pageok
pageok
pageok
[Adam Mossoff, guest-blogging, April 28, 2009 at 11:10pm] Trackbacks
Who Cares About the Invention of the Sewing Machine?

I first want to thank Eugene for inviting me to blog on the invention and commercialization of the sewing machine in the antebellum era. I've long been a consumer of the Volokh Conspiracy, and it's quite exciting to experience it from the producer's side. More important, I'm really looking forward to sharing my research and to receiving some valuable feedback. So, let's jump in!

As someone who writes on topics in legal history, I often face the question: Who cares? I appreciate this question, as I've always believed it's critical for an academic to connect abstract identifications to concrete reality. I will give an in-depth answer to that question at the end of my posts, after I have identified the facts of the invention and commercialization of the sewing machine that make it possible to infer lessons for the modern policy debates. For those readers who may not have a burning personal interest in the history as such, I hope this introductory posting will set enough of the modern policy context to keep you interested.

There is currently a raging debate in the patent literature about patent thickets. A "patent thicket" exists when too many patents covering individual elements of a commercial product are separately owned by different entities. This concept is not unique to patent law; it is based on Professor Michael Heller's theory of the anticommons in real property, which arises when there is excessive fragmentation of ownership interests in a single parcel of land. According to economic theory, the problem of such excessive fragmentation of ownership interests is straightforward: It increases transaction costs, accentuates hold-out problems, and precipitates costly litigation, which prevents commercial development of the affected property.

The debate centers on whether patent thicket theory accurately explains or predicts such problems in practice, and the empirical studies produced thus far are arguably in equipoise. In speaking about anticommons theory, Professor Heller acknowledges that "the empirical studies that prove — or disprove — our theory remain inconclusive." Nonetheless, in the patent literature and in the popular press, vivid anecdotes abound about patent thickets obstructing development of new drugs or preventing the distribution of life-enhancing genetically engineered foods to the developing world.

Given the heightened interest today amongst scholars and lawyers concerning the existence and policy significance of patent thickets, a historical analysis of the sewing machine patent thicket in the 1850s — called the "Sewing Machine War" at the time — and the denouement of this patent thicket in the Sewing Machine Combination of 1856 is important.

On one hand, it is an empirical case study of a patent thicket that (temporarily) prevented the commercial development of an important product of the Industrial Revolution. The sewing machine was the result of numerous incremental and complementary inventive contributions, which led to a morass of patent infringement litigation given overlapping patent claims to the final commercial product. The Sewing Machine War thus confirms that patent thickets exist, and that they can lead to what Professor Heller has identified as the tragedy of the anticommons.

On the other hand, the story of the sewing machine challenges some underlying assumptions in the current discourse about patent thickets. One assumption is that patent thickets are primarily a modern problem having to do with recent changes in technology and law. Professor Heller explicitly makes this point in his recently published book, The Gridlock Economy:

There has been an unnoticed revolution in how we create wealth. In the old economy — ten or twenty years ago — you invented a product and got a patent .... Today, the leading edge of wealth creation requires assembly. From drugs to telecom, software to semiconductors, anything high tech demands assembly of innumerable patents.

In fact, Professor Heller's first foray into patent thicket theory was in assessing an anticommons in "biomedical research," which he and his co-author, Professor Rebecca Eisenberg, saw as a logical result of extensive patenting of biotech research tools. Despite Heller's and other scholars' off-hand references to earlier patent thickets, the principal focus of the theoretical and empirical studies of patent thickets is on very recent inventions in high-tech and science — computers, telecommunications, and biotech.

A second assumption is that patent thickets are a property problem — too much property that is too easily acquired that results in too much control — and so they are best addressed by limiting the property rights secured to patentees. Specific proposals have called for eliminating contractual rights in patented drugs, authorizing federal agencies to terminate patent rights to avoid patent thickets, and even prohibiting the patenting biotech inventions altogether. Although the empirics have yet to confirm patent thicket theory, this has not stopped people from proposing numerous regulatory or statutory measures to redefine and limit property rights in patents.

The story of the invention and development of the sewing machine challenges these two assumptions insofar as it is a story of a patent thicket in an old technology. More important, it is a story of the successful resolution of a thicket through a private-ordering mechanism: a contractual arrangement in which patent-owners cross-license the right to use their respective patents. This cross-licensing contractual arrangement is called a "patent pool," and it so happens that the first patent pool in United States history was the one that brought the Sewing Machine War to an end: the Sewing Machine Combination of 1856. In fact, the Sewing Machine Combination successfully coordinated the overlapping property claims in the sewing machine until the last patent in the pool expired in 1877.

As I researched the Sewing Machine War, I further discovered that this particular patent thicket is an important case study because it encompasses many other issues in patent law that are often intertwined today with concerns about patent thickets. For instance, there was massive and costly litigation between multiple parties in multiple venues. One of the patentees in the Sewing Machine War engaged in an extremely costly search for evidence of prior inventions (what patent lawyers call "prior art") that would invalidate the other sewing machine patents. There was hard-fought and expensive satellite litigation in the courts and in the patent office. In other words, the Sewing Machine War reflected all of the concerns expressed today about an alleged explosion in out-of-control patent litigation, and yet one finds the same conditions existing in the patent system more than one hundred years ago.

Last, but certainly not least, I also discovered that the Sewing Machine War was precipitated by patent infringement lawsuits filed by Elias Howe, who was America's first "non-practicing entity" (NPE). As an aside, commentators and judges sometimes call a NPE a "patent troll." Interestingly, the term "patent troll" has proven extremely difficult to define with precision, and perhaps one of the lessons of the Sewing Machine War is that it is an unhelpful rhetorical epithet.

Elias Howe was a NPE, because he commercially exploited his patent by licensing sewing machine manufacturers. He also obtained injunctions against those who refused to pay him royalties, and given that Howe's 1846 patents was valid, he obtained many of these injunctions. This is interesting, because patent scholars, as well as some Supreme Court Justices, believe that "patent trolls" arose only recently and that they are undermining the proper functioning of the patent system. Such scholars and Justices may be surprised to learn how these so-called modern problems have long existed within the historically successful American patent system.

