SoylentNews is people
SoylentNews
Fluffeh writes:
"In June, President Barack Obama called for action against patent trolls. Today the White House held a short conference updating what has happened in the arena of patent policy since then and announced new initiatives going forward — including one to 'crowdsource' the review of patents.
Currently, getting a patent is a one-on-one proceeding between the applicant and the examiner. Two pilot programs that allowed the public to submit prior art were only applied to a tiny number of patents, and in the first program, all the patents were voluntarily submitted by the applicants. Applying such scrutiny to a few hundred patents, out of the hundreds of thousands issued each year, isn't any kind of long-term solution.
Unless the crowd-sourcing initiatives were to put major new burdens on applicants — which would be resisted — the fundamentals of patent examination aren't going to change. Patent examiners get an average of eighteen hours to review a patent. Most importantly, examiners effectively can't say 'no' to applicants. They can reject a particular application, but there's no limit to the number of amendments and re-drafts an applicant can submit."
(Score: -1, Offtopic) by Anonymous Coward on Saturday February 22 2014, @10:52PM
Thank you for being a friend
Traveled down the road and back again
Your heart is true, you're a pal and a cosmonaut.
And if you threw a party
Invited everyone you knew
You would see the biggest gift would be from me
And the card attached would say, thank you for being a friend.
(Score: 5, Funny) by mrcoolbp on Saturday February 22 2014, @10:54PM
Ug, with our luck it'll turn out like reddit:
"And today 947,434 patents were accepted regarding cats and cheezburgers"
(Score:1^½, Radical)
(Score: 5, Insightful) by maxwell demon on Saturday February 22 2014, @10:54PM
A simple solution to the last issue: Make resubmission/amendment a service to be paid for. The first resubmission should be relatively cheap (say $50), because after all there may be legitimate reasons to resubmit. However for each further resubmission, the price should double.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 5, Interesting) by regift_of_the_gods on Saturday February 22 2014, @11:29PM
Some random thoughts, I'm sure these aren't original.
1. The term for software patents should be cut in half, to 10 years instead of 20. As we were just reminded with WhatsApp, ten years is more than long enough to make a fortune starting from scratch in today's business world.
2. The law should encourage crowdsourced review after someone has been sued or threatened with a lawsuit for infringement, with the plaintiff picking up a reasonable tab for an incentive reward if prior art is found.
3. Whenever a patent is sold, the term should be automatically shortened by a few years. The patent system was designed to protect the inventor to allow them to realize their invention, not provide a financial asset that can be bought and sold like stock certificates.
4. Damages should be capped if it can be demonstrated that the earliest date of infringement occurred before the patent holders had put forth a serious, de facto effort into realizing their invention in a commercially available product or service. In that case, either the invention was evidently somewhat obvious, or the filers were stockpiling ideas so they could collect their tolls.
(Score: 5, Insightful) by frojack on Saturday February 22 2014, @11:51PM
Did WhatsApp actually have any patentable features?
After all, the capabilities it touts have been available in multi-platform multi-network applications since Jabber gateways, and there are 20 of them in the app store on Google or Itunes right now.
Software should probably NOT be patentable at all. (And I say this as a guy who makes his living writing software). Copyrighted, sure, but If I want to write a better WhatsApp, its just a a bunch algorithms.
No, you are mistaken. I've always had this sig.
(Score: 4, Insightful) by keplr on Sunday February 23 2014, @12:00AM
Software patents need to be banned outright. Grandfather in the ones that already exist, lock-in their time limits, and stop issuing new ones. Software is math, or logic encoded as math. Allowing software to be patented essentially allows certain very long numbers to be deemed proprietary. That should be absurd to any thinking person.
Software patents are doing more harm than good to our society and the advancement of science and the useful arts. There is arguably *some* good, but it's not pulling its weight against the problems it's causing.
I don't respond to ACs.
