The Death of Nearly All Software Patents?

An anonymous reader writes "The Patent and Trademark Office has now made clear that its newly developed position on patentable subject matter will invalidate many and perhaps most software patents, including pioneering patent claims to such innovators as Google, Inc. In a series of cases including In re Nuijten, In re Comiskey and In re Bilski, the Patent and Trademark Office has argued in favor of imposing new restrictions on the scope of patentable subject matter set forth by Congress in article 101 of the Patent Act. In the most recent of these three — the currently pending en banc Bilski appeal — the Office takes the position that process inventions generally are unpatentable unless they 'result in a physical transformation of an article' or are 'tied to a particular machine.'"

Comments

About damn time!

[neowolf]

The subject says it all.

Re:About damn time!

[colmore]

Here's an idea. Write (or type and print out) a letter to your senators and representatives and to Mr. Obama (and I guess McCain if you think he's got a snowball's chance) and tell them how important you think this is.

If you work in the tech sector, tell them that too. Super double extra bonus points if you hold a legit patent. Or heck, if you hold an illegitimate patent for defensive reasons.

Emails don't count.

Re:About damn time!

[Just Some Guy]

IT is one of the only consistently thriving segments of the U.S. economy, and the drivers of that market - Intel, IBM, Microsoft, Apple, Google, Yahoo, Adobe, eBay - all utilize and support software patents.

You couldn't be more wrong. IT is being crippled by software patents, because you can be sued for writing the most obvious things that some jackass already registered. Those companies pretty much hate software patents. Do you think Microsoft really wants Joe Troll in Texas coming after them for 20 billion dollars because he patented spreadsheets? No! They see them as an evil that they have to put up with so that they don't get run out of business by the people gaming the system.

American companies, or at least the intelligent ones, hate software patents because they're only useful against American companies. They don't do jack against the 95% of the world's population that doesn't live here, but give that 95% one hell of a big stick to beat us down with.

Re:About damn time!

[Just Some Guy]

It owns enough of them to claim that Linux is a violation of several.

I don't own any at all, but I'm going to claim that Windows violates a few of mine. I've now shown as much proof of my statement as Microsoft has of theirs.

Microsoft can buy patents.

Timeline.

Eolas.

Avistar.

Burst.com.

Gotuit.

Alcatel-Lucent.

These are all companies that sued the crap out of Microsoft over software patents. Like the saying goes, one of those companies only has to get lucky ones, while Microsoft has to get lucky every single time. Do you think they like being a target for this legal thuggery?

Re:About damn time!

[FingerSoup]

Nobody likes being a target, but if you own enough guns, I'll bet you that you've likely fired a couple of them...

Good

[Lord Apathy]

Thus is a good thing. Patenting software is like patenting a math equation. I can understand software copyrights but not a freaking patent. I wonder how this will affect the cases that where already in court.

Now if we can only get some sense in patents regarding biology. By the way my patent on the biological reproductive process in humans will go into affect today. So all you with kids prepare to cough up.

Re:Good

[Chris Burke]

Patenting software is like patenting a math equation.

It's not even "like", it is patenting math. Software is math. Someone might say that everything can be reduced to math, but the fact is that a ball tossed in the air may follow a parabola, but the ball isn't math, it is just described by math. Whereas software is math, as surely as "y = ax^2 + bx + c" is math. One is a human-readable representation of a pure mathematical concept, and software is a machine-readable representation of a pure mathematical concept. You can't patent the human-readable form of math, you should not be able to patent the machine-readable form of the exact same math.

You can patent the machine that is capable of reading and acting on the mathematical operations described by the software. But not the software itself, because that is, literally, no metaphor at all, patenting math.

Re:Good

[Klaus_1250]

By that same logic, doesn't that void patenting genes as well, as Genes are natures version of software?

Re:Good

[Locklin]

Your not patenting the patent; you are patenting the physical machine. Of course a patent is not patentable, just like software should not be patentable.

GP was NOT making an analogy, GP was making the assertion that software IS ACTUALLY, FOR ALL INTENTS AND PURPOSES math.

Re:Good

[Mouse42]

From my experience with watching how the companies I work for treat patents, they aren't used offensively. They are used to:

A) Show proof of innovation to venture capitalists, stock holders and management
B) Ensure they will have the right to use that process, so some other guy won't use a patent offensively against them.

I would say the patent you're discussing falls under A.

However, that certainly doesn't stop a patent troll from buying the company who has the patent and then using it offensively.

Re:Good

[Jerf]

I've frequently wished that the very fact that a patent troll can sue twenty other people for violating their patent was considered ipso facto proof that the patent couldn't have been all that innovative, by the very fact that several other people coming up with the same solution ought to be the very definition of "obvious to someone skilled in the arts".

Certainly when we're talking about students coming up with "patented material" that should be strong evidence that it's not that hard.

