Subject: Re: Patent-based dual-licensing open source business model
From: Scott Capdevielle <scott@syndicom.com>
Date: Tue, 12 Sep 2006 10:29:48 -0600
Tue, 12 Sep 2006 10:29:48 -0600
Lawrence,
what will the license look like? I have been researching the same 
problem and coming up with a similar solution. One thing we believe is 
that this type of license needs to have two components to have the 
desired effect. I think these address Thomas' concerns but would also 
like feedback:

1. Non discrminatory license. Anyone can obtain the license regardless 
of who they are and no exclusiviity is granted. This is important to 
truly open it up to competition. 
2. Commodity pricing known up front. The royalty must be small enough to 
look like any other cost of 'manufacturing'.  In other words, if , in 
your example it is bundled with a piece of hardware and the total 'cost" 
of this solution is $100 then the license fee should be some small 
percentage of that amount.  It is also important that the amount be 
known up front so that entrepreneurs can do business modeling and factor 
that cost into the equation. Making it a known quantity and non 
discriminatory enables a wide audience to calulate the cost to produce 
their commercial product in advance of making any investment. 

scott


> Lawrence Rosen wrote:
>
>> I'm interested in the reactions of this group to a new patent-based
>> dual-licensing open source business model adopted by International
>> Characters.
>>
>> www.rosenlaw.com/IC-Business-Model.pdf.   
>
>
> Thank you for working on such a project.   I agree with the
> broadest goals (securing royalties while contributing to an
> open source commons).
>
> Let's stipulate that the patent applied for here (a) will
> be endorsed by community/peer review (e.g., is genuinely
> non-obvious and original);  (b) will be granted and enforceable;
> (c) will attract paying licensees.
>
> First, then, the conjunction of stipulated conditions is
> unusual and will not apply to most projects.    This isn't
> a model that many others will be able to emulate.  Still,
> if we also stipulate that (d), copyright alone is insufficient to secure
> royalties to this inventor because implementations may be
> cheaply derived from published descriptions of an algorithm,
> then this form of protection is in this case necessary (if
> we agree on the broad goals).
>
> Second, best of luck in your discussions with Mr. Stallman.
> A serious problem, for GPLv3, would be that if it were
> made compatible with your license then the rights to use
> and distribute some GPL programs (those using the
> International Characters IP) would be curtailed.   For
> example, suppose that I want to manufacture TiGnu --
> a DVR with Tivo-like features running GPL'ed code in
> user-reprogrammable firmware.   Your software patent would
> prevent me from selling TiGnu machines unless I obtained
> a license first.   Your software patent would prevent users
> from using these machines in certain ways.  (A secondary
> problem is that your license helps to support the existence
> of software patents -- to accommodate you with GPL
> compatibility, RMS would have to bet that the end result
> would be that licenses such as yours are worthless because,
> in the end, the patents are worthless.   The two of you are
> in an adversarial relationship.)
>
> Third, being incompatible with GPLv3 is, as you must
> know, an uphill battle.   In many cases it means doing
> "open source" with one-hand tied behind your back (you
> may not use GPL-ed code).  Those of us who believe in
> fixing, rather than discarding, IP incentives must look
> forward to that battle but I think we make a mistake if we
> each take our individual shots in an uncoordinated way.
> At some point, we need a larger conversation about new
> licensing regimes in hopes of developing a broad consensus.
> Such a consensus would have to look beyond (but not overlook)
> patent considerations.
>
> Fourth and finally, two alternative approaches come to mind.
> One is simply to ransom the release of the patent.  The other
> is to take a two pronged approach:  (a) developing the business
> of selling licenses to establish the value of the patent as an
> emerging standard;  (b) deliberately short-terming the patent,
> establishing a near future date certain for its release to the
> public domain.    The point of (a) and (b) is to follow in the
> footsteps of RSA and expand the set of paid licensees to include
> firms aiming for open source but who want to get a head start
> on developing with the invention.    The second option, *if*
> it applies to the particular invention, is arguably superior to
> ransom in that the seller isn't setting his own upper bound on
> the amount of royalties to be collected.   Alas, both options
> are vulnerable to competition by open-source replacement.
> Compression algorithms, it turns out, aren't *so* hard to
> invent, even if individual algorithms are original and non-obvious.
> And so, confronted with too high a price to pay (in ransom or in
> waiting for the release to public domain) the open source
> community will find a replacement.    Cryptography algorithms
> are of the very rare variety that the cost of developing a replacement
> is apparently much higher than waiting out the patent.  If you're
> confident that this invention is more like crypto and less like
> compression then these alternative approaches are available,
> otherwise, probably not.
>
> Regards,
> -t
>
>
>
>
>> /Larry
>>
>> Lawrence Rosen
>> Rosenlaw & Einschlag, a technology law firm (www.rosenlaw.com)
>> Stanford University, Lecturer in Law
>> 3001 King Ranch Road, Ukiah, CA 95482
>> 707-485-1242  *  fax: 707-485-1243
>>
>>
>>   
>
>
>
>



["text/x-vcard" not shown]