Preliminary answers from SSLUG

[Erik Josefsson <sslug@sslug>]

Preliminary answers from SKÅNE SJÆLLAND LINUX USER GROUP (www.sslug.dk)
to rapporteur Arlene McCarthys questions in WORKING DOCUMENT 11 June
2002:
http://www.europarl.eu.int/meetdocs/committees/juri/20020619/471443EN.pdf

Submitted by Erik Josefsson.


Q1.
One option that I don't think has been considered is to ask the EPO
Enlarged Board of Appeal to solve the controversy:

Are "technical realizations of non-inventions" patentable?

This is remarkable, while there are many well documented cases of
completely contradictory interpretations of EPC 52, and the TBAs are
obliged by EPC 112 to forward important questions of interpretation to
the EBA:

http://www.sslug.dk/emailarkiv/patentdirektiv/2002_05/msg00038.html
 
We have a well documented case where Swedish interpretation of patent
law and EPO interpretation of EPC is completely contradictory.

This apparatus is a technical realization of the logical problem of how
to organize a queue
http://l2.espacenet.com/espacenet/drawing.pdf?PN=EP0086199&PNP=US4675647&CURDRAW=1&sslug@sslug
that was denied patent protection by all Swedish patent courts but
granted a patent by the EPO.

The Commission says that the proposed directive is needed because
national and EPO court verdicts are divergent. However, the reason for
this is not differences in law, as all EU member countries have
identical laws with respect to the patentability of computer programs.
Thus, further harmonisation of national patent laws on this issue is
actually not possible.


Q2.
I would say the objective of the directive is not completely clear. It
says it wants to exclude patents on business methods, but gives no legal
structure to achieve this.

The provision in article 3 that "computer-implemented inventions" belong
to a field of technology makes a computer implemented business method
patentable, since the performance of it involves the use of a computer
and the features of it are realised by means of a computer program,
which are the definitions of a "computer-implemented invention" by
article 2.a.

To see if the directive exclusion of business patents works, some
examples of EPO business methods should be analyzed to see if the
directive would exclude them or not. There are a lot of EPO-patents on
business methods, so the analysis should be rather easy to do.

Another objective of the directive is to stimulate innovation, but as
far as I know, there is consensus among economists that they know very
little about the effects of software patents, and if they know anything,
software patents do not contribute to knowledge diffussion (which is a
prerequisite of innovation).

Also there are indications of that the purpose of the directive has not
at all been understood. For example, the Danish government seem to think
that the directive proposal will not broaden the scope of patentability,
but rather narrow it. It is very hard to have a debate when the central
terms have not been properly defined.

And as mentioned above, the objective of harmonizing the national patent
laws cannot be met, since the national patent laws cannot be more
harmonized on this issue than they already are.


Q3.
No answer yet.


Q4.
Yes and no. You will certainly have no legal means to exclude patents on
business methods. EPO has already granted business method patents in
spite of EPC 52.2 (see Q2).


Q5.
At the IFRI conference in Paris 10-11 june, exactly this question was
raised. Some economists answered that the USA more or less stumbled into
business method patentability, and that EU should not follow a
developement that was not really planned. Also, the following expansion
of the US patent system was critizised on many levels of argumentation,
see Brian Kahin:
http://www.firstmonday.org/issues/issue6_1/kahin/index.html

There seems to be a growing concern within the US that the patent
practice has gone too far, and that something needs to be done to
rectify the situation in the US. We should learn from the US experience,
and avoid making the same mistakes.

The EU is potentially a larger software market than the USA. Perhaps EU
should take another path than the US and limit the patent system to
areas where we know more about the effects. Everybody wants to limit the
patent system in one way or the other. A reimplementation of EPC 52,
perhaps with additions as suggested by FFII, should be considered as an
option.

Another problem addressed at the IFRI conference was the lack of
classification of business method patents (a resesarch programme was
proposed), which means that it would be a really bad idea to exclude
something that we don't really know what it is. As long as business
method patents are not properely classified, we can not exclude them. 


Q6.
Regulations on regulations in not a good idea, but one measure to be
able to study abuse of the patent system is compulsory registration of
infringement claims, either at a national patent office or at an
european patent survey institution.

Perhaps that can stop fishy patent-holdup busineses.

Economists regret that they are unable to study the informal economy of
licencing, patent hold-ups and other threats. Very few patent disputes
ends up in court where we can see them.


Q7.
Many inventions are brilliantly trivial!

But the situation of an inventor is somewhat different from the daily
work of a programmer. On a daily basis a fairy skilled programmer comes
up with problem solutions that are one by one EPO-patentable, but the
result of the whole complex structure of the complete computer program
is generally not easily EPO-patentable. Many programmers think that EPO
software patents are just obvious normal problem solutions, which they
are "by default".

Trivial software patentes can be avoided only by not granting patents on
programs for computers as such.


