To the editor, New Scientist:
Fax no. 020 7611 1250
In your unsigned piece "Hypertext" on 17 June (p.60), you say, "Nelson coined the word hypertext in 1963 while working on ways to make computers more accessible at Brown University in Providence, Rhode Island." Something must have gotten lost in your word processor.
In 1963 I was a dolphin photographer in Miami, nowhere near Brown. (I had become inflamed with ideas and designs for non-sequential literature and media in 1960, my second year in graduate school, but no one would back them, then or now.) Not until years later, in the late sixties, did I spend months at Brown, with no official position and at considerable personal expense, to help them build a hypertext system. That project dumbed down hypertext to one-way, embedded, non-overlapping links. That project's broken and deficient model of hypertext became by turns the structure of NoteCards, HyperCard, the World Wide Web, and XML.
I believe humanity went down the wrong path because of that project at Brown. I greatly regret my part in that project, and that I did not fight for deeper constructs. At the time I thought of that structure as an interim model, forgetting the old slogan: Nothing endures like the temporary.
A paragraph later you talk about XML. Hooray indeed for a system "which describes information in a way computers can read. It means that with the right software, you can choose which information to make use of and how to process it. The implications are enormous." Indeed so, but those implications have been recognized as enormous since the computer field began. You could "describe information in a way computers can read" and "choose which information to make use of and how to process it" in BINARY, for goodness' sake, and hundreds of methods since. XML is only the latest, most publicized, and in my view most wrongful system that fits this description-- wrongful because it is based on one-way, embedded, non-overlapping links, continuing on the wrong pathway that humanity has gone down ever since that project at Brown. Opaque to the laypersons who deserve deep command of electronic literature and media, XML gratuitously imposes hierarchy and sequence wherever it can, and is very poor at representing overlap, parallel cross-connection, and other vital non-hierarchical media structures that some people do not wish to recognize or honor.
It is ironic to hear you say that Doug Engelbart and I invented "hyperlinks," meaning the 1-way embedded non-overlapping links of the Web, when both of us still are fighting for deeper forms of connection that we envisoned at the beginning.
Theodor Holm Nelson
Oxford Internet Institute
Personal email: firstname.lastname@example.org
Home telephone number: 01865 200 054