Copyright for innovation

[David Allouche <david@allouche.net>]

On Thu, Oct 23, 2003 at 02:43:40PM +0100, Rui Miguel Seabra wrote:
> On Thu, 2003-10-23 at 12:37, Jean-Baptiste Soufron wrote:
> > > For people wanting the freedom to create, it is a bad thing.
> > That's for sure but the author's right is not better on that point
> > either since it's also a monopoly system!
> 
> No. That's polluting the truth:
>    People can independently create their own solutions.
> 
> In the case of software this is relatively easy. If one product aims to
> be as quick as possible, you can aim to be as compliant as possible. The
> code you write will certainly follow a different path.
> So anyone has the freedom to create, just not to copy one existing
> solution and claiming it their own.

Certainly, my reflexion on this topic is not as advanced as the one of
the usual posters on this mailing list, but I have a feeling there is a
point worth discussing.

Industrials use information as their asset and want protection. The
position on that mailing list is that copyright is an effective way to
protect those assets and allow they "valorization".

For some reason that I do not yet fully understand, some industrial
actors (e.g. governmental agencies helping the creation of new
businesses or the financial market) consider patents to be more valuable
than copyright. Regardless of the problems of range of claims, this may
be caused by a confusion with the traditional production industry where
copyright does not seem to have any practical value. But there may be
more to it.

We oppose software patents because of their "scoping" (to reuse a
computer language term). It is very easy to unknowingly infringe a
software patent while only a little education is sufficient to avoid
unkowingly infringing copyright.

Let us do not discuss here of the compared economics of the information
industry and the goods industry. Also let us do not argue that any
useful "computer program machine" use hundreds or thousands of
"intellectual components", because there is no obvious contradiction
with "industrial machines".

The point I would like to discuss is the "independent reinvention"
problem.  The patent system assumes an Invention is a relatively rare
occurence.  Rare enough that all applications of the patented process
can be attributed to the Inventor with little risk of harming and
independent inventor. Whenever there _is_ independent invention, the
material costs involved make it practical to prove there _was_
independent invention.  For practical purposes, we could even define an
invention as something which has those properties:
  1. rarity
  2. practical provability of reinvention.

For reasons all readers of this list are familiar with, software does
not have those properties so we call the "progress substance" Innovation
instead of Invention. Since an Innovation does not have the properties
of an Invention, the patent system cannot be fairly applied to
Innovations.

However, we claim that copyright provides enough of a protection. I am
not fluent in the subtleties of copyright law, but clearly copyright
does not just apply to mere "expressions". Translation to other
languages are derived works. In your argument, you point out that
a _different solution_ to the same problem cannot be considered a
copyright violation because the difference in approach will cause a
substantial difference in technical means.

Clearly, that is true in the free software world, where excepted for
"educative developments" (which educate the author...), a software is
only produced when existing solutions are not considered satisfying or
when variety is seeked for its own sake (e.g. security through
diversity).

However I am not convinced this is entirely true for proprietary
software where software can be produced just because there is a growing
market. How many antivirus softwares for Windows are there? Are they all
substantially different from each other? In proprietary software there
can be an inventive for multiple actors to produce _identical solutions_
to the _same problem_.

Imagine MyStartup Inc writes KillerFeature, a nice innovative program
(which contributes to the progress of computing) but lacks the "sales
force" to penetrate the market quickly. Then BigBusiness Corp stumbles
upon the software and shamelessly clone it (without ever reusing a line
of code) thanks to its army of Programmers From MIT then uses its squads
of Marketeers in Black Suits to quickly take over the market.  The
investors of MyStartup would expect some garantee that this will not
happen and currently the only way to this end is patents. Copyright is
not enough. 

Surely, the macro-economic cost of software patents is not justified by
this kind of micro-economic scenario. However whe can expect pressure to
interpret copyright more widely.  I can imagine the copyright law being
extended to consider recreating identical solutions a copyright
violation as far as an identical solution reuses visible design
decision.  For example, "user experience" might become copyrightable, it
is already more or less...

Do we consider such an extension of copyright acceptable and what
alternative can we propose? How can we balance the "value of the design"
and the "value of the implementation"? 


My two euro cents.
-- 
                                                            -- ddaa