[787] in Commercialization & Privatization of the Internet
Re: IETF questions -- Internet growth
daemon@ATHENA.MIT.EDU (G.Huston@aarnet.edu.au)
Thu May 30 19:43:45 1991
Date: Fri, 31 May 91 09:42:07 +1000
To: com-priv@psi.com
From: G.Huston@aarnet.edu.au
Cc: ietf@venera.isi.edu
Lyle Seaman writes on the issue of internet growth:...
>In fact, most of the devices that aren't currently networkable that
>may become networked are probably not going to be running TCP/IP.
>It's my belief that the only systems which will be running TCP/IP
>are the same sort of things that are now running TCP/IP, with the
>exception that a lot of things on people's desks that use RS232
>may use TCP/IP. But then, a lot of them will be running Netware,
>too. Face it, there are going to be other protocols in use; not
>every person needs a private IP address.
I think this is glossing over a few basic issues:...
There is still the reasonable prediction that the number of systems and networks using TCP/IP _will_ grow - both in the PC area and with larger system platforms. Even techniquies of dynamic address assignment in the gateways to PC networks do not provide medium term stable answers to the diminishing unallocated Internet address pool. Sure there are/will be heaps of other protocols in use in the world, but TCP/IP has not and will not stagnate in terms of deployment as a consequence of having such a (wonderfully) diverse technology environment.
The underlying issue is that there IS a looming problem with the Internet address space, and we do need to address this issue both in terms of prompting engineering responses/development in the Internet architecture and, in the interim, putting in place appropriate network number allocation/registration practices which go some way to acknowledge the fact that the resource being allocated is right now a VISIBLY FINITE one.
--
Geoff Huston G.Huston@aarnet.edu.au
Network Technical Manager ph: +61 6 249 3385
Australian Academic and Research Network fax: +61 6 247 3425
