International In-house Counsel Journal Vol. 11, No. 41, Autumn 2017, 1

Open Source, IP and Patent Pledges:

AMANDA BROCK Visiting Research Fellow, Centre for Commercial Law Studies, Queen Mary University, UK

1.0 Introduction to Open Source This article considers the use of patent promises or pledges, as a device to counter what are often seen in the Open Source world as undesirable consequences of the operation of the patent system. They have become particularly topical in 2017, as the world’s first billion dollar Open Source company, , extended it’s patent pledge further in September 2017. 1.1 What is Open Source? In its simplest definition, “Open Source” is a term used for software, whereby the human readable source code is shared with users of the software by way of a licence. A licence enables free use, modification and distribution of that software. The human readable code is written in a computer coding language that can, like any other human language, can be read by someone who speaks that language. In traditional, proprietary software distribution, that human readable code is a closely guarded secret and not shared, as it contains the “secret sauce” of the software. Source code is compiled into machine- readable code and this then instructs a computer or device of the action to take. Without the source code it is not possible without extreme effort to maintain or modify a compiled software program hence its value and “secret sauce” nature. In the open source model, the source code is openly shared and freely available under licence allowing others to use, study, modify and add to the original open source code, and distribute the code so long as the terms of the licence on which the software is distributed are complied with (which may, depending on the licence, require redistribution under the same licence). 1.2 The History – free and Open Source software Open Source software is also referred to as “”. Richard Stallman is the Godfather of Free Software. “Every decision a person makes stems from the person's values and goals. People can have many different goals and values; fame, profit, love, survival, fun, and freedom, are just some of the goals that a good person might have. When the goal is a matter of principle, we call that idealism. My work on free software is motivated by an idealistic goal: spreading freedom and co-operation. I want to encourage free software to spread, replacing proprietary software that forbids cooperation, and thus make our society better.” “Free software” means software that respects users' freedom and community. Roughly, the users have the freedom to run, copy, study, change and improve the software and distribute it as changed, or in the original form. With these freedoms, the users (both individually and collectively) control the program and what it does for them.” The GNU web site goes on to say ““free software” is a matter of liberty, not price. To understand the concept, you should think of “free” as in “free speech,” not as in “free beer”. The essence of “free” software is enshrined in the concept of copyleft.

International In-house Counsel Journal ISSN 1754-0607 print/ISSN 1754-0607 online

2 Amanda Brock

Stallman refers to free software and not to Open Source software. He is the “father” of the ideological free software movement and the creator of copyleft. Not all code in the world of free and open source software is copyleft [add copyleft logo here]. The term is not enshrined in law, but a pun from this man, with a symbol that is the reverse of the copyright symbol ©. To make this concept a reality Stallman and his advisers developed a legal tool in the licence of the free software which is a licence provision that ensures that any user who takes and modifies the code they received under a copyleft licence will, if they distribute the modified code, be required do so under the same free licence without any additional restrictions or conditions and will release the source code in the modified code they distribute, under those original terms. The GNU General Public License is probably the best known and most used example of this type of licensing1. Each party will own the copyright in what they have created but must license any new or modified version of the code under the same terms as the code they received, ie the licence being copyleft, the rights are left in the free code and the quid pro quo for usage of the free software is that new developments if distributed, must be contributed back. The licence terms are non-discriminatory and allow anyone to modify and extend the code subject to this restriction included by the guardians of free software for the user’s own protection. It is intended to preserve the freedom under which the software was initially distributed through the use of the concept of copyleft. As the creator of the original work owns the copyright in his work, despite the work being licensed under a copyleft licence for use by others, should any user not contribute back under the same licence, then that user would have stepped beyond their rights as granted under the licence and potentially be in breach of the original copyright holder’s copyright2. As Stallman states “All code added to a GPL3 program must be free software, even if it is put in a separate file. I make my code available for use in free software, and not for use in proprietary software, in order to encourage other people who write software to make it free as well. I figure that since proprietary software developers use copyright to stop us from sharing, we co-operators can use copyright to give other co-operators an advantage of their own: they can use our code.” As a concept, copyleft takes copyright (where the creator of a work is the owner of the copyright in code they develop and third parties may use it only with the owner’s consent) and turns it around practically as well as symbolically. This is achieved by requiring that both the work licensed for free use and any and all changes made to that work by a third party must be distributed subject to the same ongoing freedoms and that this freedom cannot be restricted without the owner of the original copyright work’s consent.

