Petitioning The Government Against Software Patents
from the yes,-but... dept
Apparently I had missed that the White House set up a platform for people to petition it directly. A few folks have sent over a petition that was set up asking the administration to reject software patents, noting that they are hindering the software industry, one of the few "strong" industries in the US. The specific petition asks the government to stop issuing software patents... and to void all existing software patents.While I understand the sentiment... there's a big problem here that plenty of patent folks will point out: there's no good definition of "software patents" right now. If you just say software patents are banned, what will it mean? Chances are it will mean that patent attorneys will start writing "software patents" in a way that don't look like "software patents." I think that patents that end up covering software are a big problem, but I don't think people are doing themselves any favor complaining about software patents if they can't explain to the government what a software patent actually is.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: patents, software patents, uspto, white house
Reader Comments
Subscribe: RSS
View by: Time | Thread
Not well defined
My personal favorite right now is the Abolish the TSA petition. It's succinct but accurate and seems to get the point across.
[ link to this | view in thread ]
We hired them
[ link to this | view in thread ]
[ link to this | view in thread ]
Ok - then ban all patents.
That was easy
[ link to this | view in thread ]
[ link to this | view in thread ]
[ link to this | view in thread ]
Re: We hired them
[ link to this | view in thread ]
Get ting them to define and exclude software was the easy part.. getting them to actually pass the damn bill seems to be a lot harder. This one was already on the table, but they decided to pass a new three-strikes copyright amendment under urgency instead.
[ link to this | view in thread ]
Re:
I'm sort of curious who the "you guys" you refer to might be?
[ link to this | view in thread ]
So then Mike why don't we define it as "patents that end up covering software"?
Because enforcement is all that really matters. You can have the office issue a gazillion patents that don't look like software but would "end up" covering it. So long as the gov. makes an easy way to have a patent judged to be or not to be such a patent then you're golden.
[ link to this | view in thread ]
[ link to this | view in thread ]
Signed and Posted
[ link to this | view in thread ]
Re: We hired them
[ link to this | view in thread ]
Re: Signed and Posted
All programming code(each individual instruction) has a equivalent mathematical formula.
Ex 1: X=Y+1 ---------becomes-> X=Y+1 ....Yes I know... Obvious, but still
Ex 2: For Loops -----becomes-> Mathematical Induction
Ex 3: If A Then B ----becomes-> A ⇒ B (yes this is actually not 'quite true' but it is close)
And there are more here: List of Math Symbols, So I think it would be easy to define software patents... IF we had a patent categorization system in the first place(and we still need it)
[ link to this | view in thread ]
Also the government already has a categorical system...
http://www.loc.gov/catdir/cpso/lcco/
Now... it isn't GREAT... by any means... b/c pretty much everything to do with Computers/Computer Science is a sub-sub-sub category... BUT... it WOULD allow for proper categorization of patents (most would remain empty... like Law, but still they would be there if needed.
PLUS... Librarians work with this system already... so there is a body of people who have mastered this system... can't say so much with "all" of patents.
(I'll stop recursively commenting now)
[ link to this | view in thread ]
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Patentable Definition
This definition would also take care of 'business model' and algorithm patents.
[ link to this | view in thread ]
[ link to this | view in thread ]
Definition of bad patent
[ link to this | view in thread ]
How To Define Software.
A device shall be said to be a "Congealed Computer Program" if it can be functionally replaced by a general-purpose computer (or a network of computers) running a computer program or programs.
A general-purpose computer shall be understood to be equipped with any or all accessories which are in widespread use, and which are not mutually exclusive, eg. screen, keyboard, mouse, printer, microphone and speaker/headphone plugs, TV tuner card, modem and/or network card or WiFi. For purposes of computational feasibility, the computer shall be a million times faster than those currently in production (twenty years of Moore's Law). A reasonable complementary definition of "an accessory to a general-purpose computer" might be whether there exists a USB class library for the device in question.
A supplementary definition of "functionally replaced by a general purpose computer" might be whether the device can, in principle, be replaced in such a way that the computer inputs and outputs involve trivial quantities of energy. An engine, for example, is not a congealed computer program, because one can make program which shows pistons moving up and down on the screen, but it does not do real work, such as turning a shaft. A modem, on the other hand, is a congealed computer program, in the sense that one could cause the logic unit to emit a very small amount of power, which gets fed into an amplifier, and stepped up to line voltage.
