Re: [fsl-discuss] Re: [Patents] The politics of software over-protection -- moving forward..

[Simo Sorce <simo.sorce@xsec.it>]

I think you have a reverse thinking machine, really.

On Sat, 2003-04-26 at 21:56, Gregory Aharonian wrote:
> Protecting software ideas with patents enhances software technical
> development the same way protecting electronic ideas with patents
> enhances electronic technical development, the same way protecting
> optical ideas with paents enhances optical technical development,
> the same way protecting biotech ideas with patents enhances biotech
> technical development, etc.  In each case, the expressed logic being
> protected is the patent claims.

It is really easy to speak mr Aharonian, see! an elephant flying there! 

But come down to details and proofs please.
The fields you mentioned are entirely different, and are *NOT* at all
enhanced from patents in the same way, because the patent system is
recognized (by any economist worth of his name) to be a bad system.
It is only less bad than other known alternatives in *some* fields.

So given it is a bad system as it builds up monopolies and for a very
long time (20 years) it *must* be very carefully applied *only* when
it's damages are really much contained.

Now, industrial fields (and keep in mind the term industrial) that need
great investments simply to get to see if an idea is feasible are
generally seen as an acceptable field where to apply for industrial
patents.
Because the resulting process cost really much to be researched and
implemented, and without patents it is claimed it would be really
difficult to have any industry to do the research at all.

In other fields like software, where ideas are very much much cheaper to
make, where progress is made my micro-steps each one on top of many many
others and where what cost (not so much anyway) is implementation and
not research on process, the claim that nobody would research without
the patent system is *clearly* false as demonstrated by facts. So what
is the benefit of a patent system on software? What would a patent
system give back to society that is not yet done without it?
NOTHING

Knowledge is already spread. Society gains only monopolistic positions
in a field that is already at risk due to network effects. Society get a
field where entry barriers are raised a lot, and so competition suffer.
And as anyone know, in a capitalistic system, lack of competition is a
plague; any sector without competition simply sit down.

Such system tend to not produce anything new, as novelty is not
necessary to a monopolist to keep its primate as it have no competitors
that can take it down. The result is that research slow down a lot. This
is exactly the contrary of what a patent system is supposed to do.
And there are economic studies that already found out that sectors with
incremental grow are harmed by patent systems. (sorry I have not a link
handy if anyone has it would be nice if he can provide it).

> Now granted, the quality of issued patents in all fields of technology
> in terms of novelty and obviousness has to be improved, and the quality
> of patent claim drafting has to be improved.

It cannot, simply because software is logic underneath and logic and
ideas has no limits, so the field is so broad that nobody can ever think
of being able to really tell if the claimed patent is really novel or
non-obvious to a person of the sector, simply because there so many
sectors that meet in software that the examiners cannot claim to know
all the fields.

> But in a relative sense,
> software is just the next technology that the logistics of the patent
> system has to learn to deal with.  But not the logic - patents are for
> processes, methods, machines and designs - which is what software is.

Patents are for *industrial* process! Not for logic. Anything can be
divided in steps through the logic, but confusing industrial process
with a logical process is just a mistake and a lack of real thinking on
the whole matter. Algorithms are simply mathematics and they are divided
in logical steps as any mathematical demonstration show! So any
algorithm follow a "process"! Does it mean it should be patentable?
Of course not! Because mathematic has nothing inventive in itself, but
simply deductive steps.

Instead industrial process describe new ways to handle matter, and you
never patent the idea but the specific process you invented. Anyone else
can find out a different process to do the same thing and that would not
be affected by your patent. In software instead you are asking to patent
the idea underneath a program (explained in steps so that it form a
logical process), not the real process.

So if you would like to really make the patent system apply to software
you should make it apply to the real process not the idea as happen in
the industrial field; you claimed that, didn't you?
And what is the real process? The source code of course! Here there are
all the _real_ steps that drive the machine to do what you want. So what
should be patentable is the source code. And what would be the result?
A copyright called patent, with plus the fact that you need to pay fees
to a central office and wait them to give their approval some years
later, meanwhile your program get obsolete and will be on the market
when it is yet old.

> But copyright?  Software copyright was, is and will always be illogical.

You are really a good dispenser of affirmations, perhaps explaining also
motivation would make you more credible. Telling the Earth is at the
center of the universe is a dogma not a proof, the proof is a
mathematical demonstration based on physical laws.

Copyright is the right form of protection for software.
One of the motivation is above. Another one is that software is really
an expression of an author like any other copyrighted work, a book,
music, a film ...
(The fact that the current copyright laws are broken is another thing)

> That it ever ventured into the copyright world was due to a historical
> accident of a Supreme Court basing one of its decision partly on
> marketing considerations of an industry.

So you think a Supreme Court decision is wrong and lower courts
decisions are better?
USA must be a weird state where the more you go up in the hierarchical
scale of power the more people is stupid ... 

> Read the copyright laws

done thanks, and you?

> no copyright for ideas, processes and methods.

Given that source code is at all effects a process, copyright does
protect process!
Ideas are *NEVER* protected by copyright or patent system, or shouldn't
(read your laws Mr. Aharonian), the fact that the patent system has been
twisted and tweaked so much to effectively "protect" ideas simply
because they have been explained in steps, is a distortion not what was
meant in law.

Last but not least, if you read the main body of law concerning patents
as found in Title 35 of the United States Code you'll find also the
words "subject matter" in a lot of places, it doesn't seem OK to me to
consider an immaterial thing like software itself as subject matter.


>   Especially source codes
> - you would be a fool to rely on copyright to protect your source codes.

So fool that free software rely on it, and specially for the GNU GPL. So
fool that nobody really infringe it, and when happen nobody want to go
to courts and step back.

Please back your _opinions_ with some facts. Analogies have their limits
and cannot be pursued as the motivation for patents in software as the
software field is so much different than classical industrial fields
that deal with matter and not with logic, that most of the analogies you
presented here simply fail miserably under a closer scrutiny.

Simo.

-- 
Simo Sorce - simo.sorce@xsec.it
Xsec s.r.l. - http://www.xsec.it
via Durando 10 Ed. G - 20158 - Milano
mobile: +39 329 328 7702
tel. +39 02 2399 7130 - fax: +39 02 700 442 399

Attachment: signature.asc
Description: This is a digitally signed message part