1 GPLv2 section 6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. 2 GPLv2 section 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 3 GPL – the GNU General Public Licence, is an OSI approved standard open source licence published by the Free Software Foundation (“FSF”) on the terms of which free software may be distributed by anyone who meets the free software definition and complies with the licence terms.

Open Source 3

Copyleft code when combined in certain ways with other code may create a new collective piece of code that may be a derivative work. Understanding the derivative work concept is something best compared to cracking eggs to make an omelette. Once the eggs are combined they become one new thing not individual eggs and cannot be put back into their constituent parts easily. That doesn’t mean to say that there are not alternative ways of putting together a different meal with the eggs without mixing them and in which the eggs stay as individual eggs and identifiable as such (like gammon with a lovely fried egg on top). But, once they are mixed together to make an omelette, they are combined to create a new and distinct thing. In software terms the combination of software to create a combined new work is a derivative work, which like the omelette is not easily reversed. Where this happens in software, the new derivative work, like the original copyleft licensed component, will also be subject to copyleft terms. This impact of copyleft has been referred to, (much to the annoyance of the software development community) as the “viral effect”, where the free copyleft software, if combined with other code to create a derivative work, “infects” that non copyleft code with its freedom and makes both the original free software and the additional software “free”. However, it should always be borne in mind that this is not the normal outcome or consequence of a combination of software. This has created considerable and often unnecessary fear in the minds of owners of proprietary and is perhaps the biggest single factor that dissuades some business usage of free software. In short the owners of proprietary code are concerned to use free software due to this fear that the copyleft code will virally affect their rights in their own proprietary code. Clearly this consequence fits the ideology Stallman espouses. His copyleft ideology creates a cascade of free software and freedom. As a paternalistic concept, it protects users from themselves in the interest of freedom. The consequences of copyleft are not accidental, but a carefully crafted attempt to preserve this freedom and remove temptation. This ideology is akin to political thinking as opposed to software coding. The decision to create this was not a technological or development decision, but the expression of a belief system. That belief system is the center of free software and the free software movement but for many it was perceived as prohibitive to business and something that held up commercial usage of Free Software. The term “Open Source” was adopted in 1998, by a group on the US’s West Coast. The adopters formed a movement that rejected some of the free software philosophy. Eric Raymond and Bruce Perens founded the Open Source Initiative later that year.4 Its foundation was, to some extent at least, also a reaction against the values of free software. Like free software, Open Source software involves the distribution of the source code in the software. The human readable, “secret sauce”, that allows people and not machines to understand and read software, is freely available along with the binary or computer readable code in any software distributed in this way. To work together on code is the collaboration at the core of open source. In the same way as the FSF have a set of criteria to be adhered to in the definition of free software, the Open Source Initiative (“OSI”) created a definition that sets out 10 criteria to be met to be able to call software “open source”. Like free software there is no legal constraint on anyone calling their code either free or open. However, if you wish to comply with the true sense of either meaning then a distributor should meet the criteria set out by either

