Noah also cleared up some confusion I had about the term "hypermedia" -- in Nelson's terminology, hypermedia is not just multimedia, but multimedia the user can manipulate interactively (my paraphrase). So an online dissection kit is hypermedia, a Quicktime video clip that you simply play is not. Most of what I've heard people refer to as hypermedia is not hypermedia. Nelson's hyper derives from mathematical use of the prefix. i.e. the OED's third definition, hyper involving extension:
3. a. In various terms of modern Mathematics, as hyperconic, hypercycle, etc. (see below); esp. in adjectives applied to functions, etc., related to or resembling those denoted by the simple adjectives, but involving some extension or complication, as hyper-complex, -elliptic, -geometric (-ical), -jacobian, -spherical. See also HYPERDETERMINANT.
While I'm in the OED, I do note that its editors do credit Nelson 1965 T. H. NELSON in Proc. 20th Nat. Conf. Assoc. Computing Machinery with the first use of the term (Noah expressed frustration that more people didn't give Ted the street cred. he deserves).
Noah pointed out that the link and node model of hypertext widely understood by Web browsers was only one of several models Nelson envisioned. The OED's last citation seems to have gone there as well. 1991 Whole Earth Rev. Summer 59/2 Links connect nodes in the hypertext software by computer-supported relationships that permit rapid, easy movement across the network of nodes.
Congrats to Noah and Nick Montfort on the release of the New Media Reader. It looks great, though I still don't have my desk copy.
By the way, a hyperbaton is a figure of speech in which the customary or logical order of words or phrases is inverted, esp. for the sake of emphasis. 400 years older than hypertexts are hyperbatons.