Wireless Internet Cafe for Africa?

Not wishing to appear frivolous but you'll need:-

A connection to the internet An access point and most of all electricity.

The last one is crucial and given my trip to Africa, wasn't always available.

Do you want to charge? If so then you can buy a hotspot in a box, there are several such as that by Dlink, Phenet and other manufacturers.

If you don't want to charge then just dump an AP on the end of the internet connection.

There are other issues from a contractural and responsibility point of view but you need to crack the first three above first.

David.

Power can be provided by solar cells. I would think that the biggest challenge is internet access. How far away is the nearest location that has internet?

Mike Schumann

"Mike Schumann" hath wroth:

Does a typical African village have a coffee shop?

Perhaps it would be helpful if I offer a 2nd hand horror story to show what you'll be up against. This was setting up a radio site in central America mostly for VoIP telephony but which included limited Wi-Fi internet access.

The backhaul for the internet was provided via a satellite link. I'm not sure who provider but it wasn't DirecWay or one of the consumer satellite vendors. Satellite was a pain in the posterior because of limited performance. It was also expensive at something like $1000/month. However, attempts to do the same thing via terrestrial wireless was impossible because the location had several mountain ranges in between the village and the nearest central office. Copper wire was laid but stolen within days of installation. Fiber was tried but someone stole the repeaters. That left satellite.

In order to install the central wireless site, it was necessary to protect the installation. A concrete blockhouse was built. Tower installed. Everything surrounded by a barbed wire fence. Several families from the village were bribed into moving next to the blockhouse to protect it from vandals, thieves, and revolutionaries. Apparently, that's common practice with radio installations.

Power was all solar with some very big expensive batteries. It also ran some VHF/UHF radios so it could have been smaller if it only ran the wireless. The major drain was the satellite system which burned a continuous 80 watts and the tower lights at about 120 watts. See my Excel solar calculator:

a guess on your required battery size. Also, batteries were expected to be replaced every 5-8 years. The system has been running for about 5 years and the batteries are definitely going to need replacement shortly.

The original wireless system was 2.4Ghz based, but that didn't work well. The jungle was just too thick and too high for 2.4GHz to penetrate. The 2.4GHz cast aluminum dish antennas also corroded badly. The system was switched to 900MHz with plated and painted yagi's, which worked much better. I'm not sure who the vendor was on the original 2.4GHz system, but the 900MHz setup as from WaveRider:

In the past 5 years, literally everything has been replaced or upgraded at least once. The new LMS4000 series client radios seem to be holding up well, but the Sipura VoIP ATA adapters are constant problems. Just about anything that fails should the battery voltage be applied backwards has blown at least once. Because of insufficient ventilation in the blockhouse, everything inside has at least one layer of green slime.

Anyway, it goes on and on like that. Don't ask me about the joy of troubleshooting with broken English, non-technical, and no replacement parts. Fortunately, I'm #4 on the call list so I don't have to deal with too many horrors.

You might wanna ask the same question on the ISP-Wireless mailing list, where there are some people that have done systems in Africa.

That sounds like an awesome project, National Geographic Channel type stuff.

My experience with tactical comms, however, is leaving me to believe that you aren't going to be able to this /cheaply/ though.

How far away from the village is there an area of populace and resource? I'm inclined to believe that your best bet to bring internet in will likely be using commercial equipment (i.e., Harris Corporation) for a point-to-point microwave link between the village and a populace. Start working on proposals now because you will certaintly need government subsidy or corporate sponsors. I imagine there probably aren't any existing tall structures in this village and vegetation will also come into play, so will likely need to mast these radios as well.

Some work has been done in this area.

I'm an active humanitarian and computer scientist, but I consider this to often be "in the field, in New York" work. By that, I mean that people think that it can be done anywhere in the world, when it can't. It's a tempting thing to do because it avoids the cost of putting an engineer on the ground. The fact that we're discussing it in this newsgroup is an indication that this is the case, as is the fact that you refer it the country only as "a third world nation". The right solution will depend a lot of geographic location. I would strongly urge you to avoid this. The best story I remember along these lines was this: After the tsunami, a certain 'humanitarian' organisation (who shall remain nameless) decided that they needed to build an IDP camp in a certain area of Sri Lanka. They got somebody in, who sat in the capital, selected the site, and designed the camp.

