Why IPv4 screwed Sub-Saharan Africa

I’ll not get in to the specifics of what IPv4 is as you can read this very thorough Wikipedia article if you want to know the details. For the purposes here, all you need to know is that if you’ve ever had to deal with setting up a router at home or seen something like 192.168.0.1, then you have been in contact with an Internet Protocol address, version 4. We’ve been using this system for three decades to give a physical point where a website or piece of hardware can exist on the internet. For instance, when you type in subsaharska.com, you’re actually going to the numerical address 74.50.48.186 as is translated by another system called DNS, but we’ll stay out of that for now.
This addressing system has worked pretty well for us and it wasn’t really thought to have any problems as given that each 8 bit chunk of the available addresses can be a number from 0 to 254, which in turn gave us a total of 4,294,967,296 to play with. It seems like a lot, but most predications show that in the next two or so years, we’re actually going to run out of addresses because in addition to every website in the world needing one, every computer (other other device like a mobile phone) in the world also needs one when online. This is a problem for everyone in the world, but because of these addresses being a scarce resource, they were assigned in large blocks to each of the countries in the world and naturally, as if often the case, Sub-Saharan Africa got shafted. Here is a run down of about 40 countries on the continent and the number of physical IPv4 address that they have available: Angola: 50,432; Benin: 18,432; Botswana: 77,056; Burkina Faso: 30,464; Burundi: 2,304; Cameroon: 68,096; CAR: 3,328; Congo: 1,024; Congo DRC: 15,872; Cote d’Ivoire: 113,152; Djibouti: 12,288; Equatorial Guinea: 2,048; Eritrea: 4,096; Ethiopia: 16,384; Gabon: 155,136; The Gambia: 11,264; Ghana: 180,736; Guinea: 66,560; Guinea-Bissau: 1,024; Kenya: 336,640; Lesotho: 12,544; Liberia: 1,024; Malawi: 22,016; Mali: 24,576; Mauritania: 32,768; Mauritius: 271,360; Mozambique: 123,904; Namibia: 148,480; Niger: 17,408; Nigeria: 479,232; Rwanda: 156,672; Sao Tome: 0; Senegal: 92,416; Sierra Leone: 15,616; Somalia: 0; South Africa: 15,045,120; Tanzania: 122,112; Togo: 12,288; Uganda: 156,928; Zambia: 42,752; Zimbabwe: 43,520. Look up more if you’d like. And thanks to a bit in this article which started me thinking about this.
You can see that there are some pretty large points of inequality happening. For instance, Mauritius, which is a tiny island has over a quarter million addreses, while Sao Tome (also a tiny island) has none and Somalia (a very large country) also has none. Naturally South Africa was given about 15 million, which is the most out of whack country on the continent. Ah, but before you think that this may be unjust, how many addresses does a country like the United States have? 1,477,921,534. Yes, 1.5 billion. That’s 3,000 times more addresses than Nigeria which has a third the population of the US.

The Problem this Creates

If you revert the math (and I hope I’m doing this right) an American citizen will have access to 1,000x more things online than a Nigerian. You can see the disparity and even when taking America out of the picture, you can see how this has played out in Africa in that South Africa, Ghana, Kenya and other countries with larger IP blocks have a much more vibrant web community than say, Somalia or Burkina Faso. And as I pointed out earlier Mauritius with its large block that is out of scope, even has a datacenter, which is impressive given that it only has 1.2 million people.
Having less available IPs in a country raises the cost of getting online due to the fact that you need to have an IP and in order to get one, you will have to pay more in Sub-Saharan Africa because there simply are less of them than in the US. This is another reason why there are so many datacenters in the US, which has in turn led to people from any number of countries hosting their website there despite it incurring a slower response time and not fostering local jobs and development. It’s a fact that it’s cheaper because the US horded so many addresses when they were being handed out (along with ICAAN, but that’s another article…)

The Solution

Plain and simple, it’s IPv6 which is the next generation of internet addressing. IPv4 has 2 to the 32nd addresses while IPv6 has 2 to the 128th addresses. Basically, so many more that I can’t reasonably type that number. I’m sure that one day we’ll figure out a way to use all of these up as well, but in theory due to how the addressing system works, we can just assign massive and very much equal top-level address blocks to each country, meaning that every one of the 195 countries in the world would get their own block to exhaust however they wanted. There would also be plenty of room to add more top-level blocks if places like Darfur or Southern Sudan officially become their own countries some day.
It’s obvious to see that (I’m assuming) the US assigned address blocks based upon internet access, so it’s actually rather good timing that adoption of IPv6 is slowly happening now due to more countries the world over being online than at any other point in years past. There is little one could say to justify the US hording a third of all the available addresses like it did. Of course, they did it before, so it could easily happen again and thus vigilant we all must be as companies are already starting to buy up IPv6 blocks.
Why IPv4 screwed Sub-Saharan Africa

One Reply to “Why IPv4 screwed Sub-Saharan Africa”

Comments are closed.