[License-review] Submitting CC0 for OSI approval

[Henrik Ingo]

On Mon, Feb 20, 2012 at 6:15 PM, Lawrence Rosen <lrosen at rosenlaw.com> wrote:
> Henrik Ingo wrote:
>> The MPEG patent pool perhaps was a very easy example
>> to discuss as their actively practiced business model is precisely
>> what we want to avoid with open source.
>
> As I understand the situation, most manufacturers of commercial products
> (e.g., Android phone manufacturers) already license the MPEG patent pool.
> Why should distributors of software intended for those products need to
> worry about those MPEG patents?

Hi Larry

I don't understand this question. Distributors of Android phones seem
to always include proprietary (in the meaning of closed source)
software in their phones anyway. They are of course free to include
patented technology too if they have appropriately licensed such
patents. And they are of course also free to use those technologies in
the open source parts of their Android products, except when the
applicable license makes that problematic (GPL).

The question here is simply, what does it mean to a recipient that
some software license is an open source license? Most people expect
that if you were to give me a piece of software that is open source,
then it means I don't have to worry and can freely use, modify,
redistribute, etc... that software. However, if one takes the opinion
that OSI should only restrict itself to the domain of copyright, then
open source licenses would essentially be meaningless. From the open
source label it would be impossible to know if I can use the software
safely at all.

It is true that patent law is such that I cannot know that anyway,
since any patent held by anyone else is still a threat. But it is in
my opinion a reasonable requirement that the person/entity who gave
(licensed) me the software in the first place will not sue me the next
day for lack of a patent license (or any other reason).


(And to respond to another writer of this thread that I'm too lazy to quote...)
Yes, I can also game the patent system. I could create two independent
companies, one of which distributes patent encumbered software to
people and another one that goes after the same people and sues them.
But this is a flaw in patent law (not just software patents). It
doesn't contradict what I write about reasonable assumptions above.

> Speaking more generally, there are at least three common situations:
>
> 1. A patent is owned or licensable by the distributor of the software. Most
> mature FOSS licenses explicitly make those patent claims available along
> with the software. In others (BSD, MIT, CC0?) licenses to patent claims are
> implied (and there is possibly an unfortunate assumption by some here that
> implied patent licenses are enforceable in all jurisdictions).

And in this case it is relevant to ask whether the distributor has
licensed also the patent or not, since he has the power to do so.


> 2. As with the MPEG patent pool, necessary patent claims may be available
> from third parties so that the software distributor need not worry about
> them. This is also true for many patent claims that read on industry
> standards. FOSS licenses can do nothing about these patent claims, nor do
> they need to.
>

"need not worry" ... the distributor would probably want to cover his
own ass since he is distributing/selling the infringing software.

While it is true that this well known model is not incompatible with
most open source licenses (are there others than GPLv3?) it is imo
also not compatible with an inherent ideal in open source: lack of
centralized, proprietary ownership. Many of us believe that the power
in the open source development model comes from the extinguishing of
such bureacratic structures and simply empowering each individual
developer (or his employer) who can develop new software in a
peer-to-peer fashion.

Hence, while this model is not to me what some would say "evil",
rather the intent is actually well-meaning, it is distinctly one of
another era that predated open source. Some friction will be
inevitable.

> 3. In perhaps many cases, patent claims are owned by third parties and all
> we can do is wish for them or ignore them.

True. (Although I have been hoping for someone to invent "copyleft for
patents". I remember EFF was supposed to announce something, but
haven't heard of them since. Again, this is purely something that
could be invented for patents, not just software patents.)

The good news is these questions are still relevant in the first point
you listed.

henrik

-- 
henrik.ingo at avoinelama.fi
+358-40-8211286 skype: henrik.ingo irc: hingo
www.openlife.cc

My LinkedIn profile: http://www.linkedin.com/profile/view?id=9522559