Hacker News Reader

Comment by fnid2 | Hacker News Reader

Every time the patent debate comes up, someone links to this patent as the summit of absurdity. Why not link to this one: http://www.freepatentsonline.com/6368227.pdf

It's just as absurd as lots of software patents -- that doesn't mean all software patents are absurd. Some are genuinely novel, unobvious, and valuable.

I still don't understand why someone could patent a physical machine that computes instructions, but not software that does the same thing.

It seems like we are throwing the baby out with the bath water. I know patents cause trouble, but what about the case where a patent helps an entrepreneur get the value out of an invention when big corporations steal it, market it, and make money without giving the inventor a dime?

There's good and there's bad, but it seems like around here there's only mud slung at software patents. Some inventors of software algorithms are every bit as clever as inventors of anything else that may get patented. Why shouldn't they be treated the same under the law?

Equal rights under the law. That's the way it should be.

View on HN

Do you think mathematicians should be able to patent formulas too? Some are novel, unobvious and valuable.

The argument, I believe, is that formulas aren't invented, they're discovered. That's why I think it's totally, unequivocally absurd that you can patent genomes. Who invented the human (or any other naturally occurring life form's) genome?

Just because something's not obvious or novel doesn't mean you can patent it. You can't patent music or poetry, no matter how obvious, novel or valuable it is.

At least this example is an implementation detail. That seems more reasonable to me than, for example, UI patents. If someone patents a way to make steel, you are free to make steel any other way you can figure out. The same seems to apply with the implementation detail patents, but with the "multitouch" patent Apple's suing over, it seems that the question of how HTC has implemented the functionality doesn't matter because the patent covers the user interaction. And that makes no sense to me. If it's so genuinely difficult to make, then they can patent the implementation. But patents were meant to cover "the process P which accomplishes B from A", not "any means of accomplishing B starting with A even if totally unrelated to P."

I've heard the discovery vs. invention argument, and there is something too it, but personally, I'm still not clear on how to distinguish between invention and discovery. Wasn't steel a discovery, elements, or quantum theory?

Music and poetry are covered by copyright, currently. But the same problem exists there and Larry Lessig has done a lot to protect culture in an environment of copyright litigation.

Re: Mathematics and formulas. A lot of the discoveries or inventions in mathematics are patented. Three dimensional protein folding is a huge industry in pharmaceuticals and math is almost 100% of that knowledge. It's applied math, but nonetheless, still patentable mathematics and a lot of it too.

I think one differentiator here is that the multi-touch and the one click amazon patent as another example are something that seems obvious to try to achieve given the current environment. There are other ways as well, like a space elevator or gloves to help protect astronauts in space.

Is math a fact? I don't know if facts can be patented or copyrighted, though collections of facts surely can.

I think your assertion of patenting a way to do something vs. just patenting doing something is really the core of the problem.

Perhaps you can provide a counter example that shows the usefulness of software patents.

I personally think they are absurd, as demonstrated by the OP, but am open to changing my thinking if proven wrong.

The internet has leveled the playing field for everyone. But one of the few ways that big corporations can take advantage of their size against smaller, more innovative entities is by throwing money into a legal battle which the smaller entity can't afford to fight.

Equal rights under the law. But currently, the more money you have, the equaler you get.

Okay, then the problem is not software patents, or patents in general, but that they unfairly tip the scales in favor of the big corporations.

How do we fix that problem? That problem exists outside the world of patents and software. Monsanto patents seeds and those are just strings of DNA which are essentially just four characters repeated in a seemingly random order. And yet farmers lose their farms over those strings of four letters. Not because they were even choosing to use the seeds, corn is just pollinated by the air and some of that code got into their crop as it blew in from the neighbors' field -- a farmer who did use Monsanto's DNA. It's absurd, I know.

One case on the not absurd side of the spectrum, is, as it turns out, the original case that ruled software could be patented in the first place. It was by a guy who had to fight, iirc, all the way to the supreme court to protect his right to patent his software invention. I'll try to find a link, but is a true story indeed.

