Slashdot Mirror
Patent Office Admits Truth — Things Are a Disaster
An anonymous reader writes "For years the US Patent and Trademark Office has published data to show how well it and the patent system were running. Under new leadership, the USPTO has begun to publish a dashboard of information, including a new look at questions like how long does it really take to get a final answer on whether you will receive a patent or not? The pat answer was, on the average, about 3 years. But with the new figures, it's obvious that the real number, when you don't play games with how you define a patent application, is six years. The backlog of patents is almost 730K. And the Commerce Department under the Obama administration wants the average down to 20 months. How does this happen? Only if everyone closes their eyes and pretends. It's time to take drastic action, like ending software patents. As it is, by the time companies get a software patent, there's little value to them because, after six years, the industry has already moved on."
The true value of a software patent isn't to protect an invention. It's to have a tool for extorting others. In that sense software patents have a shelf life much longer than six years.
You know, if people would claim only what they've fucking invented on patent applications, that backlog would be much smaller. Way to go, jerkoffs.
Living With a Nerd
I have 3 fairly recent patents (one hardware and two software) - none of them took over 3 years - and two of them took multiple rounds with the patent office
Why is the focus on software patents? Doesn't this show that the patent system in general doesn't scale up and needs fixing?
They're under-resourced, but increasing their resources won't solve the social and economic problem caused by the patents they grant on software. A lot of big patent holders are saying that the solution is to increase review standards, but how would that happen? How do you put clear limits on whether something described is sufficiently innovative, or sufficiently useful?
The only simple way to reduce the workload of the patent office is to cut certain fields right out of consideration. Start with software. Making the USPTO more efficient isn't our goal, but it's a happy coincidence that there's a solution to our problem that just so happens to solve a major USPTO problem.
And it's not just the USPTO. The European Patent Office has the same problems.
Expert in software patents or patent law? Contribute to the ESP wiki!
Don't get me wrong, I'm completely against software patents, but I'm way more offended by "business method" patents. And patents on something that someone did a hundred years ago, only now someone adds the line "on a computer" and suddenly that's a new patentable event.
Jealously hoarding mod points since 2007.
That's in Germany. There are no software patents in Europe, and Germans are (traditionally) much less interested in lawsuits than US citizens and corporations. I'd also expect German patent grants to actually be valid, rather than relying on lengthy court processes to refute patents that never should have been granted due to prior art or attempting to patent laws of nature.
None of these conditions apply in the USA. It is actually to the advantage of some large companies and their lobbying organizations to keep the patent office overwhelmed and confused: they can assemble portolios of defensive patents to protect their interests, and apply those portfolios at whim against smaller, more creative, developers or businesses that haven't already invested in manufacturing or development or sales of an older product line.
I'm just a bit curious here, but have these big companies you are speaking of copied your patented techniques by looking at your product or your patent application, or do you think that they reasonably could have invented them independently?
That seems unlikely. If anything, you'd grease palms to ensure that your application was delayed. While a patent application is pending, you get most of the benefits of a patent, but none of the costs.
I am TheRaven on Soylent News
Excellent, so you're patenting not even software, but mathematics! Can this get even more broken? Of course!
It takes a man to suffer ignorance and smile
Be yourself no matter what they say
Because phrases like
"Ideas are a dime a dozen. People who implement them are priceless" - Mary Kay Ash
"Invention is one per cent inspiration and ninety nine per cent perspiration" - Thomas Edison
sound better than "first come first served".
This is BS. Because if i also put in R&D dollars and my own blood and sweet and come up with the same or similar solution, i can't benefit from my hard work because someone else also thought of it and paid lawyers.
If it was about rewarding hard work, or R&D then independent discovery would be a valid defense. Its not.
The Grey Goo disaster happened 3 billion years ago. This rock is covered in self replicating machines!
As messed up as it is, the current system creates more value for patent trolls. If it takes six years to get my patent approved, that's six years "infringing" technology getting baked into competing products.
Oh, a lesson in history from Mr. I'm my own grandpa.
