Re: IOTDB: modeling state

When I'm referring to Denon AVR and WeMo Switch, I'm actually referring to
those things. There is no theoretical aspect to what I'm talking about.

This is a well know IoT UX issue. See for example page 32 about
interstitial state:

http://environmentsforhumans.com/2015/oreilly-internet-of-things/presentations/Rowland-IoTDesignConf2015-Interusability.pdf

A list of Things I have worked with and actually coded for is here. The
list of Things I have actually worked with is somewhat longer. Except for
talking to device simulators, Arduinos, and Pin setting on Raspberry Pis,
every device has a human noticeable lag.

https://github.com/dpjanes/iotdb-homestar/blob/master/docs/modules.md

NB with low power devices - such as Bluetooth Low Energy - there's almost a
guarantee of a significant delay, as their power saving modes involve
waking up periodically to see if they need to be talking to anyone.

D.


On Wed, Jun 3, 2015 at 4:22 AM, Dave Raggett <dsr@w3.org> wrote:

>
> On 2 Jun 2015, at 19:27, David Janes <davidjanes@davidjanes.com> wrote:
>
> You always need to know when a property value has taken effect if you're
> presenting a user interface. e.g. an interface can't tell a user that the
> light has turned off when they can clearly see that it's still on! Instead
> the interface has to indicate that the system is in a transitioning state.
>
>
> Why?  In a distributed system, you can expect some lag although it could
> be quite small.  In a well designed system, this won’t be a problem.  If
> you as a designer know that there will be a long lag then you should design
> the system and the user experience accordingly.
>
> It would be very helpful if you could give specific use cases that clarify
> the issue.  I would like to hear more about the specific IoT systems that
> you’re thinking about, and why it takes so long from issuing a command to
> the change in the physical state.   The examples, I have seen, e.g.
> changing the brightness/hue of a light, or moving a servo, all seem to take
> place pretty quickly.  Rather than talking in the abstract, let’s talk
> about concrete instances.
>
> —
>    Dave Raggett <dsr@w3.org>
>
>
>
>