AOIR 5.0: Keynote by Ted Nelson,
Listening to Ted Nelson and to discussions about his keynote made me realized once again that I'm not well connected to the field of internet research yet: I had no idea who he was :) Anyway, he was talking about "my" things, so I was able to connect easily.
The main Ted's thesis is that "today's computer world is based on techy misunderstandings of human life and human thought…" He suggests that current computing is based on a paper metaphor of a document and hierarchy as a way to organize documents.
Ted says that "documents are representations of human thoughts" and that they are supposed to facilitate travel of ideas from one person to another, at the moment or over time. I can't agree more with my "artifacts as knowledge representations" :) Ted argues that using paper metaphor for designing how computers work with documents doesn't fit the way we think and communicate as "paper is a prison that holds thought". He talks about parallel thinking instead of hierarchy.
Of course, Ted goes into discussing an alternative, talking about transliterature and showing a few prototypes (I'll dig out links once I get more time online). In brief it comes to "true hypertext", with bidirectional links and transclusions.
Although I share Ted's understanding of problems with current ways of organizing documents, I don't think that solutions that he envisions would work. He is criticizing tech-geeks developing interfaces too complex for a normal people, but he does the same in another dimension (I'd call him mind-geek ;) by proposing easy interfaces that require complex mental structures. For me it comes to the discussion with Adrian Miles in Lugano, where he admitted that not many people are able to appreciate "true hypertext" literature.
Ted suggests that he aims to develop a system that allows people to use computers to support their thinking and says that it would be flexible enough to accommodate as much structure as a user would want. Again, I agree with need for more flexibility, but I'm not sure how easy it will be for people to use: Word has enough features to be flexible, but most of the users use just a few, sticking to uses and metaphors they are familiar with.
One of Ted's demos allowed you to browse through multidimensional relations in a very nice way, but then (again :) those relations has to be defined by a user explicitly. I'd say there are at least two problems with it. First, explicit effort to define relation usually do not work (people do not add metadata). Second, most relations are not explicit, but fuzzy and multidimensional as well.
And, to finish with criticism, Ted haven't said much about social life of documents and how it could be reflected in the systems he proposes.
Despite all my disagreements I like it a lot. Ted is engaging and mind provoking speaker, so I'm going to look into his work more…
This post also appears on channel AOIR