Thursday, February 16, 2006

What Cisco and the Telcos haven't accounted for


Photo
Another compelling "deal" from BellWest

Last time out, I discussed the kinds of wares hardware vendors like Cisco are hawking to cable companies and telcos. The vendors' new hardware whets the service providers' appetite for more draconian controls over content providers and consumers alike. Certainly, the telcos' top executives have served up inflammatory rhetoric of late, which only serves to confirm their desire to pile-drive the Googles of the world into submission while turning consumers' pockets inside out every time they connect.

Not a business model that makes much sense, but about what you would expect from a bunch of classic, unreformed monopolists.

Cisco's new hardware allows the service providers to limit consumer activities, throttle performance of a rival content provider's service, inspect users' packets, and perform other fairly invasive operations on traffic. Our hypothetical RBOC -- call it BellWest -- could, for instance, slow down search engine performance for Google and Yahoo while letting its own search service perform lickety-split.

That's certainly fair for them to try. Fair, but incredibly short-sighted.

The possibility that the carriers are considering these services implies a business case for implementing them. We won't be seeing that business case anytime soon, but I'll bet you a case of Guinness it's fatally flawed. The recent WSJ article on China's Internet censorship helps point out why:

1) Tor and related anonymous routing software packages prevent deep-packet inspection and disable any determination of the true source and destination IP addresses. Tor can be distributed to an infinite number of machines, making controlling a Tor network next to impossible.

2) SSL VPN technologies allow tunneling of IP traffic through SSL (and, last time I checked, no one was routinely cracking TLS).

3) OpenVPN and related technologies allow tunneling any IP traffic through UDP using strong encryption.

These are what I would consider "first-generation" anonymizing technologies that are quickly becoming mainstream. Any combination and/or improvement of these technologies would make them even more formidable. They fundamentally resist the kinds of controls that the telcos want to slap on consumers and content providers.

What to consider some possibilities? Google distributes a version of Firefox that incorporates automatic use of an SSL VPN through its enormous -- and growing -- server farm. Yahoo distributes a version of Tor integrated with its desktop search offering. And so on.

Any business case that fails to consider the impact of these "carrier avoidance technologies" is rotten to the core.

Here's a novel concept for the carriers: try competing at layers 4 through 7, where there's value to be added. Yes, that's right. I said, "value-added." A foreign concept for the telcos, to be sure, and therefore one certain to be ignored.

No comments: