[101812] in North American Network Operators' Group
RE: An Attempt at Economically Rational Pricing: Time Warner Trial
daemon@ATHENA.MIT.EDU (Tomas L. Byrnes)
Fri Jan 18 18:06:53 2008
Date: Fri, 18 Jan 2008 14:39:21 -0800
In-Reply-To: <D03E4899F2FB3D4C8464E8C76B3B68B001C87295@E03MVC4-UKBR.domain1.systemhost.net>
From: "Tomas L. Byrnes" <tomb@byrneit.net>
To: <michael.dillon@bt.com>, <nanog@nanog.org>
Errors-To: owner-nanog@merit.edu
there are already companies like Vyatta that represent the nascent part
of this space, at least on the software/equipment side.
=20
> -----Original Message-----
> From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu] On=20
> Behalf Of michael.dillon@bt.com
> Sent: Friday, January 18, 2008 1:17 PM
> To: nanog@nanog.org
> Subject: RE: An Attempt at Economically Rational Pricing:=20
> Time Warner Trial
>=20
>=20
> > The problem in the ISP industry isn't lack of usage based pricing.=20
> > It's that the going rate for basic connectivity was driven=20
> below that=20
> > which is economically sustainable by the ILECs when they engaged in=20
> > predatory pricing to drive the CLECs out of business in the=20
> late 90s.=20
> > Now that they own the market, they find that, having driven=20
> the prices=20
> > down, they can't raise them, so they are engaging in various=20
> > subterfuges that are designed to cover up the basic thing they are=20
> > doing:
> > trying to charge more for the exact same service.
>=20
> Sooner or later, somebody is going to try to apply Google's=20
> approach to hardware in a network backbone. Imagine a network=20
> backbone with no Cisco or Juniper boxes in it, just lots of=20
> commodity boxes with triple-redundancy everywhere (quintuple=20
> in NFL cities).=20
>=20
> Vadim Antonov tried to build something like this into a=20
> backbone router, but the market for IP backbone equipment is=20
> so incredibly conservative, and the pricing was up there with=20
> the big boys, so he never had a chance at it.
>=20
> I don't know if Google is doing something like this between=20
> their data centers, but I think that the fundamental price of=20
> fiber is low enough that with commodity router/switches and=20
> triple the fiber miles, we can have a reliable IP packet=20
> moving service without jacking prices up.
>=20
> Even if prices do go up, it will be a short term thing=20
> because sooner or later, Google, or somebody who thinks as=20
> bold as they do, will build a true commodity packet-moving=20
> service, and the telecoms industry will fall back into the=20
> razor-thin margin utility sector where it belongs.
>=20
> I'm sure many of you will think I am crazy because you know=20
> just how much those high-speed ports cost and you can't see=20
> any letup in bandwidth growth. But the fact is that ports are=20
> not the fundamental components of routers. Chips are, and as=20
> we all know, chips keep getting smaller, cheaper, faster and=20
> more powerful. FPGAs, SOCs, multicore CPUs and so on. The=20
> company that cracks the Internet utility problem might even=20
> design and build their own devices rather than outsourcing=20
> that, at a high price, to the benevolent vendors.
>=20
> --Michael Dillon
>=20
