Dear Mr Gilder:
I realize that you are more interested in identifying
emerging technologies (such as point to multipoint
wireless systems) than leading companies in the
markets associated to these technologies, but as
others have suggested before, you may wish to
reevaluate the competitive stance of TGNT versus
Winstar.
For the sake of full disclosure, let me indicate first
that I have an investment in WCII and not in TGNT.
First, if you examine bandwidth, WCII has approximately
750MHz in all top 50 US markets, versus TGNT's 400MHz.
It might be argued that there is actually a difference
between bandwidth at 24GHz and at 38GHz. However,
the comparison cuts both ways: for line of sight
communications, the path length attenuation varies
with the square of the frequency, so that attenuation
is higher at 38Ghz than at 24GHz. However, if one
uses transmitter and receiver antennas of the same size
at 38Ghz and at 24GHz, the higher path loss is more than
made up by larger antenna gains at 38Ghz (this is the same
advantage that satellites in the Ka band have over
satellites in the Ku band). Next, WCII has roof rights
for approximately 2,400 buildings and is installing lines
at the rate of approximately 40,000 to 50,000 lines
per quarter. TGNT is barely starting in this respect.
Both TGNT and WCII will deploy P-MP equipment from
Nortel at approximately the end of this year. However,
the big difference is that WCII has already a large
customer base. Usually, satisfied customers rarely change
their local telephone company, so being there first
matters a lot. The only advantage I would grant to
TGNT is their lower capital costs due to the NTT
connection.
If you are just interested in P-MP deployment, there is
little difference between TGNT and WCII. However, from
the point of view of maturity of their CLEC businesses,
WCII is way ahead.
Now that I got this issue out of the way, I would also
like to question your selection of P-Com as a proxy
for point-to-multipoint equipment technology. The
point to multipoint equipment business is going to be
extremely competitive. Existing suppliers and
alliance include: a) Bosch Telecom/CSCO (Bosch
for the millimeter waves gear, CSCO for the ATM
networking interface), b) LU and Stanford Telecom
(Newbridge signed also an agreement recently with
Stanford Telecom), c) Nortel (including its BNI
unit), d) Unique Systems and Bay Networks (don't
know what will happen to them after BAY's acquisition),
e) P-Com, and several others... This market will be
quite crowded. Furthermore, the high margin side of
this business will probably involve the bandwidth
allocation products and system integration (integration
of the wireless P-MP part with a fiber network), not
the millimeter waves components suppliers. Thus, although
P-MP technology is very exciting, due to its ability to
allocate bandwidth dynamically, the winners are likely
to be the usual suspects (Nortel, etc ...) rather than
smaller companies such as P-Com.
Sincerely yours,
Bernard Levy |
