Use Cases for Bridging the Clickable Web and the Semantic Web

As per my assigned action item:
http://www.w3.org/2004/09/07-rdfhtml-irc#T14-03-44

I've come up with a handful of simple use cases that motivate bridging 
the Clickable and Semantic Web. These use cases assume:

A) A URL-based document may contain semantic metadata, thus a URL *may* 
serve both as as an RDF object and a Clickable target.

B) human web browsers are not likely to interpret metadata into 
human-readable information in the near future. Special case browsers or 
plugins might. Thus, content authors will likely continue to express 
metadata in human-visible XHTML for quite some time.

C) there is significant *implied* metadata in a number of Clickable 
links on the web today. It would be very useful to leverage these 
existing links into semantic web statements, where applicable.

Keeping these in mind, here are the use cases:

1) Author Information

An XHTML document "Life as an MIT Graduate Student" is authored by "Ben 
Adida."
Human-readable information about "Ben Adida" can be found at the 
document located at http://ben.adida.net, which itself contains, in 
XHTML MetaInformation, the RDF entity "Ben Adida" including properties 
such as age, occupation, and more (see (3) for more).

The document "Life as an MIT Graduate Student" contains, in visible 
markup, a byline:

---------
written by <a href="http://ben.adida.net">Ben Adida</a>
---------

2) Licensing Information

The XHTML document "Life as an MIT Graduate Student" is licensed under 
a Creative Commons license called "Attribution-ShareAlike 2.0." This 
license is described in the human-readable document located at 
http://creativecommons.org/licenses/by-sa/2.0/ which itself contains, 
in XHTML MetaInformation, the RDF entity "Attribution-ShareAlike 2.0" 
and the properties of this license, including the requirement to 
provide attribution, the right to redistribute, etc...

The document "Life as an MIT Graduate Student" contains, in visible 
markup, a graphic of the Creative Commons logo and a human-readable 
indicator of the license:

-------
<a href="http://creativecommons.org/licenses/by-sa/2.0/">
<img src="http://creativecommons.org/somerights20.gif" />
</a><br />
This document is licensed under a <a 
href="http://creativecommons.org/licenses/by-sa/2.0/">Creative Commons 
License</a> which, among other things, requires that you provide 
attribution to the author, <a href="http://ben.adida.net">Ben 
Adida</a>.
-------

3) FOAF

The XHTML document located at http://ben.adida.net describes "Ben 
Adida" in a human-readable way, including friends of Ben Adida's listed 
in an HTML unnumbered list, with links to each friend's respective 
homepage. At the same time, this XHTML document contains XHTML 
MetaInformation expressing FOAF relationships with each of these 
friends, whose equivalent RDF objects happen to be expressed within 
their XHTML homepage.

The visible markup looks something like:

--------
My friends are:
<ul>
<li> <a href="http://www.w3.org/People/Connolly/">Dan Connolly</a></li>
<li> <a href="http://www.w3.org/People/Swick/">Ralph Swick</a></li>
</ul>
--------


All of these are quite similar in their central feature: the 
user-visible markup and the semantic information overlap. This is 
certainly not always the case. In example (1), it's conceivable that 
"Ben Adida" might have:
	- a user-browsable homepage at http://ben.adida.net/
	- an RDF/XML description at http://ben.adida.net/profile.rdf

However, if we begin to assume that all XHTML will evolve towards 
having metadata, then we should plan for the possibility that the 
user-clickable page and the RDF entity *can be* at the same URL. This 
is critical in order to leverage the existing web.

Adopting a strawman notation for clarity's sake, I will use the 
fictitious HREL attribute as a modifier on HREF, the same way REL 
modifies RESOURCE. The only purpose is to demonstrate what this would 
look like, approximately:

(1)
---------
written by <a hrel="http://purl.org/dc/elements/1.1/creator"
href="http://ben.adida.net">Ben Adida</a>
---------

(2)
Here we have two visible clickable links. We only modify the second 
one, because that is the one that seems more closely related to the 
semantics of what the user is seeing:

-------
<a href="http://creativecommons.org/licenses/by-sa/2.0/">
<img src="http://creativecommons.org/somerights20.gif" />
</a><br />
This document is licensed under a <a 
hrel="http://creativecommons.org/rdf/license" 
href="http://creativecommons.org/licenses/by-sa/2.0/">Creative Commons 
License</a> which, among other things, requires that you provide 
attribution to the author, <a href="http://ben.adida.net">Ben 
Adida</a>.
-------

(3)
--------
My friends are:
<ul>
<li> <a hrel="http://xmlns.com/foaf/0.1/knows" 
href="http://www.w3.org/People/Connolly/">Dan Connolly</a></li>
<li> <a hrel="http://xmlns.com/foaf/0.1/knows" 
href="http://www.w3.org/People/Swick/">Ralph Swick</a></li>
</ul>
--------


So, the big question, of course: what do we need to do to support this 
bridging of clickable and semantic links?

-Ben Adida
ben@mit.edu

Received on Wednesday, 8 September 2004 23:06:02 UTC