Analyst Nolle: Open Source More Likely to Solve Telecom's Problems
A veteran industry strategy consultant says the traditional standards development process isn't going to play much of a role going forward because it's too slow, and he points to the developing 5G standards as an example.
Tom Nolle, president of CIMI Corp. , and someone I frequently quote because I think he's smart, says open source processes represent a better approach to consensus on standards than the traditional processes, but quickly adds there are problems there, too.
"I think 5G is demonstrating, and NFV/SDN have already demonstrated, that we are going to have to figure out a different way to do this or we've lost any hope of relevancy," Nolle comments in an interview, following his 2018 predictions blog. "So compared with standards, open source is great."
Because of anti-trust laws in the US and Europe, an open consensus-based process is required for developing telecom specs, but in standards bodies, those processes have for too long been dominated by vendors, whose concern for near-term products and profits can skew their participation toward obstructionism, Nolle notes. There are countless examples over the past few decades of standards processes bogging down as vendors pressed their own advantage and their own versions -- think IMS, AIN and even ISDN, for those with long memories.
"The advantage open source has is, in the main, vendors are not tremendously interested in adopting open source because they don't make money on it," he continues. "So vendors have been inclined to be less obstructionist in open source areas than they have been in standards body and that reduction in obstruction from vendors is one of the factors that allows the open source to move forward."
If you sense the "but" coming...
Nolle believes that an underlying presumption of the open source process is they are addressing software-driven functionality. The success of any given open source project, in his mind, will be based on having the right software architecture. "In the main, the people involved in these processes are not software architects, however, so we don't necessarily start open source projects with the right approach," he adds.
One project that is off on the right foot in that regard, in Nolle's opinion, is the Open Network Automation Platform (ONAP) -- and yes, that is a reversal of some earlier comments he made on this project. (See ONAP Takes Flak as Telcos Prep for Release 1 and ONAP Takes Flak as Telcos Prep for Release 1.)
Nolle now describes ONAP as "the only hope remaining for successful software-driven networking in telecom," adding that this lofty position is based on four years of work internally at AT&T Inc. (NYSE: T).
"The only reason ONAP is any good is because AT&T did it internally and then made it open source," he says. "If they had ceded the concept of ONAP to open source four years ago when the process started, I think they would still be screwing around with scope and other discussions. We'd be nowhere close to where it is today."
What it all comes down to is that the industry needs consensus, and in today's world, that's easier to achieve in open source than in traditional SDOs -- assuming they get the software architecture right to begin with, Nolle says. In fact, he adds, open source approaches may be the only way to tackle the scope of the issues the industry faces.
"If you look at NFV, for example, the real problem isn't NFV anymore, it is service lifecycle automation -- and that envelopes cloud hosting, data center operations, Big Data, FCAPS, OSS-BSS -- there are so many pieces that there is probably no vendor who would try to dream of implementation that broad," he notes. "But the business case of software-defined next-gen networking demands that kind of breadth or you are not going to be able to deploy it."
Open source represents a way of developing such a "huge strategy" because "you don't have ten different vendors all duplicating efforts trying to get things done, you have a cooperative activity. If you make the assumption that they get the software architectures right then the reason why ONAP and concepts like ONAPs succeed is because they are too big for anybody to do any other way," Nolle concludes.
It's an interesting perspective on open source at a time when there is still broad industry debate on its value and whether it does speed innovation and consensus, and that's a topic that will certainly rise to the fore in the coming year, so stay tuned.
— Carol Wilson, Editor-at-Large, Light Reading