Kannan Thiruvenagadam makes an excellent point here: >Proceeding on these lines, one can conclude that placing >loooong (educational) documents on the web is no way to *use* it. >Break the information into smaller units and make them >the leaves of a tree. Each internal node doesn't have >to be a menu (like in gopher). It should be an independent knowledge >unit (which makes sense by itself). Shortness of the units >also means the unit will be over before the reader >loses interest. This is good advice! Kannan is also correct that "depth on demand is what the WWW supports." When I said that people tend not to read for depth on the Web, I was just (clumsily) reiterating my point that people tend not to read long text documents online.