[Server-sky] Communications - how much is enough?

[Keith Lofstrom]

I was talking Wednesday with Don Hoffman, who helped lead the 
Teledesic effort.  He was telling me that licensing a lot of sats
would be problematic, and that we really needed high inclination
orbits so we could talk to northerly customers (M288 can see about
60 degrees north theoretically, and practically probably 50 north).

It occurred to me that the first arrays of server-sats will be in
the computation business, not the communication business.  We 
really don't need high bandwidth global coverage in the beginning.
While we do need high bandwidth upload/download to a few ground
stations, we can do most of our computation business with low
bandwidth communication.  Perhaps we can use existing networks,
such as Globalsat and Iridium below, and TDRSS and the traditional
GEO comsats above, for the continuous communication stuff.  Further,
since the traffic to existing satellite networks is probably bursty,
and we are packet oriented, we can probably shove our packets in
between the high-paying high-QOS stuff that the existing companies
sell.  That should result in cheap backhaul for all but the highest
bandwidth high-QOS stuff.

This relieves us of any need to do complete constellations or 
high inclination orbits.  This makes the bandwidth/frequency
licensing much simpler.  We don't have to worry about array-to-
array communication.  That means we can use our poorly-cooled
radios in burst mode.  We can probably deploy with one launch.
The job gets much easier.

This is a way to get cash-starved com-sat companies on our side.
In the long term, we may (or may not) be competition, but in
the short term we pad their bottom line.  In the long run, they
might be the ones buying server-sats to expand their businesses,
so they still make big profits.  We just swap some vendor-customer
roles while maintaining productive relationships.

The above steers awfully close to business stuff that I want to 
keep private for our investors and business geeks (they like
private), but it has high impact on our technical decisions,
so I am sharing it with everybody.

Keith

-- 
Keith Lofstrom          keithl at keithl.com         Voice (503)-520-1993
KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs