I already have those links "hacked" into my documents, but they are not
part of my document and I don't want them to be printed, or saved, or
anything, other than be indicators that a client application could use
to provide navigational references.
Also, it is not a meta-document, it is a document. You can abstract the
notion of a document to be anything from an individual segment of HTML
to the entire Library of Congress and the system needs to support that
abstraction.
I am not going to hassle Marc or anyone at NCSA for this functionality;
I only offer our application of the technology as an example of the
types of hoops which people are going to put HTML/HTTP clients thru.
dcm
--------
"Peter Lister, Cranfield Computer Centre" writes:
> Well, everyone wants something different.
>
> What we want to be able to do is have information in the HTML header
> that indicates the structural aspects of an HTML document relative to
> some larger whole, e.g. pages w/in an article, w/in an issue, w/in a
> volume w/in a journal.
What you describe is different, right enough - it's not at all the
series of URLs which can be warped to directly from the Document/Manual
menu that I (and I think most others) have been discussing.
What you describe appears to me to be a meta-document - i.e. it
describes the main document, and this kind of thing can be created
easily in HTML. If I want it to look really spiffy, I can have inlined
icons of arrows, volumes, etc. I need only process the document once,
all users get the same view of the doc (given the limitations of their
viewer), and you don't have to hassle Marc A and the team to produce
run-time hackable menus (or do it yourself). Whether you prepend the
meta-doc to the real doc or not matter of taste.
PeterLister [email protected]
Computer Centre,
Cranfield Institute of Technology, Voice: +44 234 754200 ext 2828
Cranfield, Bedfordshire MK43 0AL UK Fax: +44 234 750875