A weak case for software patents

No readers like this yet.
Not about profit


For all the debate and litigation around software patents, I thought that there was at least one point on which all sides could agree: the objective of the U.S. patent system is to stimulate innovation.  A recent IP blog takes issue with that premise, and proposes an alternative objective: making money.  The blog gives a distorted view of  Red Hat's patent portfolio program in support of this argument.  The argument is interesting, and suggests that there's still a steep hill to climb to get to a rational patent policy for software.

First off, let me say that I don't see the software patent debate as a matter of good versus evil. It's about establishing and achieving reasonable public policy objectives. Patent policy relates to commerce, markets, and encouraging material improvement. When the government grants a patent, it grants a type of monopoly, and monopolies entail costs. A patent entitles the holder to exclude others from using the invention, and so that holder may use a patent to prevent competition and hinder innovation. But the widely accepted policy justification for patents is that on balance they encourage innovation and progress.

There's abundant evidence, though, that software patents serve this objective poorly. For example, as Bessen and Meurer explain in their book Patent Failure, the cost from litigation based on patents in some industries, including software, is several times the profits from patents. This means that for the software industry as a whole, patent costs exceed their benefits. This fact is still not common knowledge, but it is slowly sinking in, as patent litigation continues to grow.

Patent lawsuits (even meritless ones) can cost millions of dollars to defend. This reality creates an economic opportunity for non-practicing entities (a/k/a patent trolls) whose only purpose is to threaten and bring patent lawsuits. The large costs of defenses drive many defendants to settle by paying patent aggressors substantial sums taht are still less than the cost of defense. Such settlements may make business sense, but they divert resources that could be used for more productive purposes, like innovation.

Not everyone views it this way, of course. Those who are profiting from the existing system generally think that it works rather well. And they have some appealing-sounding arguments. For instance, they argue that patents encourage innovation by allowing lone inventors to pursue their ground-breaking dreams in the face of powerful corporations. This sort of story tends to excite emotions and hinder rational analysis. It ignores the rarity of inventors who work without significant collaboration, of inventions that are ground-breaking, and of patents that ever recover even the cost of the patent application. Dreams of getting a hugely successful patent are about as realistic as dreams of winning the lottery. Still, it's a nice, and understandable, dream.

A recent blog by Gene Quinn of IP Watchdog takes a more materialistic approach. Quinn describes himself as Patent Attorney and President and Founder of IP Watchdog, Inc. He does not mince words. In his view, those who criticize software patents are “ideological buffoons.” Quinn has this to say to the buffoons who are concerned that software patents hinder innovation: “WHO CARES?” The reason not to care, in his view, is that the only important issue is whether patents make money. Surely Quinn does not mean that  hindering innovation will result in increased societal wealth, so he must  be saying that the only thing that matters is individual money making, and not any larger conception of the public interest.

Even accepting this narrow definition of the good, though, Quinn's argument is unconvincing. It rests in part on the idea that investors believe software patents are valuable and therefore pay more for such investments. This argument is based on weak data, and it collapses entirely once the investors get around to reading Bessen and Meurer and learn the true costs of patents. Quinn's other argument is that software companies need patents in order to deter and defend against patent lawsuits. This presupposes, of course, that there is a justification for granting software patents that can be used for aggression, which is the very point under discussion. As a proof in support of the need for software patents, it's circular.

Quinn is not, however, wrong in noting that successful companies assemble patent portfolios for deterrence purposes. Given the reality of patent aggression, continued survival requires consideration of possible defensives, and deterrence is a potential defense. He cites Red Hat as an example of a company that uses this tool, which is fair enough.

It is unfair, though, for him to suggest that this is a basic ingredient to Red Hat's success. That success rests primarily on its adoption of the power of open source – not patents. It's also unfair to discuss Red Hat's portfolio without including an explanation of Red Hat's Patent Promise in which it commits not to use its patents against open source. Red Hat has been a consistent and vigorous critic of software patents, and it is misleading to suggest otherwise.