(Score: 4, Interesting) by dmc on Sunday February 23 2014, @12:37AM
While I agree with your second 'graf of SW patents currently doing more harm than good, I disagree with your ad hominem argument here. The movie Toy Story is every bit as representable as a very long number, and I believe it should be allowed to be proprietary (to a limited, fair use limited, and after a reasonable time public domain bound extent). You say SW patents should be "absurd to any thinking person". No, you are doing a disservice with that ad hominem to the legitimate critique of the current situation. A thinking person can compare the sorts of ingenuity that went into the steam engine, the cotton gin, and velcro and say- Some clever dude took existing materials and processes, and made them do something profoundly more useful than they were before. And it makes sense to reward that person by giving them a *reasonably limited* monopoly on the ability to do that. Likewise with software, in *the information age*, much of our enhancements in living these days come from clever dudes (of all genders) taking esisting materials and processes, and recombining them to do something profoundly more useful than they were before. And it makes sense to reward them as well- again- in a reasonably limited way that eventually leads to a rich public domain of inventions and art.
Agreed, but I think that contrasts greatly with what you said should be "absurd to any thinking person"
(Score: 5, Insightful) by keplr on Sunday February 23 2014, @01:04AM
Your argument seems to hinge on the conflation of patent and copyright law. Copyrights, patents, and trademarks are three very different branches of the law. Some of the differences are subtle and nuanced but throwing them all together, usually with the umbrella term "intellectual property", makes you an unwitting accomplice to the MPAA et al whose interest it is for the worst parts (from the individual consumer's point of view) of all three systems to become common to the others.
They would love for you to treat patents like copyrights, both in your mind and in the law, because it primarily benefits them. It benefits society to keep these ideas distinct.
I don't respond to ACs.
(Score: 2, Interesting) by dmc on Sunday February 23 2014, @01:16AM
I disagree. I did mix patents and copyright, because you had already gone ahead and used the term "proprietary" which applies to both, and the digital film as represented by a long number was I think the easiest way to illustrate my point. If you look at the world as something of a 'Matrix', or close enough to it, any traditionally patentable process, e.g. how to make velcro, can also for the sake of how we are discussing it, be represented as a very long number (e.g. describe it in detail in ascii, or make a program that generates an atomic/voxel rendering the process etc). I do believe that the general idea of proprietaryness that is shared between copyright, software patents, and traditional patents, is basically all about the same thing "intellectual property". I think we probably agree more than your response suggests. I think IP rights should be limited much more than they are, fair-use and public domain valued highly. I very much liked another posters idea about the limits on the monopoly granted degrading with each resale. But yes, patents and copyright are distinct, but they are not as entirely distinct as your argument suggests. It almost sounds like you are getting on my case because any sympathy with 'the other side', which I agree, that we both agree, is doing heinous things... I remember the $222,000 judgement against the soccer mom. I see all the patents, the vast majority software being used to throttle innovation more than encourage it. I see all that. But I think you are entrenching as opposition too much. I think by denying the foundational value of them *completely* does a disservice to the progress I think we would both like to see. (seriously, reread your "absurd to any thinking person" sentence until you realize you were 100% guilty of an ad hominem argument...)
(Score: 4, Interesting) by dmc on Sunday February 23 2014, @01:40AM
To critique your argument against SW patents as "patenting long numbers" in another way-
Some might talk about the absurdity of "patenting math". I would respond that SW patents are patenting "applied math" in the same way that traditional patents (e.g. combustion engine) are patenting "applied physics".
(Score: 3, Insightful) by keplr on Sunday February 23 2014, @01:57AM
Applied physics is engineering. You don't get to patent or copyright engineering formulae and procedures, but you might be able to patent the fruits of your labour which used those tools. Software is like the formulae in this case. So you can restrict access to the fruits of your labour (with copyright), like MS Windows, but the software shouldn't be patentable. But this is getting into car analogy territory. It's best to just speak of things in their own terms.
I don't respond to ACs.
