Slashdot Mirror
Federal Patents Judge Thinks Software Patents Are Good
New submitter Drishmung writes "Retired Judge Paul Michel, who served on the Federal Circuit 1988-2010 — the court that opened the floodgates for software patents with a series of permissive decisions during the 1990s — thinks software patents are good. Yes, the patent system is flawed, but that means it should be fixed. Ars Technica have a thoughtful interview with him. Ars' take: 'If you care most about promoting innovation, offering carve-outs from the patent system to certain industries and technologies looks like a pragmatic solution to a serious problem. If you're emotionally invested in the success of patent law as such, then allowing certain industries to opt out looks like an admission of failure and a horrible hack.'"
Drug enforcement agents think the war on drugs is a really good thing.
God spoke to me
"If you're emotionally invested in the success of patent law as such" - that's the problem. You should never be emotionally invested in a cedrtain law. You may be emotionally invested in a goal and thus support a law which you think helps with that goal (and revise that support if it turns out that the law doesn't help with that goal). However as soon as you are emotionally invested with the law as such, you are not any more objective about it.
If you care most about promoting innovation, offering carve-outs from the patent system to certain industries and technologies looks like a pragmatic solution to a serious problem. If you're emotionally invested in the success of patent law as such, then allowing certain industries to opt out looks like an admission of failure and a horrible hack.'"
Isn't that the actual, official, reason for having patent laws and protections in the first place?
Surely being 'emotionally invested in the success of patent law' would require you to want it to achieve what it was meant to achieve?
I think, given the number of lawyers involved and the kind of income they can make from corporates, that for "emotionally invested" read "benefiting financially". There are a few judges who, once they have a permanent appointment, suddenly start telling litigants to grow up and keep the courts out of it, but the majority are looking over their shoulders at their former colleagues and their children.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
Is this a suprise to anyone? Central planners always have an excuse for their failures and always insist they just need some reforms and tweaks to get it right. They insist the problem isn't that central planning cannot work it is just some little switch or dial needs adjusting. The fact is Central planning can never work. Free people following having the liberty to do what they want with their time and property will always work better. It won't always be successful but that is the point. The failures will simply run out of their own money. The central planners get to take everyone's money to keep funding their failures.
I love Jesus, except for his foreign policy.
Bugs are good!
-- software engineer paid for maintenance
Software patents have two main problems.
The biggest problem is, generally they aren't novel enough. Too many can be conceived by a general practioner of the art. And claiming XYZ can now be done on the internet, or on an IPad, or on 'fancy new device' doesn't make it novel enough.
The other problem is ideas can't be patented. Yet that seems to be what most patents are. They won't show the code, so how do you know if you are infringing on the patent? There are multiple ways to solve a problem. Just because I got to the same end point doesn't mean I infringed on the patent.
It's a balance between damage and benefit and the balance is firmly in the 'damage' side currently. Throwing more industries into that mess does not a fix make.
The benefit was to award exclusives in areas where invention cost was high and time to market long. Thus it enables the invention. This is not true of software, where you don't need to build a factory to make the product you just invented, and thus time from invention to market is too small.
The damage is 1) complex systems can be blocked by individual patents on individual tiny parts, e.g. blocking tablet sales in Germany. 2) none inventors can steal the profits from the inventors using wrongly issued patents, we've seen an awful lot of those. 3) Some markets are dominated by trade secrets making wrongly issued patents the norm rather than the exception. Software being an example of that. 4) If an invention requires extensive investment, it is easier for people to land mine around it.
He really has to live in the real world here. You can't pretend the conditions for one thing are the same as another just because it lets you use one set or rules to govern both. Water is not steam, and you can't use a bucket to carry steam just because you happen to like buckets. But you can get awfully burned trying to carry steam in a bucket.
Lets take a look, machine translation is done by companies like Google and Word-lens. They are the ones inventing and making products. However if you look at the patents, this is typical:
2009: "U.S. Patent # 7,610,187 - Lingual translation of syndicated content feeds ", a typical IBM patent.
Now the patent isn't enabling the invention here, IBM has just done the typical thing, looked at what people are doing and patented around it. This isn't to create things of value because IBM don't make translation software, they make patents. It's about using the weakness of the patent system to make money from companies that *are* inventing things. It adds an overhead to those companies, an extra cost in their R&D budget.
If the cost of enforcing the patent equals or exceeds the recoverable benefit, you have just conceded the fact that the benefit no longer carries more than marginal economic value to the alleged beneficiary. The best that could be said here is that it distracts a competitor. The worst (and probably closer to reality) case scenario is that the pursuit of marginally valuable patents creates a perverse incentive that distracts a company from more useful economic activities.
It's really hard to take seriously someone who says they're all about facts and figures, but then jettisons economics because the economic aspects of his preferred system are abysmal. There will come a day, at the rate we are going, where the rule of law will be formally dead in the US similar to how it is in Russia because the legal profession (and judges and prosecutors in particular) have made the cost of participation so high from various factors ranging from failing to sanction frivolous lawsuits and criminal charges, to allowing blatant corruption. As it currently stands, it's on life support.
But their duration should be shortened to 2 years to account for time to manufacture. The patent system was developed for physical devices which historically could take years to manufacture. Software is out the door in 6 months.
I don't see anything to justify "software patents" in the first place, and actually, patent law forbids it. Everyones.
Just because some idiot lawyers redefined "software" as not being "math", because they couldn't grasp the math isn't enough reason to not ditch illegally granted patents.
There is nothing to "opt out"; the situation with these illegally granted patent just needs to be resolved.
