Re: defining RDF and RESTful

Hi Sandro,

I like most of the changes you made, but wonder about the implied limitations where you say "(mostly GET and PUT)" for HTTP verbs.

Our experience with RESTful discovery of LD resources suggests the use of more verbs, especially OPTIONS for resource discovery purposes but also POST and DELETE.  We have found it necessary to balance the verb usage between them.

Please have a look at the Callimachus REST API, which shows how James Leigh and I have been enabling CRUD operations on RDF and related content for application development purposes [1].  That API reflects our practical experience and enables complete discovery of resources and descriptions via RESTful operations. 

Our three primary use cases have been:
  - AJAX operations in a browser to allow complex Web UIs;
  - Command line scripts for administration, monitoring, etc;
  - Application integration with external apps (e.g. CMSs, SAS).

I suggest that the Callimachus REST API be an input document to the LDP WG.

Regards,
Dave

[1]  http://code.google.com/p/callimachus/wiki/REST_API


On Mar 29, 2012, at 8:57 AM, Sandro Hawke <sandro@w3.org> wrote:

> In response to some comments, I've added some more introductory text to
> the draft charter and switched to "RESTful (APIs)" everywhere from
> "REST" (which may be more properly about how early Web applications
> changed state via the user navigating to a different URL -- cf HATEOAS).
> 
> Latest version:  http://www.w3.org/2012/ldp/charter
> 
> Diff: http://www.w3.org/2012/ldp/charter-history/diff-17-21.html
> 
> Comments welcome.
> 
>    -- Sandro
> 
> 
>

Received on Thursday, 29 March 2012 13:16:05 UTC