Is software too abstract to be patented? | Opensource.com

Is software too abstract to be patented?

Posted 18 Nov 2010 by 

Rob Tiller (Red Hat)
Rating: 
(10 votes)
Image by : 

opensource.com

submit to reddit

The serious problem of proliferating bad software patents was not solved by the Supreme Court's Bilski decision, but now it's looking like it may be part of the solution. The early case law applying Bilski is much more encouraging than expected. The new Bilski test focuses on whether an application attempts to patent “an abstract idea.” That new test has already been applied to reject applications for quite a few bad software patents, and with more lawsuits, it could invalidate a lot more.  To advance on this, we need to figure out out better ways to explain what software is and show when it is an unpatentable abstract idea.

When the Bilski decision was announced last June, at first, I was disappointed that the Supreme Court didn't address software patents directly, and disappointed that the Court's new Bilski test was not more clear. Instead of either adopting or rejecting the Federal Circuit's machine-or-transformation (“MOT”) test, the Court dubbed it “not the sole test,” though “a useful and important clue” as to whether a process involved patentable subject matter. The Court ultimately rejected Bilski's application on the ground that abstract ideas are not patentable.

Based on the Patent and Trademark Office's interim guidance on applying Bilski and the early case law, the test now seems to be MOT-plus-abstraction: check whether there's a particular machine or transformation, and then check whether the proposed invention is really an abstract idea. This could be good for the FOSS community and others concerned with the problems caused by bad software patents. It's easy to argue, at least as to a lot of software patents, that there is no a particular machine and no transformation of an article to a different state or thing, but instead a mere abstract idea.

That's basically the analysis of quite a few post-Bilski decisions from the Board of Patent Appeals and Interferences. I've tried to collect every one of those opinions on software patents, and came up with 25. Of the 25, the BPAI rejected software patent applications in 21. Several of these opinions state broadly that a claim that merely recites software is not patentable, because it is an abstract idea. This is helpful!

The difficult cases for the BPAI are applications that involve some sort of references to hardware as well as to software. These cases came down both for and against validity, and I couldn't discern a clear pattern. There are some decisions that take the view that a software patent cannot meet the subject matter requirement merely by referencing a general purpose computer. A few others seem to find any reference to hardware enough to satisfy the new Bilski test. In short, the BPAI cases are not consistent when there's a combination of software and hardware, but it looks like the issue is still up for grabs.

The first post-Bilski federal district court decision is also encouraging. Ultramercial, LLC v. Hulu, LLC, 21010 WL 3360098 (C.D. Cal. Aug. 13, 2010) considered the validity of a patent over a method for allowing web users to view content for free in exchange for watching advertisements. The district court recognized that the method could only be used on computers or a computer network, and considered various arguments that some part of this process involved a machine or transformation so as to satisfy the MOT-plus-abstraction test. In the end, the court found that there was no machine or transformation, and the process was an unpatentable abstract idea.

A big challenge for those facing the risk of being accused of infringing bad software patents is how to demonstrate that the patent is an abstract idea. The Bilski decision reaffirmed that abstract ideas are not patentable, but it didn't give a test for how to distinguish such abstract ideas. There are various possible paths on this. The Supreme Court has said that mathematical algorithms are not patentable, and Ben Klemens [PDF] and others have argued that software is at bottom indistinguishable from mathematical algorithms. Others have fashioned related arguments leading toward the conclusion that at least some software is too abstract to be patented. Some of the analysis of PoIR on GrokLaw is particularly interesting: Why software is abstract and An Open Response to the USPTO — Physical Aspects of Mathematics.

This is a discussion that needs to continue. I hope FOSS developers and others with deep knowledge of software technology will get involved. We need to get to a convincing explanation in terms that non-technical people (such as judges and juries) can understand of the nature of software and why it is at bottom an abstract idea. Anyone care to take a swing?

 

submit to reddit

40 Comments

tkepley
Open Minded

So I've followed the Bilski case and other cases in other courts over the years and have a few thoughts on things that need to be fixed asap.

You touch on the first thought that always enters my mind pretty well. That is, juries and judges and lawyers are made up of the general public. I know from my line of work that most people in the general public barely know the difference in Windows and Office--let alone the nuances of a closed vs open model. This problem is huge, and it's coloring legislature beyond software patents (see the problems with net neutrality and the privacy acts implemented over the past few decades).

This problem isn't unique, surely judges, lawyers and juries have dealt with unfamiliar territory before. Part of a trial is educating the members on what is being tried. The problem is that we are dealing with something that EVERYONE in that courtroom uses on a day to day basis. Yet they know nothing of how it works. (Part of this is because of problems with closed source) So they have assumptions about the tools and the concepts that are fundamentally colored and lopsided by the giants in the industry and they see it as this is just how it is. So when companies like Canonical, Red Hat, Novell etc come in that they know much less about in terms of licensing and devel models, there's an uphill battle just to show that our way is just as viable and right as any other.

The only way to fix this is to make things more transparent. It's a matter of time and this will only get better.

The second issue is very unique to open source. If I as a developer write a piece of code to solve a problem I have, it is very cost and time prohibitive of me to research to see if a patent for this exists. Even if I find one, I'd have to hire a lawyer to know if I'm risking violating something. No hobbyist developer is going to take the time to do this. So then the code is released to the public as open source. It takes hold. Now the code is on everything from personal computers, to cell phones to toasters. And then a patent troll recognizes something in the code or functionality of the software and well, there you have it.

Wow that was a long post--but I think you get the idea. Great read!

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

Software is one of the hardest technical fields to get patents in, but it’s not due to Bilski. It’s due to prior art. Even companies that file patent applications strictly for defensive purposes, like Red Hat, have trouble overcoming the prior art that the patent examiners come up with. It’s not uncommon for applications to go through 10 or more rejections before companies finally narrow their claims enough so that they can get allowed. At 6 months per rejection, that’s a five year delay in issuance.

