Then all of the subsequent ones are invalid, and can be demonstrated as such trivially. They can’t actually be enforced.
Also, based on my own experiences, the USPO actually does a fairly extensive investigation, and having aspects of your patent show up in academic literature or other patents tends to cause them to reject the application even if the overall patent is a clearly unique idea, which causes you to have to go back and explain exactly how it’s different to a crazy level of detail.
Wow capitalism sucks again…suprise.
Without solving capitalism, I would recommend that any patent should expire after a short time and become useful to anyone. Today I think there is almost no expiracy.
My employer got into a patent scuffle a while back when some other company sued us over a technique that we’d been using for ages, but hadn’t bothered to patent ourselves since it was, well, trivial (my memory of the details is fuzzy, but it was something like placing a reference to a file instead of a full copy of a file if the size and timestamp matched). We fought it and won and got their patent invalidated, but this was one small company versus another small company. I’d be less optimistic about our ability to fight off a Microsoft or IBM.
And remember that it’s not just about being first, but that it also has to be non-obvious. I think this is where a lot of the criticism is coming from, since there’s no real evidence that patent offices are even very qualified to determine what is and isn’t obvious when it comes to software patents. To my mind, sure, things like MP3’s psychoacoustic model or arithmetic coding’s mind-bending math are sufficiently non-obvious and required great research or insight, let them have their patents. But encoding motion as a series of X,Y,Z coordinates transmitted over a TCP/IP channel (I vaguely remember something like that coming up before) is bloody obvious and there should be no patent for that.
I’m going to wave my magic wand and declare that patents will expire after, oh, let’s say 20 years. Tada!
I’m going to wave my magic wand and declare that patents will expire after, oh, let’s say 20 years.
Brilliant! They should make it work like that!
NPE’s alone have cost companies half a trillion dollars in the last 20 years.
The sum for practising entities is going to be many times that, not to mention all the companies simply locked out the markets…
It has taken Uniloc 8 years to “win” their case and they still haven’t been paid their $388 million (the latest verdict threw out that sum - leaving it up to yet another lawsuit). How many small companies that actually produce software (as opposed to spending their time litigating, which seems to be the sole business of Uniloc USA) do you think can survive 8+ years of lawsuits?
As for i4i; read the patent - it is good example of everything that is wrong about software patents. Vague, over-reaching, and describing something that XML was actually designed to do in the first place. That Microsoft failed to bury this case is a pretty good example of just how broken the software patent system is.
In general though, patents tend to be valuable IP for small software companies, as they provide serious incentive for a larger company to purchase the smaller company.
In other words, the incentive for a small company is to file a lot of patents and then wait until you can sue someone for infringing your patents or be purchased by a bigger company. This furthers innovation how?
Baseless statement is baseless.
I’d elaborate, but given that you apparently think that a blinking cursor is a non-obvious invention, I doubt I’d be able to define bogus at a level that you’d accept.
Do you have any actual experience with the software patent process? Have you ever dealt with patents in any way?
Yes and yes.
Do you have any actual experience with software patents? Because from this comment, one wouldn’t think that you’d ever actually read one.
Also, based on my own experiences, the USPO actually does a fairly extensive investigation, and having aspects of your patent show up in academic literature or other patents tends to cause them to reject the application even if the overall patent is a clearly unique idea, which causes you to have to go back and explain exactly how it’s different to a crazy level of detail.
Which patent office was that? Because, judged simply from the patents that are issued, the USPO apparently can’t even figure out how to search their own patent database. Nor are they apparently able to determine obviousness for even ordinary day-to-day things (the patent for toasted bread issued in 2000 is particularly amusing) - much less stuff that is obvious to a professional in the field.
Come now, you can’t tease us with that without a link!
This one?
http://www.google.com/patents/about/6139885_Preparation_of_toasted_bread_pro.html?id=obUFAAAAEBAJ
In other words, the incentive for a small company is to file a lot of patents and then wait until you can sue someone for infringing your patents or be purchased by a bigger company. This furthers innovation how?
Well, in my particular experience, this isn’t exactly how it works.
The smaller companies create things, which they patent… In buying the company, the larger company gets the software and the IP protection of the patent.
The smaller company wasn’t just cranking out patents for the sake of getting patents… they were innovating, and producing useful technology… The patents make it such that the larger company couldn’t just steal their ideas and reimplement them without risk of legal ramifications, so they bought the smaller company.
I’d elaborate, but given that you apparently think that a blinking cursor is a non-obvious invention, I doubt I’d be able to define bogus at a level that you’d accept.
But the patent isn’t on “a blinking cursor”. It’s on a specific mechanism by which you can achieve such an effect. You can create other mechanisms to achieve the same effect, and not be in violation of the patent.
Do you have any actual experience with software patents? Because from this comment, one wouldn’t think that you’d ever actually read one.
Yep, actually have written some.