Q8.
One reason is that open source projects in Sweden and Denmark so far are
protected by copyright and that computer programs as such are not
regarded as inventions. The patent policy change is quite new and
basically initiated by the EPO. I think that Knut Blind from the
Fraunhofer Institute found that most German software developers still
think their intellectual property is protected by copyright (something
they learned from the harmonisation of copyright law in 1990) and that
it is only in the US that you can get a patent or be threatened by a
patent. The same could well be true for Sweden and Denmark.

But there are an increasing number of OSS projects being threatend and
sometimes shut down by patent holders. The most famous Swedish case is
BladeEnc: http://bladeenc.mp3.no/skeleton/oldnews.html (1998-09-20 The
Fraunhofer situation).

Unfortunatelly I do not have statistics at hand, but I think there has
been a counting of patent threats sent to the public software
developement site http://sourceforge.net/

IBM lawyer Fritz Teufel has also in public (LinuxTag 2001 Stuttgart)
declared the IBM would not hesitate to litigate Open Source projects.

One reason for the current seemingly peaceful co-existence is that that
companies know that threathening Open Source projects with infringement
claims gives very bad publicity.

The threat is there for all software developers, but only Open Source
developement has all programming secrets disclosed and can not hide.


Q9.
The question could also be asked, does copyright promote innovation? As
far as I know the software industry has not stagnated from copyright
protection. Microsoft, SAP and other big players all established their
market positions without patents. In the case of a single product,
BladeEnc is a perfect example of copyright protected innovation that
created real competition in the market place. Monopolization of basic
logical or mathematical problem solutions will stifle innovation, for
the reason that such problem solutions are by definition impossible to
"invent around". The same applies to monopolization of algorithms, or in
the BladeEnc case, file conversion patents.

The question also involves the basic premise for the patent system:
innovation through knowledge diffusion. Innovation is supposed to be
stimulated by making patented problem solutions known to the public. As
far as I know, and there was also consensus on this issue at IFRI,
software patents does not contribute at all to knowledge diffussion in
the software industry. Many patent applications are even formulated as
difficult as possible to make claims as broad as possible.

It is worth noting that, unlike other fields, in the field of software
and computer science references to the patent litterature in the
techical and scientific journals are extremely rare.


Q10.
I am not sure, but I think not, because of lack of data and the
intransparancy of the litigation business. But I think Iain Cockburn has
shown that an approved patent is not visible in stock market value (new
press release/commercial generates 7% increase).
IFRI conference:
http://cip.umd.edu/ifriworkshop2.htm

However, it will definetely bring more jobs to the field of patent
litigation. The Commissions analysis of the replies to the hearing
it held on this issue also shows that 44% of replies with a positive
attitude towards software patents come from IPR professionals.
But IPR professionals do not innovate or develop, they are an overhead
that will be needed for administration of software patents.


Q11.
In Sweden, a professional handeling of the reception alone of one patent
infringement claim (valid or invalid) per company per year in the SME
company sector 1-20 emplyees, will cost that segment 60 million Swedish
crowns per year. Even for larger companies the burdon of patent
administration is worriesome. In swedish:
http://www.softwarepatenter.dk/wmdata.html

Even more important is perhaps tax evation possabilities that may vary
from different EU-countries, but that's an issue I don't know too much
about.

For SMEs the sheer number of software-related patents issued by the EPO
is so high that it is virtually impossible for a person skilled in the
art of computer science to read them all and also have time to do real
development work. You have to have some patent staff to handle that.
Most SMEs don't.

SMEs are largely actually against the directive proposal. However many
SMEs in the software industry are very dependent on one of the big
players, like IBM or Microsoft, and may be reluctant to state their
opinion in public. This is probably another reason why so much
opposition seems to come from the Open Source movement. SMEs use
this movement to state their opinion, without coming out in the open
where they risk repercussions from large companies. I actually got some
help whith the questions from a company producing only proprietary
software.


Q12.
The impact of TRIPS, if the directive proposal is implemented, is
probably unlimited patentability. This is due to article 3 in the
directive. TRIPs does not allow member countries to have patent
exceptions, like software on a diskette or business method software.


Q13.
The construction of the EPO with the diplomatic conferences and strange
voting procedures are not compatible with the EU project. As far as I
know there are difficult unsolved constitutional questions where the
proposition to make EU a signature of EPC and thereby a member of the
EPO is a sign of this complexity. I have even heard that you cannot go
further than to a EPO TBA, not even to the European Court. That is why
adopting a praxis that has evolved at the EPO is a step in the wrong
direction. In the future the EU should be able to handle all legislative
and executive powers of an european patent system.

The implications of the directive would also be that any debate or
hesitation to legalize the EPO praxis in the non-EU EPC countries may be
interupted. It might be wise for the EU to ask these countries to state
their opinion, and consider their arguments before deciding on this
issue.


Q14.
If a completely new original and independent computer program can
infringe on a patent on a "computer-implemented invention", I don't see
how a decompiled version of a patented computer program would not.

The decompilation and interoperability provisions are inconsistent with
the whole idea of patent protection of computer programs.