Last week, I attended the OmsocomBB workshop in Berlin. Harald Welte was an excellent host and c-base was a fun venue. I got a chance to put faces to some of the names I see in email and got a much better picture of who is doing what in the world of open source cellular. I got to see Dieter Spaar show his work on writing his own RNC software for UMTS Node B equipment. It was like the early stages of OpenBSC all over again (except that the nauseating complexity of UMTS makes GSM look quaint). I finally met the Alexander Chemeris in person. And I got to spend a week in Berlin, one of my favorite cities.
One the last day of the workshop, H-Online published this article about the OpenBTS and OpenBSC, the two well-known public-release GSM projects. I assume the timing was coincidental (or maybe Jungian synchronicity?), but it was an appropriate terminus for the event.
By chance again, when I returned to California, I had a meeting with a former executive from a major cellular carrier, a big European carrier with a global reach. He told me horror stories about IMS complexity and licensing fees. He pointed me to this article about how big web service providers are buying "raw" network equipment, built to spec in Chinese factories, and then loading that equipment with their own firmware and software, usually a mix of open-source and in-house applications. This executive sees the same thing in the long-term future of telecom: commodification of hardware all the way down to the RAN head, networks based mostly on open source software and a core network protocol based on SIP but a lot simpler than current IMS attempts. In his mind, this is the inevitable "terminal state" of the telecom industry, an inevitability in which the current generation of NEPs have no place. It is a market that will be served by companies that look and work a lot more like Red Hat than like Nokia-Siemens. I see that vision too, and I see products (not projects, products) like OpenBTS and OpenBSC and yate having comfortable places in that world. If we are correct about this vision of the future, then that small gathering of hackers, freelancers and entrepreneurs in Berlin last week may have held the seeds of a revolution that will fundamentally change a multi-trillion dollar industry. That might sound very ambitious, but the software industry has seen revolutions from modest beginnings before and the telecom world is begging for this kind of disruption.
(David Burgess is a lead developer in the OpenBTS project and one of the founders of Range Networks.)
