Law in the Internet Society
Ready for 2nd review

-- NikolaosVolanis - 22 Dec 2009

Intellectual property in Open Source Software and Open Standards: Two strategies against open innovation

By picking up specific examples of misappropriation of intellectual property rights in the context of open source software (OSS) and open standards, this paper attempts their classification under two strategies: a. "Fear, uncertainty and doubt"; and b. "Embrace, extend and extinguish". Both of these strategems have as a starting point a company’s established intellectual property rights portfolio, which is then used to either discourage customers or counter the potential threat of “openness” of an open standard or the potential success of an OSS project.

a. "Fear, Uncertainty and Doubt"

The threat of software patent litigation can create an atmosphere of legal uncertainty that is detrimental to the OSS community, particularly due to its "bazaar" method of production of OSS, which may lead to potential liability for all subsequent developers (as well as users) of a specific software.

But the potential liability isn't real: no one, and I mean no one, brings patent litigation against people who cannot pay royalties. No Microsoft-or-clone can credibly threaten to sue individual free software developers, and indeed any such party will falsely trumpet its good faith by guaranteeing not to sue such developers.

Suing or threatening to sue people is definitely bad publicity and has borne minimum results, as this was aptly demonstrated in the long fight of the music industry against individual P2P? users. Still, the music industry kept fighting and fighting over a battle that was lost even before it began, one of the reasons being that there was no real possibility for recovering lost profits from the defendants. I think that RIAA was aware of that, and what they did had as its sole purpose to scare P2P? users, since the exponential increase of unauthorized copyrighted material on the net was killing the core of their business at an extremely rapid pace. Their response, I guess, was fueled by panic, not by reason. Perhaps the rapid increase in market share for OSS in the future may, in turn, lead Microsoft to panic and start suing big developers of OSS.

In the aforementioned analogy, however, there is one major difference between music file sharing and developing OSS. In the first case, a lawsuit has no other immediate impact but to the person being sued. In the case of OSS, a potential threat to an OSS developer means subsequent trouble for all the end users of the program developed by that developer. The effects of such lawsuits therefore are systemic, and even unpredictable, as they usually extend beyond the software industry, into the industries in which the end-users of OSS are active. So the question for Microsoft becomes a matter of figuring out how many enemies would such a lawsuit muster against Microsoft, and whether it is in a position to fight such a battle. Considering that OSS is currently being used by major players in sectors ranging from financial services through manufacturing and distribution to the public sector, it seems highly unlikely that Microsoft would like to fight in so many fronts. After all, many of these companies that use OSS may be Microsoft's shareholders or, at least, clients of Microsoft's shareholders.

In this context, the SCO v. IBM case is a lucid example of targeting OSS users and developers, even though it is not primarily a patent infringement case. On March 6, 2003, the SCO group filed a $1 billion lawsuit against IBM claiming that IBM has, without authorization, contributed SCO’s IP to the open source codebase, Unix-like Linux operating system. In May 2003, SCO sent a letter to members of large US firms warning them of the possibility of liability if they use Linux. Another series of letters was sent in December 2003 alleging copyright infringement related to 65 files in the Linux code tree. Following a series of lawsuits and counter-lawsuits, the cases are still pending before various courts. IBM has explicitly noted (Section E, paras 22-24) that SCO has engaged in a technique, so as to create a false perception that SCO holds the IPR rights to UNIX which permit it to control not only all UNIX technology, but also Linux. This impression has been further reinforced by various statements of SCO’s chief executive officer, regarding potential liability for Linux users.

This technique of creating uncertainty as to the legality of specific software is usually termed "FUD" – acronym for “Fear Uncertainty and Doubt”. It was first used in the computer hardware industry in 1975, as an attempt to describe IBM’s marketing and public relations policy, as a strategy aiming to influence the public by disseminating unfavorable opinions about a competitor’s product, to overstate the estimation of switching costs if current customers decide to migrate to a rival company’s products or to maintain a leverage over a current business partner who could potentially become a rival. Additionally, when said strategy is combined with the threat of enforcing intellectual property rights (IPRs) that belong to the company, then the specific IPRs are not only used to expose individual companies to legal threats, but also to create a climate of legal ambiguity which can be detrimental not only to the OSS community but the industry as a whole.