4 http://opensource.org/history

4 Amanda Brock the FSF free software definition or the OSI’s Open Source Definition (“OSD”) and there is some action to require compliance with this. There are differences in the two groups’ criteria and definitions. These differences are based on the ideological differences between the two groups. By 1999 the OSI had approved its first list of licences as meeting its OSD. Today the OSI has approved a large number of licences. Its licence review process and the OSD are often considered to provide an industry standard for open source licences with the OSI taking on the mantle of a standards body. In its work over the years, the OSI has tried to keep this simple and avoid licence proliferation5 by reducing the number of licences and categorizing licences into groups. Within the OSI’s mission statement it says “Open source is a development method for software that harnesses the power of distributed peer review and transparency of process. The promise of open source is better quality, higher reliability, more flexibility, lower cost, and an end to predatory vendor lock-in”. A commercially focused statement, geared to attract business to open source, by setting out a list of some of open source’s attributes and the associated business benefits. A far cry from Stallman’s free software ideology! This lacks the paternalistic approach of protecting both freedom in code and users from themselves. Whilst there are clear ideological differences there are not so many practical differences between the two groups. One of the key practical differences is the existence of permissive licensing through the OSD A permissive licence is one that has few requirements around how the software licensed under it may be distributed and allows wide usage of the code. It allows distribution on an open basis but does not require the copyleft or share-alike contribution-back of modifications and which may allow for modified versions of the code to be distributed under a different or modified licence. Examples of licences that are permissive include the Apache6, BSD7 and MIT licences8. Permissive licences, such as the BSD do not require derivative works to be distributed under the same licence terms as the original code and allow additional or different licence terms to be added to the modified version of the code. Whilst they require an acknowledgement of the author in the code’s notices, inclusion of a disclamer and in some cases such as Apache, the preservation of the NOTICE.TXT file, they do not meet the enforced protection on long-term freedom Stallman sought in his copyleft approach. The width of the licence means that the user has choices that may, arguably, give the individual more freedom in how they use and distribute the code, but which remove the collective freedom enshrined in the copyleft principle. The FSF was however concerned that permissive licensing does not protect freedom. In the case of BSD distributed code, the code can be re-distributed under other terms and could even be taken into proprietary code without infringing the original licence of the code. As the code may be incorporated into a copyleft piece of code, it is potentially GPL compatible and so can be combined with software meeting the free software definition. However, code released under the GPL could not be combined into a BSD licence as this would have the effect of changing the licence and be a breach of the GPL licence. Where code is copyleft, a licence cannot generally be changed without the consent9 of the creator

5 the name given to the large volume of approved licences for open source software which had been approved by the OSI, some of which were neither popular nor frequently used, which had terms which might not be compatible with other approved licences and the volume of which caused confusion. 6 http://www.apache.org/licenses/ 7 http://en.wikipedia.org/wiki/BSD_licenses 8 http://en.wikipedia.org/wiki/MIT_License 9 GPL section 10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the

Open Source 5 of that code. A significant problem in the existence of the two groups is that not all permissive licences approved by the OSI may be suitable for combination with coplyeft software licences. Ideologically the free software and open source movements saw themselves as being quite far apart. From the outside looking in this may be very confusing and even seem a little nonsensical. Both movements distribute software that has freely available source code. The difference to any user would lie purely within the terms of the licence chosen for the software’s distribution. The user may well not care what the licence terms are. They will only really be interested if the terms have any impact on them and so only users who play with the code by modifying it or combining it with other code are likely to care what the licence terms say. 2.0 Patents and Open Source The application of the patent system within the software sector has been contentious. It is not the purpose of this article to debate the pros and cons of software patents, despite the Open Source community’s obvious distaste for these. They are a reality and their existence has caused a number of major players to change policies in this area. Patent Pledges as a Response to the Open Source concern Patent Pledges take a variety of forms to and have been issued with a range of motives and goals but a single purpose of combatting patent restrictions. To understand the need for these one must understand the background of patents. A particular problem, recognised in the Open Source community, has been non-practising entities (“NPE”s); companies that seek to acquire patents not with the intention of using the technology specified but as a lever to extract licensing payments from other companies that are using the technology, ie NPEs do not operate a business that the patents apply to but the patents themselves and their licensing form the business model. Historically viewed as a US phenomenon NPEs have become an increasing issue in Europe too with Germany in particular becoming a popular venue for NPE suits. In recent times the NPEs have not been the only patent aggressors using their portfolios to attack Open Source. A number of companies’ behaviour has mimicked that of the NPEs using their patent portfolios as a lucrative source of licensing revenue. Some large players in the software industry appear to be generating more revenue from licensing their patent portfolio than from the development of software. Patents and Open Source The Open Source licensing structure or model may in fact be used as the basis of a patent pledge, with the pledge system attempting to replicate the ethos of the open source model which has historically been more (but not exclusively) copyright focused. The utility of patents in a sharing economy has been subject to challenge from many quarters. At the same time devices such as smartphones, have become ‘patent thickets’, where a single product’s components and features may be protected by a vast range of patents. In 2012 it was estimated that 250,000 US patents existed which related to smartphones10. One in six active US patents were considered to relate to this technology. It was further estimated11 that royalty payments could amount to nearly 30% of the cost

Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. 10 https://www.techdirt.com/blog/innovation/articles/20121017/10480520734/there-are-250000-active-patents- that-impact-smartphones-representing-one-six-active-patents-today.shtml 11http://www.wilmerhale.com/uploadedFiles/Shared_Content/Editorial/Publications/Documents/The- Smartphone-Royalty-Stack-Armstrong-Mueller-Syrett.pdf

6 Amanda Brock of a smartphone. This is a paradigm shift from the kinds of innovation patents were designed to protect and reward. The success of the Android operating system for mobile phones has led to challenges from the established providers of operating systems and mobile phones, who have met the challenge of a new competitor in their market by using patents as a weapon in their armoury. Numerous lawsuits were conducted between major companies such as Apple, Google and Samsung, around smart devices. Litigation is time consuming, expensive and uncertain as to its outcome. As far back as 2011, companies such as Apple and Samsung were suggested to be spending more on patent litigation that they have on research and development12. This is certainly not the outcome that those who allowed a potentially monopolistic right designed to reward invention to be enshrined in law had sought to achieve. Understandably many were left to wonder whether the patent model is a realistic one in a sharing economy where a single device might contain such a large number of patents and where no one party is dominant as the holder of those patents. Indications that even significant players in the technology sector take the view that attempting to enforce patent rights through the courts is no longer a productive or appropriate approach for that sector are becoming common and it has been reported13 that Apple and Samsung ended all patent litigation outside the United States whilst Google and Samsung have entered into a broad cross-licensing agreement14 designed to prevent future litigation. In addition to these truces amongst giants, there is an increasing use of patent non- aggression (or non-assertion) mechanisms in the form of a variety of Patent Pledges in open source. A Patent Pledge is a commitment from a patent owner that restricts how they will use their patent, and in legal speak forms an encumbrance on a patent. These are of varying parameters, and provide third parties with a degree of assurance that the third party will not face patent litigation / law suits from the patent holder(s) who have made the pledge on the patent they have made available. The pledges range greatly in both their form and content. They may be made by an organisation or individual on its own behalf or as part of a pool or group. These pledges are made in recognition that in order for the wider technology to flourish, there needs to be cooperation between different patent owners. This will remove the fear of unnecessary and expensive litigation and a more open source like model is being developed to allow for the existing patent infrastructure to be effective in such a sharing and open economy. 4.0 Patent Pledges and Promises 4.1 Unilateral assertions of non-aggression by an individual or group of patent holders: The patent holder commits to use it’s own patents solely for defensive purposes subject to certain conditions. Such a pledge may be either general in its nature - as with Tesla’s assertion that it will not enforce its patents against any other entity - or may be offered to a smaller subset of entities.

12 http://www.forbes.com/sites/davidthier/2012/10/08/in-two-years-the-smartphone-industry-has-spent-more- than-20-billion-spent-on-patent-litigation/ 13 http://www.bloomberg.com/news/2014-08-05/apple-samsung-agree-to-end-patent-suits-outside-u-s-.html 14 http://www.bbc.co.uk/news/business-25908364

