The Anti-Patent: An Alternative Approach to Patents and Open Source
At the recent IPO conference in Chicago, a panel was held entitled “The Intersection of IP and Open Source” featuring Tom Li of Adobe, Mark Radcliffe from DLA Piper, and David Marr of Qualcomm. One of the key ideas presented at the panel is that open source software and patent protection are not mutually exclusive. I want to distinguish between three related questions about the relationship between patents and open source:
- Why would a for-profit software developer want to be a member of an open source consortium?
- Why would an open source consortium want its members to hold patents?
- Why would the members of an open source consortium want to pursue patents?
After briefly considering these questions, I provide an alternative model of a defensive patent portfolio that may be relevant to members of an open source consortium.
Why would a for-profit software developer want to be a contributor to an open source consortium?
Let’s consider two examples of different business models: Red Hat and Microsoft. Red Hat makes money by installing and maintaining open source software. It supports open source software because it sells support for that software. Microsoft, on the other hand, has historically been opposed to the open source model. In fact, former Microsoft CEO Steve Ballmer famously called open source software a ‘cancer.’
However, this isn’t the Microsoft of today. In fact, Microsoft has done an about face and joined the Linux Foundation. The case has been made that their motivation for doing this was the realization that the future of their business was in cloud services like Azure and Microsoft 365. These revenue streams don’t necessarily depend on adopting a particular operating system. So, Microsoft made the decision that supporting Linux would likely lead to more revenue from cloud services than restricting support to Windows.
There aren’t a lot of successful examples of companies providing service-for-software like Red Hat, but SaaS, and cloud services in particular, are the wave of the future. It may not be an exaggeration to say that every major provider of cloud computing services utilizes open source software for at least some of their products. Interestingly, Microsoft and Red Hat have even teamed up to provide Red hat on Azure.
Why would an open source consortium want its members to hold patents?
On the face of it, it seems like patents are at odds with the objectives of open source licensing. In fact, the preamble to one of the most popular licenses, the GPL, seems pretty hostile to patents, stating “every program is threatened constantly by software patents. States should not allow patents to restrict development and use of software on general-purpose computers, but in those that do, we wish to avoid the special danger that patents applied to a free program could make it effectively proprietary.”
But in fact, an open source consortium can greatly benefit by having members with robust patent portfolios because it provides all users some defense against patent litigation. That is, while an open source license may provide the right to use particular code, it doesn’t guarantee that a non-contributor won’t obtain and assert patent rights against users. Having consortium members with robust patent portfolios can therefore be essential to the success of a large open source project.
Thus, while patents may be threat to the development of software, they may also be the only defense against this threat. If the members of a software consortium don’t have any patents, then the section of the GPL providing a royalty free patent license to all users who conform to the license is essentially useless.
Why would the members of an open source consortium want to pursue patents?
An article distributed by the law firm Fenwick and West lists four reasons that an author contributing to a GPL project may want to pursue patents as well:
- the author may plan to license the patent to others
- the author may want to assert patent rights against those who violate the GPL license
- the author may want to assert patent rights against those who are not using the GPL’d software; and
- the author may plan to also distribute a non-GPL’d version of the software.
These are all valid reasons, but patents are not cheap so anyone thinking about pursuing a patent portfolio must weigh the costs and benefits carefully. Without the obvious motivation of acquiring a temporary monopoly on the use of a particular technology, the case for pursuing patent protection is greatly diminished. While the Linux Foundation is likely overjoyed to have the massive Microsoft patent portfolio behind the project, it is unlikely that Microsoft wants to devote as much money pursuing patents on the work done by engineers working on Linux as they would for proprietary projects. That isn’t to say there is no reason for them to pursue patents, just the motivation is reduced.
An Alternative Patent Portfolio for Open Source Contributors
Despite the list of reasons above, the intersection of patents and open source is primarily based on the fact that patents serve a defensive purpose. That is, (at least in theory) you can’t get a patent on something that is already patented. Retaliatory litigation is another aspect of a defensive patent strategy, but that may not be particularly effective against non-practicing entities (i.e., trolls).
Of course, any kind of publication can be used to invalidate a patent, so a defensive portfolio doesn’t actually have to consist of patents. In fact, MIT, Cisco, Google and the USPTO have recently created a prior art archive that includes a wide variety of documents. Unfortunately, the search tools that examiners are most likely to use (i.e., EAST and WEST) are largely limited to patent publications.
Still, preventing a patent from ever being granted is far cheaper than litigating a granted patent. And while it is possible (and desirable) to monitor and intervene in the patent examination process, it would be more desirable for a defensive patent thicket to prevent new applicants from getting patents allowed in the first place (at least in areas covering open source technology).
One point that many people may not take into consideration is that a patent doesn’t actually have to be granted in order to be included in the search databases used by examiners. Thus, an abandoned patent application may be nearly as effective for defensive purposes as a granted application. With this in mind, some open source contributors may consider a strategy of preparing patents expressly for the purpose of getting them published, and then abandoning them. Such an application may be termed a defensive patent, or an anti-patent. It couldn’t be used to enforce a monopoly, but it would still be right there in the patent database ready to be found by examiners looking for prior art.
In addition to being much less expensive than a traditional patent, an anti-patent does not need to be limited by many of the considerations of a traditional patent application. For example, many patents are written in a way that purposefully obfuscates what is prior art and what is new matter. An anti-patent can gratuitously admit a wide range of prior art. Thus, while patents and open source software are not mutually exclusive, a defensive patent portfolio does not need to be limited to traditional patents. A portfolio that includes both granted patents and abandoned anti-patents can achieve many of the same objectives at a lower cost.
