Jim, I'm glad you brought up the part about your being a carpenter. It was on my mind
earlier, but I didn't think anyone would be interested. The fact is, however, that
mounting the older variants of infrared transceivers used to be a bear. Precision of
orientation and long term stability was paramount, and freeness from instantaneous
vibrations was equally crucial. Each time that I recall systems being installed they
required expert carpentry (especially in well appointed offices), and/or the coordinated
mounting of two inch pipes which served as masts into floor slabs in order to ensure
that these qualities were met.
To make matters more interesting, when these are mounted in open office spaces they
often require some form of baffle in order to hide them from open view, or to prevent
them from clashing with the decor.
One other piece of trivia before we get to your questions, is the amount of speculation
that these things can cause if they are not explained adequately at the time they are
installed:
Pregnant women will immediately question the effects of these devices on their yet
unborn children, and jokes about sterility are bound to proliferate, as well as some
more serious concerns on the parts of employees about eye safety, just to give you an
idea of some of the potential fun that awaits the HR folks in early adopter firms. All of
these examples are legitimate experiences I've had, by the way. No kidding around
here.
----------
The amount of hops from click to central node shouldn't matter theoretically, since the
potential exists for each hop to be sped along at the speed of light, literally. But these
enhancements will not matter here, any more or less than over other forms of media, if
the new i-r provider doesn't adequately size their pipes and processing capacity to the
larger net.
They can have all the bandwidth in the universe available to a given building, and if they
don't have a commensurate scale of bandwidth pointing to the core, it will be useless
for all situations except those where the other end point of the connection is also on-net
to the IR cloud.
Even here it would depend on how good of a job they did when provisioning their own
central node plumbing: their switch pipe and port sizings / router processors /
switch-blocking factors / buffering / caching/ etc.
The patterns (topologies) that they will have to choose from in big city settings may
more often than not be chaotic in nature, rather than being coherent ones, except for
those nice round circles we see on paper indicating where their node sites, or cells as
they call them, will be located. And yes, they will resort to in-building and
building-to-building repeater links, reflectors, and even cutting deals with landlords and
adjacent tenants to allow fiber runs between different parcels of real estate in order to
get their prospect du jour onto the "good side" of the building.
------
Do the RF folks have a lot to worry about? Here's a message that I wrote on the
Gildertech Forum earlier today that addresses some of these concerns.
=======
posted: Sunday, March 12, 2000 03:05:55 PM --
author: Frank C <Fcoluccio@dticonsulting.com>
subject:GG: Demise? Or, a lifting of all boats?
Hello George and All,
This is in reply to both EJB's earlier question concerning the demise
of existing technology plays, and to your assessment of what the rise
of Terabeam would mean in this same context, per this month's GTR [report].
I've just replied to a similar question which was posted earlier today
concerning what the effects would be to MFNX if Terabeam were to
be successful. The same reply will hold true here, and that is that all
boats will rise above the banks of the river if and when this free
space, or any other last mile/last 1000 ft. technology, is universally
successful to the levels that have been implied.
I understand your enthusiasm here, but I believe that you may have
taken this a bit too far when you begin to predict the demise of other
established technologies at this time. Speaking as a networkologist, it
would be very narrow and somewhat reckless for me and my
colleagues to follow any blanket assumption that all Terabeam users
would be conversing ONLY with other Terabeam users. Especially
during the next several years, prior to their having time to penetrate
widely, assuming, for the sake of discussion, that they will. Things just
don't work that way:
For every yottabit of incremental traffic that is added to overall Internet
traffic -- enabled, say, only by the cheaper bandwidth afforded by the
Terabeam system -- there will be some corresponding increase in
terms of terabits of capacity added to backup --or load sharing
systems-- which might not have happened at all in the absence of a
Terabeam primary route, or some similarly abundant source of
bandwidth which might eventually be enabled by a wdm arrangement
over dark fiber, when it becomes more accessible to more users.
Here, the determination of which will be a primary or contingency
route will inevitably wind up being a judgment call, based on the
criticality of the applications being supported, and other factors
having to do with the specific technological attributes of any proposed
solution, such as distance, unobstructed views and weather issues, in
this case.
If the type of traffic activity in question can withstand falling back to
lower levels of capacity periodically, based on criticality assessments
--which will usually mean falling back to some more expensive route
per unit of capacity, albeit at lower capacity, when the primary-- then
the Terabeam system would be the primary, and fixed wireless or
fiber would be the secondary. But it would call for an appropriate level
of risk analysis to sort this out for any given user's situation. Given
how businesses value their communications links now, I'd suspect
that in many cases the infrared solution will prove in as a backup
solution, more often than as a primary. For residential? Maybe an
even toss at first, but I suspect primary over time, where it is available
for entertainment and casual surfer needs, for the cost conscious and
bandwidth starved if they can get it.
Getting back to the prospect of business/e-commerce users
abandoning alternative means, conversely, it is likely that until TB is
truly ubiquitous (assuming, again, that this would ever be the case),
most of their users would be conversing with web sites and other
private users who are "NOT" on Terbium's system. What do you
suppose these other end users will be using if they are not using
Terabeam, in the face of this projected incremental tsunami of traffic
which will be engendered by Terabeam's presence where it "IS"
available? Run the numbers, and then you tell me who is going to be
toast. I'll take mine with jam.
The implications of universal free space infrared system coverage
could only be covered by many discussions over a wide array of
topics that would be beyond the scope of this board. Suffice it to say
that I have been a believer and a user of line of sight infrared systems
going back many years. I have lived with both its joys and its perils in
both brokerage and educational settings. I strongly recommend it for
backup scenarios, based primarily on statistical unlikelihoods, and
would be very cautious with it where mission critical systems were
involved, except possibly in the role of a secondary, or even a tertiary,
backup utility.
Some of the reasons for this stem from the actual physical
characteristics of these systems in the face of sometimes hostile
surroundings, and some of the reasons have to do the perceptions of
upper management, and what will or wont pass the audit criteria
found on the controller's and corporate security officer's checklists.
FACts |
