Update 9/2/2010 - Hank Hultquist posted another follow up on this that is very interesting.

Hank Hultquist of AT&T wrote this interesting post summarizing AT&T’s letter to the FCC defending the right of “end users” to buy premium networking services i.e., Paid Prioritization.  Hultquist made a number of key points:

• Free Press made the claim that paid prioritization has never been contemplated by the standards bodies, doesn’t exist on the Internet and that it is to be used for nefarious purposes, and will only be affordable to the larges companies.
• But the Internet’s standard body created the DiffServ standard that clearly contemplated and permitted fee based premium services.  These premium services are widely available and purchased by businesses and organizations of all sizes.
• Free Press makes the strange assertion that DiffServ prioritization is harmful to regular Internet traffic while “geographic prioritization” i.e., Content Distribution Network (CDN) prioritization isn’t.
• However, empirical evidence (based on research conducted by Digital Society) shows just the opposite is true and that CDNs are more harmful to regular Internet traffic than higher class DiffServ traffic.

Derek Turner and Free Press fired back against AT&T calling their letter misleading.  However, they only addressed some of the issues raised by AT&T with some questionable arguments.

• Free Press argues that AT&T is conflating the sale of enhanced QoS to business customers with “the anti-consumer practice of speeding up and slowing down Internet traffic according to which service provider pays more”.

I hate to break the bad news to Free Press and Derek Turner, but selling enhanced QoS (Quality of Service) to businesses is precisely the act of speeding up and slowing down Internet traffic according to who pays more.  It’s the concept of “getting what you pay for” or “pay more get more”.  It’s the concept of “equal service for equal pay” which Free Press distorts into “equal service regardless of how much one pays”.  If businesses didn’t get faster, more reliable, and less congested Internet services, why in the world would they pay 10 to 40 times more per Mbps of bandwidth than consumers?  Isn’t that the epitome of “priority service”?

• Derek Turner then makes the argument that AT&T increased its capital expenditures in 2006-2007 which was a time when AT&T was obligated to live under the terms of a merger condition with the FCC in which it refrained from selling paid prioritization to content/application/service providers.

There are serious problems with this line of reasoning which I’ve pointed out in the past.  Those merger conditions were nowhere near as extreme as the ones proposed by congressman Ed Markey, didn’t outlaw existing business models, and didn’t apply to wireless networks by pretending that wireless networks operated under the same technical and economic constraints as wired networks.

Furthermore, content/application/service providers are merely another type of business.  More specifically, these are businesses that offer products (applications or content) and services to consumers over the Internet and these businesses are commonly referred to as business to consumer (B2C) sites.  Why is it that a business selling to other businesses should be allowed to buy priority routing but business that deals with consumers shouldn’t?  And if they are prohibited from buying priority routing, why should they be allowed to buy priority by leasing or building more network infrastructure or buy CDN services?  Our research already showed that Free Press’ argument that router prioritization is harmful but CDNs are not is completely backwards.

The fact that companies like Linden Labs wants to outlaw their competitors like Blizzard from paying for priority because they don’t want to look inferior doesn’t make outlawing priority routing services good public policy.  Just because one economist at NYU believes that services which enable winners and losers should be banned doesn’t mean that this is the accepted legal and economic definition of what constitutes a bad form of discrimination.

This debate over paid prioritization has always been the essence of the fight over Net Neutrality.  To have a meaningful debate, we must ignore the bogus arguments that distort the meaning of free speech or the silly “all sites load equally” arguments.  We must debate whether we want to preserve an open bandwidth market or if we want to outlaw cheaper and faster paid peering services that may one day threaten the dominance of companies that built their own private priority Internet like Google.