Home » Internet, Wrong On The Internet

More misguided representation of the Internet standards

By 13 September 2010 One Comment

Last week we called on Free Press to stop misrepresenting the Internet standards in their efforts to push for Internet regulations.  This week we need to point out more misrepresentations of those Internet standards.  ArsTechnica’s Matthew Lasar pondered whether the founders of the Internet anticipated paid prioritization in this Wired article, but the article confuses some key points and lacks sufficient research.  The issue at hand is over the established technical standards of the Internet and not what “the founders” thought, and the comments that Lasar cited from the Center for Democracy and Technology (CDT) were inaccurate and do not offer sufficient insight on the relevant standards.

Lasar claims that the DiffServ standards merely authorizes “system administrators” to implement DiffServ on the “intranet” and questioned whether this really meant that the standards were “about generating money for Internet Services”.  This arbitrarily narrow interpretation of the word “administrators” is misguided, and the standards aren’t just limited to an intra-company intranet.  The relevant standards documents like RFC 2474 and RFC 2638 use the word “administrator” in the context of an “administrative domain”.  Domains encompass the entire company or organization and not just a single employee working as a system administrator.  Even if those documents only referred to the system administrator, those employees don’t operate in a vacuum without regard for their organization.

The Internet standards clearly accomodates differentiated pricing

The CDT argues that a single phrase in the DiffServ standards about “differentiated pricing” doesn’t justify payment schemes, but there are several places in the DiffServ standards that make it clear that they fully contemplated payment schemes.  In fact we can point to four different explicit examples.

RFC 2744 – “Differentiated services enhancements to the Internet protocol are intended to enable scalable service discrimination in the Internet”

RFC 2475 – “permit differentiated pricing of Internet service”

RFC 2638 – “It is expected that premium traffic would be allocated a small percentage of the total network capacity, but that it would be priced much higher.”

RFC 3246 – “To protect itself against denial of service attacks, the edge of a Differentiated Services (DS) domain SHOULD strictly police all Expedited Forwarding (EF) marked packets to a rate negotiated with the adjacent upstream domain.  Packets in excess of the negotiated rate SHOULD be dropped.”

From the examples above, it is clear that the Internet standards contemplated higher priced priority service and offers a standardized method to enable it.  The Internet standards don’t set specific pricing and it would be highly inappropriate (and illegal) if it did, but it states the obvious that the market will likely price premium priority traffic a lot higher than normal traffic and it offers a mechanism for enabling and enforcing it.

Russ Housley keeps confusing the issues

Much of this confusion started two weeks ago when the group “Internet Society” paid a public relations firm to plant the story in the media that the IETF chairman Russ Housley discredited AT&T.  Housley quickly wrote an email to the IETF that he was not speaking for the IETF or IETF consensus and reiterated this point in a story published at Broadcasting & Cable by John Eggerton.  In a notably different tone, Housley admits that AT&T made many correct points but he continued to jab at AT&T that they have “jumbled some things together” and that they give readers a “distorted impression”.

Russ Housley continues to assert that “The IETF does not make statements about prices for network services.”  But if we read the four examples above, that’s precisely what the IETF standards do.  They don’t set specific price levels, but they do acknowledge that there will obviously be higher prices for higher quality of service.  So Russ Housley is the one giving all the distorted impressions of the Internet standards.  While it isn’t a knock on Russ Housley that his expertise is in the field of security and not network engineering, it is important to point out that he is speaking outside of his expertise and that his assertions clearly conflict with the IETF DiffServ standards.

Not only is Russ Housley presenting a distorted impression of the Internet standards, he is giving a distorted impression of the policy debate.  Housley asserts that everyone within the same tier should be treated equally but that different tiers can be treated differently, yet differentiated tiers is precisely what Net Neutrality regulation advocates oppose.  Housley has created a red herring that critics of Internet regulation are advocating different prices within the same service tier when they are advocating different prices for different service tiers.  The reality is that it is the advocates of Net Neutrality regulation that want to outlaw differentiated tiers, and that position is inconsistent with the history and the standard of the Internet.

One Comment »