(This is a summary of the argument, there are details and nuances, etc. But I think the root idea is sound.)

Re:Good

[johannesg]

You know, this argument always comes up and I firmly believe it is rubbish. Software is NOT math, and the fact that computer science gets linked to math courses in almost every university is actually harmful (because it scares off people who might otherwise have become very good computer scientists, and because it propogates this "software is math" fallacy).

Granted, math is a tool used in software engineering to achieve certain effects. That does not turn software into math. And you can use math to describe (some) software. That also does not turn software into math.

The essential difference is this: software is an engineering discipline, while mathematics is a science. Therefore they cannot possibly be the same thing. And until I see a proper software science, I will not buy into the "software is math" fallacy, and neither should anyone else.

Hans

Re:Good

[azgard]

We don't know if information contained in DNA is actually sufficient. It may well also be that DNA catalyzes some processes, and then the invention (of nature) is embodied in a physical "product".

That being said, I disagree with patenting of DNA.

Re:Good

[gnick]

I don't like automobile patents because my favourite car won't legally be able to include that superior suspension system.

Yes they will - If they decide that the value of that superior suspension system exceeds whatever the patent holder is charging to license its use.

Patents are being badly abused, but they're not all bad. If car manufacturers knew that pouring $$ into R&D would aid their competitors evenly with themselves and provide them with no market advantage, they would all eliminate R&D, wait for their competitors to innovate, and then start using the new design (should it ever emerge). Unless tinkerers were developing stuff on their own time and donating it to the car companies, neglecting patents all together would ensure that you would never see a superior suspension system coming from anywhere.

Re:Good

[aproposofwhat]

Hans,

At the level at which you (possibly as a software engineer) deal with it, the mathematics behind the software is normally so abstracted as to be almost invisible.

That does not change the fact that at the most basic level all computer programs are just mathematics - a Turing machine manipulating program and data.

And mathematics isn't a science at all - it is a logical discipline, where (all useful) theorems are either true or not true, unlike science where hypotheses are there to be shot down, but so long as they work better than existing hypotheses they are held to be contingently true.

Fact is, it is possible to prove the correctness or otherwise of a computer program, so it's mathematics.

Bri.

Re:Good

[Schadrach]

Let's look at this from a different angle than "Comp Sci is usually in an engineering department at uni and Math isn't".

A computer is, essentially by definition, a machine that computes. As in, it performs mathematical operations (essentially nothing else besides math, input of data and output of data). Those mathematical operations are fed to it through a machine language (which is specific to the type of computer), and thus machine language is a direct statement of a series of mathematical operations, and thus is math (whether you write your derivatives in Leibniz notation or not does not change whether or not they are math). A high level language (the sort you generally actually write code in), is directly translatable into machine language and is thus, wait for it, shorthand for math!

Unless, of course, you can demonstrate for me where in the chain of events the math suddenly stops being math?

Re:Good

[Chris Burke]

Math, can represent what is happening, and it can be used to determine what will happen, but software isn't math.

Yes it is. Software is nothing but a series of mathematical operations expressions. It's in a binary format a computer can read, but it is, literally, actually, and only, a series of mathematical statements. It starts as a series of human-readable mathematical statements, and is translated into machine-readable mathematical statements, yet the resulting binary remains nothing more than a series of mathematical statements. The ASCII code that you see here: "x := a + b" is no different in any way from the ASCII CODE: "add r1, r2, r3" is no different in any way than the binary equivalent of that instruction for a given ISA. What's the difference between the same statement, one expressed in ASCII and one in MIPS? Nothing. They're exactly the same math statement. They're both math. One cannot be math, yet the other not, because they are the same statement.

You liken it to a car engine, the behavior of which is described by physics, but which isn't itself physics. The car engine here is the computer which is capable of reading the binary math statements, and controls millions of tiny electrical switches based on that input. The operation of that computer can be described by math, but the computer isn't math.

The instructions that the computer reads? Those ARE math, literally. They are literally a language that expresses mathematical operations. Not math as embodied in the behavior of a physical entity. Math itself.

Re:Good

[Opportunist]

What bothers me most about those patents, what if, for some odd reason, your lifestock has the patented gene?

Human has bred animals for centuries and millenia to make them perform better. More milk, more meat, more eggs, more ... anything. Now, of course, what companies will patent is genes that offer the same, after all, who'd want to buy a patented cow that gives less milk and beef?

What happens if "natural" breeding (ya know, the kind you can find on youtube) happens to yield the same result as years and decades of research? May I not sell my naturally created beef anymore because someone suddenly patents what I had for years, I just never bothered to patent it?

I see a new problem on the horizon, and it won't be pretty.

Re:Good

[seanadams.com]

Maybe 20 people are infringing on the patent because the claims of the patent are public knowledge.

