Can we drop the religious zealotry in Net Neutrality?
After reading this summary of the Net Neutrality debate from David Post, I’ve decided to write the following open letter to him. [Update - David Post has graciously responded in the comment section]
David, you sound like a very reasonable and intelligent person who can be reasoned with. But since you’re admitting to some religious zealotry when it comes to the supposed “architecture of the Internet”, can I ask you to set that zealotry aside for a moment and debate the facts surrounding the actual history of the Internet?
You stated:
“I’m a believer — indeed, I’m a religious zealot — when it comes to e2e design; if you want to know the details, read my book. E2E means, simply, that the network itself does virtually nothing besides transporting bits from one place to another, leaving all processing to take place at the “ends.” E2E has “non-discrimination” built into it; the network can’t discriminate among the different packets of data (to put those originating from Bill Gates’ machine, or from Verizon premium subscribers, or from Google users, at the head of the transport queue, for instance) because to do that it would have to analyze the packet (to figure out where it came from), and that violates the E2E principle.”
The E2E arguments never called for a restrictions against intelligence in the core (I have to credit Richard Bennett for pointing this out to me). The End-to-end principle was always meant to be a pragmatic engineering philosophy for network system design that merely prefers putting intelligence towards the end points. It does not forbid intelligence in the network itself if it makes more practical sense to put certain intelligence in the network rather than the end-points. David D. Clark along with Marjory S. Blumenthal wrote in 2001 in “Rethinking the design of the Internet: The end to end arguments vs. the brave new world”:
“from the beginning, the end to end arguments revolved around requirements that could be implemented correctly at the end-points; if implementation inside the network is the only way to accomplish the requirement, then an end to end argument isn’t appropriate in the first place.”
RFC 1958 in the section “Architectural Principles of the Internet” points out that:
“In searching for Internet architectural principles, we must remember that technical change is continuous in the information technology industry. The Internet reflects this. . . .In this environment, some architectural principles inevitably change. Principles that seemed inviolable a few years ago are deprecated tomorrow. The principle of constant change is perhaps the only principle of the Internet that should survive indefinitely.”
Furthermore, DiffServ is defined in the Internet Standard RFC 2475 and it states:
“This document defines an architecture for implementing scalable service differentiation in the Internet. A “Service” defines some significant characteristics of packet transmission in one direction across a set of one or more paths within a network. These characteristics may be specified in quantitative or statistical terms of throughput, delay, jitter, and/or loss, or may otherwise be specified in terms of some relative priority of access to network resources. Service differentiation is desired to accommodate heterogeneous application requirements and user expectations, and to permit differentiated pricing of Internet service.“
Now as for your assertion that the Internet can’t handle QoS in real-time, that argument hasn’t been an issue for a decade. Cisco has even testified to the FCC that they can’t sell a router that doesn’t have QoS capability. Furthermore, the lone paper that everyone cites as evidence of why the Internet can never implement QoS is widely misunderstood due in large part that the paper’s conclusion is inconsistent with its own content. I detailed both of these points here.
I’d love to hear your response to these criticisms. I’ve written my own analysis of the Google-Verizon proposal here and I welcome any comments you may have on that as well.
[Note to readers: George sent me these comments via email, and I responded as follows:]
George:
First off, thanks for these comments – I am a pretty reasonable guy, and am always happy to have my errors or over-simplifications corrected . . . I suspect that law profs do about as good a job understanding engineering principles as engineers do in understanding legal principles ;). I also think that I understand the point George is making (and I am familiar with the Clark/Blumenthal paper you cite), that the E2E principle can perhaps better be stated not in absolute terms (such as those a religious zealot might use), but with qualification: “push applications to the edge and away from the network, unless there are really good efficiency benefits to be gained from not doing so.” But my attitude about that is a bit (actually, a lot) like my attitude towards the principle of free speech. “No government restrictions of any kind on speech” — now, I know that too is a bit oversimplified (crying “fire!’ in the crowded theater, conspiring to commit murder, etc.), and that a “better” articulation of the principle might be “No government restrictions of any kind on speech, except where they’re justified — in legal jargon, where there’s a “compelling governmental interest,” and the restriction is the ‘least restrictive alternative’ available to the government to further that interest.”
I really do get it that the more nuanced version is the better of the two. But I stick, in both cases, with the absolutist version, because I think the principle (in both cases) is so important and I want the burden of justification to be a very, very difficult one to meet. If nobody articulates the absolutist position, what happens over time is that there’s ‘creep’ — you get lots and lots of violations of the principle, because it becomes a balancing test, weighing costs versus benefits, and there are always lots of good reasons to think that in THIS case the benefits will outweigh the costs, and before you know it there are lots and lots of speech restrictions in place and the basic principle has all but disappeared. So it’s a bit of a rhetorical pose that I’m taking, I admit – as long as there’s a significant portion of the debate taking an absolutist position, the violations will hopefully be few and far between —
David
The thing about end-to-end is that it was extremely pragmatic. The principle of pushing things out to the edges to ensure minimal overhead (they used the word “cost” which is tech jargon that may be lost on non-engineers) and maximum versatility isn’t quite as sacred as protecting free speech. At the end of the day, we’re talking products and services and that’s ultimately what the Internet is. It’s certainly a revolutionary commercial service but it’s not this scared “public network”. It’s a network of private networks that benefits most from the network effect which is to include as many players as possible so it naturally wants to be inclusionary rather than exclusionary. However, it’s always had a commercial price tag and it’s always had multiple tiers.
The whole Net Neutrality thing gets distorted by extremists like Free Press into an endangerment of free speech which can’t be further than the truth. If the NN advocates aren’t claiming that ISPs are trying to collude with large websites to shut smaller competing websites down which is all but officially illegal because we know any ISP that tries it is suicidal politically and PR wise, they’ll resort to the argument that all websites are supposed to be equal which we know is ludicrous.
If it isn’t that, they’ll claim that large Hollywood hates Net Neutrality because they want to see independent film makers shut out from the Internet because they can’t afford to be on the Internet. Well guess what? It turns out that independent film makers really can’t afford to host their own content today, but that’s totally irrelevant because large sites like Google and Vimeo give bandwidth away for free because they desire all the free independent content. In fact they want it so much that Google will even share revenue with independent film makers even though Google still loses money on YouTube. As for the vast majority of content on the Internet which is not video, the cost of bandwidth is so cheap that it’s almost irrelevant. $50 a month for server colocation service allows Digital Society to reach over 10 million page hits a month but our challenge is attracting even a fraction of those page views in the first place.
So all these worries about the arrival of the tiered Internet are just silly because we’ve always had a tiered Internet and this equal outcome Internet they’re trying to preserve is based on fantasy. Only the largest companies like Google and Facebook can afford to buy and/or lease their own network and CDN infrastructure. Google and Free Press will claim that CDN is a different kind of “geographic prioritization” that is harmless to the competitors vying for the network, but my data shows that CDN is much more harmful than router prioritization technology and that CDNs produce similar route prioritization like DiffServ.
The smaller startups have plenty of opportunity to demonstrate their stuff on a smaller scale e.g., Flipboard and the VCs will go nuts throwing money at them. The barrier to innovation is getting the smart people and engineers to write your software and develop your hardware. The cost of the initial bandwidth needed to demonstrate viability is cheaper than their electric bill and substantially cheaper than their servers and human resources.
Lastly, One of the reason I get so worked about people talking about principles of the Internet is that the Internet works on a implement and deploy first and then standardize later. This is why you see so many commercial finished products with “draft” standards like 802.11n. You can implement any new standard you like though the Internet community is free to reject or accept your proposal (usually with changes requested by the community). The Internet does not work on some sacred principle/architecture handed down by God and disseminated to the masses through great prophets like Lessig, Wu, and Crawford.
The Internet was unique in its freedom to experiment. It was free from government restrictions as well as architectural restrictions that people wrongly perceive to be the end-to-end “principle”. It is one of the reasons that the IETF RFC model beat out its competitors. You can basically implement anything you thought was a great idea and then try to formalize it through an RFC and hope to gain support by attracting support from other companies. Any unworthy or outright bad ideas simply never get adopted.
Anyhow, I’ve ranted long enough and I’ve posted my comments on the Google-Verizon proposal.
Why does the end to end principle, however we define it, have anything to do with Internet policy? It’s never been the aim, as far as I can tell, for any sort of regulatory system to preserve the intellectual purity of any particular technology. So what if the historical Internet was based on the end to end arguments, among other things? The policy job is to stimulate it toward certain public ends and to either permit or forbid it from being used in particular ways that are independent of its internal organization.
The Internet should not be used as an instrument for identity theft, copyright infringement, or terrorism, regardless of how it was put together.
The argument against placing functions in the lower levels of the system that aren’t used by all higher levels – a latter day extreme addition to the original statement of E2E – is based on a nonsense guess that the circuitry necessary to provide such functions would impair performance, which it demonstrably doesn’t in most cases. The E2E arguments are more a guideline than a principle in any case, and a weak guideline at that because they make predictions that aren’t borne out by the facts.
So what relevance does E2E have for the ongoing policy debates over the Internet, given that E2E is false to the extent that it can be verified and outside the scope of regulation otherwise.
Once upon a time, the Model T Ford followed the principle of the single color: any color you want as long as it’s black. Cars now come in a variety of colors. Should car regulators in the 20s have stepped into the market and prevented this deviation from the purity of the Model T color model?
Good points from Richard above, as well as in his piece on prioritization in hightechforum:
http://www.hightechforum.org/a-question-of-priorities/
[...] [...]
[...] [...]
Leave your response!
Twitter Feed
About Us
Digital Society is a digital think tank that believes culture and commerce are inseparable, that the digital economy flourishes when people are free and rights are secure, and that free markets free people.
Digital Society is an independent 501(c)3 non-profit organization, funded by donations from Jon Henke and from Arts+Labs. We advocate for a pro-culture, pro-commerce digital society through research, analysis and debate on emerging technology issues.
Reply Comments
Transparency and interactivity are trademarks of the Internet era, and we aim to foster them here at Digital Society. It is inevitable that some people will disagree with the technology policy positions we take. We want to have that constructive debate.
The Reply Comments feature gives our critics a chance to respond to our viewpoints and the Digital Society audience convenient access to competing arguments. Any time we directly challenge the views of an individual or a group on this site, the party in question may substantively respond in a guest post.
Please contact executive director Jon Henke by e-mail.
Subscribe
Daily Digest Email
Recent Posts
Recent Posts
Most Commented
Most Viewed