[Air-l] A definition of the internet

Kevin Guidry krguidry at gmail.com
Tue Oct 17 07:51:53 PDT 2006

On 10/17/06, Michael Maranda <mmaranda at afcn.org> wrote:
> Yes, in my brief phrase "Internet as an agreement(s)"  I think we lead
> towards an understanding of this ... it's an agreement around TCP/IP ...
> that TCP/IP be a protocol over which we can define additional protocols, and
> around peering agreements by which traffic will be carried through networks
> held by others.

   Is TCP/IP really a core, defining feature of the Internet?
Wouldn't it still be "the Internet" if we could somehow replace TCP/IP
with something functionally equivalent?  I have a difficult time
making the current networking protocols selected out of the middle of
the networking stack a defining feature of what I view as a merely a
means of linking disparate networks. Couldn't we rightfully say that a
device on the other side of a bridge linking a non-IP network to the
Internet is just as much on the Internet as a device on an IP network?
   I certainly don't deny (a) the major historical role that TCP/IP
has played and continues to play in the development and operation of
the Internet and (b) the dominance of TCP/IP in the majority of
computer networks.  But it just seems to me that TCP/IP as the
underlying protocol of the majority of the Internet is a historical
accident.  Other protocols that provided the same functionality could
just as easily been used.  An Internet with some other
functionally-equivalent protocol would still be the Internet just as
my LAN at home is still essentially the same despite moving from
Ethernet over Cat5 to 802.11g over the air.  It also seems to me that
the slow-moving-but-supposedly-imminent move to IPv6 provides some
support for my argument since we are replacing one of the core
protocols used on the Internet with another protocol but it's still
going to be the Internet.
   I am also uncomfortable conflating the effects and uses of the
Internet with its definition.  Your initial thoughts related to
"agreements" seems to me to be the closest to what I would call a
"good" definition.


More information about the Air-l mailing list