, I believe they are worth retaining (absent someone convincing me they cause problems I don’t currently see).
Consider this the latest and greatest version of the license.  I’ll let any further comments come in for a few weeks, then post something on the list indicating what the final version of this license is and asking for its formal approval for adding to the list.

McCoy

=========================================================================================================================================================

Copyright (c) <YEAR> <COPYRIGHT HOLDERS>

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

1.      Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

2.      Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

Subject to the terms and conditions of this license, each copyright holder and contributor hereby grants to those receiving rights under this license a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except for failure to satisfy the conditions of this license) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer this software, where such license applies only to those patent claims, already acquired or hereafter acquired, licensable by such copyright holder or contributor that are necessarily infringed by:

(a) their Contribution(s) (the licensed copyrights of copyright holders and non-copyrightable additions of contributors, in source or binary form) alone; or

(b) combination of their Contribution(s) with the work of authorship to which such Contribution(s) was added by such copyright holder or contributor, if, at the time the Contribution is added, such addition causes such combination to be necessarily infringed. The patent license shall not apply to any other combinations which include the Contribution.

Except as expressly stated above, no rights or licenses from any copyright holder or contributor is granted under this license, whether expressly, by implication, estoppel or otherwise.

DISCLAIMER
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

=========================================================================================================================================================
From: License-review [mailto:license-review-bounces at opensource.org] On Behalf Of Smith, McCoy
Sent: Monday, March 13, 2017 2:50 PM
To: License submissions for OSI review
Subject: Re: [License-review] Approval: BSD + Patent License

A good comment, and it raises a somewhat interesting point.
I replicated that language from Apache.  In the original draft, I had taken it out, given that I interpreted it to only apply to the defensive patent termination provision in Apache, which I of course removed from this license to ensure GPLv2 compatibility.
Someone (I think Jim Wright) pointed out that having the patent license be revocable when the other conditions of the license are not satisfied (which in the case of BSD, I think would mean only the attribution and license notice provisions) might be a bug, as someone could argue that they receive the benefit of the patent license without having to comply with the (fairly minimal, in the case of BSD) obligations of the license.  As a result, I put the conditional revocability back in, but using the same language as Apache.  Which as you point out, is ambiguous or doesn’t capture the intended meaning (which is as described above).
You are correct in the intent with that parenthetical.  Your language might be a solution, although I might see if there is another way to phrase it.
From: License-review [mailto:license-review-bounces at opensource.org] On Behalf Of Luis Villa
Sent: Monday, March 13, 2017 1:42 PM
To: License submissions for OSI review
Subject: Re: [License-review] Approval: BSD + Patent License

Hi, McCoy-
One question: "irrevocable (except as stated in this section)" - I don't see any possible revocation clause in that section (paragraph?). Did you intend something like "irrevocable (unless the license's conditions are breached)"?
Luis

On Sun, Mar 12, 2017 at 5:19 PM Smith, McCoy <mccoy.smith at intel.com<mailto:mccoy.smith at intel.com>> wrote:
All:

It has been over a year since the last discussion on this proposal;  I would like to revive it again in view of the following:

1.  I have gotten some feedback from the FSF on the license.  At the time of initial submission of the license, several on the list suggested that, given the special purpose of this license -- to provide a version of the BSD license that was both GPLv2 compatible and included an express patent license, I should get confirmation from the FSF that it is indeed GPLv2 compatible.  I have heard from the FSF and they have confirmed the present draft [appended and discussed below] is GPLv2 compatible.

2.  I have also gotten feedback from others on the last draft (on-list, and privately), as well as from the FSF,  which has resulted in me reconsidering the last draft and tightening up and/or altering the text somewhat (not to change the substance, just to change ambiguities or other drafting issues).  So the latest version (reproduced below, as well as appended in color-coded format to show the origin of the text) is in this e-mail.

A few drafting notes on this version:

A.  There are two provisions in this draft that are in square brackets ([]).  Those represent provisions that I have added in response to the comments of others on this list, but which I haven't yet decided should be in the final version.  The first provision (that the licensed patent claims are those "already acquired or hereafter acquired") comes from GPLv3 -- but is not in any of the licenses (BSD, Apache, Eclipse) from which my license is derived.  That provision is intended to make explicit that the license is intended to cover patents acquired by the licensor after the grant has been made.  I think that that is inherent in the grant language of Apache and Eclipse (and the Apache FAQs say it is explicitly), but the language does have the benefit of making it explicit and part of the license text itself.  The second provision is an explicit disclaimer of all other licenses beyond those expressly granted.  That language comes from Eclipse, with modification.  I'm now of the mind to keep in the bracketed provisions, given that I think being express/explicit is better, even if there is an argument that these provisions are somewhat unnecessary given the unbracketed language already covers the concepts.  I'll leave these provisions open for additional comment by anyone who wants to convince me there is a good reason to leave them out.

B.  The patent license grant is on behalf of "copyright holders and contributors."  I have used that language as a result of its existing use in the disclaimer section of 2-clause BSD, which I interpret to be intended to cover those who make copyrightable contributions as well as those who might make contributions not subject to copyright (an issue discussed in some depth in separate threads about military and/or government open source licensing, particularly those not subject to US copyright rights).  In order to make that distinction more clear, the patent clause now has an embedded definition of "Contribution(s)" which intended to now capture both types of contributions -- copyrightable and non-copyrightable.  Although I tend to disfavor embedded definitions in a license of this type (since BSD is free of them), this was one definition that seemed to be needed as a point of clarity.

Further comments are welcome.  Now that the GPLv2 compatibility issue is resolved, I'll leave this one open for additional comments from the member list for a bit,  in case anyone sees further drafting improvements they'd like to suggest -- or express objections as to why the license itself should not be approved.  I will then indicate that the text is finalized and ask for Board approval to add to the OSI list.

McCoy Smith
Law & Policy Group
Intel Corporation

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensource.org/pipermail/license-review_lists.opensource.org/attachments/20170314/87d58a7a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BSD + Patent markup vs. BSD & Apache & Eclipse v 4.odt
Type: application/octet-stream
Size: 30779 bytes
Desc: BSD + Patent markup vs. BSD & Apache & Eclipse v 4.odt
URL: <http://lists.opensource.org/pipermail/license-review_lists.opensource.org/attachments/20170314/87d58a7a/attachment.obj>