<<< Date Index >>>     <<< Thread Index >>>

[IP] more on Worth reading Wall Street Journal on fragmentation of the Internet





Begin forwarded message:

From: Karl Auerbach <karl@xxxxxxxxxxxx>
Date: January 19, 2006 4:44:05 PM EST
To: David Farber <dave@xxxxxxxxxx>
Subject: Re: [IP] more on Worth reading Wall Street Journal on fragmentation of the Internet
Reply-To: Karl Auerbach <karl@xxxxxxxxxxxx>


On Thu, 19 Jan 2006, David Farber wrote:

From: Fred Baker <fred@xxxxxxxxx>

What would concern me more is balkanization at the network layer.

Fred's right - fragmentation at the IP layer would have rather greater consequences than the existance of competing, but essentially consistent, systems of DNS roots.

Two points: As for DNS the issue is that of *consistency* not of *singularity*. The US government seems not to comprehend this simple distinction.

First, We know that we can do name-to-phone-number lookups using any number of phone books from different providers, CD-ROM databases, websites, PDA's, etc. The issue for competing DNS roots is exactly the same - what matters is consistency of the answers not the source of the answers.

That leaves the question of what is "consistency".

Some believe that consistency requires that there be only one set of top level domains that are provided identically by all systems of roots.

Some of us believe that consistency is to be measured not by an absolute degree of perfect equivalency but rather by there being a core of names and TLDs that are perfectly identical surrounded by an additional set of boutique TLDs that are not seen by every internet user via every root. These boutique TLDs are the newcomers that are aspiring for greater visibility and greater market share.

In other words the real difference of thought is between centrally allocated TLDs versus a more distributed entrapreanural approach to TLDs.

Two: The potential for IP address/IP-layer fragmention is very real. The use of NATs has already made IP address space re-use and translation a routine fact of life for many, if not most, residential and small business users of the net. A super-NAT, one that could map the traffic load of a small country, is not inconceivable.

We are already having a head-to-head collision between NAT unfriendly protocols, such as SIP+RTP/RTCP (VOIP), and uniformity of address space. As those issues are worked out there will be less and less reason for countries to carve off their own IPv4 /8's and NAT their way to the outside. And, from the point of view of a repressive government, having a NAT-ed country makes it harder for dissident voices to communicate without going through the government-approved filters.

        --karl--









-------------------------------------
You are subscribed as roessler@xxxxxxxxxxxxxxxxxx
To manage your subscription, go to
 http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/