RE: Question on the boundaries of content negotiation in the context of the Web of Data

________________________________
From: Ian Davis [mailto:me@iandavis.com]
Sent: 13 February 2009 13:01
To: Williams, Stuart (HP Labs, Bristol)
Cc: Michael Hausenblas; www-tag@w3.org
Subject: Re: Question on the boundaries of content negotiation in the context of the Web of Data


On Fri, Feb 13, 2009 at 12:24 PM, Williams, Stuart (HP Labs, Bristol) <skw@hp.com<mailto:skw@hp.com>> wrote:

Sure... but we're punning here on the word document (representation or resource sense). The bits that came over the wire are awww:representation of a graph and of an image that either depict or describe a common subject (a house), but neither graph nor the picture are that subject - and neither are their awww:representations awww:representations of the same thing... one is an awww:representation of a graph whilst the other is of a picture.


Would you agree that by that logic an SVG representation that reproduces exactly the pixels of the image cannot be used in content negotiation for that resource? The SVG document is a representation of an xml infoset

Ian

:-)

Nope...

You'll see from other messages in the thread that I there is a mention of a graph using some appropriately grounded vocabulary to encode the same image - in which case I could be persuaded that (at least for the purpose of rendering an image to a user) a turtle representation and PNG representation could be equivalent. Point is there can me be many different descriptions/depictions of the same thing - that what they describe or depict is the same doesn't make them the same. Things are also constrained such that any fragmentIds that resolve in multiple representations should be referring to the same thing regardless of representation - though not all frag Ids need to resolve in all representations [1]

Stuart
--
[1]  http://www.w3.org/TR/webarch/#media-type-fragid