The critique is a bit unfair because the Internet had clear design goals and criteria with which to apply rigorous engineering practice. The web by contrast was a essentially a document sharing and linking protocol and document format. It looks bad mostly because it outgrew it's design, but at the same time if it had been designed correctly for its eventual use it probably never would have taken off to begin with. Granted, it has fundamental warts as well, but I ascribe to that to the relative uncertainties of designing a higher level application that has at once more constraints and unanticipated usage.
I don't know, perhaps Kay is right, being that he is 1000 times the engineer I am, but the web just seems such an amorphous thing compared to TCP/IP.
I don't know, perhaps Kay is right, being that he is 1000 times the engineer I am, but the web just seems such an amorphous thing compared to TCP/IP.