Linked by Thom Holwerda on Sat 30th Jun 2012 19:34 UTC
Legal Yesterday, we were treated to another preliminary injunction on a product due to patent trolling. Over the past few years, some companies have resorted to patent trolling instead of competing on merit, using frivolous and obvious software and design patents to block competitors - even though this obviously shouldn't be legal. The fact that this is, in fact, legal, is baffling, and up until a few months ago, a regular topic here on OSNews. At some point - I stopped reporting on the matter. The reason for this is simple: I realised that intellectual property law exists outside of regular democratic processes and is, in fact, wholly and utterly totalitarian. What's the point in reporting on something we can't change via legal means?
Thread beginning with comment 524747
To view parent comment, click here.
To read all comments associated with this story, please click here.
Neolander
Member since:
2010-03-08

Let's start with three fundamental principles which I strongly hope we will agree on. 1/IP law is here to address the problem that creating new stuff comes at a cost and brings little rewards in itself, by offering various legal ways through which innovation can be indirectly financially rewarded. 2/This protection of existing innovation should not come at the cost of hampering future innovation by putting ridiculously high demands on new actors of a given market. 3/The legal protection of a given innovation should be proportional to the extent of the innovation that is being protected.

Given this, I can start to address your first question : what is it that makes software patents so dangerous as compared to the copyright that protects novels and music ? Basically, it's the granularity of each legal construct.

In most countries that implement copyright, there is a fair use clause in the law that states that people can freely copy, say, a few pages of a book, or 30 seconds of music. Such a legal provision is necessary because otherwise, people would be able to get IP protection on single language constructs (words, sentence, 5-tune musical arrangements), which are not innovative in themselves. This would violate principle #3, and ultimately principle #2 since innovators would end up stepping on copyrighted constructs by accident, allowing the IP owner to act like a dick and sue them for 5 notes of music.

Software patent law, however, often has no such clauses, which is made worse by the fact that patents apply to nonexistent products. It allows IP holders to patent language both figuratively (as in the case of multimedia codecs, which are languages that multimedia devices use to communicate with each other), and literally (as in the case of Apple's patents on single man-machine interactions such as slide to unlock, multitouch gestures, or scrolling mechanisms). Now, IP protection of language is a huge violation of #2, because new market actors come with a huge handicap if they have to spend years enumerating patents and pay absurdly large sums of money just to make their products work together with existing products and user habits. And even if you don't agree with this moral principle, basic concerns of usability and interoperability should have been sufficient to get this abuse of patent law outlawed. So far, probably due to financial pressure from big companies, it obviously hasn't.

In the case of software patents, the ability for market actors to patent language has already been greatly used in litigation (Unisys' patents on GIF, Apple's overlapping windows and touchscreen interaction lawsuits, Microsoft's FAT long file name patents), which is why the tech world is aware of that issue. But similar issues can also exist in hardware, as soon as you bring design patents to the mix. Individual features of hardware design, such as the use of specific materials or simple geometric shapes, can play just as much of a vital role in device engineering as language plays a role in man-machine and machine-machine interaction. Thankfully, I believe that the former is illegal. The latter, however, obviously isn't, since Apple managed to get a patent on what is basically a parallelepipede with rounded corners not so long ago (and as any hardware engineer will tell you, rounding edges is a vital feature of any device which does not want to hurt its users). Can you imagine what would happen if someone patented such a core element of everyday hardware as Phillips screws ?

So basically, when people say that they are against software patents, they are not necessarily against the concept that software should be able to receive legal protection like any other innovative kind of engineering work. They are maybe just against the current usage of software patents, which is to patent minor details of a product instead of patenting a complex whole (as in your example of a whole, existing as opposed to virtual, shoe design).

So basically, I suggest to outlaw any direct patent on language, whether it concerns man-man, man-machine, or machine-machine interactions (I'm actually surprised that this is not the case already, since we forbid patents on mathematics which is one kind of language), or alternatively to add a legal clause that allows any patent to be bypassed at no cost for the sake of interoperability.

Patent legislation also desperately needs coarser granularity like what is done in the realm of copyright, so that obvious details cannot be patented, but I'm not sure how this could be expressed in legal terms. An option could be to limit the amount of patents which a company, no matter how big it is, can hold, by making patent registration fees directly proportional to the income of the registrar. This would also be more fair to small market actors than the current fixed-rate model. However it would also require provisions against compagnies that create unofficial subsidiaries whose sole purpose is to manage patents and litigation (hello, Apple and Microsoft !)

Edited 2012-07-02 06:10 UTC

Reply Parent Score: 3