- From: Wilde, Erik <Erik.Wilde@emc.com>
- Date: Thu, 21 Feb 2013 05:36:10 -0500
- To: John Arwe <johnarwe@us.ibm.com>, "public-ldp-wg@w3.org" <public-ldp-wg@w3.org>
hello john. On 2013-02-13 14:09 , "John Arwe" <johnarwe@us.ibm.com> wrote: >We had trouble enough getting consensus amongst our own product groups >(Erik's "very bad" flaw being one I raised) about the paging approach >generally that we chose to simplify by omitting it from the submission. >We can't > solve everything in a single go and still have it finish soon enough to >be relevant, so we put it on the "good follow-on discussion" pile. just as a follow-up: while client-controlled paging (requesting specific page sizes and specific pages) gets a little more involved when it should be done RESTfully (because these request parameters probably should be exposed via URI templates), the "cheap" solution is much easier. for server-controlled paging, there simply can be "next" and "previous" links (and a few more) exposed in a way similar to http://tools.ietf.org/html/rfc5005 . in this case, no templates are required, but clients still can happily page through results (but they cannot control the paging process, or request specific pages). cheers, dret.
Received on Thursday, 21 February 2013 10:37:03 UTC