@ ads I would be happy to debate the point with you.
We need penetration % to get real growth. This will never happen as the vast majority of our population does not have the income to even afford the most basic of internet connections. This is why we will never have the scale of internet that exists within Europe or the USA. Simple fact of life. 90% of our population is battling to survive. WHy would they want internet ?
The funny thing is, a few years ago, you'd hear exactly the same thing said in India (where the inequities in society are much the same as South Africa), but it doesn't stop them from thinking of themselves as (and being) the IT leaders in the world, and doing all the right things to get Internet to people. Even today, they have only reached 7% Internet penetration, and broadband is much lower still. Only a very small part of the population is actively engaged in high-tech industries, or even in service industries at all, and income in India goes down to levels way below that of South Africa, so it can't just be about that, even if there are large numbers of people to buy.
I think you'll find, if you do a few comparisons globally, that South Africans, on average, don't get close to "battling to survive" on average. Try looking across the border at Zim, for example. High unemployment and poverty are still better than actually dying of starvation. If there's no income out there at all, who's using all the cell phones? Starving people don't pay even 99c for a starter pack.
I think our problems run deeper than that. Almost any idea that may really make a difference is hijacked by other, supposedly higher priorities. We cannot simply create a model that gives incentives to private investment - it has to be the "right" (left?) private investment, and it has to fit in with particular political logic. Why, for example, does the whole spectrum policy apparently hinge on poor rural consumer needs, and the issuing of spectrum depend on who is applying, rather than the value it would bring? Countries that have got spectrum policy more or less right (and here I exclude India, that has messed up its hopes of growing 3G with bad spectrum policy) actually value the science, technology and associated knowledge that is required over half-baked economics and political favours. The greatest spectrum needs are urban, and one has to understand all the markets, not just poor consumers.
Of course, we also have to get over the insidious tendency to insist that everyone has the same needs. The most successful efforts to extend Internet to larger populations have been through shared models - Internet cafes etc - just as we've seen with community pay phones for voice here. The pay phone model works here because there really is money in it for all concerned, from the shop owner to the network owner, although it was dressed up as a "community service obligation". If there were similar incentives for Internet e.g. you can have all the spectrum you asked for, and do what you like with it, provided you build 1000 Internet cafes, and the government will give you a subsidy for each one you build it's still there after a year. Models like that work in some places, and fail in others, but why not try? In South Africa, they wouldn't get past the starting blocks because there would be so many conditions and vested interests, political twists and incompetent government officials in the way as to make it impossible to start.
Your point about "why would they want it?" is trickier to answer, but maybe, just maybe, it's got more to do with lack of availability than lack of need. We actually have some of the best government and related web sites in this country, and yet they are largely wasted on the few who can access them. Your average non-Internet user (for want of a better term) doesn't actually know what they are missing. Personal Internet may well be beyond people's reach, but it must surely be cheaper to get services at a local Internet cafe than catching a taxi to stand in a queue for a few hours. There's plenty of this kind of small scale entrepreneurship going on in other ways, so what's the real barrier here? Nigeria doesn't have this problem - they have thousands of Internet cafes just like this.