[pp.int.general] software patents

[Reinier Bakels]

>
>    - It is not difficult to design a mechanical invention that only 
> depends on
>    some geometry.
>
> That's no counterexample to my argument.  I think you have focused
> on a part of the argument, and not on the overall point.
Strictly speaking, you are right. I attended some of your presentations and 
I recall that you argued that a software designer is faced with a machine 
that has a strictly predictable behaviour while mechanical and electrical 
engineers are faced with a physical reality which has often surprising 
side-effects (hope I phrase your argument correctly).
I would say that:
- mechanics (for instance) sometimes is quite predictable
- software sometimes inot predictable
- if some field of technology is really easier, inventions in that field are 
likely to be "obvious to a person skilled in the art", so they are not 
patentable.
>
>    - Software design for control enginering and signal processing 
> applications
>    is often non-trivial.
>
> Signal processing interacts with physics; it is typically part of a
> specific physical system.  So it does not really contradict what I
> said.
Perhaps I should have said: *digital* signal processing. I wonder whether it 
relates more to the physical world than other software, e.g. an editor. The 
European Patent Office would have said that signal processing software 
(perhaps) brings a "(further) technical effect". It is the slippery slope 
that eventually allows (for instance) a new malloc implementation to be 
patented, basically because programmers consider memory management a 
"technical" problem.
In the debate on the former EU "CII" directive, the (pro) argument was "no 
this is not about software patents, but about patents in 
computer-implemented inventions" (computer controlled conventional 
technology likes TVs or X-ray equipment). This is a departure from e.g. 
Parker v. Flook in the US (1978) or Dispositionsprogramm in Germany (1976): 
decisions that both distinguished the "abstract" processing of data 
(inventive but not patentable), and the conventional technology part that 
was identified as "post-solution activity", and which was considered obvious 
anyway. (Flook was about a chemical plant)
In practice, digital signal processing is an important type of sw patents: 
the very first EPO sw patent case, VICOM, was about signall processing. A 
similar US decision is Alappat in the 1990s. Nowadays, there are often 
police raids on trade fairs because a (usually Chinese) manufacturer ignored 
the MP3 patents.
>
> But even if there are some specific areas of software which are
> exceptions to the general conclusion, that is of small importance
> compared with the rest of software, so the conclusion is still valid.
> We're talking about public policy, not mathematics.
> A small exception does not make a policy argument invalid.
>
> I'm not going to defend non-software patents.
> If society is ready to get rid of them, I won't stand in the way.
> But it is very useful to be able to show why software patents pose
> a special problem.
I think the type of "special" problem of swpats pertainms to other fields as 
well. I may be useful to consider "the swpat problem" in a historic 
perspective.

The delimitation of "pure" software inventions vs. "computer implemented 
inventions" always posed problems (even in your own argument). The franmers 
of the EPC in 1973 did not know how to handle it, do the decided to adopt a 
limited exclusion of sw, "software as such". Nobody know (and knows) what 
that means. It was about the time when independent sw providers emerged: 
until well into the 1970s, IBM provided operating system software for free, 
because it was written for IBM argcitecture machines, that were only 
provided by IBM. When (Japanese) "plug compatible" mainframes emerged, 
running IBM software, if was obviously in IBMs interest to charge the use of 
software, specifically if it ran on their competitors equipment.

In rhe 1970s, patent offices were very restrictive in allowing swpats, but 
in the 1980s they became more responsive to (alleged!) "market needs". A 
landmark decision is T 1173/97, in 1998, about the same time as "State 
Street". The European Commission wrote a write paper on (software patent 
policy), also in that timeframe. In 2002, the EU Commission issued the 
infamous "computer implemented inventions" directive proposal, which was 
rejected in 2005. Which does not mean that sofwtare patenting is not (or no 
longer) allowed! It only means that EPOs permissive case law is not 
codified. In EU, there are just two countries that really matter for 
patents: Germany and the UK. EPO and Germany usually agree (EPO Boards of 
Appeal reside in Munich), the British foster their own "stubborn" approach 
(incidentally not just on swpats but also on verious other aspects of the 
patent system). The Directive basically should have aligned the British with 
the EPO and Germany. After the directive was rejected, the British gradually 
adopted EPO rules voluntarily, because they believed that harmonisation is 
an objective by itself.

A parallel development is the emergence of Open Source software. Software 
patents are incompatible with Open Source software. Which was (probably) the 
reason why patent policy making was eventually politicised.

And now it is 2010. Software patents are no longer a political issue. Still, 
patents in general are very much a political issue: presumably the climate 
summit in Copenhagen failed due to patents:on the one hand the perception of 
US firms that patent exceptions will allow firms abroad to "steal" US 
technology, on the other hand the awareness in emerging lesser developed 
countries that patents basically are a new form of colonialism. Climate, 
health, food: patents affect many aspects of our life. And frankly, swpats 
are not that much of a social problem.

Should the pirate party movement advocate the abolition of patents? The 
question is similar to the question whether it should advocate the abolition 
of copyright - which kills the Open Source model as well. In my perception, 
a first step is to get the perception accepted that patents are *NOT* a 
means for the "commodification" (or: appropriation) of any form of useful 
knowledge (which is very much the perception of the US patent system today). 
Patents should be considered *exceptions* to the rule of freedom of 
competition - which is the basic principle of a non-communist society. And 
patent policy making should be considered part of economic policy making. 
Perhaps in the US it is better, but in Europe, the EPO is an island, 
completely detached from actual economic policy making. An exception is 
perhaps DG COMP in the European Commission, but its criticism eventually did 
not affect DG Market's "intellectual property" policy very much. And DG 
Market is a dedicated follower of corporate US interests. They listen to 
"rent seeking" firms. You may recall that professor Hugenholtz was very 
angry that his advice (on copyright policy) was simply ignored by DG Market, 
even though it had been commissioned by the EU!

The faulure of "Copenhagen" is a prime argument to revisit the patent 
system. If the UNITAID initiative fails, it is another argument: the 
perception that patents *even* don't forster pharma innovation could be 
crucial to create public awareness that patents are BAD for society.

reinier