The Sewing Machine War came to an end with the voluntary formation of the Sewing Machine Combination. In this respect, the Sewing Machine Combination reveals how patent-owners have substantial incentives to overcome patent thickets, even in contexts in which so-called "patent trolls" are exploiting the strong enforcement of the property rights in patents. The Combination was formed though pre-existing private-ordering mechanisms, such as contract and corporate law, and not through judicial decisions, PTO regulations, or statutes that limited or restricted the patent-owners' property rights. In sum, the Sewing Machine Combination reveals the innovative ways in which patent-owners can rescue themselves from commercial gridlock, and in so doing, unleash an explosion in productivity and innovation.

In my next posting, I will discuss the invention of the sewing machine, which was an effort that took place over a hundred years (approximately from the mid-1700s to the mid-1800s). The result of this lengthy inventive process was that the sewing machine — a tremendous technological accomplishment — was developed through incremental inventive steps. By the early 1850s, the inventive contributions by so many different people over so many years resulted in a product in which there were many differing (and legitimate) property claims. This set the stage for the Sewing Machine War, and thus to understand how this thicket arose, it is necessary to identify its first cause. I will discuss this part of the story in my next blog posting.

(If anyone is interested in seeing the sources for the quotes and other factual claims in my posts, please feel free to download my paper, A Stitch in Time: The Rise and Fall of the Sewing Machine Patent Thicket, where you can see the sources in all their glory. As everyone who reads law journal articles knows by now, law professors love citations so much, we put them on almost every sentence, phrase and word. :) )

Bruce:
"[V]ivid anecdotes abound about patent thickets obstructing development of new drugs or preventing the distribution of life-enhancing genetically engineered foods to the developing world."

Well, that settles it then! More seriously, this is a fascinating project. You may address this later, but it occurs to me that Heller still has an out after your paper: namely, that you've found one earlier instance of a patent thicket, but that the problem he is identifying is one of scale. Namely, that modern technological developments have made it much, much more likely that patent thickets will evolve covering a wide number of products. I think this will come down to how typical you think the sewing machine situation was during the industrial revolution.
4.28.2009 11:54pm
Adam Mossoff (mail):
Hi Bruce,

That's a very interesting idea about how the anticommons theorists might respond to this historical case study.

If they do take that tack, then they'd have to back it up with solid empirical evidence on the exact ways in which modern technology differs from old technology, and then show the causation (not just correlation) between these differences and the problem of patent thickets today. I'm skeptical that such an empirical project could be done, but I don't think it's impossible. I actually think such a research project would be very interesting, and it would do much to advance the policy debates in this area.

BTW, I do think there are some important differences for patent-owners between the Sewing Machine War and today, but you'll have to wait to read my final thoughts on the policy pay-offs. Hah! How's that for a teaser? :)

Thanks again for the feedback!
4.29.2009 12:36am
Grigor:
I am hoping that these, er, threads may answer what to my mind has always been among the greatest of mysteries: why is it that shops that repair sewing machines always -- always -- repair vacuum cleaners also? Is there a legal reason? Did vacuum cleaners emerge from the same patent thicket as sewing machines, back in the primordial ooze of household appliances? Or is it an engineering reason on which the law can shed no light?
4.29.2009 1:12am
Frater Plotter:
One major way in which modern technology (notably software) differs from older technology (e.g. sewing machines in 1850) is in the availability of invention and manufacturing to the typical hobbyist.

The tools and resources needed to build a sewing machine in the 1850s are radically different from those available to a typical user of sewing machines. A maker of sewing machines needs machine tools and a foundry, as well as a ready supply of raw materials and fuel. In contrast, a person interested to write and publish software today needs precisely the same tools that are needed to use software: namely, a computer -- and, in this day and age, an Internet connection.

A typical personal computer can be equipped with the software tools needed to develop software for zero additional cost. No foundries or machine tools are involved. Thus, there is no capital outlay involved in creating software. Moreover, there is no incremental labor or materials cost: once you have produced the first copy of your software, making the second and the millionth copy costs nothing. You don't have to buy steel and coal to run your software factory; you just upload it to a Web site and let the Internet make copies.

This has two consequences regarding patents:

First, having to pay any per-copy fees, such as patent license fees, is especially burdensome because it changes a zero-cost transaction (making additional copies of a piece of software) into a costly transaction. If you have to pay patent fees, you don't have the option of giving your software away for free -- which is the typical behavior of many, many software producers, be they Mozilla (Firefox), shareware games creators, or Linus Torvalds.

Second, the potential targets of patent law are many more in number. The number of people in the world who can raise capital to start a sewing-machine factory is very small. The number of people who can write and distribute software is much, much larger. Because there is no capital outlay involved, software (as a field) naturally invites hobbyists, part-timers, academics whose work might otherwise stay confined to journals, semi-professionals with mundane day jobs, and self-taught tinkerers learning by doing. All of these are deterred by the threat of being liable for patent violations in the stuff they write and give away for free.

A knock-on effect: Both innovation in the field of software, and the development of the skills of its practitioners, rely heavily on the accessible and ad-hoc nature of the field. There are schools which teach "software engineering", but it is widely known in the field that a graduate of such a school who does not also have independent work to show for themselves is counted as vastly inferior to one who does have such work. And it is that independent work which is imperiled by the enforcement of patents.
4.29.2009 1:23am
MCM (mail):
What about airplanes in the early 20th century? From what little I know, it seems we have another example of a patent thicket that required strong government intervention to resolve.

Elias Howe was a NPE, because he commercially exploited his patent by licensing sewing machine manufacturers. He also obtained injunctions against those who refused to pay him royalties, and given that Howe's 1846 patents was valid, he obtained many of these injunctions. This is interesting, because patent scholars, as well as some Supreme Court Justices, believe that "patent trolls" arose only recently and that they are undermining the proper functioning of the patent system. Such scholars and Justices may be surprised to learn how these so-called modern problems have long existed within the historically successful American patent system.


I think you're ignoring a few elements that are commonly (but not universally) ascribed to patent trolls:

-Trolls don't innovate. I assume Elias Howe personally created the inventions covered by his patents. Trolls arguably encourage innovation by raising demand for patents, but trolls are generally perceived to acquire patents, not create them.

-Trolls don't license in good faith, if they do license. They acquire patents in order to prevent competitors from accessing a market, or demand confiscatory licensing fees (to make licensing unfeasible). Elias Howe apparently did license his inventions at marketable rates.

