Re: don't rely on bnodes when updating an Annotation or a Reply

On Fri, May 09, 2003 at 01:02:34PM -0400, Ralph R. Swick wrote:
> 
> While debugging new Annotea server code, Eric Prud'hommeaux
> and I noticed that Amaya did not exactly conform to the
> Annotea protocol for updating an Annotation or a Reply.
> 
> Per the examples in
> 
>  2.4 Updating an annotation
>  http://www.w3.org/2001/Annotea/User/Protocol.html#Updating
> 
> and
> 
>  3.4 Updating a reply
>  http://www.w3.org/2001/Annotea/User/Protocol.html#UpdateR
> 
> an Annotea client is expected to not use bnodes in the PUT
> message body when referring to the existing Annotation or
> Reply or their bodies.  Amaya was failing to include these
> URIs in the PUT body.  The Annotation or Reply URI is the
> PUT URI but it should still be given explicitly in the
> contained RDF.  The server might be able to deduce a
> body URI from this, but should not be relied upon to do so.
> 
> I have updated the Amaya CVS sources with a fix to this, so
> that Amaya will not use bnodes when updating an Annotation
> or a Reply.  For backwards compatibility with older clients,
> server developers might want to consider doing what the
> current annotest server does, which is to deduce the URI
> from available information when the client does not
> specify it.

server does this with the annotation or reply object. not yet with
the body, but i have to pack and get on a plane.

-- 
-eric

office: +1.617.258.5741 NE43-344, MIT, Cambridge, MA 02144 USA
cell:   +1.857.222.5741

(eric@w3.org)
Feel free to forward this message to any list for any purpose other than
email address distribution.

Received on Saturday, 10 May 2003 14:24:35 UTC