[License-review] Approval: BSD + Patent License

Kevin Fleming kevin+osi at kpfleming.us
Thu Mar 30 20:41:04 UTC 2017


Yes, that's frequently how list archives operate.

On Thu, Mar 30, 2017 at 4:37 PM, Smith, McCoy <mccoy.smith at intel.com> wrote:

> OK well then maybe the attachments are just not getting posted to the mail
> list archive (which makes more sense to me).
>
>
>
> *From:* License-review [mailto:license-review-bounces at opensource.org] *On
> Behalf Of *Kevin Fleming
> *Sent:* Thursday, March 30, 2017 1:05 PM
> *To:* License submissions for OSI review <license-review at opensource.org>
> *Subject:* Re: [License-review] Approval: BSD + Patent License
>
>
>
> The attachments were *not* scrubbed by the mail server; both are visible
> in my copy of the message.
>
>
>
> On Thu, Mar 30, 2017 at 3:35 PM, Smith, McCoy <mccoy.smith at intel.com>
> wrote:
>
> All:
>
> The BSD+Patent license is in a state where we would now like to request
> Board approval (preferably, at the April board meeting).  We have
> incorporated changes to the license (originally submitted on the list in
> January, 2016), and have received many helpful comments which have resulted
> in changes to the original draft.  Some comments on the latest version, and
> the text of the latest version, are below.  We have also attached an odt
> copy of the license with color coding indicating the origin of the language
> in the license (as well as a "clean" version of the license language in a
> separate document), although I think that document might be getting
> scrubbed by the mail server (if there's a good way to post the doc so that
> doesn't happen, let me know).
>
> Given that the original submission was so long ago, I've also included
> answers to the original questions (per the requirements of the approval
> process) on the license that were answered upon original submission.
>
> RATIONALE:
>
> This license has been created to address a specific problem that has been
> encountered with existing licenses:  the desire of certain organizations to
> have a simple permissive license that is compatible with the GNU General
> Public License (GPL), version 2, but which also has an express patent grant
> included.
>
> PROLIFERATION CATEGORY:
>
> This license should be categorized as a “Special Purpose License,” or,
> alternatively, as an approved variant of the BSD 2-clause license.
>
> DISTINGUISHING FROM OTHER OSI-APPROVED LICENSES:
>
> The BSD 2-clause license is currently categorized as one of the “Licenses
> that are popular and widely used or with strong communities.”  The BSD
> 2-clause license is also considered by the Free Software Foundation (FSF)
> to be compatible with both GPLv2 and GPLv3.  However, some organizations –
> particularly those with large or valuable patent portfolios – are reluctant
> to use the BSD (and MIT) licenses because of the lack of an express patent
> license grant in those licenses.  Because of that issue, many patent
> holders prefer to use the Apache 2.0 license when they wish to license code
> permissively, since it has an express, and well-written, patent license
> grant.  Unfortunately, the FSF has determined that the Apache 2.0 license
> is not compatible with GPLv2 because “it has some requirements that are not
> in that GPL version. These include certain patent termination and
> indemnification provisions.”
>
> There is currently only one GPLv2-compatible permissive license that
> includes an express patent license – the relatively-recently approved
> Universal Permissive License (UPL).  In the past, this has created a
> problem for some authors wishing to license their code permissively, but
> also including an express patent license grant, and also having the license
> be GPLv2 compatible so that users have the ability to integrate that code
> into GPLv2-licensed code (including, for example, the Linux kernel).
> Although the UPL provides such an option, we believe that users will still
> wish to have a license with more familiar, and widely accepted licensing
> language (like a BSD-variant) that solves the issue of an express patent
> grant and GPLv2 compatibility.
>
> We believe that this license is non-duplicative (for the reasons set forth
> above), as it solves an issue all but one OSI-approved license solves
> (permissive, express patent license grant, GPLv2 compatibility).  The
> license itself is a combination of language from the BSD 2-clause license,
> the Apache 2.0 license, and the Eclipse Public License (all of which are
> already OSI-approved, and all of which are in the category of “Licenses
> that are popular and widely used or with strong communities”), with a
> handful of editorial changes for consistency and clarity (and to remove
> features that would make the license GPLv2 incompatible).  It therefore
> should meet all the criteria of the OSD.
>
> LEGAL REVIEW:
>
> The text of the license is reproduced below.  We have also included an
> attached mark-up of the license showing where the language comes from BSD
> 2-clause, Apache 2.0 and Eclipse.  The patent grant from Apache 2.0 and
> Eclipse were used because the combination of the two was more
> tightly-worded and fit better with the existing language in the BSD
> 2-clause license.  Note that we have removed the “patent
> termination/retaliation” language from both of these licenses because this
> language is one of the reasons why Apache 2.0 has been found to be
> incompatible with GPLv2.
>
> In keeping with the License Approval Process submission guidelines:
>
>         1. We have read the Open Source Definition and ensured that this