More important, it is a story of the successful resolution of a thicket through a private-ordering mechanism: a contractual arrangement in which patent-owners cross-license the right to use their respective patents. This cross-licensing contractual arrangement is called a "patent pool," and it so happens that the first patent pool in United States history was the one that brought the Sewing Machine War to an end: the Sewing Machine Combination of 1856. In fact, the Sewing Machine Combination successfully coordinated the overlapping property claims in the sewing machine until the last patent in the pool expired in 1877.


This is interesting, I thought, because it immediately brought to mind how computer manufacturers deal with patent thickets today. There is a great deal of pooling, because there are so many patents intertwining that nobody could possibly do anything if anybody enforced their respective patents.

Doesn't this maybe suggest that our patent system is outdated and badly in need of reform? Just because parties can contract around stupid laws doesn't mean that the law isn't stupid.
4.29.2009 1:38am
Splunge:
Goodness, discussion of an event that actually permanently changed peoples' lives for the better, strewn no doubt with concrete facts. I await the denouement with bated breath, humming catchy themes from the overture while I dogear my program hunting drollery in the bio sketches.

As one with a long interest in the history of science and technology, I find the "assembly" argument silly. Relative to the technology, Notre Dame was a hell of a lot harder to assemble than a 747 is today. And so forth.

The only obvious distinction I can imagine, off the top of my head, between today's patentable tech and yesteryear's is that today's patentable physical engineering tech has reached a point of sophistication (e.g. takes place on nanometer scales) that only monster amounts of capital can exploit it. Really, with a mere handful of exceptions, it's hard to imagine an individual being able to exploit a modern engineering patent on actual machinery, a chemical compound, or some other actual device.

On the other hand, the explosion of general-purpose computing resources has lead to a very ripe field of invention in the area of intangible engineering -- algorithms and programs, for example -- and this is one area where individuals and small firms can exploit patentable tech with mere M$, not G$, of capital. From your precis above, this arguably sets the stage for "patent thickets," inasmuch as the "property" here is cheap in tangible terms to "acquire" (you just need to think of it first).

My vague impression is that the PTO has responded to this situation by allowing intangible tech patents to encompass surprisingly wide claims, e.g. it appears possible (IANAPL of course so this may be wrong) to patent an algorithm based largely on its function only, so that someone who comes along later and implements the same function but far more efficiently is mostly SOL. Contrast this with, say, patents in chemistry, where altering a drug molecule to merely reduce side-effects (i.e. improving efficiency without touching the function, nor even the process by which the function is achieved) seems a perfectly ordinary way to defeat a patent. This seems...inconsistent.

Perhaps at some point your disquisition on the tricky question of how people successfully drew intellectual property lines then will be illuminating of how we might draw them now.

Phoo, I see the conductor and first violinist have gone off together arguing. I'm going to fetch Jujubes and Diet Coke from the snack bar.
4.29.2009 1:41am
Frater Plotter:
On the other hand, the explosion of general-purpose computing resources has lead to a very ripe field of invention in the area of intangible engineering -- algorithms and programs, for example -- and this is one area where individuals and small firms can exploit patentable tech with mere M$, not G$, of capital.

If by "M$" you mean "thousand$", you're close. If you mean "million$", you're way off.

These days, it requires only hundreds of dollars (would that be "C$" or "H$"?) of capital, since that's the price of a personal computer ... plus somewhat more in yearly expenses for Internet access and a Web site to distribute the software from. The costs of human time and education involved vastly dwarf these outlays ... but, of course, one reason that individuals set out to write software is to learn how or to prove that they can.
4.29.2009 2:08am
Fub:
Splunge wrote at 4.29.2009 1:41am:
Really, with a mere handful of exceptions, it's hard to imagine an individual being able to exploit a modern engineering patent on actual machinery, a chemical compound, or some other actual device.
Maybe this is too old for your "modern" criterion, or maybe it's one of that handful of exceptions. But this (USPTO #4,429,609) was a modern engineering patent when invented, and arguably still is modern engineering. Its lone inventor realized it in hardware, and exploited it through licensing of rights to manufacturers. I saw the prototype in 1981. Devices using it still sell well.
4.29.2009 3:15am
Donna B. (mail) (www):
I think most of the commenters so far have confused incremental progress with leaps in progress. If I'm not mistaken, the leap in sewing machines was the interlocking stitch. Incremental improvements on that are numerous. I welcome corrections.

Where computers are concerned... software, in that is it is mathematical in origin came first, did it not? Hardware was simply a way of computing faster and more accurately. If software is a mathematical derivative... can it be patented?
4.29.2009 3:47am
Duffy Pratt (mail):
Goretex and the stent are two fairly modern examples where individuals did rather well for themselves.
4.29.2009 3:48am
Montana:
"These days, it requires only hundreds of dollars (would that be "C$" or "H$"?) of capital, since that's the price of a personal computer"

I would even argue that it doesn't cost anything. Saying that you need to buy a computer to write software is like saying that you need to buy a pair of shoes to walk to the store. Technically it may be true, but I reckon most people already have one.

But it seems like hobbyists are one of the contributing factors, not the real driver. The real problem seems to be that the pace of innovation has exceeded the ability of patents to expire before becoming a thicket.

Imagine the problems that would arise if patents lasted 100 years and IBM tried to introduce the personal computer in 1981: Mechanical typewriter patents would cover the keyboard. Electrical typewriter patents would cover the printer. Television patents would cover the monitor. Transistors, electric fans, molded plastic, magnetic storage, LEDs, voltage regulators, printed circuit boards, electrical plugs, bus multiplexers, the list goes on.

In software we see an analogous amount of innovation occurring in less than 20 years and the same result presents. By the time the patents on the fundamentals of the field expire, the field has fundamentally changed and there are new patents on the new fundamentals.
4.29.2009 3:59am
markm (mail):
Adam: You may have overlooked some issues:

1. Patent pools are common nowadays, but as usually implemented they only solve the "patent thicket" problem for the incumbents. The incumbents gain licenses to all the pooled patents, but don't ease the problem a new entrant would face in negotiating licenses with each incumbent. It isn't in the incumbents' interest to allow new competitors. Did the Sewing Machine Combination provide a single point to buy licenses? (Elias Howe would have different interests in this respect than incumbent manufacturers.) Did any new manufacturers enter the business between 1856 and 1877?

2. Furthermore, modern patent pools are often effectively eternal - before the old patents expire, new patents have been added.

