Thanks, Dave, for taking the time to respond to my comments in such an honest and forthright manner. If it's okay, I'd like to continue our little debate -- not forgetting, of course to point out where I agree with you and where we have common ground.

But first, where we differ.

You say, in reference to my point that Microsoft has never unilaterally sued anyone for patent infringement, that "The patent business has evolved to a point that it is far more efficient to threaten to sue than it is to sue."

Ah, if only that were true. The reality, of course, is that the patent litigation business is alive and well and growing (generally speaking) a lot in recent years -- a fact which both Open Source and proprietary software developers note with some concern. Obviously, there would be no patent suits if they were not useful and necessary to those companies willing to spend the average $3.5 million it costs to bring one.

Even in your own article, you hint that you don't really believe that point about the threat to sue being just as chilling as an actual suit, as when you state that "the intellectual property system is causing dollars to be spent defending against nothing."

So either patent suits are "efficient" and necessary, or they are not. Which is it?

As for IP and innovation, you simply avoided answering my request for evidence -- any evidence at all -- that patents are stifling innovation, especially in the software industry. You cite exactly ZERO facts, and instead put forward only an assumption -- i.e., that "Any dollars in a software development business that are diverted from innovation cause innovation to suffer." You may feel this is a logical assumption, but it is only an assumption, not a piece of evidence.

And in point of fact, it is NOT a correct assumption. Indeed, all available economic research -- not most, all -- demonstrates conclusively that time and money spent developing patents and intellectual property directly stimulate knowledge-sharing, innovation, and economic growth.

(Dave, I'm sending you a version of this comment with footnotes attached, so you can check out the research yourself.)

A December 2006 survey published by the French economists Francois Leveque and Yann Meniere, for example, found that 88% of all U.S., European and Japanese firms rely principally upon the information disclosed in patents to keep up with technology advances in their industries and to direct their own R&D efforts. Japanese companies and inventors, in fact, reported that patents were the number one information channel for learning about and acquiring new technologies, whereas in the U.S., respondents said patents came a close second -- after scientific publications.

That study further noted that on average across all industries, time and money devoted to the creation of intellectual property directly leads to an average boost in R&D investment of 6%. Patented technology innovation accounts for over half the growth of the U.S. economy (and some studies suggest it accounts for closer to 80% of economic growth). And as one might expect in a knowledge economy, the trade in ideas through patent licensing is today growing at twice the rate of the trade in goods.

Worldwide, in fact, economists have found that it is not capital resources but the strength of a country's intellectual property system that is the principal spur to technology development and economic growth. As one study from the National Bureau of Economic Research noted, in the absence of strong intellectual property rights, “the leading countries have insufficient incentive to invent and the follower countries have excessive incentive to copy” rather than invent for themselves.

I could go on and on, but I think you can see that there is at least some reason to believe -- and evidence to support -- the fact that IP promotes rather than stifles innovation.

And again, please show me ANY evidence that patents are stifling innovation. It's certainly not the case in the software industry, where as I've noted, the industry is less concentrated, more productive, and more encouraging of the emergence of startups and independent inventors (including Open Source inventors) than it was before patenting took off in the industry ten years ago.

But where I totally agree with you is when you state: "The reason that we open-source people don’t like software patents is because there are so many examples of poorly issued patents."

I'll go you even one better and offer this quote: "The quality of patents has suffered, many are neither novel nor useful. And the courts are overwhelmed by patent infringement and validity suits."

Do you know who said that? No, it wasn't an Open Source critic of the IP system. That statement was actually made in 1836 by Senator John Ruggles, who authored the patent reform act of that year.

There's a point here, of course. And that is that whenever the United States has undergone a major industrial renaissance -- such as during the 19th Century when the steam, telegraph, telephone and electric power industries emerged -- patenting rates have skyrocketed, as have concerns about patent quality and increased IP litigation.

Today's renaissance boom in information technology industries is no different. Once again, the patent office has been slow to adjust to the demands of new technologies and new industries. But just as we did in 1836 and again in 1870, there is every reason to believe that we will once again reform the patent system to deal with this latest upsurge in crappy patents and litigiousness.

Once again, I'd like to say that I support the Open Source movement. I believe -- and the facts show -- that it offers a viable alternative to the proprietary model of software development and plays a critically-important role in many business and social applications.

The simple truth is that Open Source and proprietary software co-exist already and will continue to co-exist. Open Source would not survive without the financial support of large, patent-holding firms such as IBM. And proprietary software appears to simply not work as well as Open Source in certain applications and roles. So each has a role to play, in my opinion.

Anyway, thanks for continuing the discussion, Dave.

Can I ask one more question, Dave, that just stumps me?

Why does Microsoft get all of Open Source's ire, whereas IBM gets a free ride?