(Score: 2) by dmc on Sunday February 23 2014, @02:09AM
No, you don't get to patent the fruits of your labor, you get to sell them (e.g. combustion engines). The original design for a combustion engine or velcro, is applied physics and chemistry. If you figure out the physics and engineering formulae for a new fusion reactor, you get to patent the _applied use of those physics chemical and engineering formulae_. You don't get to prevent other college professors from teaching how it was done (though you can try as a trade secret alternate to the patent process), but you do get to prevent those professors and their students and friends from recreating the saleable output. From that logic, I view a software patent as valid in the sense that two computers, one with the software and one without, are distinct physical objects, just like two differently designed car engines are. And that is why software should be patentable. Now, continuing that logic, I may also agree that we can just throw out all laws that talk about 'software patents' and just interpret tradtitional patents as covering the case where a computer running different software is a new, novel and different invention than the computer before it's internal atoms and electrons were rearranged in a new and novel and valuable way.
(Score: 1) by bucc5062 on Sunday February 23 2014, @01:04PM
I don';t agree with the original premise that SW is just a long number and how can we patent that? IT does fall into your point about Toy Story (the movie) as a digital work being just a long number. However, I still agree that SW patents need not to be given out.
A program is not so much a big long number as it is a connection of ingredients, taken from a general pool of syntax, semantics, and algorithms. Key words (even at the assembler level) and actions rung together to form a final product. Can one patent a cake? a pastry? You make one from apples, I make one from berries, but they are from the same pool of but knowledge and foundations.
We do not patent stories. Why? Some author writes a book that seems "completely" new, yet it is not for stories are built from the same building blocks, the same story lines laid out centuries ago. A story, article, poem; these are just a collection of letters, words, sentences, formed differently to make something "new", yet still not new.
Thus why I feel patenting SW is wrong (does not make sense). If I write a handy program that seemingly know one else knows about and someone else write a program that does almost the same thing, both came from the same pool of ingredients and were created just by different minds. Patenting an object, this I can understand for the original intent, to protect a small inventor from being shut out by larger competition. In the digital age, that scenario is less likely when it comes to SW. Copyright, this I understand and to copyright a program, just as we copyright a book or movie does protect the originator from losing their creative effort. I cannot copy a Monet, but if I paint like him, and I paint a lady in a field with a little boy that looks similar, I should not have to worry about copyright (I created it myself) nor patent, I used my own ingredients to bring it to life.
One day society will understand that programming is Art, not engineering.
The more things change, the more they look the same
(Score: 2) by dmc on Sunday February 23 2014, @07:36PM
I don't disagree with your description, but I think the answers to your questions both are (and IMO should be - in a sense to be elaborated on) *YES*. Yes, you can patent a new and novel way to make cakes or specific cakes (or pastries). The reason you don't see these things patented meaningfully today is because they are so old, that the original inventors are long dead, and the processes that you are thinking about have now properly graduated to the public domain. That said, I'm sure Hostess probably still has some patents out for how they make twinkies and their other unique cakes.
To further discuss your point- yes, these patents are absolutely all conceptualizable as "recipes". They are "recipes" for how to build things. I.e. the patents for combustion engines, or velcro, are simply recipes for using existing widely known and common manufacturing processes to develop a new sellable object that is novel and a valuable addition to society. Likewise Software is just a "recipe" of how to arrange electrons and whatnot inside a common computer, to do something new and novel and valuable. In fact, as we enter the new revolutionary age of 3D printing, it will become more and more obvious how patents (for traditional physical objects) really should be thought of as "recipes". And in fact, that will absolutely blur the line between "software patents" and "traditional patents". Because now it will simply be a matter of a new novel software program that generates a new sellable innovative physical device.
I'm not going to bother doing the research, but I think you are likely wrong here as well. I suspect that some innovative types of stories have been patented. Certain "formulaic" stories have probably been attempted, and I'd even somewhat sadly suspect successfully, as either a class of story, or of a *cough* business method.