3. Did Elias Howe offer new product designs and improvements to manufacturers for a fee, or did he wait for others to do the work of designing a producible product somewhat resembling his patent, and then sue? That's the difference between a legitimate NPE and a troll.
4.29.2009 7:18am
Mark in Texas (mail):
Professor Mossoff

Could you tell us what sort of differences existed between the situation where a patent thicket restricted further development of the sewing machine until patent pools got around the problem and Eli Whitney's invention and patent of the cotton gin which was widely copied and the patent ignored.
4.29.2009 8:32am
jviss (mail):

Really, with a mere handful of exceptions, it's hard to imagine an individual being able to exploit a modern engineering patent on actual machinery, a chemical compound, or some other actual device.


This is a failure of imagination, not a reflection of reality.

There are key phrases and words here the place this perception squarely in the realm of unwashed masses:

- individual
- actual machinery (versus virtual? or software?)

It is quite reasonable for an individual, acting alone or as part of a team or consortium or company, to invent something that has a mechanical realization, and then "exploit it;" whether that means going into production, selling the invention, selling the company that was established to produce the invention, or even just collecting the invention incentive award from one's employer and listing the patent on one's resume.

There are approximately 3500 new patents issued weekly. A small, but clearly nonzero number represent this type.
4.29.2009 8:51am
PersonFromPorlock:
Splunge:

Really, with a mere handful of exceptions, it's hard to imagine an individual being able to exploit a modern engineering patent on actual machinery....

I have an improvement to lawn tractors, which I built in a day with ordinary hand tools several years ago, which has been proving itself very useful ever since. So far as I know it's novel and would be easy to incorporate into existing lawn tractor designs at essentially no cost.

So what's holding me back? Simply the sheer cost of applying for a patent, and the thought of further costs to defend one should it issue. I suggest that these factors inhibit individual inventors to such an extant that the appearance is that only corporate players are left; but that the limiting factor, in many cases, isn't the complexity of the technology but the cost of the patent process itself.
4.29.2009 9:01am
geokstr (mail):
An analogous situation occurs in the entertainment field, specifically movie and television productions.

More than one studio may own the rights to a given picture or series, and the proceeds of the sale or usage fee of every unit have to be meticulously accounted for, because there are also dozens of separate contracts with various talent used in making the original. Actors, producers, writers, composers, musical artists, directors and others can have contractual rights to part of either the gross or net proceeds, which can vary for each "window of revenue" for each individual participant. Even the definition of cost and revenue varied for every participant based on their bankability or negotiating power.

For instance, a typical movie for theatrical release goes through a number of stages timed to generate the maximum revenue at each stage before entering the next: domestic theatrical release, international theatrical, video for rental, video for sale, pay-per-view TV, subscription cable TV, standard cable TV, free TV, some of which can go on forever. At each stage, there are different costs involved in production, manufacture, distribution, promotion and overhead. In the area of video especially, there were a multitude of ways to release, for example as part of franchise set or a genre collection, where the revenues and costs would have to allocated to the individual titles. Every one of the participants has the right to independently audit all the records to ensure they are getting their contractual shares.

As accounting manager for 20th Century Fox home Video for 7 years, I can assure you that most of the participants exercised their rights to audit. I was also in charge of the audit support, and heavily involved in the computer systems development.

This added tremendously to the accounting and legal costs involved. One movie title could have hundreds of different methods of release, multiplied by the dozens of participants. The complexity added to the computer systems because of this was astonishing.
4.29.2009 9:15am
corneille1640 (mail):
I would like to know how (or whether) antitrust laws would change the way patent pools could work and how these laws might affect the "problem" of patent thickets. I guess I'll have to look forward to reading the future posts!
4.29.2009 9:21am
DDG:
Corneille,

Antitrust laws (and the doctrine of patent misuse) are significant limitations on patent pools and related structures (such as standard setting organizations) Unfair competition law is one of the reasons why we should be careful when drawing analogies between the modern day and the mid-19th C. The fundamental patent-antitrust cases are from the early-mid 20th C.
4.29.2009 9:48am
Liberal Libertarian:

It increases transaction costs, accentuates hold-out problems, and precipitates costly litigation, which prevents commercial development of the affected property.


Minor point: don't you mean optimal development or "highest and best use" development, not just commercial development?
4.29.2009 9:59am
PA (mail):
MCM said:"-Trolls don't license in good faith, if they do license. They acquire patents in order to prevent competitors from accessing a market, or demand confiscatory licensing fees (to make licensing unfeasible). Elias Howe apparently did license his inventions at marketable rates. "

You obviously don't know much about trolls. Do you seriously believe that anyone asserts patents just for fun? Well, patent lawyer, maybe ... but with OPM...

Whether developed or acquired, a patent owner's incentive is to profit. If I am a troll, what benefit do I get from keeping you (the producer) out of a market or otherwise "mak[ing]licensing unfeasible"? I want producers to enter markets and make profits so there is a financial incentive to take a license. The worst patents from a patent troll's perspective are those that read on an unprofitable product, where the potential damages are far too low to justify the risks of litigation.

The patent troll has every incentive to license as quickly and as efficiently as possible ... They license almost exclusively through litigation for a variety of reasons, but that alone doesn't make it "bad faith."
4.29.2009 10:06am
Ken Arromdee:
There are approximately 3500 new patents issued weekly. A small, but clearly nonzero number represent this type.

That doesn't follow. The fact that there are a lot of them doesn't necessarily mean there's even one of any specific type.
4.29.2009 10:06am
Bob White (mail):
Glad to see others have already raised the antitrust issue-not a problem for the sewing machine as it antedates the Sherman Act, but a real problem for a patent pool these days. A course I took in law school looked at the DVD patent pool, which IMO was clearly illegal and economically efficient, and which was not prosecuted by the (Clinton) DOJ.
4.29.2009 12:03pm
MCM (mail):
MCM said:"-Trolls don't license in good faith, if they do license. They acquire patents in order to prevent competitors from accessing a market, or demand confiscatory licensing fees (to make licensing unfeasible). Elias Howe apparently did license his inventions at marketable rates. "

You obviously don't know much about trolls. Do you seriously believe that anyone asserts patents just for fun? Well, patent lawyer, maybe ... but with OPM...

Whether developed or acquired, a patent owner's incentive is to profit. If I am a troll, what benefit do I get from keeping you (the producer) out of a market or otherwise "mak[ing]licensing unfeasible"? I want producers to enter markets and make profits so there is a financial incentive to take a license. The worst patents from a patent troll's perspective are those that read on an unprofitable product, where the potential damages are far too low to justify the risks of litigation.