The solution, in my opinion, is outside review by experts before filing. These experts really know what is out there and can help an applicant set realistic expectations on what is genuinely new and inventive in their innovations. This can be expensive, but it more than pays for itself in timely issuance of genuinely worthwhile patents.

At least that's been our experience.

Vote up!
0
Vote down!
0
Jose_X

That patent applications get rejected so much shows to what degree greed knows little bound in patent applicants trying to get the broadest chunk possible (but the reward to them personally, a 20 year monopoly, is more than enough to lead those that have the money to spend and spend).

It also suggests that those applying are not our smartest bunch (they fit into the category of "non-obvious" that defines the middle of the skill level bell curve), yet these monopolies will lead to the smart people (finding these inventions obvious) to be stifled.

As for covering the prior art, tell me what fraction of the hundreds of millions of lines of open source each or any patent examiner has ever studied carefully?

It's impossible to know the prior art or come close, and so law should not allow the burden of proof to shift automatically to the defendant just because a patent examiner spending minimal time on that patent was not able to find prior art.

BTW, the low bar to getting patents is the tool used by very wealthy companies to outspend all their small rivals. This can only happen if there are a great many more engineers that can meet the patent inventiveness bar than the opponents can afford to buy off. .. And of course, unlike copyrights, our very expensive patent system is a two-tiered system inaccessible to any real degree to the majority of inventors, especially to those that use open licenses and love to develop inventions and so have no plans on finding or spending funds to litigate to any significant degree (or to patent any of their many many own invention ideas that meet the low legal inventiveness bar).

BTW, changing to first-to-file is yet another way to facilitate robbing from society using the ideas from these open source inventors.

Here is an interesting essay: http://philip.greenspun.com/business/internet-software-patents .
Also: http://www.philsalin.com/patents.html (note that State Street has reverted)
And: http://www.gnu.org/philosophy/software-literary-patents.html (sw patents are clearly too broad for "processes" and algorithms).

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

As for covering the prior art, tell me what fraction of the hundreds of millions of lines of open source each or any patent examiner has ever studied carefully?

0%

What we need instead is someone to invent a way to automatically compare patent claims to source code. As I understand it, open source agreements and copyrights in general are easily enforced since you can automate the search function and look for exact matches.

Tougher for patents. Claims themselves are abstract. Perhaps someone can invent a new type of claim that is sufficiently specific so that comparison with prior art can be automated, but sufficiently abstract so that the underlying concepts of the invention can be covered.

Vote up!
0
Vote down!
0
Jose_X

http://en.swpat.org/wiki/Tandberg_Telecom_AS

What is the price of justice? Easy to attempt to rip off FOSS, yet millions USD in cost for defendants to prove innocence even when they have managed to find prior art many years after the fact? How much more costly if they don't find the needle in a haystack or take longer to find it, perhaps because they don't attract some public support?

The result is no justice. It's patent owners shaking people down who may very wall have no real choice in most cases but to pay the arbitrarily high fine or change business/product (and that won't end possible future litigation in new areas). http://opensource.com/law/10/11/software-too-abstract-be-patented#commen...

Microsoft is "crying" over costs of burden of proof, yet they have ridiculously large resources. Is there justice for any (let's assume, for dramatic effect) *genius* FOSS developer or those small entities using that work or ideas in 10 or more years if that developer attracted patent applicant copycats?

Why struggle for new ideas when so many good ideas are lying all around you? Does it make any sense to give one person a monopoly since we all use others' ideas (never mind potentially copy the vast majority of details from others)? Patents are not for markets with many participants. We already have the much better (yet imperfect) copyright for that. Patents are for high cost markets and where participants are in very short supply. Otherwise, patents won't stand a chance to promote the progress and will likely also violate free speech (software is writing.. how the computer absorbs that writing to carry out the instructions/algorithms is irrelevant).

[And changing the burden of proof scenario still leaves the fact that we'd have to do the impossible of reading everything carefully on a per patent basis to have a chance to say we've covered the prior art. This approach is flawed in contrast to copyright where the prior art must be produced as evidence. To prove a patent follows from the prior art, even when true, can be arbitrarily difficult and costly. This system is very unjust to innocent parties and to many inventors not able or willing to afford patents (at least for software) in addition to being potentially extremely stifling. So costly to prove innocence! And frequently so stifling even when you are technically infringing (eg, because independent invention has not been a clear or useful defense and software patents are quite broad)!]

The SCOTUS is hearing the i4i/Microsoft case: http://www.techdirt.com/articles/20101129/17123112046/supreme-court-will... . In addition to this costs issue, it might be a great idea to petition for a fair use that at least avoids patent infringement in areas that overlap with copyright.

Vote up!
0
Vote down!
0
Unidentified

This would require a considerable amount of artificial intelligence, and is not possible at our level of technology - remember, the concept is abstracted within the code.

It would also be necessary for all proprietary code to be made public in order to be able to search the current known art.

Until and unless both of these are possible, the USPTO should not be issuing software patents.

Vote up!
0
Vote down!
0
Unidentified

The real problem with software patents is that the stuff that is currently patentable (the concepts and algorithms) are completely abstracted behind the code. It is basically impossible to search for prior art without undertaking a massive reverse engineering exercise that would take many times the effort in producing the code in the first place, and orders of magnitude greater than spent on thinking up the idea being patented. What is more, most prior is proprietary software which is secret and therefore unsearchable.

I object to software patents on the grounds that because of the abstract nature of the way it is implemented into code - in other words it is not a physical thing that can easily be compared with existing art - it is not possible or economical to properly search for prior art, and as such and until all proprietary code is published publically, it is idiotic to grant a parent on ideas and concepts for which the patentability cannot be fairly, properly or reasonably be assessed by the patent office and by the courts.

Another objection I have is the length of patent protection for software. Patents are supposed to be time limited monopolies granted to permit an invention to be developed before others can also implement it. It is not intended to be a grant of a permanent monopoly for the lifetime of the invention. A 20 year monopoly may well be reasonable for other areas of invention, but in the case of software which develops and is obsoleted very quickly, 20 years is way more than the useful lifetime of any software innovation.

