He says:

A lot of attention and money is being poured into Health care by US Department Of health and Human Services, who are promoting standards for Electronic Health Records systems, and a National Health Information network. The belief being that standards will promote interoperability and create a frictional-less flow of health care information between standard stores of information. This slippery flow of information will then open up heath care and enable a new mass market in which consumers now empowered by their own portable information and widely disseminated provider quality information, can shop for care on a level playing field.

… [It] would be a simple, secure, open extensible protocol for the exchange of health care information.

… It’s no accident that the most successful Internet protocols have the letter “S” (for simple) in their names. By being open, its specification would be widely disseminated and available on the Internet so that anyone could implement it. By being a simple protocol and one that leveraged existing protocols, its implementation and development would be simple. Implementation in a broad set of languages would increase the velocity of its uptake and use. Being open, it would not be owned by any special interest, and if its creation followed an open source, beneficial dictatorship model its creation could be marshaled and be “architecturally consistent.” Concentrated leadership and vision would speed its creation in contrast to a standards body based solution. Its openness would endow it with a level of trust by its users who could see into its structure and governance. This openness could be carried through into its implementation so that the contents of a transfer would be open to all parties in a specific transaction.

It would need to be extensible so that a small subset of the protocol could be immediately launched and put into use. This small subset would form the core of the protocol which could be extended by its users to bootstrap an exchange. It has always annoyed me that standards take so long to emerge because the participants want to get it right first time. In reality this is impossible since the real-world is simultaneously evolving and providing new and enhanced meaning to the standardized concepts.

Go ahead and read the whole piece. So, Adrian, are you going to propose such a protocol?!?