What does that mean? A text includes a lot of questions. It's theese question that the user should be able to ask and get a good answer. An easy way to ask the question is to klick on the part of the text the user want explained. This could be done for the phrases in the text. But questions about the text as a whole should be stored separatly from the text, i.e. not explicitly at any specific position on the presentation screen.
What attributes should the link objects be capable of
All texts should be categorized in a number of dimensions. This categorization could be made both by the author, and through analysing the context of the use of the text, and automaticly through word-counts.
The question of creation and modification dates is complex if you want to compose presentation from many sources. Tha familys or classes or subparts of the document could have been changed. For exampel: the text could belong to a category that has some properties. Changes of theese properties could change how the text is presented, what it links to, and what you can do with the text. There could also be changes of only one scentence in the text.
It's going to be great!
The weaknes compared to spoken language or traditional books are that the hypermedia space under some conditions could degenerate rather than evolve. A continious growth of the stored information require a continious rearangement of the information structure. A book can be read as as singel object. A hypertext page must be read in its context of other pages.
I know what direction I would like hypermedia to go. But I don't know how long it's going to take. For one thing: I would like to replace my almost 50 cards in my wallet, to one transponder in a bracelet or in the hand or head.