Information Economy and Swpat Conference Paris 20020610-1

The Conference
Priorities according to Hartmut Pilch
Relevant Work, Background Documents

The Conference

CIP/IFRI 2002/06: Conference in Paris on the Impact of Patenting Immaterial Innovation: Institut Français des Relations Internationales (IFRI.org) and Center of Information Policy Research at Mariland University (CIP.umd.org) co-organised in June 2002/06 a transatlantic conference on information economy and in particular on the limits of patentability as well as the problems in neighboring areas such as database exclusion rights and copyright.
Paris 2002/06: Papers for "Frontiers of Ownership in the Digital Economy": Papers presented by the participants at a seminar held in Paris in June 2002, among them many economists and social scientists from the US.
Paris 2002/06: Background Materials for "Frontiers of Ownership in the Digital Economy": Links to documents in the WWW which are related to patents on immaterial innovation (rules of organisation and calculation, software & business methods) and to the European Commission's recent push to sanction such patents by a directive.
CIP/IFRI 2002/06: Program of the Software Patents Workshop: Schedule of the talks planned for the IFRI conference.
DG IV Bakels 2002-06-19: The Patentability of Computer Programs: This paper was "opensourced" by us shortly before the conference and published by the authors, who attended the conference, a few days later.

Priorities according to Hartmut Pilch

Any discussion should be conducted on the basis of example patents, as outlined in the Software Patentability Legislation Testsuite approach.

The causality between law and various economic effects is an indirect one: the legal system produces patents, and the patents produce the effects. The example patents are an intermediate link in the causality chain which people investigate, and this link is often missing from the investigation.

The European Commission's Directive Proposal displays a few formulas, such as the "requirement of a technical contribution in the inventive step", which suggest that a limitation on patentability is intended. Upon close reading, it however turns out that this proposal legalises all types of patents which are generally deemed undesirable and removes all limits on patentablity by 4-5 layers of safeguards, designed to ensure that patentability can in no way be limited and that lawcourts can no longer find any support in the law for refusing even the most undesirable kinds of patents on rules of organisation and calculation (information innovation, business methods).

Studies such as that of Fraunhofer/ISI have not made it clear which patents, granted according to which patentability rules, they are talking about and yet investigated industry opinion about such patents / patentability rules. Such a flaw can seriously reduce the value of a study. Several recent government-ordered studies from Europe seem to suffer from this flaw.

Even some of the best studies from american academia don't make it very clear which sets of patents, granted according to which rules, they are investigating. The distinction between "industries" or "fields of technology" is not very sharp. A sharper delimitation is possible and should be attempted.

In particular, the concept of "technical invention", as employed by German courts and legal literature throughout the 70s and 80s (and partially until today), is founded in non-legal categories that are successfully used in the philosophy of science and serve to define sharp boundaries. I would like to recommend its use to economic researchers, particularly from the USA, where this concept was not as well established and has been forgotten for a longer time.

Some other ideas which I may like to propose to the conference are found in the "background documents".

