The Role of the Patent Trolls
On Wednesday, the Financial Times published a piece by Alan Cane called Trolls control the rickety-rackety bridge of intellectual property. In the article, Mr. Cane points to the cost of patents in products and draws attention to patent trolls that "participate in or monitor standards-setting processes without revealing, until it is too late, that they hold patents essential to the standard." While he finds that these activities can be problematic, he concludes by saying that "[p]atents are property and can be bought and sold. There is nothing wrong with holding an asset from which you hope to profit."
Although patent trolls can pose problems within standards bodies, they are a far greater problem outside of a standards body than if they are a member. Most standards organizations are based on RAND concepts, but each has a different set of rules governing its operations. Those rules are generally designed for the particular standards context and the types of activities that the organization engages in.
For example, some standards organizations require that their members positively disclose any necessary patents they may have. Other organizations require members to only notify the organization of any necessary patents that they are unwilling to license on RAND terms. Without debating the merits of each approach here, the important point is that these obligations are imposed upon members of the standards organization. To go back to a point I made in an earlier post, standards members generally cannot refuse to grant a necessary patent license, and that license must be on reasonable and non-discriminatory terms. Even patent trolls that are standards organization members are bound by this obligation.
Although disputes can and do arise as to whether patent licensing terms are indeed reasonable and non-discriminatory, those disputes are exceedingly rare. There are hundreds of RAND based standards organizations, which publish thousands of standards, which are implemented in millions of products. Yet the number of cases brought each year over RAND licensing terms can usually be counted on your fingers.
It's also important to keep in mind that even in the standards context, patent licensors operate in the marketplace. If their patent royalties are too high, the market will not adopt the standard and the patent owners will not make money. That pressure tends to result in patents being reasonably valued. In the patent pool process, it's very common for the pools to lower their fees in response to market forces.
Despite a handful of high profile issues with patent holding standards members, the problems with patent trolls tend to arise outside, rather than inside, standards bodies.
Legal Overlord? I Prefer Benevolent Counsel
James Governor, an analyst with RedMonk, has welcomed the new Legal Overlords to the conversation, for which I thank him. I absolutely agree with James that lawyers need to be part of the conversation.
There seems to be a perception that standards are developed by groups of engineers working in harmony to reach the optimum technical solutions. The reality is that standards are venues for both cooperation and competition between rival interests. Making a standard successful in this environment often requires the involvement of individuals from many disciplines. Technology, business, and legal must all work together to ensure that the right technology is solving the right business problem using the right legal framework.
Lawyers play an important role in this process. In some standards bodies, the only thing that is perhaps more complicated than the technology is the legal framework that governs it.
As several of James's commenters noted, lawyers have certain constraints when speaking publicly that others do not. As a lawyer, I have legal and ethical responsibilities to my client, Microsoft, and often to the standards bodies I work with. These obligations will guide my writings on this blog. This isn't a fine line test. Rather, if there is a topic that can even be remotely construed as interfering with my legal and ethical obligations, I simply won't discuss it.
While smart minds can differ on a conclusion, it's important that the conclusion is founded on appropriate facts, background, and knowledge. It's my goal that through this blog I can provide a legal perspective and background to the standards conversation. That leaves us with lots to talk about.
My Welcome to the Blogosphere
In my first 24 hours in the Blogosphere I've received a warm welcome (thanks Jason) and, well, a not so warm welcome from Dana Blankenhorn over at the ZDNet Open Source Blog. Despite Dana's urging me to stay offline, I figured it might be more interesting to respond to his comments.
Dana first takes issue with my characterization of RAND licensing practices:
"First Rudin describes the controversial RAND (Reasonable And Non-Discriminatory) licensing schemes that some proprietary firms have been trying to force into the Internet standards, in order to make it a permanent toll road, as a universally-accepted concept. I think that's pulling a fast one, but I'll let it pass."
