Re: LDP feedback ( LC-2812)

(changing lists for possible WG discussion)

If we believe that to be true (for either RDF or HTML, frankly), we should 
be articulating those requirements.  Section 4 in the editor's draft has 
everything I'm aware of, and once you note that all the MUSTs are 
qualified (so they're not really Musts, in some sense), I'm not sure what 
a generic HTML and/or RDF client can't do or needs to do extra for LDP.

4.1.1-4.1.3 don't affect HTML (but not RDF-aware) clients; RDF clients 
should already be doing those things or they have an RDF-level problem.
4.1.4 boils down to a restatement of HTTP PUT (don't use it for *partial* 
updates unless you preserve all the stuff you're not intending to update).
4.1.5 is a MAY
4.1.6 mostly boils down to a restatement of the prefer spec; it only adds 
that (LDP) clients can't use "preference not applied" as an excuse to go 
casters-up.

Note that this section was born (as was 7) largely based on Mark's and 
TimBL's comments.  In the drafts they read, it might have been impossible 
for them to figure out that "this is all there is" for clients.

Best Regards, John

Voice US 845-435-9470  BluePages
Tivoli OSLC Lead - Show me the Scenario




From:   Mark Baker <distobj@acm.org>
To:     ashok.malhotra@oracle.com, 
Cc:     public-ldp-comments@w3.org
Date:   02/12/2014 01:33 PM
Subject:        Re: LDP feedback ( LC-2812)
Sent by:        mark@coactus.com



On Wed, Feb 12, 2014 at 1:03 PM, Ashok Malhotra
<ashok.malhotra@oracle.com> wrote:
> Mark:
> Is your point that the client cannot be a vanilla HTML client but must
> be aware of special LDP considerations?

Assuming you meant "RDF" instead of "HTML" there, that's correct.