What constitutes an original patentable idea in software?

Question

We regularly hear a lot of news about software patents and how certain companies own or are buying gobs of patents in order to stave off potential law suits.

What constitutes an original patentable idea in software?

Answer 1

I can only speak to the United States and actually that is somewhat all that matters as it is very difficult to get software patents in other countries.

In the US, 35 U.S.C. 101 states that "Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title."

So, what does that mean exactly? The attorneys at the invaluable website bitlaw.com have complied a list of relevant cases concerning software patents. There are a few take away points that can be summarized:

• You can't patent algorithms in their purest sense detached from the physical world. The relevant case is Gottschalk v. Benson. They must be attached to something in the physical world.
• You can patent software that is a part of a larger invention connected to the physical world. The first case on this issue was Diamond v. Diehr. Specifically, if what you are applying for performs a function that transforms or reduces an article to a different state or thing it generally would be afforded patent protection.
• However, the test mentioned in Diamond v. Diehr is not to be used as the sole test for patentability as pointed out in the recent Bilski v. Kappos case. The court in Bilski stated that the machine-or-transformation test is not to be the sole determinant for patentability.

Which brings us to the present day. It is now unclear what the particular standard is that software needs to meet to be patented. If it is connected to a machine, or is involved in transforming something else then it is probably patentable. So what you will see is patents for software that have some sort of effect in the physical world like e.g. video compression software that produces high quality video on a display screen to a user or something.

For a much more thorough treatment of the topic and far better insight than I can provide I highly recommend the blog Patently-O.

Answer 2

ihtkwot's answer to this question is accurate but it focuses only on the notion of "patentable subject matter." When I consider the patentable subject matter doctrine, I like to imagine a really ingenious idea and then ask whether that is the type of idea that should be amenable to patent protection. Subject matter eligibility is also sometimes about how concrete the idea needs to be in terms of its implementation.

Although the public debate on software patents tends to focus on subject matter eligibility, that doctrine actually does very little work in the patent system. Rather, the focus of patent attorneys and patent examiners is (1) whether the invention (as claimed in the patent application) is a significant step beyond what was known and (2) whether the patent application specifically and concretely describes and claims that invention. Step (1) above is embodied by the doctrines of novelty and nonobviousness. Step (2) is embodied by the requirements of definiteness, enablement, and written description.

Nonobviousness: In terms of actually obtaining a software patent at the USPTO, the most important and most used doctrine is that of nonobviousness. When someone comes to the USPTO with a patent application, the examiner asks the hypothetical question of whether the invention would have been obvious at the time the application was filed (or sometimes at the time of the invention). Importantly, the hypothetical question employs a hypothetical Person Having Ordinary Skill In The Art (PHOSITA) who is familiar with all the relevant prior art.

In the early 1900's, courts began requiring a "flash of genius" as part of the patentability test. That approach was rejected by Congress in 1952 Patent Act and replaced with this ordinary thinker - PHOSITA. To be clear, PHOSITA is not a creative genius -- and thus the law allows for lots of ideas that are non-obvious. (This helps explain why there are so many patents.)

One important element of the obviousness analysis is that the patented invention must be considered "as a whole." Very often, every component element of an invention is already well known. In those cases, the invention is found in the new combination of features or steps.

One problem with obviousness is that the results always require a judgment call -- this results from the fact that PHOSITA is a figment of our imagination. This inexactness is by design. It allows the law to be more flexible in response to new innovations, but it also makes resolution more contentious and less predictable.

Finally, the law begins with the presumption that a patent application properly claims an invention. The USPTO Examiners have the burden of providing evidence that shows the invention is obvious. We are all good at hand-waving, but the law requires evidence (combined with common sense).

Lets look at an example - I just randomly pulled up U.S. Patent Application Pub. No. 20120240162 that was published last week, titled "Content Provision" and owned by Zeebox. That application has been applied for, but is not yet patented. Claim 8 reads as follows:

1. A method of operating a television system, comprising the steps of:

receiving live television at a user device which includes a channel selector;

displaying the received live television at the user device;

selecting a channel using the channel selector; establishing a direct connection between the channel selector and a server device;

controlling channel selections at the user device by sending instructions from the server device which is operatively connected to the user device, but is remote therefrom.

So, we basically have four simple steps that control the interaction between (1) a user device that shows live TV; (2) a server that is remote from the user device; and (3) channel selector that directly connects with the server. Has this been done before? How would you prove that it would have been obvious?

Answer 3

Sadly, and incorrigibly, I think the safest answer to this is:

Whatever survives litigation.

Answer 4

In addition to what @dennis-crouch said (and I read his blog -- its excellent), recent case law (Alice and its predecessor cases going back to State Street Bank) teach that you will need to define your "invention" -- i.e., write claims -- that demonstrate a particularly useful (transformative?) result. Just "computer-enabling" a known process is likely to not meet that standard (since, after all, the result is the same as not computerizing the process, even if the computer is faster and more efficient). The "computerization" itself must lead to a useful, new and non-obvious benefit. A skilled patent attorney with many years successfully drafting software patents will not try to formulate broad abstract claims that don't meet these tests, but will rather focus on the computerized steps in the process where computerization provides a benefit not otherwise (easily) obtainable.

So, in the example, I don't see anything in the claim itself that meets this test -- what benefit is obtained by having the server change the channel on the remote device? But, importantly, there may be aspects of the disclosure that say what the benefit is -- and I didn't read the disclosure. Maybe the remote device is one of many devices that just passively receive the content or changed channel content and allow multiple displays of the same content. Maybe the direct connection between the channel selector and a server device includes a time synchronization that improves the battery life of the channel selector. There could be other benefits to computerizing this client-server method. We'll see as the application goes through prosecution, etc. (has a patent issued from it yet?)