They hired local contractors who built it, at a cost of (I guess, looking at the photos) about a million dollars. Somebody from another organisation went to have a look. The camp was badly designed, in common ways (poor design of the buildings, buildings far too close together, lack of facilities etc), but these issues are unfortunately pretty normal. The visitor then walked down the road, found a small village and asked the people why nobody was already living on this seemingly excellent land. Answer: it's on a flood plain. The camp is now 12 feet underwater. Nobody ever lived in it.

Learn from this mistake: VISIT THE SITE AND LOOK AROUND. Only then can you think about the best way to deal with the problem.

I will, however, give a few general tips. I can give you more tips if you tell me where in the world we're talking about. To start with you need to consider four areas:

1. Technical.
2. Legal.
3. Anthropological.
4. Education.
5. Maintenance.

Don't forgot about the last four.

Technical. Why are we having this discussion in alt.internet.wireless? Have you discounted wired? The cost of laying wires is usually very low in these places, and fiber may well be feasible. If you live with high latency links (because you're only providing email, etc) then a guy on a bike carrying some storage device to another village may be a "link" in the case of a particularly small or hard-to-network village. People often think of solar power as being sensible in this situation, but it's very expensive; a petrol generator often makes far more sense, and people could use it for other things too.

Legal. Are people legally allowed to access the internet at all? I guess you checked that, right? :-) Is access to certain information not allowed, and would you get in trouble for not censoring it? Political stuff? Pornography? Child Pornography? What are the laws regarding radio transmissions?

Anthropological. Will people accept it? (It should be asked because it's easy to ask, but I can't think of anywhere that would answer: no) How are they likely to use it? Where should the computers providing access be physically located? Does a "cafe" make sense socially? How do people normally gather socially? Where do people normally location communal tools? Are you going to annoy anybody (the guy who currently is employed to deal with all the communications, passing messages, mail, etc)? Do you care?

Education. How will you ensure that people will actually be able to use it? Is information on the internet generally avaliable in a language that they speak/read? Can they speak/read the language of the operating system (bear in mind that variations in the language can be very large - you may best off taking a screenshot and asking people)? Is an appropriate alphabet supported?

Maintenance. What happens when things go wrong? Who pays for it? Can replacement hardware be sent? How can people inform you that it's broken if their only major method of communication is now broken? Could you train a local to fix the stuff? Physical problems: water proofing, etc.

I hope this provides some information (or at least raises some of the right questions), but I didn't have an awful lot to go on. If you provide me with more information, I may be of more help.

Alun Harford

Oops.

Good advice. One minor addition. Make sure that the local government hasn't sold monopoly business franchises to someone. I ran into that one in India. Only the unerfunded and seriously incompetent local monopoly holder could provide internet access. Nobody noticed because they weren't doing anything useful. Some high level strings were pulled to end that impediment, but it could turn into something serious and should not be ignored.

I don't travel anywhere but do advise on such projects. While officially, I'm only involved in the technical aspects, the ability to make spectacular mistakes seems to be epidemic. In particular, maintenance is often performed by someone who's primary mode of repair is to pound on it with a hammer.

In many parts of the world, getting one's hands dirty is considered beneath the dignity of the upper classes (and sometimes the educated classes). One US group (name withheld) found it best to prototype the system in the US and invite the tribal chief or war lord for a visit to the US for a week to see how it works. The chief never shows up because he's always afraid that there's going to be a revolt back home in his absence, but someone (usually a local skool teacher) can usually be found. The trip to the US (including the requisite Disneyland diversion) is generally more useful because changes can easily be made in the US.

Light reading and clues (how it's done in India, not Africa). Note that it's not free, but just very cheap.

Microsoft takes Internet kiosks to rural India Program offers localized content and applications

Women lead rural India's internet rush Bringing IT to Rural India One Village at a Time

Power helps if the customers have it too otherwise it could be a bit of a pointeless exercise. I'm not just talking about the coffee shop here but the whole village.

David.

Gasoline or diesel generation is probably cheapest way to go. Solar for any real power is very expensive and in poor/impoverished countries, subject to growing legs and walking off. Not that a generator can't grow legs, but it can more easily be secured.

fundamentalism, fundamentally wrong.

I have assisted in only one of these projects and then only on the edges. Solar worked and was not too expensive as most state side designers assume

24/7 where as the village needed the thing to work only for a few hours a day and not every day. In the event there was a need the local one eyed (in the land of the blind the one eyed man is king) technical type was available to flip the switch.