But it's not a way to stop the programs, only a way to make money for trolls. There are currently a couple of trolls trying to take bites at big commercial Linux users. At least one, like many trolls I know, is a patent lawyer who helped to get a pretty sketchy but not entirely bogus patent through, and has now invested in trolling it. Maybe the people who he's hitting up will settle with him, or maybe they will join with parties upstream to destroy his patent if they can, or maybe both. But whatever happens it's not going to stop people from using the kernel. There are many reasons Microsoft doesn't go to war on its patents, of which not the least important is that they couldn't stop anything with them anymore.

As I read your comment I understand that, indeed, the troll threat that OSS is facing does not jeopardize OSS adoption by a wider user base; it is mainly a problem of increase of overall costs of OSS use by software users, as trolls take big or small bites at commercial OSS users, who, either because they are bullied through FUD or because they cannot afford a arduous and costly legal battle, decide to settle "amicably" with a troll. Logically however, since this technique did not work in the past, when the market share of OSS was smaller, it will not work a fortiori today, where the OSS user community has grown and established itself in all sectors of the economy. In this sense, people will not stop using OSS because OSS is constantly becoming better as it increases its share ("enough eyeballs" argument) and therefore it is becoming more important for them. It is a sort of a virtuous circle.

Since the 1990s, the term has been used to characterize a facet of Microsoft’s response to the open source movement, which has been implicitly acknowledged in the company’s internal “Halloween Documents”. The Halloween documents comprise a series of confidential Microsoft memoranda (drafted by Microsoft employees Vinod Valloppillil and Josh Cohen) on potential strategies relating to free software, open-source software, and to Linux in particular; and a series of responses to these memoranda. Both the leaked documents and the responses were published by Eric S. Raymond, an open source advocate and co-founder of the Open Source Initiative. Marked "Microsoft confidential", these documents identify open-source software, and in particular the Linux operating system, as a threat to Microsoft's dominance of the software industry, and suggest ways in which Microsoft could disrupt the progress of OSS. According to the Halloween Documents, OSS is identified as a “long-term credible” product and thus “FUD tactics cannot be used to combat it”. Still, over the following years, Microsoft has made various announcements regarding the potential dangers of developing or using OSS software, particularly with regard to the General Public License’s (GPL) “viral nature” which “_[...] poses a threat to the intellectual property of any organisation that derives its product from GPL source_”, and to the potential liability for users of Linux, since the latter software allegedly infringes 235 of Microsoft’s patents. This acknowledgment has been recently followed by relevant legal action taken by Microsoft against a company using Linux as the software platform for their products. The latter case was eventually settled out of court.

TomTom's decision to settle with Microsoft may serve to show that Microsoft patent enforcement agenda may not be obsolete yet. But in light of the foregoing findings regarding the importance of some OSS commercial users, this agenda is definitely undergoing a thorough review. In one strange twist, strategic litigation may have acquired a new meaning and goal: in the case of the music industry, strategic litigation aimed to cause the widest possible impact to the community of P2P? users. In the context of patent infringement, Microsoft's strategic litigation is probably aiming solely at maximizing profits from their IP portfolio, without stirring the waters of the ecosystem that includes developers, users, clients and investors who are directly or indirectly in favor of OSS. TomTom? seemed like a good target for Microsoft, being relatively deep-pocketed but also not a key or significant player in the market.

b. "Embrace, Extend and Extinguish"

A different strategy has been followed in the course of open standards (HTML 4.0 standards in particular), in the seminal antitrust case between Microsoft and the U.S. Department of Justice. As referenced in the proposed findings of the Department of Justice , “[...]_ Microsoft’s response to the browser threat was to “embrace, extend, extinguish”; in other words, Microsoft planned to ‘embrace’ existing Internet standards, ‘extend’ them in incompatible ways, and thereby ‘extinguish’ competitors_.” (Section V.A.3.b. para. 91.3.2). In this context, the first step involves the development of software which is substantially compatible with competing products and which implements the public standard, the second step refers to the adding or promoting features which are not supported by competing products or part of the standards, and thus creating interoperability problems for customers who attempt to use the standard without said additions or features, whereas the third and final step involves the marginalization of the competitors, by the time these additions or features become a de facto standard because of the company’s dominant position in the market.

