Re: Regular Web of Things call on scripting API

On Mon, Nov 7, 2016 at 7:43 PM, Dave Raggett <dsr@w3.org> wrote:

>
> On 7 Nov 2016, at 13:12, Kis, Zoltan <zoltan.kis@intel.com> wrote:
>
> We agree that the mother of all is the REST API.
>
>
> In my personal opinion, I think that requires some qualification.  The
> interaction exposed to applications (properties, actions events) should be
> decoupled from the communication patterns and protocols, given that the
> requirements vary so much from one domain to another.  The Web of things
> needs to work with existing platforms, and this includes those that use
> REST, but we shouldn’t exclude platforms and protocols that aren’t based
> upon REST.  That said, it is pretty clear that we need good ways to express
> REST in the communications metadata, and to characterise how the lifecycle
> and operations on things map to the REST primitives.
>

Right, that is the full context.



> It isn’t clear to me that we need a REST API though.
>
>
We need to have something, either scripting APIs, or REST APIs, or
definition of data models and some general directions on how to work with
them.

Perhaps we should collect concrete use cases that illustrate both REST and
non-REST applications, and see how can we solve them with various
approaches.

So far I have encountered use cases that can be solved in a RESTful manner,
which is nice for simplicity and compatibility, but we may well need to be
able to express certain use cases in other ways. Let's turn that into
practice and discussion by examples. I wonder if we have a document
summarizing such classification of use cases.