Would the creation of a novel sorting algorithm (presumably something significantly faster than a Quicksort) really help in terms of attracting attention to your product, or would it be better to either publish that algorithm with the ACM Journal (giving you guys prestige and helping with recruiting new employees to you're company.... saying "come work for us where we invent cool stuff") or simply keep it as a trade secret (giving you a competitive advantage).
The largest problem I have with software patents is the business of prior art, where algorithms are patented that have already been invented or are trivial constructs that almost any software developer would have created given the circumstance. The "1-click purchase" button is an example of that.
BTW, I find that it isn't just software patents that are overly generalized but nearly all patents. This is also by design. In theory, the proper role of a patent is to record knowledge for future generations that would otherwise be lost. There are several devices and processes that we know about from history that simply weren't recorded in terms of how they were put together... or in the case of a metallurgical process what the steps were for making the items. A Stradivarius Violin is a prime example, and those are even still in use, as is something like a Damascus Steel. The problem with this philosophy is that I fail to see how the information given in a patent application can ever possibly be used in most cases to recreate the process.... even for somebody "skilled in the technology". I've looked at several software patents over the years and for many I would be at a loss in terms of how to recreate the algorithm that is being described. At best the patent description would only cover a class of algorithms like sort algorithms in general, not something specific like a Quicksort or Bubblesort.
but in all "fairness" if someone had the idea first then why shouldn't they get some benefit from it ?
Because it isn't "fair", whatever that should mean? Neither is it supposed to be the reason for the patent system in the first place.
The test for non-obviousness was supposed to make patents innovative beyond mere ideas, ie. full documentation of implementations which otherwise would be lost in trade secrets and obfuscation. However, non-obviousness tests are seldom used for anything else than delay a certain application, until it is reworded enough to be granted. This makes sense to the patent office and state, which earns Big Money from granting a mind-numbing number of patents each year. It also makes sense for huge mega-corporations, because they get defensive and offensive patent portfolios to squash lesser competitors with. It even makes incredible sense for patent-trolls, as they can push/buy up patents from dying companies, and extort money, without risking anything themselves, as they are producing nothing of value themselves, only sue successful businesses through courts out of the remains of dying businesses..
This all works splendidly on the cost of everyone else: inventors who are restricted in arbitrary fashion and customers who are forced to buy inferior products at exessive prices. It makes any business a risky operation, because at any time, you can be sued into oblivion, despite otherwise successes in the marketplace. Thus, the state monopoly-granted patent system works against the free market.
If it was "fair", then if someone has an idea, they shouldn't be sued into oblivion when implementing their idea as a business or "free software", just because someone "thought of it first", which is not even proven beyond any reasonable doubt. If everyone gets the same idea, or if the patent is just a physical process translated into the world of computers and software, then it shouldn't be patentable at all, since it is an obvious invention, a natural evolution of software to scratch an obvious itch.
Of course, only big corporations have the money to build a huge patent portfolio, and then use it as a defense mechanism, or even aggressively attack GPL, BSD, open source and free software. You can bet your sorry ass, Mozilla Firefox, Linux and most complex software out there, already violates hundreds of patents. It's just because of bad PR, the dogs have been kept in leash, but we remember SCO, and it is not far-fetched some dying corporation with real ownership of patents, could go for licenses instead of competing in the marketplace.
Just because nobody has patented it yet, doesn't mean nobody has thought about it. Just because nobody has started doing business around it, doesn't mean there are 20 competitors working on it already. Patents usually just gets in the way and squash the little inventor trying to do business themselves. They then have no recourse but to find a huge corporation sugardaddy to implement their idea at all.
Patents were never supposed to cover ideas themselves, but certain implementations thereof. The problem with software, is that copyright already protects software, so there if you're going to cover something more, you need to rape your constitutional forefathers, as USA, land of the "free", is doing now.
Evolution, will sort itself out though. USA will go bankrupt into its own corruption, greed, war-mongering and neglect of the environment and its own citizens. Somewhere, in the free world, some country will ignore software patents, and through that gain competitive advantage.
http://www.debunkingskeptics.com/
If something can be made in 1 week by a teenager, on no pay or salary, then it obviously is not worth protecting with hundreds of thousands in court fees to make greedy lawyers rich, at the expense of society at large and more pressing cases.