I mean, talk about the "threat to sue" and FUD! What do you think powers IBM's $2 billion/year patent licensing business except FUD and the threat to sue? And by the way, that $2 billion a year represents more profit just from licensing patents than all but the top 65 Fortune 500 companies make from ALL sources!

In fact, IBM makes more money from patent licensing in three or four days than Microsoft will make all year.

And as for actually suing, well, IBM does it all the time. As noted previously, Microsoft has never unilaterally sued someone for patent infringement.

Sure, IBM gives away a few patents in order to buddy up to the Open Source movement. But IBM sure isn't giving away any IP in its critical areas of business, such as Websphere or services or hardware, are they?

I just don't understand why so many Open Source people think IBM is some anti-capitalist paragon.

Try crossing IBM where it makes its money and just watch what happens!

I still believe that it is more efficient to negotiate rather than to sue for all parts of the legal system. However, the efficiency of negotiating is predicated on the willingness to go to court. Larger companies have the wherewithal to compel negotiations. A perfect example is the ongoing RIAA actions against alleged distributors of copyrighted works.

As with most civil legal actions it is hoped that a simple demand letter will suffice. However, as with all bluffs, it is necessary to be willing to go "all the way" with a threat. When a company holds a shaky patent then they will be extra patient with their negotiations vs. when said company feels extraordinarily confident that their patent is solid and the alleged infringer is infringing.

I would like to see an intellectual property system where true innovators are rewarded. I'm not yet convinced that the software patents I've seen actually represent innovation.

My comment about dollars being spent defending against nothing represents my feeling that most of the patent claims are unsupportable. Just because something is unsupportable doesn't mean that a court won't support it. I remember having a discussion years ago with someone on this subject and my comment to them was, "Do you want to trust a jury with your well articulated technical argument?"

I believe that patent suits are less efficient than patent negotiations but patent negotiations have no meat without a real threat of a suit.

Evidence, evidence, evidence. It's not just assumptions on my part. The best example for me is Amazon's one-click patent imbroglio and the way this issue cluttered web-development and the courts for years. Of course, my own experiences and context have also led me to my conclusions and, yes, assumptions. I've seen many an unfounded claim over the years that cost a lot of time and money, which I would have rather spent elsewhere.

I could probably be convinced that the problem isn't the patent system but, rather, the awful patents that are issued, the lack of a true innovation requirement or the willingness to patent obvious and trivial ideas. It took me a long time, after sitting through many meetings, to discover that it's the actual claims of the patents that matter rather than all the other matter thrown in. By matter I mean the introduction to the patent and the marketing messages. I don't think it helps that in software it seems that everyone is an attorney and we have much confusion as a result.

My main issue with the velocity of innovation point you raise is my belief that the software industry would have innovated faster had it not been for these poorly issued patents. The problem with these studies is the impossibility of having a control group. I'm reminded of the centenarians who, when asked for their secret to a long life, spout off smoking, drinking and other vices as the secret to longevity. I think these people were predisposed to live 105 years and lived that long in spite of everything they did. Did the software industry innovate faster or slower than it's natural rate of innovation? I think it's impossible to answer that question.

Your quote from Senator Ruggles in 1836 could very well be deja vu all over again. It's quite likely that we're making the same mistake that past generations have made.

I believe that many government sponsored things go through a life cycle that resembles: correct a wrong, cause overcorrection, incidentally create a market for the corrective action and, finally, protect the market. In this sense, I believe that the business of IP is being protected rather than innovation itself. That worries me because it causes all of us to forget why the original protections were created. I'm further concerned that the issuance of weak patents will cause a widespread disrespect for all intellectual property laws, further undermining their original purpose.

The other question, why do I think Microsoft suffers so at the hands of the open-source community compared to IBM? Microsoft is such an easy target and they do it to themselves.

Many see Microsoft not as an innovator but, rather, as a company that takes the ideas of others, develops those ideas, over-markets the products and then seeks innovator status when attempting to protect their products. Their behavior towards open-source has also been astoundingly bad. I remember years ago visiting Microsoft because they were interested in having me do some work with them. In the introductory meeting, the only one I had by the way, I told them that their biggest mistake is hiring "consultants" to write white papers either defending Microsoft or trashing open-source. I felt that Microsoft should look towards how to work with open-source. Other companies, after all, had succeeded in doing it so why couldn't Microsoft. They then asked me to write a white paper saying how Microsoft Windows was better than Linux. Like I said, it was my only meeting with them and it was clear that they really didn't get what was happening in the market or to the technology.

I don't think that IBM has crossed the line when it comes to protecting their intellectual property. If IBM were to suddenly attempt to enforce patents on how a GUI menu unfolds against everyone then that would be a different story. IBM recognized long ago where they make their money, hardware and, more importantly, services. Enabling their customers to run all software on IBM hardware would give IBM a bigger portion of the pie. So, if someone were to get an itchy trigger finger in the software division they would need to understand how pulling that trigger would impact the other divisions. I think IBM has found a good way out.

