They state that patents should not affect the majority negatively in order to be valuable. Patents should apply to scarce resources (to motivate the creation of more of them).

They also mention the ways the system is abused: patents that are essentially obvious and patents where an observer simply puts on paper the current state of the art (common practice) perhaps at a time when little prior art exists (which they can work around) so as to be able to take an industry hostage.

They mention that testing obviousness by mostly only looking at prior art (much of which is missed anyway), the USPTO fails to account for many things that are considered so obvious and standard that they are never written down but are known by everyone in the industry.

Also, they justify their submission because of the importance of having laws that are consistent across nations and also that SCOTUS ruling will be taken into consideration by the EU so much care should be taken to be fair and correct.

[I read the brief entirely first and them mostly paraphrased without reviewing all the details, so some of the above statements might be a bit off. It's a great read for anyone watching the software patent battle closely.]

It’s offers a very focused point from the legal point of view.

It has a great section on the differences between software and hardware. These include obvious things we all recognize that differentiate quite clearly the ease by which numerous things can be done with what is intangible but which cannot be done with what is physical.

It addresses IBM’s contention that reverse engineering gains makes not having source code a small issue.

It addresses the lie that patenting opens up inventions to the world, and states that open source (through copyright) do fulfill this openness and very successfully.

It also hits on the economic problems/costs with patents and that software patents won’t promote progress. [Patent supporters talk about the positive side of the ledger (more jobs, blah blah), but ignore the costs and inefficiencies of such a system. Would you pay 1 million dollar for a cheeseburger? Why not? You'll gain a cheeseburger. Don't you want to gain something? Ah, but at what costs?]

It too takes the position that SCOTUS and other important courts have been almost clear that software is abstract and cannot be patented (and loading it onto a machine is a trivial step) and have drawn an almost clear line between what is patentable and what is not, but suggests an improvement to the Machine/Transformation test from Bilski (leveraging European experiences that stopped those that try to game the system) so that lower courts and the USPTO can more easily avoid future mistakes. The improvement is as follows: to reject abstractions outright within claims rather than do what is currently done of allowing these abstractions if the abstractions are a subset of the whole.

It gives historical European examples of lessons (eg, for business patents) we should keep in mind if we don’t want to repeat the mistakes.

Great exposure for Red Hat (and taking share from Microsoft), Linux, FOSS collaboration (great for security, privacy, and low cost), and the problems with software patents.

This video is probably a wake-up call to groups betting on Microsoft.