Law in the Internet Society
Current mass-market communications technology is very insecure. User adoption rates for encryption of emails, instant messages, and phone calls are for all intents nil. Even the use of ad-free, cookie-free, untraceable internet browsing is on the order of 5%, despite its extreme ease. Legal protection for the privacy of such communications is scarcely any better. For example, the NSA wiretapped and obtained incriminating evidence about a high-ranking Democratic congresswoman who was perhaps uncoincidentally in line for the chairmanship of the House Intelligence Committee. Authority for roaming, warrantless wiretaps has been claimed by presidents from both political parties. The dominant usage of technology in America has destroyed the expectation of privacy that people used to enjoy effortlessly. Perhaps we can convince people to take matters into their own hands.

I initially proposed the PATRIOT phone, envisioning this. It'd feature RSA cryptography with no back doors or key escrow. With the right marketing strategy, a satisfactory product, and reasonable price points, it shouldn't be difficult to sell several thousands of such devices. At that point, the device becomes an impediment to large scale surveillance. If even 1% of network traffic were encrypted, it would be impossible to capture that volume of traffic for cryptanalysis -- there simply aren't enough supercomputers.

Such consumer technology does not exist. Many phones that employ RSA or similarly strong encryption, but they are not interoperable and cost $1500 or more. This is puzzling. Crytographic algorithms, qua algorithms, are not patentable subject matter and to whatever extent such monopolization was possible (i.e. patenting cryptographic math done "on a computer" or "a microchip that performs cryptographic math"), those patents should have long since been expired. See 1, 2. This academic paper suggests a means for implementing RSA into programmable logic chips that cost less than $10, and provides sufficient computing power to encrypt voice data in real time. In fact, that paper is five years old: the technology should now be trivial.

The US cellular handset market is highly concentrated. The top three companies account for two thirds of the market, and the top six account for almost 90%. Part of the problem seems to be that US consumers buy handsets bundled with phone service, whereas in most countries the two products are separate. Nokia, for example, has 40% of the global handset market but only about 8% of the US market, purportedly because they refuse to 'play ball' with the large telecommunications companies. The cellphone market is similarly concentrated -- 4 companies control 81.2% of the market. High concentration indicates barriers to entry (i.e., large fixed capital costs). Thus, a would-be wireless player must build its own network or play nicely with a party that already has one. In other countries, an innovator might be able to rely upon the essential facilities doctrine of antitrust law, or open network access provisions, but not here.

Recent telecommunication innovators like Skype avoid the cellular network altogether. Skype users may make encrypted calls (only to other Skype users) using 256 bit AES encryption, which, while not impenetrable, would certainly thwart all but highly motivated eavesdroppers and would prevent wholesale data mining. However, Skype itself is not open source, and it is widely believed that some parties (governments, at a minimum) are given "backdoor" snoop access. A number of open source VoIP suites exist, but none of them seem to feature PC-to-landline/cellular calls without an additional account with some form of interconnection service provider that charges a few pennies per call.

Well, indeed. You have to get to that phone using the closed network to which the phone is attached, and that network provider's interconnection charge is inescapable. But because switching services and wholesale bandwidth are extremely competitive in the global market, the price charged by the VoIP? supplier will be only infinitesimally greater than the monopoly interconnection fee at the other end. So even the consumer, using Skype's service to reach landlines and cellphones, gets within three significant digits the same price that the smartest business VoIP? purchaser can get. A free software user running Asterisk in her laptop and having an account with Junction Networks or some other well-run VoIP? supplier would get a global best price to within four significant digits on every call.

Incumbent telecommunications carriers are unwilling to allow access to their network for free, unlike the rest of the internet. To the extent that they allow Skype and Google Voice to exist, they may view them as interesting experiments in marketing and distribution, much in the manner that they experiment with giving customers "night and weekend" minutes or a "circle of five" or other such nonsensical demand metering masquerading as bandwith pricing.

No, they pay their way as interconnecting foreign telephone companies do. Skype passes that cost along.

I would be surprised if Ebay viewed Skype as long-term strategic fit--perhaps one of the four companies that control 82% of the phone network might be interested in buying it in the future. The only real solution is to abandon the closed networks of the telecommunications monopolists entirely. This would require adoption of an addressing system more like instant messenger than a phone directory, but if the reward were free or near-free phone calls, people might be convinced. Cell phones would fuse with tablet PCs, and access the internet over municipal wifi.

The use of open networks to transmit voice data would make encryption nearly mandatory, since otherwise anyone situated between two callers could simply sniff packets and eavesdrop entire conversations. Due to open network architecture, there are a lot of people so situated, whereas on closed phone company networks, the only person between you and the person you are calling is Ma Bell. People are not generally convinced of the need conceal their conversations from her, which is just as well, because as long as you're on her network, the only way to do so is to buy $1500 phones, and to insist that your banks, doctors, and phone sex operators all use them.

Because of open-source VOIP suites and other free software, these problems don't really exist for a small, educated, and technologically proficient subset of society. When those people speak to each other, they do so with a degree of privacy that has not existed in at least a century. Everyone else has to pay $50 to $75 a month for a portable phone featuring less privacy by the year. It is no solution to condemn those people as stupid or lazy for their consumer choices. Marketing need not be exclusively for the purpose of making people want to buy things that they don't need. Social attitudes about communication are malleable, and if a physical device can be constructed to wean people off of the incumbent telecommunications monopolists, it would offer a value proposition that people of all levels of sophistication would appreciate.

Agreed. Hence my proposal for the Freedom Box, which would put a privacy-engineered telecommunications server inside everyone's house for a price less than half that of the world's cheapest cellphone..


Webs Webs

r10 - 07 Sep 2011 - 00:44:10 - IanSullivan
This site is powered by the TWiki collaboration platform.
All material on this collaboration platform is the property of the contributing authors.
All material marked as authored by Eben Moglen is available under the license terms CC-BY-SA version 4.
Syndicate this site RSSATOM