We invite the engineering community to join this debate in the discussion thread below. But first, some background.

Bernard Bilski filed a patent in 1997 on a business method for hedging financial trades, but the Patent Office rejected it on the basis that it was not tied in some way to a physical device.

A Federal court upheld the decision and laid down a controversial test for any patent: it has to be tied to a device or transform something physical. (Wikipedia lays out many of the details of the case.)

The Federal court decision has sent waves of concern across the industry where more and more innovation is in software or hybrid mixtures of systems, software and business methods. Dolby, Google, the IEEE-USA, Medtronic, Microsoft, Philips, Pitney Bowes, Red Hat and Yahoo were among those who sent briefs to the Supreme Court arguing their opinions on the case.

"The new test discriminates in favor of users of hardware technology and against users of software technology," said Philips.

"The decision is overreaching, works an unnecessary sea change in deep-rooted principles of patent law, and will necessitate a massive revaluation of America's intangible technology assets [that] disrupts our research and business environment," Philips added

Medtronic gave several examples in genetics and other medical fields where the decision could harm innovation. "For companies focused on medical technology innovation, the prospect of a revised patent eligibility standard that is less inclusive presents the grave concern that the development of critical lifesaving medical technology will be impeded," it said.

Microsoft was among a handful of companies who suggested a solution, in its case a return to a requirement that patents involve something physical.

"This Court has long recognized that a patent-eligible method must involve one or more disclosed physical things—that is, it must describe a series of steps that use physical means to produce a result or effect in the physical world," Microsoft said. "This test for patent-eligibility has withstood the test of time, and it is sufficiently flexible to accommodate technological advances," it said.

The IEEE-USA said the Federal court's decision "creates new confusion" that is causing "collateral damage to software-based patents, with the Patent Office denying patents and district courts striking down granted patents." Like Microsoft, it recommended returning to an older and broader patent test involving "machine, manufacture, or composition of matter" the "provides a good and clear line for even the most cutting-edge technology," it said.

Open source advocate Red Hat argued a bolder position against software patents generally. The fact that 200,000 or more software patents have been granted since the practice was encouraged by Federal Court decisions in the mid-1990s hasn't helped advance the industry, it argued.

"Such enormously successful software products as Microsoft Word, Oracle Database, Lotus 1-2-3, the Unix operating system, and the GNU C compiler all date from the 1980s or earlier—well before the proliferation of software patents," Red Hast said.

We'd like to hear what the community of engineers thinks. You create the technology behind so many of today's patents. What do you think should be the test or definition of a patent?