I was sent a private note from a student that strikes me as raising questions of general interest.
> Do you share the notion, that I've read so much, that telecom can enable Africa to "leap-frog" in
Not exactly, though it depends on what aspect of development you mean. You'd almost have to define
"development" to answer that question, if only to know what it is that Africa is supposed to
I've more often heard the question asked this way: "Will Africa's telecoms development leap-frog
over intermediate technologies?" Perhaps that's more what you mean.
With respect to purely technical development, then sure, I think Africa will take advantage of the latest
computers and software. I don't expect people to buy an old IBM AT before surfing up to a Pentium
system. I'm generally of the school that believes that second-hand computers and old software are
a waste of money. Some projects distribute 2400bps modems. That's regrettable. I don't know of
anyone who is buying them, so why should they be given away?
On the technological side, which I take to mean the nexus of technical and organizational, then
"leap-frogging" probably isn't the right metaphor. It's probably not a good metaphor for general
The notion of "leap frogging" implies a linear path over which the Internetworked nations have
already passed, and over which Africa will follow. I personally believe technological development in
Africa will be quite different. And I doubt it will take place along a linear path.
Even technological development in America has been anything but linear. For its lower income
citizens, development in America is and will continue to be different than for those generally
wealthier persons at its universities and large corporations.
If you check my notes about the "Wiring Africa" workshop in Washington last week, Sheila Young
talked about "equipment drops". I think she's talking about this same issue. Sure, Africa should
jump right to a snazzy Unix server with banks of dialup modems and a VSAT link to the Internet.
That's the technical side. On the organizational side, however, who's going to "hold" the server,
pay the recurring costs, set prices, facilitate training for users, promote local network
development, and so on? These are the tougher questions.
And the answer to these questions may reflect back upon choice of technical configuration. Linux
might be better than Unix, based on PCs rather than Sun Sparcs (unless Sun buys Apple, in which
case all bets are off). Rather than a Unix web server, a combo server on a PC might be a better
choice -- have a look at Galacticom's WorldGroup, for example, or Mustang's latest Wildcat, which
offer an interesting alternative for all the existing Fido systems in Africa.
Stepping back from the individual system and thinking about national and regional network
development in Africa, generally speaking, what's pushed on Africa for Internet access are the
larger and more expensive systems that require a high level of technical competence, extensive
foreign training, and so forth. In parallel or instead, however, we might think about smaller and less
complex server systems for accessing the Internet.
Consider the implications for centralization and control, network redundancy, and so forth. Think
about a country somewhere in which there's a single Internet link, which publishes something
controversial from a local newspaper. Government knows right were to go to pull the plug.
Farfetched? I seem to recall hearing about something recently...
Imagine how different it might be if there were dozens of small systems in the country, run on
simple PCs, interlinked but not wholly dependent on any one other system. Imagine if dedicated
routings to a central international gateway could quickly be converted to batch telephonic routing
of newsgroups. What if the government were informed that all the publications were already
mirrored on a server elsewhere?
There are interesting democracy and governance issues stemming from network organization, and those
focusing solely on technical aspects will fail to see them. A motivating factor in structuring
America's network development was the threat of nuclear war. I dare say the motivating factor will
be different in Africa, but it might just favor the same kind of decentralization.
Jeff @ Washington, DC USA
AfricaLink (Opinions expressed are solely the author's.)