Vote up!
0
Vote down!
0
Jose_X

When you create a machine claim in a software patent, you do not have a valid claim if you refer to a computing system generally and to things like document (meaning e-document) because none of that is novel.

Eg, an e-document (see i4i patent in Microsoft suit) is giving a new name to an existing item. When we call a piece of paper with something written on it a document, we are not creating a new invention but a new label. Yes we have modified the paper by writing on it, but the abstract aspect comes in here because the writing itself does not create a new physically functional effect (that is why I call it merely a label). It's new usefulness exists only in the intellectual (abstract) realm.

The same exact situation holds with an e-document, it is nothing but a relabeling of a part of an existing machine, of a portion of storage of the "computing system".

That the computing system is able to read and act upon the contents of its parts based on algorithms fed to it has nothing to do with creating a new machine.

As for process patents. My argument there is that these are unconstitutional at least to the degree many inventors hands are bound. You can't promote the progress because the liabilities are too great (the loss of progress of each individuals and of their collective synergies) to be overcome by any possible motivational gains such a patent might have produced.

I believe process patents were not put into the Patent Act until much after the Constitution was written and Patent Act introduced.

Process patents on an "industrial process" at least involve or should the requirement for a very large capital expenditure; however, even for this, I think it should be noted that it is *much* easier in many cases today (vs 1800) to manufacture, get financing, and target an audience to make back investments. Patents are for that purpose, a (presumably needed) lever to enable creation and dissemination of an invention that otherwise would not make it out the door or earn enough to have been worthwhile for the inventor. -- I think Bill Gates and the Microsoft clan have made enough money without leveraging software patents. This should be a tip-off that software might just be very different.

I have written the vast majority of a long essay (to be hammered into an amicus curiae brief afterward) that goes into more detail. The focus I took was in introducing "fair use", and I explain why when many hands would be bound, patent law is very crude and fails miserably in its intended purpose. We have copyright law (itself not perfect) which sets a standard of sort for respecting free speech and trying to avoid stifling. Patents fall very short of meeting the copyright standard. For starters, we know software can be copyrighted and patented, yet patents completely pre-empt the copyright claims. [The inventiveness bar to patents is a very low standard, "non-obvious to a PHOSITA", guaranteed to lead to stifling (see "bell curve") over even much less than 20 years in almost all likely scenarios (a broad patent pre-empts a genius patent in all likelihood because it is much easier and lucrative to write it so as to stay as broad as possible.. not to mention that those likely to find it non-obvious are not among our smarter inventors). The possible exception for patent use not stifling is if very few independent people could participate were the patent not there. This is where the high capitalization costs come in, and where we might have hope that progress will in fact be promoted.]

It is important to point out that the computer is a tool and hence stifling it (through a process patent) is equivalent to stifling the experts and the industry that uses it. There is no creation of a machine and the process is on cheap parts. It should surprise no one that many software developers come from other fields and sometimes even wear a dual hat or otherwise the software developer works very closely with experts in other fields. Software is the intellectual component of arguably the most powerful tool able to advance other fields, fields where patents are not in use and would have stifled.

The Court should not have worried (as it did in Bilski) that patents might be necessary to create life-saving devices but should have feared instead that the patent monopolies are limiting the capabilities and peer-review quality of the "brains" used in these life-saving tools.

Hope this helps, and I will introduce the essay later on in some more public fashion. Thanks for this discussion and for fighting stifling patents. Please feel free to use any of this argument. You will likely find that, like almost anything anyone produces, it exists thanks to the social context. We hardly add anything ourselves to any social context (even Einstein had to rely very largely on the work of others.. and he was no stranger to patents yet avoided them.. and his best contributions are way above the contents of patents). Patent monopolies, to the degree they stifle one person, approach the disgusting. For software, the disgust doesn't end.

Vote up!
0
Vote down!
0
Jose_X

Have people considered the vast difference between the prior art in 1800 and today, especially in light of the introduction of process patents and the digitalization of machine logic?

And adding patent examiners doesn't even come close to addressing the problem because all of these examiners are not put together in a room to address each and every patent. They also haven't been added at a quick enough pace. I mean, if you added as examiners all of the people that have contributed significantly to FOSS over the years, they would still miss and forget things on their own projects and that number is over a hundred thousand I suspect. And this is only for FOSS prior art.

All of this very reasonable conclusion is supported by research that shows that the more a patent gets litigated the greater the chance that it will be overturned. One simple contributing factor to this is the recognition that over time the people that remember the prior art get sued or come into close contact with a lawsuit. There is so much prior art that it sometimes takes multiple litigation attempts to uncover prior art (not to mention the many that have no funds to get involved in this or stay far from the whole patent discussion but could point out prior art).

Vote up!
2
Vote down!
0
Jose_X

>> the digitalization of machine logic

"Digitalizaiton" is the abstraction that we create, like any language, musical notation, mathematics or logic notation, etc, where the representation is very different from the represented, the representation is robust to variations in detail in the represented, and the representation is potentially able to capture only the essentials needed to form that representation (ie, only the essential detail is kept and everything else is removed).

One other example of digitalization is our digits/fingers. It doesn't matter the size, color, or texture of our hands, all that matters and is readily identifyable is the number of limbs we hold up.

Digitalization enabled us to take control functions of machines from a physical sciences experimentation approach to an entirely abstract intellectual approach with absolutely no loss of information, predictability, or accuracy except from the limitations of the models we decide to use.

Digitalization created software and simultaneously took it off the market from patenting as it made the design of control a purely intellectual process based entirely on idealized abstractions (eg, we code to an ideal abstract (virtual) general purpose computing machine).

When someone wants to patent a machine and they refer to computing systems, they are trying to rob from society. They are trying to stifle progress in various intellectual disciplines for their own profit. They do not need leverage to acquire $1 billion dollars for capitalization purposes. They want to hand-cuff potentially millions of those developing software using cheap tools (nowadays, almost as cheap as pen and paper) so that they can have a walk in the part monopoly with no further need to compete or even innovate.

