Google’s Internet “Fast Lane”

by on December 15, 2008 · 17 comments

The Wall Street Journal reports today that Google wants a fast lane on the Internet and has claimed that the Mountain View based giant may be moving away from its stance on network neutrality:

Google’s proposed arrangement with network providers, internally called OpenEdge, would place Google servers directly within the network of the service providers, according to documents reviewed by the Journal.

The problem with the Journal piece is that OpenEdge isn’t exactly a neutrality violation, or maybe it is.  As Declan McCullagh at CNET has pointed out in his post “Google accused of turning its back on Net neutrality,” figuring out when a neutrality violation has occurred is a little tricky:

The problem with defining Net neutrality so the government can regulate it is a little like the problem of defining obscenity so the government can ban it: You know it when you see it.

Well, Google says that it knows a neutrality violation when it sees one and not surprisingly it doesn’t see one in its own actions.  It’s defense essentially boils down to them pointing out that OpenEdge is caching.   It’s more of a warehouse than a fast lane.  Besides, anyone else can do the same thing, so Google isn’t using any ISP’s “unilateral control over consumers’ broadband connections” to their advantage.

Interestingly, however, the same Google’s Policy Blog entry defends other companies that engage in the same sort of caching, including LimeLight.  But LimeLight Networks isn’t just a data warehousing company,  they combine caching with real fast lanes.

LimeLight was the platform used to distribute video for the 2008 Beijing Olympics.  This architecture involves ISP-based servers similar to Google’s, but also involved piping in video directly to those ISP-based facilities via LimeLight’s proprietary fiber network that spans the globe.

LimeLight’s sales materials point this out:

Our dedicated high-speed network spans the globe, interconnecting all of our regional data centers over a fiber-optic backbone. This allows Limelight Networks to deliver the most massive of files. And even the most obscure objects in your library can blast their way to your users at CDN speeds — just as fast as your most popular objects.

That’s a genuine, very non-neutral fast lane and it allowed me to watch the opening of the games from my laptop.

We should be welcoming this kind of innovation, not shunning it.

Hey, even Google itself has constructed a series of fast lanes for routing its internal data around.  In fact, it’s the first non-ISP to own part of an undersea cable so that Japanese servers can stay current with their US counterparts.  This isn’t web traffic, so it’s also viewed as neutrality compliant, but it still gives Google an advantage, one based on Google’s own unilateral control of its own network.

But we should want companies to compete over providing us not only content, but content with an efficient delivery mechanism.  So I welcome these types of experiments in content delivery.  I don’t know that they’ll be widely adopted, or even needed if broadband speeds continue to climb.  But, in the interest of determining the right distribution method through experimentation, I say we allow these companies to press on.

Though the Journal article was technically incorrect on this one, I was heartened to hear that many companies (Microsoft, Amazon, Yahoo!) are softening their stance on neutrality.  Even Larry Lessig seems to think that neutrality doesn’t always make sense, the Journal quotes Lessig saying:

“If everyone had to pay the same rates for postal service, than you wouldn’t be able to differentiate between sending a greeting card to your grandma versus sending an overnight letter to your lawyer.”

This softening of the stance of the pro-neutrality side is heartening, but so long as someone stands to gain from having neutrality regulations on the books, I’m sure we’ll be hearing more about the virtues of this networking philosophy for years to come.  After all, where there’s a corporate interest, there’s the political will for a new law.

Previous post:

Next post: