Patents Are Evil

In TwoIrreparableMistakesOfTheSoftwareField, WilliamGrosso claims that patents are evil. Is this a commonly held view that goes along with ExtremeProgramming or SoftwarePatterns?

--GarthDickie

I think it is quite orthogonal from ExtremeProgramming and SoftwarePatterns. IMHO it would be a grave mistake to think that people adhering to any one of these must necessarily adhere to or agree with at least one of the others. I'm sure there are plenty of people in both camps who feel that way about patents. And I'm sure there are plenty who don't.

Sounds reasonable to me. Thanks.

Actually, I don't really believe that. I think they're an interesting idea. And I think they can be a very useful device. But I also think that, in a litigious society mostly unconstrained by notions of good faith (especially one where there is as much knowledge floating around as ours has), the whole system seriously needs revamping (something along the lines of "filing a frivolous patent is a felony" is a starting point).

I have a friend who's busily patenting a whole bunch of ideas. She was telling me about them. And I kept saying things like "That's been done before. Here's an example." And she kept saying "Doesn't matter. We're getting these patents to show the VC we have intellectual property, and to make it harder for other people to enter the space. Not because we're doing anything original." -- WilliamGrosso

This points up the real flaw in the patent system: the existing laws aren't being enforced. Or at best they are being enforced by the court system rather than the patent office. It's supposed to be very difficult to get a patent. You have to prove that the idea is useful (which only excludes inventions that can't possibly do anything, like perpetual motion machines), original, and non-obvious. Virtually all software patents that have been issued to date should not have been because they violate one of these two rules. The fact that companies set quotas on the number of patents they expect to produce per year shows that patents are way too easy to get.

Patents are supposed to protect inventors who have truly novel ideas from being ripped off. I think that the software industry is too young to be able to tell the difference between what is novel and what isn't. We are discovering too much too fast to be able to tell the difference between ideas that are novel (in the sense that they aren't obvious) from the ones that are merely new. Lots of new ideas can be generated by applying existing technologies to problems that they haven't been applied to yet. If the solutions that are generated through this process are the same as the solutions someone else would generate through a similar application then the solutions aren't novel enough to warrant patents. If there isn't enough prior art to know whether or not your solution is the same as someone else's would be then there isn't enough evidence to prove that a patent should be awarded and therefore none should be issued. That would still be an economically feasible solution because being first to find an idea in an evolving field of technology affords advantages that are similar to those that patents grant to those who find novel solutions in more stable fields. Patenting an idea that is merely new and not novel can actually slow down or prevent innovation by preventing the idea from being fully explored by competitors who might find truly novel applications for the idea. -- PhilGoodwin

To me, patterns literature seems like a good way of demonstrating prior art, but it is a DoubleEdgedSword sword. It is only a matter of time before someone attempts to patent a pattern. -- MichaelFeathers

This is incorrect:

: Technically, prior art has to be more than one year old before it can prevent a patent from being filed in the United States. This sucks, because it provides a one-year window during which you can be ripped off -- unless you can either (a) get your idea timestamped and sit on it for a year before you publish it, or (b) patent it yourself. But filing a patent costs around $3,000.
: Corporations, which can easily afford this, must file patents as self-defense moves: even if they had no intention of collecting a dime in royalties or stopping anyone from implementing their idea, they'd still have to file the patent, to keep someone else from filing it and halting their operations, or claiming royalties from them. -- EdwardKiser

If you publish something, it immediately stops someone else from patenting it. But, it allows you 12 months to patent it yourself. (This one-year window of opportunity generally doesn't apply for foreign patents, but I have an AmericanCulturalAssumption.)