In the end, Quinn's weak arguments aren't likely to persuade anyone not already inclined to agree with him. But his frank indifference to the public interest is unsettling. Thinking that short-term gain is the only meaningful value is part of what led to the recent financial meltdown, and it may account for some of the difficulty in advancing patent reform. Quinn's narrow view is a reminder that a lot of grassroots educational work needs to be done to arrive at a more reasonable patent system for software.

User profile image.
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.


Everyone opposed to software patents should read Quinn, because it shows what a challenge we face. He and many lawyers like him are getting rich on the back of the software patents industry, and if lawyers are good at one thing, it's convincing people who aren't expert in the subject area to agree with their views.

Quinn is a little extreme in characterizing those with contrary views as dunces and communists, but it shows the lengths to which people will go when their livelihood is threatened.

when the Framers wrote both to allow inventors to patent their materials and that it should be for a limited time, and that it is for the public good, it isn't necessarily clear whether its the patent/monopoly that is the public good, or if it's the time limitation causing the idea to revert to public domain that is the actual public good.

The "Free Culture" folks (and I'm largely in that camp) would suggest it's the limitation and subsequent public domain status that provides the benefit to society, while people who are in favor of patents in general and software patents in particular would suggest it's the monopoly that is the public good (since it gives people an incentive to invent).

While I'm personally somewhere between those two poles, I definitely lean toward the time limitation and subsequent public domain being the better of the two.

but somehow in all the heated (and not so heated) discussions of patents, this tension is very seldom brought up for discussion. It's as if those doing the discussion see no need to even admit that the opposite concept exists, much less has any validity.

Great write-up.

If eating babies is a modest proposal, then perhaps we would not be going far at all in legalizing state hanging and "Wild West" justice as two more efficient means of stopping those believed to be infringing patents.

Quinn might agree that this would enable some individuals to get even more wealthy than they can under the current system. Quinn might also agree that this will serve as greater motivation (for some) to invent/innovate than serves the current system.

Maybe Oracle's recent lawsuit against Google (for java patent/copyright infringement) demanding injunctive relief will become the SCOTUS case to end most future software patent threats against FOSS. We need to address the "promote the progress" failure of software patents (especially when aimed at FOSS) to end this nonsense.

And while at it, we might as well bring up the abridgment of free speech. Software is communication and expression. Patents are broad so prevent development of ideas, even of ideas independently developed.

freedex, reason and data shows monopolies are not desirable, especially when you turn a significant amount of unlimited good into a very scarce good. The framers weren't universally for monopolies and hence the qualification and criteria "to promote the progress".

Relatively to 1790.. the tremendously improved means of collaboration; the very much improved infrastructure, access to capital, and speed of commerce; the insanely low levels of capital investments required to invent, produce, distribute, etc software; a very large number of inventors being negatively affected by software patents; the high degree of complexity, interdependence, and incremental development nature of software; the informational nature of software (separate from all physical laws and limitations, like literature or math); the very low bar the USPTO places on inventions ("non-obvious" to a PHOSITA rather than "of break-through quality" or "deemed revolutionary by the overwhelming majority of geniuses" -- and note that even the high bar examples would have prevented further revolutionary science and math if applied in those fields); the very broad reach and generality allowed for patents by the USPTO; the nonsense that being second or third or fourth should give you zero for 20 years; the inequity that the many who are contributing ideas and inspiration to the patent authors get zero; the inequity that the many that can't afford the time and/or dollar costs of patenting are third class citizens in the patent game since they have no real leverage, can be blocked from growth, and can have this happen by patents created off inspiration from their works; and the first amendment rights violation allowed by not recognizing independent invention ..should have been reasons for the length of typical software patents today to be but a couple of years if not zero years, zero days, and zero seconds.

Quinn is a patent attorney. The only argument he is concerned about is whether eliminating software patents would be good for him personally. Of course the answer is 'no' because it's one less thing for him to litigate about. Anything else he says is just so much hot air. He's hoping that there really are a lot of "buffoons" out there who will accept what he's saying at face value.