That is not the case in the overwhelming majority of computer hardware/software patent lawsuits today. More often than not, the people who are sued have independently developed their supposedly infringing design well before the patent was granted and made public. Also, the grantee of the patent usually doesn't actually get around to shipping their product until late in that time frame, so it's not available to be reverse engineered either.

It takes a long time, often three to four years, between the critical filing date and the date the information is made public. Indeed, the fact that others in the field made the same invention at the same time should alone stand as proof that the invention was obvious, but unfortunately the courts haven't seen it that way.

Basically, the vast majority of stuff that is patented today in EE/CS is stuff that comes about as soon as either a) the prerequisite, underlying technology becomes available/affordable or b) demand materializes for the thing. Almost never is it precipitated by any actual invention.

Re:Good

[HiThere]

Go back and look at how this got started.

Intel wanted to patent the code to a process, which it sold as a ROM. The ROM was only different from several other ROMs because of which particular code it contained. So the court decided that the particular configuration of switches inside the ROM was patentable. Everything else grew from that original decision. Including most micro-processors.

I don't know whether it was a good decision or a bad one. I didn't know at the time, and I still don't. Some of the follow-on decisions, however, were frankly stupid. They were done, generally, to make the decision tree simpler...and at the time they were made, they did, indeed, simplify the law. Oversimplify. Then other situations would arise, and decisions were made to be consistent with the previous decisions. UGH!! So we ended up with a bunch of spaghetti law that nobody understands (though some people can argue about very well). Nobody can understand it because it isn't self-consistent, but it's so complex that it's nearly impossible to prove that it's inconsistent.

The whole patentability of software needs to be redone practically from the beginning. Probably code that's embedded in ROMS should be patentable, but I'm not sure. Consider FPLAs, EPROMS, and EEPROMS. Those all look like their code should be copyrightable, but not patentable. But if ROMs aren't patentable, why would the equivalent circuit implemented with hard wiring be patentable?

Possibly one could say "binary code can be patented and not copyrighted, but source code can be copyrighted and not patented". But then what about the various kinds of interpreter, and what about assembler code, Hex representations, and all of the intermediates.

Probably the simplest reasonable answer is to just say that software cannot be patented. Everything else leads into a tangled jungle.

Re:Good

[HiThere]

Any particular instance is an implementation using physics. The program itself is math. As such it can be implemented on multiple different substrates. It can even be implemented isomorphically if the substrates have an identical interface. (Which is what, e.g., a java interpreter or a C compiler attempts to provide [Java slightly more successfully than C].)

This is why the same programs can be run on a Mac and a MSWind box and a Linux box. (If, or course, you choose a program whose complete interface needs are handled identically on all three systems.)

A part of the problem here is the use of the term "is", which has multiple different meanings, none of which precisely fit the universe in which we live. However, let me try...

A program implements a design in symbolic logic extended with various useful operations that were not included in, e.g., Russel's Principia Mathematica or Bool's Modern Logic. With the inclusion of certain extensions and the alteration of certain syntax rules conventional symbolic logic we call the resulting code C code, or Ada code, or Java code, or code written in some other conventional amalgamation of syntax rules and extensions to symbolic logic.

Binary code describes code used by some particular design of mathematical structure descendant from the design called "A Universal Turing Machine". (Note that calling a design a machine does not entail it being one.) Such a design can only act in the material universe when implemented in material form. It is intended that an isomorphism exists between the design and the implementation. Deviations from this isomorphism get called bugs. Binary code can be understood both as a large number expressed to base two and as a representation of a sequence of instructions to be executed by the instance of the design.

Note here that the program is math, but the execution of the program uses physics (and that what's being executed is not the program itself, but rather something intended as isomorphic to it).

Now consider a ROM. In a ROM the mathematical structure of the program is fixed into a physical form. (Math describes a series of relations between different entities, possibly otherwise undefined. Physics is differently constrained.)

I'll believe it when it happens, not before...

[mark-t]

I don't care who's reporting it or how reliable the source, the news that software patents would be invalidated, at least to me, and I'm sure a great many others, is something that is far in excess of too good to be true, so I'm gonna wait and see what happens.

I really wish I could believe that this were possible, but I think too many people with very deep pocket and friends in the right places would get screwed over by this sort of thing to ever allow it to happen.

Re:I'll believe it when it happens, not before...

[Alexpkeaton1010]

You are exactly correct Sir. There is so much money (read: lobbyists) involved, that even if the Patent office has good intentions, I won't believe this until it actually happens.

Re:I'll believe it when it happens, not before...

[greenguy]

Precisely. Patent abuse is both a cause and effect of corporate power. Money speaks, and I think I can hear it clearing its throat even now.

Somebody post on this after it's happened, and a long list of major technology corporations has not only acknowledged it, but acted on it. Until then, I'm not getting my hopes up.

Re:I'll believe it when it happens, not before...