Software lowers the bar of innovation, so yes, nothing in software is really worth protecting. Software is already protected by copyright, which should provide sufficient protection, without hindering the free market to unfold itself.
There is a reason programmers in the field are called "code monkeys". After 4-5 years, if you haven't moved on or up, you are either a real geek, or just love mind-numbing work. There's usually not much innovation going on, just translation of real world processes into the world of computers. At any time, you can usually be replaced by another guy, don't fool yourself. Same with patents. It is not unusual for many people around the globe to get the same ideas at the same time, because the bar to software is so low..
http://www.debunkingskeptics.com/
So, extending this distrust to all government agencies is a natural thing to do, whether it is true or not. Try bribing a policeman - a large percentage of the time, you'll end up in jail. Try bribing a politician... oh, wait.
The patent system is made up of scientists and engineers, and the attorneys all have these backgrounds as well.
If that is the case, then why do they mess some patents up by not seeing prior art? Yes, mistakes happen and I suppose there is a process to work it all out. But while that process is rolling, the patent holder is out there drying up the coffers in attorneys fees of all potentially infringing users.
Are you saying software simply can't be inventive?
Not at all; software can be marvelously non-obvious, novel, inventive, useful, etc.
The problem isn't software per se, it is the software industry, and frankly, it's not really a problem, either.
Patents are intended to promote the progress of the useful arts. This is accomplished by encouraging inventors to invent, disclose, and bring to market, inventions which are useful, novel, and non-obvious, when they would otherwise not have done so, for the least cost in terms of restrictions upon the public. However, we know from history that some inventive activity will occur even in the absence of patents as an enticement. That, then, is our baseline. Whether or not any particular patent system is efficacious can only be measured by whether or not it encourages more invention, disclosure, and bringing to market than would occur if it did not exist (and remembering to take into account that in the absence of a patent, it needn't be the same person to do all three of those things), where the benefit of those things is not outweighed by the cost to society of burdensome monopolies. Likewise, alternative patent systems and reforms to the law may be compared by weighing them against each other in the same manner: the best is the one that produces the most public benefit for the least public cost.
This more or less works fine for many inventive industries. But software is an odd duck.
It seems very likely that the amount of invention, disclosure, and bringing to market that would happen in the field of software in the absence of patents is just as great, or perhaps even greater, than under the current patent system. That is to say, granting patents in this field may actually be harming the progress of this useful art. That's directly contrary to the purpose of patents. After all, there are great incentives to, and low barriers for, the invention of new software, and for bringing them to market. And often the interesting part of software is easy for those skilled in the art to discern without the disclosure requirement. The infamous One-Click patent tells you everything you need to know in the name, for example. Everything beyond that core idea are just implementation details that any PHOSITA can manage.
So why are people -- especially those in the industry -- opposed to software patents? It's not because they don't think software is inventive. It's because they think that the patents are a drag on the industry. That they're pointless at best, and actively harmful at worse. And comparing us to our foreign rivals who lack these patents seems to confirm this.
It isn't inevitable that this has to be so; perhaps someday in the future, natural incentives to inventors in the software field may decrease, and the added artificial incentive of patents may be very useful in keeping things going. But until that happens, we really ought to abolish patents in this field, since they are apparently not able to do their job.
-- This and all my posts are in the public domain. I am a lawyer. I am not your lawyer, and this is not legal advice.
I'm not going to criticize your knowledge of the patent system and process, because obviously, as a practitioner, you are obviously more knowledgeable than I.
What I will criticize is your bias as a legal professional.
What I find most interesting is that its biggest proponents [of getting rid of software patents] are people within the software industry itself
What I find interesting is that the biggest proponents of the current patent system are not industry professionals, but patent attorneys. I consider that damning evidence of who truly benefits from the patent process.
I think you're missing the perspective of industry practitioners, but not the ones you tend to meet in your career--the relatively small group who were told by their company to apply for patents on anything and everything they can think of. The vast majority (I claim) of software developers are not interested in "protecting" their toolbox of clever little three-line inventions, and just want to get on with be inventive without worrying about walking through a mine field.
