home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: alt.hypertext
- Path: sparky!uunet!wupost!spool.mu.edu!agate!linus!alliant!mydual!olson
- From: olson@mydual.uucp (Kirtland H. Olson)
- Subject: Re: Definition of Hypertext ?
- Message-ID: <1992Dec25.143905.9672@mydual.uucp>
- Organization: The Harvard Group, 01451-0667
- References: <6761@tuegate.tue.nl>
- Date: Fri, 25 Dec 1992 14:39:05 GMT
- Lines: 89
-
- In article <6761@tuegate.tue.nl> toin@blade.stack.urc.tue.nl (Toin Bloo) writes:
- >I know only a little of hypertext, but I need to use a kind of definition
- >of it in a report about "electronic script".
- >Can someone post a nice and sound definition of the "idea" of hypertext?
- >
- >Thanks in advance!
- >--
- >Toin Bloo member of: The Young Scientists Eindhoven
- >+31 40 622681 (home) Frederiklaan 163 5616 NE Eindhoven
- >toin@stack.urc.tue.nl The Netherlands, Europe
- > (toin%storc@dutentb.et.tudelft.nl)
-
- "By 'hypertext' I mean non-sequential writing." --Theodor H. Nelson
-
- "_Basic_or_chunk_style_hypertext_ offers choices, either as
- footnote-markers (like asterisks) or labels at the end of a chunk.
- Whatever you point at then comes to the screen."
-
- "'As We May Think' (_Atlantic_Monthly_, July, 1945) is most notable for
- its clear description of various hypertext techniques--that is linkages
- between documents which may be brought rapidly to the screen according
- to their linkages. (So what if he [Vannevar Bush] thought they'd be on
- microfilm.)"
-
- Quotations from Dream Machines, pp 44-5, copr. 1974, Theodor H. Nelson
-
- Although Nelson called it non-sequential writing, I find it easier to
- understand as multi-sequential writing--there are many orders available
- and I may follow one until another seems more attractive.
-
- I've posted the following material before, but see the section
- HYPER[whatever]--a structure of linked nodes.
-
- I think these are the definitions I actually use in thinking about
- hypertext and hypermedia. My internal picture of the structure is a
- finite state diagram in which each node represents one state of the
- system labelled by the collection of values that uniquely address that
- state no matter what the incoming path. In my mind the nodes are
- arranged so that each is adjacent to those whose state labels differ by
- exactly one unit in the particular variable.
-
-
- NODE--the end of a link.
-
- Just the end of a link, no notion of content or capability.
- Sort of like an airport--maybe nice, maybe great, maybe just a dirt
- strip. Node names a class of things that occur at the end of links.
-
- LINK--a transition between nodes.
-
- Although links have some physical representation for the engine,
- to the reader links equal action. Link names a class of actions that move
- the reader in the structure. I could as well think of this as
- 'changenode' without distrurbing my picture of the process.
-
-
- HYPER[whatever]--a structure of linked nodes.
-
- Often the [whatever] attempts to distinguish node content in
- some gross and approximate way. Hyper[text] may mean text only,
- anything that can be shown on a text-only display (including character
- graphics) or anything that doesn't move or make noise--except maybe
- sound effects or ... You get the picture. Similarly full motion video
- plus text might be hyper[media] as well as full fidelity audio with
- photorealistic slides. But all these share the concept of a structure
- within which one can traverse links to arrive at nodes.
-
-
- HYPER[whatever] ENGINE--software to implement the structure on a
- platform.
-
- HYPER[whatever] TOOLS--software to control the engine.
-
- I know the reasoning seems circular (I prefer to think of it as self-consistent)
- but this is the most complex structure into which I can fit the systems
- I know of. Any attempt to make more specific definitions gets me in
- trouble with one or more hyper[whatever] systems. For example, HyperRez
- does not separate nodes that contain only a list of choices from those
- that contain no further choices at all--both are files, addressed in the
- same way, equally visible to the reader--they just don't differ. Other
- systems seem to keep those meta-nodes in separate databases, but I reach
- them through a link. Thus I can't fit both into one node class more specific
- than "the end of a link."
-
- Similarly, if I were to say links connect two nodes, the systems that
- use AI to alter the links on the fly will not fit.
-
- --
- Kirtland H. Olson olson%mydual.uucp@alliant.com
-