Internet traffic exchange: not many people understand it fully (even those who make the deals). And when they say they do, they often diverge on fairly fundamental points? Perhaps the OECD can help. I.D. Scales reports.
Perhaps not so coincidentally, the OECD, like the Internet Society, has also ridden to the intellectual rescue in the run-up to the WCIT in December (see - The Internet works just fine... hands off!). Its report (“Internet Traffic Exchange: Market Developments and Policy Challenges” ) is not so much about Internet technology as about the Internet's commercial models which the OECD seeks to defend and advance, as it has been doing for the last 20 years or so.
The horror of a telco-style "sender network pays" environment (see -
US delegation prepares to play Whack-a-Mole at ITU meet) is not so much over the funds that might be extracted, as it is about the detailed billing and accounting requirement that would flow - turning a brilliantly simple stream of "handshake" agreements into the complex bureaucratic nightmare that we all know and love in the telco environment. This recently posted
TelecomTV interview on TEM (telecoms expenses management) provides an inadvertent glimpse of the world into which network peering would plunge.
Meanwhile, back on the Internet, the OECD points out that "a survey of 142 000 peering agreements conducted for this report shows that the terms and conditions of the Internet interconnection model are so generally agreed upon that 99.5% of interconnection agreements are concluded without a written contract."
The OECD goes on to assemble all the arguments for the Internet model.
There's a welter of statistics and similes designed to make the reader marvel at the Internet's dazzling growth in terms of traffic and costs. Here's the one I think is the most stunning: "Today, twenty households with average broadband usage generate as much traffic as the entire Internet carried in 1995." It might have added that, even in 1995 with its piddling little capacity and traffic total, the Internet was under attack for being unsustainable and about to grind to a halt.
The OECD points out that one really magical, self-balancing aspect of the Internet has been the way the system shrugs off attempts by players to extract profits above and beyond a competitive return based on the value they actually provide. When so-called tier 1 players attempted a bit of a takeover a few years back the result was simply that backbone traffic moved off to other networks, including content distribution networks (CDNs). More on this in the report.
The problem is that even today, a full 20 years after the invention of the World Wide Web, there is a widespread inability in our particular part of the broad IT and communications market to understand the nature of the settlement free peering and the fact that the Internet hasn't prospered despite this feature but because of it.
That inability has culminated in this year's ETNO proposal for a new IRT (International Telecom Regulation) to overturn or supplement (the proposals are very unclear on detail) settlement-free peering with telco-style 'sending network pays'.
So why does the drum need to be banged on a regular basis by the likes of the OECD if the evidence for the Internet model is just so overwhelming?
My pet theory is that it's so highly abstract a topic that it's difficult to compare it to anything else in the commercial world without spending more time explaining the validity of the metaphor than the illumination it might, or might not, shine on the issue.. if only you could remember what the issue was. Perhaps the OECD paper can help.
please sign in to rate this article
49531
