Re: HTML 2.0 specification

David C. Martin (
Tue, 6 Sep 94 13:46:45 EDT

The question of LINK and GROUP is one of semantics, not syntax. We
chose to implement the elements such that LINK elements are only
allowable in the header of a document and describe the linkages to
other documents (both static and dynamic). For example:

<GROUP ID="Issue">
<LINK ROLE="partof" HREF="http://.../NEJM09011994/">
<LINK ROLE="partof" HREF="http://.../NEJM09081994/">

while in the header of a document the link tag could be explicit, e.g.:

<LINK ROLE="Next" HREF=""http://.../NEJM09011994/">

with the explicit LINK elements in the header considered static
structure of the document and the GROUP encapsulated LINK elements
considered as dynamic and not part of the document (ala HyTime ILINK).

We have modified Mosaic to have a set of buttons for the static LINK
elements and a dynamic menu for the dynamic LINK elements, with the
buttons greyed for non-specified LINKs and the menu built on the
encapsulating GROUP elements.

Our canonical example is a set of articles that have some static
structure, represented in the static LINK elements, as well as some
dynamic structure, e.g. a list of "hits" from a DB search. This would
allow the user to read the article using the static LINKs and also go to
the next, previous, etc... articles in the hit list.

Our implementation of EMBED, as documented in our paper "Integrated
Control of Distributed Volume Visualization Through the World-Wide-Web,"
to appear in the proceedings of IEEE Visualization 1994, is as follows:


<EMBED HREF="http://.../volume.hdf"

The EMBED element arranges for the execution of a external application,
based on the MIME content-type, to take the data specified in the HREF
and perform its actions appropriately. The browser allocates a window
in the document for the application's output (400x400 requested) and
maintains a dialog w/ the external application, informing it of state
changes (e.g. change page).

I would like to discuss more with you how we can work together to get
all three ELEMENTS defined and implemented for the next specification.



Assistant Director			mail:
UCSF Library & Ctr for Knowledge Mgt	at&t: 415/476-6111
530 Parnassus Avenue, Box 0840		 fax: 415/476-4653
San Francisco, California 94143		page: 415/719-4846
Dave Raggett writes:


> We have implemented several features from HTML+ that we would like to > discuss having in the HTML 2.0 specification. Specifically, we have > implemented the EMBED, GROUP and LINK elements.

Its too late to put these into the 2.0 spec, particularly as the spec aims to rationalize the de facto situation *NOT* future enhancements. However, I am the editor for the 3.0 draft spec which the group will start working on once the 2.0 spec is signed off. The plan is to base the HTML 3.0 spec largely on HTML+, and I am currently working with CERN to build a public domain X11 browser for HTML+/HTML 3.0 to support this process. This browser will be made available later this Fall.

The EMBED and GROUP elements refer to an older version of the HTML+ spec. I backed off <embed> for technical reasons and now am following a path which generalises inlined material by using OLE/OpenDoc style application embedding in HTML documents. Can you tell me more about how you are using these elements in your implementation?

You can read an overview of the latest state of the HTML+ spec, as presented at WWW'94 in:

The formal document type definition is at:

> Please let me know how we can participate in the discussion.

The working group is open to all people interested in participating.

Best wishes,

Dave Raggett

----------------------------------------------------------------------------- Hewlett Packard Laboratories email: Filton Road tel: +44 272 228046 Stoke Gifford fax: +44 272 228003 Bristol BS12 6QZ United Kingdom