Open Source 7

“The Patent Pledge”15 which has been signed by more than 30 companies – unlike Tesla’s individual pledge this is an example of a group making a pledge and including significant players such as DropBox and many small companies- is the shortest form, pledging “No first use of software patents against companies with less than 25: people.” 4.2 Contractual Effect of Patent Pledges Whether such a pledge is contractual or non-contractual in nature will have a significant impact upon the extent of liabilities that may arise from relying on it. Small undertakings are at greater risk than larger companies from the activities of patent trolls and the intent behind the pledge is clearly to give a measure of re-assurance to such undertakings. In terms of scope, the pledge is wide and, in the words of it’s author, are modeled on Google’s corporate motto - “Don’t be evil”. The commentary accompanying the pledge states that the intention is that it is “not legally binding”16. This issue arises frequently in the context of Patent Pledges of this nature and indeed more generally. In contractual terms, for an agreement to have legal effect there needs to be an offer, acceptance, consideration (in some jurisdictions) and the intention to create a legally binding agreement. It is questionable whether the Patent Pledge providers could place confidence in the commitment in the Pledge having no legal consequences. Further, the application of the doctrine of estoppel may operate to prevent a pledger going back on its promise to the world. An additional source of ambiguity within the Patent Pledge relates to the definition (or lack of) of the term ‘software patents’ themselves. Contracts may be bilateral or unilateral, where only one party commits to do something. An offer of a reward for the return of a safe pet would be an example of this. Some patent pledges can be viewed as being contracts, but being unilateral not bilateral contracts. In the case of the , its “Open Specification Promise” states that: “No one needs to sign anything or even reference anything. Anyone is free to implement the specification(s), as they wish and do not need to make any mention of or reference to Microsoft. Anyone can use or implement these specification(s) with their technology, code, solution, etc. You must agree to the terms in order to benefit from the promise; however, you do not need to sign a license agreement, or otherwise communicate your agreement to Microsoft.”17 Although this would likely be sufficient to establish a legal relationship between Microsoft and the User of the patent, despite there being no formal acceptance, it may be viewed as a unilateral contract. However it may be a challenge to prove knowledge of and agreement to the terms of the Pledge. 4.3 Restrictions and the Open Source Definition in Patent Pledges There is invariably some form of restriction in a patent pledge on the use that may be made of the patented technology. In most cases this is linked to the pledge allowing use of the patent with respect to the development of open source software. The “Computer Associates PNA” has such a restriction and continues to define the concept of Open Source software in the following terms: “Open Source Software is any computer software program whose source code is published and available for inspection and use by anyone, and is made available under a license agreement that permits recipients to copy, modify and distribute the program's source code without payment of fees or royalties. All licenses certified by opensource.org

15 http://thepatentpledge.org/ 16 http://paulgraham.com/patentpledge.html 17 http://www.microsoft.com/openspecifications/en/us/programs/osp/faq/osp-general/default.aspx

8 Amanda Brock and listed on their website as of August 1, 2005 are Open Source Software licenses for the purpose of this pledge.” IBM has been one of the most active players in the field, consistently topping tables for the number of patents awarded in a year.18 In 2012 alone it was awarded nearly 6,500 United States patents. In its current incarnation, the IBM pledge operates in relation to open source software; defined in the following terms: “Open Source Software is any computer software program whose source code is published and available for inspection and use by anyone, and is made available under a license agreement that permits recipients to copy, modify and distribute the program’s source code without payment of fees or royalties. All licenses certified by opensource.org and listed on their website as of 01/11/2005 are Open Source Software licenses for the purpose of this pledge. “ The IBM pledge continues to refer to and validate all licenses certified by opensource.org and listed on their website as of 01/11/2005. This is a precise formulation but other pledges refer only to the general notion of open source (or free) software and this may cause definitional problems in the future. Clearly a wise drafter of a pledge will tie its parameters to a clear definition and probably the Open Source Initiative approved licences. As mentioned in the opening paragraph of this article, Red Hat amended their patent pledge in 2017, to extend to “Covered FOSS” which is explained in the FAQ o the pleadge as follows “While the old Promise covered approximately 35 percent of open source software, the new version will cover more than 99 percent. It applies to all software meeting the free software or open source definitions of the Free Software Foundation or the Open Source Initiative and listed by the FSF or OSI.” 4.4 Networked or Pool non-assertion policies Networked or pool approaches do not fall into a single category. They generally involve the work of two or more organisations’ collaborative efforts to establish templates for patent sharing or licensing agreements (including PNA provisions) that may be used by companies or groups of companies collectively. 4.5 Networked or pool mutual non-aggression in the form of a defensive patent pool A second form of networked mutual non-aggression comes from defensive patent organisations like “The Open Invention Network”, www.openinventionnetwork.org”. OIN is a collaborative effort applicable to a specific field of use of the patents, which is set out in its “ System Definition” (a list of software packages which is updated on a regular basis). The Open Invention Network (OIN) was established in 2005 as a response to concerns that the increasing number of patent lawsuits was threatening the open source software community. OIN is the biggest patent non assert organisation in the World, founded and funded by 6 initial Members, IBM, RedHat, Philips, , Suse and latterly joined by Google as a 7th Member, and Toyota with TomTom and joining as Associate Members. There are circa 2200 OIN licensees who have signed up to the OIN licence agreement at the time of writing this article. OIN operates on the basis of a written licence agreement that defines the terms of licence and which must be signed by a party wishing to take advantage of its provisions and become a licensee. The licence may be entered into by any individual or entity at no financial cost to them. The members agree to a free “mutual hold harmless” type of