Though there are certainly those such as Dana that don't agree with RAND licensing, most standard organizations operate on a RAND basis. ECMA comes to immediately to mind. IEEE and OASIS allow licensing on RAND terms, with or without royalty. Even the W3C, which is responsible for a great number of Internet standards, operates on a royalty free RAND basis. No pulling a fast one here. The point is that whether or not a patent royalty is allowed for a given standard, most standard organizations operate on a RAND basis.
Dana then takes issue with me calling patents a property right. Fortunately, I can let the U.S. Patent Office speak for me on this one:
What Is a Patent?
A patent for an invention is the grant of a property right to the inventor, issued by the United States Patent and Trademark Office....
The right conferred by the patent grant is, in the language of the statute and of the grant itself, “the right to exclude others from making, using, offering for sale, or selling” the invention in the United States or “importing” the invention into the United States.
Dana then moves on to copyrights, a separate intellectual property right from patents. Unlike patents, which protect inventions, copyrights protect the expression of ideas. The reason I bring this up is that it's important to understand the roles of patents and copyrights in standards.
Standard organizations generally publish technical specifications for others to actually implement. Those specifications, being expressions of ideas, are protected by copyright. Standard organizations have the right distribute those specifications, sometimes for a fee and sometimes at no cost (running a standards body isn't cheap and they have expenses they need to cover).
Getting the specification on its own, however, does not necessarily give people the right to implement that specification. That's where patents come in. Under RAND policies, patent holders in those standard bodies agree to give up one of their primary patent rights - the right to exclude others from making use of that patent.
It's important to keep in mind that standards licensing is very different from commercial software licensing that most people are familiar with. When you license commercial software, you get the rights to use that software. Those rights include both copyright and patent rights. When you want to implement a standard, the standards organization can only grant copyright rights to the specification. Implementers are still required to get the necessary patent rights, if any, from patent holders. Sometimes this means getting the rights from the patent holder itself, other times it might mean getting the rights from a group of patent holders through a patent pool.
What Microsoft has done with its Open Specification Promise goes beyond its RAND licensing commitments. It say that implementers of the covered specifications don't need to come to Microsoft to get a license - we've already given you the rights to use our patents.
Microsoft’s Open Specification Promise
A fair amount has been written today about Microsoft's Open Specification Promise. Publications like Eweek have discussed it, and my colleague Jason Matusow has provided links to commentary across the net. From all I've read so far, Microsoft's promise has been well received.
The question I'm going to address is how is this approach different from current industry practices. To those involved in the standards setting world, terms like RAND and RAND RF have intrinsic meanings. To everyone else, its just another acronym in a strange standards world of organizations made up of yet other acronyms (IEEE, IETF, W3C, AACS, DVD CCA, etc.).
Most standards bodies require their members to make licenses to their necessary patents available on what are known as RAND terms. So what does RAND mean? RAND stands for "reasonable and non-discriminatory." Although the particular rules vary from standards org to standards org, RAND means that standards participants are required to grant standards implementers patent licenses for those patents that must be infringed to implement that specification.
To put it another way, a patent is a property right and, like other property rights, the owner can exclude other people from using it. By joining a standards organization and agreeing to grant a RAND license, a patent holder gives up its right to exclude others from using its necessary patents within the scope of the specification. Moreover, the terms that govern the use of that patent must be "reasonable and non-discriminatory."
RAND-z or RAND-RF add another import restriction a RAND license- the patent owner cannot charge a royalty fee for use of that patent. What is still technically required, however, is that implementers of the standard must still obtain a license from the patent holder.
What makes Microsoft's Open Specification Promise different from RAND licensing approaches is that for those specifications covered by the promise, an implementer doesn't need to come to Microsoft to get a license. Instead, Microsoft has promised not to assert its patent rights against implementers using Microsoft's necessary patent claims in connection with the specifications covered by the promise. Nothing needs to be signed and we don't need to know you're taking advantage of our promise. As long as you adhere to the promise's limited rules, Microsoft will not take legal action against you. It's not a free pass, since there are limited conditions that still apply, but it does ease the legal load for implementing standards.