Re: [Iotsi] interactive vs. programmatic IoT

[Dave Raggett <[e-mail]>]

That sounds like a question for innovators to explore according to the circumstances.  One example is the way that device UIs can indicate that an action has been accepted but not yet acted upon. e.g. when waiting for a sleeping device to wake up and deal with the action. Standards are important for enabling a marketplace of apps that can be installed on gateways for greater flexibility.  User interaction is needed here for managing services and their permissions. This could be through a web application hosted through the gateway or cloud. For me the question is what kinds of standards are needed to encourage an open market of devices and services. 


> On 24 Mar 2016, at 17:57, Michel Kohanim <[e-mail]> wrote:
> 
> Hi Dave,
>  
> Thank you and not only I totally agree with your comments but I subscribe to them wholeheartedly.
>  
> My main questions (as the subject of this email chain suggests) is how does a developer (such as our company) figure out “interactive” vs. “programmatic” IoT. Where is that boundary? Is it organic? If not, what are the constraints?
>  
>  
> With kind regards,
>  
> ******************************
>   Michel Kohanim
>   CEO
>  
>   (p) 818.631.0333
>   (f)  818.436.0702
>   http://www.universal-devices.com <http://www.universal-devices.com/>
> ******************************
>  
> From: Dave Raggett [mailto:[e-mail]] 
> Sent: Thursday, March 24, 2016 10:54 AM
> To: Michel Kohanim <[e-mail]>
> Cc: Subramaniam, Ravi <[e-mail]>; Kreuzer, Kai <[e-mail]>; Andy Bierman <[e-mail]>; [e-mail]
> Subject: Re: [Iotsi] interactive vs. programmatic IoT
>  
> Late binding is necessary where the data is not fully defined in advance. The recipient may have complete knowledge of the range of types that are late bound. If not, it may be sufficient to pass the data on to another agent that does. Alternatively, the recipient may in some cases be able to extend itself so that it now is capable of dealing with the data. This could just be a matter of referencing an external library and is in reach of gateways and cloud based systems that run relatively powerful software, e.g. based upon NodeJS. 
>  
> Another perspective is to consider the process of defining a distributed service composition involving a suite of devices. Metadata about the capabilities of individual devices can be used to match tasks to devices according to their known capabilities. An example is a smart factory where you want to allocate production cells to an order book for bespoke products. Dynamic reconfiguration is necessary when problems arise on particular cells or there are hold ups in the supply chain. This can involve cooperation with the human workers, e.g. to review proposed solutions.
>  
> —
>    Dave Raggett <[e-mail] <mailto:[e-mail]>>

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