It isn't really "open source," the way that phrase is accurately applied to software development. The founding principle of the O-RAN Alliance (where "O-RAN" stands for "Open Radio Access Network") is for enough stakeholders in the future of wireless communications to agree upon 1) the identities and functions of the individual components of a wireless network and, 2) the requirement for just one, non-proprietary means of making these components interoperable in a telecom system. That seems noble enough.
It also isn't really 5G Wireless. For the O-RAN concept to work, it must make itself downwardly compatible with as many pre-existing networks as possible. Theoretically, a 2G GSM system should, at some point, be capable of being retrofitted with O-RAN architecture.
What O-RAN actually is, as much as it can be amid the current state of the telecommunications market, is the framework for a strategy. When "disruption" is a good thing, it's a way to make room in a market for new competitors. The Radio Access Network is the part of a cellular communications system that enables a wireless device to join a network and participate, as the network maintains that device's mobility and likelihood of transitioning between coverage areas, or "cells." More simply put, it's the front door. If the whole point of a wireless network is to maintain accessibility, and the optimal wireless network is one that follows a global standard, then you would think an open architecture for front doors would be a premier necessity.
When a strategy needs backing and support, it should become a cause. When that backing is being beckoned from a country's government, it must become a cause célèbre.
"We must continue to deploy 5G networks as quickly as possible, in order for the United States to maintain its position as the global leader in wireless communications," remarked Rep. Greg Walden (R – Ore.), the outgoing ranking member of the House Energy and Commerce Committee. "By offering nearly free equipment, Huawei and ZTE have made it difficult for trusted vendors to compete in the marketplace. As a result, upgrading equipment and software has become an unnecessarily expensive… process."
Rep. Walden's recorded comments came at the opening of a virtual panel conducted Tuesday by the Open RAN Policy Coalition. You may recall, ZDNet introduced you to this organization last May. It's a group of major telecommunications stakeholders — most notably including 5G leaders Nokia and Ericsson — who advocate the type of interoperability and collaboration that O-RAN is working for, although there is a thick layer of abstraction between "O-RAN" and "Open RAN." One is a technology; the other is a rallying cry.
Ostensibly, the O-RAN Alliance — which deals strictly with architectures and interface specifications — tries to stay out of geopolitics. Indeed, two of its founding members are AT&T and China Mobile, whose initial collaboration on the concept of moving RAN functionality off of towers and onto cloud platforms, is said to have been the catalyst for 5G.
But Walden is co-sponsoring a bill introduced last April that would set up a government-managed grant program for the express purpose of promoting and supporting the use of interoperable standards in wireless networks. The bill mentions O-RAN specifically, while leaving the door open for any other alliance that may happen to come along. During Tuesday's meeting, Walden and others specified the grant fund's value as $750 million. (Back in January, before the pandemic, the number being bandied about was $1 billion).
Rep. Doris Matsui (D – Calif.) describes the bill as a means to restore supply chain diversity to the 5G market — diversity she believes has been directly threatened by Huawei, which she describes as spreading like some sort of virus.
"A more robust supply chain will also produce gains for national security," remarked Rep. Matsui. "Strong state support has allowed [Huawei] to undercut competitors, and integrate its equipment throughout the US, Canada, Europe, and emerging markets. The prevalence of this equipment is a significant risk for data security and critical infrastructure."
US Government support could become a lifeline for organizations such as the O-RAN Alliance, which may need greater participation from smaller companies, upstart vendors, and research institutions, if they are to cement O-RAN's position as a global standard. But to ensure that support is forthcoming, the Open RAN Policy Coalition finds itself courting a benefactor that would leverage the emerging interoperability framework as an exclusionary weapon against China, in the country's newly rebooted Cold War.
With respect to the United Kingdom's decision to reverse its stance on Huawei, ordering a purge of that company's equipment from its networks by 2027, Doug Brake, director of broadband and spectrum policy for the Information Technology and Innovation Foundation — told the Open RAN Policy Coalition panel he believed there may actually be an economic benefit, at least in the longer term, for operators forced to make the purge, if O-RAN or something like it enters the picture.