Many software developers are disgusted with software patents. [How many hours have I wasted thinking about and posting about this garbage when I would have much preferred to have been writing code or developing a business.] It is an insult to all your peers. We hardly, as individuals, contribute anything of value in the big picture yet receive a huge amount in return, and some people want to mock society's gifts and contributing communities. That they want to do this meeting the "non-obvious to a PHOSITA" bar rather than say the Nobel Prize bar is only fitting for the filth it is.

Vote up!
0
Vote down!
0
Jose_X

If I sound a little angry in parts of the above comments, it's because the patent system (for software anyway) is little but a system for restricting masses of people. This is so contrary to say the effects and goals of open source (to enable people).

Everyone that participates doesn't carry out the damage to the extent allowed by law (and some do very little with their patents but to build leverage for defensive purposes with even some support for the wider community of developers), but it is still an offensive regime. This stifling aspect is what I am protesting.

Anyway, the key to "software as abstract" (and maybe a key to knocking down all or classes of process patents) might be seen through its digital nature. Language is digital. It's a model. It's ideal. The representation does not match the represented in sensual likeness. We thus achieve the ability to describe to any detail we want simply by extending the intellectual model without a need to maintain any physical relationships or deal with costly physical laws of nature. This potential for creation through perfect precision and predictability was made possible by parallel successes in creating circuits that ignored the precise values of signals to focus on the extreme conditions of either "on" or "off". Mother nature, through practical digitalization, got factored out of the new virtualized worlds we had gained the power to create. Add in the Internet and cheap computing from this century, and this power to create virtual machines (also leveraging wide-scale collaboration synergies) is now accessible to anyone.

I think a more fundamental question than "software being abstract" is over promoting the progress, and the key for that is the number of inventors negatively affected; however, "abstract" is in fact one way to help us nail down many cases where progress is not likely to be promoted. "Abstract" also happens to be a precedent the Court has set and find agreeable and convincing.

Vote up!
2
Vote down!
0
Florian Mueller

I have a copy of Red Hat's amended US patent application No. 10/450.581 (Amendment dated June 30, 2006, Reply to Office Action dated December 30, 2005).

The USPTO had on a non-final basis rejected a Red Hat software patent application. The examiner argued with failure to show a useful, concrete and tangible result. However, Red Hat -- very contrary to what you are writing above -- then argued against the rejection. In other words, Red Hat argued that a software patent was NOT too abstract but pretty much a tangible invention, more or less. That's how Red Hat tried to push that application through.

Red Hat cannot seriously expect to impress, let alone convince, the USPTO if it proposes one legal interpretation in the public debate (the one posted above) while actually taking the absolutely opposite position when its own patent applications are at stake.

By that I do not mean that one can't be against software patents and apply for them at the same time. That is always an option. But in this case we're not talking about the creation of new law versus the application of existing law. We are talking about what is a reasonable interpretation of the existing law in both cases.

Unless Red Hat generally argues that software patents are too abstract -- including when Red Hat's own patent applications are concerned and, such as in the case I mentioned, on the verge of rejection --, I suspect that Red Hat isn't serious about change but more than anything else seeks to curry favor with the open source community in a way that would have to be described -- sorry to say so -- as disingenuous.

Vote up!
2
Vote down!
0
Jose_X

I understand why a company would want to seek patents today as leverage, but your point is basically correct.

Perhaps they can say, such and such patents were granted and ours is just as qualified as those.

Vote up!
0
Vote down!
0
Jose_X

I've thought about seeking patents as a way to try to keep balance in the system (essentially as a troll against these large firms supporting sw patents), but the deck is simply too stacked in favor of the wealthy (never mind the other sacrifices that would have to be undertaken). It is clear to me that the system is extremely unjust and very likely unconstitutional if something is not curtailed noticeably. I think it is possibly much more useful to argue against this system repeatedly before courts and legislators and of course before the public (helping them amplify their disapproval of the system that would rob them of FOSS) than to turn towards trolling (that option will be the nuclear option).

Vote up!
0
Vote down!
0
Jose_X

Red Hat can maintain that the system is a failure and they are only trying to prevent getting slaughtered from monopoly abuse by getting a few strategic patents (at least until the referee starts making the fair calls). This position will strengthen their overall case.

They are probably a bit divided right now.

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

I have a copy of Red Hat's amended US patent application No. 10/450.581

Florian,

Red Hat has several hundred patent applications on file, all of whom they think cover valid, patentable inventions. What made the '581 application stand out?

Vote up!
0
Vote down!
0
Florian Mueller

Simply that the original application was rejected by the examiner on the grounds of not being tangible enough, and Red Hat then argued against that reasoning in very broad and general terms and, thereby, against the reasoning presented at the top of this page here.

Red Hat's amended patent application, which was meant to salvage the application, made a number of general points (not just specific to that application) that were in favor of the idea of software NOT being too abstract to be patentable.

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

Well here’s where it gets tricky. The first independent claim that was allowed was:

26. A system for accessing data comprising….physical metadata elements….

So technically, they have patented a machine (i.e. system) that has physical metadata elements (i.e. memory bits) configured in a certain way. The infringer would not be someone who wrote software, but someone who ran software on a machine.

You can argue that this is a ridiculous distinction, and I would agree. Machines, however, are firmly patentable. So if you want to insist that software shouldn’t be patentable, that’s fine. The consequence, however, is a bunch of patents that claim physical machines and where it is that much harder to tell if you are infringing.

Vote up!
0
Vote down!
0
Florian Mueller

You say that with patents claiming a machine it is "that much harder to tell if you are infringing." I don't see that. It's true that Bilski related to method patents, not machine patents, but in both cases the key patentability question is how "particular" the machine must be that is used to solve a problem. In its defense of that patent application, Red Hat argued for a standard that is met by any run-of-the-mill computer, and arguing for a higher standard in public debates than in the pursuit of one's own patents is inconsistent and therefore can't convince the decision makers.