Microsoft is in the software business and they see open-source as a threat. Their actions, therefore, tend to be a tad more hostile than the actions of IBM.

Well, you'll get no argument from me that sometimes Microsoft is its own worst enemy when it comes to IP and its image.

Be that as it may. You say that the software industry *might* have innovated even more without patenting. But how can anyone know that? All we can judge is the world that actually exists, and in that world, there is no evidence that patenting is stifling or slowing down innovation, as some people claim.

As for patent reform and improving quality, my point about similar problems existing in the 19th century was just to show that, like most areas of law and business and public policy, continuous reform and recalibration keeps the system more or less healthy and adaptive to new conditions. I'm sure that community patent reviews, post-grant opposition, and other reforms now being proposed (and supported by IBM and Microsoft, by the way), will impropve things, just as other reforms in 1836 and 1870 did.

What's really interesting to me is how people assume that there's some genetic or characterological difference in Americans that has made us the most innovative and productive nation on earth. It's true that there are not a lot of successful Italian or French software programs. America is clearly the leader in this essentially knowledge economy sector. But in my view, it isn't our character -- it's the unique history of our IP system.

Although most people don't realize it, the American patent system was as unique in the world -- and as revolutionary -- as the Constitution itself. Indeed, the leaders of our young nation rejected the Old World view that genius was the exclusive preserve of privileged elites. Their democratic instincts told them that genius could, if given the chance, be the province of everyman. And they knew that any nation that could unleash the ingenuity of millions of its citizens (as opposed to only a few) would inevitably lead the world in economic advancement.

So the Founders very deliberately set out to construct a patent system that would stimulate the inventive genius of the common man. Unlike the British patent system, which charged exorbitant application fees equal to 10 times the annual per capita income of its average citizen, U.S. patent fees were reduced to a level that even ordinary workers and farmers could afford. Administrative procedures were also simplified. And through a host of other means as well -- including allowing anyone applying for a patent by mail to do so postage free -- the patent system encouraged innovation on a mass scale.

The results were immediate and dramatic -- as Thomas Jefferson put it, the new American patent system "has given a spring to invention beyond my conception." Only 13 years after the first patent law was enacted by Congress, the United States had already surpassed Britain -- until then the acknowledged leader in the industrial revolution -- in the number of new inventions patented. By the 1860s, the number of new inventions patented in the U.S. was an astonishing seven times the number in Britain.

One very important reason for this dramatic surge in American innovation, of course, was the fact that by design the American patent system encouraged a much broader range of creative individuals to take part in inventive activity than was the case in Britain or other Old World countries. Whereas most British inventors were of privileged status, the vast majority of America's new inventors came from humble beginnings -- farmers, factory workers, carpenters and other artisans for the most part.

According to the economists Kenneth L. Sokoloff and B. Zorina Khan, who studied patent records and biographical data from the period, nearly 70 percent of the 160 "great inventors" of the 19th century had only a primary or secondary school education. Half had little or no formal schooling at all. And many of the most famous names in American invention -- men such as Matthias Baldwin (the locomotive), George Eastman (roll film), Elias Howe (the sewing machine), and Thomas Edison (the electric light and the phonograph) -- had to quit school at an early age to help support their families.

But the American patent system did not simply encourage the masses to participate in inventive activity. It made it economically feasible for them to do so. Indeed, by granting secure property rights to inventions for a limited time, the patent system enabled innovators to make an actual career of invention through the licensing of their discoveries. This in turn created the world's first national market for technology innovation, which was critical to powering America's emerging industrial economy to a position of world dominance by the end of the 19th century.

As Abraham Lincoln (himself a patentee) noted, the brilliance of the U.S. patent system was that it "added the fuel of interest to the fire of genius."

It wasn't until the end of the 19th century that most European nations amended their patent laws to match the innovation-enhancing character of the American system.

It was this hundred-year lead in democratizing our patent system, more than any supposed special ingenuity of the American people, that made the United States the "arsenal of democracy" in the 20th century and the economic powerhouse that we are today.

"By the 1860s, the number of new inventions patented in the U.S. was an astonishing seven times the number in Britain."

"just as other reforms in 1836 and 1870 did."

So how many of the patents issued in the 1860's before the reforms of 1870 were junk patents like many of those issued today?

software should not be patentable - it is like patenting art, math, or a cooking recipe.

they are just stupid and to me are all obvious.

the only people that I know of that have defended patented software are the patent lawyers.

glad the xerox park folks weren't greedy when apple stoled their interface and then microsoft stoled apples - guess if software was patented back then we wouldn't be very far today would we?

David B asks:

"So how many of the patents issued in the 1860's before the reforms of 1870 were junk patents like many of those issued today?"