Software patents are fundamentally flawed anyway. Patents are supposed to be on specific implementations of ideas. In software, implementations are already covered by copyright, so patents end up being patents on the ideas themselves. Whitney's patent on the cotton gin was not a patent on the general idea of 'a machine that removes seeds from cotton.' It was a patent on the specific implementation of that idea that he invented and the specific method his machine used to do this. If someone could come up with a better way to remove seeds from cotton, they were free to patent that. Patents are supposed to be about the vehicle, not the destination. Read a few software patents and you will see that they are vague. They are about <em>what</em> is accomplished rather than <em>how</em> it is accomplished (other than it being through software). The specifics are in the code, and that is covered by copyright. (There may be some exceptions to this rule, such as perhaps patents on compression algorithms, but they are few and far between.)

As to what part of patent law was supposed to be about the public good, the answer is both the monopoly and the time limit on that monopoly. The monopoly by itself only serves the inventor's good. It being time-limited changes it to serve the public good.

According to the patent idea (and similarly for copyright), granting a monopoly provides for the public good by creating incentive for people to work toward progress because they can capitalize on their ideas. Making the monopoly temporary assures that all will eventually be able to benefit from that progress. So without the monopoly, there is less progress to benefit from, and without the time limit, the benefit never completely reaches the general public. Both parts of the idea would be considered essential to accomplish what patents were supposed to accomplish.

<cite>...as we enjoy great advantages from the inventions of others, we should be glad of an opportunity to serve others by any invention of ours; and this we should do freely and generously.</cite>
-- Benjamin Franklin, in response to the Governor of Pennsylvania on why he did not want to patent the Franklin Stove.

Gene Quinn is the lawyer who defends patenting everything. During the debate about genes BRCA1 and BRCA2 he argue that is acceptable to let women who can't afford to pay the extortionist price set by Myriad Genetics to die.

Just watch starting at 5 min and 50 sec until 8 min and 30 sec.

Dodo birds of a same feather, it is said, flock together.

We should all be like Dilbert:

and blindly obey the pointed hair man.

After all, a cubicle is a comfy cave to spend the rest of your life in. Why bother having an original thought of your own when you can simply parrot the bird songs of your fellow cubicle dwellers?

Polly no wanna patent. Squawk!
Polly no wanna patent. Squawk!
What? Me think for myself? Squawk!

I'm not sure of your motive for posting that strip, but the strip itself is quite funny (after you've read over a few software patents anyway).

Are you actually trying to claim that anyone who doesn't like software patents hasn't thought the matter through, or simply isn't capable of thinking it through? The people who aren't capable of an original thought aren't even aware of software patents, and couldn't care less about this issue.

If you want to respond to reasoned arguments against the practicality of software patents, then I suggest you actually use reasoned arguments instead of ad hominem attacks by just calling anyone who disagrees with you a "dodo bird" and a "parrot." I know Quinn isn't a very good example, but that's no excuse.

What is sad about Dilbert is that he is a bright Engineer-head who falls time and again for the management antics of CatBird and of the pointy haired (no-name) boss.

That is why Dilbert lives forever in his cubicle, grinding out meaningless equations, useless designs and waiting to be called to the next dreaded meeting where it will all go up in a puff of managerial smoke after first bouncing around a bit between the mind-bending mirrors.

Software developers who fall for the "software patents are bad" line of management are just born-again Dilberts. They don't understand that voting against software patents is voting against their own economic interests. What value is their IP work if it can be freely copied by the big boys? Without patents, you can never leave that cubicle and strike it out on your own. Why are so many of you so blind you can't even see that simple angle?

Dilbert himself is generally correct when he makes an analysis. The problem is that the management <strong>ignores</strong> him and his advice, which is actually what you are suggesting that we do.

Also, the current system of software patents favors established firms against other practicing entities. Even if you have a revolutionary idea, the 'big boys' will have enough mundane patents to have the upper hand.

