Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?

Comment Re:You know where it went.. (Score 1) 234

A recent study by the Richmond School of Law found that the USPTO's actual grant rate is currently running at about 89%. In 2001, it was as high as 99%. See http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2225781 page 9. In 2001, it didn't matter if an application was overbroad, obvious, trivial, a duplicate, or unreasonable, they ALL got granted.

On the contrary, 80-90% of patents are initially rejected.. Yes, the rate at which patents are eventually granted is 90%, but that's after they've been substantially amended and narrowed. Pointing at the eventual grant rate and claiming "it didn't matter if an application was overbroad, obvious, trivial, a duplicate, or unreasonable" while hiding the rejection rate is disingenuous.

Furthermore, the fact that you mention "duplicate" indicates you may not actually know what you're talking about. There are no duplicate patents, as double patenting is a rejection under 35 USC 101. There are continuation applications, in which one application gets filed and then later, another application with the same specification, title, and figures gets filed, claiming priority to the first one. So, to someone who doesn't understand what's going on, they may look like duplicates... however, (i) the claims are different, and (ii) the second application, even though filed later, expires on the same day the first one expires. They're really just offshoots of the same application, and not "duplicates" at all. My guess is that you heard that terrible episode of This American Life where they mentioned that several applications (all continuations of a first application) had the same name, and were as equally confused as their "journalists".
This also explains your later complaint:

6) Finally, I suggest that it is a mistake to allow patent applicants to modify or extend their patents after submission. This complicates the patent pipeline. It facilitates ‘submarine’ patents. It enables capturing Standards. It also enables gaming the patent system. Reform must simplify and reduce the patent process. Patents should be quickly evaluated. Most should be denied. If an applicant wishes to modify a denied patent, they should alter it, resubmit, and pay a new filing fee.

None of this is true - patent applications cannot be extended after submission, and any entry of additional material is "new matter" and rejected immediately. Applications can currently be altered, resubmitted, and filed with a new filing fee, and then they have the same title, as mentioned above when you were complaining about duplicates. Submarine patents are an entirely different thing, where a patent is granted and first published years after filing. Patents are now published after 18 months, and so submarine patents are mostly a thing of the past.

Most of your complaints are based on misunderstandings of the process and half-truths or outright falsehoods spread by poor journalists. A few minutes of research before you post would clarify them.

Comment Re:You know where it went.. (Score 1) 234

You're missing the point. A lot of this was unintentional. They made the USPTO run on fees that were charged for patents which gave the USPTO and incentive to rubber stamp patents while not receiving sufficient funding to cover the cost of having patent examiners that could do the investigation that they used to do.

If it was true that the USPTO was just rubber stamping patents, then you'd expect to see close to a 100% allowance rate, no? Instead, you get about an 80-90% rejection rate, depending on art group.

In reality, it's a bit more complicated than you think: the USPTO runs on fees, but those fees are for application filing, searching, examination, continued examination after final rejection, issuance, and maintenance fees. Only the last two give an incentive to rubber stamp patents - the others give an incentive to reject patents.

Furthermore, the USPTO is full of humans, with their own personal interests. Specifically, the Examiners are graded for promotion/bonuses on a point system in which they earn points for examining an application, points for rejecting it, and points for issuing it - they have an incentive to initially reject patents, even just to allow them after getting a request for continued examination, because they get an extra couple points. That also means extra fees for the PTO, incidentally. The Examiners, however, get no points for maintenance fees, so their interests are in drawing out prosecution to a certain point, rather than immediately rubber stamping everything allowed.

What needs to happen is that the USPTO needs to go back to being a government service the user fees need to be based upon the amount of time and energy it takes to deal with the application.

They are - see the bit about continued examination fees. More time spent = more money.

And while we're at it, the duration of the patent period should go from the point where the first application is received to a reasonable period after that. For technology 7 years is likely more than adequate as a lot of that IP is no longer of value several years later.

The problem here is that now you need Congress passing different laws for patent terms for every single industry. And when there's a new industry - like online retail - is that really a "new" industry that should deserve a short 5 year patent term; just an offshoot of an older internet industry that deserves a 10 year patent term; an offshoot of computers in general, deserving a 15 year term; or an offshoot of electrical engineering, deserving a full 20 year term? And how do you determine this on day one so that you can put it in the statutes, rather than waiting 5 years to determine how quickly the new industry advances?
Similarly, say you invent a new computer control sequence for car engines that significantly improves mpg efficiency: is that an automotive advancement with a 20 year term, or a software advancement with a 5 year term? And while maybe you can pick an answer and justify it, what's the general, objective rule that you can apply to any new invention to determine what its term should be?