> Okay, then the problem is not software patents, or patents in general, but that they unfairly tip the scales in favor of the big corporations.

Agreed. Legal issues (patents included) tend to favor big corporations because you're more likely to win a case if you spend more money on it. I don't know about all the factors contributing to this, but I think a big one is that law (in the U.S., anyway) is difficult to understand for laypeople, and so you need lawyers to defend you, and the lawyers that cost more have higher success rates.

More likely, lawyers with higher success rates cost more.

Sure, but it's beside my point, which is that the small g(uy|al) would be better off if lawyers were either obsoleted by a ``Law 2.0'' or made into a purely public resource.

I don't think it is fair to criticize patents on DNA and defend patents on software with the argument that DNA is just a combination of four characters - you could say the same thing about any piece of encoded information, software most notably in binary.

Dude, do not argue that software patents are ok because genome is patentable. That's going in the wrong direction.

Well, perhaps to add more content here w.r.t. the Monsanto case, I mean, the court should have been able to determine if the wind just blew the pollen in to pollinate the plants. IANAL, perhaps an IP attorney who is knows what they are talking about can explain why that case worked out the way it does.

Anyway, it's pretty unfortunate how that one example worked out, but the point I failed to make, was that the engineers who invented that wonderful corn that is immune to disease and predation and grows happily almost anywhere in the world and brings great value to the future of the world are no more or less clever than someone who assembles for loops, encryption functions, and database calls into something just as valuable to the future of computing. In a way, those are even more letters and numbers and perhaps physical systems too, just represented in code, and therefore should be even more patentable than just 4 letters repeating in a sequence.

Perhaps it is another type of discrimination I am addressing. A discrimination against a new kind of intelligence. A discrimination against machine intelligence. The artificial brains and neural nets and perhaps years of training are somehow different than a solar panel that can follow the sun. I don't see a difference and I don't see any reason the law should make such a distinction. That's something that may come up in the future, but it's not really a problem now. Somewhere along the way though, We will probably start talking about giving rights to machines who have claimed their identities. We let corporations have all the rights of humans, why not individual sentient robots? But again I digress.

My argument is this: If we are going to do away with software patents, we should do away with patents all together. I don't know what that world will be like and if that is the solution to these kinds of patent problems, not just for computer science, but for all fields, then I say let's do it, but don't single out software patents and axe just those.

Once the patent to the DNA was granted, the suit against the farmer in the neighboring field was logically necessary. It's the way patents work: you either sue to protect your monopoly whenever you're aware it's being infringed or you lose it. Nations suffering from starvation have refused patent encumbered food, because it would have meant that their farmers, who would plant some of that grain in the same way they have for thousands of years, would no longer have access to European markets, and would be liable for patent infringement. http://www.independent.co.uk/environment/us-policy-on-aid-is...

While the outcome of the Monsanto case was beyond absurd, that wasn't what I was referring to wrt genome patents. If you invent a lifeform, then I think it's fine that you patent it.

What makes no sense to me is how the genomes of naturally occurring life forms can be patented. By the same rationale, Clyde Tombaugh could have patented Pluto, and Galileo would have patents to Jupiter's moons. It makes no sense whatsover.

Nobody "invents" lifeforms. They hack a few strands of dna cobbled together from other sources and wait until a batch actually grows.

Which is what the plants do all by themselves. Copyright, maybe. But not patent.

In this case patent's are less harmfull than Copyright. With a patent your idea becomes public domain farily quickly with Copyright 100 years from now the same DNA would be protected.

How do we fix that problem?

Loser pays.

With loser pays the big corporation can always generate such legal fees that, even if you have a 70% chance of prevailing, you'd be an idiot to risk bankruptcy.

And why is this worse than the current situation in which the same thing can happen but there is no chance of a recovery by the loser?

When working correctly, patents allow information that would be a corporate secret to be more openly disclosed. In my career, I've been able to discuss technical details more frankly in blog posts and academic publications because we know that our large, well-funded competitors would not be able to abuse that disclosure to harm our business.

