Bruce Perens has an article over at Gigaom about the lack of transparency in the recent settlement between Software Tree and Red Hat. Perens worries about whether the open source community is “in peril” as a result. “Peril” may be a bit melodramatic, but Perens’ concerns are sensible:
Should Acacia decide to prosecute the patent against them, open-source developers’ own standing has likely been damaged by Red Hat’s action. According to Heather Meeker, author of The Open Source Alternative: Understanding Risks and Leveraging Opportunities and A Primer on Intellectual Property Licensing, “Once Red Hat settles and pays for a license for the patent, any subsequent defendant will find it more difficult to convince a court that the patent should not be enforceable.”
Perens comments echo the concerns of developers that create products and business models around open source software. Open source developers rely on software licensing to provide rights to software that they use and incorporate into their own software products. Provided the developer follows the rules set forth in the license, they need not fear the author(s) of the underlying code will someday seek royalties or injunctions halting the developer’s business.
Unfortunately for the developer, the open source proprietor can only convey that which it owns. If third parties, not party to the open source licensing, happen to own patents relating to the developer’s products, the fact that the software was obtained through an open source license is of little, if any, use (barring any specific commitments by the open source proprietor to defend the developer in such an event).
In this specific example, Perens echos the concern that Red Hat’s actions may have strengthened Acacia, and placed developers building JBoss related applications in a worse position by “validating” Acacia’s patent. Of course, the real issue is that no one knows (other than Acacia and Red Hat). The consequent confusion and uncertainty
is benefits no one. For example, its possible that Red Hat obtained a license that specifically authorizes developers to build applications without fearing legal claims from Acacia. Its also possible that the legal doctrine known as “exhaustion” bars any relief that Acacia might have against downstream developers.
In the most recent case on the issue, Quanta v LG, the US Supreme Court denied a patent holder the right to enforce its patent against a downstream product developer. However, in that case, the product developer purchased components “substantially embody[ing ] the patent” from a provider that had a broad license specifically authorizing it to sell components free of the plaintiff’s patent claims.
In the case of Acacia, I haven’t reviewed the pertinent patents, but whether an exhaustion defense would arise from a Red Hat license may also depend on what product the developers receive. If that product itself doesn’t “substantially embody” the patent (for example, the developer must introduce some substantial component on its own in order to come within the boundaries of the patent claims), then Red Hat may not be a purveyor of a licensed product.
The JBoss developers, meanwhile, apparently have no information about the scope of the license (if any) Red Hat obtained from Acacia, which eliminates even the option of obtaining a legal opinion, however expensive it may be. I would have to agree that a little sunshine would go a long way. Dr. Roy Schestowitz summed it up nicely:
Next time around, Red Hat ought to put transparency first. The person to push regarding the secrecy turns out to be Tiller, not Fontana. He may not be able to unsign an NDA, but the same mistake oughtn’t be repeated at a later date.
Also, Dr. Schestowitz’s article suggests the possibility that Acacia paid Red Hat, as an alternative explanation for why there has been silence on the matter.