[94840] in North American Network Operators' Group
Re: death of the net predicted by deloitte -- film at 11
daemon@ATHENA.MIT.EDU (Frank Coluccio)
Sun Feb 11 21:42:39 2007
From: Frank Coluccio <frank@dticonsulting.com>
To: nanog@merit.edu, "Geo." <geoincidents@nls.net>
Reply-To: frank@dticonsulting.com
Date: Sun, 11 Feb 2007 20:41:42 -0600
Errors-To: owner-nanog@merit.edu
I believe that the element that has been missing in this discussion thus fa=
r has
been the source (content) players, and where they are hiding. CDNs, a la Ak=
amai,
Limelight, etc., will take up some of the slack and mitigate much of the ba=
ckbone
burden where legitimate ISPs are concerned, as will hierarchical caching fo=
r the
newbie carriers-that-came-to-be-called "ISPs" -i.e., the MSOs and Telcos. P=
laying
the Pareto, the higher the demand (95/5) for a title, the closer it will be
stored to the user community, and the longer the tail (5/95) of a title, the
farther its storage from the user community. My point is, CDNs and hierarch=
ical
cache must be inserted into the calculus, because one, they are already bei=
ng
used, and two, their use will only increase with time, fwiw.
Frank=20
ps - I've had some issues with my email editor of late. If anyone notices a=
ny
artifacts or extraneous characters in the delivery of this message, kindly =
email
me off list and I shall be indebted to you, tia.=20
On Sun Feb 11 19:22 , "Geo." sent:
>
>
>> do what google is presumably doing (lots of fiber), or would they put
>> some capital and preorder into IDMR?
>
>IDMR is great if you're a broadcaster or a backbone, but how does it help=
=20
>the last 2 miles, the phoneco ATM network or the ISP network where you hav=
e=20
>10k different users watching 10k different channels? I'm not sure if it=20
>would help with a multinode replication network like what google is probab=
ly=20
>up to either (which explains why they want dedicated bandwidth, internode=
=20
>replication solves the backup problems as well).
>
>Also forgetting that bandwidth issue for a moment, where is the draw that=
=20
>makes IPTV better than cable or satellite? I mean come on guys, if the=20
>world had started out with IPTV live broadcasts over the internet and then=
=20
>someone developed cable, satellite, or over the air broadcasting, any of=
=20
>those would have been considered an improvement. IPTV needs something the=
=20
>others don't have and a simple advantage is that of an archive instead of=
=20
>broadcast medium. The model has to be different from the broadcast model o=
r=20
>it's never going to fly.
>
>TIVO type setup with a massive archive of every show so you can not only=
=20
>watch this weeks episode but you can tivo download any show from the last =
6=20
>years worth of your favorite series is one heck of a draw over cable or=20
>satellite and might be enough to motivate the public to move to a differen=
t=20
>service. A better tivo than tivo. As for making money, just stick a=20
>commercial on the front of every download. How many movies are claimed=20
>downloaded on the fileshare networks every week?
>
>Geo.=20
>