From an enforcement and evidence point of view, if the standard is as low as suggested by Red Hat in that defense of its application, it doesn't represent any noteworthy hurdle to demonstrate an infringement of a machine patent.

Also, you talk about the "first independent claim that was allowed" while I was talking about Red Hat's 2006-03-60 filing that sought to push almost the entirety of the claims (many of which had been amended) through.

Vote up!
2
Vote down!
0
Mark Nowotarski
Community Member

Well you’re right. I’m probably overstating things a bit.

But let me ask this. Is there a better practical alternative?

Vote up!
0
Vote down!
0
Florian Mueller

Suggestion #1 is to be realistic and honest. The Bilski ruling didn't raise the bar for abstract subject matter. It said that previous decisions have provided examples of what is too abstract and what isn't. The ruling made it clear that only legislation can restrict the scope of patent-eligible subject matter -- not the judges. In other words, it was a ruling that stated the expansive design of US patent law.

So let's all tell the open source community the truth: after BIlski there's no hope that the courts would do away with software patents at any point in time in the foreseeable future. Legislation would be the only option, and that one won't happen for lack of support for the cause (and of evidence that software patents do harm, given that hardly any businesses take action against them but many -- including all large players -- support them).

If one comes up with a way to demonstrate serious resistance by businesses against software patents, and if that opposition is stronger than the support for them, then we have a new situation. I doubt we'll see it anytime soon, if ever in our lifetime.

It's just a PR stunt and currying favor with the community to pretend that there's anything to be gained that way. The SCOTUS saw the Bilski submissions and the complaints and determined to support expansion, not restriction. It won't even hear another case on patent-eligible subject matter for years, or even many years.

I think it would help a lot to tell the community what the situation is. Websites like GroklXX (ZDNet's open source blog just wrote about its manipulation tactics on Friday) do the community a disservice by raising completely false post-Bilski hopes. I liked Brad Kuhn's recent blog posting in which he said the same thing: post-Bilski, forget about help from the courts -- only legislation would change things.

Making people realize this would help focus on pragmatic approaches to the issue, such as embracing FOSS-compatible FRAND licensing and focusing on really bad ways in which patents get used, including some of what Red Hat's strategic partner IBM does.

The second suggestion is to be consistent between political/community-oriented statements like the one at the top of this page and documents submitted to the USPTO in connection with applications.

Red Hat can't have its cake and eat it. If it argues that software is too abstract, it should tell the USPTO the same belief even when its own applications are at stake, and should then (which Jose_X accurately suggested here) base its argument entirely on an entitlement to equal treatment. Such a message would look like "you (the USPTO) granted patents X, Y, Z to companies A, B, C, so we want ours granted, too, because it isn't less tangible than those are, even though we'd prefer an approach that would prevent the issuance of patents like X, Y, Z as well as the one we seek here. We only obtain patents defensively and would rather not have to. But as long as we need them, we can't accept to be disadvantaged compared to others."

I know that this approach would be strongly discouraged by the lawyers trying to push the application through. They would say that this is counterproductive and unusual. But integrity is the combination of multiple consistent approaches, even if there is a price to pay for it.

I make these suggestions and I voiced my criticism because I think the open source community should know the truth, face the facts, and Red Hat shouldn't tell it anything different than it tells the USPTO when it defends its own patent applications.

I know I make myself unpopular among a lot of people for telling what the post-Bilski situation really is about. I believe that time is on my side because people will see, for years to come, ever more such patents granted and enforced in courts, and a few people may then remember that I told them the truth. So I trade in short-term popularity for long-term credibility. I think companies could and should do the same.

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

That's always a good idea.

Vote up!
0
Vote down!
0
Jose_X

>> The ruling made it clear that only legislation can restrict the scope of patent-eligible subject matter -- not the judges.

Easily with a different SCOTUS makeup you get a different practical result.

The fact is that Constitutional rights come first in priority over other laws. The fact is that the Court is the entity that decides if a particular law is in conflict with the Constitution. The fact is that courts are made up of humans that develop different opinions and biases, and tomorrow's Court (even with the same people) might decide that in fact the Constitution would be violated if they interpret certain current laws a certain way. The Court added fair use for copyright before that legislation was passed. The Court has not considered a great many questions concerning the patentability of software running on a computer, for example. You can even go to the very top of this blog posting and see that the appeals process has already decided against software patenting as expressed in patent claims. The fact is also that a number of justices disagreed with the majority opinion in stating that business methods likely have not been written into the law. To extend this, software has not even been written into the law. Obviously "everything under the sun" is not a consistent approach when you consider the courts have specifically taken some things off the table and the Patent Act is not above the Constitutional rights and requirements. If we can stop a dam from being built to protect the legal rights of a few (not even humans), we can certainly stop a patent system gone insane to protect the Constitutional rights of many (humans).

As far as convincing legislators, the opinions of individuals and businesses matter. This is why it is important to get as much support against software patents as we can find (now and as long as this is an issue).

It is very important to communicate clearly the threats posed by software patents.

Vote up!
0
Vote down!
0
Jose_X

>> So let's all tell the open source community the truth: after BIlski there's no hope that the courts would do away with software patents at any point in time in the foreseeable future.

I can't think of anything right now from the Bilski ruing that would lead me to think they view a person using a PC or otherwise engaging with software is in risk of violating a patent. In one part (maybe minority opinion, though can't remember), they clearly stated that nothing on this brief should be interpreted to suggests that software is or is not patentable. I didn't read the whole thing carefully, but it would be useful to me for you to point out what exactly about that ruling makes you thing software running on a computer is patentable?

I think we each recognize that a fair degree of uncertainty exists here, so I find it troubling that you chose to take the position that software patents are good to go into the foreseeable future. You advocate that we strike patent deals with those that have amassed many software patents. You support the status quo, and you spend much time not speaking of the ills of the system but about accepting it. If you are unpopular (your words), it's because these people really do not like software patents and your position of being against software patents seems remarkably inconsistent with your actions.

