Simple explanation of difference between XLink and XPointer ?

[Daniel B. Austin]

James,

    My attempt to provide answers to your questions. Hopefully others will
correct my mistakes.

-----Original Message-----
From: James H. Blackwell Jr. <jim.blackwell at gsfc.nasa.gov>
To: XML Dev <xml-dev at ic.ac.uk>
Date: Tuesday, July 07, 1998 7:53 AM
Subject: Simple explanation of difference between XLink and XPointer ?


>Hi all,
>
>Can anyone out there give me a simple explanation of the differences
>between XLink and XPointer ?  Also, how do they relate to XML-LINK ?

    XML-LINK and XLINK refer to the same XML dialect. The name is
still being finalized by the WG.
    XLINK and XPOINTER are different however. While your distinction below
is correct in that XPOINTER refers to specific locations in documents, XLINK
is not confined to documents but can be used to point to any resource
whatsoever.
    XPOINTER can be conceived as a logical extension to the familiar
'named anchor' syntax used in HTML documents. Instead of the author being
required to manually place anchors within the document at the desired
reference point, the structure of the document, which must be known to some
extent beforehand, is used to locate the reference. Thus we can locate the
third subheading of the second section of the first chapter of the XML
specification easily, merely by knowing that the document's structure
follows rules which require that chapters have sections and sections must
contain
subheadings. (The XML spec contains no chapters - this is only intended as
an example.) We add the reference to this document element to the end of the
URI for the document in much the same way that named anchors are currently
done.
    XLINK is more complex and difficult to imagine in execution because it
does not resemble closely the current hypertext linking model with which we
are all familiar. The common hypertext link is the simplest case of an
XLINK. More complex link structures can be developed that serve purposes
beyond one-way resource links with XLINK. A multiple ended link is a link
that points to more than one resource - perhaps a reference in a paper that
points to several documents that are used as reference material. This might
appear as a drop down menu that is displayed when the user moves the mouse
over the linked text in the document, offering several destination choices.
Links can also be made bidirectional (or multidirectional) in the sense that
the link can be traversed in both directions, from document A to documents
B,C, & D, and also from B, C, & D back to A. Reciprocal links of this nature
provide a contextual net for the information that provides greater control
over the document's use and better linking for the user. Also with a
bidirectional link, the well known and despised error 404 will cease to
occur, since the link cannot exist unless the correct resource is available
at both ends.
    It's difficult to provide examples of either of these at this time; the
concrete syntax and use are still being determined by the WG.

Regards,

D-

****************************************************************************
****
Daniel Austin, Director of Development, Creative Services, CNET
daniela at cnet.com    415-395-7800 x1438
"To change the old into the new, and the shapes of things to come..."



-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3065 bytes
Desc: not available
Url : http://mailman.ic.ac.uk/pipermail/xml-dev/attachments/19980707/94d905ae/smime.bin