Back at the office

Now, back at the office, was able to start to organize my thoughts, notes and research from the trip.

Was able to reach a satellite teleport operator in Florida that does a fair amount of business in the West Africa region. He was very helpful in filling in some details on what the various satellite options were and what the relative costs of each were. Also, a US based teleport has a distinct disadvantage of European teleports in that these guys end up doing a double-hop satellite bounce to make it happen. This sounded like a tough business for US businesses, as it effectively doubles the cost of transmission...

I'm currently working on getting some information from PanAmSat as well, hopefully they'll be able to continue to fill in the gaps that I have as far as current capabilities and cost models for Internet access in the region.

Clearly, this multi-hop satellite based Internet solution is sub-optimal on several dimensions: cost, bandwidth, latency, capacity, reliability, locality.

As to cost, for example, a 256kbps dedicated both ways, C-band setup costs about $4k/month. I can get a full dedicated T1 (1.54Mbps) at just $400/month and that's not even the best deal. Latency -- I wish I had measured my latency from Abuja, but I can't imagine it being real good!

And then try to cover the $4k/month costs using the current market price of N100/hour at cyber cafes. At 100% capacity, each PC can generate some $500/month -- just to cover a decent dedicated network, one would need 8 full time fully utilized PCs. Since that just won't happen, what appears to be happening is the providers are opting for shared bandwidth (perhaps a 35:1 sharing ratio) to get the costs of bandwidth down, with a complete loss in quality of service and anything approaching broadband speeds.

Not the best situation, and since the connectivity continues to be fragmented with each provider solving the same problem over and over again, their is no market strength for the African providers...