First off, I think that making information available inexpensively to
people in the third world is probably the most important thing that we
can do for them - technologically, culturally, politically. Ideally with
as little censorship and the like as possible.
I find the comments of some of the responders to this thread interesting - and in many cases I feel that they reveal as much about the poster as about the problem.
In particular, mentions of music sharing and porn.
In this vein,
I just heard part of a program on NPR that talked of how porn has been a major part of every emergent medium since Gutenberg. So, if there's some porn floating around but it spurs development of the information infrastructure, so be it.
When someone says "get medical care to the people first", they dont say who will do it, how, nor who will pay - but with good information access, I can envisage a system in which students would be given a year or so of education in basic medical care and moved to villages and areas with none at all. If they had good communications (internetish, say) they could continue to study while there and even consult with doctors on an "as needed" basis. This would be much cheaper and more effective than trying to import trained medical care, or than trying to send students abroad.
I don't know that much about the technical details of wireless
networking (indeed am struggling to get an 802.11 usb port working in
linux), but I know a bit about networking.
Within a village or town, wireless is easy enough to build and will likely
grow - I don't know how it would work, but something like cellular routing with
each machine contacting others in range and sharing traffic with them certainly
sounds feasible. And it could likely be made nicely redundant and tolerant to
the loss of access points.
At the big city level, there is likely to be communications infrastructure available.
So the real problems lie in connecting the villages and otherwise geograpically
isolated stations with each other or to some node serving as a hub.
At the outset, uplink speed is not likely to be that important so
satellite downlinks are tempting. Of course we need to consider the
cost of the satellite, which could be shared with other countries and
is likely to be more than reasonable when amortized over the
population it serves. Third world countries are likely to be less
than enthusiastic about such an outlay, but it would likely be cheaper
than building on-the-ground infrastructure.
I've come to think that even "receive-only" access
could be useful if there is storage to save information for a few days
(or longer). News, weather, music, video, lessons for the schools,
commodity pricing information, and so on could all be sent out, stored
on a local machine or network, then viewed and browsed more or less
interactively by users over the next few days. The utility is less
evident, but it could even mirror interactive discussions (ala irc).
This could provide a much better window onto the information world
than may be currently available, and might further serve as an
incentive to develop more interactive access.
If receive only access is useful, then any layer of uplinking is
likely to make it more so. Indeed if it takes a day to send a message
upstream in an otherwise receive-only situation, the answer is still
likely to arrive more quickly, and with more information, than it
In this light, even very low bit rate communications upstream (via
radio links, slow phone lines or whatever) could be combined with
efficient downstream links to provide decent information access to
those who would otherwise lack it.