[mr_mischief]

There is also a large amount of money in the software markets held by people who compete with patent trolls. RIM, Barnes and Noble's web design team or anyone who knows how to put payment information in a database, Novell, IBM, and a great many other companies will be glad when obvious, common-sense methods are not being used to fish for huge settlements.

Re:I'll believe it when it happens, not before...

[steelfood]

Let's see among the major corporations in the technology sector and outside, who will defend this new position, and will speak out against it (or not speak at all, for that matter). Then we'll really see which companies are evil, and which ones truly have large software patent portfolios only for self-defense purposes.

Google, Microsoft, Apple, Adobe, Oracle, IBM, HP, Cisco, TI, Intel, AMD, Nvidia, Sony, Nokia, Nintendo, I'm looking at you guys.

Re:Patent Pending

[Presto Vivace]

Actually I assume that this ruling will be litigated.

What about compression algorithms?

[Spy der Mann]

Does compression of data count as "physical transformation" (IMO it's not, but I wonder what the USPTO thinks about it)?

And if this turns out well, does that mean that the MP3 and MPEG4 formats will no longer be patent encumbered?

Re:Can Someone Please Speak English?

[rah1420]

You don't need to be "in the know."

The patent office is tired of BS patents for ideas, and is telling inventors that it has to transform a Lumpy Object (to quote Tom Peters) or be part of a process that is inextricably tied to the operation of a machine (for example, a process to get an internal combustion engine to get 100MPG.)

Simply coming up with a software algorithm or something in the abstract won't be patentable.
(And, IMO, shouldn't be patentable. But of course, it's easy for me to say that because I don't hold any patents, least of all a software patent.)

Mixed Blessings

[danaris]

First of all, can I just say, WOO HOO! This has been far too long in coming!

If this is what it sounds like (and no, I didn't RTFA; way too many links that look like they're probably rich in legalese!), it could pull the rug out from under many patent trolls, and allow a lot more innovation to come back into the US software world.

However, precisely because it has been so long in coming, it could mean a major shakeup of a number of things. One important example is listed right in the summary: Google's PageRank patent. With that invalidated, other search engines can legally use PageRank, without giving Google a dime, which could give them the same searching power as Google.

I can't even begin to speculate what the fallout of this would end up including, but I think it's important not to underestimate it. However, even if there are some short-term negative consequences, I think most of us here will agree that in the long term, at least, this is a big win for everyone (well, everyone but the patent trolls, that is!).

Dan Aris

Re:Mixed Blessings

[Narpak]

If this were expanded to count DirectX I would be pleased. Opening up DirectX would be a good way to reduce Microsofts dominance of the computer gaming platform. And hopefully make games run better and faster. Then again, that is something I can't see happening regardless of the letter of the Law. Microsoft will fight hammer and nail for all of its patents I am sure. Of course, is such a reformation is executed, Microsoft, and others, could take a heavy hit to their stock price.

Re:Mixed Blessings

[lintmint]

I can't even begin to speculate what the fallout of this...

Dan Aris

It's called competition, don't be scared, it's a good thing

Re:Mixed Blessings

[Qzukk]

If the patent is really like that it is useless

Ding ding ding! PageRank's patent is simply one of thousands upon thousands of useless patents exactly like this. Take, for instance, this lawsuit over this patent. Read the line items there, and tell me how one would go about creating a "video codec" using "a single semiconductor chip". I'm almost willing to bet that this "Advanced Video Technologies" couldn't tell me either, but I'm sure they thought that it sounded like it would be a good idea.

Retroactive?

[rumblin'rabbit]

The article uses the workd "invalidate", which sounds to me like it would cancel existing patents. My question is - would this new set of criteria be retroactive, or would it apply only to new applications?

The PTO changing the rules to cancel previously approved patents would generate massive legal problems. In particular, companies have spent billions of dollars to register patents, only for the PTO to say "Oops, just kidding. Jokes on you."

No matter where you stand on software patents (and I'm against them if they can be restricted sensibly, BTW), that's no way to run a patent office.

Re:Retroactive?

[pavera]

I doubt it would immediately invalidate existing patents, however, it would certainly open the door to A LOT of patent challenges, and certainly greatly reduce the value of people's patent portfolios. If you implement page rank, and Google sues you, you have a clear cut defense now it appears. In your case you can argue that page rank is not patentable material, and it appears you would win based on these recent decisions.

It would certainly take the bite out of the patent trolls, as soon as they sue, their patents would be invalidated by this rule, and they'd lose.

In short, I don't think it would invalidate any patents immediately, but it would greatly reduce the ability of companies to leverage their software patent portfolios against competitors (think Microsoft's patent threats against Linux...)

Re:Retroactive?

[DragonWriter]

The article uses the workd "invalidate", which sounds to me like it would cancel existing patents. My question is - would this new set of criteria be retroactive, or would it apply only to new applications?

