I originally tried to tackle Government ICT policy issues directly (eg, contributing to the SA position paper ("The Information Society and the Developing World: A South African Perspective", 1996) for the ISAD 1996 conference called by then vice-president Mbeki). It did enable us to develop a new perspective regarding the "Information Community":

The information age has not so far contributed to a reduction of inequalities between people, regions and countries. Where it has entered the social arena, IT ha s not operated on a communal basis but rather on an individual one: personal computers rather than community access. To date it has served to concentrate wealth and power. (p 3)

Information infrastructure is not an end in itself, but simply a means to an end. Appropriate, timely, culturally relevant content must be made available. Indeed, the creation of content is the raison d'être for the development of infrastructure.

Ultimately, however, I found that experience frustrating and without practical consequeces.

Instead I have decided to focus on developing small telling examples of ICT applications that expose ICT possibilities for development and at the same time highlight policy issues. A good example of this is Chetty, Blake, McPhie “VoIP deregulation in South Africa: Implications for underserviced areas”, 2006. This connects with the work I have done on use of ICT for development, empowerment of deprived people and indigenous knowledge. The most spectacular example of this is no doubt the CyberTracker system which has attracted a great deal of media coverage and has a worldwide impact in terms of downloads and use of the software (see CyberTracker website to get an idea: www.cybertracker.org). I published the account of that development as Blake “A field computer for animal trackers” 2001.

In reflecting on this and subsequent projects I have been struck by the need to rethink the way we do SE in a country like South Africa. SE needs to step outside the traditional methodologies and involve much more of the social context in which the software is placed. If we do not, we run the huge risk of having no useful impact - as has happened so often. The methodology I favour has grown out of the realization that there is a parallel between the cycles of newer SE methods and the cycles of (Critical) Action Research. The development has lead over the years to Socially Aware SE, eg, Blake and Tucker “User interfaces for communication bridges across the digital divide”, 2006.