[cc-community] The Question of Manufacturing Copyleft

Terry Hancock hancock at anansispaceworks.com
Fri Nov 2 08:37:37 EST 2007


Greg London wrote:
>>Greg London wrote:
>>How about this ... the license could simply make this requirement --
>>take boilerplate from one of the "OSD approved"/"non-free talk-back"
>>licenses.
>>
>>THEN, it could EXEMPT anyone from that requirement who meets the other
>>parts of the copyleft requirements.
> 
> I'm not sure what the "other parts" of the requirements are.

Bad word choice, sorry. The "other parts" are an alternative for the
licensee, not additional requirements -- IOW, he can escape the forced
publication clause if he just treats the hardware as an analog to
binary distribution under the GPL.

It will take some clever writing to make this both clear and enforceable
of course, but it should be possible.

> What you want is a license that prevents Alice from keeping
> a derivative private and does not hinge on distribution of a
> derivative. As long as the license requires all derivatives
> to be made public, then Alice can't keep her changes private
> and hide them and distribute them in the form of an asic.

Right. But we can be a little more generous than that and give Alice a
set of conditions under which she *doesn't* have to publish the
derivatives (for example, not until she makes and distributes parts from
the design).

The result leaves my "hard copyleft" intact, but (potentially) avoids
the level of restriction that most free software advocates find
offensive about talk-back requirements in OS licenses.

For example, the requirements can be written to suit the "desert island"
and "dissident" tests that Debian legal likes: respectively meaning that
a person who is not able to publish their changes (e.g. no web access)
doesn't have to and that only people who receive the product have to
receive the changes (notionally for political security reasons). Both
are satisfied if the source is included with the product.

Which suits me fine. I don't think developers will have a problem with that.

I'm not sure it actually matters if it is written in a negative way "you
must publish source UNLESS you do X or are otherwise EXEMPT according to
section Y". Which sounds complicated when written like that.

But it might be the clearest way if most users are just going to publish
the source online and not worry about the exemptions. I think this is
quite likely in this case -- particularly for parts that are inexpensive
in mass production, because they're not going to want the hassle of
including media disks with each little $1 part! And frankly, I think
it's better for users, too. I'd rather be able to download the specs
than to have a huge collection of disks with tiny amounts of data on
them stacked in the back of my workshop.

> Enforcement is always optional. And more than likely,
> you won't care about all the intermediate derivatives they
> made to a core, you just want the version that was used
> in their latest asic. And if they release that version under
> your hardware copyleft license*, then no need to go after them
> for not making all the intermediate derivatives public too.

Right, that's the sort of thing I'm talking about exempting.

I don't want to do this through selective enforcement, though -- I want
the license to be quite clear about what IS permitted.

> (*) Hardware Copyleft License. Is that it's name?

Oh, it's way early for naming.

At this point, I'm still trying to hash out what's possible. Then I'll
write a proposal, then that will get shredded by my colleagues, etc.,
etc. Then we'll probably publish it for the public to shred. Then tape
the bits back together and have a lawyer review it. Possibly, we have to
repeat this cycle a few times. You gotta love that community process! ;-D

But complicated or not, I think this needs to be done. Right now there's
an accumulation of work under soft copyleft licenses like the GPL that
don't provide the protection that most developers think they are
getting. At some point, a manufacturer is going to get a "brilliant"
revelation about this, and then it'll be like the whole "Tivo-ization"
debate, all over again. People will get snippy. There will be lawsuits.
Open hardware will be declared a failure. All die. Oh the embarrassment.

Cheers,
Terry


-- 
Terry Hancock (hancock at AnansiSpaceworks.com)
Anansi Spaceworks http://www.AnansiSpaceworks.com



More information about the cc-community mailing list