It’s been said that code is the coin of the realm. In the world of networking, code has taken center stage in recent years, shining in the spotlight of open source projects like OpenDaylight and Open vSwitch. The big idea is that if you can make networking work and you’ve got the code to prove it, you can move ahead with productization. Get enough people behind the project and community using the code, and all of a sudden, there’s a de-facto standard.
The key here is that code moves along quickly in well-organized groups who are heading towards a common goal. By contrast, standards bodies are moving more slowly. For example, I’ve both heard and witnessed that it can take two years or more to move a project from draft idea to ratified standard. Why? Bureaucracy. Politics. Process. Trivial use cases treated like major ones. And so on. This begs the question…why should a company trying to get something done wait for a standards body when code already exists?
Well, there are reasons. And we’re not predicting the imminent death of the IETF or IEEE. But maybe something’s got to change. In this Packet Pushers, we address this issue with Dave Ward, along with several other top of mind topics related to the future of networking like OpenFlow 2.0, PCE, I2RS, and NSH.