A Congealed Computer Program shall be considered as software for purposes of establishing "machine or transformation." Returning to the modem, what you would have to do to get a hardware patent would be to exploit some hitherto-unforeseen property of the wires connecting the modems, or something like that.
I think this is a reasonably workable definition. It does not have to cover fundamental components, such as flip-flops or operational amplifiers, because those are deep in prior art, long before the patent horizon.
[ link to this | view in thread ]
Re: How To Define Software.
The best definition of software uses computation theory, much like TechnoMage did in his post, but more formal, using equivalence to a computation model.
Groklaw had a few epically long posts explaining this in detail. Here they are:
* An Explanation of Computation Theory for Lawyers
* An Open Response to the USPTO -- Physical Aspects of Mathematics
* Why Software is Abstract, by PolR
* 1 + 1 (pat. pending) — Mathematics, Software and Free Speech
* A Simpler Explanation of Why Software is Mathematics by PoIR
[ link to this | view in thread ]
Govt Branch
[ link to this | view in thread ]
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Re: Patentable Definition
If it can't be "translated" into a "blueprint", then it's not patentable.
AND the patent only covers that specific "blueprint", not other things that looks like it, leaving the door wide open for independent inventions.
[ link to this | view in thread ]
Rather than ban s/w patents
The arguments in Europe over s/w patents (and the shenanigans of those who have attempted to evade the ban) do show that Mike is correct here - so moving the definition of s/w from the arena of the patent itself to that of the alleged infringement may be a better route.
[ link to this | view in thread ]
another biased article
The Constitution says “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”. Therefore, if an invention is useful and promotes science, it should be patentable. It’s that simple.
[ link to this | view in thread ]
All business methods are algorithms. A rose by any other name would smell as sweet. It doesn't matter what you call it, it's still an algorithm. And unpatentable.
[ link to this | view in thread ]
Re:
It is sad that you live in an us versus them world.
[ link to this | view in thread ]
Re: Re: Patentable Definition
While we are at it, limit all IP to 7 years, no exceptions.
[ link to this | view in thread ]
Re: Re: Signed and Posted
[ link to this | view in thread ]
Re:
I can't speak for everyone, but I can say with a certainty that I do. I think almost everyone is aware that congress is the entity that writes and passes legislation in this country. However, to "talk to congress" on this issue would require contacting 538 different offices.
If the president asked congress to "take up bill XYZ", then it immediately becomes part of the national conversation. The president is a shortcut to call all 438 representatives, and 100 senators in one petition. As Aaron Sorkin puts it in the The West Wing the president is "a human starting gun".
So, yes, I think we are talking to the correct branch of the government.
[ link to this | view in thread ]
Re: another biased article
Not quite... if the PATENTING of that invention would be useful and promote science, it should be patentable. The argument about which patents promote science and which hold it back is the point here.
[ link to this | view in thread ]
Is software patents really the issue?
If I have invented a widget that does something useful, novel and non-obviousness to a skilled practitioner in the art, why does it matter if it is build with gears, levers and springs or with a machine that includes software or firmware?
Software is not an algorithm ... it implements an algorithm. So do mechanical machines. Take for example US patent # 6,183087 B1 granted on 6 Feb, 2001.
http://www.google.com/patents?id=xQwFAAAAEBAJ&printsec=abstract&zoom=4&source=gbs _overview_r&cad=0#v=onepage&q&f=false
From the abstract:
"A driving mechanism such as a Geneva Mechanism which allows a load, such as film in a motion picture projector, to be intermittently advanced with a controlled acceleration."
I leave it to you to read the rest but it would be hard to deny that this machine implements are rather simple algorithm.
The standard Geneva Mechanism is not new ... they reference it as prior art. What they did was make some small changes to the shape of the star wheel and drive pin to reduce the intermittent motion time, reduce the load on the mechanism and the load. (From the Summary of the Invention" section.)
Another way to put it is, they changed the algorithm the device implements.
It is the same as a car transmission ... it isn't an algorithm but it does implement an algorithm. If I invent a way to make car transmissions better are you saying that if I implement this improvement with machine parts it's patentable. But if I can get the same result by microcomputer (with software in it) hooked to the transmission then you're all saying that it should not be patentable?
If that's what you're all contending then I disagree.
If you're saying that patents that are vaguely defined and be construed to cover a broad range of things the original inventor never intended, then I might agree.
[ link to this | view in thread ]
http://www.plotpatents.com/
Did this guy ever get his patent?
The argument for patents upon written material is absurd.