> license complies with it (and in fact, this license is an amalgamation of
> already OSI-approved licenses).
>         2. This is an Approval submission, as this is a new license
> (albeit derived from existing OSI-approved licenses).
>         3. We have appropriate standing to submit this request, as this
> license was created by us (again, derived from existing OSI-approved
> licenses).
>         4. We are subscribed to license-review.
>         5. This communication is our formal request to license-review.
>
> Please approve this license for inclusion on the OSI approved license list.
>
> Commentary:
>
> 1.  A primary goal of this license is to reproduce, identically, the text
> of BSD 2-clause given the popularity of that license for those wishing to
> grant permissive rights.  Other than the patent clause, and two small
> editorial changes (the adding of a "DISCLAIMER" header above the disclaimer
> paragraph, and correction of a typographical error in the disclaimer
> section to recite "CONTRIBUTORS" instead of "CONTRIBUTOR"), this license
> includes an exact reproduction of BSD 2-clause.  Several commentators have
> pointed out ambiguities in the language reproduced from BSD 2-clause; to
> the extent those exist, those exist in BSD 2-clause and this license will
> also contain such ambiguities.   Those ambiguities have not impeded the
> adoption and success of BSD 2-clause.  Other than to include an express
> patent license that will be GPLv2 compatible, it is not our goal to
> "improve" the language of BSD 2-clause.
>
> 2.  In crafting the patent grant, we have tried to use, as much as
> possible, terminology that already exists in BSD 2-clause.  Thus "copyright
> holders and contributors" and "this software" are used in the patent grant
> as a result of the use of those terms in the disclaimer section of BSD
> 2-clause.  We have adapted the language of the Apache and Eclipse patent
> license grant, upon which the patent grant is based, to conform to that
> terminology.
>
> 3.  The patent grant is primarily based on the language in Apache.  We
> have, however, included some language from Eclipse that we thought useful
> to tighten up or clarify the Apache grant:  1) the clarification in Eclipse
> that the patent grant for combinations is measured at the time a
> contribution is added; 2) the disclaimer of a grant for other combinations;
> 3) the general disclaimer of other licenses.
>
> 4.  One feature we added to the Apache patent grant from GPLv3 is the
> recitation that the license patents are those "already acquired or
> hereafter acquired."  Although Apache doesn't explicitly say this in its
> text, the FAQs on Apache say that those are the patents that are licensed (
> http://www.apache.org/foundation/license-faq.html#PatentScope Answer #2),
> so we thought adding that clarification in the language was within the
> spirit of Apache's patent grant.
>
> 5.  We have not added an express right to sublicense (several on the list
> believe the license should include that).  Neither BSD 2-clause, nor Apache
> (in its patent grant) nor Eclipse (in its patent grant) recites a right to
> sublicense.  The license itself is styled as a direct grant from the
> copyright holders and contributors to those "exercising rights under this
> license" so sublicense rights should not be needed.  Given the long history
> of BSD 2-clause and its use in permissive licensing (including conversion
> of BSD 2-clause code to other licenses, including GPLv2), we did not
> believe reciting a right to sublicense was needed or within the spirit of
> reproducing, as much as possible, the language of BSD 2-clause (and the
> patent grant of Apache/Eclipse).
>
> 6. As some suggested upon the original submission of this license, we have
> received confirmation from FSF that this license is GPLv2 compatible.  Once
> the license is approved by the OSI board, we will be sending that approval
> to the FSF for consideration of adding the BSD+Patent license to their list
> of GPL compatible free software licenses.
>
> McCoy Smith
> Law & Policy Group
> Intel Corporation
>
> License text reproduced below:
>
> ============================================================
> ============================================================
> ============================================================
> =====================================================
> 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.
> ============================================================
> ============================================================
> ============================================================
> =====================================================
>
> -----Original Message-----
> From: Smith, McCoy
> Sent: Sunday, March 12, 2017 5:18 PM
> To: 'License submissions for OSI review' <license-review at opensource.org>
> Subject: RE: Approval: BSD + Patent License
>
> 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.
>
> _______________________________________________
> License-review mailing list
> License-review at opensource.org
> https://lists.opensource.org/cgi-bin/mailman/listinfo/license-review
>
>
>
> _______________________________________________
> License-review mailing list
> License-review at opensource.org
> https://lists.opensource.org/cgi-bin/mailman/listinfo/license-review
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensource.org/pipermail/license-review_lists.opensource.org/attachments/20170330/e3a51912/attachment.html>


More information about the License-review mailing list