The public good here is hard to measure from the other side, as a reader of those blog posts and papers; if it were not for IP protection, I would not willingly endanger my job or business to discuss the technology so openly.

> Perhaps you can provide a counter example that shows the usefulness of software patents.

There are some interesting optimization methods that probably wouldn't have been made public if it weren't for patents.

I want all useful algorithms to become publically known. Patents help accomplish that.

How do you propose to get people to forego trade secret protection?

There are some interesting optimization methods that probably wouldn't have been made public if it weren't for patents.

Which ones?

I want all useful algorithms to become publically known. Patents help accomplish that.

I don't think that's true. If you invent an algorithm that is genuinely novel and you get a competitive advantage from it, patenting it is the last thing you want to do. Keep it a trade secret.

Why? Because as a legitimate company you can't enforce the patent. If you try, you'll just be counter-sued for infringing someone else's patents.

PKZip was one that really moved the compression industry forward. JPEG is another. The current H.264 debate is a sound one and the owners of that patent have even publicly stated that it'll be free to use for the next 5 years or so. Imagine how many better methods may be discovered to compress and stream video than H.264 over the next half decade. That's a lot of time in this industry!

> There are some interesting optimization methods that probably wouldn't have been made public if it weren't for patents.

The one that I was thinking of was http://en.wikipedia.org/wiki/Karmarkars_algorithm .

> If you invent an algorithm that is genuinely novel and you get a competitive advantage from it, patenting it is the last thing you want to do. Keep it a trade secret.

You're assuming that the competitive advantage from using the algorithm exceeds the value that one can get from licensing. That's not necessarily true.

> Because as a legitimate company you can't enforce the patent. If you try, you'll just be counter-sued for infringing someone else's patents.

That's not necessarily or always true. Take drugs - the majors don't clone one another - they honor each other's patents.

For example: I come up with a system where basically I define 1 xml file which is then read by 3 different mechanisms to generate the m/v/c of a web application. the definition significantly saves times on web development when using the system.

Now look the argument is "oh its just a few algorithms used in a certain way, that should not be patented". however how is that any different than creating a ammo holder for .45mm bullets which can be put into any gun that can use .45mm bullets? Its just taking some metal and springs and putting them together in some novel way which has not yet been done before.

Remember once you see the invention that is TRULY innovative AND simple but nobody quite thought of using it that way before, it becomes a no-brainer. You start to think "yea given a few hrs I can make one of these myself without any special training or anything, without even looking at schematics, or even taking the one I have apart" but the fact is that nobody thought of it before.

The fact is, it was mentioned in a '96 letter, we must decide what part of programming is the "words" and what part is the "innovation" so that people can't do the equivalent of copyrighting words, but they can copyright books (the end product). (yes I am aware that patent != copyright, but try to see the analogy)

"I know patents cause trouble, but what about the case where a patent helps an entrepreneur get the value out of an invention when big corporations steal it, market it, and make money without giving the inventor a dime?"

Patents don't protect against that. Problem is, filing a patent takes time, money, and attention that are in very short supply in a startup. You can file to protect your inventions - but if you do, that takes time away from more important things like getting the money you need to stay in business. And patents generally need to be filed before the invention becomes public (you have a year grace period in the U.S, but no such boon in the EU). If you gate your launch process on the speed of the patent process, you won't be launching much.

That's why Google's first patent wasn't filed until 2003 (http://www.google.com/patents?vid=USPAT6526440 - the PageRank patent is actually owned by Stanford, since they were Page & Brin's employers when it was invented). And many of the initial EBay inventions - online auctions and reputation, for example - were never patented, because the 1 year grace period had expired by the time they got around to it. Patents are something you get once you become big and have extra money lying around, not something you can afford as a startup.

I don't think that is exactly true. A good patent from a very good firm in L.A. will run you about $20k. Yes, it's a lot, but I a computer engineer could sock that much away in a year. Work on the software at home at night, patent it, sell the patent to Apple or Microsoft or another big company for a couple million dollars. The corporation makes $100 Million of the device. Great, the inventor got $2M and can innovate for the rest of his or her life.

That's a good story in my opinion. Is that any better or worse than a social media site that gets 1,000,000 users and sells itself to google or whatever? How is that better?

Seriously, and with all due respect, a lot of the social media sites that do that add little value to anyone but the investors. Sometimes they take money from their customers in nefarious ways that maybe the world would be better off without -- especially right now.

So I'm a staunch supporter of software patents, because I see them as a path to freedom and a mechanism to publish ideas into the population and allow them to be built upon and improve the quality of life in the world.

I think the world could use a little improvement in its quality right now. There are probably millions of people who will agree with me.

When it comes down to it -- you gotta make money. That's what it comes down to and patents, any kind of patent, whether it is for wind shield wipers or for a software revolutionizes the solar energy system.

If something like that saves the world, I don't see a problem with an inventor getting a little bling out of it.

When it comes down to it, I just don't see the difference between gears spinning in the real world and gears spinning in a 3d model. That's every bit as much software as the algorithms and hardware behind some of what the iPhone does.

Some little company patented a film that goes over glass to make it much much stronger, but still transparent with UV protection and over a decade later, a car company used that material in its windshields without ever compensating this small individual owned business. It just turned out that when the material was invented, there was no market for it. No one could figure out what to do with it.

But it was within the 20 year period for the invention, which is essentially an arrangement of atoms in space, which could all be described by software. But I digress.

"Work on the software at home at night, patent it, sell the patent to Apple or Microsoft or another big company for a couple million dollars."

Your employer will usually own the patent then, as part of your invention disclosure agreement. California's the lone state that doesn't let employers claim ownership over inventions done in your own time, but there's a big exception for inventions that relate to the current or forseeable businesses of your employer. If you work for a big tech company, that could be practically anything.

Okay, I'll go further then to say, save up $40,000, then quit your job. Create a company called whateveryouwannacallit.com and start inventing stuff. New algorithms. Optimize compression. Optimize Video streaming.

Then spend $20,000 to have an attorney draw up a patent for you. I know some. I can connect you. Then, market the patent to corporations. Start a bidding war. That's what it all boils down to.

It's hard work to feed yourself and clothe yourself and be an entrepreneur and build a company that solves real problems for people that are willing to pay money for it. Those entrepreneurs should be rewarded by the system just like corporations.

Corporations are people too. They are composed of employees who are their employees and feed beautiful families with children living good lives. Yes, some of them make billions, like Steve Jobs, but some of them live modest lives in an old neighborhood near downtown just like lots of other people.

I think there is a lot of value here. Good value for hard work and innovation and I think there need to be more rewards for innovation that makes the world better. This is just one way it can be done.

It happens genuinely everywhere. Maybe someone gets good at writing music and sells millions of albums. That's just code or could be represented with code, yet if I write an algorithm, how am I going to share that algorithm with the world and also receive the benefits. If you argue copyright, then perhaps I could copyright it, but that means the whole world has it and if the whole world has it, then the inventor gets essentially nothing. It's like open source.

As I see it. Patents are the only protection for open source. When including the source code with the patent, you ensure the idea can be built upon and improved by a community of passionate people who can really do a lot to build on the myriad patents out there.

A copyright doesn't do this. It's not sufficient. My argument is that Monsanto should be able to patent DNA and DNA is no different than the source code behind an algorithm. They are both equal under the law and that is the way it should be.

I mean, in essence, opponents of software patents are arguing for discrimination under the law. I don't understand why computer scientists and computer engineers who build novel software systems that move stuff around and store it on physical disks aren't every bit as patentable as a crane or a type of diesel engine.

"Then spend $20,000 to have an attorney draw up a patent for you. I know some. I can connect you. Then, market the patent to corporations. Start a bidding war."

This doesn't work either (unless you have an ex-Microsoftie whose initials are N.M. backing you). Think of it from the corporation's POV. They could buy the patent now. Or they could wait - since you just spent $20k patenting your invention, you're probably down to a year or less of runway left. When you run out of money, then they buy your company (and the patent with it), at fire-sale prices.

