There is a contradiction, however, between GII-GIS rhetoric, which focuses on deregulation and unfettered mobility, and its more specific policy objectives, which stress the need for 'infrastructure' and 'networks'. These infrastructural policy objectives imply support for fixed invest- ments and communication between fixed points (that presumably lie within state territory). However, the GII-GIS concept, while making use of these territorially situated and state-bound networks that interconnect, avoids referencing the unevenness, nodes and unique dynamics of infra- structure in the networked economy or network society. GII-GIS rhetoric assumes that by loosening market forces from the restrictions of regula- tion, investment in new technologies will emerge to meet market demands. This argument is reminiscent of the liberal view of space and its concep- tual distinction between production/territory/fixity and trade/non- territory/mobility. According to this rhetoric, the state territories within which productive activities occur exist logically prior to trade, and 'free' trade will flow across the seamless non-territory between states, if only states would remove regulations that presently hinder this trade. This imagery notwithstanding, historical research on economic development of countries in the global periphery has noted the importance of transpor- tation and communication infrastructures designed to export staples or commodities to the metropole in shaping the geography and the path of national development or underdevelopment (Innis, 1995). In 'staples' theory, the overcapacity and rigidities associated with infrastructure investments designed for connecting to the world economy contribute to a continued cycle of economic crisis that stifles diversity in economic growth. In fact, GII-GIS' dual mandate of fostering cyberspace both as a space of flows (mobility) and as a space of development (fixity) is fraught with contradictions. The US effectively has recognized this contradiction in GIl- GIS' dual mandate by funding construction of a second Internet system, the Next Generation Internet (NGI), a high-speed network devoted solely to non-commercial activities (MacKie-Mason and Varian, 1996; White House, 1996). By creating a second Internet, the original Internet (tellingly referred to as 'the commodity Internet') increasingly may lose its identity as a free space of flows where a lack of barriers maximises communication efficiency. Instead, the 'commodity Internet' can be reconstituted as a space wherein those making fixed investments may reap profits under the protection of state regulation while the NGI retains its friction-free purity. Over time, however, it seems likely that commercial interests will demandaccess to the higher-quality connectivity of the NGI, generating yet another round of tension within the fixity-mobility dialectic.