Re: Comments on Annotea server and protocol

Le ven 29/08/2003 à 18:48, Marja-Riitta Koivunen a écrit :
> We did think carefully about the wordings but, if you can give me exact 
> places where you would prefer that we use stronger keywords that would be 
> helpful. Often it is that the protocol doc might not be the exact right 
> place for the strong requirements (although some are there now), we should 
> have another doc for explaining the implementations.

A few examples of what I meant about RC 2119 keywords usage:
"When a client creates a new annotation it posts RDF statements in an
XML body that describe the new annotation resource to a selected
annotation service. The client constructs an HTTP POST message
containing the annotation as an anonymous RDF resource."

In this paragraph, it is not clear what is simply descriptive and what
is assertive; reformulating it as follows:
- To create a new annotation, a client MUST post RDF statements in an
XML Body that describe tne new annotation resource to a selected
annotation service
- the client MUST send an HTTP POST message containing the annotation
and the contained annotation MUST be an anonymous RDF resource
clarifies what are the real requirements set on what actors of the
protocol (note that I made up which of those were actual requirements,
and I also made up the force of the requirements).

Idem for
"The annotation service is responsible for allocating a URI for the
annotation".
-> The annotation service MUST allocate a URI for the annotation. 

HTH,

Dom
-- 
Dominique Hazaël-Massieux - http://www.w3.org/People/Dom/
W3C/ERCIM
mailto:dom@w3.org