18 http://www.theregister.co.uk/2013/01/10/ifi_uspto_patent_issue_list_2012/

Open Source 9 relationship to each other, covering patents held by them that read against the Linux Definition. The pledge is irrevocable but does allow licensees to assert an election and revoke the patent pledge on a go-forward basis, although this will not remove past participation and obligations to the group of licensees at the point in time of withdrawal. OIN’s website states that: “Open source software has been one of the greatest sources of innovation. Open source developers have built excellent software solutions for business, education and personal use. Free and open source programs give companies, schools, governments and users more choices, ensuring that they are getting the best possible technology for their needs. Unfortunately, the last decade has seen an enormous rise in software patent suits. Open source developers aren’t any more immune to this threat than other software vendors. However, the culture and innovation modality of open source software, based on engagement and sharing, made it natural to build a collective defensive solution to protect and enable it.” OIN additionally actively acquire patents that might be used by NPEs to threaten companies seeking to develop in the Linux environment and has taken on patents from entities that cease or plan to cease to exist or conduct business. Specific provision is made regarding choice of law issues with the agreement providing that it “shall be construed in accordance with the laws of the State of New York as such laws apply to contracts entered into and fully performed in the State of New York.” License on Transfer Network (LOT) is a smaller and newer organisation, following closely on the OIN model. Its website19 states: “Had we formed LOT Network in 2005, and if every operating company had joined that year, approximately 10,500 defendants in PAE litigations using acquired patents may have collectively avoided more than $100 billion in costs. These litigations have slowed the pace of innovation, increased prices to consumers, lowered returns to shareholders, and disrupted our best technologists and senior management from moving our technology and our national economies forward. Instead of spending valuable time doing innovative and productive work, our employees are frequently pulled into deletions and court defending our companies against patent lawsuits that often lack any merit. These litigations impede every aspect of our businesses.” The LOT network claims to coordinate 405,493 patent assets, including over 111,000 US issued patents. It currently has almost 70 member companies including a number of very major players20: The LOT licence is perhaps the most extensive licence developed by any organisation. It extends to 21 pages and almost 12,000 words. It provides that members will grant each other provisional licenses in respect of the use of the protected technology. Unlike other agreements there is no provision for members to exclude certain patents from their commitments – which may account for the large number of patents existing under its umbrella. The major distinguishing aspect of the LOT network is that for the majority of time and in presumably the vast majority of cases the agreement has no impact upon its members’ activities. The philosophy behind the network is broadly supportive of the patent concept but the concern is that rights may be transferred from practicing to non-practicing entities and the key provision is that in the event any patents are transferred to a non-member, the

19 http://www.lotnet.com/index.cfm 20 At the moment, the LOT network has nine member companies: available at < http://www.lotnet.com/learn- more/members.cfm>

