Re: [Iotsi] interactive vs. programmatic IoT

[Dave Raggett <[e-mail]>]

> On 24 Mar 2016, at 21:24, Clarke Stevens <[e-mail]> wrote:
> 
> For a start, maybe we can look at it in this way:
> 
> A human should be able to control things in real time in a logical way through some UI (e.g. push a button on a smart phone to turn on a light).
> A human should be able to specify automated rules in a logical way through some UI (e.g. set a rule to turn on the porch light when I get home).
> Semantic interoperability should take care of most other tasks (e.g. How do I recognize a light? How do I make functionally similar lights from multiple manufacturers and ecosystem all look like logical lights? How do I recognize when the user is home?)
> 
> There is a lot of functionality that is left out here, but I think this is plenty difficult as a starting goal.
> 
> -Clarke

Semantic interoperability makes it practical to relate information from different sources. This is key to realizing the value of the IoT in the broader sense. In the absence of standards, we are left with the problem of needing to know direct mappings between each of the sources, which would act as a brake on innovation and market activity.

W3C’s work on the resource description framework provides a scalable basis for an interlingua for translating between the semantic representations used by different platforms. The challenge is then how to encourage innovators to publish their semantic models and enable re-use where it makes sense.  For instance, could we agree on standard that enable search engines to index models published via the Web?  Can we encourage royalty free use of published models? Can we decouple application logic from the complexities of the underlying platforms and protocols?

—
   Dave Raggett <[e-mail] <mailto:[e-mail]>>