[1895] in Commercialization & Privatization of the Internet
T-3 Eggs in the Right Basket
daemon@ATHENA.MIT.EDU (Jordan Becker)
Mon Jan 6 15:48:34 1992
Date: Mon, 6 Jan 92 15:44:44 EST
From: Jordan Becker <becker@ans.net>
To: cook@tmn.com
Cc: com-priv@psi.com
> <<MESSAGE from>> Gordon Cook 03-JAN-92 21:04
> cook@tmn
> Let me retarget my question. IBM and ANS say that they are making great
> strides into the future of Networking by putting what will likely be 2
> years work into DS-3 technology - a technology that some technically
> knowledgable people say is "dead end." Why the effort on what appears to
> be an extension of the T-1 technology when again some knowledgable people
> seem to think the network could have survived on T-1 for a few more years?
The network could very well "survive" for a few more years through focused
re-engineering and improvements to the T1 network infrastructure. But is
"survival" what the future of networking is about? I thought it was about the
creation of new applications and services that broaden its impact on society
and hopefully improve peoples lives.
The technical work that led to the current and future shape of the T3 network
began in 1988. At that time it seemed like a very realistic and fruitful next
step to complement the T1 network technologies that have also improved since
1988. I would not call these technologies "dead-end" since they continue to
evolve and serve end-users at all levels of the bandwidth hierarchy (300
baud-> T3).
> Of the three SMDS, Frame Relay or ATM lets pick ATM as likely to be the
> most significant long term? What was the reasoning that dictated the T-3
> now strategy rather than the ATM 2 years from now strategy? And while
> this question COULD be asked about the decision made a year ago, does it
> make any sense to ask now about decisons being made now?
As for the three technologies you mention, these technologies dont replace T3
(or lower speed) leased line services. Among other things, they allow
dedicated tariffed circuits to be shared among different virtual private
closed user groups. Don't mis-interpret me, they are very important
technologies that many service providers are working with including ANS (we
just announced two T1 SMDS trials with NYNEX and PacBell). However the right
question in this context may be T3 versus SONET OC-3, OC12, etc.
Jordan