The reforms in the Patent Act of 1836 eliminated many of the junk patents of that era. These reforms mandated, for the first time ever, that patents could only be issued for novel inventions that were deemed to be novel by trained PTO examiners. Prior to this act, there was no independent examination of patent applications, which led to great uncertainty as to validity and a coprrespondending rash of litigation. In fact, 2% of all issued patents were litigated during that era -- a rate that would give us 34,000 patent suits today if it still held true (instead of the 2,300 or so suits now before the courts.

The 1836 patent reforms created what we now know as the modern patent system, where independent examination and novelty are requirements. The reforms of 1870 further refined these reforms, instituting new requirements re: obviousness and greatly expanding the resources of PTO examiners to deal with the huge increase in patenting that occurred during the emergence of the railroad, electric power, munitions, and telegraph industries.

The reforms needed today are probably a mix of both revoltionary and incremental, to deal with the recent similar surge in patenting. Here I'm referring to proposed reforms such as community-based peer patent review, post-grant opposition, major increases in PTO resources (especially of trained examiners in new fields like software), etc.

As for James comment that "software should not be patentable - it is like patenting art, math, or a cooking recipe," it is worth remembering that during the time of Alexander Graham Bell and Thomas Edison, there were also those who claimed that electricity and telephony ought not to be patentable.

Their views were of course rejected, owing not least to the fact that what was actually being patented was not electricity or telephony, but novel and useful devices and results that employed these elements and processes.

So far the courts have taken a similar view of software -- i.e., that what can be patented is not algorithms or ones and zeros but products and creations employing mathematical computations that produce specific and useful results.

This is pretty much the same approach, by the way, that for over 100 years has governed the patenting of natural chemical processes that are combined and configured in novel, non-obvious ways to produce specific useful results.

Now, if you want to talk about non-novel, obvious software that ought not be allowed to be patented, then I'm with you all the way. Some studies suggest 30-50% of software patents are at least arguably invalid. That's something for serious patent reform to solve, just as such reform in the past solved (or at least significantly reduced) similar invalidity problems.

Reform is all the more important today because in many ways we are now re-creating the independent market for invention that existed in the 19th century and that powered all of industry.

Did you know, for example, that prior to the 1930s, there was really no in-house corporate R&D? All major companies, including the original American Bell Telephone Company, purchased the vast majority of the technology they used in their products on the open market, from independent inventors able to make a full-time career of inventing thanks to the existence of that independent market.

Today, with the emergence of Open Source and other forces and trends (and thanks to the Internet), that independent market is beginning to be recreated. In today's hetereogenous technology world, few if any of even the largest corporations hold all the pieces of their own product technology in their own hands. That's why patent licensing has grown from $15 billion to $110 billion annually of late.

This is to everyone's benefit, I believe.

One interesting side-note to the 19th century independent inventors market is how corporate R&D departments in those days viewed their job.

Consider the annual report of the American Bell Telephone Company's patent department to its board of directors for the year 1894. That year, the patent department evaluated 73 patents submitted from outside inventors and only 12 patents brought to its attention by employees.

This corporate focus on externally-sourced innovation was no accident, but rather the result of official corporate policy. As T.D. Lockwood, the chief of American Bell Telephone's patent department, wrote in 1885: "I am fully convinced that it has never, is not now, and never will pay commercially to keep an establishment of professional inventors in [our firm's] employ."

And Lockwood was right -- at least until the 1930s, that is, when the rising cost and complexity of technology combined with one rather earth-shaking legal change to finally create the conditions for concentrated in-house R&D within corporate America.

What was that legal shift? Simple. It wasn't until the 1930s that most large companies got their employees to assign the patent rights to their innovations to their employers.

Prior to that time, it was well established in case law that the mere fact of an employment relationship did not entitle a firm to an employee's inventions, even if those inventions were developed at company expense and on company time. As the Supreme Court decided in the 1893 case Allen G. Dalzell et al. v. Dueber Watch Case Manufacturing Company, "in the absence of an express agreement" by which the employee promised to assign all his patents to the firm, the company could not claim ownership of the patents.

So forget Henry Ford and all the other big business magnates of the early 20th century. It turns out -- incredibly! -- that it was actually the lawyers who were responsible for ushering in the modern era of large-scale in-house corporate R&D. By creating patent-assigning employment contracts that could pass court muster, corporate lawyers diminished (but did not entirely kill) the independent inventors' market.

But that independent market is clearly coming back -- and has been since even before Open Source emerged in the late 1990s. In 1972, for example, only 5% of patents went to small firms and first-time patentees. Within two decades, however, the number had skyrocketed to 23% and is still rising today.

Good news, in my opinion.

"So either patent suits are "efficient" and necessary, or they are not. Which is it?"

defense is costly, even if the suit eventually is not filed.