[NNagain] The rise and fall of the 90's telecom bubble

[odlyzko at umn.edu]

Dave Taht's question about all the redundant fiber that
was put down in the telecom bubble is a very interesting
one.  It would be nice if some folks on the list could
provide some solid information, even if only for one
large carrier.

My impression, from communications with various folks,
is that much of that fiber from around 2000 was never
lit.  The reason is that better fiber came on the market.
However, what was used (at least in some cases, again,
this is something I would love to get real data on) was
that some of the empty conduits that were put down then
were used to shoot the new generations of fiber through.
(It was quite common for carriers to put down 4 conduits,
and only pull fiber through one of them, leaving the
other 3 for later use.)

Concerning the Doug O'Laughlin post that Dave cites,
it is very good.  For more on the myth of "Internet
doubling every 100 days," my paper "Bubbles, gullibility,
and other challenges for economics, psychology, sociology,
and information sciences" published in First Monday in
Sept. 2010,

    https://firstmonday.org/ojs/index.php/fm/article/view/3142/2603

But O'Laughlin is too hard on Global Crossing, for example,
when he says it "was essentially a fundraising scheme looking
for a problem."  Global Crossing had a real business plan,
it was the first transatlantic cable that was not built by
consortia of incumbent telcos, and it planned to take
advantage of the rising demand for transmission by offering
capacity to new players, who would otherwise be gouged by
incumbents.  (It did get into accounting shenanigans later
on, as competition arose, but that was later.)  What is
most interesting is that their business plan was based
on an assumption of demand about doubling each year (which
is what was taking place), not doubling every 100 days.
(This I learned when I was consulted on some of the
litigation after the telecom crash, but by now the
information is publicly available.)  What killed them
is that their assumption that it would be difficult for
others to get the (special undersea) fiber, the cable-laying
ships, the permits, ..., turned out to be wrong, and so
a slew of competitors, inspired by the myth of astronomical
growth rates, came on the scene.  (Global Crossing's expansion
into terrestrial fiber networks was also a major contributor.)

One of the astounding observations is that while Global
Crossing was assuming 100% annual growth rate in traffic,
the industry as a whole (as well as the press, the FCC,
and so on) were talking of 1,000% growth rates.  And the
only observer that I was able to find who noted this in
print was George Gilder, who drew the wrong conclusion
from this!  (Details are in the paper cited above.)

Andrew

P.S.  Some interesting materials from telecom bubble era
are available at

     https://www-users.cse.umn.edu/~odlyzko/isources/index.html