Trolls Attack Innovation: Panic in Corporate Parks!

TrollPatent trolls are becoming a nuisance for technology companies. These trolls are companies/entities that develop or acquire patents for the sole purpose of suing companies that unknowingly infringe on those patents. The infringement lawsuit brought by NTP against Research in Motion (RIM) received a lot of attention in ‘06 when the ruling court threatened to shut down RIM’s Blackberry service. RIM eventually settled by paying NTP over $600 million. The motivation for such patent trolling is clear.

In the June 2008 issue of the Harvard Business Review (HBR), management professors Henkel and Reitzig have an article that offers their suggestions for how businesses can protect themselves against these trolls (which they call “sharks”). In my opinion, their recommendations are unrealistic in that they idealize the willingness of competing companies to cooperate with each other and to forgo competitiveness in technology development in order to protect themselves against the trolls.

The first half of their article is informative and makes very good points about the general situation that exists today. The recommendations they make, however, are not so good. I’ll address each of their five recommendations here and suggest what I believe are more realistic solutions below.

1. High-technology firms should move away from building huge patent portfolios for the purpose of cross-licensing with competitors
The authors correctly point out that one reason companies generate patents is to trade them for patents needed from other companies. Such trading is done because the patents behind key components in complex technical products are usually distributed across all of the major companies in an industry—no one company owns all of the patents necessary to produce a product. Because no company can produce a product solely on technology from their own patent portfolio, companies trade their patent licenses for licenses to the other companies’ patents. The cellphone industry is a typical example of this, e.g., Nokia trading patent licenses with Motorola and Samsung. The authors suggest that this strategy of building a strong portfolio and trading licenses is no longer valuable to companies and should be stopped because it doesn’t protect them from trolls. Indeed, it doesn’t—it protects them from their competitors, who are ultimately more threatening than patent trolls. Just because a patent strategy does not affect patent trolls does not mean that it isn’t worth doing. If I were Nokia, however, I would certainly try to convince Motorola to follow this advice to stop building up Motorola’s patent portfolio.

2. Companies must simplify technical standards and create more-modular designs
These two points are reasoned as follows. The authors assume that simpler technical standards would reduce the number of patents that teach the technology in a patent, thereby producing less opportunity for patent trolls to own the patents on technology required by the standards. The authors assume that modular designs for products would allow a technology component to be swapped out of a product if a patent troll claims that the component infringes on their patent; in other words, redesigning the product to respond to the troll’s threat would be easier with the modular design. Well, easier said than done. The nature of standards committees is such that they are not going to adjust their ways to account for a trend in patent litigation abuse that may or may not impact their standard—to make such adjustments would be considered by committee members as compromising the integrity of the standard. As for the modularity recommendation, a company has many other motivations to modularize their designs; if companies don’t modularize then it is because they can’t or have reasons not to (manufacturability, cost). The assumption behind the authors’ modular strategy is that patented technologies can be easily worked around by swapping different components, while in fact patents can be quite broad and difficult to design around. Whether RIM could have designed around the NTP patent is unclear to me, but there are many patents that are so fundamental that they embody the general feature of a product (e.g., an lcd screen on a remote control, ringtones on a cellphone) and to avoid the patent is to eliminate the feature. Finally, the bottom line is that companies will design products that best serve their customer in a way that best meets the needs of the company (e.g., profitability). The financial threat from patent trolls to a company is not (yet) so great that a company with risk revenue and profitability by changing their whole R&D approach to product design, particularly given that it’s unclear whether the proposed solution will, in fact, eliminate threats from trolls.

3. Companies must begin cooperating with their competitors early in the R&D process
The authors argue that the process of a company developing technology in secret, without the knowledge of their competitors, is “outdated”. The authors acknowledge that companies would be “uncomfortable” to share knowledge with their competitors early in their technology development process, but they argue that by doing so patent trolls can be abated. While the authors state that lawsuits from trolls “could change if more firms started to disclose early-stage information,” they do not explain why this the early-stage sharing of secret information would help against patent trolls. I can’t explain why either. Again, the authors seem to be suggesting that companies would be willing to forgo their competitiveness in order to avoid patent trolls, joining with their competitors in a spirit of innovation cooperation and protection. Again, if I were Nokia, I would strongly recommend to Motorola to share with me their early-stage technology development to guard themselves against patent trolls. I would go even further by offering to sacrifice Nokia to patent trolls by not sharing our technology secrets with Motorola and thereby drawing the trolls towards Nokia and away from Motorola.