10 Amanda Brock other LOT members will immediately obtain a royalty free, non-exclusive right in respect of that patent. The licence will extend only to the patent(s) that are involved in a transaction and will not extend to any other patents in a member’s portfolio and the licence effectively requires any transfer to be subject to an encumbrance of the patents for the benefit of the LOT members. 5.0 Difficulties of Pledges and Open Source licensing One of the difficulties inherent in this field is the style and content of the pledges. This is not a new issue in the world of open source as many of the licences drafted in the early days of open source were drafted without legal advice. There is obvious merit in having a pledge written in a style that can be understood by non-lawyers and in particular Open Source developers, but this may give hostages to fortune by allowing for ambiguities that may subsequently form the basis of legal proceedings and confusion. Another issue identified by Google in the FAQ attached to its PNA is the relationship between patents covered within the scope of a PNA and others which may have been dealt with on a different basis. Two points are of particular relevance. “Existing open source licenses with patent provisions, such as Apache License 2.0, Public License 2.0 and the GNU General Public License v321 are principally copyright licenses. As such they are typically used in connection with software source code distributions. Google releases a lot of open source code under such licenses. The OPN Pledge is oriented to open source software to which Google has contributed little or no code, or has not otherwise incurred any patent license obligations under patents it owns. The OPN allows us to do more than conventional patent rights transfer mechanisms around free or open source software would allow. The Google PNA should therefore be seen as an addition to existing licensing provisions, a point that is considered further in the FAQ. Given that Google has released lots of code under open source licenses with patent provisions, such as the Apache License 2.0, one may question what impact the OPN Pledge is intended to have on such releases? Google has released and will continue to release OSS code under a variety of license agreements, including those with patent grant provisions. The OPN Pledge is not intended to conflict with or supersede those more specific arrangements. In particular, the patents selected for inclusion in the OPN should have little to no bearing on patent rights that Google may have granted under prior OSS releases.” This is a significant issue. In addition to issues regarding the scope of individual PNAs, the interaction between the PNA and any applicable licence(s) must be considered. Layers of consistency may arise also in another respect. PNAs and licences are not static documents The General Public Licenses (GNU GPL) developed by the Free Software Foundation is another example of a licence that was originally designed to deal with copyright issues but which has been extended in recognition of the potential impact of software patents. The licence has gone through several incarnations and the current licence family is GPLv 3. Discussion focused on this and the perceived problems with version 222 that it sought to address. In discussions of the draft of GPLv3, Richard Stallman, commented regarding the intention behind the new provision. It is a standard feature for PNAs to provide that they

21 See section 4.2 22 Version 2 of the GNU licence was adopted in 1991 and predates many of the issues relating to the award of software patents.

Open Source 11 will cease to apply in the event that a beneficiary initiates action of some sort. The notion is that it is legitimate to use patents as a defensive tool rather than as on offensive weapon. As was stated: There are two kinds that we said are ok. One kind is where retaliation only occurs against aggression. Party A sues B, B is most likely to counter-sue. We've decided we want retaliation only against A, not against B, even if he counter-sues. Retaliation is only against those who commit the aggression and not against those who are themselves retaliating. So we have a way to make the distinction.23 However, would a PNA in fact be a further restriction and void under the GPL? Conclusion: Open Source is not the natural home of patent ideology, but from that friction the seeds of change have been planted through open source Patent pledges and a model has been formed not only to support the sharing economy but which can be replicated in other areas which might experience patent thickets. *** Amanda Brock, European Representative, Open Invention Network Sept 2017 A senior in-house lawyer, Amanda was most recently Associate General Counsel, Digital Financial Services working in emerging markets at the MNO Veon and as a European Representative of the defensive patent pool, the Open Invention Network. Having been General Counsel, Canonical – lead commercial sponsor of the open source operating system Ubuntu and open stack cloud partner - for 5 years, she set up and managed the global legal function. Her tech and digital experience is diverse and includes software, cloud computing, big data, IOT, new governance models in the sharing economy, e-business, data and privacy, FinTech and device manufacture and distribution. Her experience of digital and IT law spans more than 15 years and a range of sectors, with roles as European Manager at DSG International where she was the first lawyer at the ISP, Freeserve (UK’s first and biggest .com IPO) and as part of the Digital Transformation team, UK Legal Director, Aramark and General Counsel Nicole Farhi & French Connection and General Counsel of a major Data Centre provider, her experience lends itself to her very commercial approach to legal advice. As a solicitor, she is admitted in England and Wales and in Scotland and has a Masters in IP and IT law from Queen Mary and a Masters in Comparative Jurisprudence from New York University. Amanda has been a Visiting Research Fellow at Queen Mary and was a founder of the QM Legal Incubator Q:Legal. A regular speaker at international academic, legal and business conferences she has also written extensively on open source, tech and commercial law, and is the author of “E- Business: A Practical Guide to the Laws” (2 editions) and the commercial agreements section of the Oxford University Press’ “Free and Open Source Software, Policy, Law and Practice”. The Open Invention Network is a shared defensive patent pool with the mission to protect Linux. Launched in 2005, OIN has strong industry support with backing from Google, IBM, NEC, Philips, Red Hat, Sony and SUSE. Any company, project or developer that is working on Linux, GNU, Android or any other Linux-related software is welcome to join OIN, free of charge or royalties www.openinventionnetwork.com

23 http://fsfe.org/campaigns/gplv3/patents-and-gplv3.en.pdf