The patent troll has every incentive to license as quickly and as efficiently as possible ... They license almost exclusively through litigation for a variety of reasons, but that alone doesn't make it "bad faith."


I think you missed my point. Entities sometimes acquire patents solely to ensnare's a competitor's products.

I make product A using patent X. My competitor makes a competing product, product B using patent Y. If my competitor doesn't outright own patent Y, I acquire it and get an injunction against my competitor. Or maybe slightly earlier patent Z implicated product B/patent Y, so I acquire patent Z.

Now I basically withhold the license either by refusing to license or demanding a rate that makes product B commercially non-viable. Of course the incentive is product the whole time.

If I am a troll, what benefit do I get from keeping you (the producer) out of a market or otherwise "mak[ing]licensing unfeasible"?


You keep the market to yourself for your own competing product, based on a different patent.

Seriously, did you need me to explain that it's a rough world out there and that people don't always play nice?
4.29.2009 12:20pm
MCM (mail):

Of course the incentive is product the whole time.


product -> profit
4.29.2009 12:22pm
Bruce Hayden (mail):
There are possibly de facto patent pools in the semiconductor and electronics industries. There are definitely thickets. At least when I was involved there a decade ago, you couldn't build an IC or a computer without infringing the patents of dozens of companies, and, likewise, they couldn't do so without infringing yours. The result at times was a Mexican standoff, and at others, cross-licensing.

This is where the difference between NPEs and manufacturers came in. In the case of other companies in the same industry, litigation was uncommon simply because they could shut you down as quickly as you could shut them down (except when they filed in the TX Rocket Docket and you didn't). The NPEs didn't face this, and so would resort to litigation first, not last.

BTW - recent case law has significantly limited preliminary injunctions, etc. for the NPEs, and I think that good. It is very hard to see irreparable harm when the plaintiff doesn't compete, but merely collects revenues.
4.29.2009 1:22pm
Bruce Hayden (mail):
I still am unconvinced that the problem with software in terms of patents is how easy it is to create software, and think that at least some of the problem is that of the prior art. The USPTO was dragged kicking and screaming into the area, and as a result, much of the previous art in the area never made it into patents and patent applications, which is still the almost exclusive source for prior art by examiners.

BTW - we are starting to get Bilski rejections of software claims. Yes, it wasn't supposed to happen. But we all know that it would. some examiners are now taking the position that loading different software into a general purpose computer no longer makes it a specialized machine, and storing it on a floppy disk (CD, etc) doesn't make that novel either.
4.29.2009 1:29pm
Ken Arromdee:
Some of the problem with software is prior art. Other problems include the fact that "obvious" for software patents means something utterly divorced from what a human being would find obvious. Often, something that's obvious in any normal sense, and would be independently "invented" by anyone in ten seconds, hasn't existed yet because it isn't practical until technology advances. Then it gets patented, and the fact that nobody used it before is taken to be a sign that it's not obvious rather than that it's blatantly obvious once you have cheap computers but computers only got cheap enough yesterday.
4.29.2009 2:38pm
einhverfr (mail) (www):
In general, patent thickets are liveable problems in pharmaceutical and biotech industries. It is rare that you have more than a few patentholders involved in most of these products or projects, and hence even a major patent thicket there is not a tremendous obstacle. Furthermore, where the patents exist, they tend to be of reasonably certain scope so it is reasonably possible to know in advance whether or not a given product infringes.

However, my industry (software development) is completely different. There are millions of vague patents of uncertain scope of which any piece of software might be argued to infringe. Most software patents are sufficiently vague I can't read them and know if anything I have written actually infringes. Furthermore, given the complexity of systems, I can't know whether something in my dependency tree infringes even if my code doesn't and even if the patent is clear as to what it protects. This is a problem which is of a severity that is unparalleled in our history.

One of the fundamental issues here is that patent enforcement becomes fundamentally arbitrary. Many engineers such as myself just ignore the problem. It is no use worrying about a problem one fundamentally cannot avoid.

Not surprisingly, the IEEE (of which I am a member) a couple years ago did a breakdown of which companies supported and which companies opposed patent reform. They were able to show that the IT companies generally supported it (both hardware and software) because of the complexities of dealing with patents in those environments, while biotech companies typically opposed reform.

I personally thing the solution would be to exempt certain types of products from patent enforcement, and these should include software (at a minimum), and possibly integrated circuits as well (though actual manufacturing processes of microcircuitry might be patent-worthy).
4.29.2009 2:45pm
Harry Eagar (mail):
Curse you, Grigor. Now I know I'm not going to be able to sleep tonight.
4.29.2009 4:48pm
Gregory Howe (mail):
I noticed this story and thought I would chime in. You see, my great great great grandfather was none other than the New England machinist Elias Howe, the inventor of the interlocking stitch. We have seen none of his great fortune down through the years. He seems to have spent most of his court winnings around the time he won them. Some of my family still lives in New England, and we're all still engaged in crafts and technology (furniture design, art, VOIP, expert systems, and web applications). I don't have anything substantive to add to this fascinating discussion, but I will say on a personal note I've recently wondered what would have happened had he not taken that fateful trip to England to sell his ideas. It is a shame that communication
4.29.2009 4:48pm
Gregory Howe (mail):
...had not reached today's levels. He might have gotten wind of what Singer was up to back in the states.
4.29.2009 4:50pm
Splunge:
I see some of the responses to my essay have forgotten the old maxim, the plural of 'anecdote' is not 'data.' It would be interesting if someone could provide an example of a concrete device technology game-changer like the sewing machine or the cotton gin that was invented five years ago and is being exploited by an individual right now. Good luck with that.

It's one thing to observe that individual inventors can still nibble around the edges of the known art, and invent small widgets that make small but useful (and profitable to the inventor) changes in our lives.

But from where do we expect the big changes that will reshape life in the 21st century? I would hazard mostly in biotech and bioinfotech, possibly in materials, barely possibly (but not likely) in electronics.

