China's Proposed Internet Wrests Control From Users
The new template could split the Internet under the guise of innovation.
First of a multipart series.
The next-generation Internet proposed by Huawei and supported by the Chinese government would provide a platform for revolutionary capabilities while implementing repressive measures that would eliminate today’s open communication. At worst, it would place control of Internet content in the hands of a few masters. But even if it does not subsume the entire Internet, it would cripple the interoperability that has characterized the network’s value as an economic growth engine by creating separate and unequal Internets.
China is pushing its new Internet architecture both through market penetration by its own industries and by influence among international standards bodies. Ultimately, the only way to prevent this methodical takeover would be for governments to provide financial and logistical support to their own companies while those companies improve efficiencies to compete with China’s telecommunications firms globally.
Vint Cerf, an Internet pioneer who is vice president and chief Internet evangelist at Google, notes that China’s own implementation of the Internet—“The Great Chinese Firewall”—indicates a high degree of sophistication in the ability to filter content. It has shown no reluctance to clamp down on social media viewpoints that run counter to the government’s positions.
This sophistication reflects the dramatic investment China has made in this Internet implementation, and it has supported the growth of large companies such as Alibaba and WeChat that are comparable to Western counterparts. At the top of the investment pyramid are companies that make communications and networking products, such as ZTE and Huawei. These companies are the tip of the spear in China’s thrust to dominate global networking and communications, and they are slotted to play a large role in seizing the digital high ground.
Cerf notes that their products are favorably priced and have carved out a significant market share, especially in Africa and elsewhere in the developing world. China’s M.O. is to loan money to countries for purchase of products and services. This has great ramifications when it comes to telecommunications networks, he allows, as loans can be called for lagging payments with China taking ownership of telecommunications infrastructures. It’s consistent with the plan behind the Belt and Road Initiative (BRI), in which China lends money for projects that it profits from either immediately or over the long term.
“The Chinese are playing a very long game here,” Cerf states.
China aims to introduce a new Internet design, which it calls the New IP, for Internet protocol. The New IP would represent a radical departure from the status quo.
Cerf relates that when a Huawei representative introduced it as an Internet Research Task Force (IRTF) development proposal, the IRTF evaluated it and stated that some aspects were not in its scope of responsibility, and it only would consider those aspects that were in its area. The Huawei representative said it was an “all or nothing” proposal and promptly went to the International Telecommunications Union (ITU) and the European Telecommunications Standards Institute (ETSI) to push for support.
“The Chinese seem to be launching a significant effort to alter the course of the development of the Internet,” Cerf says. “Where they can’t do it by those overt things, they also are populating many of the standards groups, including those that are part of the IRTF to bring as much suasion as possible to the Chinese point of view.”
The Internet’s architecture has remained largely unchanged over the past 45 years, even while protocols have been added to all layers to advance its capabilities, he adds. “I am unpersuaded that the desirable properties that the Chinese argue for are not possible to be implemented within the framework of the evolving Internet,” he declares, drawing from his pioneering experience with the Internet. “We have retrofitted a lot, and we’re doing what we’re doing now because the Internet’s capacity has increased mostly because of the speeds of optical fiber and routers that go with it have been able to keep up with the demand.” Cerf has watched this progress, sometimes with surprise at the way technology has kept up with new capabilities offered in the market.
“We don’t need to invent something brand new,” he allows. “What we need to do is ask, ‘What are the requirements, and can they be retrofitted and met within the general Internet framework?’”
Ted Hardie, self-described Internet boffin at Google, concurs. “You [incorporate new advances] in a way that remains interoperable with the existing system in order to make sure that what you’re doing is growing the system rather than creating a fork,” he states. “Once you start forking a system, the user base goes off into the two different forks. They have to be managed separately, and you end up with higher costs if you’re trying to service both of them. And, in a lot of cases, the interoperability after that point becomes extraordinarily fragile, and it only works in a limited number of cases for a limited amount of time.”
A gateway would be needed to transition between the two forks, and this indicates the deployment pattern, Hardie says. First will be New IP in China, followed by the gateway system that links the two. Eventually, all the traffic that flows through those gateways will need to undergo a translation process.
Hardie points out that this gateway takes the place of a firewall system or other method of monitoring. It has access to all the traffic that is translated between the two systems, thus adding “a very competent point of control,” he warns. “It’s important for us to recognize that this consequence is something that would be a result of the success of such an approach.”
Cerf believes that the chance of a fork is more likely as Chinese technologies become more ubiquitous around the world. “If you can get enough equipment in the field that is compatible with everything else, and then you send out a whole new software load that does something different, you could succeed in injecting the new capability without having to even fight with the standards community,” he points out. “We’d still end up with instant non-interoperability with the systems that haven’t ingested the software and its new functionality.
“This is the worst possible outcome,” he continues. “Ending up with a non-interoperable system where you have to build translating gateways wrecks end-to-end security, exposes everything that is going on to the translating gateway, and so it destroys the global interoperability of the system we have today.”
Cerf is blunt about the consequences of this approach. “If you fork, you’re forked,” he declares.
In Part II: How this fork would work, and what can be done to avoid it.