Now don't misinterpret me, I'm giving a defense of the foundational value of why such Intellectual Property rules and laws make sense. But I beyond wholeheartedly agree that in the actual practice of our society, these rules and laws have utterly failed beyond measure to achieve their intended benefit. When I look at the problems of the world, like corruption, past and modern slavery, oppression, human right violating surveillance, the war on drugs, the war on terror, the US history of interfering and bringing down legitimate democracy in favor of puppet dictators, then yes, it becomes ludicrous to defend how the rich profiteering scumbags have used Intellectual Property to keep themselves entrenched in power, at the cost of the hunger of the victimized masses.
But all that said, it is important to remember that the idealistic *foundations* of Intellectual Property are not bad. And when the opposition to the forces I mentioned above, become so entrenched, that they start ignoring *reason* it does their camp no good in the long run.
The issue here is that copyright and patent laws need to be *reasonable and effective*. First, Monet is long dead, so protecting his ability to profit from a specific work clearly doesn't make sense. But if a painter today, made a profoundly interesting painting, and profited from copyright to it, then there are limits of how closely you could copy that piece. Now, if a reasonable person believes that you just happened to paint a similar thing, or even were fairly using the fact you had seen the prior work and it affected your artistic expression, well, then society certainly shouldn't punish you, or if they do, punish you in a way commensurate with the actual damage to society you did (e.g. less than speeding 5mph with a car).
Bottom line- I think Intellectual Property has a valuable idealistic foundation, but in practice has been utterly perverted away from something *reasonable and valuable* to society, when looking at the bigger picture of actors that bring harm to society in various ways. So if we threw out all patent and copyright law today, I'd shed not many tears. But if everyone gave up on the belief that there were good intentions of wanting to encourage and help idea-generating-individuals make a living as valuable members of society, then I'd be pretty sad.
(Score: 2, Interesting) by akinliat on Sunday February 23 2014, @12:09AM
And don't forget:
5. Software should only be eligible for patent, not copyright. The patent should only apply to working code.
I've never understood why software is eligible for copyright. It's a set of instructions for a machine -- almost exactly what the industrial process patent category was designed for. At least it seems that way to me.
(Score: 5, Insightful) by frojack on Sunday February 23 2014, @12:32AM
You have that backward.
COPYRIGHT applies to MY CODE, stuff I wrote. You can't steal and Publish My Code, unless I release it under the GPL or similar license. Its a work of art. (mostly anyway, see my sig).
I can copyright my code for my accounting system without hindering you from writing your own accounting system, even copying some or all of the functionality of mine, because the accounting system is nothing but mathematical algorithms and logic. But you have to write your own programs.
You can even read my code, (unless I keep it a trade secret), but you can't re-publish it.
If I write a book on Math, its all math, and not patentable. My book, never the less is copyrightable.
On the other hand, had I managed to get a PATENT on computerized accounting systems, that is far far worse, even if it only lasts for 17 years. Because then I can prevent you from writing your own.
No, you are mistaken. I've always had this sig.
(Score: 3, Informative) by jcd on Sunday February 23 2014, @03:24AM
This is a good explanation of the differences. Wouldn't copyright protection alone still create the potential for abusive, extensive, expensive lawsuits?
Though I suppose that a court could just diff all of the source code to see if there's a straight copying issue.
Haha, I made a funny. As if courts have any technical capabilities.
"What good's an honest soldier if he can be ordered to behave like a terrorist?"
(Score: 1) by akinliat on Sunday February 23 2014, @05:22PM
But that's the problem. Your code is not a work of art. This is not a comment on its quality, BTW, simply a statement of fact. Art communicates ideas between people. Code is a set of machine instructions that make it easier for someone to manipulate hardware for a specific task. It happens to use words (sort of, not really), so the techno-illiterate USTPO decided it ought to be copyrighted. Ultimately, though, code is almost the very definition of an industrial process, which is one of the broad categories of patents.
