kdmoen@watcgl.UUCP (11/05/87)
In article <299@PT.CS.CMU.EDU> ns@CAT.CMU.EDU (Nicholas Spies) writes: >Not to slight V. Bush for his theoretical contributions, until Ted Nelson >they were largely just that. <amusing anecdote deleted> >The moral? Ted Nelson ALMOST invented vaporware, too! Speaking of vapourware, has anybody heard any recent news about project Xanadu? The last time I saw Nelson (July 1985), he was losing hope of finding a large industrial sponsor for implementing Xanadu on a mainframe, and was making plans for implementing "personal Xanadu for the IBM PC". -- Doug Moen University of Waterloo Computer Graphics Lab UUCP: {ihnp4,watmath}!watcgl!kdmoen INTERNET: kdmoen@cgl.waterloo.edu
david@daisy.UUCP (David Schachter) (11/09/87)
In article <2259@watcgl.waterloo.edu> kdmoen@watcgl.waterloo.edu (Doug Moen) writes: >Speaking of vapourware, has anybody heard any recent news about >project Xanadu? The last time I saw Nelson (July 1985), he was losing >hope of finding a large industrial sponsor for implementing Xanadu >on a mainframe, and was making plans for implementing "personal Xanadu >for the IBM PC". I went to a talk at Stanford Univ. a few weeks ago, given by Ted Nelson. The following OPINION is an OPINION which is not fact, because it is OPINION. He is a con man. His original thought, "hypertext", was interesting a decade ago and is still interesting now. However, Xanadu, as he described in his talk, is merely a filesystem with links and facist user contracts. There is no search facility. There is little, if any, flexibility. According to Nelson, these are not deficiencies, they are attributes. His "project" is the creation of a database and he leaves the front-end software to customers. Xanadu is a system for storing files and for storing links between files, where a "link" is not just a file-to-file pointer but a byte-to-byte pointer. (Actually, the link connects a span of bytes in one file to a span of bytes in the other. This mitigates but certainly doesn't solve the update problem: what to do with a link when the pointer at one end becomes invalid?) I went to the lecture expecting to find some interesting implementation of hypertext or hypermedia. Instead, I shelled out $25 for a book which he will mail to me, someday. (Haven't gotten the book but did get the cancelled check back in my monthly bank statement.) He has a few programmers working part-time on the project. He has a "proprietary" technique for hashing (or some such drivel; his lack of specificity was excused with the "it's a trade secret" argument.) Xanadu won't succeed if it is only a bad database back-end package. What Nelson described in his talk was not a useful system. In my studied OPINION, which might very well be WRONG. Oh, the "facist" nature of Xanadu to which I allude above is in the nature of the contract one must sign to get material put into a Xanadu database. The contract, as described by Nelson, is quite restrictive. Nelson claims this is necessary because of the uncertain nature of copyright as applied to computerized databases. Perhaps. But commercial information providers don't seem to have such problems. -- David Schachter The OPINIONS expressed above are my own. They are not reflections of the policy of my company. They are probably random line noise, come to think of it. "This microphone amplifies all sound by a factor of one to the fourth power."
david@hpsmtc1.HP.COM (David Williams) (11/10/87)
I heard and talked to Ted last month at Stanford. Xanadu is alive and well. I believe they have it up on a Sun and will be porting it to the Mac II shortly. I'll let you know whats going on when I talk with him again, as to progress.
munck@linus.UUCP (11/13/87)
>The following OPINION is an OPINION which is not fact, because it is >OPINION. > >He is a con man. This is also OPINION: it's comforting to know that Ted hasn't changed over the 20 or so years since he hung around the Hypertext project at Brown. That project was the foundation of many things, including one of the first WYSIWYG editors (running on about $3 million worth of equipment per user), hypermedia and their current Intermedia, and a whole bunch of great students now doing graphics and other good stuff all over the place. No, hypertext came neither from Apple nor from Nelson, any more than the mouse came from the Mac or from Xerox or outliners from Thinktank. Heck, one of the facilities on the old Hypertext was the ability to put numbers in rows and columns and have the machine sum them. Then you could change an entry with the light- pen and the sums would change. I believe that Bricklin was in high school at the time.