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

The Bilski decision affirmed that if a process satisfied the “machine or transformation” test, then it was patentable subject matter. The court did not elaborate, however, on what the “machine test” exactly was. It said that would have to wait until more test cases were placed before it.
"This Court's precedents establish that the machine-or-transformation test is a useful and important clue, an investigative tool, for determining whether some claimed inventions are processes under 35 USC 101."
In the meantime, the USPTO has taken the position that a process running on a computer passes the machine test, as long as the machine is not merely doing “insignificant post solution activity”. The current debate being worked out at the Board of Patent Appeals and Interferences is just what is “insignificant” post solution activity. The Board is taking a more conservative view since the Bilski decision. Skilled patent attorneys and agents, however, are usually still able to draft acceptable claim language, but some cases are getting stuck in multiple rounds of rejection. I suspect that one or more of these will get appealed back up to the Supreme Court.

If you have a specific area of software you are concerned about, I would be happy to point you to a few examples of what has been allowed since the Bilski decision.

Vote up!
0
Vote down!
0
Jose_X

>> In the meantime, the USPTO has taken the position that a process running on a computer passes the machine test, as long as the machine is not merely doing “insignificant post solution activity”.

Thanks for the heads up, but I am not shocked they took that position... even though you are just writing out the algorithms and then feeding it into your existing glorified calculator. That's a new machine just like how I became a new person when I start "executing" a new task.

>> If you have a specific area of software you are concerned about, I would be happy to point you to a few examples of what has been allowed since the Bilski decision.

Again, thanks, but all software areas are of interest to me, and at this point in time, I am not going to go seeking "help" from patents.

In other news, two giants that gain from software patents decided to try and establish some court precedent http://finance.yahoo.com/news/OracleSAP-trial-goes-to-jury-rb-2470484165...

> For his part, lead Oracle attorney David Boies said the high-profile case was about the sanctity of the software industry's intellectual property, not just a highly public battle of two technology heavyweights. "Protection of intellectual property is at the heart of the software industry," Boies said.

> From the outset of the trial, SAP acknowledged that a now-defunct subsidiary, TomorrowNow, stole Oracle software.

Large firms are the winners in this game, and they are the ones able to easily take a suit far into court (in addition to buying up or creating an armament of patents).

From http://en.swpat.org/wiki/Inequality_between_small_and_large_patent_holders

> Small patent holders have a weak negotiating position

> A very clear example is Andre Geim's discussion with a large electronics company. Geim won the 2010 Nobel Prize for physics for his discovery of graphene, but he didn't patent it. Graphene is hardware, not software, but this situation shows how weak the position of an individual with a patentable idea is. Geim explains:

> >> We considered patenting; we prepared a patent and it was nearly filed. Then I had an interaction with a big, multinational electronics company. I approached a guy at a conference and said, "We've got this patent coming up, would you be interested in sponsoring it over the years?" It's quite expensive to keep a patent alive for 20 years. The guy told me, "We are looking at graphene, and it might have a future in the long term. If after ten years we find it's really as good as it promises, we will put a hundred patent lawyers on it to write a hundred patents a day, and you will spend the rest of your life, and the gross domestic product of your little island, suing us." That's a direct quote.[1]

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

There may have been many important reasons for Dr. Geim did not seek a patent on graphene, but lack of funds and/or intimidation by a large corporation should not have been either of them.

He was a member of the faculty of the University of Manchester at the time he developed graphene. They routinely file patent applications on inventions made by their faculty and would have borne the full cost. They also have a formal technology transfer office, “University of Manchester Intellectual Property Ltd”, so they would have been fully aware that the threatening remarks by the large corporation were meaningless. Technology transfer officers of major universities know how to license patents and enforce them effectively, no matter how large or how many patents an infringer has.

Vote up!
0
Vote down!
0
Jose_X

As explained below ( http://opensource.com/law/10/11/software-too-abstract-be-patented#commen... ), when we use an existing computer (computing system) to effect the same behavior of a "new" machine that is the object of a patent under consideration, we are not creating that patented machine, but are merely simulating its effect using an existing machine.

And if the patent makes specific vague reference to such an existing "computing system", then the patent is likely not describing anything new and particular (unless it describes new physical parts).

That the USPTO has set down guidelines for its patent examiners to consider these software algorithms running on an existing computer to be a new machine shows that the courts still have work they can do to help right a grievous wrong, the creation of many thousands of broad stifling monopolies yearly that should never have been granted.

Vote up!
0
Vote down!
0
Jose_X

>> That the USPTO has set down guidelines for its patent examiners to consider these software algorithms running on an existing computer to be a new machine

I wanted to clarify that it's very possible that the major problem is not the granting of invalid patents but of the owners of those patents wanting to enforce them against people using computers to run software that mimics the patented machines; however, I looked at the i4i patent (No. 5,787,449) that is the subject of the Microsoft suit and see initial claims (eg, #1) that simply rename existing parts of an existing machine and don't specify anything that isn't already a part of the typical "computer system". Thus, we have the problem of infringement claims that should be dismissed quickly but might not be as well as the problem of patent grants that should never have taken place.

["Metacodes," "map," "document," "addresses," "mapped content," and "menu," are just labels applied to subsections of the computer storage based on the expected usage of the values that will be held in these areas. See http://opensource.com/law/10/11/software-too-abstract-be-patented#commen... .]

Vote up!
0
Vote down!
0
Jose_X

>> So let's all tell the open source community the truth: after BIlski there's no hope that the courts would do away with software patents at any point in time in the foreseeable future.

You say this as if the Court's position now is clearly more in favor of software patents than was the case immediately after State Street.

Here, let me add some "evil" from groklaw.net: http://www.groklaw.net/articlebasic.php?story=20100522202326865

> She never paid it, either. She next petitioned Congress, asking them to undo what had been done, and they refused to help her at all, believe it or not. The law was the law. One might conclude that the legal system didn't work very well. If you were commenting on the case back then on Groklaw, if such had been possible at the time, some of you'd tell me that I was a dope to put any confidence or trust in the legal system, which clearly wasn't fair or working well at all.