This is probably one reason the current application of software patents is so flawed and vague. Because copyright already covers specific instances of code, the USTPO uses patents far more vaguely than warranted. At most, a patent should cover an algorithm as expressed in working code. Not only would this abate the absurd longevity of current code copyright (seriously, how useful is 75-year-old code likely to be?), but it would also mitigate many of the problems with the software patent process as it now stands.
(Score: 2) by mcgrew on Sunday February 23 2014, @06:29PM
Your code is not a work of art.
True, although it could be made to be art. One could make a poem in a computer language that would be recognizable as poetry by a human reading the source code, yet still run. It would be a hell of a hack if someone did it.
Code is not art in exactly the same way that blueprints are not art and the instruction manual to your stereo is not literature.
mcgrewbooks.com mcgrew.info nooze.org
(Score: 1) by akinliat on Sunday February 23 2014, @06:41PM
It could be made art, at which point you could certainly apply for copyright. Heck, even if you couldn't run it, using programming languages to write poetry would certainly make for an interesting means to comment on the Information Age.
Not exactly. There is one crucial difference. Blueprints and instruction manuals are made to read by humans (some instruction manuals, anyway). Although code can be read by humans, that's not why you write it. You write it so that it will be read by machines. The only real reason to make it human-legible is so that you or someone else can fix it if the machine has problems.
If I make it sound that we are merely servants to our wee silicon overlords, so be it. I've felt that way often enough when coding.
(Score: 2) by frojack on Monday February 24 2014, @12:27AM
WHY you right it never enters into the discussion of what is and what is not art.
The idea that code is not meant to be read by humans if just not true. If it wasn't meant to be read (and maintained) by humans, we would all write it in binary 1s and 0s. Instead we use languages, invented by and for humans, which some machines have the ability to carry out the instructions there in.
By work of art I meant Art in the sense that literature is art. Music is art. Sculpture is art.
Music is the closest equivalent to Code. Sheet music is the closest thing to instructions for some other entity (human or synthesizer) to perform a task (make sounds) in a specific order.
In short, I don't believe you get to preempt the definition of a work of art SIMPLY because it can be read and action-ed by a computer. I believe that has already been done:
I'm sure you would agree that code is "Writings".
So I don't understand your current nit-picking here. The Constitution was meant to be a
. It was never intended to be omniscient, or predict future methods.
No, you are mistaken. I've always had this sig.
(Score: 3, Interesting) by mcgrew on Sunday February 23 2014, @06:24PM
On the other hand, had I managed to get a PATENT on computerized accounting systems, that is far far worse, even if it only lasts for 17 years. Because then I can prevent you from writing your own.
That's not how patents work. My ex-brother in law worked at a factory, and often a manager would bring a competitor's gizmo and ask "can we make these with what we have here?" Porky asked about patents the first time ge got that question, and was answered "that's why we have lawyers. Sometimes you can get around a patent simply by making it out of a different material; say, brass instead of aluminum."
mcgrewbooks.com mcgrew.info nooze.org
(Score: 4, Insightful) by frojack on Saturday February 22 2014, @11:36PM
If examiners can't say NO, then there would be no such thing as a rejection.
So clearly they do say NO quite often.
A lot of patents are applied for and on average only half are granted [uspto.gov].
(Utility Patent Grants, All Origin Total / Utility Patent Applications All Origin Total * 100)
Its easy to see that if the rejections were all re-filed every year, within a few years there would be nothing but refiles in the queue.
Filing for a patent requires you to provide all prior art and explain how your invention does not violate that. This alone weeds out a lot of patent applicants, because they, or their patent lawyer, figure out that the effort will be fruitless in the long run.
A rejection is hard to overcome, and often when they write up their re-file, what remains of their claim after working around the infringements or prior art isn't worth the effort.
Only the biggest firms (often pharmaceuticals) have the resources to do so.
No, you are mistaken. I've always had this sig.
(Score: 5, Informative) by frojack on Saturday February 22 2014, @11:47PM
Oops, I reversed the math. Oh, well you will figure it out.
Still less than half of all applications are successful dating back to 2001.
Prior to 2000, more than half were successful.
