Almost everyone agrees that the current patent system isn't working well, but the question is, what do we do about it?
I received an email from Craig A. James, a software architect, who wishes to express an alternative view on patents. Because he believes it is unrealistic to ask that there be no software patents, he suggests alternatives. His reform proposals make some sense if you posit, as he does, that there are inevitably going to be software patents, so I agreed to publish his article, even though I personally hold the view that software patents are not a good thing.*
Craig specializes in software design and architecture for scientific systems. His best-known project was a special-purpose database specific to chemistry that revolutionized the cheminformatics industry. This link takes you to the meat of the matter. He has developed software that received a patent, U.S. Patent #4837434, in mass spectroscopy for high-performance detection of chemical compounds, and normally he is writing articles with titles like
"CHORTLES: A Method for Representing Oligomeric and Template-Based Mixtures,"
but he sees that the current system needs reform.
I asked him what his patent does, and here is his reply:
"The patent is for an algorithm to collect data in a quadrupole mass spectrometer two to ten times faster than was previously known, without modifying the hardware. (That's 'quadrupole', not 'quadruple'.) Since mass spectrometers start at about $50,000 for the 'cheapies' and go way up from there, scientists value the speed and sensitivity of the instruments highly. This algorithm gave them huge improvements in both with no increase in cost."
If his sensible suggestions for patent reform were applied, the worst abuses in the current system would be ameliorated, but ideally, no patents on software would be, in my view, the better choice.
Concern about software patents is growing, particularly in the wake of the Sun-Kodak settlement. You might find this thoughtful Information Week article of interest. It says that the settlement covers not only Sun but anyone who licenses from them. HP, IBM and Microsoft already have licenses from Kodak:
"Just how far the settlement will protect all Java users isn't clear. Large companies--heavy users of Java for internal applications--typically don't seek a license from Sun. Also, Java is a hidden component in an increasing number of products and services, from cell phones to truck-scheduling systems. Deciding where a given company makes commercial use of Java might depend on how aggressive Kodak plans to be going forward. Kodak declined to comment on the possibility of more Java-infringement lawsuits. . . .
"Some point to the U.S. Patent & Trademark Office as a big part of the problem. The Patent Office's process of issuing software patents has all the virtues of an 18-wheeler with no brakes, says John Rymer, an analyst with Forrester Research. While the government agency is charged with protecting innovation through patents, it's issuing patents on software code that applicants claim are unique software processes without much basis in fact. 'Process patents are fairly easy to get,' Rymer says. 'There's no rigorous standards for issuing them. The result is you have conflicting and overlapping patents that aren't worth much.'"
Since there seems to be consensus that the patent process is broken, the question is, now what do we do about it? Craig has some ideas he would like us to consider.
* [Note that I altered the wording in the article to make it clear that I have not changed my views on software patents, after receiving some email asking me about it. I never intended that publishing this article be viewed as a change in my position. That is why the title is "An Alternative View", meaning alternative to mine. I hope the clarification makes it crystal clear where I stand, and I'm sorry if I gave you a fright. It isn't always possible to predict how your words will sound to others, and in this case in particular, I didn't write clearly at all.]
Patents - An Alternative View,
by Craig A. James
Groklaw and other FOSS sites have recently advocated essentially abandoning patent protection for software, either by getting rid of software patents altogether, or by exempting FOSS from patent enforcement. This seems like a case of throwing the baby out with the bath water.
Software patents are being abused, but patents are important and useful, and with sensible reform to patent laws, patents could encourage innovation, which is the fundamental purpose of patent law.
The problem with current patent laws are many, but to list a few:
- Patents are granted frivolously.
- Litigation is heavily weighted in favor of large corporations; in fact, individuals and small companies simply can't litigate.
- It is almost impossible for an innovator to know if there is a patent already covering his/her innovation.
- Patents can be used to block use of an innovation rather than encourage its use.
- Holders of "submarine" patents can deliberately lurk, predator-like, waiting for honest, unsuspecting innovators to invest huge amounts of time in already-patented technology.
Straightforward and fair changes to patent law would completely change the landscape of patent litigation and innovation.
1. Require publication and public review of all software patent applications prior to issuance, regarding the key elements of a patent -- prior art, innovation, "non-obvious" technology, and that it's an actual invention.
2. Change the law, so that individuals and small businesses can investigate and rectify actual infringements with modest legal fees, and so that wealthy corporations can't force a settlement simply because their victims can't afford to litigate.
3. Shift the burden of discovery of patent infringement to the patent holder. A patent holder must actively monitor products and technology, or forfeit his/her patent rights.
