The patent systems of the world are functioning at a historically high level. This is good because, among other indicators, intellectual property (IP) makes up the bulk of S&P 500 market value.

Leading patent offices are skilled in examining most technologies. They search and examine for novelty and non-obviousness (or inventive step). Meritorious inventions in most technical fields—most industries—are rewarded with patents. A patent, of course, conveys the exclusive right to practice the claimed invention for a period of 20 years from the filing date. Patents translate into market leadership, revenues, and margins. They build capital valuations. Patents attract investments and sometimes even a buy-out of the company.

At the same time, patents have become targets for scorn. Are patents good or bad? It depends on the eye of the beholder.

In the camp viewing patents as “bad” are many engineers and coders who want to be free to play with all the toys in the sandbox. Defendants in patent infringement suits rarely appreciate the patents and suits asserted against them.

In the camp viewing patents as “good” are employers, who generally place a high value on inventions made by their employees, and who often seek to patent such inventions.

Much of patent law and practice is pretty well settled. But software inventions are singled out for special scrutiny and are treated inconsistently within the U.S. Patent Trademark Office and in other patent offices.

Some say that software is a “process” (or method) eligible for patent under the same criteria applied to other technologies. See AIPPI resolution.

Others say that software should be “categorically not eligible for patent.” See Intellectual Ventures v. Symantec (Fed. Cir. 2016) (Mayer, concurring).

In Alice v. CLS Bank, the U.S. Supreme Court struck down a patent on a sophisticated, computerized method of international currency hedging—holding that such hedging is merely an “abstract idea” to which the inventors failed to add a nebulous “something more.” It is no invention, held the Court, to take a known algorithm and run it on a computer.

The U.S. Patent Office has been issuing one set after another of “guidance,” rejecting one application after another on the basis of Section 101 (eligibility). Many of these patent applications ultimately are issued as patents, but unpredictability, inconsistency and cost are drags on applicants and the system.

Post-grant challenges in the Patent Trial and Appeal Board (PTAB) and the courts are running roughshod over the software patents that do issue and that are successfully commercialized by the patent owners and/or their competitors. If a patent survives, its owner is sometimes bruised badly.

For its part, the European Patent Office (EPO) usually finds software or “computer-implemented inventions” (CII) to be eligible for patent, but, unfortunately, after the EPO surgically removes everything not technical, it finds nothing left that could be considered inventive. That EPO approach ignores the principle of examining the claimed invention as a whole (deeply engrained in 35 USC 103 and U.S. case law), yet it’s probably working better than current U.S. approaches post-Alice.

Patent offices around the world are struggling to define their standards for patent-eligibility of software inventions. In the U.S., claims directed to laws of nature are rejected as ineligible for patent. In Japan, in direct contrast, patents are rejected as ineligible unless they are based on a law of nature.

The U.S. approach to examining software patent applications is broken. All inventions can, at some level, be reduced to an abstract idea. The elusive “something more” is impossible to quantify.

See “Hot Issue in International IP Law: When are Software Inventions Eligible for Patent?” presented by Richard Beem at the College of Engineering of Iowa State University (90 minute HD video published on Vimeo) (Oct. 12, 2016).

Further attempts to define a category of “technical,” e.g., in contrast to a category of “design,” are unlikely to be helpful. Also unlikely to be helpful are the ever-growing lists of “this is eligible, and that is not.” Nor would it help to vary by statute the lengths of patent term for various categories of technologies.

What would work? Well, it would be helpful to focus on the statutes as written. Section 101 could be read and applied to look for processes (such as code) that are useful, i.e., applied inventions. The “useful arts” mentioned in the Constitution are applied technologies. (The word “technology” had not been coined in the late 1700s.) Renewed efforts could be refocused on novelty and nonobviousness.

This would be preferable to the current practice of conducting an inquisition to find and destroy the notorious “abstract idea” unless it can be saved by a successful search for the holy grail of “something more.”

If the U.S. courts cannot save us from the effects of Alice, a legislative solution might become necessary. Such need should come to a head within a couple of years if the courts and the Patent Office continue to be paralyzed. Further damage should be avoided. It would be desirable for such an act to legislatively overrule Alice.

In the meantime, how then shall we live? Inventors—those who solve annoying problems using technology—should continue to prepare, file and prosecute patent applications. After all, the only sure way not to get a patent is to refrain from filing an application for patent. New hope for software patents can be found, for example, in the recent Federal Circuit en banc decision in the Apple slide-to-unlock case (which is driven mainly by software though applied in a physical device). Similarly, Enfish, DDR, and McRo, and even the USPTO guidance, are encouraging for software patentees and applicants. Amdocs (Nov. 1, 2016).

If you drop out of the software patent arms race now, you may have a big hole in your IP protection within a couple of years. Patent life is long, at 20 years from filing. When the worm turns, make sure you have applications ready to pursue the upward potential for software patents.

Here’s a further practical tip: Don’t put all your eggs in the patent basket. Long before software patents became cool, Microsoft and others protected their code primarily by agreements. Copyright registrations can be valuable these days. And don’t forget to register each new version.