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
[ link to this | view in thread ]
Re: another biased article
Software commands what a machine or person can do, we don't patent protocols, we don't patent business practices and we don't patent laws do we now?
[ link to this | view in thread ]
Re:
I think we should reward people doing manual labor, like warehouse pickers, farm labor, factory workers, because nobody wants to do those jobs. Artists and inventors create because they enjoy it so pay them less.
[ link to this | view in thread ]
Re:
Plots are just like directions or maps they tell you how to do something is like graphical software today that interpolates X,Y and Z coordinates from one point to another and let you plot a path, it does make sense once you look at it, but it should not be granted any privileges because it harms everyone else.
In the past authors could not compete with big publishers, to publish something there was a need for big machinery and a lot of expenses, this not true anymore, today others can compete and everbody can have their slice of the market pie it doesn't make sense anymore to "protect" people like it was done in the past and it doesn't make sense to protect plots either today.
Although this could be very attractive to studios and book publishers and it is based on the same assumptions that others need to come up with new plots.
[ link to this | view in thread ]
Re: Re: Signed and Posted
Equations that define how substance reacts can be defined by math, turbulance in fluids can be define by math,
Cell interactions all of those things can be defined by math.
The problem with being to vague is that today it can be used to exclude something but also in the future it could be used to include things if it is ever to change.
[ link to this | view in thread ]
Re: Is software patents really the issue?
[ link to this | view in thread ]
Re: Re:
The branch of govt he is referring to is, the one that requires you to have a friend, that hooks you up with a lawyer, that gets you in contact with a lobbyist, that gets your message to the government, that then requires you to fund several politicians, who then give you money to do your own research, and then you are allowed to write the laws they will sign.
Mike I thought you were smarter than that !!! Sorry you were talking about "You guys" never mind.
[ link to this | view in thread ]
Re: Govt Branch
There was central planning before Marx, and unlike Marx, it was oriented more towards protecting existing businesses against new competition. In other words, the kind of central planning you’re looking for is not Marxist, but mercantilist.
[ link to this | view in thread ]
definition
[ link to this | view in thread ]
Re: Re: We hired them
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Re: Re: Is software patents really the issue?
I don't like patents either. But your answer is boooogggguuuusss.
1. "not so with programming languages for which you have a few which all translate to a basic language and that is the language a machine understands, there is no search for the perfect combination, there is no difficulty in finding a solutions to that problem"
Clearly you're not a programmer. Take for example the LZH algorithm used for data compression. I was astonished at how really simple and elegant it is. I never would have tried it because it is completely surprising, at least to me, that it would give you some much compression for so little work.
2. All mechanisms are made up of six classical simple machines. So if your argument that programs "all translate to a basic language and that is the language a machine understands" holds any water then I can say the same for all mechanisms; they all reduce to six simple machines ... just like your description of a computer.
http://en.wikipedia.org/wiki/Simple_machine
In summary, I find your arguments without merit.
[ link to this | view in thread ]
Re: Re:
... and it too should be allowed a patent.
ftfy
[ link to this | view in thread ]
An attempt at a definition
A software patent is any patent that would be infringed upon by a machine which consists of a computer running software.
[ link to this | view in thread ]
Re: An attempt at a definition ...
[ link to this | view in thread ]
Charles Babbage ...
http://www.i-programmer.info/news/82-heritage/3101-babbage-archive-digitized.html
The line between hardware and software really isn't that distinct.
[ link to this | view in thread ]
Re: Re: Govt Branch
[ link to this | view in thread ]
Re: Re: How To Define Software.
This would create one test of sorts that could be used to kill many patents (eg, process patents on computing machines) or at least to restrain their applicability (eg, "fair use").
[ link to this | view in thread ]
Re: Re: Re: We hired them
[ link to this | view in thread ]
Re: Re:
Sort of. Sometimes. (the writing part I mean)
[ link to this | view in thread ]
Re: Re: Patentable Definition
Why not actually require it to have been made? That may be the only way to be sure you're patenting an invention and not an idea.
[ link to this | view in thread ]
Re: Re: Re: Signed and Posted
Equations that define how substance reacts can be defined by math, turbulance in fluids can be define by math,
Cell interactions all of those things can be defined by math.
Are you arguing that shorelines, turbulence, and cell interactions should be patentable? Because if not, I don't understand your objection.
[ link to this | view in thread ]
Re: Is software patents really the issue?
[ link to this | view in thread ]
clever drafting
[ link to this | view in thread ]