Ah, that's good then. -- EdwardKiser

I concur with Phil: The crux of the matter is that the PatentOffice? doesn't understand software. I would include a quick coverage of the issues most important to me that relate to patents, but people who didn't agree with 90% of my views on patents wouldn't see them as important issues in any case; there are better forums for such topics. -- DanielKnapp

Patents are supposed to protect inventors who have truly novel ideas... -- uh, according to the UnitedStatesConstitution, the patent system is to "promote advancement in the arts and sciences." On that basis (which is a good basis, although it can argued that patents don't do that), I think finding "new" uses for an invention is worth encouraging. If nothing else, it (InTheory) promotes interchange between various specialties.

Further, the one-year rule regarding prior art is an anomaly of the US patent system. In the civilized (civilised?) world, any prior publication invalidates the patent.

-- BillTrost

Is the one-year rule truly the case? I knew the US patent system was stone cold bonkers, but that would be truly insane.

From the ACM TechNews? service:

http://www.acm.org/technews/current/homepage.html

"Patently Absurd" New York Times Magazine (03/12/00) P. 44; Gleick, James

: "The patent system was originally designed to protect inventors and thereby encourage innovation, but in today's e-commerce economy the system has mutated into a form that more than anything else threatens the very entrepreneurial spirit it was meant to safeguard. To qualify for a patent, an item no longer needs to be a tangible product with form and substance--thoughts and abstractions are now equally eligible. Because of this, large firms have begun taking fundamental practices and business methods that have been in existence for years and incorporating them into technology-based applications, while claiming the repackaged products are innovative items worthy of patents. The U.S. Patent Office issues 10,000 patents every three weeks, and its examiners appear to be operating under the misleading impression that more is better. In fact, software patents have become the most frequently issued patents, and many corporations are wielding such patents as weapons against competitors, hoping to stifle development of rival technologies. As the blanket of patents grows denser, consumers must pay higher prices and individuals genuinely interested in creating something new and unique must surmount more regulatory hurdles. Patent Office officials argue every patent represents a good idea that merits reward, but corporations managed to survive just fine without them during the previous two decades when PCs, the Internet, and related software and hardware were in their infancy. The corporate legal skirmish over patents only truly began in earnest in 1998. That year, a ruling by a Massachusetts federal court that stripped a company of its patent for a data processing method was overturned by the Court of Appeals and the patent reinstated."

  http://www.nytimes.com/library/magazine/home/20000312mag-patents.html

InanePatents

I don't think PatentsAreEvil, just confused. I think the problem with US patents is that it purports to identify the first inventors. Inventions don't really work like that. Multiple inventors can have the same idea at about the same time. The rest of the world awards a patent to the first inventor to file a patent, which translates to me to be a license for that invention, which is fine. A government should be able to award business licenses to businesses. But when a government tries to legally say "these are the people who invented such and such first", that shouldn't be a government's power. Just my 2cents -- MikeHo

Patents were intended to *spread* information, not restrict it. The motivation driving the patent system was a desire to assure an inventor reasonable compensation for publishing information that would otherwise be proprietary. It was thus intended to promote the sharing of information, not to restrict it. That is why the patent application requires explanation of how the proposed device works, that's why the patent has a lifetime, and that's why the government attempts to enforce compensation for patent violations.

At the time of its creation (late 18th century), the duration of a patent was viewed as short enough to be a minor hindrance to the public acceptance of the technology being patented. The goal was to move knowledge from the private into the public domain.

In a similar vein, *ideas* have never been, nor are they now, treated as "patentable". Any idea must be reduced to "practice", and it is the practice that is patentable, not the idea behind it. This partitioning worked well enough until computing systems were discovered. The rub is that "software" bridges the gap, and makes the distinction impossible to draw rigorously. The water is further muddied, in the software domain, by the distinction (if any) between "patent" and "copyright". The patent generally applies to the operation of the code. The copyright generally applies to the source code.

Whether or not PatentsAreEvil, they are what they are, and should be understood, for better or worse, in that context. Let's not distort the discussion by confusing each other about what they do and why.

