Last week I attended Africa Peering and Interconnection Forum; it brought together ISPs, content provider and regulators from 20 African countries but the Communications Commission of Kenya was conspicuously missing.
It is only after attending such meetings that you understand some of the reasons the cost of internet will not drop for the individual, who is probably looking forward to the day we pay $20 per month for limited connectivity.
First, peering means exchange of content between ISPs at a central point like the Internet Exchange point (IXP). Think of an IXP as a hub in airline terms. Nairobi is the regional hub and most arlines with no connections to the region from Europe usually get their passengers to Nairobi, then they are taken over by KQ or another airline.
For instance, if you are travelling from Windhoek Namibia to Lome in Togo, you will have to buy a ticket from different airlines or from one airline that goes to Lome, which is likely to give the cheaper option. If you decide to travel SAA then KQ, there is no doubt the ticket might be higher; this is Africa!
The same principle applies, if your content is hosted abroad and you are in Kinshasa, you have a choice of ISPs that will charge the cost of international connectivity because at some point, the data will leave your ISP network to another, and they have to pay. Assuming that content was hosted locally, it means that cost of international connectivity would be eliminated.
But the internet costs remain the same whether local or international, so why bother? Thats a question that many people ask and the answer is simple; we do not generate enough content to spur competition in hosting services and lower the costs. Abroad, people can charge lower because maybe the server is in a garage somewhere, under the bed or in a company like Yahoo, that attracts 10 million sites a year, so it can afford to charge 10 dollars and still make money.
Ever wondered why a ticket to Malawi costs $800 and one to South Africa costs $400 yet it may seem logical that the reverse be the case? Well, it is argued that the Jo’Burg route has more flights and airlines compared to Malawi and therefore more competition. Indeed, if you are going to Lilongwe via Kenya Airways, you might be taken on a round trip of your life but if they attempted that on the Jo’burg route, they may lose the customers.
Anyway, if the ISPs in the region do not see the need to share content and so long as we continue hosting abroad, the cost will remain high. Of course those are not the only issues, there is licensing and the desire to get a quick return on investment, maybe the period may be made longer.
There are ways content providers can reduce costs by setting up POPs like in Nigeria and ISPs evolving to be more of regional carriers and reduce backhauling costs like Orange is doing in the region
There is no doubt that the industry is getting better; maybe most us host 9 out of 10 websites locally but it will probably get to a time when ISPs say that if you access websites that are local, then the cost is low and flat, maybe like they do with on network calls.
It may look simplistic but we all have a role to play in making internet cheaper. Follow-up piece can address licensing issues and benefits we can expect from having Kenyan Internet Exchange Point, Mombasa Internet Exchange Point and a POP hosted at Internet Solutions in Kenya.