Comment Re:The bogus patent in question (Score 1) 65

This bit, gleaned from the linked slashdot interview, sums it up:

Can you explain? by trcooper

Can you explain, in terms I could tell the average person, how your patent is novel enough that anyone who wants to distribute audio over the internet should license it from you? I'd appreciate it if you could address how the distributions of podcasts today widely differs from downloading audio files in 1995 and how your patent help change this.

Logan: Trcooper, this is one of those of questions that could get me in a boatload of troubleâ"with my lawyer, that is. Any comments I make regarding the claims and how they are different from previous systems, can and will be used against me in court. So Iâ(TM)ll have to take a pass on this one.

This bit, gleaned from the recent television interview, sums it up:

Can you explain why you claim you're innocent of the murder you've been accused of?

Defendant: My lawyer has told me not to answer any questions, because anything I say may be twisted or used against me in court. So I'll have to take a pass on this one.

Ah ha! Clearly he did it and there's no need for trials!

... at least, by your stilted logic.

Comment Re:I don't understand (Score 1) 111

The process of vulcanization of rubber is approximately "raw rubber plus sulfur plus heat equals material with new properties". You have created something new by this process. The new physical material has different properties.

Sure, but just so that you're not moving the goalposts, realize that we're talking about a patent on the process of vulcanization, not the physical rubber. And that process is simply a mathematical algorithm, right?

Gear ratios are properties of physical objects that have a mathematical relationship. You don't patent a ratio, you patent a mechanism that may use those ratios to some effect.

You can also patent the method of transferring power via those gears (if it's a new and nonobvious process, of course). It's not just the machine that's patentable, but also the process... even though the process is merely a mathematical algorithm.

If the mechanism you have invented is a universal gearing system, then should someone who picks a particular gearing ratio because it is useful for some purpose that nobody noticed before deserve a patent on that ratio? I don't think that they should.

Why not? If they invented a brand new use for the ratio that no one ever noticed before, why shouldn't they receive a patent? They say build a better mousetrap, and the world will beat a path to your door... you're saying "pfff, it's just wood and metal and springy bits. Nothing new about that."

...Good thing no one writes a patent claim that says "A method, comprising controlling computer hardware, period, the end" then, eh?...

It's extremely frustrating to me that people *do* write those claims...

[Citation needed].

... and don't have anyone pointing out that general purpose computers are designed to be able to carry out *any* calculation.

Sure, a Universal Turing Machine can carry out any calculation, by definition. So why didn't we have GTA V thirty years ago? Was it because people hadn't figured out to make those calculations yet? And it's not mere hardware limitations, or there'd have been no advances in game technology from the very first PS3 games to the newest ones, for example. No, people have figured out new and better ways to use the hardware that were never conceived of when the hardware was first built.

The processors, memory controllers, caches, etc that make up the computer are all patentable. They do stuff. Software is just math that controls how they turn a given input into an output.

A furnace does stuff. The timing algorithm for vulcanizing your rubber is just math that controls how that furnace will turn a given input into an output. But you agree that that's patentable.

I write software every day.

That was obvious. I'll never understand why programmers are so disparaging of their own inventions, but it's definitely common.

It's not magic, and absent a computer, you could still generate the output of a program, just not nearly as fast (which is why the general purpose computer is such a great invention).

Sure, but it's like that old joke about going to the doctor with an injury, he hits you with a hammer and cures it, and gives you a bill for $500. $5 is for the hammer, the other $495 are knowing where to hit. Without the novel and innovative algorithm, that computer - or you with your pen and paper - just sit there doing nothing. So, while the computer, or your pen and paper, can be used to calculate the most amazing things ever, the mere existence of those tools doesn't mean that the resulting calculations are obvious.

...Here's a simple question - imagine the most novel, non-obvious bit of software ever. Like, some brand new algorithm that can losslessly compress even random data by 99%. Go nuts with imagining the details - point is, it's a revolutionary bit of code that no one has ever thought of, and it wins billions of awards and the keys to the city. Should it still be unpatentable, merely because it's software?...