I would agree with the above if it were true. In practice, patents are applied as licenses to monopoly. A monopoly concentrates income to the holder of the monopoly at the expense of those who do not hold that monopoly. As long as the non-holders are your competition this is good, otherwise this is bad. This is a balance of scales. Software and business process patents, since they are only respected in the US, are good for the individual business but bad for the US as a whole. As long as the rest of the world laughs in the face of "one click" patents, US innovation stagnates. Sorry WTO but no thanks.

Where does the WTO fit in here? (It's the first mention of it on this page).

One of the planks of the WTO is to create a worldwide system of intellectual (read commercial) property law. This is (or was) predicated on respect for existing patents, of whatever quality. Most of the news is about bio-piracy (Monsanto et al) but they are only a small part of the picture. It is a form of capitalist imperialism. Scary stuff. We'll see whether our Euro parasitic class understand the threat or not.

So far, the "Euro parasitic class", as your balanced phrase has it, are doing quite a good job rejecting the patentability of software and business processes, which is fine by me.

Yes but there is a tension here. Heres some primary source from the TRIPS agreement which is already accepted. There are caveats throughout but I think the intention is clear:

TRIPS: AGREEMENT ON TRADE-RELATED ASPECTS OF INTELLECTUAL PROPERTY RIGHTS Part I : General Provisions and Basic Principles:

2. For the purposes of this Agreement, the term "intellectual property" refers to all categories of intellectual property that are the subject of Sections 1 through 7 of Part II.

3. Members shall accord the treatment provided for in this Agreement to the nationals of other Members.

Article 28 Rights Conferred

1. A patent shall confer on its owner the following exclusive rights:

(a) where the subject matter of a patent is a product, to prevent third parties not having the owner's consent from the acts of: making, using, offering for sale, selling, or importing (6) for these purposes that product;

(b) where the subject matter of a patent is a process, to prevent third parties not having the owner's consent from the act of using the process, and from the acts of: using, offering for sale, selling, or importing for these purposes at least the product obtained directly by that process.

--RichardHenderson. (please help me format this right, copy/paste hell :(. ).

Are patents evil as such or are they a GoodIdeaBadlyImplemented? I think they are a GoodIdeaBadlyImplemented. A better patent system is possible. -- EdwardKiser

Patents are evil as such, period. The whole idea to them is the monopolization of practical knowledge and know-how. This is evil.

But far worse than this, patents are evil because they do not work. This deserves repeating so PATENTS DO NOT WORK. Everyone in industry understands this fact. With a very few exceptions, patents are only ever used in a purely defensive manner. You get patents as counter-measures against someone who uses their own patents against you, and you only use your own patents against someone as a negotiating tactic to cross-license thus nullifying the thread of both of your sets of patents.

So what we have here are evil second order effects (inane patents, submarine patents and truly evil patents) on top of a first order non-working system designed to create evil. And as the final straw, this non-working system adds its own overhead to the cost of doing business.

The only exception to the above, that patents simply do not work, is the chemical industry (with the biochemical industry as a subsector). Patents in the chemical industry work (they successfully promote evil) because:

they are unusually powerful (you can cover millions of different chemicals in one patent)
you don't need to know what you're patenting (the chemicals you patent might not even do what you claim)
the field isn't understood sufficiently to engineer around patents (it's generally impossible to create a chemical with known properties in a systematic manner)

-- RichardKulisz

Good arguments can be made that there are InanePatents, but it shouldn't logically follow that PatentsAreEvil. A CeeCorporation?, a common corporate legal structure used in the US, is taxed at a 40-50% rate. The government, therefore, becomes an entrepreneurs largest business partner.

You can either fight them or use their protections. As part of their partnership, the US government offers patent protection. You'd be wise to figure out a way to use it. --MichaelLeach

See http://www.infoanarchy.org/wiki/wiki.pl?Totally_Evil_Patent. This page explains the maximal level of evil that can follow from a single valid patent. It also includes some suggestions for alleviating the possible existence of such patents.

Enough negativity! HowToImprovePatents?

CategoryEvil CategoryLegal