Ohne einen starken Schutz durch Patente könnte es sein, dass sich die Software-Industrie in einem gewissen Sinn “zurück in die Zukunft” bewegen wird und neue Firmware-basierte Systeme, ähnlich den bereits für Breitband-Modems und andere Hardware-Peripherie benutzten, entwickelt. Aus Die Zukunft der Softwarepatente in Europa, einem Plädoyer pro Softwarepatente von Joel Buchner vom Institut für Unternehmerische Freiheit, das als IUF weekly comment 17/05 im englischen Orginal erscheint. Da Sascha T. sich aktuell in Russland herumtreibt (nein, nicht um Chodorkowski zu befreien ;)), trifft mich das schwere Los, den Comment hier zu publizieren. Als entschiedener Gegner der Patentierbarkeit von Software würde ich eine lebhafte und fundierte Diskussion zu diesem Text sehr begrüssen …

The Future of Software Patents in Europe
The battle over intellectual property in Brussels continues with few signs of ending anytime soon. The Council of Ministers took a pass on the issue of software patents allowing the EU to continue granting software patents in cases of unique innovation. Microsoft is now suing the European Commission to protect its server protocols from being released to its competitors. Future EU parliamentarians, however, may restart the legislative process to end software patents. In the long run, the momentum of the European debate seems to be behind Michel Rocard and his band of socialist, technology-loathing MEPs. The lack of comprehensive research on the issue, combined with the grassroots prowess of European Linux users has not favored those who want stronger protections of intellectual property.
Michel Rocard and other MEPs have made clear their distaste for granting software patents and intellectual property of all types. Leaning toward the far left, Rocard believes that anything “intellectual” is part of a worldwide store of knowledge. Rocard states the following: A civilization should be preserved where the place of the world outside the market and of the human intellect is respected. Rocard elaborates in a 2003 translation of an interview with the French magazine Liberation where he specifies: Software´patents should be permitted only if something uses the forces of nature or acts on matter Where one draws the line between software and the “matter-based” hardware, however, is not as clear as one might think. It is only fair to note that Rocard admits what he is not: one of the generation which has an easy facility with the computer. As a president of the Committee for Culture in the European Parliament, he had to plunge himself with an evil madness into the issue of software patents. A year before the interview Rocard made clear his lack of understanding of the issue, stating that software patents were unknown to me. Indeed, intellectual property is one of the most complex and technical policy issues debated today.
Rocard puts his finger on the crux of the issue. Are devices (such as computer hardware) that act on matter so different than devices created by written code? Rocard claims that “invisible” devices do not merit patent protection, but there is no clear difference hardware and software at the most elementary level. Computers are complex systems made of hardware that is inextricably intertwined with operating software. Drawing a line between software and hardware seems to hinge on whether one views programming code more as instructions (or schematics) to control and in a sense “build” a computer, or part of some amorphous realm of free knowledge.
Cryptographic systems engineer Terry Ritter clarifies the issue further in his enlightening 1991 paper from Midnight Engineering, “The Politics of Software Patents”: So called “low level language” or assembly-language software consists mainly of symbols or specific words, each of which directly selects a particular operation for the physical machine to perform. Such software can scarcely differ in intent from the sequencing performed by a piano roll, a music box cylinder, a clothes washer selector or dishwasher selector, or any other machine sequencing device which is normally considered a part of the machine.
Basically, Ritter finds that the assembly language of computer software does little more than flip transistor switches on a processing chip albeit in a highly sophisticated sequence. Perhaps viewing the code not so much as words or symbols, but as the contents of a hard drive further clarifies the point. Why should patent authorities not view the magnetic image of a computer software program as its primary, physical representation?
At face value, viewing assembly code as the “schematics” for a magnetically-based system that operates computers seems more valid than Rocard?s simple claim that Software is nothing more than a collection of mathematical formulae. Moreover, why should mathematical formulas belong to the world’s store of knowledge when schematics to “physical” devices as simple as the cotton gin are easily patentable. At some level, all devices are intellectual in nature, because they are based on complex schematic drawings. It is these drawings and instructions that are stored in government offices and form the basis of actual patents.
