My alternative busines model

[Niall Douglas]

On 3 Dec 2002 at 15:11, Marcus Brinkmann wrote:

> If I got it right, your idea is basically this:  Everything's fair
> game as long as you don't charge.  If you charge, you have to pay
> royalties to all the copyright holders.

That's right!

> I want to give you an equally abstract reason why what you try to
> achieve will never result in a free software license.  It might not be
> your intent to provide a free software license, but I hope that the
> reason is good enough for you to see why we don't think it is an
> alternative to free software either.

No it isn't an alternative for free software - it's an alternative 
for proprietary. I would fully expect and indeed encourage free 
software to continue doing what it's doing.

> The reason it is not a free software license is that it forbids
> commercial redistribution without negotiating and getting back to the
> copyright holder. In other words, if I want to sell a copy of the
> software (with modifications or without), I can not do this unless I
> have negotiated royalties with the copyright holder and paid those
> royalties.

Only if the copyright holder has set a price. If they said it's zero 
cost, it becomes like BSD. If they set a price of 100 billion Euro, 
it becomes very like GPL. You're right in that the code it itself 
reuses may come under a different price, but my intention was that if 
this were introduced legally it wouldn't make free software in any 
form illegal. It would however make current proprietary software 
illegal.

> Where is the practical problem?  First, it means that if I can not
> reach the copyright holder (ie, he ignores me, or he is deceased, or
> just became a juggler in a chinese circus, or whatever), I am not
> allowed to copy for charge at all, until the copyright expires.

No, I explicitly give a month for the holder to reply. If they don't, 
you can carry on regardless though the copyright holder has the right 
to retroactively apply for their royalties. Here's where it gets 
complex, because at this point the law really needs to come in with 
adjudication services.

> Second, it means that if I want to combine several programs, I have to
> contact and negotiate royalties with every contributor.
> 
> Now, imagine something like Debian.  Debian does not sell the
> software, but CD vendors do.  A CD vendor which would try to sell a CD
> collection with 4000 software programs under your license would have
> to try to negotiate royalties with 4000 copyright holders.  This would
> basically mean that distributions of software are economically
> impossible, as nobody would come up for the costs.

That's a good point, but where in the proprietary world do you hear 
of 4000 pieces of software bundled on a CD? The answer is outside 
shareware, you don't - it's a unique ability afforded by the 
economics of free software.

> Now, to the other end of the spectrum, the individual programmer.  A
> programmer who would want to write a program and reuse, let's say,
> code from 6 other programs.  He would need to negotiate with 6
> copyright holders.  I can imagine that the royalties would make the
> whole project financially unviable, because the profit you could
> potentially make does not cover the royalties.  This assumes that all
> copyright holders would be cooperating. There is always the chance
> that a copyright holder just boycotts certain projects.

There's nothing stopping reimplementation. In fact, I'd suspect it 
*encourages* reimplementation for cost reasons with the view that 
innovation will result. Where it's cut & dried that a library is the 
best tool for the job, it'll get reused verbatim but where it isn't, 
I expect improvements to be made which then could be reused by others 
ie; it's a perpetual cycle of improvement.

> No, sorry.  Your idea doesn't fly.  In fact, I am positiviely sure
> that if we were forced to use it, it would destroy the free software
> environment. The only options that are feasible with this kind of
> license is always gratis reproduction ("no commercial use") or
> everybody writes their own stuff without copying from anybody else,
> and takes high royalties ("proprietary").  Ie, it's the worst from
> both worlds.

No, you've not read the formulae closely enough. It's like writing 
C++ in that it's not a good idea to inherits from too many things - 
same will go for this system. The likelihood is even on a large 
project, you'd pay royalties to a maximum of two or three people. 
Remember - it's the *code* which is protected not the algorithm.

Also, no one can prevent reuse because the royalties are set to a 
maximum of 90% of the sale price. While someone couldn't make any 
money reselling the same product verbatim, they could with 
significant extra features say if the original provider lost interest 
in development.

> Frankly, I can not imagine how any extra innovation would come from
> your model. We already have lots of experience with gratis software
> and with proprietary vendors (after all, it's very common for
> proprietary software vendoprs to license products from others for
> inclusion in their code).  It's certainly not something the free
> software movement needs.

No it's not what free software needs but then I do think free 
software is 98% best for the industry. It is, however, very much what 
proprietary needs whilst keeping compatible with free software. I 
also think it provides the entrepreneurial force missing with free 
software.

Cheers,
Niall