[1] From: John Unsworth <jmu2m@virginia.edu> (32)
Subject: Re: 10.0712 TACT problem? Cartesian questions?
[2] From: Lou Burnard <lou@vax.ox.ac.uk> (28)
Subject: RE: 10.0712 TACT problem? Cartesian questions?
--[1]----------------------------------------------------------------
Date: Tue, 18 Feb 1997 11:47:41 -0500
From: John Unsworth <jmu2m@virginia.edu>
Subject: Re: 10.0712 TACT problem? Cartesian questions?
Leaving aside the problem that the browser would have to be modified
to read these coordinates, I think there's a larger problem here, namely
that the x,y coordinates would change if the document were resized by
the browser. It would seem to me that the only reliable way to specify
this kind of location would be with a different set of coordinates that
referred to the beginning and ending characters, specified as an offset
number counted from the beginning of the file (counting tags? not counting
tags? Dunno).
John Unsworth / Director, IATH / Dept. of English
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
http://jefferson.village.virginia.edu/~jmu2m/
--[2]----------------------------------------------------------------
Date: Tue, 18 Feb 1997 12:43:18 +0000
From: Lou Burnard <lou@vax.ox.ac.uk>
Subject: RE: 10.0712 TACT problem? Cartesian questions?
Re your strawberry-eating correspondents'
|We concluded that every spot in a document is locatable but may not be
|marked. This was arrived at by abstracting the case of time code on a
|video tape and cursor position in a text file to the general case of
|coordinates. So we asked ourselves how could the fragment identifier <NAME>
|attribute of the <A> element in HTML be used to find an unmarked-up
|spot in a html file.
|
| <A href="url#y,x">...</A>
|
|y = vertical position or line number
|, = separator
|x = horizontal position or character number
|
How profoundly mistaken is this linearity! Consider: a video is made up of
frames, each of which is a complete 2-d image. An image exists in a Cartesian
space: if you address its space, you address the image. That all makes perfect
sense.
But texts *are not* images! Texts (even those on the Web) are made up of
non-visual signifiers (words, to you and me) whose signification lies in far
more than their character position and line number -- which are mere
accidentals of the way the page happens to be rendered. Consider what happens
when the recipient of your web page decides to use a different font to display
it. Or even when the window in which it is displayed gets resized.
No, if you want "all-points-addressability", the only reliable method
is to mark-up explicitly the addressable components of your document. Use
SGML.
-------
Lou