Parenthetically, I know that last bit seems a bit heretical, but personally I think the Golden Age of Computing has concluded, or is about to. That may seem crazy, given how amazing is the level of computation you can get on your cell phone or PC these days, or the fact that any high-school kid can and often does write near leading edge software on his Linux PC at home. But I see these as the equivalent of observing how fast and relatively cheap clipper ships were in the 1890s, evidence that the low-hanging fruit in a formerly rich area of exploitation has all been picked, and further advances will be slow and painful. The fact that the leading edge in hardware is now only slightly in advance of what's in everyone's pocket (as opposed to the light-years ahead it was in the 50s and 60s) is evidence of how much slower the leading edge is moving.

It's quite otherwise in, say, biotech. Here, the ideas and tools being bounced around in labs are light-years from what's in clinics and hospitals, evidence that the leading edge is moving fast and there's still plenty of easy fruit to pick.

But to return to the point, I just can't see any individual inventor doing very much to exploit by himself major technological advances that deal with molecules and medicine. You just need too much capital, the cooperation of too many other experts. (Some people above have confused the inventor profiting by his invention, which I hardly deny, with the question of whether an inventor, unsupported by a large team or large amounts of capital, can develop his idea into practicality.)

On the other hand, this seems clearly not true in the area of algorithm design, where an individual can readily develop his idea himself. (Although the increasig dependence of modern software on things like libraries is an argument against that position.) That is, a key fact about 21st century tech in some areas is that we've successfully separated, in some areas, pure function and implementation, by designing general purpose machines, like computers. This allows an inventor to "realize" his invention almost just by thinking about it -- without having to build things. It's almost as if Edison could invent the light bulb just by imagining it, without having to build one, because there was some kind of "universal machine" that would implement his concept of light-emitting device, once it was "programmed" to do so.

SF writers in the 50s used to imagine a similar thing when they thought the world would be populated with robots. Instead of people figuring out new devices to, say, mine or manufacture or whatnot, they would figure out new programs for the robots to implement to do those things.
4.29.2009 5:29pm
einhverfr (mail) (www):
Splunge wrote:

Parenthetically, I know that last bit seems a bit heretical, but personally I think the Golden Age of Computing has concluded, or is about to. That may seem crazy, given how amazing is the level of computation you can get on your cell phone or PC these days, or the fact that any high-school kid can and often does write near leading edge software on his Linux PC at home. But I see these as the equivalent of observing how fast and relatively cheap clipper ships were in the 1890s, evidence that the low-hanging fruit in a formerly rich area of exploitation has all been picked, and further advances will be slow and painful. The fact that the leading edge in hardware is now only slightly in advance of what's in everyone's pocket (as opposed to the light-years ahead it was in the 50s and 60s) is evidence of how much slower the leading edge is moving.


I think this is a slight-bit premature for a couple fo reasons.

1) Different sorts of thinking skills peak at different ages. Some sorts of software may be susceptible to a brilliant high school student's abilities to write (Math computational software comes to mind). Others, I can assure you, are not. Bright high school students are best able to tackle highly technical but discrete problems. Bigger picture problems (a payroll software program capable of being loaded with easy-to-develop rules for different locale's taxes, for example) tend to be more problematic. Pure logical computation tends to peak in most people at around 22. Wider-picture processing tends to peak at about age 36, and those where wide experience is key tend to increase on average until at least age 60.

2) There are still important technical issues to be solved regarding computing. These include more general usability of quantum cryptography. Most often the boundary today is both physical (in the sense of the difficulty in delivering keys in a QC system), computational, etc.

3) I am willing to bet that computers in 40 years will be even more ubiquitous, and may even be as unrecognizable today as our current netbooks would be to the computer techs of the 1960's.

However if by "the golden age is over" you mean "the most central accomplishments are done" then that is true. I am not sure that is a very useful measure though.

As for biotech, I do agree that this area is still growing, and some areas of it show great promise (particularly in medicine). I am not convinced of the widespread value of genetically modified (and hence subject to much more intrusive patent controls than typical plant patents allow) foods. But the medical area is an area which has REAL potential for revolutions in all sorts of care.
4.29.2009 6:16pm
Fub:
Splunge wrote at 4.29.2009 5:29pm:
I see some of the responses to my essay have forgotten the old maxim, the plural of 'anecdote' is not 'data.' It would be interesting if someone could provide an example of a concrete device technology game-changer like the sewing machine or the cotton gin that was invented five years ago and is being exploited by an individual right now. Good luck with that.

It's one thing to observe that individual inventors can still nibble around the edges of the known art, and invent small widgets that make small but useful (and profitable to the inventor) changes in our lives.
That certainly moves the goalposts considerably from here:
The only obvious distinction I can imagine, off the top of my head, between today's patentable tech and yesteryear's is that today's patentable physical engineering tech has reached a point of sophistication (e.g. takes place on nanometer scales) that only monster amounts of capital can exploit it. Really, with a mere handful of exceptions, it's hard to imagine an individual being able to exploit a modern engineering patent on actual machinery, a chemical compound, or some other actual device.
And, if one knows the technological history of instrument tuners (hint: electromechanical strobes), then it would be obvious how the patent I cited changed the game entirely. It was not "nibbl[ing] around the edges of the known art." It changed the art fundamentally, and the marketplace recognized that.

Granted that it didn't feed the world's starving masses. Instead, it fed the inventor's family and helped large numbers working musicians feed themselves better, by providing a much cheaper, more accurate, more portable and easier to use device necessary to their livelihoods.
4.30.2009 1:34am
Passing through:

I personally thing the solution would be to exempt certain types of products from patent enforcement, and these should include software (at a minimum), and possibly integrated circuits as well (though actual manufacturing processes of microcircuitry might be patent-worthy).



Why should software and integrated circuits be exempt from patent protection and not other technologies? What is the overarching principle that declares software and integrated circuits so special that they should be exempt from patent protection?
4.30.2009 10:49am
LaoK:
Splunge writes:


But to return to the point, I just can't see any individual inventor doing very much to exploit by himself major technological advances that deal with molecules and medicine.


The closest example of a game-changing individual inventor in the area of molecules and medicine that I can think of is the invention of the polymerase chain reaction (PCR) by Kary Mullis (for which he subsequently won the Nobel Prize).
4.30.2009 12:42pm
Fub:
Passing through wrote at 4.30.2009 10:49am:
Why should software and integrated circuits be exempt from patent protection and not other technologies? What is the overarching principle that declares software and integrated circuits so special that they should be exempt from patent protection?
IANA patent attorney, but I can offer some rationale for exempting software (distinct from integrated circuits).

