Re-posting to License discuss
On 2022-12-13 2:26 p.m., Mike Milinkovich wrote:
On 2022-12-12 5:40 p.m., Pamela Chestek wrote:
On 12/12/2022 2:01 PM, Mike Milinkovich wrote:
Pam,
Just to pipe in as a practitioner, not a lawyer. One of the major
differences between EPL-1.0 and EPL-2.0 was the removal of the
license's choice of law provision. We spent something like 15 years
arguing that the certainty provided by the choice of law provision
was valuable. I don't remember winning that argument once in all
that time. In my experience, both adopters and contributors viewed
the choice of law as a negative.
I would also add that if you sort open source licenses by usage you
will find that something like 95%++ of all free and open source
software is currently made available under licenses which do not
have a choice of law provision. So as a purely practical matter I
consider this debate settled in favor of do not have a choice of law
provision.
I don't disagree that it seems disfavored, but I was curious why,
especially after someone challenged me and I didn't have a good
answer. And the reason seems to still be ... just because?
Pam,
I am coming at the question purely as a guy who for many years was
essentially the steward of a license that had a choice of law
provision and found it an unpleasant experience.
In my view, the purely pragmatic answers to "why are choice of law
provisions in open source licenses disfavored" are:
1. Many lawyers don't like them. In my experience there were lots of
lawyers who found the EPL-1.0 USA-centric because of its choice of
law provision and avoided it as a result. E.g. why would a German
automaker want to contribute code under a license that stipulates
US law when they go to great lengths to shield their company from
US law? Telling them that the lawsuit could still proceed in a
German court did not give them much comfort.
2. Many lawyers don't pay attention to them. For example, I can think
of multiple instances where lawyers insisted that the EPL-1.0 was
a strong copyleft license because it relied upon the US Copyright
Act's definition of derivative work (as indicated by the choice of
law provision) rather than defining the term in the license. So
instead they read the EPL-1.0 using the FSF's interpretation of
derivative-work-includes-linking. Suffice it to say we vigorously
disagreed with their interpretation, but we were never able to
change their minds, even after pointing them directly at the US
law. So for EPL-2.0 we cut-and-paste the definition of derivative
work from the US law into the license to fix that. Go figure.
To summarize: as a practitioner, I found the EPL-1.0 choice of law
provision to be a barrier to both contribution and adoption. Because
of that direct personal experience, I believe such provisions are a
bad idea.
And at the risk of belaboring the point, I do think that think the
fact that 95%++ of all FLOSS code is published using licenses without
a choice of law provision is a valid point and should not be dismissed
as a "...just because". Further, I agree with Bradley on his point
that if someone thinks that there are issues with the existing major
licenses under German law we should be fixing /that /problem. I think
approving a new license in 2023 that includes a choice of law
provision is ... quaint. IMHO, it is an anachronism. Probably not a
fatal flaw for approval because the OSI and/or this list has never
come out firmly against such clauses. But I am sure that the Open
Logistics Foundation relies upon large heaps of software using
licenses which do not specify German law. Relying on that software
while believing there is a fundamental flaw in their licenses seems
like a contradiction, no?
_______________________________________________
The opinions expressed in this email are those of the sender and not
necessarily those of the Open Source Initiative. Official statements by the
Open Source Initiative will be sent from an opensource.org email address.
License-discuss mailing list
License-discuss@lists.opensource.org
http://lists.opensource.org/mailman/listinfo/license-discuss_lists.opensource.org