Changes to the scope of matters to which patent protection applies would affect existing patents, since those patents would no longer relate to patentable subject matter and thus could not be enforced in court.

The PTO changing the rules to cancel previously approved patents would generate massive legal problems.

The PTO is not changing the rules, the PTO is arguing to the courts that the law has been incorrectly applied (including by the PTO) in the past, and that the law should be correctly applied now and in the future. Since patent rights are not inherent rights, but privileges granted by law, there is no basis for protecting them other than the laws passed by Congress authorizing and limiting them. If those laws are incorrectly applied to restrict freedoms of others in ways that the law does not authorize, it is a violation of the Constitutional rights of every person affected (specifically, its a deprivation of liberty without [procedural] due process).

In particular, companies have spent billions of dollars to register patents, only for the PTO to say "Oops, just kidding. Jokes on you."

If the PTO is correct, those companies spent billions of dollars to exploit a misapplication of the law to which from which they were not entitled to benefit but did, in fact, benefit. Since they will neither be recompensed for the expenditures nor forced to disgorge their already-realized ill-gotten gains, I don't see the particular problem here.

This seems to be a fairly routine controversy over what the law means and how it should be applied vs. how it has been applied, not some kind of unusual power grab that you are characterizing it as.

Not good

[5pp000]

I don't agree. Once again, patent policy is being set by people who obviously don't understand the technology, and so, having lurched from one extreme to the other back in the 1980s, we're now going to lurch to a new extreme that is also not going to make sense. If you read TFA closely to the end, you'll see that somehow two connected computers constitutes a "particular machine", where one does not. This doesn't make any sense, and is going to result in an arbitrary selection of which patents are valid and which aren't.

I understand that many people feel that software patents are so broken they should be thrown out. I don't agree. I think the problem with software patents is that the PTO never has had adequate expertise concerning prior art in the industry, and largely as a consequence, the bar for obviousness has been set about two orders of magnitude too low.

Re:Not good

[malkavian]

Bear in mind obsolescence and market saturation times as well.
Patents were developed with a long time to market and market saturation time (i.e. several years to ramp up production, then about 10-15 years to get a market using this as almost a standard), which ate up about half of the patent time. So you had about the same time again to enjoy the benefits of a stable market before the floodgates were opened, and everyone could make it.

In the software world, a technique can have the development time of hours. Market saturation can happen in weeks/months.

If software were to be patentable in its current form, I'd say 5 years would be a good ballpark figure. Like all things, this would have to be hashed out sensibly, so it'll likely never be implemented in a workable form.

Re:Not good

[betterunixthanunix]

The problem with software patents is that they are too easy to unwittingly violate. Even a very specific sounding patent could wind up being violated, just because it is easy to unknowingly embed one system in another. This is where the mathematical roots of CS show through. Worse, the existence of prior art may not be so easy to determine -- the prior art may itself be embedded in a larger system, hidden from view.

Re:Not good

[SpinyNorman]

I understand that many people feel that software patents are so broken they should be thrown out.

The purpose of patents is meant to be to encourage innovation by protecting investment in innovation, but by that standard the concept of software patents is indeed broken.

Software is not like other fields where innovation occurs relatively infrequently and often at considerable cost of time and money. In the software field, there are two contradictory forces at play that capture the essence of the field:

1) Writing software is an inherently creative / innovative process. Every day you are innovating - sometimes coming up with a design takes longer than others, but innovation is essentially a daily and cheap process.

2) Software inherently requires reuse. As the realities of design patterns (formalized or not) and libraries attest, even programming languages themselves, software is inherently about applying a limited set of tools and approaches to solving the unique task at hand.

Consequently, and correspondingly:

1) Software doesn't need patent protection because innovation is not an optional investment - it is a fundamental daily practice part of the field.

2) Software is hampered by protecting "innovation" (i.e. other's software designs) since it is the nature of software that at a certain level of abstraction there are only so many ways of doing things and so many types of functionality that are needed (design patterns and libraries). If software patents are allowed it is inevitable that other software developers, on a daily basis, will need to keep redesigning the wheel, since all software needs wheels. Look at the GNU compiler set as an example - there are only so many types of code optimization techniques that make sense, and due to the patent office having allowed these "wheels" to be patented, every compiler designer, GNU team included, need to find less optimal and obvious ways of doing optimization than the obvious approaches that suggest themselves though the normal discipline of software design.

Previous lawsuits from frivolous patents?

[Jinky]

If this is actually true and at least the frivolous software patents are going to be removed, what about those who have been sued by the patent owners and lost? Will they be entitled to receive anything back from the patent owners as the patents are now invalidated? I know if I lost out hundreds of thousands of dollars to one of these BS patents, I would want my money back. Hell, if I lost $5, I'd want it back. I only read part of TFA, so maybe I'm missing something. Patent discussions are boring :o

Google will be fine.