4.Firms must foster interdepartmental and intercompany cooperation
Here, the authors discuss the value of getting patent lawyers involved early in the patent process. No argument from me here, although I’m not sure what they has to do with interdepartmental and intercompany cooperation. They then suggest that companies should cooperate with each other, and they give an example where a patent troll bought patents from company A and then sued company B for infringement. The authors suggest that company B would have been better off if company A had talked to company B first and offered a license to them rather than selling the patent to the troll. Indeed! In other words, please don’t sell to trolls. What the motivation is for company A to do so, however, I don’t know. This seems to me to be an unrealistically idealistic hope for how companies could work together.

5. Companies must stop flooding the patent office with insignificant applications
Now, here I agree with the authors. The patent office in the US is overwhelmed, making them more prone to issue patents that are not novel and thereby not valid. Inventions that lack novelty but are erroneously issued can become legal weapons against companies that have used the technology in a product. Also, as the authors point out, a flood of frivolous patents in a field makes monitoring patent activity difficult. The suggestion that companies should unilaterally decide to reduce patent activity for the good of the whole (and their competitors) is, however, unrealistic. Again, as Nokia I would strongly urge Motorola to reduce their patent activity as to not overburden the USPTO and my patent managers who monitor Motorola’s activities.

As can be seen, the solutions that the authors suggest simply aren’t realizable in a competitive world such as today.

I believe that the solution to this problem to is to de-fang the trolls by

  1. making their threats less likely to succeed in court through the combination of new federal laws and Supreme Court guidance,
  2. improving the patent office process in a way that reduces the number of illegitimate patents that patent offices issue.

To the first point, the US Supreme Court has already ruled in such a way that the bar is raised for obviousness. This was done in the KSR case, which I have previously blogged about with respect to the Court’s interpretation of innovation. The authors of the HBR article do mention the KSR case and others, and they clearly support this idea. What remains to be seen on the impact of KSR is if its precedence will have an impact on future court decisions. If it does, the validity of frivolous patents wielded by trolls will be easier to attack.

To my second point about reducing the number of illegitimate patents, a test program (the Community Patent Review Project) is in place in the U.S. to assist the USPTO in finding and analyzing prior art so they can better assess whether an invention is novel or not. In this program, companies are able to see patent applications earlier than normal and to offer prior art and commentary for patent applications in their field of expertise. There are potential problems with this approach that I won’t go into, but trying new ideas to help the USPTO is critical for mitigating the effectiveness trolls.

Finally, there is a patent reform bill in Congress that has been trying to gain momentum for quite a while. One effect of this bill, if passed, would be to restrict the ability to sue over patent infringement. Obviously, this would de-tooth the trolls quite a bit.

Supreme Rulings on Innovation

The Supreme Court ruled this week on a case that impacts past and future patents while also commenting on the nature of innovation. I’ll get to the innovation part in a second.

Briefly, the Court threw out the previously used test for whether an invention was obvious or not. Most new ideas do not pop out of the air but are combinations of two or more disparate concepts that, when put together, create something new. One of the most critical questions before patent examiners and patent trial judges has been whether this new creation is nonobvious. A patent attorney that I used to work with explained the concept of nonobviousness by holding up a pen and saying that his pen has never been made in the color chartreuse, so such a pen would be new/novel but would also be an obvious modification of currently available pens—there would be nothing nonobvious about it. A chartreuse pen would not elicit a, “Wow, how did you think of that?” response.

In a nutshell, this week’s Supreme Court ruling makes it easier to denote an invention as obvious. One interesting aspect of the ruling for this blog is that the language of the Court ruling speaks to the definition of innovation. Several blogs on innovation, including this one, have attempted to define what innovation is and is not (see the left column of Broken Bulbs, for example). Surprisingly, there is disagreement on the matter, with some people claiming that an invention must be successful in the marketplace in order for it to be innovative. I’ve argued against that definition.

One can look at quotes from the Court’s ruling to infer the judges’ opinion on what defines innovation.

Justice Kennedy wrote,

Granting patent protection to advances that would occur in the ordinary course without real innovation retards progress.

Here he says that innovation is more than just the “ordinary course” of development of something new. Now, how one differentiates an “ordinary course” from an “extraordinary course” is another matter (presumably, one has to rely on the expert opinion of someone knowledgeable within that field of technology).

Interestingly, the court also wrote that ordinary innovation is not entitled to patent protection because it “does no more than yield predictable results.” So, what exactly is ordinary innovation? My best guess is that it is synonymous with the more commonly used term incremental innovation which is differentiated from radical innovation. The former is the common process of improving a product or process through regular R&D development and results in small and predictable improvements. The latter requires creativity and produces unanticipated inventions, resulting in often dramatic changes to product or process performance and often gives the inventing company a significant competitive advantage.

