Once upon a time, folks who made content for hypertext systems were called authors. Because they wrote (hyper)text. The move to ‘web design’ — as with the move to advertising-based revenue models and dynamic content — was fatal blow for the universe of serious hypertext we could have had, and occasional holdouts like wikipedia serve only to underline how fucked everything is.
The difference in feature sets between the web and xanadu bears this out.
The web has no permanent addresses and no guarantees that document contents are permanent. It has no transclusion support. It has no first-order versioning. It has only jump-links.
In 1984, the protagonist works at a department that modifies all archived copies of historical events and destroys the old versions.
The web not only supports but incentivizes a decentralized version of that.
There is no basic, built-in mechanism for making sure the thing you are linking to now is the same thing you linked to six months ago.
There’s no mechanism to reference a section of a page without copying and reformatting it, so there’s no proof you are not misquoting.
Clicking a link is like playing russian roulette. You cannot see what’s on the other side of the link before you click it. You cannot trivially keep both sides of the link in view in order to compare them.
Just as walking through a door causes us to drop some of our working memory and lose context, so does clicking a jump link.
(Adapted from a thread)