Open-standard bearers close ranks

Everyone loves open standards, but few agree on what they are. New principles lay down the law
Written by Rupert Goodwins, Contributor

All good politicians know that the best way to deal with an opposing idea is to ignore it. If that doesn't work, ridicule it. If it's still there, then subvert it. Claim it was yours all along while working to change what the words themselves mean. Once you've won the words, you've won the game and the original idea is lost.

Proponents of open standards and open software know this all too well. Nobody on the planet has a word to say against open standards: like liberty, the family and good television, they are universally popular. You just have to agree on how to define them.

Take IBM and Intel's recently announced open standard for blade servers. With physical, electrical and signalling components firmly defined and open for anyone to use, what could be a better demonstration of those companies commitment to the greater community? A standard that didn't require you to buy your server racks -- at a very plump mark-up -- from the standard-makers: sure, anyone can make the blades; it's just nobody else can make the boxes they go into. Can a closed box hold an open standard?

More urgently, take the anti-spam Sender ID proposals. Bill Gates has made it his personal mission to eradicate spam within two to three years. To that end, Microsoft proposed to make its Sender ID technology -- which is subject to a patent application -- available to the IETF for use in an open standard. For Microsoft, open meant free to use provided you sign a licence with the company that restricts what you could do. That level of control was unacceptable to the IETF and others. According to The Guardian, the company's refusal to make Sender ID available without restriction came from Bill Gates himself: the company told the newspaper that "a small vocal minority of about 20 firms have been influencing the IETF, and those firms have no concept of intellectual property."

That's an interesting stance, given that free and open-source software has unimpeachable intellectual backing from people like Professor Lawrence Lessig of Stanford Law School, and practicing lawyers like Lawrence Rosen (what is it with people whose names begin with Law?). True, Bill is the son of a lawyer, yet although he got an early introduction to the benefits of open development -- "The best way to prepare is to write programs, and to study great programs that other people have written. In my case, I went to the garbage cans at the Computer Science Center and I fished out listings of their operating system." he said in Programmers at Work, a Microsoft book -- I'd still take the word of qualified people on this one.

Fortunately for you, me and Bill, Larry Rosen has been at work studying the definition of open standards and has brought a lawyerly eye to the problem. At the Open Standards Alliance conference, Open Source, Open Standards: Maximizing Utility While Managing Exposure, in Arizona last week, Rosen presented five principles of open-source software alongside five compatible principles for open standards -- a handy, if somewhat portentous, 10 commandments for the open source/open standards movement. Boiled down, they say that in the case of software, people shall be free to use it for any purpose whatsoever - explicitly including combining it with non-open software, distributing it in any form as they wish, and making derivatives.

For open standards, the principles are analogous, but with a sting in the tail. The standard shall be free to distribute under an open-source licence. Anyone can create software based on that standard under unconditional licences to any patent claims in the standard. Anyone can give away -- or sell -- that software with an unconditional licence, the only exception being for reciprocal licensing of patents in that software. A licensee who sues anyone else on a patent claim that's part of the standard immediately loses their licence to use that patent.

These principles are the DNA of open development. They emphasise that open doesn't damage intellectual property, but instead works within the existing legal system of contracts, licences and the rights of property owners. They unambiguously lay out what it is to be open, what it means for software and standards that claim that title and how they interact with the non-open world.

Any company or organisation that claims that it alone understands how IP and standards work can now be asked to demonstrate how its ideas correspond to the basic rules of the game: anyone who flies under the open flag can be made to demonstrate they know what the colours mean. The words are safe: the arguments can move on to matters of fact.

Editorial standards