Re: Fw: Comments on LDP-Paging - the remainder

On Thu, Jul 31, 2014 at 9:21 AM, John Arwe <johnarwe@us.ibm.com> wrote:
>
> Getting this on-list.
>
> wrt when an in-sequence page would be removed, simple example is that
data is deleted from the paged resource (>=1 entire page's worth), after
the traversal has passed that page.  If the client "comes back again", and
especially if it cached/is using the page links from its original
traversal, it can observe a smaller set of pages.  It's just the symmetric
case versus the "added data => larger and/or new pages" case we've focused
on.
>
After thinking about this some, I can see some cases where server would
stick with the current sequence and a way clients can deal with it.

> wrt the 404 question, BP has NOT yet been published so it's TR-space URL
still 404s, as of yesterday.  I begin to wonder if we should just point to
a list hung off the WG wiki page (or that page itself) and get of South
Park shop class.
>
Right, was thinking we could just point to editor's draft, but BP is fairly
close so worthy of waiting but it is pretty easy to update the reference
when we need to update for CR, ...

All the best,
Steve

> wrt Preference-Applied, your inserted contexts match my intent
>
> Best Regards, John
>
> Voice US 845-435-9470  BluePages
> Cloud and Smarter Infrastructure OSLC Lead
>
> ----- Forwarded by John Arwe/Poughkeepsie/IBM on 07/31/2014 09:05 AM -----
>
> From:        Steve Speicher <sspeiche@gmail.com>
> To:        John Arwe/Poughkeepsie/IBM@IBMUS
> Date:        07/30/2014 04:31 PM
> Subject:        Re: Comments on LDP-Paging - the remainder
> ________________________________
>
>
>
> > On Wed, Jul 30, 2014 at 1:41 PM, John Arwe <johnarwe@us.ibm.com> wrote:
> > > <#terms-from-paging>
> > > "Paged resource"
> > > "In-sequence page resource"
> > > "In-sequence page resource", "fist page link", "last page link"
> > All these should be addressed now; SS, see if you agree.
> > changeset [2] / editor's draft [3]
> >
> +1
>
> > > <#ldpr-pagingGET-sequences-change>
> > Should have been addressed via "the Sandro-motivated changes"...
changeset [4] / editor's draft [3]
> +1
>
> Though I still have a hard time understanding when an in-sequence page
would be removed and the sequence would be maintained, I would think server
impl would would initiate a new sequence.
>
> > > <#ldpp-client-paging-incomplete>
> > Another thread handling this [5]
> (responded there)
>
> > > <#MISSING?>
> > (two of those in [1]) Another thread handling this [6]
> (responded there)
>
> > > <#intro>
> > Being unwilling to add a 404 link in an LC draft, carrying as TODO in
HTML comment at top of file.
> I wasn't suggesting pointing to non-existent docs, both these documents
exist.  I guess if there were separate documents for paging that were on
our todo list, then yes, they'd be 404s.
>
> > > <#ldpp-ex-paging-303>
> > yes
> (added in missing context/question)
> >> Examples 6, 8, 10, 12: Is this missing?
> >> Preference-Applied: return=representation
> implying "no"
>
> >> I see in 6.2.4 says it can be omitted if client can determine from
other
> >> values, so I assume that is why it is not there?
> implying "yes"
>
> +1
>
> > > <#ldpr-impl>
> > No objection from me; asked those authors offline IIRC, no response yet.
>
> Ok
>
> Thanks,
> Steve
>
> > [1] http://lists.w3.org/Archives/Public/public-ldp-wg/2014Jul/0059.html
> > [2] https://dvcs.w3.org/hg/ldpwg/rev/34d96d2d6359
> > [3] https://dvcs.w3.org/hg/ldpwg/raw-file/default/ldp-paging.html
> > [4] https://dvcs.w3.org/hg/ldpwg/rev/76e5cbeb6b36
> > [5] http://lists.w3.org/Archives/Public/public-ldp-wg/2014Jul/0136.html
> > [6] http://lists.w3.org/Archives/Public/public-ldp-wg/2014Jul/0138.html
> > Best Regards, John
> > Voice US 845-435-9470  BluePages
> > Cloud and Smarter Infrastructure OSLC Lead
> >
> > From original [1]