By leading Dilbert on into believing his ideas are being <b>"ignored"</b>, management is sending a few simple subtexts to Dilbert:

1. Your ideas are worthless
2. You are no more than the sum of your ideas and therefore you are worthless
3. Since you are worthless, you do not deserve a raise
4. You need to keep working harder and harder, faster and faster, to deserve even the meager pay we give you now
5. Because they get more pay than you, management is clearly smarter and has more of a clue than you. You need to go back to your Batboy Cubicle and meditate in the darkness over that until you come into full acceptance of that Nirvana-ic Truth

Before you toss the above aside as being merely a cartoon, pay attention to what is being said here in these very comments:

1. The measly few lines of code you individually produce are insignificant microbes in comparison to the mounds of bloatware your corporate overlords produce (by efforts of their other worthless serfs) and therefore you are worthless
2. Just be glad you have a job and we don't ship yours out to India
3. Never ever think about getting patents because those would ruin our plans to ship your job to India anyway so that we can justify the absurdly large salary your do-nothing corporate overlord (aka CEO) makes.
4. Also, if you start thinking patents might be good, where is that going to lead to? It's a disastrous slippery slope. Next you'll think you can leave the master's feudal castle and start a company of your own and protect your innovative new software with these patent things. This is exactly why patents must be destroyed. And who best to do the destructive deeds than the worthless Dilberts who cannot for the life of them see how the bigger picture is put together?

(It was Catbird's idea to have the Dilberts shoot themselves in their own economic interest toes and he is laughing himself hysterical that they, the "freedom" programmers think they are working on behalf of their "freedom" instead of for their continued serfdom.)

Large corporations where they have 'Dilberts' are in the strongest relative position among practicing entities when software patents are in place, and I'm pretty sure any corporation with any business sense has their employees sign work for hire, assign copyright, a moonlighting clause, and all that jazz to effectively make anything patentable they produce on or off the clock be assigned to the company. So, the startup is at a disadvantage because Path-E-Tech Management Corporation has a nuclear arsenal of patents, and the Dilberts have already signed away their rights. The only party that can get a benefit is a patent troll or their lawyer, they probably don't make money most of the time (okay, the lawyer probably does), and even the catch of the day is chump change compared to what the big boys make. So who exactly are you speaking of that is missing an opportunity if software patents go away?

Let me add that an important compensating lever available to every single small business entity (producer or consumer) is open source. The only real supporters of software patents are those who are among the most desiring of becoming very wealthy. Most creators don't have the desire to follow that path through its many negatives and obstacles. Hence we get to the real purpose of these broad patent monopolies: to allow a small few to become very wealthy at a very real cost to everyone else (who otherwise would have a fair shot at making a decent living based on his/her skills or of enjoying a larger set of quality goods at overall lower prices and higher rate of improvement).

The big boys can outspend you anyway. With software patents they can stop you from even trying. I don't know what your idea of small is. most small businesses don't have the money / time to file , much less defend patents. Maybe that is why literature or math or not patentable. Software patents are just another bubble in the US economic decline. All short term , with no consideration to the larger ecosystem. This lawyer is not even a part of the ecosystem. after the software industry is destroyed he will move on to another host.
he is the best argument FOR communism and the worst of capitalism. And this is why US economy is on a decline. Look no further.

>> The big boys can outspend you anyway.

Isn't it "great" that the big boys can outspend you? Many month's salary to us is a few seconds or minutes cost to them. While they can attempt to patent many many "non-obvious" ideas per year at that rate, we can patent possibly not even one per year. Then we have even larger defense costs to deal with as well.

But weren't patents to protect against "the big bully imitators" and to encourage some small degree of contribution of society by revealing an "invention"? [Note that trade secret is still allowed, so patents can be taken out to stifle competition without any need to reveal the really good secrets.]

>> Maybe that is why literature or math or not patentable.

There are just too many people who would be denied creative output, honest business, learning, progress, and speech if we allowed patents on math and lit. There would be an uproar by the People, as there should be eventually against software patents if these continue to get in the way.

>They don't understand that voting
against software patents is voting against their own economic interests. What
value is their IP work if it can be freely copied by the big boys?

