The driving force that propelled the Internet of the 1990s across the Atlantic and expanded into the earth-encompassing information network that we have today, was a particular distributed application called the World Wide Web (a very basic Internet-based hypertext system). Today the Internet has become a vital infrastructure for developed countries that is similarly important as running water, sewage or traffic. Despite this, there are still large parts of the world that are not connected to this monolithic global information network, namely within underdeveloped regions.
The WWW, however, is more and more recognized as a means of cheap access to (self-organized and localized) education, as well as an incentive to pursue it. Education, in turn, is considered a pivotal factor for the democratic, social and economic evolution of underdeveloped regions. In addition, mobile ICTs may be a seed for the formation of small local businesses like food cooperatives, microfinanced enterprises etc.
Despite the obvious utility of wireless digital networks for underdeveloped regions, the introduction of network infrastructure within these seems like a chicken-or-the-egg problem: There is no immediate incentive for network providers to introduce wireless networks to areas where nobody has a connected device, yet. On the other hand these devices are useless, when there is no network infrastructure available.
However there is a way to introduce digital networking to underdeveloped regions that doesn't require rigorous initial investments in network infrastructure: Cheap Android-based tablets are useful even in the absence of wireless communications. In addition they may contribute to network accessibility by serving as a node within an ad-hoc network.
In order to provide an incentive for the use of this feature, it is necessary that information can be disseminated across communities of users without the need of a permanent connection. (E.g. Information can be passed on in direct communication, like you would do with USB-drives, while the ad-hoc wireless module is kept on line, mainly in order to find other users well before the critical mass for an ad-hoc infrastructure is reached).
In order to make this happen, it takes applications that support this notion of low-entry-threshold usability. So, a crucial Problem that obstructs the dissemination and evolution of ad-hoc digital networks is the lack of applications that implement similar use cases as applications on the Web. Most Web applications in turn only work as long as an Internet connection is available, which is something that is usually not given within underdeveloped regions (or disaster sites). So most Web applications are not portable to ad-hoc Networks.
This calls for a general framework for hypertext and hypermedia applications which is tailored to the special circumstances of ad-hoc networks. These are:
o High churn: there is a constant in and out of peers.
o Decentralization: No nodes can be assumed to be fixed. Hence there is no client-server dichotomy.
o Geographic proximity: Most peers will be close enough to engage direct communication. Ad-hoc applications are also appropriate for tasks that comprise both, direct and digital interaction between peers.
o Reduced anonymity: The fact that most peers are likely to be somewhere near has manifold implications: On the one hand things like spamming or scamming are less likely to occur (and in a smaller scale) within ad-hoc social networks. On the other hand the policy and safety of personal information needs to be reconsidered.
There already exists a model that gets close to the needs for decentralized information sharing and dissemination, namely that of peer-to-peer (p2p) file sharing. So the overall vision is to merge the idea of p2p file sharing with that of a general hyper document (or rich client) framework in order to realize a p2p Web that works both, within localized ad-hoc networks and over the Internet where Internet access has only the effect of a broadened access to other peers (and maybe trackers).
There are good reasons why inhabitants of an underdeveloped region should not be confronted with the whole World Wide Web from the start, but rather interact within their neighbors and real-life acquaintances, first. In addition, these kinds of p2p Webs could make up for a restricted or censored Internet access during times of political instability. When p2p Webs fall apart they still make up a sub-Web that contains left-over information and the means to communicate over these (e.g. through comments).