Re: ConNegging on Description Pages (was RE: Southampton Pub data as linked open data)

Hi, Tom --

* Tom Heath [7/30/08 9:19 AM +0100] wrote:
> If I've understood you correctly, you're suggesting that the HTML
> document about a pub should 303 to the RDF/XML document about the
> pub if RDF/XML is requested, and vice versa. (please correct me
> if I've misunderstood)

The above is correct -- if I ask for RDF/XML and the server cannot
provide that but *can* provide HTML, it should redirect me (303) to
the HTML (which I may decide I don't want to retrieve!).  If I ask
for HTML and the server cannot provide that but *can* provide RDF
(whether RDF/XML or Turtle/N3 or ...), it should likewise redirect
(303) to that alternative (which, again, I may not pursue).

I should never get a `200 OK` delivering a document format not in
my Accept: header.  This really applies whether or not you, as the
page author, have control over the web server.  If it cannot provide
the content form requested by a client, the server should *always*
say so, offering whatever other form(s) it might know about as
alternatives via 303.

(And of course, the information in all 303-associated formats should
be the same, though it be presented differently.)


> Is conneging on description pages desirable? i.e. if I request
> http://revyu.com/people/tom/about/rdf
> in my regular browser (e.g. vanilla Firefox), should I be
> redirected to http://revyu.com/people/tom/about/html ?

Yes!  Because *Firefox* wants HTML.

Firefox doesn't generically know how to handle application/rdf+xml,
nor application/x-turtle, nor application/turtle, and really only
pretends it knows how to handle text/rdf+n3...

It's important to use the right tool for the job.  I can't very
well drive screws with a hammer, nor nails with a screwdriver.
Similarly, an RDF browser doesn't do well on HTML; an HTML browser
doesn't do well on RDF.


> I don't think so, for the simple reason that I might be a developer
> wishing to study/debug the RDF.

If I'm a developer wishing to study/debug the RDF, I should use
a tool which explicitly requests the RDF serialization(s) I want
to study -- which might be Turtle/N3, or RDF/XML, or... -- or
at least indicates Accept: *.

(Note -- this tool *might* be an *extended* Firefox, which *might*
be able to handle RDF/XML, in which case that should be included
in the Accept: header issued for relevant requests.  You specified
vanilla Firefox in your example, so that's what I pursued.)

Be seeing you,

Ted





-- 
A: Yes.                      http://www.guckes.net/faq/attribution.html
| Q: Are you sure?
| | A: Because it reverses the logical flow of conversation.
| | | Q: Why is top posting frowned upon?

Ted Thibodeau, Jr.           //               voice +1-781-273-0900 x32
Evangelism & Support         //        mailto:tthibodeau@openlinksw.com
OpenLink Software, Inc.      //              http://www.openlinksw.com/
                                 http://www.openlinksw.com/weblogs/uda/
OpenLink Blogs              http://www.openlinksw.com/weblogs/virtuoso/
                               http://www.openlinksw.com/blog/~kidehen/
    Universal Data Access and Virtual Database Technology Providers

Received on Wednesday, 30 July 2008 17:26:00 UTC