You are 100% wrong here. The economic value of software (as well as the cost) does not come from one or two "non-obvious" ideas among 100,000 lines of code: it comes from the efficient and effective implementation of the 98,000 lines that are not patentable, and would never be patentable even if Mr Quinn had his way. The proof is that the vast majority of useful and effective programs have no patent protection. People buy software because it is useful, not because it is innovative.

In fact, I would advise any developer nowadays to think hard about that remaining 2,000 lines and try and rewrite it in a less innovative way, to remove the risk of being sued. However novel you think it is, the chances are someone else had a similar idea in the past, and they might have patented it. And whatever you do, don't apply for a patent on it: that will only bring yourself to the attention of the patent lawyers working for your competitors. (A friend of mine made this mistake and essentially went out of business as a result.)

But the real point is that software patents are bad because they are not in the public interest. Innovation is in the public interest, and innovation in software works best when innovative developers are constantly leap-frogging each other - taking each others' ideas and improving them. The patent system is designed to prevent that way of making progress.

Two other points: (a) phrases like "they don't understand" are incredibly condescending. Do you really think the people with this view are morons? Some of us have been writing innovative software for 35 years. (b) "voting against their own economic interests" - you're on a forum here where you're probably talking to many people who try to consider the interests of society at large, knowing that they will thrive if the economy thrives.

@Michael Kay

With due respect, I know exactly how they think because I used to be one of you (... before I joined the dark side).

I am not trying to be "condescending". I am trying to slap you silly on your face so that you wake up and smell the Matrix.

For once in your scared-of-it life, take the red pill.

You no doubt applauded when Oracle/Sun said a few years back they would never ever on-their-grandmother's grave, use patents as offensive weapons. Where is your faith in your overlords now? Has it not yet been shaken? If not, under what circumstances will it ever be shaken?

There is a large and vocal class of people out there who believe in their right to carry a gun; to "bear arms". None of them are saying they plan to kill their neighbor tomorrow. But if push comes to shove they "can" and they just may because they have the gun in hand, loaded and ready to use. In a strange (and MAD) way, that helps keep the peace and helps keep everybody respectful of the next guy.

Software patents are sort of like that. They are weapons (usually small power ones) that you can keep loaded at your side or under your pillow. But possession of such a weapon presumes that you hadn't first gone the religious fanatic believer route and had disclaimed all rights to defend yourself (with a patent weapon) and that you didn't blindly throw yourself prostate and in a state of total fidelity at your overlord's feet.

No one says you have to shoot ever, or shoot first when you own a patent weapon. But if you disavow even the idea of ever having such an IP legal weapon, and you convince your trusting brethren to do the same, well then; all the people out there who harbor nothing but the bestest of intentions for you and your brethren, know that they can step on you all and crush you all like bugs should the mere desire to do so ever arise in their heart of hearts. (You know, their corporate hearts. The ones that beat with pure green greed. Makes you feel all kind of warm and humanitarian inside, doesn't it? Keep on believing in "them". Catbird loves you.)

What is your chances with single six-shooter against army with thousand guns in any shape and power?!
Please don't bother to answer. I know you have even more ridiculous answer.

ps. I feel dirty for feeding the troll. Especially the patent troll.

Sal, Sal,

I've sat eyeball to eyeball at the negotiation table with some of these I've-Got-1000-Guns sombrero's.

All I had to say was, You and what army, Amigo?
Then they crumbled.

The reason is that 99.99% of their so-called sharp shooting guns and nuclear bombs are duds.

You can't have it both ways here. Either you say many of those "software patents" out there are just plain silly and will whither under the light of morning sun or not.

And if yes, why are you so frightened of their 1000 guns all filled with dud gunpowder? All you need is one good, well aimed bullet on your side.

But you're telling your software-crafting brethren they shouldn't even think of packing a one bullet pistol under their shirt sleeve, huh? Sound advice.

As I recall it, David had only one pebble and a cheap sling shot. But your take away from that story is that he should have used the sling shot to quickly hang himself on the nearest tree rather than daring to fight back against Goliath.

Let's look at two core problems.

