Thoughts: some DOM some general

Personal opinions from another relative DOM novice:

>I have always found Dynamic HTML to be a confusing term.

I don't think the DOM can do anything about this. Contact the HTML
committees?

>settle on a compressed transmission protocol

I think that's outside the scope of the DOM, which only deals with how to
manipulate the document while it's in memory. (And the DOM provides only
the essential functions; applications will either extend the DOM or wrap
other code around it to add the behaviors they need.)

>what are the procedures for registering that a DOM implementation
>matches the spec?).

It _would_ be nice if there was a publicly-available regression test;
that's a good suggestion. Anyone have one they want to contribute, or is
the DOM still sufficiently in flux that it's premature to attempt one?

>I have always wanted hypertext to have a built in conditional door

Sounds to me like you want something approaching a MUD environment. I'm
sure such an application could be built on top of the DOM -- but in fact
the DOM doesn't even address hypertext, beyond its ability to store HTML.
Even following normal HTML links is the application's responsibility rather
than the DOM's.

>It should be possible to request a graph summary of the structure
>of a document

It's not to write a tool which walks the DOM and generates this. In fact,
one implementation of the DOM that I've worked on also implemented Java's
TreeNode interface, so it could be directly displayed in a JTree; that only
takes a few hours of hacking to bring up a basic version. But display is
not something the DOM itself should deal with; as far as the DOM is
concerned, it _is_ the graph.

Received on Sunday, 2 August 1998 02:20:42 UTC