[11511] in Commercialization & Privatization of the Internet
Re: The whole CIX concept is flawed
daemon@ATHENA.MIT.EDU (Dick St.Peters)
Mon Apr 4 23:42:17 1994
Date: Mon, 4 Apr 94 18:18:35 EDT
From: stpeters@bird.crd.ge.com (Dick St.Peters)
To: sean@dsl.pitt.edu
Cc: com-priv@psi.com
Reply-To: <stpeters@dawn.crd.ge.com>
>From: Sean McLinden <sean@dsl.pitt.edu>
>There is a huge digital data network out there that has very low startup
>costs, no membership, and a large volume. All you need is $6 bucks a
>month for dialtone and you've got FAX; store a bunch of phone numbers in
>the thing and you've got routing. Replace FAXes with FAX modems,
>computers, and OCR and you have FAX-email gateways. Still no membership
>fee. Still at dialtone + usage-based pricing (and if it is in the
>metropolitan network, it is flat rate).
Sean, this is a good description of just the kind of thing I want not
to happen: creation of disjoint, non-interoperable networking because
it's all people can afford.
Permit me an analogy, made with full understanding that analogies are
only an aid in explanation, not a valid basis for argument. When the
telephone was invented, business was used to doing business by
telegraph. Some early business customers rented one phone and put it
in their old telegraph room. Their vision was that employees would
carry messages to the "telephone room", where the "telephone operator"
would deliver the message to the receiving site's operator. The
telephone's advantage was seen as simply that it didn't require an
expensive skilled operator.
About all that's left of this applying the telegraph model to the
telephone is the term "telephone operator." What the telephone really
meant was a whole new paradigm of doing business, one in which
communication is no longer between firms but penetrates into them and
permeates them, having become communication among individuals within
different firms.
This was an unsettling concept when it first appeared. Management did
not like the notion of unsupervised, unapproved, and even unrecorded
communication directly between their employees and those of other
companies. However, it was such a tremendous lubricant of business
that the new paradigm swept the business world anyway.
I think we're going to see another similar change, one in which the
communication extends beyond the people to the machinery that runs
businesses. The systems at a company will be in touch with those at
the company's suppliers, at its customers, at its bank, at the company
that made them, at the company that services them.
Yes, this kind of integration is a long ways off, but the first pushes
in that direction have begun. I've previously mentioned that GE's new
line of scanners use tcp/ip for connections among subsystems, so they
could in principle be distributed around the Internet. Larry Walker at
GE Medical is taking the first steps toward implementing some of the
potent service concepts that this enables.
Nobody knows what new business and other implications such distributed
systems can have. Perhaps large organizations buying and leasing big
systems will partly give way to small groups or individuals buying and
leasing out subsystems. Your medical scanner or factory control system
or business information system might be a fluid, dynamic entity
assembled from component pieces obtained as needed from around the
network.
Or perhaps not. But something, many things, will be different that we
cannot foresee. I think the fundamental difference will be people and
equipment at one place communicating directly with, even operating,
equipment elsewhere. This needs real packet-passing networking.
--
Dick St.Peters, Gatekeeper, The Pearly Gateway; currently at:
GE Corporate R&D, Schenectady, NY stpeters@dawn.crd.ge.com
