Hi Juan,
On 15 Jul 2009, at 15:43, Juan Sequeda wrote:
and the objective is not to start another long philosophical
thread :P and
it may be a very dumb question
What are the drawbacks of this simple solution.
in PHP for example:
if($_SERVER['HTTP_ACCEPT'] == "application/rdf+xml" ){
header('Content-type: application/rdf+xml');
echo "......."
}
else{
echo "...."
}
A typical Accept header sent by an RDF client can look like this:
text/html;q=0.3, application/xhtml+xml;q=0.3, text/plain;q=0.1, text/
rdf+n3, text/n3, application/n3, application/x-turtle, application/
turtle, text/turtle, application/rdf+xml, text/rdf, text/rdf+xml,
application/rdf, application/xml;q=0.2, text/xml;q=0.2
A typical Accept header sent by a Web browser can look like this:
application/xml,application/xhtml+xml,text/html;q=0.9,text/
plain;q=0.8,image/png,*/*;q=0.5
As you can see, your code above will not work. You really need a
proper implementation of content negotiation rather than such
simplistic hacks.
Vapour just checks some very simple cases. A green light from Vapour
does not necessarily mean that your content negotiation works with any
real client. (Unfortunately! I wish it was better at validating conneg.)
(The former header is from the any23 library in default configuration.
The latter is from Safari. Note that modern RDF clients can consume
several RDF syntaxes, including RDFa.)
About the "/id/" vs. "/id" thing. This behaviour is more or less
hardcoded in Apache and similar web servers. After you follow the 301
redirect at "/id", you get the same 200 at "/id/". Essentially this
means that "/id/" and "/id" identify the same resource, and that
resource is a document.
Best,
Richard
I did this at http://www.juansequeda.com/id/
However, there is a difference when it is /id and /id/. When I
dereference
http://www.juansequeda.com/id I get a 301 (Moved Permanently) but with
http://www.juansequeda.com/id/ I get 200 (and everything validated by
Vapour!).
As this ever been discussed? I can obviously see the drawback of
having /id/
vs /id .
Comments?
Juan Sequeda, Ph.D Student
Dept. of Computer Sciences
The University of Texas at Austin
www.juansequeda.com
www.semanticwebaustin.org