The flaw is with a lack of a rigorous model for determining what is and isn't obvious; the difficulty of truly understanding, without hindsight bias, what the level of ordinary skill in the art is.
I definitely agree with this assessment, however you left out "conducting an honest and thorough search for prior art". Perhaps you can educate me about how rigorously companies search for prior art that, if discovered, would ruin their chances of profiting from their patent submission. Or how rigorously attorneys conduct these searches, that if successful, would cut short the process (along with their hourly fees). Or how thoroughly and carefully the patent office conducts these searches, with 3-6 year backlogs and pressure to cut those backlogs drastically.
I submit that we have the mess we do because there is no force anywhere along the process that would motivate any of the involved parties to deny a patent.
The patent system is made up of scientists and engineers,
I have seen patents which would call that into serious doubt.
What I find most interesting is that its biggest proponents are people within the software industry itself, but usually not the real innovators.
And how do you judge whether or not they're "real innovators"? Let me guess, by how many patents they've filed?
Are you saying software simply can't be inventive?
Hardly. In fact, the problem seems to be largely that patents are holding back inventions. From the patents which currently exist, there is simply no way for me to avoid infringing on someone's patent and also write any software at all, without an army of patent lawyers to dig through all those patents and tell me what not to write -- a process which would significantly slow innovation, if indeed I could write anything new at all.
As it is, the way large corporations seem to handle this is to file their own patents as fast as they can, so that when (not if) they infringe on someone, there's hopefully enough mutual infringement that they can work out a licensing arrangement. It is, in other words, a sort of software MAD.
What I find truly disgusting is that software patents, unlike copyrights, can be vague enough that I could legitimately invent something, in a cleanroom, without being aware of the patent I'm infringing, and be hit with a patent lawsuit for some mathematical truth I end up using. That's right, it's not just invention, but underlying principles of the universe which are protected here.
And that's a worst-case scenario -- but what if I want to support something for the purposes of interoperability? Consider H.264, and tell me HTML5 video is not innovative. Tell me Firefox is not innovative. But because of the patent issues surrounding H.264, Firefox refuses to support it, and certainly, if they were to implement it themselves -- even if they wrote their own decoder from scratch, without looking at any of the code for the official H.264 decoders -- they'd still be in violation. It's not terribly hard to find similar examples where, only because there weren't sufficient patents (or because companies chose not to enforce them), we have interoperability -- would OpenOffice be where it is today if the Microsoft Office formats were patented?
Stop for a moment and consider what the world would look like if web standards were patented. If HTML, CSS, JavaScript, and so on could only be implemented by those willing to pay a licensing fee to a central authority. If any third-party re-implementation of TCP/IP would likely result in a lawsuit. Would you really want to live in such a world?
Well, you're a patent attorney, so don't answer that. But why should any sane person who is not a patent attorney want to see HTML patented?
So the inventors can be rewarded... really? Do you really think Tim Berners-Lee has received no rewards for his efforts?
That you can't possibly think of something in software that anyone else couldn't have thought of, even given the exact same problem set?
That's sometimes a good idea, but not always. Because Adobe either doesn't have patents on PDF, or doesn't use them, people can send me documents and expect me to be able to read them with any of a half-dozen PDF readers I have, and I can expect to be able to work with them -- chop them up, rearrange them, print them, and so on. Certainly, there are other ways of solving the same problem, such as PostScript, but that doesn't decrease the problem -- if some people can read and write PostScript, and others can read and write PDF, we can't communicate, no matter how "innovative" one of them might be.
Also: Why would you want to force people to reinvent the wheel? One of the largest problems in software today is NIH syndrome, and you want to increase that? Seriously?
The reality is that we d
Don't thank God, thank a doctor!
The problem with your argument is that in many cases the government has given us a multitude of opportunity to be suspicious of it. There has always been a distrust of government in the US.
So, extending this distrust to all government agencies is a natural thing to do, whether it is true or not. Try bribing a policeman - a large percentage of the time, you'll end up in jail. Try bribing a politician... oh, wait.
