Slashdot Mirror
SFLC Tells SCOTUS, "Software Patents Are Unjust"
H4x0r Jim Duggan writes to inform us that the day after Red Hat advised SCOTUS that software should not be patentable, the Software Freedom Law Center filed its amicus brief in the Bilski case. "In this closely-watched case, the Supreme Court will decide whether the Court of Appeals for the Federal Circuit was correct in restricting patentable processes to those 'tied to a particular machine or apparatus,' or which 'transform[s] a particular article into a different state or thing,' a conclusion which if fully implemented could bring to an end the widespread patenting of computer programs. ... This case gives the Supreme Court a chance to reaffirm what its past cases have held for more than a century: that no patent law consistent with the US Constitution can permit the monopolization of abstract ideas." Groklaw is running the usual cogent gloss with the full text of the SFLC's brief.
yeah good luck with this
This is one where I believe that no amount of logic and reasonable precedent will matter. The court will simply not invalidate the basis of an existing industry and it has nothing to do with corruption. No quantity of shine can alter the fundamental nature of this complaint. It won't happen, don't get excited.
I would go further to speculate that the patent system as it is harms technological advancement more than it encourages it in most industries. The patent system was established with the intent to create temporary monopolies for inventors in order to encourage the development and dissemination of that R&D throughout society. The problem is that too often, it's used to destroy competitors. The court costs and inequality of enforcement associated with the system defeat most of its purpose as specified in the US constitution. Software patents are only the tip of a very large iceberg.
Sigs are too short to say anything truly profound so read the above post instead.
Solution: store all the payment and shipping information before hand (90% of responses)
Patent solution: store all the payment and shipping information before hand
Patent is obvious
REJECTED
But it would show that there is valid justification for their existence. Who started up a campaign to end software patents? It wasn't Novell, or Red Hat, or the Open Source Initiative; it was these folks. The FSF is willing to take unpopular hardline pro-freedom positions long before they become politically correct enough for the "respectable" FOSS groups to hop on the bandwagon.
Sort of analogous to how I think there's a good reason for the existence of folks like Dennis Kucinich, Ron Paul, and other people pursuing principled but somewhat inflexible positions. I might not want them as president of the world, but I like them being around.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Being the first time someone has done something doesn't make it non-obvious.
Problem: Nobody has ever jumped off the roof of my apartment and survived.
Solution: use a parachute.
Is this patentable? No, because people have jumped off high places before, and that's what their answer was. A patent's purpose is to promote the useful arts and sciences. This is their purpose under the constitution. The problem that must be non-obviously solved is therefor the technical one, not the business desire. His statement of the problem was correct- how do you perform purchases without a postpay or confirmation step. The solution was to save that information server side, which is what 90% of programmers would have told you in under 30 seconds of thought. It is therefor an obvious solution.
I still have more fans than freaks. WTF is wrong with you people?
There is a faint (very faint) glimmer of hope that this will be the beginning of the end of stupid software patents. I have a bad feeling though, that.....
This is not the end. Its not even the beginning of the end. But it is perhaps end of the beginning. -Sir Winston Leonard Spencer Churchill
This isn't so much a business methods case as it is a case of Bilski's patent. The CAFC rejected Bilski's patent and installed a test, and some of the patents that will fail that test are what we call software patents.
There's no "business methods" line that the CAFC or SCOTUS have to stay within, and FSF's request is just for upholding of the CAFC's test and a clarification about when a computer (such as is used by Bilski's patented system) is or isn't a "particular machine".
I take your points, but I still see this as by far the biggest chance we're getting this decade to fix this problem.
Please help publicise swpat.org - the software patents wiki
The judicial power shall extend to all cases, in law and equity, arising under this Constitution, the laws of the United States...
Notice how the Supreme Court has jurisdiction over Constitutional questions, in addition to just being able to "interpret laws"?
To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries;
And notice that the scope of patentability is a Constitutional issue?
35 USC 101 allows patenting of processes
US code doesn't allow patenting of business processes unless the US Constitution allows patenting of business processes.
"I assumed blithely that there were no elves out there in the darkness"
So no one in the history of mankind ever had a store account ? Identify yourself and the payment is taken care of discretely and automatically ? Just because it's "on the internet" does not make it patentable. And people were doing it before, and people are still doing it now. One Click is just a marketing phrase describing a widely held practice, one that was widely held before the patent was granted. Why do you think there was such a fuss when the patent was granted ?
