Software Is Not Necessarily Business Method
Section 101 of the patent statutes (Title 35 of the U.S. Code), broadly outlines the categories of what is patentable subject matter. These include "any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof." A business method falls into the "process" category, but claims in patents directed to software can be written as a process, machine, or manufacture. However, people commonly confuse software patents with business method patents. These are different things, and although the distinction can be difficult to discern because businesses today rely on software so much, the distinction is likely to become important over time because of the U.S. Supreme Court opinion in Bilski.
Most people understand that a claim directed to software can be written as a process. They are commonly drafted as "method" claims, which is probably why many people confuse these with "business method" claims. A business method does not necessarily involve software at all. For example, a method of hedging energy risks does not necessarily require software. People were hedging risks relating to commodities over a hundred years ago -- well before electronic trading devices were invented -- by buying and selling commodity futures.
Claims directed to software can also be written as a machine or manufacture. Patent attorneys commonly draft claims directed to computing devices and computer-readable storage media that implement or embody the software. We do this to provide broad protection to our clients so that someone who makes, sells, or imports infringing software can be liable in addition to a user of the software who only infringes when the software is run and it performs the claimed process.
The media equally confuses software and business method patents. For example, an article published in the San Jose Mercury News last week equates software with business methods. It provides examples of patents owned by software companies (one of which was successfully procured and litigated by Perkins Coie) that are drafted to include aspects of computing. However, these are not pure business method patents: the claims are drafted in terms of what a computing device does and not what a business does. Although many business methods today are performed using computing devices, it is likely that those same methods could be performed without the computing devices. On the other hand, most software inventions would be nearly impossible to implement without computers.
The article goes on to state:
Folks such as the Electronic Frontier Foundation have been railing against the evils of such patents for years. "The patents can be so broad that they cover a lot of activity that people have been doing for a long time," said Michael Barclay, an attorney and EFF Fellow. "The result is a lot of patents that get allowed that shouldn't get allowed."
This statement is a red herring. Once the United States Patent & Trademark Office (USPTO) determines that an invention falls within one of the eligible categories outlined in Section 101, it has to determine whether the claimed invention is novel (Section 102) and not obvious (Section 103). However, patent examiners are not perfect and can award patents that should not be awarded because they do not have infinite time to analyze all the prior art even if all of it could be identified. However, this is a problem in all areas of innovation and not just software (or business methods, for that matter). If a patented claim covers "a lot of activity that people have been doing for a long time" as Mr. Barclay suggests, such a patented claim is invalid under either the novelty or nonobviousness requirements and cannot be successfully enforced. Just because the USPTO granted a patent because it had inadequate resources to properly and fully examine the underlying patent application, it does not mean that all innovators should be barred from receiving patents on their innovations.
The article states that because the U.S. Supreme Court opinion concluded as it did,
the folks at the patent office were feeling the relief of someone who had just avoided a neutron bomb going off in their house. Rather than descending into chaos that would have come with reviewing thousands of patents, patent officials were thinking about issuing follow-up guidelines at some point.
This reflects another misconception a lot of people not familiar with the patent system have. The USPTO does not review previously granted patents unless someone requests a reexamination, e.g., because of newly found "prior art." On the other hand, the law evolves continually and patents that were valid one day may become invalid the next because the law changed. As far as I know, the USPTO does not have the authority to revoke a patent sua sponte, and a defendant's recourse is to request a court to find that an asserted (or threatened to be asserted) patent is invalid as a matter of the new law.
Luckily for the software patent bar, the Supreme Court Justices wisely recognized that tests to determine whether an innovation was patentable during the Industrial Age are irrelevant during the Information Age. (U.S. Supreme Court Slip Op. at 8-10.)