One could also argue that hard drives, with their magnetic images of computer code, are as much a hardware-based and patentable device as a computer chip. Computer chips have instructions built into them and, similarly, one could see a tangible magnetic image of the assembly language on a hard drive with a very good microscope. However, we could not comprehend the instructions in the magnetic image. If no person can read the assembly language in any form, why should it be described as knowledge?
Further, if all knowledge is a commons, why would Rocard not eventually reject the granting of patents to ?physical? devices? It seems possible then that any device worthy of a patent is nothing more than an abstract and subjective notion of a highly unique design without´prior art. Software, for instance, can create devices such as a calculator. Had a software version of the calculator been invented before the hand held version, or even the primordial vacuum-tube calculator created and patented by Babbage, would it not have merited a patent? Would a hypothetical hand held calculator invented after a software version also have merited a patent? Is it right to grant two different patents of the same virtual device if one is based purely on code for a computer and another on schematics for building hardware? More importantly, who has answered these complex questions? Without a more robust intellectual foundation for patents of all types, it seems valid to speculate as to what might happen if a future EU parliament and Council of Ministers agreed to end all software patent protections. One result could be a race to develop new code-embedded hardware protected by patents under most intellectual property laws.
Without strong patent protection, the software industry might go “back to the future” in some sense and create new firmware devices similar to what is now commonly used to operate broadband modems and other computer hardware accessories. Hardware routers, networking cards, and modems, of course, can be patented. The firmware loaded into the memory of these devices can also be patented. Would the EU have the gall to eliminate patents for firmware? In this worst case scenario, consumers could end up buying hardware keys or even application chips to plug into their computers. A hardware-based coding industry, however, might require new economies of scale that would penalize smaller software firms. Although the cost of selling software without “hard” media disks has approached zero with the advent of DSL and cable modems, only hardware keys or application chips can thwart pirates from cracking software applications in a world without strong intellectual property protections. There is already evidence that weak IP protection is leading firms back down the road to pure hardware-based coding instructions.
Software development firms selling low volume, high-cost software for the music and design industry are already using hardware keys to protect their code. Many music synthesizers sold today are nothing more than housings for software- and firmware-based computers designed to make it nearly impossible to steal the underlying code, despite the fact that a more profitable, pure software version of the same device could easily be produced. Expensive design software often requires a hardware Universal Serial Bus (USB) key for operation. If patent law is evolving into some sort of race between bureaucrats, pirates and industry, who stands to benefit from this race other than the government? Governments could just as easily consider making patent and copyright protections stronger to reduce the need for code locking devices.
Critics of software patents might be appeased by efforts to streamline the process of obtaining intellectual property. Many smaller software development firms support the elimination of patents, because they feel they cannot participate in the current system due to the costs of lawyering and administration of a very bureaucratic process. Fairness and the need for competition demand that patent applications be simple and require little in the way of financial resources. Easier access to patents and copyright protections would promote the growth of small software firms, but would also make the institution of intellectual property stronger by creating more owners of IP.
In an environment where political outcomes are so unclear, one must ask whether Rocard’s belief in the “intellectual commons” actually has any bounds. Although Rochard claims to support patents for physical devices, why should anyone believe an avowed socialist advocating a right to property?
One thing is certain, though: Nokia, SAP, Bayer and other European corporations create thousands of research jobs based upon patent protections for devices critical to the medical, chemical, computer, transportation and other industries. For the sake of protecting these jobs as well as jobs in the software industry, the EU must avoid any rash decisions on the issue by commissioning more research on software and intellectual property. Decision makers must be better informed in future decisions regarding this complicated technical issue.
The dearth of research on the topic of software patents combined with the number of plausible views on the issue of intellectual property has created a confusing and disorderly debate, making it difficult to draw the line between what Rochard and industry proponents actually think about the issue. Ultimately, the greatest threat to intellectual property and software patents is the complexity of the issue. There does not appear to be an easy way to end the debate over intellectual property so it will take years of research and further debate before any well-baked ideas evolve. For now, the maxim “don’t fix it if it isn’t broken” seems the prudent course until the EU and business-interests lay a more robust foundation of legal and technical research.