I appreciate a healthy distrust in government. I in no way advocate trust in government, or in anyone or anything in particular. But there's a big difference between being skeptical and inflammatory without any reasonable basis. It's an issue I take to heart in the legal profession, when I see baseless accusations of fraud (or inequitable conduct in patent law, fortunately not first-hand) made without consideration of the damage it does.
And if the assumption is that everything the government does is tainted with fraud, then there's no hope you could ever do anything to fix it. Nor will the government ever be able to prove clean hands.
If that is the case, then why do they mess some patents up by not seeing prior art? Yes, mistakes happen and I suppose there is a process to work it all out. But while that process is rolling, the patent holder is out there drying up the coffers in attorneys fees of all potentially infringing users.
Well, first, there's no obligation to search prior art by anyone other than the examiner. Some very diligent clients do ask for a search, and the patents that emerge from that practice are usually downright bulletproof. But it's expensive, and with no other obligation not many people are willing to pony up. A start-up would rarely want to spend several extra thousand dollars to do a search.
Second, as a patent practitioner, it should be obvious that not being in the trenches day-to-day in the precise technical area of the inventors I work with (and the technology space is *vast*), I personally won't know off-hand if there's something else out there. Generally, the inventors would have the best idea of what's going on with the competition. And while there's no obligation to search, there is an ongoing obligation to the duty of disclosure, which means if anyone involved with prosecution of the patent (inventors, in-house counsel, myself, etc.) is aware of something relevant, it must be cited.
Third, there is a resolution process, and it's called reexamination. You don't have to litigate a patent to invalidate it, you can pay a fair fee, file your paperwork, and tell the PTO what they screwed up. This procedure has been available forever, and is vastly underused.
The patent office is working diligently on the quality problem, with pilot programs to obtain input from the general public, experts, and so on. Measures have been proposed to require searches. It's all on the table.
We all know what is and isn't broken. And even as a patent practitioner: I hear you guys and feel your pain. But, as with everything, there's a right and wrong way to go about fixing problems, and so it's frustrating to me to see the problem dismissed as *fraud* with a +5 vote on what should be a website for technical thinkers and problem solvers.
Tim Berners Lee? John Carmack? DONALD FREAKING KNUTH? These people aren't innovators? REALLY?
Hell, if anything, software shouldn't be patentable because Knuth probably already published the damn algorithm 30 years ago!
No, what I'M saying is that math shouldn't be patentable. Anyone who says otherwise is probably a patent attorney or an MBA, not a Computer Scientist.
Stop expecting me to support a system that directly threatens my very livelihood. Carving up portions of a domain of math and saying that use of them without a license fee is illegal is disgusting. It's even gotten to the point now that MPEG-LA triumphantly claims that no one can create any video codec without infringing upon their patents. How is this encouraging and protecting innovation? It's actively preventing it!
Maxim: People cannot follow directions.
Increases in truth directly with the length of time spent explaining them
Well, setting aside the goal of disclosure, I think that if no software patents were granted, you'd see just as much, if not more, inventive activity and bringing to market. The natural incentives in this field are so great that there's no need for the artificial incentive of a patent, especially given that the negative effects of patents would probably harm the software field, no longer being outweighed by their positive effects.
Disclosure is desirable, but I suspect that in most cases, software is easy enough for a PHOSITA to inspect and reverse engineer that the benefits of disclosure by the inventor would be fairly minimal. (Especially if we were to make much-needed copyright reforms that mandated putting complete and well-documented source code and other supplementary information in the Library of Congress as part of a revitalized deposit requirement) The one-click patent wasn't obvious except in hindsight, IMO; I raised it because it is a good example of disclosure not being enough of a reason by itself to continue to have software patents.
So if you'd have invention, and bringing to market anyway, and disclosure largely takes care of itself (along with some copyright reforms), what benefits are we getting from software patents that we couldn't have otherwise? Likewise business method patents. I realize that this might reduce 'breathing room' for small inventors in the software field, but I think that the advantage of being first to market, combined with trade secrets, NDAs, etc. to protect against unscrupulous business partners, is probably adequate. In any event, I'm prepared to take a chance on it.