Which patent office was that? Because, judged simply from the patents that are issued, the USPO apparently can’t even figure out how to search their own patent database. Nor are they apparently able to determine obviousness for even ordinary day-to-day things (the patent for toasted bread issued in 2000 is particularly amusing) - much less stuff that is obvious to a professional in the field.
Like I said, I’ve had experience where they’ve come back with specific citations of related technology cited in patents and academic literature. I’m not sure exactly how they conduct their searches, but they obviously do something which turns up material that is relevant (in the case of some of our patents, it’s come up with other patents BY US on related technology).
Recalling some research I did on software patents a while ago. A couple of things stood out.
First, software patents were the second most likely category of patents to end up in a lawsuit. They trail only business method patents. This is no surprise–poor patent quality will cause patents to end up in suit, and software patents are necessarily much more vague than, say, chemical patents.
Second, set aside chemical and pharmaceutical firms. Looking at firms in other industries, by the year 2000, the aggregate US litigation costs to alleged infringers were more than triple the worldwide profits from the associated patents.
(Got those figures from this book, if you have questions about the source.)
As for how the mythical small software maker in the big scary world is supposed to survive in a world without software patents, might I suggest that copyright, trademark, trade secret, network effects, superior product, and superior service might be able to help? Is that really so difficult? How many patents hath Facebook (or, more saliently, how many patents did Facebook have in its early years)?
And might I note that I find it hilarious that one of our most rabid libertarian trolls finds it in his heart to argue on behalf of software patents, which are a wildly anti-competitive intrusion of government into the wonderful world of the free market, because he perceives a threat to his livelihood without such intervention. There’s a lesson in that.
To be fair (heh), among the purposes of patents was, in fact, the obviation trade secrets.
That doesn’t make software patents any less bullshit.
True, but only insofar as it would contribute to the progress of science and useful arts. The hideous increases in patent applications, application pendency, and litigation costs certainly call into question whether the system is doing that.
There are some funny patents in the gaming industry. My favorite is Ridge Racer – they patented the idea of mini games during loading screens.
Check it: http://www.freepatentsonline.com/5718632.pdf
I love how the intent of a patent is to get the inventor the document how something works for the common good, so it doesn’t linger in secrecy – but in practice, patent writers write the convoluted and least useful explanations they can.
An interesting proposal for software patents: http://www.youtube.com/watch?v=32LRK0hzEmc&feature=player_detailpage#t=2440s
The other funny thing, is that the more generic and non-novel a patent is, the more valuable it is, as more people are likely to infringe on it.
Yea, but there are very few patents in the games industry. The quickest way to get blackballed is to start filing for patents.
Copyright, trademark and even trade secret disputes? Accepted*. Patent? Erm…
(*Mostly. Unless your name’s Tim “The Troll” Langdell)
So, do you believe in patents at all? If I invent some new physical device, for instance, do you believe that it’s ok for me to be able to patent that, or should we do away with those patents as well?
And might I note that I find it hilarious that one of our most rabid libertarian trolls finds it in his heart to argue on behalf of software patents, which are a wildly anti-competitive intrusion of government into the wonderful world of the free market, because he perceives a threat to his livelihood without such intervention. There’s a lesson in that.
Patents are not some new grand intrusion into our lives. They’re an aspect of government which has been in place for a few centuries.
The idea of protecting someone from theft is not an anti-libertarian idea. Quite the contrary. And it’s been an accepted point of law that scientific discoveries and innovations are property, which those who discover them are entitled to benefit from.
Do you dispute the very notion of intellectual property?
The guidelines for patents seem pretty reasonable; the way they’re getting applied to software doesn’t seem reasonable at all.
And it’s been an accepted point of law that scientific discoveries and innovations are property, which those who discover them are entitled to benefit from.
Nope.
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.
Those two phrases aren’t even close to each other. They don’t describe them as property, where elsewhere in the document other things are described as property. They’re a concept explicitly granted to get the results hoped for the first half of the sentence. By contrast, other property and concepts throughout the document are described as “inalienable rights” and such. If you read the history of the concept it’s pretty much explicitly state planning created a right out of thin air. It was mostly used for early industrial planning to encourage specific industries and inventions; “those who discover are entitled to benefit” is backwards. See history of US law.
A state big enough to enforce property rights on patent law, especially non-physical-goods patent law, is pretty much in contradiction to most of the founding principles of libertarianism. One of the reasons “propertiarianism” would be more accurate.
See, I think the core issue is this…
There may be a larger subset of software patents which are “bad”. In that they are vague, or overlap, or whatever. In many of the cases folks have mentioned, they would simply be invalidated if someone actually tried to enforce them (the cases where someone else had already invented something, for instance, but didn’t patent it). This is where you had crazy situations where lawyers would dig through old emails and notebooks and crap.
However, there seems to be an idea being presented here by some, which I find disquieting, that software patents are inherently worthless.