> Or one might notice that in the end, 14 years after Anthony died, women won the right to vote despite it all and you might realize that what happened in this case no doubt contributed to that end result. That's why they published the pamphlet, after all, and included in the book a speech by Ms. Anthony she gave prior to the trial on this very topic of women's suffrage. After her death, Wikipedia notes, the New York State Senate passed a resolution remembering her "unceasing labor, undaunted courage and unselfish devotion to many philanthropic purposes and to the cause of equal political rights for women."

> Law is a process, not a day's event, a marathon, not a sprint. Hopefully, both cases will be decided as you think they should be, but if not, I hope you will remember Ms. Anthony's peculiar trial on the crime of voting "while she was a woman".

No one is being asked to end their life "for the cause", but I'll be darned if I am going to roll over and acquiesce to software patents abuse.

I don't like closed source software, and I don't like patents. These are different things, but they are each tools used to unfairly stifle competition and take advantage of consumers. They aren't needed to make a decent living doing something very interesting with your time, and they are each very antisocial. The patents (to the degree enforced for such purposes) are worse obviously because they stop other people from creating and leveraging the fruits of their own honest labor. It's honest to sit down and come up with ideas and source code useful to help other people. It's dishonest and stifling to write a skimpy and merely non-obvious patent claim to be used against others, in particular to whatever degree it was used to increase resistance and difficulties for open source software.

Here are a few opinions from a very well respected programmer:
http://en.swpat.org/wiki?title=John_Carmack_on_software_patents&oldid=18369

"The idea that I can be presented with a problem, set out to logically solve it with the tools at hand, and wind up with a program that could not be legally used because someone else followed the same logical steps some years ago and filed for a patent on it is horrifying."

Vote up!
0
Vote down!
0
stites
Open Source Champion

" Is there a better practical alternative?"

No. Software patents are unworkable. The effort spent in trying to make them work is wasted effort and would be better spent in trying to abolish software patents.

Another way of looking at your question is to point out that software patents are a huge expensive drag on the software industry. Not only that but the current software system is unworkable. Mark Nowotarski's solution is to create a huge mountain of work to search for prior art which will add an even larger financial burden on the software industry. This huge expense would reduce the number of software patents issued but the smaller remaining set of software patents would still be an expensive unworkable mess.

-----------------------------
Steve Stites

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

Are you referring to my comment on using experts in examination?

Vote up!
0
Vote down!
0
Jose_X

The job of granting patents could be much simplified if there were fair use zones that adequately allowed the small business (large population working under "primitive" tools) to create and have basic commerce. You should not restrict areas where access is already fairly widespread. Fair use would have to be wide enough and it would cover software use for many cases, in particular, when used as a process on accessible items (a computer and related peripherals would certainly qualify as accessible). Digital could be a major clue to something being abstract or not (abstract counting as a process). And for any invention, low scale commercialization or even greater scale commercialization under some conditions (eg, certain use research or where the research was contributed to society) would be allowed. Open source components would work in favor of immunity towards patents at any scale. Other factors to take into account would be independent invention, free expression, and the costs to acquire and maintain patents (the latter, multiplied by some factor, serving as a safety zone in order to avoid a situation where the wealthy get free access to monopolies over others who can't afford them to any significant degree).

In short, patents would be granted but not usable under conditions likely to be stifling. There would be an implied fair use restriction on the scope of all patents.

Additionally, patent examination could become as a tool to help the court but assumed to have potentially many oversights and contradictions (patent examiners are only human and searching properly is not possible at this point in time). This would affect burden of proof in court and simply serve as a bit more evidence as well as research already carried out that the patent owner could leverage for the price already paid.

It has been argued that all patents are bad for society, but a balance that would negatively affect very few could certainly be struck and likely not hurt too much. The large firms can still deal among themselves (and even have some loopholes against "trolls" depending on the use), but would otherwise not have much opportunity if any to enforce leverage against small firms and individuals.

Under the above scenario, it would be hoped that the groups most heavily affected (large entities) would then lobby to raise the inventiveness standard for getting a patent granted.

[I hope other people push forward with some of these ideas.]

Vote up!
0
Vote down!
0
Jose_X

How you implement an invention matters. Eg, if the resulting product has much software, that could end up qualifying for fair use no matter the scale. Similarly, if low cost manufacturing is used, that may also allow fair use widely (though perhaps not at all scales). In contrast, the case of a particular or more optimized machine and perhaps using sophisticated manufacturing processes would very likely demand large scale commercialization in order to turn a sufficient profit and would be denied any fair use over most of that scale. Meanwhile, a firm that already had the manufacturing in place (or could adapt it cost-effectively from an existing plant) could likely turn a profit with small scale or else gain extra leeway in scale by adjusting some of the components to be less "optimized" or otherwise contributing research to society.

Vote up!
0
Vote down!
0
Jose_X

With a safe bounds fair use in place, the public would be much more willing to help patent applicants and the USPTO find more prior art. Otherwise, why help establish these monopolies?

And if the inventiveness bar were higher, the odds would also improve that the key prior art would be found (especially if the public were participating, including competitors). The USPTO would also have fewer patent applications to grant and so could allow examiners to collaborate more and/or research more.

The other side of the coin is that patents would be much less powerful and hence fewer would feel threatened by them or participate. But that is probably a very good condition. Patents are too course and are likely to be positive tools only in sophisticated situations among sophisticated competitors (without removing the ability for the determined and "needy" inventor to step up as one of these sophisticated players).

Vote up!
0
Vote down!
0
Mark Nowotarski
Community Member

Patent lawsuits typically cost $1 million or more. That’s not to say that you should deliberately infringe if you are a small entity, but it does help put the risk of a lawsuit in perspective.

