The use of broadband co-axial cable as a medium for data communications is a very attractive proposition. The "institutional" cables, available principally in the larger cities, are natural candidates for this service, and even those business users who have access only to "entertainment" cables will find that, in most cases, a sufficient amount of unused bandwidth is available to carry a large number of data channels. However, the implementation of these services has proceded in a most desultory fashion, and one technical reason for this is the lack of precise knowledge, both theoretical and empirical, of those characteristics of the broadband medium which are likely to affect the transmission of data and which may cause reciprocal effects upon video signals.
While it is not possible to exactly characterise all cable systems, bearing in mind the diversity of topography and gross performance specifications, some general theoretical and practical observations can be made here which may smooth the way for a more enlightened approach to this application of cable TV networks.