"There's a real challenge in trying to get interoperability between existing Huawei 4G equipment and new 5G deployments," said Brake. He continued:
So for any country, particularly in Eastern Europe, that has widespread existing deployments of Huawei, there's a huge economic cost to try to transition away from that in a sort of hard-cut, unless we're able to successfully open up those interfaces in a way that has real ease of interoperability. I think that will continue to be a major driver in different countries' decisions, in which direction they go. And that's why, in my mind, insisting on a security threat, and trying to strong-arm countries into coming along with the US in foregoing Chinese equipment. . . it would be much better if we were to go this route and try to promote these open interfaces, and try to find a way to have interoperability between existing 4G networks and new 5G equipment, in a way that doesn't have quite as much of a performance impact as it does today.
Huawei's technological arguments against an open, modular RAN are already on record. It claims that an open system would, by nature, have too many components. Even with open interfaces, the complexity of their interconnections would work against the objectives of high performance and reliable security. But Nokia's objections to an all-out embrace of O-RAN is also a matter of record, with company representatives warning Congress against mandating O-RAN as a national requirement as recently as last March.
If Nokia's position can pivot so quickly, even if just part-way, then why couldn't Huawei's? What's to stop Huawei from obtaining a license for the same open standard everyone else is entitled to? ZDNet asked the Open RAN industry panel on Tuesday.
"I would honestly try to encourage a global marketplace," responded John Baker, senior vice president of business development for mobile network software maker Mavenir. Baker continued:
As soon as we try and put barriers up, effectively, between countries and companies, essentially, to lose what was set out to be. . . I was lucky enough to be here at the beginning of GSM. It was all about generating a global marketplace. Clearly, we've done that, and we shouldn't necessarily try and tear it apart. I think all we're trying to do here is, open interfaces and widen the supply chain again, to the extent that there's fair competition on the marketplace. I'd argue that opening these interfaces would put competition in China as well. Let's look at it the other way around. Let's encourage more companies in China to compete with Huawei and ZTE. It's really about how to broaden that ecosystem.
"I think the other side of it is, if you're talking an international standard," responded Chris Boyer, AT&T's vice president for global security and technology policy, to ZDNet's question, "by definition, that means it's international. So there's no reason to exclude any one particular entity. And there actually are Chinese entities looking at the dimensions of some of these technologies. It's a question of, though, if it's standardized and open, then it creates an environment where multiple entities can play in that space. That's the key. You're not trying to foreclose someone's opportunity, but as an operator, it gives us a lot more options."
Openness as a weapon
In any market whose principal revenue stream is intellectual property, every player has to be extremely careful about how it uses "openness" as a weapon for competitive advantage. In the mid-2000s, prior to the consolidation that left us with Amazon AWS, Microsoft Azure, and Google Cloud Platform as the "Big Three" cloud service providers ("AMG," as some call them), the key players in computing all tried to use their existing position in conventional computing to stake some kind of pre-existing claim to the cloud market.
All the major players tried some form of "open" policy play. In 2007, IBM allied itself with Google to form the "Blue Cloud Initiative." As the cloud platform OpenStack evolved from the early work of NASA's Project Nebula and the company now known as Rackspace, IBM staked a further claim to openness by promoting an "Open Cloud Manifesto." Amazon and Microsoft refused to sign on.
In his 2009 objection to IBM's move, then-Azure Product Manager Steven Martin wrote, "We love the concept. We strongly support an open, collaborative discussion with customers, analysts and other vendors regarding the direction and principles of cloud computing. . . It appears to us that one company, or just a few companies, would prefer to control the evolution of cloud computing, as opposed to reaching a consensus across key stakeholders (including cloud users) through an 'open' process."
The fact that most, if not all, of the various consortia and industry coalitions rallying around the causes of openness, fairness, and consumer choice in the cloud computing market, are about as well-remembered today as medieval folk songs, stands as testament to the fact that they served their purpose for their respective memberships, and were then discarded.
"Openness" typically has two purposes in information technology:
- As an effort to disrupt a market dominated by a monopolistic player, by consolidating multiple, small bargaining positions into a single, bigger one;
- As a way to funnel more customers into a business model or intellectual property, by popularizing or expanding the reach of a technology that's dependent on it.
Neither of these two methods is necessarily deceptive. Red Hat earns its revenue from service, not software, and that's by design. OpenStack expanded the cloud by eliminating the market value of competitive technologies designed expressly to maintain their own exclusivity.