And using the fact that no-one did it before as a non-obviousness filter is stupid. Most sensible companies didn't do it because of the extra security concerns involved, not because they didn't think of it. As an e-commerce developer at the time, I didn't do it because I didn't like the idea of my details being held by an invisible third party. I therefore didn't place others in that situation.
A patent was originally defined as legal protection granted to an individual for creating a device that does something unique and useful. A temporary monopoly was granted by the government to allow that inventor who created that unique and original device to recoup the R&D costs associated with creating that device in the first place, noting that R&D for something genuinely original often has dead end roads when trying to get something to work and is generally quite expensive to perform. Once somebody has established a good template for how something can work, many others are willing to copy that device since the hard work has already been accomplished.
The slippery slope from this definition and the one given above (that a patent is about solutions instead of physical devices) is precisely what has led us to software and business method patents. R&D costs, while not completely missing, are orders of magnitude lower and really not a significant problem. Furthermore, previous tests measuring the validity of patents have included an explicit exclusion for mathematical theorems and formulas.
To put it another way, all computer software can be reduced down to a single number. Perhaps a rather large number with thousands, millions, or even billions of digits, but a number none the less. What a patent essentially does is to give people exclusive copying authority over certain classes of numbers and makes it illegal to publish that number without their permission.
Other arguments can be offered here, but the point is that patents don't cover a mere solution or abstract idea. The original intent of the framers of the U.S. Constitution and based upon the previous English Common Law precedent over previous patents... including abuses of patent law in England that the founders of the American Republic were trying to avoid... was intended to be narrowly defined to cover strictly physical devices precisely so patent law wouldn't become perverted to censor speech and political philosophies. A broadly construed patent philosophy can be a tool to pervert other aspects of the constitution including the 1st amendment and other areas of the constitution as well.
Horrible software patents include such things as the LZW algorithm patent that somehow landed in the hands of Unisys... where they asserted the patent to extract royalties for those companies wishing to use the GIF image standard in their products... including web browsers. The argument that payment of royalties to Unisys for the use of this algorithm is somehow going to encourage Unisys to invest into its R&D program seems absurd. Besides, in spite of the millions in royalties that were paid to Unisys for those who did pay the extortion tax here, very little can be said to have come from it other than the employment of a few lawyers who were involved in setting up the royalty schedule and sending out the cease and desist letters. In other words, even in this clear-cut example of a software patent that was granted, the use of the patent was to stifle innovation and progress rather than to encourage further development of software ideas.
Actually, the enforcement of the LZW patent did encourage the development of other methods of graphical image display formats: It created the PNG format that was explicitly established as a way to legally work around the LZW patent so it would never have to be used in the first place. Those involved explicitly were involved in establishing a graphical image standard that would not be encumbered by patents or require royalty payments for its use and application. It also was a general improvement upon the original GIF standard as well, allowing for greater color depths and even improved data compression compared to the GIF images that were originally being protected. That was useful, but it wasn't the patent itself that was proving to be useful in this case... it was merely a roadblock that needed to be overcome and ignored. It has become ignored for the most part as well.
For software, that's really easy: let reality run its course so that the inventor is first to market.
In 1800 that didn't work very well, because of the nature of the inventions. Without inventions, the inventor had to keep a secret, snail mail investors who took a 4 week journey to come see it in the lab, get the money together using the archaic financial systems, get the gold onto the stagecoach with enough guards, then the newly-formed company has to hire a workforce and fabricate the new parts in secrecy, and get them onto the stagecoaches out to market. Then customers see it and word about the awesomeness of the product slowly spread by word of mouth and snail mail.
Now the inventor goes from the invention to a product that uses it, immediately, and customers can pay for it and download it immediately while a firestorm sweeps through Twitter. By the time someone can make a competing product that uses the same idea, the inventor already has his money back.
Ok, you've convinced me: they should have a 4 day monopoly. I would have said 3 days but that's not fair to people who invent things right before a holiday-adjacent weekend.
No, they didn't. They just took a different compressor off the shelf. The only things in the compressor that they "looked long and hard at" was whether they would be allowed to use that algorithm, and how well it performed compared to others. In 1995, compression algorithms superior to LZW were a dime-a-dozen thanks to the compression fad that swept through hacker society 5-7 years earlier, and JL-G's zlib was ready and waiting for them.