Relevant Work, Background Documents

Software Patents: Questions, Analyses, Proposals: Wie wirkt das Patentsystem auf die Wirtschaft im allgemeinen und wie auf die Informationsökonomie im besonderen? Warum sind Softwarepatente so trivial? Wo liegen die Grenzen des Patentsystems, wie wurden sie früher formuliert und wie haben sie sich verschoben? Unter welchen Einschränkungen bewegt sich das Patentsystem? Welche Handlungsoptionen haben wir? Mit den hier gesammelten Artikeln wollen Mitglieder und Freunde des FFII Antworten geben.
Patentability Legislation Benchmarking Test Suite: In order to test a law proposal, we try it out on a set of sample innovations. Each innovation is described in terms of prior art, a technical contribution (invention) and a small set of claims. Assuming that the descriptions are correct, we then test our proposed legislation on them. The focus is on clarity and adequacy: does the proposed rule lead to a predictable verdict? Which of the claims, if any, will be accepted? Is this result what we want? We try out different law proposals for the same test series and see which scores best. Software professionals believe that you should "first fix the bugs, then release the code". Test suites are a common way of achieving this. Pursuant to Art 27 TRIPS, legislation belongs to a "field of technology" called "social engineering", doesn't it? Technology or not, it is time to approach legislation with the same methodological rigor that is applicable wherever bad design decisions can significantly affect people's lives.
CEC & BSA 2002-02-20: proposal to make all useful ideas patentable: The European Commission (CEC) proposes to legalise the granting of patents on computer programs as such in Europe and ensure that there is no longer any legal foundation for refusing american-style software and business method patents in Europe. "But wait a minute, the CEC doesn't say that in its press release!" you may think. Quite right! To find out what they are really saying, you need to read the proposal itself. But be careful, it is written in an esoteric Newspeak from the European Patent Office (EPO), in which normal words often mean quite the opposite of what you would expect. Also you may get stuck in a long and confusing advocacy preface, which mixes EPO slang with belief statements about the importance of patents and proprietary software, implicitely suggesting some kind of connection between the two. This text disregards the opinions of virtually all respected software developpers and economists, citing as its only source of information about the software reality two unpublished studies from BSA & friends (alliance for copyright enforcement dominated by Microsoft and other large US companies) about the importance of proprietary software. These studies do not even deal with patents! The advocacy text and the proposal itself were apparently drafted on behalf of the CEC by an employee of BSA. Below we cite the complete proposal, adding proofs for BSA's role as well as an analysis of the content, based on a tabular comparison of the BSA and CEC versions with a debugged version based on the European Patent Convention (EPC) and related doctrines as found in the EPO examination guidelines of 1978 and the caselaw of the time. This EPC version help you to appreciate the clarity and wisdom of the patentability rules in the currently valid law, which the CEC's patent lawyer friends have worked hard to deform during the last few years.
Call for Action: The European Commission's proposal for the patentability of software innovations requires a clear response from the European Parliament, the member state governments and other political players. Here is what we think should be done.
Fraunhofer/MPI 2001: Economic/Legal Study about Software Patents: In 2001-01, the German Federal Ministery of Economy and Technology (BMWi) ordered a study on the economic effects of software patentability from well known think tanks with close affinity to the German patent establishment: the Fraunhofer Institute for Innovation Research (ISI.fhg.de), the Fraunhofer Patent Agency (PST.fhg.de) and the Max Planck Institute for Foreign and International Patent, Copyright and Competition Law (MPI = intellecprop.mpg.de). The study was largely concluded in 2001-06 and preliminary results were presented to a selected audience. The final report was published by the BMWi on 2001-11-15. The study is based on an opinion poll answered by several hundred software company representatives and independent software developpers, conducted by Fraunhofer ISI. Most respondents have had little experience with software patents and don't want software patents to become a daily reality like in the US. The poll also investigated the significance of open source software for these companies and found it to be of substantial importance as a common infrastructure. Based on these findings, the Fraunhofer authors predict that an increase in the use of software patents will put many software companies out of business and slow down innovation in the software field. The study then jumps to conclude that software patents must be legalised and SMEs must be better informed about them. This surprising conclusion is drawn by the patent law scholars from MPI. The MPI's legal study does not explore any ways to redraw the borders between patents and copyright but just takes the EPO and USPTO practise as an inevitable reality. They find that the EPO's caselaw is contradictory and chaotic and blame this on Art 52.2c EPC, which they say has failed to provide clear guidance and should therefore be deleted. Business related algorithms are, they say, less likely to be patented at the EPO than algorithms that "stand in a tradition of engineering". The MPI writers however do not try to provide a clear rule for distinguishing the two, and they oppose the idea of drawing a line between the physical and the logical ("technical inventions" vs "rules of organisation and calculation") as done by lawcourts in the 70s and 80s, asserting that information is also a physical phenomenon. They propose that all legislative power concerning the limits of patentability be handed over to the EPO, which should then, at its discretion and as far as Art 27 TRIPs allows, consult experts of interested parties for regular rewriting of its Examination Guidelines. Art 27 TRIPs demands that patents be "available in all fields of technology", and the MPI understands "technology" as "the useful arts" and is careful not to mention Kolle and other European theoreticians of the concept of technical invention. Summarily the study can be summarised as "Fraunhofer: software patents are unpopular in the software industry and dangerous to innovation and competition. MPI: Fine, so let's legalise them quickly."
Patent Jurisprudence on a Slippery Slope -- the price for dismantling the concept of technical invention: So far computer programs and other rules of organisation and calculation are not patentable inventions according to European law. This doesn't mean that a patentable manufacturing process may not be controlled by software. However the European Patent Office and some national courts have gradually blurred the formerly sharp boundary between material and immaterial innovation, thus risking to break the whole system and plunge it into a quagmire of arbitrariness, legal insecurity and dysfunctionality. This article offers an introduction and an overview of relevant research literature.
Software Patentability with Compensatory Regulation: a Cost Evaluation: Europe is preparing major changes in its patent system. The European Patent Office (EPO) has proposed to remove limitations on patentability, such as the exclusion of computer programs in Art. 52 of the European Patent Convention (EPC). A report by the French Academy of Technologies supports this proposal but suggests additional regulation measures in order to reduce potential abuses of software patents. In this article, we try to assess the costs of such compensatory regulation. They add up to an estimated 1-5 billion EUR per year for the European Union. Various regulation approaches and cheaper legislative approaches are compared in this article.
Political Economy of the Patent System: the Mechanisms of Patent Inflation: During the last 200 years, the patent system has continually expanded. This expansion is not so much the result of conscious economic policy but rather to some automatic dynamics of political organisations, similar to monetary inflation or to the arms race. This article identifies several causes of patent inflation, traces their evolution and tries to look into the future.


Software Patents > Events > 2002 > Paris 020610
2002	Kiel 021126	EuroParl 0211	Amsterdam 020830	Brussels 20020619	Paris 020610	Linuxtag 0206	EuroParl 2002-05-15	Madrid 020506	EPC 20020610	GUUG 020301	CIP 020128