[SanityInAnarchy]

That blog seems to want software patents to continue -- not surprising, really, given that it's a "patent law blog", and lawyers are the ones with the most to win from the cottage industry of software patents.

While I doubt this ruling will stand, I hope it does. Google has an enormous amount of manpower to throw at this kind of problem, most of it highly intelligent. The only way this hurts Google is if a competitor is able to implement PageRank (and other features) so much better than Google that people start to switch -- and I doubt Google will be standing still as this happens.

The fact is, software patents have had an overwhelmingly detrimental effect.

Does anyone really believe that, for instance, h.264 would never have been invented, were it not for patents? It would either have been open, or some interested party would have paid for the development.

As it is, while it's relatively cheap to obtain the computing power needed to, say, transcode a large library of video to h.264, the licensing cost can end up being something like $2500 per machine used in this capacity. And because of the longevity of patents, it seems unlikely that it will expire before a better encoding option surfaces.

Yeah -- ever wonder why YouTube took so long to convert everything to hi-def, when they have the computing resources of Google available? I think we know now.

I can imagine software patents being a good thing, but not in their current form. Getting rid of the 15-year-monopoly on an ephemeral idea or a mathematical function can only be a good thing for society as a whole.

Tied to a particular machine...

[gmuslera]

how much particular? What about i.e. iPhone? There are a lot of patents around it, probably several related with software. If "tied" means "no way you can implement a similar algorithm if is in a different hardware" should be mostly ok.

This could be the silver bullet needed to kill all patent trolls.

Re:Can Someone Please Speak English?

[Chris Burke]

(And, IMO, shouldn't be patentable. But of course, it's easy for me to say that because I don't hold any patents, least of all a software patent.)

Sure, but for anyone coming from the other side, who does have software patents and is thus in favor of keeping them, all I can say is this: You would be nowhere and have nothing if patents had been allowed in the first thirty years of electronic computing. All the sorting algorithms, all the OS scheduler algorithms, all the compiler technology, all the things you take for granted every day, would have been locked up and all the amazing development that required freely taking these basic ideas as building blocks for more ideas would have faced repeated decade-long roadblocks. The environment in which you are creating your software patents would not exist if they had been able to place those roadblocks to progress just as you are doing today.

So sucks to be you, Mr. Software Patent Holder, but the health and development of the industry requires you to take down your toll booth.

New strategy

[Captain Spam]

s/process/physical transformation/g

Before: "A process by which text may be inserted into a 'base' of 'data' in a format easily searchable by electronic means."

After: "A physical transformation by which text may be inserted into a 'base' of 'data' in a format easily searchable by electronic means."

And presto, patent granted! Cue the next round of the trolls! :-)

Oh, that's right, it's an election year

[realinvalidname]

Those parties materially impacted by this policy, for good or ill, will presumably need to get out their checkbooks and start making contributions to candidates who will legislate/administrate in whatever manner suits the contributors.

You can't have it both ways

[JSBiff]

You've got to decide whether software patents are good or bad. If it's good for Google to patent the idea of page ranking, then software patents, in general, must be a good idea. If software patents are a bad idea, then allowing Google to have a software patent on the PageRank algorithm is a bad idea.

  In the future, if software patents are basically denied altogether, Trade Secret law will used to protect this sort of thing. Unfortunately for the many companies like Google, who've already been awarded patents, the algorithms are already disclosed. Which is why you will probably see some sort of transition period where currently existing software patents aren't just immediately invalidated, but I suspect will be grandfathered in - a basic principle of fairness is you can't change the rules after someone has already upheld their end of the bargain - the patent bargain is that you publically disclose your 'secrets', so that other people can *eventually* use them, but get legal protection on those secrets for a limited time. Telling people who've made disclosure that suddenly they get no protection on their disclosured algorithms is something I don't think is gonna pass - there will be too much resistance from companies on legislators to get protection for this sort of thing.

Personally, I think search engine competition is a good thing. I think competition in general is a good thing. My only concern with outright gutting of the patent system, is that now much knowledge that would have been disclosed in patents, will now remain locked up as trade secrets and NOT eventually become generally available to practitioners of software engineering, and so will, long-term, hold back the progress of computer science. Of course, we all know that right now, patent trolls are holding back the progress of computer science even more. Sort of a lose-lose situation. I guess gutting software patents is, really, probably the lesser of two evils here.

Re:Can Someone Please Speak English?

[Spy der Mann]

Sure, but for anyone coming from the other side, who does have software patents and is thus in favor of keeping them, all I can say is this: You would be nowhere and have nothing if patents had been allowed in the first thirty years of electronic computing.

Maybe, but these first thirty years of electronic computing are gone. Right now you practically can't design any new software because you might be infringing on a software patent. And it's worse if you research because you might find out you ARE infringing on it, and the responsibility is much greater.