Vote up!
0
Vote down!
0
Jose_X

I have not been involved, but it's probably a fair guess that the initial costs to threaten and/or start litigation is probably much less than $1 million for the plaintiff, and the defendant also incurs these costs (and the defendant most likely has no economies of scales that a plaintiff seeking targets for licensing might have). Also, I think the defendant has the higher burden of proof (and so perhaps higher costs) if the patent smells legit.

Anyway, for every company that caves in to a settlement favorable to the patent owner, that owner has extra funds they can afford to spend in order to be sufficiently aggressive to get more stubborn firms to also settle. Settling likely makes up for costs and they can keep going after more firms (where they have a legit case) each time with more leverage (at least up to a point or until they hit someone very stubborn and with enough money to set them back if the patents are invalidated).

Like I said, I have not been sued so have no direct experience, but the fact patents exist on much software is itself stifling because of the potential lawsuits. And it's not just the condition today but where we can end up if the tide turns significantly in favor of software patents.

.... Here is another example. If you can sue a company out of business or challenge them enough to abandon a market in which you have interest, then you will have effectively bought them up for the costs you spent on the rounds of litigation, and that might be a very good price. And if they settle, they you got leverage on them (especially if the royalties or terms are very favorable to you) and use their sales for your own growth to later challenge them more aggressively. If you have several patents or deals with "non-practicing entities" or anyone else, a series of hits against these smaller or less serious firms can very well drive them out of business. The costs are made up by those that settle and the gains from having more of the market accessible to you. The short story is that the patent threat can be very credible once that broad patent has been issued, and this raises the cost of doing business and the hurdle that must be overcome.

....Another example is that by focusing on customers that are significant users of software sold, serviced, etc, by smaller firms, a large firm can starve those smaller competitors of revenues if the large firm can get those customers to switch for apparently no extra charge to their products, services, etc. The customer has the motivation because the large firm might otherwise sue them and perhaps repeatedly (or use proxy partners to sue). Thus the patents serve as a way to simply take customers and revenues from smaller firms that lack the patents and don't pose the same threat as does the big firm. Yes, large firms can always use their size as leverage, but patents can be a very powerful tool and becomes yet another hurdle small firms face when competing against well capitalized larger competition.

Vote up!
0
Vote down!
0
Jose_X

http://en.swpat.org/wiki?title=Software_does_not_make_a_computer_a_new_m...

> A full working computer can be made from a house with a great many doors and door sensors, a central clock, lots of wires, and a motor per door to open and close that door. This house is the computer, made and patented once. From that point forward, every single case of this house "becoming" a new machine is absolutely nothing but a reconfiguring of its doors into the open or closed position. That's it. We aren't creating a new machine. We are simply opening and closing the doors to match the new program this house will run to effect a new set of logical steps needed to effect a new computation (ie, to simulate a particular analog machine that might be built just for the occasion).

> Does it make sense that by changing which doors are open and which are closed (and doing nothing else), that we are creating a new machine? Maybe it takes creativity, luck, analysis, etc, to find a good set of doors to open and close (for example, mathematicians and physicists try to solve new problems by coming up with such a configuration), but we are most certainly not creating a new machine. We are reconfiguring the state of the machine, reconfiguring its doors to be either opened or closed. And, further, the machine, in any of its states, does nothing but process information in a way any human could by knowing the initial state of the doors. Since this is just information processing (digital in nature) coupled with ordinary conversion to (from) analog form through standard peripheral devices (see below), we do not need a slow human or a bulky house for the digital processing, but can instead use a modern digital calculator (the "computer") which uses very tiny parts that use up very little energy and have very little mass so can move very fast.

> This modern computer already exists and can be bought very inexpensively in many stores around the US. It's just a glorified pocket calculator. That same modern computer can run essentially an infinite number of distinct algorithms. We just have to set the initial configuration of "open or closed doors" appropriately and the computer will do the rest automatically. We are not creating a new machine. We are configuring an existing machine's "doors" to the *exact* same opened or closed position as we would if we were dealing with a computing house.

> To see the results and interact with the gigantic computing house more easily we take a standard display monitor, keyboard, mouse, etc, and attach it to the doors that are responsible for holding the values inputted into and outputted from these peripheral devices. We would use ordinary peripheral devices for this in the expected way and for the designed purpose. These all might have been patented. What goes in and out of them is just data in the proper understood format, eg, to be directly displayed on the screen as colors. Every image can be trivially digitized to be seen or vice-versa to be marked as door open/close configuration. At the time we built (and perhaps patent this computing house), we also create the adaptors that change the electrical signals of these peripheral devices to signals that drive the motors corresponding to the proper set of doors. We note that because the house will be very slow in comparison to a modern computer, the display screen will be updated very slowly (so we won't be able to watch a film except over a long time like perhaps months). We note as well that we would need to use at least about the number of doors as we might find inside all the houses that exist in the US today (and this would run only very crude programs with low resolution graphics).

Vote up!
0
Vote down!
0
dabble53

Of course software is abstract. It's merely a fairly concise way of expressing a complex mathematical algorithm or "equation".
Look at what a computer does....it does simple math, boolean operations, and load/stores. Software does nothing more than direct the order of those operations. They could all be done by pencil and paper (albeit much slower). Still, nothing innovative in the process of substituting shorthand and a pen for long hand and a pencil. (Ok, the PEN is innovative to a degree - but then, that's hardware, not software.)

Vote up!
0
Vote down!
0
aikiwolfie

Software is nothing more than a collection of mathematical algorithms containing on occasion, embedded data. As such software is an expression of pure, raw knowledge. Which so far as I'm concerned should never be patentable.

Vote up!
0
Vote down!
0

Rob Tiller is vice president and assistant general counsel for Red Hat, where he manages patent, trademark, and copyright matters. He is a frequent speaker and writer on open source legal issues. Before coming to Red Hat, he was a partner with the law firm of Helms, Mulliss & Wicker, PLLC, where he specialized in commercial and IP litigation. He is a graduate of the University of Virginia School of Law, and a former clerk for Justice Antonin Scalia of the U.S. Supreme Court, and Judge

Reader favorite