Open hand, clenched fist
O-RAN has had a legitimately "open" purpose: to ensure that no single vendor in the telecom space can build a system whose components are so exclusively dependent upon one another that they disable the introduction of competitive alternatives. But in a market whose consolidative forces have been so great that eight telecom giants became three in just the last decade, it's easier for one of those three to become the one left out. With Ericsson having signed on at the beginning and Huawei declining early on, O-RAN was looking like Ericsson's very own, exclusive customer-generating machine.
Which is why the response of Brian Hendricks, Vice President of Policy and Public Affairs for Nokia Americas — the US division of the current holder of the Bell Laboratories' patents for telephony itself — to our question on leaving the O-RAN gateway open for other players such as Huawei, was so very telling, on a number of levels.
"That's another reason why I think Nokia's participation is important," responded Hendricks. "Although there's a great deal of marketing around our Chinese friends, we have more than our share of intellectual property as well."
Nokia, Huawei, and Ericsson are all very substantial holders of intellectual property (IP). It's IP that gives these companies bargaining rights when engineering standards. Companies that can't come to the bargaining table with IP for table stakes, may consolidate their power through some kind of "open" coalition. The O-RAN Alliance gives its members the hope of some degree of accessibility to the technologies they need to conduct their marketplace, without having to face licensing fees they would have no way to reasonably afford.
What the future of 5G will come down to, explained Nokia's Hendricks, is the continuations of these tenuous cross-licensing arrangements for the use of IP protected by so-called essential patents. To the extent that these essential patents are implicated in the engineering of 5G, he said, the cross-licensing arrangements will be the key to the industry's future. His suggestion — despite everything Reps. Walden and Matsui said at the opening — was that if that arrangement falls apart, Nokia may find itself in an indefensible position. Hendricks' answer here is less that of an engineer, and more that of an attorney.
"It will be difficult — not impossible to imagine, but difficult, I think — for the Chinese vendors to pull their IP, which is one thing that I've heard expressed," Hendricks stated, "because they rely on those cross-licenses from us, and from others as well. So I think that danger, to the extent that the question is being asked, is mitigated to some degree by the support of other vendors who have vast patent portfolios. And then. . . there's a great amount of proprietary patents that are held by the other companies who are looking to jump into this space as well.
"So I'm not as worried about an escalating patent conflict being an impediment here," he added. "I think there's too much to be lost on all sides, but certainly from the Chinese side."
With respect to the issue of whether Huawei would, or should, come around to supporting O-RAN, Hendricks very surprisingly remarked, "I think our question has come back to, do they even need to? Because I don't think we know the answer to that question." Hendricks continued:
I think it's a pretty safe bet, after spending the money and the time that China has in cultivating domestic champions, it's not about to just lay down. So if they need to embrace Open RAN technology in order to remain competitive, I think it's fair to say they'll probably do that. But again, I don't want to ignore the very real, non-domestic side of this equation, which is that, they may not have to, because they'll still be able to offer extremely attractive packages to people — heavy subsidized rates, commercial dumping, deferred payment terms, and things that will make their classic stack extremely attractive. We have to be cognizant that simply opening interfaces is not a magic bullet, in terms of scale, and being able to provide a check on Chinese dominance. I think too often in the O-RAN discussion, we are ignoring that very real threat: that there needs to be far more tools than just the R&D support and the interface specing [developing specifications]."
The Nokia VP's answer actually went two steps beyond the question. Clearly, Hendricks understood that O-RAN could be a gateway to enable greater numbers of customers to utilize IP protected by essential patents. To the extent that it excludes two essential patent holders in a three-player market, Nokia could not afford to be excluded.
Regardless of how Reps. Walden and Matsui opened Tuesday's panel, the executives of the telecom stakeholders (plus one Washington think tank representative) found themselves inexorably drawn toward the opposite conclusion: Openness only works openly, in the absence of clandestine motives. Perhaps Huawei can afford not to participate, as Hendricks asserted. But when openness is applied properly, it forces the exclusivity player to withdraw unto itself, to waste its energy on self-sustenance — to do what Linux did to Windows Server in the enterprise computing space. The same efforts that openness would achieve through coalition and cooperation, for the withdrawn and heavily fortified market player, become uphill battles in downhill territory.
There's a proverb in there somewhere. Let's see whether Nokia or Huawei writes it first.