Patently Absurd

There's been an ongoing issue with software patents. Software is, after all, nothing more than a set of instructions - a procedure for a machine. Human "calculators" can carry out the same set of instructions, albeit inefficiently and prone to error. If software can be patented, then can you also patent the instructions for fixing a leaky faucet? Shouldn't, by the same logic, mathematical proofs be allowed to be patented? What about the equation that generates a novel fractal image? Arstechnica tells us that the US Patent and Trademark Office has been wrestling with this problem and, after a series of missteps, may be finding their voice.

In an article last week at Patently-O, law professor John Duffy warns that the Patent Office has staked out positions that, if accepted by the courts, would amount to the de facto abolition of software patents. He's right that the Patent Office has become increasingly hostile to software patents in the last couple of years. However, it's far from clear that the end of software patents is imminent. And Duffy is dead wrong to suggest that fewer software patents would be bad for innovation.

The problem is so bad that even woo has been patented. Take, for instance, patent no. 7,357,638 - a computer program for generating astrological horoscopes. The abstract reads

The invention provides a zodiac sign horoscope mapping device, locating the coordinate positions of the twelve zodiac signs, in a manner accurately reflecting the precession of the equinoxes. The device divides the ecliptic coordinate sphere into twelve uniform longitudinal arcs, aligning an origin by a zodiac sign to degree, and thereby locating the twelve zodiac signs along the sphere. A dependent device divides the ecliptic coordinate sphere into twelve non-uniform arcs, each arc isomorphic to the longitudinal span of one of the twelve traditional zodiac constellations, and thereby locating each zodiac sign as isomorphic to its named constellation. The invention provides an astrological horoscope device, which provides the calendar dates of the year for each zodiac Sun sign. The invention provides a horoscope generating method, which can comprise a horoscope computer program's functional specification. The invention provides a horoscope generating computer program.

The patent points out the absurdity of horoscopes found in newspapers, in which the population is divided into twelve groups based on their astrological sign.

Common sense states that, to provide each twelfth of the human population with the exact same horoscope, it can not accurately predict or render a unique and suitable horoscope for each and every individual.

So how does this program "render a unique and suitable horoscope for each and every individual"? The program divides them into a larger number of groups!

Personalized astrology is based on the unique natal data of the individual, comprising the individual's birth date (day, month and year), birth place, and birth time. These are the input requirements for generating an individual's unique astrological data (the exact locations of the ten "planets" of astrology, which comprise the Sun, the eight non-Earth planets of the solar system, and the Earth's Moon), from which charts, tables and aspect grids for the individual can be accurately manufactured. Hence, a daily horoscope for each person should also be generated along these lines, using each person's unique birth input data.

Supposedly common sense has been restored, but I haven't quite figured out where. Close inspection of the patent shows that, even though time of birth can be input, it is optional and not used. Besides, the program doesn't look at the time the horoscope is output - just the date. So "birth time" is irrelevant.

But are the results individualized and unique? NO! The program can call upon a database of "interpretations" given the aspect of orbs (i.e. 'planets' in the parlance of woo). The particular interpretation it provides is entirely random. While the results may be unique (as determined by the size of the database), they are certainly not individualized.

So why did the USPTO grant this patent? Who knows. This one is tame compared to the universe of weird patents, but it serves to illustrate that woo encroaches everywhere, even in our beleaguered patent offices.

For more great info on this topic, see last year's post at the Coding Horror blog.