Unfortunately, software patents hurt innovation much more than they foster it. In other words, they fail at doing what they were made to do. And perhaps the same is true for other kinds of patents - but in the information era, the impact of software patents is critical.

Which part don't you agree with?

[DamnStupidElf]

Do you not agree that software is equivalent to mathematical formulas, or do you think that software patents (an arbitrary subset of mathematics) should be patentable for some reason?

In the latter case, do you believe that, say, Andrew Wiles proof of Fermat's last theorem could be translated to a typed lisp expression (or any equivalent in some other language) and patented as a method of proving a certain fact about integers, and then authors of any proofs using Fermat's last theorem as a lemma could be sued for violating the patent? What about the four-coloring theorem which requires a programmatic proof? In short, what criteria would you use to distinguish unpatentable mathematics from patentable software?

viewpoint of the article

[bzipitidoo]

I suppose the author, Mr. Duffy, thinks he wrote a fair and balanced article. He mentions several times how valuable certain patents are, and that this change "threatens" many patents and will "destabilize" patenting. There is no mention of the gains to society, this seems to be purely a lawyer's outlook on the immediate consequences to patent attorneys and their clients. Clients will lose a lot of patents, oh no!

There's no mention of how this will greatly help everyone against patent trolling. The closest he gets is a mention that this is "cause for celebration among those ... philosophically opposed to property rights in innovation", as if there are no real benefits to be had, as if that's a point that is of no interest to any except a few theologians of some obscure religion, He writes that this is a departure from over 2 centuries of tradition. We think otherwise. The tradition of patenting is over 200 years, yes, but the tradition of patenting software is perhaps only 30 years old, not 200. And the "system is supposed to be designed to encourage ... the innovative", but there is nothing about the means or any studies about the system's effectiveness. Are we supposed to take it for granted that because the system was designed to accomplish a purpose, it does so? Apparently we are supposed to take that view. How else could a "contraction of patent eligibility" be seen as "very troubling"? To the contrary!

He concludes that figuring out answers to these "fascinating questions" will keep patent attorneys employed. I would add gainfully to the front of employed, but I'm doubtful that these questions are "fascinating", any more than a "debate" over the merits of Young Earth Creationism vs Old Earth Creationism is interesting. We know the answers already, it's that some don't like the answers and have tried to create debate and controversy out of nothing.

Re:This violates my patent

[Kingrames]

Well then we're safe then, since we're not funny.

But don't forget Turing..

[mengel]

Who proved that computer programs are equivalent to Turing machines, which can be actual mechanical devices.

So computer programs are both equivalent to a physical machine, and an abstract mathematical proof, at the same time.

That is, you can argue it either way, and they're both right.

That said, I maintain that you should have to acutally build the Turing machine to get the patent :-)

Re:But don't forget Turing..

[gtall]

Turing machines assume an infinite tape. So when you build one, please get back to us.

Gerry

Just waiting for the counter strike....

[plsander]

You just know someone is going to sue claiming that abolishing software patents is a "Taking" under the fifth amendment and that they (and all the other patent trolls) are due compensation.

Very close - though I would alter it a bit...

[CFD339]

...I would say it is much more akin to authoring music or text. You have a limited (though sometimes wide) vocabulary and syntax. Many authors will put the same vocabulary and syntax together in a strive for similar goals, yet the combination used is generally unique. The differentiation between them come out in skill and subtlety as well as application.

Copyright, IOW, is very appropriate. Patent is entirely not.

Business process falls into the same category IMO. If the drive-through window were invited in the 90's and not the 40's, only the first bank or fast food restaurant to come up with the concept could use it and all others would have to pay royalties. Then the owners of toll roads would sue for prior art, etc. etc.

Re:Patent Pending

[phoenixwade]

Actually I assume that this ruling will be litigated.

Yeah, that's a "well, Duh!" comment - there is too much money involved here for it not to be litigated.

If this interpretation is upheld in litigation, you can bet that congress will get involved and fix it so that software patents are retroactively reinstated.

I'd like to see big money lose over the interests of the people, but I doubt our system could ever allow that.

My dad can beat up your dad...

[RingDev]

Actually, all that math is an abstraction of electronics, and electronics is just an abstraction of physics.

So "researchers" are no more Computer Scientist that "programmers" are because neither of them comprehend the entirety of the physics that are driving the electrons that make it possible for that computer to do the math that has been abstracted to a high level 'English-like' readable language. /sarcasm

Really, intra-software industry pissing matches are just humorous. There is no significant difference in skill levels of programmers and researchers. Only the knowledge of the application's focus changes. And for as complex as modeling some new mathematical equation may be, I can present you with a mile long list of tax codes, international trade law, federal regulation, incentive programs, etc... that can make a "simple" accounting program just as complex.

-Rick

Re:My dad can beat up your dad...

[The Dancing Panda]

Those are all programming problems, not CS problems.

