Why Network Neutrality Will Take a Beating

Sign in to queue


A couple of months ago, Jeremy Geelan at SYS-CON asked me for my predictions about the tech industry in 2008.  They just published my predictions in their end-of-year issue, highlighting the prediction that "Network Neutrality Will Take an Even Worse Beating in 2008".  You can see my other predictions on their site, but I'd like to go into more detail about network neutrality here.  After reading this post, you should have a clear picture of how network neutrality affects you, and how Microsoft and others in the industry think about network neutrality and the upcoming 700MHz spectrum auctions.

What is Network Neutrality?

Any time you read a blog post, send a twitter, or check your e-mail, you're depending on two very different types of businesses:

  • Software or content businesses write the services or create content for you to enjoy.  A few random examples include Wordpress, Yahoo! Finance, XBox Live, and Twitter.  Since software and content can easily be created and copied, these businesses need to be creative about how they protect against competition.  New competitors can pop up any time.  Example strategies for protection include copyright and patents, hiding the software behind a service or inside hardware, or establishing moats based on profile data.
  • Bandwidth providers enable you to access to the services and content.  Verizon and Comcast are examples.  Bandwidth is a scarce physical good similar to real estate, limited by basic laws of physics.  New competitors cannot create bandwidth the way they can create software or content.  If you want to connect from a certain place, you need to connect through the person who owns the bandwidth.  Obviously, making profit from a physically scarce good is very different that making profit from software or content.

You are just a serf on the bandwidth provider's land.  Every time you read a web page, you are using a physical good which they own.  Every time you put up a new web site for others to enjoy, you're relying on the bandwidth provider's largess.  Of course, the bandwidth providers wouldn't make much money without cool services and people to use them, but the point is that it's their bandwidth -- not yours, not Microsoft's, and not Google's.

Now, when you own a piece of property, you want to control how it's used.  For example, you might happily let your neighbor plant flowers in your yard, but you might charge him for the right to grow a vegetable garden, and you'd just say "no" if he asked to raise pigs in your yard.  Likewise, the bandwidth providers want the kind of traffic that's the most convenient and profitable for them -- and they want to exclude or charge a premium for traffic that is less convenient.

Since companies like Microsoft, Google, and Yahoo make a living on the bandwidth providers' land, we depend on them being as "neutral" as possible toward us.  We ask two primary things of bandwidth owners:

  • Do not discriminate against data traffic based on the source, application, or company.  For example, if Comcast developed a proprietary e-mail system, and then started charging triple for all web-based e-mail traffic, that would be bad for Hotmail.  If a backbone provider in China found it profitable to redirect all Google search traffic crossing their network to Baidu, that would be bad for Google.
  • Offer access as uniformly and universally as possible, and don't exclude people in rural areas.  Bandwidth is infrastructure service, like mail or electricity.  American history would have been rather different if people in rural areas had to pay more to receive mail, or if the government had not subsidized deployment of telephone and electric transmission to rural areas.

Microsoft and Google are pretty much on the same page regarding network neutrality.  So, besides pleading and cajoling, what are Microsoft and Google doing about network neutrality?  To answer that question, you need to understand the upcoming 700MHz wireless spectrum auction.

The 700MHz Auction

When the FCC recently announced that it would be auctioning off a huge chunk of 700MHz spectrum, people were excited.  This is probably the last big auction of bandwidth, beginning in a couple of weeks and shortly after.  People became even more excited when Google announced intentions to bid on the spectrum.  In fact, Microsoft has been lobbying Congress to open up this spectrum, specifically because 700MHz can be used to provide broadband in rural areas.

Microsoft's interest in the spectrum, followed by Google's interest, has led many to speculate that our companies want to become bandwidth owners.  While Microsoft's motives were less ambiguous, many are still convinced that Google intends to win some serious spectrum in the auction.  Only in my wildest dreams would Google actually bid high enough to win, and then be saddled with a business they know nothing about.  Not long after I sent my predictions to SYS-CON, Om Malik got it right, explaining that Google doesn't actually intend to win in this auction.

Google's bid was essentially a PR stunt engineered to get the FCC to impose neutrality constraints on whoever wins the auction (likely Verizon will be a big winner).  The stunt worked, sort-of.  First, the FCC agreed to some of the constraints.  Then, Verizon announced that they would pledge to adhere to some neutrality principles on their own.

So, if the 700MHz auction bodes well for rural broadband, and if Google succeeded in driving some modest pledges of neutrality in the 700MHz auction, why do I say that things aren't getting better?

Trending Away from Neutrality

It's true that things could have been worse, but the gap between bandwidth haves and have-nots will only get worse from now, and discrimination based on traffic type will only increase.  Note that this analysis is U.S.-centric, but there is some applicability to international as well.

For starters, this is the last big auction, and thus the last opportunity for the FCC to intervene in this way.  And the federal government seems to have less interest in pushing rural bandwidth than they had in rural electricity and telephone.  Distributing ultra-fast fiber in densely-populated urban areas is far cheaper than deploying fiber to rural areas, so companies like Verizon are being very selective about where they deploy this capability.  Today, if you are a banker in Manhattan or a computer engineer in Seattle, you can get high-speed FiOS at a price that would bankrupt a poor Mississippi farmer -- but if the fiber ever makes it out to the Mississippi farmer, it will cost him a LOT more than it costs you.

And even within urban areas, the penalty for being poor is high.  Let's say that you pay $120 per month for the "unlimited" data plan on your cell phone.  You'll probably use 1GB of data for that $120.  In contrast, people paying per-SMS message are paying somewhere between $500 and $2000 per 1MB of data transfer.  This is more than 5,000x the rate that you pay for your mobile data.  The same sort of disparity will emerge as fiber is selectively deployed to people who can afford it.

In addition to the growing gap between haves and have-nots, the bandwidth owners are becoming more bold about hijacking services like search, and many bandwidth owners are already engaging in tricks to slow down people who use protocols like bittorrent.  The bandwidth owners have no responsibility to tell you if they are doing this, and the techniques are designed to be pretty much undetectable.  Your downloads just run slower or crash frequently, and you eventually get frustrated and do something else.

The bandwidth providers argue that such "traffic shaping" is necessary for the continued survival of the Internet, and seem to have convinced at least one "cute" reporter at the Economist.  But it's difficult to see what the content and service providers can do about it anyway.  There is a limited amount of bandwidth available, and the moment that people watching mobile video on their iPhones (for 12 cents per megabyte) start to compete for traffic with SMS (which makes $500 per megabyte), the iPhone video is going to suddenly get really unreliable.  People who use large amounts of bandwidth to download movies, while paying the same amount as the guy next door who uses 1/10th the bandwidth, will have to get used to an unreliable connection or else upgrade to FiOS.

Other than an occasional PR stunt or congressional hearing, I don't see any major changes on the horizon; so we can expect things to continue on the current trend for at least the next year.

The Discussion

Add Your 2 Cents