[Patents] translation - complete version

[John Tobey spam@john-edwin-tobey.org]

This is an English translation of Hartmut Pilch's recent post titled
"brevets et logiciel libres (version elaboree)".


Sun, 5 Sep 1999 16:08:40 +0200 (CEST)
PILCH Hartmut phm@a2e.de
[Patents] brevets et logiciel libres (version elaboree)

patents and free software (extended version)

office@breese.fr (Virginie BARBET ) wrote:

> I do not agree with the arguments set forth.  The anti-Americanism
> makes me laugh: the free model seems to me to have brought profit to
> the American company Red Hat, which has gone public thanks to the
> naïveté of developers who let themselves be seduced by the idea of
> freedom.

Red Hat's success in the market shows that one can run a large company
completely without intellectual property.  Do you think that Intel is
naive?  They invested in Red Hat!

Besides, the value of free software to society is independent of the
success of companies that represent it in the market.  These companies
are not the principal force behind free software, but they profit from
it.

> I also noticed that most of the signers came from public research
> institutions, funded by our taxes, and therefore lacking the concern
> for profitability of a commercial enterprise.

I see that many companies, not just Red Hat, but also proprietary
software companies, profit from the existence of free software, and,
more generally, from the existence of a public and open information
infrastructure, and, of course, from an infrastructure of culture,
civilisation, education, etc.  It is false to say that these
infrastructures are funded "by our taxes".  One could as rightly say
that private profit is made possible "by our public sector".

> It is easy to give to the public domain that which is paid for by
> the citizens, and which they would normally own (cf. the
> rulings [?(dispositions)] concerning inventions and software created
> by public employees, even outside of their working hours).  At best,
> the free model looks very much like a booby trap, in which the
> consequence, if one were to follow its proposals, would be a greater
> difficulty for creative European companies to find their place in
> the global market.

Even if your argument may be valid, it is too simplistic.  Business is
not everything.  The prosperity of the information sector does not
rely solely on software companies.

But your argument is not in line with the reality of most European
businesses.  These are small companies with 5-10 programmers who make
a good living from copyright and don't want to get themselves sued by
big corporations capable of maintaining legal teams and patent
departments.

> Let's not get hung up about this, let's trust in European creativity
> and learn how to encourage competitive activity without dogma...

Good suggestion, but the dogmatism seems to be mostly on your side,
see below.


> From: jv@breese.fr (Julien VILLA ) <comments>

> Most of the discussion on this list shows a great misunderstanding
> of the patent system.  A patent is only valid if the invention is
> new and inventive.

... and, don't forget, has an industrial application.

Yes, there are some comments on our letter that ignore these criteria.
But really, the enforcement of these criteria by the patent
authorities is very unsatisfactory and unmeasurable.  They have not
found a way to prevent the majority of software patents from being
patents on trivialities, obvious to the programmer's way of thinking,
but not to the more stringent requirements of the legal system.

> Algorithms, software, "ideas" that already exist or are obvious are
> therefore not patentable, and patents granted on them would be void.
> Furthermore, a patent does not prevent scientific exchanges, on the
> contrary.  It simply allows the inventor to control the profits that
> result from its use.

Control = forbid others to use it.

Programming is a cultural practice, which can very well exist without
any form of organisation as an industrial practice.

Industrial organisation often produces monopolies and worse results.
Unfortunately, the current form of patent rights supports some
behaviors that aggravate these problems, such as exclusive licensing.
And a multitude of international treaties, which sovereign nations no
longer have the right to change, prohibit us from improving the patent
system and adapting it to the special needs of software.

> Lastly, be aware that the principle of the patent as we know it came
> out of the French Revolution, and, contrary to what one reads
> sometimes, software is patentable in Europe.  The EPO counts 13,000
> patents granted for software.

Yes, but fortunately not yet as such.  One cannot yet, therefore, sue
a publisher for direct infringement.  And many of the software patents
granted now may be challenged under the current legal system.  This
system upholds our position, but practice, under the influence of the
proprietary lobby, has strayed from the law.

> From: yas@breese.fr (SAUVAGE YVES-ALAIN )

> The impassioned remarks that I read on this list are surreal!
> Software IS PATENTABLE, and IS effectively PROTECTED by COPYRIGHT.
> The fact that some creators choose to renounce the use of their
> rights is up to them.

Free software authors do not renounce it.  But they renounce trade
secrecy.

> But to try to impose this choice smacks of totalitarianism.

Free software authors do not want to impose this choice on anyone.

But they do not accept being barred from programming.  That is what
the patent system does.  Several free software programmers have been
sued in the United States on the basis of a patent that they didn't
know about and from which they did not receive their inspiration.  In
the area of cultural practice, it is normal for many people to have
the same ideas independently.

