Well, what do I know about software?
I am the sole author of a ca. 10 megabyte software adventure program called Infinity One: The Secret of the First Disc (version one), which was submitted to Microsoft in 1994 for a finding that it was Windows Compatible, which it was, resulting in a 1994 license agreement with Microsoft (May 27, 1994) for the use of the Microsoft Windows Compatible Logo on the product packaging.
Accordingly, I know something about software, having programmed that software from scratch, originally for Atari, and then converting it to Windows. I did it all as a beginner, without ever consulting anyone on programming "methods" or "algorithms", having written the program mainly to "learn how".
Did I violate any software patents? Who knows? and who cares!
Software programming is all just math and logical reasoning.
That is the software programming reality.
But it is certainly not the current "legal" reality relating to software and patents.
The U.S Supreme Court has now (December, 2013) put a software patent case on its docket by granting certiorari in Alice Corporation Pty. Ltd. v. CLS Bank International, CLS Bank Int'l v. Alice Corp. Pty. Ltd., 717 F.3d 1269, 106 U.S.P.Q.2d 1696, 2013 ILRC 1851 (Fed. Cir. 2013) [2013 BL 124940]
The writ of certiorari raises the following constitutional question:
"QUESTION PRESENTEDThe case is being appealed from a Federal Circuit Court of Appeals decision issued by a bevy of allegedly technically-trained judges (how many of them have ever written ANY software?), who in this case were able to reach a majority agreement only on the Per Curiam Judgment, i.e. "the result", but not on the underlying legal standards applicable to software patents.
Whether claims to computer-implemented inventions — including claims to systems and machines, processes, and items of manufacture — are directed to patent-eligible subject matter within the meaning of 35 U.S.C. § 101 as interpreted by this Court?"
The case involves the question of a computer implementation of a method for exchanging financial instruments, which an initial panel of judges of the Federal Circuit found patentable, but which the Federal Circuit sitting en banc found unpatentable, finding the method to be merely the computer implementation of "an abstract idea".
The patent system has been broken for years, and up to now the Supremes have been content to issue patchwork decisions that leave many of the major questions raised by software patents up in the air.
Whether the Justices of this particular U.S. Supreme Court can do any better in coming up with a sensible standard in this case is questionable.
We therefore reiterate our stand on this issue -- a position that we think is the only sane patent standard possible to avoid the kind of chaos in which patent law currently finds itself.
If it is deemed that software can be patented (frankly, we would not permit it), then only a specific complete software program should be able to obtain a patent, NOTHING MORE.
A good example here would be a software program that does your taxes.
If you were to write a program called John's Tax Program, there is really no reason not to issue a patent for THAT program in its entirety as a particular solution or "invention", including as it does not only the actual software programming, but also the design of the user interface, etc. i.e. no one should be able to duplicate that software program exactly and sell it under that or another name. YOUR product should be protected. OK. No problem.
But if another person writes something similar called William's Tax Program, it should not be possible for John to keep William from patenting or selling his alternative tax calculation program, even though they both might use the same method or methods for tax calculation, because such software methods are mathematical by nature. They involve NO REAL INVENTION, but merely apply mathematical principles to implement abstract ideas that prevail throughout the legal, business, financial and accounting world.
Indeed, given the laws of mathematics and the demands of calculation generally, there are only so many options available, and at the simplest level, there are as good as no options at all: e.g. "if N=1 and P=2 then N+P=3, go to 4". Mathematical solutions are not inventions, computer or no computer, but are merely creative applications of "tools" that are given.
One could compare them to food recipes.
Creative? Absolutely.
Patentable? No.
People claiming exclusive, monopoly patents for any kind of mathematically based methods implemented in electronic instructions -- software -- are thus trying to exclude legitimate competition from using the "tools" of the trade.
EVERY software patent increasingly takes certain tools away from other software programmers, and that is why software patents are so insidious.
In seeking such patents, patent applicants and holders are not protecting THEIR invention -- rather, they are trying to obtain unconscionable monopolies to keep others from using essential abstract ideas as tools that inhere in mathematics -- tools that should be available to all. The fact that a given mathematical implementation is, for the time being, unique -- does not make it patentable, Diamond v. Diehr notwithstanding (that was a stupid decision).
The same reasoning also applies to software implementations such as the infamous "bounce back" patent of Apple -- essentially stolen from the Atari "Pong" game -- and issued absurdly by incompetents at the USPTO.
In the case of the "bounce back patent", its regrettable grant by the USPTO and its upholding by uninformed courts reduces for other software programmers the ability to implement elementary software programming code (we paraphrase) -- "start at A, go to B, if you get to C, go back to A".
The U.S. Constitution protects YOUR invented product. It says NOTHING about keeping others from producing similar, competing products, which is the norm in the commercial world.
YOU should be able to build and sell YOUR mousetrap. Others should be able to build and sell THEIRS.
YOU should be able to build YOUR mailbox. Others should be able to build and sell THEIRS.
By their very nature -- as in the case of software implementations of abstract ideas -- such products by their very nature MUST be similar, because there are only a few possible ways to catch a mouse and only a few possible ways to build mailboxes, and only a few obvious ways to build handhelds.
To permit patenting of the METHOD, e.g. by "enclosure" of the mouse, or via "a slit insert for letters", whether at your front door or in your electronic mailbox version, is just legally IDIOTIC.
But that is what 99% of patents are today -- abstract ideas implemented on computers or handhelds or other electronic devices, or implemented in media such as the Internet ("one-Click" patents and such nonsense) -- which should NEVER be patentable because NOTHING is invented. Simple abstract ideas are merely being electronically implemented.
Many of the surely well-meaning but erring judges on the Federal Circuit still do not get this and we hope that the Supreme Court does better, but we are not optimistic, given the near-sighted Supreme Court precedent in Diamond v. Diehr and its consequences, which have been catastrophic, creating a "patent industry" far beyond what the U.S. Constituton envisioned and permitting an arguably "decaying" society chained by its ill-conceived monopolistic intellectual property laws, which are not promoting invention and discovery, but rather are monopolizing the fruits of human ingenuity in the hands of the "clever" patented few who know how to use foolish laws to their advantage.
For the story, see:
- Lyle Denniston at Scotusblog in Court to rule on patent rights
- Daniel Fisher at Forbes in Supreme Court Prepares To Wade Into `La Brea Tar Pits' Of Software Patents - Forbes
- Ashby Jones at the Wall Street Journal in Can Software be Patented? Supreme Court to Decide