If you reject the notion of patents entirely, I guess that view is at least consistent (albeit foolish in my opinion). But some folks seem to think that patenting a physical object is ok, but patenting something like software is bad. That doesn’t make any sense to me.
My inventions are just as “real” as they would be if they were physical objects. They involved creativity to come up with. We iteratively create and refine ideas, as any inventor does.
And for the same reason that physical devices can be patented, we should be able to patent ours… Because, frankly, it’s far easier to steal the ideas for software than it is to steal the ideas for physical objects.
If you design a new engine, and I want to copy it, I have to actually retool a manufacturing facility to do so… If you design software, it’s much easier for me to copy it.
Now, there may be all kinds of problems with the way the patent office functions. Certainly, in many cases we’re talking about stuff which is beyond the grasp of many of the clerks there I would suspect, as software covers a far broader realm than mechanical engineering of physical devices as would cover patents in the old days.
But the answer isn’t, “Just stop allowing people to patent their ideas.”
I don’t think they’re worthless, just that the system has clearly run off the rails. The blinking cursor, for example, strikes me as an absolutely asinine patent.
What are they patenting? Remember patents are on implementations, not concepts, that are supposed to be new, non-obvious, and useful, to further the public good. They have nothing to do with “what’s creative” or “what you make”; if you create a useless idea, you can’t patent it. Or at least you’re not supposed to be able to.
What on earth is being patented for a blinking cursor?
All of those strike me as such nuts. Seriously, everyone else in software is supposed to pay royalties to make a fucking cursor blink? How does that further the public good? Then there’s the concept of “implementation”, where software is far closer to abstract math, which is unpatentable for the usual reasons, then they are car engineering. Hardware engineering, at least, leaves open the option of alternate ways to do things and pushing science forward. There’s only one way to write a lambda function from a mathemetical sense.
What on earth is being patented for a blinking cursor?
The thing which was actually patented was the process of XORing a cursor onto the display, and then XORing it again to make it blink off.
It wasn’t actually patenting the abstract notion of a cursor. It was patenting that one specific mechanism by which to implement it.
Seriously, everyone else in software is supposed to pay royalties to make a fucking cursor blink?
You can either do that, or you can design around it. For this specific example, there are other designs which would achieve a similar outcome.
How does that further the public good?
This is a tiny patent. The same type of thing could be said for, say, the patent on highway dividers. Or little garbage toys that someone makes.
Hardware engineering, at least, leaves open the option of alternate ways to do things and pushing science forward. There’s only one way to write a lambda function from a mathemetical sense.
As pointed out above, in the case of the XOR patent, there are other ways to achieve the same effect. The same goes for software.
The type of stuff contained in a software patent is equivalent to the stuff you’d see in a mechanical engineering design patent. You’ve got various components put together in a specific way.
You’re not allowed to patent something like a lambda function. But you can’t just say, “Oh well all software is just math, so you can’t patent any of it.” Modern software exists at a level where it is MORE than just the math. You’re not writing software from the perspective of the mathematical operations that are being done by the processor’s registers. You’re creating a complex system of component parts, just like building any other type of system, whether it be physical or virtual.
If folks are interested, here’s an old ACM article that folks might find informative.
How many ways are there to make a cursor blink? 12? What, exactly, is the point of the patent? To encourage people to exhaust the tiny search space of how to make a cursor blink?
All software that exists can be described as rearranged in Turing machine terms; all software is math. The only patentable angles that make sense to me are specific implementations for specific goals.
This is a tiny patent.
That’s not an answer to the question. Highway dividers have a very wide range of design possibilities, involve engineering and production work, and have multiple design constraints about cost, safety, etc., and the design has significantly improved over the years.
In that case, your experience isn’t related to software or it’s very old. Because what you describe is not how it has worked for the past 6-7 years.
Writing software today without infringing on someone’s patent is becoming increasingly difficult in the US. There is a reason why Microsoft et al. are involved in an arms race to file as many patents as possible - it gives them ammunition with which to smack anyone who tries to threaten their turf - in addition to which they can use the patents to kill any small companies who might think of actually competing with them.
Which is why the actual incentive today is to create NPE (Non-Practicing Entities) as I described. If you’re not actually doing anything, you can’t be counter-sued. It’s all win.
But the patent isn’t on “a blinking cursor”. It’s on a specific mechanism by which you can achieve such an effect. You can create other mechanisms to achieve the same effect, and not be in violation of the patent.
Again - this is not the form of software patents today. Many - if not most - are written to be as broad and non-specific as possible, so as to be usable in litigation. That’s the purpose after all - a patent is only as useful as the claims you can make with it.
If you design a new engine, and I want to copy it, I have to actually retool a manufacturing facility to do so… If you design software, it’s much easier for me to copy it.
In which case you are protected by copyright and can sue for copyright infringement.