Re: licensing and CLA

On Thu, 14 Aug 2003 21:08:58 -0700
Zach Welch <[email protected]> wrote:

> Hi all,
> 
> Here are some brief and jumbled thoughts I wanted to share, after the
> initial (and mostly negative) responses to the idea of dual-licensing
> and CLAs.  If the rest are silent supporters, that's great, but I'd
> love to actually hear you help me make the case - or simply voice your
> support.
> 
> For those of you waiting for development to happen, that's at the top
> of my list and we will have something soon.  Graham and I have been
> making progress on gathering the full requirements for Repoteus, but
> this message popped out in between my hacking and IRC chats.
> 
> * Ideally, I wish all the software could be free software.  Even if
> such becomes the reality, the cost of developing that software will
> never be free, and development organizations will always need to
> derive revenue somehow.

Definitely, otherwise I won't get paid for doing what I enjoy and would
have to find a real job instead. :-)

> * A non-profit can be run like a business, just the revenues go
> exclusively to further the cause, not help line shareholders pockets.
> This foundation will be driven like a business; its business plan will
> be continually update with all of the traditional aspects used to
> drive an organization forward.

I agree with you. Also, like any other organisation it should try to
always have a cash float available.

> * Any relicensing would take place in the future, could be required to
> be voted upon by the community of members.  I do not want the
> foundation to be able to arbitrarily relicense the technology I help
> develop for it any more than I expect others would want that to
> happen.
> 
> * I also own a embedded Linux consultancy.  I want to be able to say
> that non-GPL is an option - and that is an option for those works that
> I create by myself from scratch and without outside contributions.

No problem as far as I am concerned.

> * For Zynot technology, I would happily arrange to pay a per-unit
> royalty (or just regular wads of cash) if a client insisted for
> non-GPL terms.  This is not an unreasonable scenario for an embedded
> OS vendor, though one I would fight against (and would turn down if
> enough pure open source contracts were available).
> 
> * Such terms would be approved by the membership when that day comes.
> Such would be above and beyond licensing of our yet to be developed
> brand name and image.

I think this is the big thing.

ISTR people talking about not including the "or later versions of the
GPL" bit of the GPL license, however it was worded. I raise this because
it is related. *ANY* change to the licencing, whether providing a
non-GPL licenced version or changing from GPL version 2 to a later
version of the GPL should require the approval of the membership,
although the board should fell free to *recommend* such changes.

> * While the reality of business is that other licenses are viable
> (even those labeled as "closed"), the foundation will only fund the
> development of open source software.  Non-GPL is only a licensing
> option that gives us revenue, support would obviously be only under
> the GPL.

Any non-GPL license should be worded so as to give Zynot as much freedom
as possible. Maybe a clause saying something along the lines of,
"purchasing this license does not grant the right to exclude others from
developing similar changes and enhancements to the software to any
changes that you make." Just to make it clear that if close sourced
company X develops a closed source enhancement they can't sue us if we
develop something similar. I am not a lawyer, so consider the idea
rather than the specific wording.

> * The ability to relicense under non-GPL terms is only one reason for
> using a CLA; it also provides a means for accountability with regard
> to IP disputes, either with closed use of GPL or with contributions
> with submarine patents.
> 
> * I do not believe that having a free-for-all copyright scheme (which
> is what you have without CLAs) is either prudent or acceptable for a
> business to operate.  Such scattered control prevents the code from
> ever being re-used in non-GPL contexts.

I agree with you in principal. I've not read the CLA yet so I can't
comment on the specifics.

> * The recent SCO lawsuit serves as a sign that we ally ourselves
> though such community organizations such as Zynot, providing united
> accountability of our actions.  The CLA serves as this foundation for
> better protecting the community.

Agreed.

> * There has been an annoying trend of certain embedded device vendors
> not immediately providing source code for products that obviously
> contain GPL source code.  While many cases get resolved amicably, the
> foundation should be empowered to protect that works derived from
> Zynot remain open.

Agreed.

> * Hypothetically speaking: If re-licensing were taken out of
> consideration (i.e. Zynot is GPL-only), would I still want CLAs?
> Absolutely, as the language goes beyond the realm of licensing;
> however, the title becomes conspicuously inappropriate.

I agree that a CLA is good even for a pure open source product. It
clarifies the relationship and means you have bits of paper for lawyers
to play with if anyone outside the organisation gets awkward. Without
that, someone outside could try claiming that Zynaut did not own the
code it was trying to protect and that it was therefor up to the
individual authors. I'm not saying any such claim would be right, but it
does help preempt any such attempt.

> * I've heard rumors that rewording of the CLA might be happening soon,
> from which Zynot could benefit.  I am not in a hurry to see these
> signed, as the terms do specify all past contributions.
> 
> * I am trying to convey what I perceive as a real need for the
> organization, bearing in mind that I personally feel this is the right
> thing to do for open source in the long term.
> 
> I hope these spark more pointed lines of conversation that can help us
> move these topics along to some sort of conclusion.

Well, I found time to respond this time round anyway. :-)
-- 
Mark Gordon
Paid to be a geek and software developer.