A similar example is also the contest between Sun Microsystems and Microsoft, whereby, Sun accused Microsoft of attempting to use the same technique to “extinguish” its Java cross-platform language. Sun allowed users to freely download the tools needed to read and write programmes in the Java language, which has become popular because of its cross-platform interoperability. In this context, Sun licensed Java technology to Microsoft for its inclusion in the Windows platform. However, Microsoft applied a stratagem of “embrace and extend”, by implementing additional features in Java which were not part of Sun’s standard. As a result, if developers wrote software that took advantage of the extra features, this software would only run in Java running on the Windows platform, thus “extinguishing” Java’s cross-platform compatibility that Sun was aiming for. For this reason, Sun sued Microsoft based on patent infringement and antitrust violations. In 2003, the 4th U.S. Circuit Court of Appeals reversed the decision of a lower court, by deciding that Microsoft did not have to distribute a version of Java endorsed by Sun, but on the same time affirmed that Microsoft violated Sun’s copyright by distributing its own version of the language for the Windows platform. In this context, Microsoft was not burdened with a “must-carry” obligation, but simultaneously was prohibited from distributing any version of Java other than that licensed to Microsoft in a 2001 Agreement with Sun. Finally, the companies reached a settlement over Sun’s antitrust claims in 2004.

Here I think you are not giving sufficient credit to Microsoft's attitude, which was that control of APIs meant control of developers which meant ultimate control of the software environment. They needed to keep developers coding to their APIs. We changed that more fundamentally, by providing other ways for programmer brains to be used.

Controlling software's "means of production", i.e. programmers, can indeed be achieved through a number of ways, one of them being setting the rules (and consequently the limits) of what can be programmed under a given set of restrictions. Microsoft's attempt may be considered, I reckon, as a different sort of "walled garden", and as a predecessor to Apple's current attempt to indirectly control software production through its app-store. Of course, in an App-store environment, the extermination of third party unwanted code is implemented from day one, embedded in the systems architecture: There is no need to figure a strategy to embrace and extent.

Final Thoughts

I suspect that the mere presentation of these two strategies helps little to address the underlying questions regarding the use of intellectual property rights in the context of OSS and open standards; to conclude that the aforementioned strategies are based on the misuse of IPRs - and that proper use of IPRs would resolve this tension - would be a rather short-sighted remark, since the effect of IPRs on these areas is systemic. Moreover, this remark overlooks the fact that the judgment on whether IPRs (especially patents) are used or misused depends essentially on the standards of granting a patent by the administration and upholding it by the courts (in this sense, if an abstract software patent (see p.194 etseq.) is granted based on current legislation and upheld by the patent court, its enforcement by its holder would not constitute a legally prohibited abuse stricto sensu, despite its detrimental effect to the software community. In any case, used or misused, IPRs have an impact on innovation fostered by the open source community as well as standard setting organizations that promote open standards. The strategies that we may devise to overcome the potential threat of IPR enforcement (e.g. collecting and keeping prior art information in order to invalidate patents, establishing mechanisms that facilitate better examination and quality of patents, patent pooling, open source insurance, code scanning / code "management") may help mitigate the risks deriving from IPRs, but do not tackle the problem at its source.


This was an interesting read. I have no substantive comments, I'm mainly leaving this comment because I changed the second section's heading from "Embrace, Extent and Extinguish" to "Embrace, Extend and Extinguish" and I thought that, even in a wiki environment, it would be inappropriate for me to adjust even something so minor without telling you (especially if I was wrong about the change). Again, a good read and I appreciated the linking for additional information.

-- BrianS - 17 Jan 2010

Thank you Brian! indeed, mea culpa

-- NikolaosVolanis - 17 Jan 2010

I think this is a rather surprising combined discussion of a number of matters I usually think about separately. You can see your disparate illustrations as joined by the theme of use of concentrated intellectual property rights to compete at an advantage against free software production. But they aren't very successful for those purposes anymore, so we have evidently created strategies of our own that are less pithily described. And the proxy war that is SCO, the MS Win95/IE disaster investigated by the Clinton Justice Department, and the Java wars (which will now be conducted between IBM and Oracle rather than between Sun and Microsoft) seem to me more different than they are similar. We have, at any rate, moved pretty far beyond that frame for the industry, it seems to me. It's time, I think, to look at the evolving issues, which involve a step back to an older unfree software industry through the App Stores, and a step forward in network operator misbehavior, as the operators try to destroy the Internet itself.

thank your for the comments, Professor. I'll attempt to integrate them, taking into account our recent conversation.

-- NikolaosVolanis - 31 Jan 2010



Webs Webs

r12 - 07 Sep 2011 - 00:53:28 - IanSullivan
This site is powered by the TWiki collaboration platform.
All material on this collaboration platform is the property of the contributing authors.
All material marked as authored by Eben Moglen is available under the license terms CC-BY-SA version 4.
Syndicate this site RSSATOM