You are, of course, welcome to try. But typically people find they have more success by keeping the invention secret, then either finding customers or drumming up enough attention that they can get investors, then marketing the company to corporations while they're in a position of negotiating strength. Ultimately, you need money to stay in business - if you have patents but no money, you'll quickly find that someone with money gets the patent, but if you have money but no patents, you can get somebody else's patents pretty easily (either through employment or acquisition.

There are several options at this fork. A) go back to work with owning a patent on your resume. Lots of companies would hire you. Use the patent to make money and support yourself and avoid the need to sell at fire sale prices.

More success keeping the invention secret is precisely why patents are a good idea. It lets the inventor release the idea and still protect it.

And what do you do if a large corporation uses your invention anyway, and doesn't tell anyone they're doing so? Firstly, you have to find out they've infringed. Secondly, you have to sue for protection.

How will you accomplish either of those steps, let alone both?

Monsanto can patent dna, because they got those atoms in a particular sequence. I'm willing to stipulate DNA is no different than the soruce code behind a program, but no way is that the same as an algorithm.

an algorithm is a set of rules for solving the problem. i can follow those rules, i can pay someone to follow them for me. i could have a computer follow those rules, i could embed those rules in dna, do some chemistry i don't understand then extract the sequence that followed the rules i care about. The algorithm is independent of the medium.

You can patent an engine. Heck, you can patent a few. You can't patent the idea of engine.

Just to be clear, you're arguing software needs to license patents to simulate patented objects?

By the way, you forget that the purpose of patents is not to allow you to get rich, it is to "promote the Progress of Science and useful Arts". The relevant question is if you think progress of science is promoted by the current system, or if progress is hampered because software patent litigation puts a damper on it?

The multitouch patent, for example: Do you think Apple wouldn't have bothered with the iPhone if they couldn't patent their "invention" (which wasn't theirs either, but that's a different issue)? In that case, the patent has promoted progress. But it's sure standing in the way for ubiquitous competition and rollout of touch devices now.

Well, actually the purpose of patents IS to allow you to get rich. Why? Well, by protecting the invention, you allow the inventor to make money with it for some time. This means that if you put millions in R&D, you have a guarantee that if there's a result, you can also win back the investment. Do you really think pharmaceutical companies would pour so much money into research if whatever they invented was immediatly copied by another company which doesn't put any money in R&D.

But I agree that patents are sometimes absurd, software patents are quite absurd most of the time.

This one is even cooler, patenting patent trolling:

http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Se...

Absurdity is ubiquitous. Computer science is no exception. I know we think we are above it, but we aren't. We're just like everyone else. People don't have to be mechanical engineers to pattent something physical. They don't have to be computer scientists to patent an algorithm either.

Maybe that person who patented the method of swinging really thought the idea was great and could save the world. I don't know. But really, patenting inventions is almost a right.

We argue for copyright protection, or the abolishment of it in the same mind. If there are no patents for software and no copyrights period, what is there to protect the risk innovators take to make the world better?

It's true of corporations as well, because they spend a lot of money on Research and Developmennt. R&D budgets are in the billions. If a company, like Apple in the current debate, can't protect that investment, a lot of people who participate at Hacker News wouldn't have jobs. They'd have to go do something in some mundane role that has negligible impact compared to something, like the iPhone.

We all cheered when it arrived and now that Steve Jobs wants to get his money for it, his investors money for it, and his employees satisfaction for a job well done, he has to fight for that value.

If, in fact, HTC did steal some ideas from the iPhone, why shouldn't Apple get compensated for the lost value? Those are phones Apple could have sold and why shouldn't Apple have every bit as much right to make a positive return on their investment in the R&D for the iPhone and the parts of the iPhone that make it special to Apple, to Jobs, and for consumers?

why shouldn't Apple get compensated for the lost value?

Because most of the cited patents are probably just restatements of research done at universities and research labs (like Bell Labs) in the 1960s and 70s. Note that I haven't read the claims to avoid polluting my own work. Most patents don't seem to explain anything, they're deliberately obscure, when they're supposed to be teaching future inventors how to reproduce the invention.