First, most engineers prefer to create things (and society is thankful for this) rather than to live on litigation, which is the business plan you propose. If you want to create, their thousands of "duds", army of lawyers, and billions in the bank stand in the way. They will complicate your life, very possibly pressure you out of the market, and stifle you significantly, as among these "duds" is much that has been granted holy powers by the USPTO. [There is a very low inventiveness bar to being granted holy patent powers over others and a very high bar to the presumed guilty defender to showing that even such a bestowed patent did not pass that low bar. This is a broken and unconstitutional part of our patent system.] We know there are many broad patents difficult to avoid while still producing an engineering top or even competitive product, no matter how creative or otherwise brilliant and hard-working you are.

Second, as Red Hat recently showed with their successful defense, if you opt for the path to litigate and persevere throughout, you can spend a bunch of money and still end up empty-handed or in the red at the end.

[Ie, as Red Hat recently demonstrated to a stubborn attacking patent troll....]

David had a sling, not a slingshot. Nebecanezzer had the patent on slingshots so David had to use the old fashioned sling on which the patent had expired.

Steve Stites



However, a slight fact check problem. David is alleged to have lived 1040–970 BC

While Nebeebee was around 605 to 562 B.C

So Doovid bubbaleh was first to invent and publicly use the shepherd's "sling", thereby invalidating Nebeebee's later-in-time patent.

On the other hand (there's always an other hand in these Talmudic discussions), the patent codes of Hammurabi and Babylon did not recognize technology disclosures written in foreign tongues such as the Hebraic one. Thus the Supreme IP Court of Babylon ruled that David was an abstraction and he had never sling-bladed Goliath down with a mere single pebble. It was a rural legend.

Fact check sources:


Many people who help companies such as Red Hat build defensive patent portfolios have an uneasy feeling at the same time. "What if we get acquired by a company hostile to our ideals and plans about a defensive-only portfolio?"

That appears to be exactly what is happening right now with the situation of Sun patents acquired by Oracle. This article summarizes James Gosling's blog post:


In the original post (http://nighthacks.com/roller/jag/entry/quite_the_firestorm ) Gosling reveals that Sun engineers had a contest amongst themselves to make patents out of clearly unpatent worthy processes in order to see who could get the most ridiculous patent through. Gosling points, as an example, to one of his patents that describes an un-novel power switch.

So the engineers were making defensive-only patents in an environment that actually eschewed them, and they unwittingly set up the current situation with Oracle owning a large patent portfolio filled with potentially unwarranted simple-machine software patents and a willingness to use them as business leverage or for straight lawsuits (i.e., offensive instead of defensive.)

If you were in the position of advising a group of open source engineers working at a company such as Red Hat who were concerned about this, what would you recommend they do and think?

Do you think there is any value other than the good story aspect to Gosling revealing that some of the patents currently under litigation might have been intended as jokes instead of representing serious innovation?

Red Hat's patent promise seems like a good place to start, although there may be quite a bit of room for improvement.

"Step back": You're saying, get yourself a gun before someone shoots you. Well, you may or may not be right (I don't think you are). But that's totally irrelevant to the discussion, which is whether software patents should be permitted. You seem to agree that software patents can hurt innocent people, and that would suggest you agree that they are bad policy.

I won't bother replying to the other points you made, where you attributed opinions to me that I don't hold and then refuted them. That's the silliest debating game in town.

All I'm saying is to not wipe all your MAD options off the table before the negotiations even begin. ;-)

step back, I think your best long term play, if you really represent engineer-trolls rather than large corps in disguise or patent attorneys at large, is to seek limits to patent law where certain groups are immune from software patents but others are not. Your proposals here are not profitable or enjoyable to the majority of engineers and hurts consumers. On the other hand, you might garner sympathy and some degree of support if you turn your guns unambiguously at the largest of closed source companies (including, eg, monopolists), as these are the ones less likely to have the population rush to their defense as the points of facts are made known through growing public debate. Support, in the minimum, safe harbor for open source and perhaps as well for the smallest of entities no matter closed or open source be their business dealings.

On second though, being as unreasonable as possible might just work!

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License.