Software is essentially an abstract idea, a series of instructions or operations to be performed, rendered in a fixed medium, very similar to writing a recipe or a procedure for changing a tire. The abstract idea is not a physical thing, and can be expressed in a myriad of linguistic ways.

So, a patent on a computer program would have to be limited to a particular language on a particular computer, in order to be physically realized. Otherwise, the patent is on an abstract idea.
4.30.2009 12:44pm
rosetta's stones:
Passing,

Because it's so simple an exercise that nobody should be allowed to cast a net, and haul it in just for themselves?

I was once working on a project to provide zebra mussel control on a City of North Chicago water intake, out in Lake Michigan. Seems the mussels were catching on everywhere, and the solution was to eject a low dose of chlorine at the mouth of the intake, to discourage the little buggers.

Pretty simple deal, chemical feed, basic hydraulics, but when it came to lay out the diffuser ring out at the mouth of the intake, the partner in charge asked me to lay it out in a precise way, so as to avoid a patent some guy somewhere was holding (and running around the Great Lakes cashing in, allegedly).

A patent on a diffuser ring? Plastic pipe and fittings... standard pipework materials and methods... in a municipal public utility application? How does somebody manage to shanghai a diffuser ring out of the public domain? They might as well patent the tomato plant layout in my garden.

Software is getting like Home Depot home repairs. Everybody's doing it, even a dummy like me. The saleable feature is the guy/company doing the software, and their support, not the software itself, it seems to me. And in his post just above, einhverfr eloquently speaks of the value of those people.
4.30.2009 1:06pm
einhverfr (mail) (www):
Passing Through:

Why should software and integrated circuits be exempt from patent protection and not other technologies? What is the overarching principle that declares software and integrated circuits so special that they should be exempt from patent protection?


Should as in public policy and that the law should be changed to make it work this way.

Simple reason is that the patents are found in these areas to discourage the useful arts and sciences.

I can't avoid possibly infringing on huge numbers of software patents in my work and my only reaction is to ignore the problem and figure that there is sufficient risk to anyone who might sue me that it won't happen.
4.30.2009 1:07pm
Passing through:

Software is essentially an abstract idea, a series of instructions or operations to be performed, rendered in a fixed medium, very similar to writing a recipe or a procedure for changing a tire. The abstract idea is not a physical thing, and can be expressed in a myriad of linguistic ways.


Abstract idea? Anything can be considered an abstract idea. An automobile engine is the abstract idea of the Otto cycle embodied in a construct of steel/aluminum/etc. Most inventors form the "abstract idea" of their inventor in their head before implementing their idea. As such, you haven't shown anything about software that makes it so special from other technologies.


So, a patent on a computer program would have to be limited to a particular language on a particular computer, in order to be physically realized. Otherwise, the patent is on an abstract idea.


Does a patent to a new automobile engine require (for example, the face of the piston being shaped like a hemisphere) that the engine be constructed by a particular material? or be machined/casted using a particular type of machine?

Why is it neccessary to limit a computer program to a particular language on a particular computer?
4.30.2009 2:24pm
Passing through:

Because it's so simple an exercise that nobody should be allowed to cast a net, and haul it in just for themselves?


So, you are saying that patents shouldn't be given out to simple inventions? What if the patent on the software is something that took 20,000 man-hours from a PHD to create? Would it pass your "simplicity" test then? How would you treat the difference between a patent to an improved bird feeder that took somebody 10 minutes to create versus an invention to software that took 1 year to create?


A patent on a diffuser ring? Plastic pipe and fittings... standard pipework materials and methods... in a municipal public utility application? How does somebody manage to shanghai a diffuser ring out of the public domain? They might as well patent the tomato plant layout in my garden.


Really, he patented a diffuser ring? What I think you meant to say is that he patented a diffuser ring in combination with a whole bunch of other items. As such, he didn't take a "diffuser ring" out of the public domain. He took the diffuser ring, in combination with all these other elements, out of the public domain.


Software is getting like Home Depot home repairs. Everybody's doing it, even a dummy like me. The saleable feature is the guy/company doing the software, and their support, not the software itself, it seems to me. And in his post just above, einhverfr eloquently speaks of the value of those people.


Next time, after you've written your "Home Depot" software and find that Microsoft has bundled it in with their latest OS, don't complain when everybody is going to MS for their support and not you.

Also, where does that leave people that write great software but aren't great with support or aren't great at marketing their support or don't have a recognizable name in the industry? I guess that answer to that would be "out in the cold."
4.30.2009 2:36pm
Passing through:

Should as in public policy and that the law should be changed to make it work this way.

Simple reason is that the patents are found in these areas to discourage the useful arts and sciences.

I can't avoid possibly infringing on huge numbers of software patents in my work and my only reaction is to ignore the problem and figure that there is sufficient risk to anyone who might sue me that it won't happen.


I work in the field of biotechnology -- there are too many patents there, thus, biotech patents should be eliminated.
I work in the field of traffic lights -- too many patents, so they should be eliminated as well.
... and so on and so forth and so on and so forth ....

There are innovators, there are copiers, and then there are some that do both. It seems that the copiers are the ones who always complain about patents.

Again, I asked the question, what makes software so special from other technologies? Is it just because there are more people whining about software (i.e., the squeeky wheel gets the grease syndrome).

BTW -- how many times have you been sued for patent infringement?
4.30.2009 2:46pm
rosetta's stones:

What if the patent on the software is something that took 20,000 man-hours from a PHD to create? Would it pass your "simplicity" test then?


Well first, the magnitude of the effort would imply it not to be "simple", wouldn't it? But ignoring that, assuming somebody else put in 20,000 hours and developed a system performing similarly, why would you deprive that guy of the fruits of his work? It's like you built a masonry wall, takes a lot of time to build that wall, and now you're claiming that others can't build walls? Wall building is a simple matter, if time consuming, just like coding, isn't it?



What I think you meant to say is that he patented a diffuser ring in combination with a whole bunch of other items.


No, I think I meant to say what I said. The general arrangement of the diffuser ring was removed from the public domain, not the diffuser ring and something else. That general arrangement was the only parameter in the project on which I received special guidance.

A diffuser would be a tube with holes in it, just to clarify, and there are no other "elements" to such a diffuser. I performed all hydraulics and specified the material and sizes... that is to say, I did the professional work required for the installation. The patent holder did nothing, and provided nothing... not to me or the client anyway.