I see in the comments of intellectual property lawyers a kind of
totalitarianism.  They seem to think that everything of value in the
world comes from the private sector and that we should privatise all
aspects of life, even those where the concept of property obviously
can not be easily applied.  In order to apply it anyway, they create a
bureaucracy of prohibitions.  And if this system becomes more and more
costly and impractical, too bad.

What is this, if not proprietary totalitarianism?

Personally, I see in this system all the symptoms of the Soviet
disease:

- dogmatism
- prohibitions
- rigid institutions, removed from democratic control (in this case by
  international treaties)
- expansionism (created by the dynamic of the legal-industrial
  complex)
- wooden language (always the same formulaic statements without
  individual thought)
- self-perpetuation far removed from economic and technological
  reality

> (this assumes that these authors who pride themselves on giving up
> their rights really do own these rights: often, they give up rights
> that in fact belong to their employer or to the state.  It is easy
> to let everyone use something that belongs to someone else...)

That is true, but it is not relevant to our discussion.

For universities, it is normal to publish the results of research so
that everyone may profit from them.

Even if commercialisation is possible, it is often a bad choice,
because

- the commercialised product is secret and therefore removed from the
  cycle of research and improvement by colleagues in the field
- commercialisation is complicated and expensive.  It creates
  administrative costs -- especially when there is a patent system in
  place.

> If you wish to let just anyone come in though your window to live in
> your apartment, you are allowed to do so, and it is generous.  But
> from there to claim that property rights don't exist, there is a gap
> which intellectual honesty forbids one to jump.

I don't know against whom you make this argument.  People have often
tried to incriminate Richard Stallman for usurping the fruits of
others (such as AT&T) in creating "only" clones.  These people
forget/ignore that

- cloning is done for interoperability more than for lack of
  inventiveness.  The "market" of proprietary software is full of
  barriers to entry.  You have to be "compatible" if you want to gain
  access.  All new developments start out as cloning.
- in many areas of programming, the work is in the details, not in
  some "basic research" separate from the "product" itself.
- there is a lot of inventiveness in the GNU system.  Also a lot of
  new specifications formed in the process
- It is thanks to GNU that Unix continues to be an important force in
  the market and commercial Unix companies can make money.

> These are exactly the same arguments that underdeveloped countries
> use against the general principle of patents.

Please cite an example!
Anyway, we are not against patents in general.

> Are we so unsure of our innovative capacity in Europe?  Would Europe
> become the third world of software creation?

No, Europe has many small small companies that are very advanced in
their fields.  And these companies feel threatened by an
American-style patent system.  cf http://www.freepatents.org/pr1.html

> Isn't it preferable to enter the game of competition on equal terms?
> It has been proven that innovation comes much more from new
> companies than from well-established companies.

Perhaps.  But innovation is not a scarce resource in the field of
software.  It is in most cases an everyday activity.  In these cases,
patents serve only to hinder competition.

To try to hinder competition is a normal impulse of every business,
but the public must grant privileges only if it receives something
important in return.  Everyday innovation does not merit such
privilege.  Often, it would be better to punish it, because it serves
to subvert standards.  Look at Microsoft's arguments in court for the
"right to innovate".

> New companies must still be able to combat the economic power of
> more established, often less creative, companies.  Patents and IP
> rights are the only tools that permit them to fight the imperialism
> of corporations that dominate the market with mediocre products.

That may be possible in the case where the small company concentrates
all its effort on one single technology and has no need to program in
fields outside of that one.  It is a rare case.  In most cases, a
program is a complex system that touches on many patents at once.  So,
most small European businesses such as exist would instead be
pressured to submit to a large corporation with a large patent
portfolio.

> The free model is nothing but a temporary illusion, whose
> limitations the RedHat experience well illustrates.

What do you mean?
The Free Software Foundation was founded in the early 80s, and before
then open source software played an important role.  With regard to
Redhat, see above.

> Is it not significant that the Free Software Foundation - a private
> American company - presents itself as the objective owner of the
> rights to so-called free software?

No, it does not hold those rights but it is a collective agency that
defends those rights.

There are many agencies of that sort.  We might also need such a
collective agency for the interests of groups threatened by the patent
system.

I hope that we can arrange a conference to study the possibility of
limiting the negative impact of the patent system on software
development.

I also hope that my explanations incite you to explore with a little
less prejudice the economic reality of free software and of software
in general.

Cordially,

Hartmut Pilch <phm@(a2e.de|ffii.org|suse.de)>
PGP key at pgp-public-keys@keys.net.pgp

-
translation from French by John Tobey.