A prototypical CS problem is "Traveling Salesman". Writing a program to route a salesman over a series of cities would be a programming problem. The CS researchers problem would be solving the underlying mathematics of it in polynomial time. CS Research can be done sans computer, programming cannot (at least not usefully). There is a huge difference.

Also, your assertation that math is an abstraction of electronics is just weird. Electronics is a specific field of physics. Physics is (in essence) a specific field of math.

Re:My dad can beat up your dad...

[RingDev]

A nice response, I agree with much of what you said, with one exception:

and mechanics and automobile designers both work with cars.

I don't believe that accurate describes the correlation between Programmers and Computer Scientists.

The 'Computer Scientist' of the automobile industry are the material engineers that determine what compounds to make components out of to best perform their specific duty. For instance, designing a ceramic piston to replace a traditional steal one in high compression non-nitrous engines, or developing an aluminum block with steal sleeves to replace traditional cast iron block, or designing a 2nd catalytic converter to take advantage of a lower temperature with a different chemical reaction to reduce emissions. None of those tasks require any significant knowledge of the automobile as a whole, but do require significant knowledge of their specific scientific field.

The 'Computer Programmer' of the automobile industry are the vehicle designers. They take the tools and devices the 'Computer Scientists' have made available to them and build something whole and functional out of it.

The 'Application Administrators' and 'Support' are the rough equivalent of the Mechanics of the automobile industry. Sure, if there's a design problem, they'll kick it back up to the programmers, but for configuration and maintenance, they handle the bulk of it.

My apologies for being confrontational, I've met my fair share of egotistical 'computer scientists' who were so proud of their brand new' life altering' linked list logic... So usually when people pull out the "programmers are lesser than computer scientists" argument, I jump to the defensive side ;)

-Rick

Re:Patent Pending

[afidel]

Actually with all the patent trolls big money is starting to realize that software patents are doing way more harm than good, even if you have a large warchest of patents a patent-troll can still cause you a large amount of cash and time. It's a bad situation for big money since they have no leverage against the patents trolls as the trolls only goal is to extort a jackpot out of big money.

Re:This violates my patent

[erroneus]

The question is then "should software be a product?" Initially, it wasn't. Software was just something that was needed to show that the computer worked. People either wrote their own software or borrowed it from someone else. It was never really considered a product all by itself.

Efforts to change this has resulted in quite a mess we have today. Among these are stiffled innovations, advancements and imporovements in software technologies. And as can be seen, treating software like hard copies or an object is a truly broken idea since software doesn't exist in the physical world -- physical paradigms don't work on it. Copyright is the most applicable intellectual property protection mechanism for software. Patents just don't fit.

The fact is, patenting software is nothing short of anti-competitive behavior. The very idea that someone might be able to do the same thing better, faster, cheaper or more efficiently is put on hold while the patent holder, who may or may not necessarily have a working model of his own, sits back demanding money from everyone else who is actually capable of moving forward with ideas. In the case of copyright, you have to have actually written and produced something to have it copyrighted. This is certainly more appropriate for software.

In fact, the only way software patents have inspired people to be creative is in finding ways AROUND existing patents. PNG is a perfect example of this. I think it's hard to agree that software patents are good or appropriate. Designing a new bearing, or a new type of chair might be worth patenting. Software isn't... it's a list of instructions. Patenting a list of instructions shouldn't be permitted.

Re:This violates my patent

[cHALiTO]

And then there's the issue of Software being seen as a product, but not being sold, and it being -licensed- instead (even tho you usually don't get to read the license until after you've bought and installed the copy).

Of course it's debatable, but personally I HATE the idea of licensing software. I think a Software Product should be sold as a copy, protected only by explicit copyright law (i.e.: you can't redistribute it without permission), period. None of this nonsense of dictating how you can or can't use it. If I want to use it for business, I will, or at home, or for whatever purpose the author didn't intend. It's my copy, and with it I should be able to do absolutely anything I want to as long as I don't redistribute it without permission.

I want to BUY a COPY, not just a limited set of permissions from someone to only do a couple of things with it.. where did this idea that just because someone's the author of some work he gets to dictate what others do with it once it's been paid for?

If I buy a copy, that copy is mine, I do whatever I want with it (as long as it's not explicitely forbidden by law).

This is all fucked up.

Software should be no more patentable than a poem

[dontmakemethink]

Software is a series of words that instruct a computer to perform specific functions. Copyright is all that's necessary to protect the intellectual property of software developers.

And I don't mean copy protection of the specific code, obviously there are many ways for software to achieve the desired effect. It's the effect that should be subject to copyright, which is what leans software developers towards the patent office.

However, if you translate a poem to french, which would entail significantly changing several phrases to maintain a rhyming scheme, the original author still has copyright protection.

L'auteur n'en apprendra jamais probablement*, but the software developer probably will learn of competitive products that violate copyright.

* The author would probably never learn about it