Would anyone care to comment on this post which I placed in the Gorilla thread?
Message 10293096 :
-------
To: +CRUSADER4TRUTH (3222 )
From: +Frank A. Coluccio
Sunday, Jun 27 1999 9:20PM ET
Reply # of 3225
Hi Byron, Glad to see that you are not asking too much of me, this time. All you want me
to do is identify the next paradigm shift?
I take it you are asking me what I think the next shift will be in telecomms? Correct?
It'll manifest, IMO, in the utility derived through the manipulation and routing of
wavelengths and subdivisions thereof (lets call them wavelets, for now), in combo with
millimeter wireless systems and intelligent optical nodes (and, let's call those IONs, for
ease of recall). How's that for a mouthful?
The optical component will bring massive amounts of information to the general vicinity,
and the wireless component will bring selective information to the device. At the node
nexus between these elements there will be a photonic switching matrix located in the
ION to solve who gets what, and who is admissible on the return. Whose code will be
used? I don't know. It probably wont be CSCO's IOS, though, although it might resemble
it in some future macro way. If anything, unless CSCO is willing to forego control on the
embedded tens of billions it will have in place throughout all networked spaces, by then,
they might actually be loathed to suggest such a model, early on.
It's one thing to be untethered at 9.6k through a couple of Mb/s, under ideal, static
conditions. It's something else, altogether, at a gigabit or higher, to be wired anywhere,
anytime.
The optical accomplishments will be very much akin to the parameters which were set
forth by George Gilder's fiber sphere article, in 1992. In '92 those views were deemed to
be twenty or more years off into the future, by most measures of the day. In '96 those
same views were suddenly only ten years off. Today, well, we're already seeing the
beginnings of its manifestation and an evolving deployment curve, in some crude ways.
But all of the focus has been on the higher capacity flows thus far.
Who will be the company to perfect, and cost-effectively produce the
dynamically-variable, narrow line-width optical devices, capable of switching and routing
mini-wavelengths or wavelets the way we now switch and route 100 Mb/s Ethernet
frames and OC-48 IP packet flows, respectively? That's the question I'd like someone to
answer for me, while we're on the subject.
Thus far, all of the to-do has been about massively parallel OC-192 flows aggregated on
DWDM systems, seeing how far, how many wavelengths could be supported, and that is
not merely fine, but it's an imperative, a "must do" in the evolution of optical development.
But at some point we have to start looking the other way, to the point where individual
devices can harness optical, as well, say, at the level of I/O competence of every
networked user device in an enterprise environment, and similarly through wireless, both
on the road, and in the residence. Variations to each of these settings apply as well.
That will probably mean dealing with lower denominations of optical delivery through
some period, at least, lest we remain bound to the artificial 100 meter limitations imposed
by Category 5 premises cabling. If you can identify this company that can harness and
economically mass produce narrow line width optical networking devices, then you have a next generation gorilla on your hands. Anyone?
Regards, Frank Coluccio |