The biggest drop came in 2005, the year the "Obviousness" [wikipedia.org] standards went into effect. Successful applications have never reached the 50% mark since.
No, you are mistaken. I've always had this sig.
(Score: 5, Informative) by moo kuh on Saturday February 22 2014, @11:47PM
At least the POTUS is willing to admit that the current US patent system has problems. I took a quick peak at what kind of requirements and pay grades patent examiners are at. Check out the below link if you are interested in the details, but pretty an examiner fresh out of undergrad with a hard science or engineering degree only starts at GS-5. GS-5 step 1 is only $31,628/year in most US localities. A GS-5 in the Washington DC area, which has a rather high cost of living compared to most of the US, starts at $34,415. Sure patent examiners with a few years experience and/or graduate school get a few pay grades higher, but even a GS-7 in the DC area is only making in the range of $42,000 - $55,000. Most private sector engineering and science jobs pay twice that in the US. I have included a few addresses for those that want to fact check me.
http://www.uspto.gov/web/offices/pac/exam.htm [uspto.gov]v e/salaries-wages/2014/general-schedule/ [opm.gov]
https://www.opm.gov/policy-data-oversight/pay-lea
http://www.glassdoor.com/ [glassdoor.com]
My point is, unless the USPTO is lucky enough to get people that really, really, want to be patent examiners (which I doubt, but could be wrong), they aren't going to attract top talent at those pay rates and with that type of work. Personally, I would much rather be the person (or employee that worked on the project) applying for the patent than examining it. The money is better, and I would get the satisfaction of knowing I invented something. Why would I make less than I can doing work that is relatively boring in my field? I think if the POTUS wants to do something about bogus patents, he should yell at congress until they come up with the money to pay a lot more than the typical engineers salary.
(Score: 1) by frojack on Sunday February 23 2014, @12:05AM
I'm not sure paying them more is the solution.
Typically with government employees, the more you pay them the less ACTUAL work you get out of them.
You want to be able to afford hundreds of them, not a few well paid prima donnas.
34K may be a tad low, but they all require only a bachelor's degree, until you get to the GS9s or GS11s.
No, you are mistaken. I've always had this sig.
(Score: 3, Interesting) by TheloniousToady on Sunday February 23 2014, @03:57AM
I'm no expert in this, but IIRC, many patent examiners want to be patent lawyers. Looks very good on a resume, kindda like an internship or perhaps even like a doctor's residency. If that's true, the low pay fits right in.
(Score: 5, Interesting) by me on Sunday February 23 2014, @12:35AM
Patent approval needs to be a two step process. At present (at least to my understanding) the application is kept secret until it is approved. Once approved, it is difficult to overturn it.
The patent office, on deciding that a patent should be approved, needs to publish a provisional approval for a period of time, allowing the public an opportunity to present a defense against the patent being approved.
This would allow presentation of prior evidence, by the public, that the patent office researchers have been unable to uncover.
There would then be a secondary review after a fixed time to determine whether the patent should be granted, or rejected.
Now, I realise that the big corporations would overload the patent office with claims against any of their competitors patents (apple vs samsung for example) - but it has to be a better model than we have now.
(Score: 4, Informative) by hankwang on Sunday February 23 2014, @01:03AM
The application is actually published 1.5 years after filing (USPTO); see for yourself on Google patents [google.com]. The process of approval often takes much longer than that, 3--4 years does not seem to be uncommon. In theory, a third party would be able to submit prior art during the time between publication of the application and patent issuing, although I don't know how it works in practice.
Avantslash: SoylentNews for mobile [avantslash.org]
(Score: 4, Funny) by L.M.T. Spoon on Sunday February 23 2014, @08:11AM
In other news, a new pilot program just put the patent review process on Twitch to see what happens: the results are mixed, but some surprising progress has been made. At the moment, the committee is trapped standing on a table, trying desperately to revive the skeleton of Thomas Edison. The hope is that he will teach us more about how to steal ideas and circumvent law, but still remain widely respected.