Other sources offered their opinions on innovation in this case. According to the New York Times,

Pharmaceutical and biotechnology industry groups…argued that innovation would suffer if patents became too hard to defend.

This argument was against the Supreme Court’s final decision, which will make it easier to file invalidity claims against issued patents. The biotech industry was worried that if straightforward tests for determining obviousness were eliminated, then innovation programs would be difficult for companies to justify because of the potential for increased uncertainty in issuance and litigation risk. This is counter to many recent arguments that the current state of patents has stifled innovation by creating such a minefield of potential litigation that companies have difficulty doing anything without infringing on weak patents and potentially facing frivolous litigation.

In contradistinction to the biotech group, Microsoft and Cisco filed a brief that was in support of how the Court finally ruled, stating bluntly, “The Federal Circuit’s current test for obviousness hurts innovation” and “has had a stifling effect on true innovation.” Their argument was simple:

Defensive, large scale patenting drains resources away from real innovation: scientists and engineers must spend time working with lawyers and patent agents to file patent applications where their time would be better spent on product development and research.

In their decision on this case that included discussion of obvious innovation, as discussed above, the Supreme Court seems to be aligning the legal assessment of patentability with the concepts of incremental vs radical innovation that have been developed over the past decade. There’s no doubt that the Court accepted the case due to their frustration with the current state of patents (given the unanimous decision, any guesses if the judges are all Blackberry users?), but it’s interesting to see their decision’s collateral contribution to the innovation discussion. I would say that this Supreme Court decision solidifies Innovation as the key business concept of this decade.

I will leave the final words of this post to the following three quotes on innovation from Justice Kennedy in the opinion of the court:

…When there is a design need or market pressure to solve a problem and there are a finite number of identified, predictable solutions,
a person of ordinary skill in the art has good reason to pursue the known options within his or her technical grasp. If this leads to the anticipated success, it is likely the product not of innovation but of ordinary skill and common sense.

This is so because inventions in most, if not all, instances rely upon building blocks long since uncovered, and claimed discoveries almost of necessity will be combinations of what, in some sense, is already known.

We build and create by bringing to the tangible and palpable reality around us new works based on instinct, simple logic, ordinary inferences, extraordinary ideas, and sometimes even genius. These advances, once part of our shared knowledge, define a new threshold from which innovation starts once more. And as progress beginning from higher levels of achievement is expected in the normal course, the results of ordinary innovation are not the subject of exclusive rights under the patent laws.


Defining Innovation

Geoffrey Moore’s blogging dialogue with Tom Forenski highlighted to me my own vagueness over what, exactly, is meant by innovation and innovative. Tom Forenski said, "I think innovation *always* has to have the quality of disruption" in response to 10 myths about innovation posted by Moore on the Sandhill website.

Moore (correctly, in my opinion) argues that there are varying levels of innovation, many of which are not disruptive. Moore’s arguement is made clear by the very fact that the term "disruptive innovation" was the focus of Clayton Christensens book The Innovator’s Dilemma: if all innovation were disruptive, then disruptive innovation would be redundant. I doubt that Christensen would have wasted his time detailing and defining a redundant concept, anymore than I expect to see a leading chef publish a book detailing how we should focus our culinary resources towards “edible food.”

Still, where do you draw the line on what is and isn’t innovative? I believe we can draw some guidance from the decades of thought that has gone into determining what is patentable. One of the criteria for whether an invention is patentable is whether it is novel or nonobvious. By this is meant whether someone knowledgeable in the field of the invention would consider this to be an obvious creation or not. A patent attorney at one of my previous companies liked to demonstrat this concept by pulling a pen out of his pocket and saying, “This type of pen may have never been produced in the color red, but that does not mean that a red version of this pen is novel or nonobvious and can therefore be patented. To anyone who knows the field of pens, the creation of a red version of what I am holding is an obvious creation.”

This criterion, perhaps, could be part of the definition of innovation. To use the example above, a red pen might be a new product, but it would not be innovative, unless there was something unique about that color which causes it to succeed in the marketplace far beyond the expectations of one who knows pens. Then, the development of a red pen is only innovative in the selling and marketing of it.

As detailed by Moore, innovation can take many forms, and can include incremental improvements that cause no disruption whatsoever. But, the development of a new product, or the sales of a new product, should not be described as innovative unless there is something special about the product or the result in the marketplace that is not obvious to someone expert in the field of that product.