4. Prohibit damages in arrears. If a patent holder discovers infringement, royalties could only be collected once the infringer has been notified and given reasonable opportunity to remove the infringing code, or enter into a royalty agreement with the patent holder. The exception would be when it can be shown that the infringer must have known or did know about the patent.
5. Require that a patent holder use the technology or lose the patent. A patent could not be simply held indefinitely. This would put an end to both "submarine" patents, and to "patent holding companies", both of which stifle innovation and impede commerce. It would also end the practice of obtaining a patent to block technology -- patents could only be used to develop and create products. Patents could not be used as a speculation commodity, and innovators would be free of unknown and unknowable legal traps.
6. Shorten the life of a software patent compared to traditional patents, to reflect the nature of the business itself. Seventeen years (now twenty) may have been appropriate for the color TV tube, but it's an absurdly long time for a software innovation.
Some of these would be tricky, and surely require more careful definitions. For example, what constitutes "use it or lose it"? Guidelines would help, for example three years might be a "grace period" where the presumption of a valid patent is with the patent holder, but after three years with no product on the market, the burden of proof of ongoing activity would shift to the patent holder.
Patents have historically served an important need. Huge investments in research are often required, whether in semiconductors (the transistor), electronics (the color-TV tube) the drug industry, and many, many other industries. Including software -- I myself received a software patent for an invention that required months of research and years of development. Once the product was on the market, anyone could have copied it.
Ironically, I am also the victim of one of the most notorius software patents: The infamous "XOR Cursor" patent, #4,197,590, filed in 1978 and granted in 1980. Way back in 1976, while a student at UC Davis, I built a computer terminal for NASA that used an XOR to move the cursor around the screen. The work was published in an obscure NASA journal. Only recently did I learn that Cadtrak has collected large sums of money and successfully defended patent #4,197,590 against a number of claims, on something I invented as a sophomore computer-engineering student. Talk about "obvious to anyone versed in the art." Had our work for NASA been more widely published, or if I'd worked in a job where I might have run into the Cadtrak controversy, Cadtrak would probably have lost the patent. Instead, I only found out about the XOR patent last year, after it had expired.
There is no doubt that software patents are being abused.
But to throw the proverbial baby out with the bathwater is to ignore the critical role patents have played in some of the greatest inventions in history. Take one example: Zantac, sold by GSK (formerly Glaxo). Zantac was said to be the single most lucrative product in the history of commerce in the world (it may have been passed by ibiblio by now...) Before its invention, stomach-removal surgery was the most common surgery in the United States, because bleeding ulcers were often uncurable and fatal. Now surgeons hardly remember how to perform the operation - it's only done for morbid obesity. Zantac is a virtual miracle drug, vastly improving millions of peoples' lives. And we're not talking about headaches here. We're talking life or death.
Zantac would not have been developed without patent protection. The cost to bring a drug to market is measured in the hundreds of millions to billions of dollars.
Unfortunately, the drug industry is abusing the patent process. For example, they often file ten bogus patents for every bona-fide patent, in an attempt to obfuscate what they're really working on. They file blocking patents to prevent new drugs that would compete with their lucrative products. They've been accused of using their monopolistic position for a particular disease to reap huge profits, to the detriment of society and sick individuals.
A software invention doesn't require the same massive investment as a drug, but it's not free, either. A true innovation, one that's worth patenting, should come from hard work, insight and genius. And the inventor should be able to reap the benefits of his or her invention, by receiving some period of exclusive use.
While I sympathize with those frustrated by the abuse of software patents, I believe it is naive to think any legislative body would vote to eliminate software patents. Reform requires realistic goals and sensible proposals. If we want to be ignored as the left-wing fringe, then we can keep on advocating an end to software patents. But if we want to be taken seriously, we should propose and lobby hard for reasonable changes to patent laws.
Biography: After dropping out of college at the age of 17, four years pounding nails in the hot sun and teaching sailing to Silicon Valley executives convinced me college wasn't so bad. I'm a graduate of UC Davis (BSEE/CS, 1978), and Stanford (MS Computer Science, 1985). My professional career has been devoted to scientific software. I am currently self employed, consulting on cheminformatics and molecular-modeling software, and I have partnered with a colleague in the UK to work on chemical-registration systems that store detailed information about the millions of chemicals that a pharmaceutical company must track.
I live near San Diego, California with my wife and teenage children, and enjoy music, juggling and bicycling. My wife and I also are semi-professional screenwriters.
This article is released into the public domain by the author.