We know that currently, there are more than 4.6 billion
mobile phone users around the world, and in South Africa 90.16 people in every hundred use a mobile phone. But how many of these people are using the mobile Internet?
At one stage, there were discrepancies of 9.5 million in estimates of the number of South Africans using the Internet on their phones. In 2008, Rick Joubert, head of mobile advertising at Vodacom claimed that the number of unique South African users accessing the mobile internet using WAP was almost double the number of South Africans accessing the Internet via fixed lines. He had extrapolated data about the most popular mobile Web destinations for South African users (e.g. 2.3 million Vodafone Live! users and the then 9 million registered MXit users) and estimated that by early 2009 there would be more than 10 million mobile Web users. At that point, the number of South Africans using fixed lines to access the Internet was estimated at 5 million, and so this was a claim that grabbed headlines and gained a great deal of attention. Joubert speculated that the online media industry did not cater adequately for most of these users, since up to 70% of them might not have any other form of Internet access. He also pointed out South African media brands did not feature at all in the top 50 mobile sites.

At a Netprophet 2009 talk, Arthur Goldstuck argued that these estimates of mobile Internet should be questioned. Notably, he pointed out that the mobile advertising industry has ‘a vested interest in persuading corporates to market to 10 million people on their cellphones’. Goldstuck pointed to proprietary research by his company, World Wide Worx which estimated a considerably smaller number of mobile Web users (500 000) , or 180 000 people who use their cellphones as their primary form of Internet access.

It is possible to reconcile these two sets of figures if we develop a more nuanced model of South African mobile Internet use. Jonathan Donner and Shikoh Gitau differentiate between mobile primary and mobile only Internet use (Donner and Gitau, 2009), a model which recognises that many people have some kind of access to the Internet on computers, but that they are more comfortable using their phones, or that they have to use their phones most of the time.

My MA student Tino Kreutzer did some excellent research for his MA dissertation (2009), where he  differentiated between mobile Web use (the use of mobile browsers to visit websites and WAP sites on mobile phones) and mobile Internet use (a far broader category which includes any use of the Internet protocol, such as the use of Internet-based mobile applications such as MXit). Goldstuck suggests further categorisation, which would allow a sense of the range of audiences now available to online marketers via mobile sites, as well as the new groups of users who can now potentially be reached in different ways, via MXit and other Internet platforms.

How we understand ‘Internet use’ is obviously crucial to this debate. Technically speaking, the ‘Internet’ is the transport mechanism – the network of networks which links devices and the TCP/IP protocols that facilitate data transfer around this network. What people call the ‘Internet’ depends on what applications they use, and the platform to which they have access – while students and office workers might use Facebook and Twitter, and might be able to check Google for every query, other South Africans might only know MXit, wap download sites like zamob.com, and while some may well have used the ubiquitous Google for an occasional query, they may not even realise that all these things they do require the Internet to work.

In common lingo, ‘Internet’ is most often used to refer to the media transferred via the Internet – the graphical interface, or the Web, most often viewed via a browser, which has traditionally been used on a computer. Most South Africans have no experience of the Web, and consequently associate ‘Internet’ with computers, rather than with the mobile applications such as MXit which are more widespread, but which only work on phones with Internet access (e.g. GPRS or 3G).

Many people use applications such as MXit, but they don’t always know that they are using the Internet. For example, the  mobile-centric teens I’ve interviewed may often refer to MXit as a ‘game’, since it is stored in the same folder on their phone where their games are stored. The need to define ‘Internet’ does not only relate to the reporting of statistics, but also extends to the way in which data is gathered in surveys of Internet users.  Tino’s dissertation (2009) makes the point that the word ‘Internet’ bedevils the usefulness of many surveys where people are asked whether they use ‘the Internet’, more particularly when this question is used as a filter for further questions in the survey (such as in the AMPS survey, for example).

Goldstuck rightly explains  that the nature of Internet use is very different for users who primarily use their cellphone’s Internet access for WAP downloads, or to IM with their friends on MXit. This is obviously very important for marketers. At the same time, Joubert’s point, that the online media industry does not understand the needs and interests of the new group of mobile-centric web users, remains entirely valid. We could even start talking about different ‘Internets’, given the different socio-economic circumstances, technology, display capability and bandwidth available to people in our country. Consider two South Africans. One might use a 2 year old handmedown Nokia to access MXit, and treasures the tiny amount of precious on-board memory on the phone where she can store only a limited number of pictures and music while her access depends on prepaid airtime and electricity to charge the battery. Another wealthier South African might own a desktop and Mac Powerbook which hold terabytes of data, and is connected via a broadband contract which allows a sense of always-on connectivity, interrupted only by the occasional Eskom power heist or MWEB capping message. He or she may have Internet access on a smartphone as well, but  is seldom motivated to use it. These two people thus have an almost entirely different experience of connectivity, which leads to distinct concepts of the Internet, and it certainly means that their demands and requirements for their phones are very different. And we haven’t even started to address the other differences that would come into play. We should thus beware of the assumption that everyone’s ‘Internet’ looks the same as our own.

I recently asked MXit how many of their 15 million registered users are South African, and the answer was 13  million. This is particularly impressive if you think how complex it can be to download and install an application on a cellphone. As Vincent Maher said at a recent talk at UCT,

[MXit] has prepared about 5 or 6 million young people for the process, very painful as it is, of downloading and installing an application on their phone. The Americans wouldn’t do it. They needed the iPhone to come around before they would do actually bother to have Internet on their phone, because it was just too complicated. But thanks to MXit (and I think it has around 12 million people now) there is an entire generation of South Africans who understand how to interact with the operating system on their phones.