Nov 21 2007
Isn’t The Web Built From Links?
If my shadow web post hasn’t convinced you then try this thought experiment:
You want to link from your webpage to Tim Berners-Lee’s URI <http://www.w3.org/People/Berners-Lee/card#i>, except you can’t because that link points to something that can never contain the #i fragment in its HTML. It can only ever link to RDF because Tim is relying on RDF’s semantics for the meaning of the #i fragment. Tough luck if you can’t read RDF or don’t want to have to learn.
6 Responses to “Isn’t The Web Built From Links?”
Now try this actual experiment. Get the tabulator extension and try the same thing.
I don’t want *you* to learn RDF. I want your browser to!
Sorry the tab’r is very alpha. The online script version (development trunk) is more advanced in several ways.
Man, what crappy kind of browser do you use? When I click that link, I get a nice mug shot of the man and all sorts of interesting information. His assistant is cute!
C’mon, Firefox and the Tabulator extension. All the cool kids use it!
I’m looking forward to your post that explains how getting everyone to upgrade their browser is impossible. That’s why CSS is never going to work and you have to design for Netscape 4.
Richard: my link was to the fragment on that page which cannot exist in an HTML representation
Tim: the point is that URI links to a specific representation format (i.e. RDF) because of the fragment. That’s ok if we only ever want the semantic web to consist of RDF representations. However, I’m quite sure we will want to be able to cross-relate semantics from all kinds of representation formats in the future. URIs with fragments are an impediment to that IMHO. Potentially interesting non-RDF formats that come to mind are RDDL and Topic Maps.
Ah cool, now I understand what the tabulator extension is for. Nice.
But I understand Ian’s point. The idea is that in the rdf representation
http://www.w3.org/People/Berners-Lee/card#i
denotes Tim, whereas for the html representation, if there were one, that URL would denote apparently a portion of an html page. Does the html spec really say that though? Does it not just specify some behavior a typical html web agent should display, namely to display that portion of the text that is located close to that id? In which case Tim could also create an html representation for his rdf and have a section with id “i” in it, which would lead to the correct behavior.
This is similar to the sentence
“Tim is at MIT”
We can say that “Tim” above is the first word in that sentence, and that Tim refers to a real person. In html the #i works in a syntactic fashion, and so can be used to identify a position in the html.
And if this is not the case then perhaps this just indicates that something needs fixing in the definition of html. I suppose the RDFa folks are onto this. There you clearly can use the #i to refer to Tim, and you can indicate a piece of the html as being the one preferred piece that people should go to if they are interested in him.
The example is a good one that should help clarify a problem in the specs somewhere.
Henry: have a read of Tim’s message at http://lists.w3.org/Archives/Public/www-tag/2007Oct/0140.html for more clarity on this. His view is very relevant to this problem: “allow the same fragid to be used as an ID for an anchor and an ID for a thing, with RDF clients and HTML clienst doing different things. I think that this path leads to madness”