The February issue of the Harvard Business Review has an article on Management Innovation, and in that article the author unintentionally provides his own description of what makes something innovative. I have slightly modified one of his statements on management innovation to speak to innovation in general:

innovation can be defined as a marked departure from traditional principles, processes, and practices or a departure from customary forms that that significantly alters functionality.

The author later states that (management) innovation stems from “unorthodox thinking” and “wisdom from the fringe”. This surely separates innovations from simply good ideas. Introducing a de-featured, lower cost version of a product may be a very good idea, but I would not classify this as an innovation.

Gordom Graham makes valid points on distinctions between different types of innovation over at Broken Bulbs. I invite readers to comment on, and bloggers to post and trackback their own thoughts on, what is meant by innovative and innovation.

The Value of Patents

Patents have come under attack recently. A recent report by Booz Allen Hamilton (discussed in a previous post) indicates that patents have no value to corporations anymore and have produced comments that perhaps patents are no longer valuable. On top of this, the RIM-NTP debacle, also discussed here and here, has shone a light for the general public onto the dark and brutal patent underworld in which companies compete in an arena where customers are nowhere to be seen.

I’ve told many people over the years that the most surprising aspect of my move from academia to industry was my discovery of the enormous effort that companies expend towards IP, and how lucrative or damaging patents can be. A well-written patent can protect the value of a company’s technology or it can force other companies to "pay to play" if they want to sell similar technology.

As an aside, you are probably wondering why any company would license a patent to their competitor and thereby enable them to compete. The reason is a pragmatic one: the claims of most patents are not air-tight and if a company tries hard enought it can figure out a workaround, although they still risk litigation. Often, the easiest solution is for both companies to come to an agreement where the patent owner licenses the patent to its competitor for an amount both sides can live with. A common strategy for large companies is to trade patent portfolios with their competitors: Nokia with Samsung, Philips with Sony. These large companies usually all have patents that could tie each other up in courts for years and grind their whole industry into a litigious standstill. Better to trade and allow each other to compete in the marketplace rather than decend into a mutually-assured destruction scenario. Guess what happens, though, to a company that doesn’t have any patents to trade. And if that poor company is a publicly traded one, their stockholders will not be amused.

Therin lies one of my many difficulties with statements that patents have no valuable impact on a company. BAH’s report looked at the relationship between number of patents filed and a company’s financial success and found no correlation. This has been interpreted by many as meaning that patents have no value to a company and that they should stop wasting their time with any patent strategy. I bet, however, that you could walk into any popular lunch spot in Palo Alto and find several people whose companies have either significantly benefited or been hurt financially because of patent issues.

So why did BAH find little or no correlation? I believe that the proper interpretation of their data is simply that focusing on the number of patents filed is the wrong approach to success. Throwing money at developing a patent portfolio does not in itself add value to a company. Patents must be developed strategically and thoughtfully. A little luck doesn’t hurt either.

In my experience, the least valuable patents often emanate from the most prolific patent filer: for them, the slightest notion of novelty (to them) is enough to generate 43 claims. But these patents have received little deep thought and produce little value. Just because an author has published a huge number of books does not mean that they are a great author, and there is probably little correlation between the number of books an author has published and the winning of a Pulitzer/Nobel/Booker Prize. But, this lack of correlation does not mean that there is no merit to writing books.

In general, innovators like to invent, to see their ideas developed. They often have to be forced at gunpoint to work on writing a patent application–there’s very little fun to be had in that activity. Additionally, the most brilliant innovators often fail to see why their ideas could be patented at all–the inventions seem too obvious to them to be patentable.

So, back to my the point that I originally wanted to make. I’ve already pointed out that patents are valuable for trading. Patents are also highly valuable to startup companies. I was at a startup where we filed close to two dozen patents in our first two years. Having a phat portfolio bought us a lot of milage with the Sand Hill Roadsters and allowed them to cross off a gating item for funding from their VC Checklist. And if anyone is wondering if patents can be valuable, feel free to send your question to the CEO of RIM, whose company was recently thought to be in danger of going under and Blackberry service shutting down thanks to patent litigation.

Yes, the patent office is vastly understaffed right now (with 400,000 applications filed last year), and the vast majority of patents issued have no value whatsoever (thus the BAH result: more patents does not equate with greater value). But meaningful patents protect a company’s right to sell technology, prevents others from selling that technology, can be sold and traded, and provide a level of comfort to external investors and analysts. Ignoring wise patent development is done at a company’s peril.