-- This and all my posts are in the public domain. I am a lawyer. I am not your lawyer, and this is not legal advice.
Self-interest is natural. It's totally understandable to want to protect one's profession. It makes total sense that by and large, the only vocal supporters of our patent system are attorneys and large firms that hold war chests of patents.
As a software professional, patents are detrimental to my livelihood. I oppose them pretty much across the board (including non-software patents). It's a self-interest-based position as much as it's a moral position. People should be allowed to innovate and invent without fear of walking through a mine field. You don't deserve exclusive right to make something just because you happened to file some paperwork first. These are opinions based on my morals and based on my desire to be able to go to work every day.
If patents disappeared tomorrow, people won't all of a sudden stop inventing things. The motivation is always there. There's so much profit in bringing to market novel inventions--with or without patents. As evidence look at the number of profitable companies that hold no patents. What WOULD stop overnight is companies using their war chests and lawyers (rather than their smarts) to bully other companies out of the market. And before you suggest it, I believe the societal "disclosure" benefit of the patent process is overblown in this day and age. Society no longer needs a government patent list to figure out what the latest advances in science and technology are and build upon them.
You mentioned earlier that the you are puzzled as to why so many people within the software industry itself want to do away with software patents followed by an unsupported dig at those obviously uninventive and uncreative people.
how the answer isn't obvious to you boggles the mind but lets assume you've never made the slightest effort to understand the culture in the software industry nor made the slightest effort to understand the nature of writing software.
the culture was shaped in decades past by hordes of the kinds of people who are willing to sit on their own reading manuals all night long, this leads to a certain bias towards the little guy
1:
Anyone can do it, that's one of the best things about software.
if you have a working brain you can create useful software.
Most of the time no factories are needed, no massive capital, just enough for a cheap laptop and cost of living for a few months.
Patents being stupidly expensive to obtain are as such nothing but a barrier to entry to the small time programmer.
I could live for a year on the cost of pushing through a handful of patents.
2:
It's utterly impossible to avoid infringing on at least some patents if writing any large piece of software and unless you happen to have a legal department and millions of dollars there's no way to be sure.
And I can never know for sure, if I create something useful and give it away free out of the goodness of my heart or sell it I could wake up to a lawsuit that could cost me everything I own.
And there is absolutely nothing I can do to protect myself other than to not create useful things and not sell them or give them away.
Does the patent office offer any system where for a fee that wouldn't cripple a normal person working a normal job I can submit my code to be compared to existing patents and receive a list of all patents I'm infringing and also receive complete immunity from any claims from the owners of patents not listed?
If it does I'll happily remove this complaint.
If it does not why am I expected to be able to do what the patent office cannot do itself?
3:
Programming attracts maths geeks to whom programming is merely an extension of mathematics, the mere idea that you can patent doing a certain type of calculation is absurd to them.
4:
At the other end of the scale programming attracts artists (who can oddly overlap with the above) who view a piece of software like a piece of music, to them patenting a particular algorithm is as absurd as patenting writing a piece of music in 4½/4 time.
5:
The software industry is distributed.
In a centralized industry like auto-mobile manufacture where there's a handful of really big companies, expecting each one to have a legal department which can wade through the recent patents in related areas is somewhat reasonable.
Expecting every developer in the world to keep track of every patent is absurd.
Especially when they can get fined triple the amount for trying to do that kind of research themselves and failing.
which effectively forces you to work for someone else if you're poor or hire a legal department if you're not.(which also fosters feelings that it's just corruption and lawyers making utterly useless and wasteful jobs for other lawyers)
and many many other reasons which other slash-doters will berate me for not including.
Your snide and insulting remark that the programmers who complain about software patents are the uninventive ones only displays your complete and utter ignorance of the issue.
Some of the best and brightest minds in the industry including Knuth http://www.pluto.it/files/meeting1999/atti/no-patents/brevetti/docs/knuth_letter_en.html (who quite literally wrote the book on algorithms) and John Carmack( Ask a gamer if you don't recognise the name, recently moved into rocket science after reaching the top of the field in graphics engine programming and getting bored) are opposed to patenting algorithms and software.