Assignee Name and Address: Halliburton Energy Services Inc.

Wow, Halliburton is just synonymous with evil.

I can hardly believe that this exists... but there it is. I think this is the only evidence needed to establish the fact that our patent system is hopelessly and irreparably broken.

That's an application, though, not a patent.

These two patents don't mean that all software patents are absurb, but it means that the current notion of software patents itself is absurd.

No, I don't think it does and I don't think simply stating that it is without providing any evidence for your case is going to change my mind or anyone's mind who agrees with me, but it looks like I'm the only one. Anyway....

The legal system is there to figure out if a patent is absurd and if you read the whole patent for the swing, at the end, it was ruled out. At least claims A, B, C and D. That same review process exists for software patents too. The patent office is taking steps to fix some of the problems, at least it is being pragmatic about pre-approval review processes that still protect the inventor.

Lots of arguments all the way to the echelon of the judicial system have taken place over software patents and it was ruled they are valid, so if a claim is to be made that they should be abolished -- it better be a very good one.

>The legal system is there to figure out if a patent is absurd

If the legal system was free and fast, you would have a point. However, the legal system is expensive and slow. A system where a patent as clearly absurd as this one can survive to the point of requiring a legal challenge to be overturned is simply unacceptable.

A patent also takes 8 years to come to fruition. There is a huge backup. Startups can be invented and sold in that amount of time or fail completely, so if it's just one person in a garage, there are bigger issues, but still, some of those people are saving up a lot of money right now selling iPhones. Even kids in malls and hackers in their basements.

Perhaps I am being too egalitarian in my stance.

I agree. The patent system is too slow. I'm not sure what your point is, because that only seems to support the argument that the current system of software patents is absurd. Despite the slowness and expense of the review process, crap like this gets through all the time, then requiring an even slower and even more expensive legal challenge.

I think the patent laws and the patent office should figure out if a patent application is absurd. Absurd patents should never exist, and the patent laws should be very clear and very strict. (Actually, I don't think we should have patents period.)

> Why not link to this one: http://www.freepatentsonline.com/6368227.pdf

This is a mock patent. It was obtained by an IP lawyer for the exact purpose of demonstrating the absurdity of the state of the patent system in the US.

The legal counsel person in one my previous companies was a close friend of this patent author. That's why I know.

Well, yes, of course it is a mock patent. The point is that it was even accepted in the first place.

The problem isn't "is it a good idea?"

The problem is, "Is math part of the structure of the universe?" Is pi a discovery that is fundamental to the existence of reality, or is it just a useful shorthand for thinking about things we don't understand?

If math is discovered, and algorithms are math, then algorithms can't be patented. You may reject the notion that math is discovered, or that algorithms are math, I disagree.

I'm fine with patenting a specific sequence of instructions on a particular chip, because they're extra fast or something. that's an invention.

You can't patent an electron, or the idea of an electron. You can patent a device for moving an electron around.

True, it doesn't mean all software patents are absurd, but it does mean something is wrong with the system.

Wow. That method of swinging on the swing actually is innovative. At least for me. I never did that by pulling one chain and the the other. I always sort of balanced my whole body to start and keep swinging.

I'm eager to try this new method. Should I apply for some kind of licence to that guy? I don't live in US is that patent valid in other countries?

When did anyone bring up inventors? It would be nice if they had rights, under the current system there is no inventor protection. Any idiot who can imagine something can get a patent, and then turn around and sue the hard working engineer who puts together a practical product.

Nah. This is easy to beat. Hang the swing set from anything besides a horizontal tree branch.

Do you think you should be able to patent a sentence? Even if its a good one? If not, why?

patents apply only to things that are: anything made by man new non-obvious commercially useful

i think a sentence, even Shakespeare, fail one or more of those qualifications. which is why we have copyright for sentences.

It depends on the sentence. And I know where you are going.

The patent was canceled according to page 8.

The patent system worked in this case.

Reply on news.ycombinator.com