Re: <a footnote="proposal">

Foteos Macrides (MACRIDES@SCI.WFBR.EDU)
Tue, 15 Oct 1996 20:22:48 -0500 (EST)


Date: Tue, 15 Oct 1996 20:22:48 -0500 (EST)
From: Foteos Macrides <MACRIDES@SCI.WFBR.EDU>
Subject: Re: <a footnote="proposal">
To: murray@spyglass.com
Cc: www-html@w3.org
Message-id: <01IAOMM0LI9E00BRTK@SCI.WFBR.EDU>

murray@spyglass.com (Murray Altheim)
>I got in a disagreement several two years ago with Dan Connolly over
>head and tail. He corrected me, as apparently there is some history
>in the lex/parsing terminology (or somewhere) that describes them about
>*exactly* the opposite of the way I thought it would work. I can't remember
>his reference, but it seems you click on a tail and it locates the head.
>Yup. Weird. Don't try this with snakes or mollusks.
>
>From HTML 2.0 (RFC 1866):
>
>    hyperlink
>            a relationship between two anchors, called the head and
>            the tail. The link goes from the tail to the head. The
>            head and tail are also known as destination and source,
>            respectively.
>
>I've just avoided the terms entirely, as they give me headache,
>particularly since from my way of thinking (which I admit is possibly
>flawed) this seems counterintuitive.

	Here's my take on this headache.  The HREF always is the head,
according to both the HTML 2.0 RFC and the HTML 3.0 draft.  For a
LINK, the tail is the entire document whose HEAD contains that LINK.
For an Anchor, the tail is its content (and if you forget the </A>,
it's the rest of the BODY 8-).

	Unfortunately, both the HTML 2.0 RFC and the HTML 3.0 draft
then attempt to explain REL and REV via references to A pointing to
B versus B pointing to A, instead of references to head or tail, or
destination or source, so you need to know already what they intend
to say to understand what they're saying.

	One can infer, however, from:

	<LINK HREF="mailto:author@host" REV="made">
	<LINK HREF="contents.html" REL="ToC">
	
that REV associates the head to the tail (author made this_document)
  so REL associates the tail to the head (this_document's ToC is at
  					  destination content.html)

 and <A HREF="contents.html">Table of Contents</A> has an implied
        REL="ToC"
	
Thus, <A HREF="footnote.html" REL="footnote"><SUP>1</SUP></A>

seems correct, and a display engine which recognizes "footnote" as
a REL naming convention shouldn't need the SUP, and if it's a GUI
should fetch footnote.html and render it as a "popup" or "temporary
frame" superimposed on the current display for the current document,
that disappears when you hit the "back button".

	Now all you need is that naming convention, and some display
engine maker to make a display engine which does that. :) :)


	For:
	
	<A HREF="#one"><SUB>1</SUP></A>
	
where the document contains, among other things, this head for that
hyperlink:

	<FN ID="one">This is the first footnote.</FN>
	
what you need is some display engine maker to make a display engine
which hides FN content (be nice if it hid SCRIPT and STYLE content,
too, as well as the content of any OBJECT with the DECLARE attribute)
and on activation of the 1 tail rendered the FN content as a "popup"
or "temporary frame" superinposed on the current display for the
current document, that disappears when you hit the "back button".
But at the very least, the display engine could display the FN
content where it appears in the document, but scroll to it as a
head when the 1 tail is activated, then scroll back to the tail
when the "back button" is activated. :) :)

				Fote

=========================================================================
 Foteos Macrides            Worcester Foundation for Biomedical Research
 MACRIDES@SCI.WFBR.EDU         222 Maple Avenue, Shrewsbury, MA 01545
=========================================================================