[License-review] Submitting CC0 for OSI approval
Clark C. Evans
cce at clarkevans.com
Mon Feb 20 02:57:58 UTC 2012
On Sun, Feb 19, 2012, at 08:08 PM, Russ Nelson wrote:
> Thanks for pointing that out, Richard. I think we dropped the ball
> with that license approval. It would have been a *good* thing to have
> a freely copyable implementation of patented software.
Patent holders are more than welcome to release reference works
with a license that deliberately witholds patent rights. However,
it doesn't mean the OSI should be approving these sorts of things.
> And Carl is right ... our policy should be reflected in the OSD
> rather than being something you find out only when you submit
> your license for approval.
If patent licensing is off the table for the OSI, it becomes an
irrelevant Ostrich. I look to the OSI to review licenses for me
that are "safe" to use so that I don't have to be an attorney as
well as a software developer.
> Patent licensing is a complicated topic which deserves its own
> committee. Am pretty sure that tying a patent license to a public
> domain declaration is not a good solution. Nor is holding up approval
> of this license waiting for that committee a good idea.
Most of the people who would use CC0 are trying to make an outright
gift to society, that they are permitting their work to be used without
any threat of infringment. To this extent, including an abandonment of
patent rights for the work they are releasing seems quite appropriate.
It's just horrible for a patent armed author that wishes for an OSI
approved license to push their "reference implementations" with.
On Sun, Feb 19, 2012, at 08:10 PM, Russ Nelson wrote:
> Clark C. Evans writes:
> > The problem with the CC0 is that it could possibly be a trojan
> > horse... otherwise appearing to be a gift license but lacking a
> > criticial component needed to be viable glue for the constrution of
> > permissive works.
I'm sorry for using an obtuse (and perhaps paranoid) analogy. I'm
referring to the Gift Horse that the Greeks used to breach the walls
of Troy. Not that open source software has walls, mind you. However,
I do see the CC0 as a being a "gift" packaging with an extra special
patented place for infringment claims to hide.
OSI approved licenses are essentially kinds of acceptable gifts
that developers use in a open source community. The OSI reduces
the transaction cost for the construction of open source works by
permitting a developer to look up a license and decide to "let
the code in" without having to be a software licensing specialist.
The CC0 seems like it'd be an exceptionally permissive offering,
especially if you read the absolutely wonderful preamble. Since
the legal language would be approved by the OSI, developers should
rightly assume that they could incorporate it into their works and
use it without any threat of infringment by the author of the work.
Trusting the OSI's judgement, I'd incorporate a CC0 work into my
software, making dependencies and perhaps basing key design decisions
around the work so that I could save time. I'd then release my work
under a permissive license, and my friends would build on that. They
would probably incorporate those into other works, and so-on.
Then, after a critical adoption mass is there, and night has fallen,
what I had presumed was a Gift becomes a Lawsuit. Exactly the sort
of thing that I trusted the OSI to protect me from.
I hope this helps explain.
More information about the License-review