www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alex Harui <aha...@adobe.com.INVALID>
Subject Re: [jira] [Commented] (LEGAL-304) BSD3 with nuclear clause
Date Tue, 16 May 2017 15:27:19 GMT
FWIW, when I read it quickly, my first impression was that it was an FOU
(as in you 'must not use' in a nuclear facility), but if I know it was
meant to be a liability disclaimer, I can read it again and see how it
isn't saying you 'must not use' it, just that they 'was not designed or
intended' and don't want the blame if it causes a problem.  However, I
have to wonder why the standard liability disclaimer (which is not one of
the 3 clauses but follows it in BSD3) wasn't good enough for the author.

But even if it we can convince the same lawyers Jim talked to, or a
court/jury, that it is just a liability disclaimer, the PMC should be
considering whether this license will affect the "consume-ability" of
their release.  This clause is the sort of thing that will keep popping up
over and over again on their dev@ list and cause some customers to shy
away.  It might be lower impedance to treat it as an optional dependency,
and/or request the dependency to adjust the wording on their license or
drop it, or switch to another license that has a better liability clause
built in.

IMO, it seems like the main problem to decide on is whether we will allow
PMCs to elect to have non-OSI approved licenses in their dependencies, and
secondly whether there is special wording that must go along with certain
of these licenses.  I think there is already consensus that PMC's can
choose non-OSI licenses.  Can we close on whether there is special wording
and close out all of these threads?  Then maybe we don't have to debate
the actual meaning of this nuclear clause.

My 2 cents,

On 5/16/17, 7:13 AM, "Jim Jagielski" <jim@jaguNET.com> wrote:

>The thing is that whether or not it is a FOU restriction will be
>decided by the courts. It will be determined there. So the opinions
>of FOSS-saavy lawyers on whether or not it be determined and decided to
>be a FOU restriction seem quite pertinent to me.
>Of course, all this matters iff we consider OSI approval as a condition,
>which it seems that we don't.
>That also is something that those self-same lawyers were "shocked"
>that we allowed for licenses that were not OSI and/or FSF approved.
>> On May 15, 2017, at 3:57 PM, Ted Dunning <ted.dunning@gmail.com> wrote:
>> A field of use restriction would say something of the flavor of "This
>>license is restricted ..." or "You may not exercise any of the rights
>>...". The clause in question doesn't do that.
>> What this says is that you acknowledge that the software was not
>>designed or intended for use. That doesn't say that you can't do it, nor
>>does it say that the license is limited to non-nuclear use. It merely
>>says that you say that you know that the original authors didn't design
>>it for such use and, by implication and conjunction with the liability
>>limitation, it is on your head if you cause a problem by so using this
>> I hate it when people say "Trust me", but in this case my source is
>>good enough that I really can say trust me on this interpretation.
>> On Mon, May 15, 2017 at 7:16 AM, John D. Ament <johndament@apache.org>
>> Just to make sure its on the thread, here's the text:
>> You acknowledge that this software is not designed or intended for
>> use in the design, construction, operation or maintenance of any
>> nuclear facility.
>> That is a field of use restriction.  Similar to what happened w/ the
>>JSON license, We are explicitly saying that the software cannot be used
>>as a part of a nuclear facility.  I read that as if the nuclear facility
>>published statistics to a tomcat server, tomcat cannot leverage software
>>with this license.
>> On Mon, May 15, 2017 at 10:10 AM Ted Dunning <ted.dunning@gmail.com>
>> Those lawyers didn't read the text very carefully.  This is NOT a field
>>of use restriction.  It was intended (according to my source who was
>>there at the time) as a liability warning.
>> On May 15, 2017 4:59 AM, "Jim Jagielski (JIRA)" <jira@apache.org> wrote:
>>     [ 
>>EfL7F4lCyel5JfNpooY8IuKYqwTcp9pgxLk%3D&reserved=0 ]
>> Jim Jagielski commented on LEGAL-304:
>> -------------------------------------
>> Oops. Seems I was premature. As OSCON I spoke w/ several lawyers about
>>it. Unanimously, every one considered it a FOU restriction since the
>>clear intent of the author/licensor is to prevent said s/w from being
>>used in a specific way.
>> > BSD3 with nuclear clause
>> > ------------------------
>> >
>> >                 Key: LEGAL-304
>> >                 URL:
>> >             Project: Legal Discuss
>> >          Issue Type: Question
>> >            Reporter: Tim Allison
>> >
>> > On LEGAL-44, a question was asked about whether BSD-3 with the
>>nuclear clause was acceptable?  Two conflicting opinions were expressed,
>>and the issue was closed because of a change in the license.
>> > On TIKA-2338, we'd like to move a a portion of a dependency that was
>>restricted to test-scope (according to LEGAL-37) to our regular
>>distribution because that portion has been moved to BSD-3.
>> > However, we noticed that this is BSD-3 with the [nuclear
>>E.txt].  Can we include this in our distribution under ASL 2.0?
>> > Is this a "field of use" restriction (which would lead to a "no"
>>answer) or is this an "acceptance of no liability" (which would lead to
>>a "yes" answer)?
>> --
>> This message was sent by Atlassian JIRA
>> (v6.3.15#6346)
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: legal-discuss-unsubscribe@apache.org
>> For additional commands, e-mail: legal-discuss-help@apache.org
>To unsubscribe, e-mail: legal-discuss-unsubscribe@apache.org
>For additional commands, e-mail: legal-discuss-help@apache.org

View raw message