Greetings Afrik-ITers!

David Dion writes:

>         I would suggest that you investigate radio modem
> technology. While this may mean somewhat more investment up front
> than normal telephone modems, it is a lot more sustainable because
> you have no transmission costs.

Ah, radio.  My favorite subject!

David, there are I think actually a few hidden recurring or
operational costs of radio systems.

Transmission speeds for long-distance radio (more than, say, 100km I
guess, requiring HF radio as opposed to some other kind) are quite
slow, less than 2400bps on the WFP system in Kampala.  The radio
and computer are thus occupied some 12 times longer than would be the
case for transmitting the same data over a 28,800bps telephone modem.

One commercial service in East Africa levies a volume charge to
cover the cost of having its system occupied with data transmissions
for long periods.  This volume charge is reportedly US$0.30 per
kilobyte. An organization sending and receiving on average about 10
pages of data daily (20k) would have to pay about US$230 per month
for that volume.

If the base station is operated in-house, there is a cost for the
system operator and technical support for users.  WFP in Kampala had
a staff of 2, with oversight time required by others.  Staff are not
cheap.  And available time slots on an in-house server will fill up
12 times as fast as on a telephone based system, perhaps
necessitating purchase of additional transceivers and modems on a
high-volume system, with the technical problems I'm told that may
arise when more than one transceiver is operating in a given

Equipment costs for HF (high frequency, for long distances) systems
are about $8000 per site connected, including a base station.  Base
station costs are higher of course if there are multiple
transceivers and radio modems.  There may be onward forwarding
charges from the base station to the Internet.  If the commercial
provider is the base station, then these forwarding charges are
included in their charges.

The technology is not ubiquitous, so technical support is not as
readily available as with standard telephone/PPP.  Reliance on highly
paid consultants is common, and by definition costly.

Having said all that, I still believe there is a role to be played by
this technology, and am enthusiastic about it.  But the costs are
problematic.  I am presently trying to identify sites where the value
of information is sufficiently high to justify those costs, a place
where USAID might then invest in radio connectivity.  My thought is
that radio connectivity to an institution with many scientists, or to
a kiosk with many clients, makes more sense than connectivity to a
single individual or office.

A few references:
and search on Uganda.  Check results for Bushnet.
for a brief introduction to radio email.


Jeff @ Washington

[log in to unmask]
1111 North 19th Street Suite 210
Arlington, VA 22209 USA
Tel 1-703-235-5415  Fax 1-703-235-3805