Well, yes, it should be unpatentable. Just because my nifty new bit of math can be implemented quickly on a general-purpose computer, that doesn't make it an invention.

Well, by definition, it is an invention. You're just saying it shouldn't be protectable.

That's the thing about math...we *discover* mathematical ideas all the time. In software, we make use of them to some effect. That doesn't mean that because they're more complicated than a certain gearing ratio that they suddenly create a new machine when used.

No, I agree that the concept of "poof, it's a new machine" is a stupid one that the courts came up with to try to get around some earlier case law. But if we're talking policy and what the law should be, I think software should be patentable, with the carve-out that it shouldn't create thoughtcrimes by performing the same calculations mentally or on paper. This is what they were really trying to get to in Bilski, but they haven't figured it out completely yet: if you can infringe a claim mentally, and the patent owner can get an injunction to stop you from infringing, how do we force someone to stop thinking of something? It's not possible, nor should it be even up for debate. Diagnostic claims are similar - determine patient X has symptom Y, realize they have disease Z. That could be infringed merely by reading the patent, if it describes an example of patient X, since you'll realize that hypothetical patient X with symptom Y has disease Z. So we can't have that. But by requiring physical steps or devices in the claim, you can't possibly infringe mentally. This also fits with other areas of the law, such as criminal law, where thought must be coupled with action to create liability.

But you disagree... Why shouldn't your work be protectable from people wanting to duplicate it? And before you say "copyright", realize that only covers people who want to copy your specific code - if they want to reverse engineer it or code the same thing, copyright is useless. And before you say "first to market", realize that it's easy to hire a team of programmers to reverse engineer any product in two to three weeks, even if it took you years to write and debug.

Of course, there's another way your work could be protected - non-disclosure agreements, heavy handed licenses, cloud-processing everything so that no user has access to the full code, etc. I think you'll agree that that's a terrible result.

Comment Re:I don't understand (Score 3, Informative) 111

...The law is quite clear - if you come up with a new and nonobvious use for a known machine, such as a computer, then you have invented a patent eligible process.

Right. Except that computers have a single purpose...they do math.

This is not a new use of a known process.

You mention vulcanizing rubber. Furnaces have a single purpose - they heat things. Therefore, vulcanizing rubber isn't a new use of a known process?
Also, "doing math" isn't what's claimed, but rather specific operations that haven't been done before. They are, by definition, new uses of the older technologies.

This is saying "you're not allowed to write a computer program that does something in a certain instance". The courts are (again) failing to grasp the way that computers function. Vulcanization (of rubber) is a process. A computer program is an algorithm...math.

You're apparently not a physicist. Vulcanization of rubber is an algorithm... math. Gear ratios are algorithms... math. Every process or machine in the physical universe can be described mathematically. But that doesn't mean that a patent on an engine is attempting to claim all math, any more than a patent on a software method is attempting to claim all math.

The fact that you can control computer hardware (like a network connection) with software is about as far from non-obvious and new as you can get.

Good thing no one writes a patent claim that says "A method, comprising controlling computer hardware, period, the end" then, eh?

But really, this shows that your understanding of the issues is confused. Patent eligibility under 35 USC 101 is different from novelty under 35 USC 102 or obviousness under 35 USC 103. They're three different, independent requirements. You're trying to claim that all software should be exempt from patent eligibility under 35 USC 101... because it's all obvious under 35 USC 103? That simply mixes up two different statutes, two different bodies of case law, two different logical tests, etc.
Here's a simple question - imagine the most novel, non-obvious bit of software ever. Like, some brand new algorithm that can losslessly compress even random data by 99%. Go nuts with imagining the details - point is, it's a revolutionary bit of code that no one has ever thought of, and it wins billions of awards and the keys to the city. Should it still be unpatentable, merely because it's software? That's the issue about patent eligibility under 35 USC 101: should software, as an industry, be ineligible for patentability, regardless of how innovative or inventive something is.

Comment Re:I don't understand (Score 1) 111

> while most legal people are not conversant in technology, all patent attorneys are - the proper scientific or engineering background is a requirement to sit for the patent bar exam.

You're effectively trying to claim that ANY engineer is qualified to be an expert witness on ANY engineering discipline.

Except for the fact that I never said that and it's impossible to read my words to say that, and the fact that my post was specifically arguing that most PEs would not be qualified to discuss networking, yes. You sure caught me.

Comment Re:So does anybody... (Score 1) 111

And that says "The first spanning tree protocol was invented in 1985" which predates the earliest patent, therefore is prior art and invalidates the entire lawsuit. Why is this still in court?

Does the spanning tree protocol include each and every element in the patent claims? If you can literally make a table with each line of the claim on one side and a corresponding quote from the spanning tree protocol documentation on the other side, then it does invalidate the patent. However, if there's anything missing, then while it's relevant prior art, it's not anticipatory prior art.

Comment Re:I don't understand (Score 2) 111

Am I going to look at those patent numbers? No ! I don't want to get get a migraine headache since many modern IT patents are written in "legalese" such that someone with a Professional Engineering (yes a real one) background who is actually conversant with the field has a hard time understanding the words. Of course the opposite applies since most legal people can understand the words but not the context.

Considering that most PEs are civil engineers and that while EE is a valid discipline for the PE, they tend to be electrical system designers, I'm not sure why you think that that's a helpful background here. For example, very few IT people are PEs, simply because (a) it doesn't apply; and (b) they can't hack the necessary math and physics. Additionally, while most legal people are not conversant in technology, all patent attorneys are - the proper scientific or engineering background is a requirement to sit for the patent bar exam.

I particularly like fig 1c which basically a stylised drawing of the USA (well worth the LOL look) - see here and click on fig "1c" of Images which is the second image from the left.

Not sure why you think that's lol-worthy. The patent describes a system for self-healing networks such as the nationwide telecommunications backbones that go from major city to major city... Accordingly, any geographical network diagram of an example network will inevitably look like the country it's using for the example. Are you suuuure you're a PE?

Comment Re:I don't understand (Score 2) 111

Trying not to have an aneurysm...*why* in the world do courts continue to view software as patentable subject matter. The communications nodes are general-purpose computers with networking hardware attached. They have an operating system that allows for the switching of communications lines. Now there's a rash of lawsuits because the courts are too dimwitted to understand that software plus computer != invention/new machine.

No, the courts are following the law. A new and nonobvious use of a known machine is a patentable invention. Specifically, 35 USC 101 says "Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title." and 35 USC 100 defines a process as "The term “process” means process, art, or method, and includes a new use of a known process, machine, manufacture, composition of matter, or material."

The law is quite clear - if you come up with a new and nonobvious use for a known machine, such as a computer, then you have invented a patent eligible process.

Comment Re:Idiots are against Golden Rice (Score 1) 400

You linked to a Score:0 Anonymous Coward, who I think was probably making a sarcastic joke that they should obviously just eat more carrots. However after you posted we got at least two certifiable cases:
Score:5, Insightful ranting we should just "encourage them to grow more sweet potatoes" and
Score:5, Informative ranting they should just eat sweet potatoes and moringa tree leaves.

If they are growing rice, there are probably reasons that they grow rice. Maybe sweet potatoes don't grow in that soil and climate. Maybe they can't obtain an economically viable yeild. Maybe moringa tree leaves taste like shit. (WTF is a moringa tree anyway?) Maybe there are cultural reasons. It doesn't matter what the reasons are, it doesn't much matter if they are good reasons. The fact is that six hundred thousand children are DYING each year and another half million are going BLIND each year, and yelling at people in extreme poverty to just "eat some goddamn vegetables you idiots" is not a particularly successful solution. I kinda suspect that curing blindness and vitamin-deficiency disease might provide a teensy-weensy bit of help for them to climb themselves out of poverty.

If I found a mutant variety of rice out in the wild which contained Vitamin A, these GMO ranters would embracing it because it's a NATURAL mutant. We knew we wanted rice containing Vitamin A, we looked for it, we didn't find it (yet), that mutation might pop up in a field somewhere tomorrow..... but somehow it magically becomes daaaaaaaaangerous if we make it ourselves rather than waiting around looking for that "natural" mutation. Because a natural mutation potentially making cyanide in your food is somehow better than intelligently, carefully, deliberately putting vitamins into food.... because if we smartly do it that's unnaaaaaaatural.

There's the old Luddite saying: "If man were meant to fly he would have been born with wings"
Fuck that. If we weren't meant to engineer artificial solutions to problems we wouldn't have been born with brains. Well, some of us anyway.


Slashdot Top Deals

One good suit is worth a thousand resumes.