Yes, MS leverages their size, no doubt. Don't have an answer for that, other than to throw in with them. All your base are belong to us.
4.30.2009 3:04pm
Passing through:

But ignoring that, assuming somebody else put in 20,000 hours and developed a system performing similarly, why would you deprive that guy of the fruits of his work?


The same can be said in any technology, so why is software special?

In all technologies, some patents are developed through a lot of hardwork and some are developed through luck/happenstance/etc. What is it about software that distinguishes it from other technologies?


Wall building is a simple matter, if time consuming, just like coding, isn't it?


My guess is that many coders would disagree that it is a simple matter. Regardless, the code from a programmer, the brick from a mason, the sheet metal form a welder, or the chemicals from a chemist can be transformed into something entirely new. Why should we treat the intellectual efforts provided by the programmer any differently?
4.30.2009 3:28pm
rosetta's stones:

What is it about software that distinguishes it from other technologies?


First, it's not hard product, it's soft. It's something "other". What other is it? Good question.

My answer is that it's thought, expressed in a language. Now, we can copyright thoughts expressed as a language, in a specific form, as many writers do. I can't steal your writing and publish it for myself. I can use the same thoughts, however. And I can use them without ever knowing you had them. Because it's common, like the diffuser ring.

That's why you hear somebody above saying they don't worry about it anymore, because they know it's not worth fishing through potential conflict issues, because they're so common. That should be an indicator, if industry professionals tell you that. Somebody's choking off the process, and hindering it, in an inefficient if not illegitimate manner.

Now, if you can catch somebody stealing your thoughts expressed in your specific form, i.e. stealing your property, then you have recourse to that theft, I would think, but not prior to that theft through the patent process. If I did my work, I wanna be paid, and I don't wanna pay somebody else for my work.





My guess is that many coders would disagree that it is a simple matter.


Simple doesn't imply easy. There's work involved, no doubt. And skill. But my advice to the coder is not to expend 20,000 hours on any application of their skill, unless it's directly billable. Don't bet-the-come.

The "thickets" that Mr. Mossoff speaks of are likely composed of many types of vegetation. Best if some of those types are never allowed to germinate, so the thicket isn't so thick.
4.30.2009 9:18pm
PA (mail):
MCM -- You're right, I completely missed your point about how you were using NON PRACTICING ENTITIES and "elements that are commonly (but not universally) ascribed to patent trolls" to make a point about PRACTICING ENTITIES and how they might use patents to their advantage. What an idiot I must be.

May god have mercy on your soul.
5.1.2009 9:03am
Passing through:

First, it's not hard product, it's soft. It's something "other". What other is it? Good question.


OK. If I have two black boxes -- one is hard-coded with a particular circuit that provides certain outputs based upon certain certain inputs, and the second is a general purpose processor running software that permits the general purpose processor to produce the same certain outputs based upon certain inputs.

The first box is purely hardware (i.e., your "hard product"). The second box is a known general purpose processor in combination with something "soft." Would you be comfortable in saying that the first block box is patentable but the second black box is not? If so, why?


That's why you hear somebody above saying they don't worry about it anymore, because they know it's not worth fishing through potential conflict issues, because they're so common. That should be an indicator, if industry professionals tell you that. Somebody's choking off the process, and hindering it, in an inefficient if not illegitimate manner.


Interesting, . If most people are ignoring patents anyway, how is it hindering anything? Also, I find it hard to believe that the software industry is somehow being choked off. For anybody who has been using computers the last couple of decades, it is unquestionable that the software that exists today is magnitudes more powerful than the software that existed in the past.
5.1.2009 1:42pm
rosetta's stones:
If the first black box is designed with industry standard materials and methods... it shouldn't be protected... just as my diffuser ring should not have been. So my "hard product" test is just a first qualifier. Now, don't let me go much further, because I haven't put much thought into this besides these few posts.

And basically, you're talking to the wrong guy, if you're asking me to locate where the line is to be drawn, because the only thing I want to do with that line is erase it. Not every line, in every case, but again, I believe the thickets are created because we're allowing them to be created.

I'm not familiar with this subject other than a few anecdotes, as you've likely noted. I suspect you're correct, that software is/has been advancing at equivalent speed with or without patent process.
5.1.2009 2:28pm
Passing through:

If the first black box is designed with industry standard materials and methods... it shouldn't be protected... just as my diffuser ring should not have been. So my "hard product" test is just a first qualifier. Now, don't let me go much further, because I haven't put much thought into this besides these few posts.


Interesting -- my response would be the following quote, which can be found in many places:


At some level, all inventions are combinations of old elements. As Chief Judge Howard Markey of the Court of Appeals of the Federal Circuit once observed, "virtually all inventions are 'combinations,' and . . . every invention is formed of 'old' elements' . . . Only God works from nothing. Man must work with old elements." Howard T. Markey, "Why Not the Statute?," 65 P.Pat.Off.Soc'y 331, 333-34 (1983).
5.1.2009 9:10pm
Breadcrumbs (mail):
Passing through,

Let me add a quote that I will steal from the Patent Prospector Blog:

"The fact is that one new idea leads to another, that to a third, and so on through a course of time until someone, with whom no one of these ideas was original, combines all together, and produces what is justly called a new invention."
- Thomas Jefferson, Director of the 1st U.S. Patent Board -
5.2.2009 8:53am

Post as: [Register] [Log In]

Account:
Password:
Remember info?

If you have a comment about spelling, typos, or format errors, please e-mail the poster directly rather than posting a comment.

Comment Policy: We reserve the right to edit or delete comments, and in extreme cases to ban commenters, at our discretion. Comments must be relevant and civil (and, especially, free of name-calling). We think of comment threads like dinner parties at our homes. If you make the party unpleasant for us or for others, we'd rather you went elsewhere. We're happy to see a wide range of viewpoints, but we want all of them to be expressed as politely as possible.

We realize that such a comment policy can never be evenly enforced, because we can't possibly monitor every comment equally well. Hundreds of comments are posted every day here, and we don't read them all. Those we read, we read with different degrees of attention, and in different moods. We try to be fair, but we make no promises.

And remember, it's a big Internet. If you think we were mistaken in removing your post (or, in extreme cases, in removing you) -- or if you prefer a more free-for-all approach -- there are surely plenty of ways you can still get your views out.