"The more prohibitions there are, The poorer the people will be" -- Lao Tse
For a judge who served on the court that "opened the floodgates for software patents," this guy knows remarkably little about software. He (self-admitting) doesn't even know anything about the software industry or its current disregard for patents. How can we take any of his comments seriously? The interviewers did ask some thoughtful questions, but I wish the interviewers would have mentioned that the current approach in the industry uses terms like Mutually Assured Destruction.
"If software is less dependent on patents, fine then. Let software use patents less as they choose," Michel said.
"Yeah, if countries didn't like the negative impacts of nuclear bombs, they shouldn't have produced so many during and after World War II." The problem is it only takes one Nazi Germany to scare a country into producing such a thing before the other, and one Stalinist Soviet Union to scare them into continuing to produce them. That's the problem with the software patents - everyone has to arm themselves against everyone else who isn't looking out for the good of the software industry. History may judge nuclear weapons as a great human mistake, and I suspect software patents also. Besides, software patents were NOT allowed to be patented before the Federal Circuit. It's not like that situation is without precedent.
You appear to have responded to my post without reading it. I suggested that it is the lawyers who benefit financially, and I didn't suggest anywhere that "law firms" are incorporated.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
The software industry is indeed different.
There is a reason why people say an Internet technology year is like 7 years in another industry.
If patents are intended to protect inventors while commercializing their inventions, then current patent policy is a grevious failure and harm to inventors, and must be scrapped or greatly reformed.
Following are some key points:
- Huge number of obvious patents
- Large companies forced to buy tens of billions of dollars in patents as insurance against mutually assured destruction. This warfare means a single inexperienced jury can greatly impact trillion dollar multinational business strategies and a great segment of the global population, while making further invention exponentially risky.
- Smaller companies are unable to defend themselves in this warfare. They are periodically destroyed by large companies wielding patent weapons.
- U.S. inventors are put at a disadvantage by patents / legislation due to the immediate nature of software / Internet / speed of development overseas
- Mathematical nature of software languages and code
- Cooperative nature of software repositories, libraries, class hierarchies and APIs
- Revolution of the software industry, as a simultaneously cottage industry and international in nature
- The nature of software and the Internet means code can be transparently executed on servers in other jurisdictions.
- Legislation is both hidebound, slow and naive while having a permanent and disproportionately large impact on the software industry. A quickly reacting and quickly editable legal board is probably necessary if laws on software are to continue realistically.
As other industries become more and more dependent on software, they too will become more endangered by software patents, and by Internet-style information technology based disruptions. As it currently stands, individuals are at a severe
disadvantage in patent wars and on a global stage due to the USPTO's spamming of software patents with a total lack of responsibility for the massive losses in time and money required to justly determine the patents' validity after the development of critical infrastructure using them.
And who is going to overhaul the system and eliminate the garbage? I don't know about the USA, but in this country whenever a "system" needs "overhauling", all of a sudden the Government seems to employ a lot of lawyers on long contracts. Whereas reverting to a state in which neither algorithms nor their implementation in software could be patented would have the reverse effect.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
IANAL but I am a programmer. Judge Paul Michel fails to notice that he is not a software developer, and lacks perspective on the software industry as a whole. Here are four reasons to reject software patents:
Modern computers are general purpose machines - hence BASIC (Beginners ALL-PURPOSE Symbolic Instructional code). All programs are therefore written within the specifications of the hardware designer. This makes ALL software predictable by those versed in programming and not patentable in the first place.
Since all software runs on hardware that only understands the values of 0 and 1, it is all reducible to math. Anyone who has taken a digital logic class can attest to this. What you see on the screen is a representation of that math. Dump the contents of the RAM in binary if you want to prove it to yourself. Math is discovered, and therefore not patentable.
Software patents typically contain no code. The "Inventor" fails to disclose their invention, which should justify the patent being thrown out for lack of documentation. The patent holders, which are increasingly attorneys, are typically unable to actually implement their own patents. This practice discourages innovation.
Software patents typically make no sense to programmers. If a programmer can not understand the patent, then it does not describe a program. On that basis it should be thrown out.
We programmers are sick of being harassed by patent attorneys. They are leeches on our business, and have served to stifle innovation in the industry. It is time to fight back. We should earn triple damages if we successfully defend a suit based on bogus patent claims. For instance, Google should be paid $3 billion by Oracle ($1 billion *3) if they win their case. That would put the trolls back under their bridges.
http://www.ted.com/talks/drew_curtis_how_i_beat_a_patent_troll.html
http://www.thisamericanlife.org/radio-archives/episode/441/when-patents-attack
I'm seeing the usual anti-patent rants here, and many of the usual mistakes. Some corrections:
There are some thing universally accepted as not being patent-able: Natural Law, Physical Phenomenon, Abstract Ideas and from these we also get Mathematical Algorithms. These are the components of Software. There are natural laws and physical phenomena that apply to the creation and use of abstractions, and in this case the abstraction is often perceived in terms of mathematical algorithms. http://abstractionphysics.net/pmwiki/index.php
In order to ban software patents, one must first define software patents. I challenge anyone in favor of banning software patents to come up with a definition of the term that is neither under- nor overinclusive, can be easily and unambiguously applied, and cannot easily be gamed. Here are some example technologies to think about as you develop your definition:
1. A machine that cures rubber by heating and cooling it, controlled by hand.
2. A machine that cures rubber by heating and cooling it, controlled by a computer program using a new, nonobvious, and useful application of a mathematical equation.
3. A new and nonobvious kind of rubber curing machine that uses such a complex curing process that only a computer could control it, resulting in a significantly superior product.
4. A computer program for controlling the rubber curing process using a new, nonobvious, and useful algorithm that could only be carried out by a computer, resulting in a significantly superior product.
5. A computer program for modeling the rubber curing process using new, nonobvious, and useful applications of mathematical equations.
If you want to ban software patents, where do you draw the line?