Re: [Iotsi] acceptance notice and a request

[Barry Leiba <[e-mail]>]

> This is to let you know that you have been accepted for the workshop.
> For your information due to the popularity of the workshop we've been
> forced to limit the attendance and reject a fair number of papers. But
> it is important that the workshop is seen as a joint discussion
> between many people and organisations. As a result, we're asking all
> ADs and IAB members who desire to attend to submit position papers
> (see below for a message sent to the others). In your case, per prior
> agreement, you have already been accepted, but we’d like to request
> you to submit a position paper nevertheless.

Thanks, Jari.  Below is my contribution:

As the Internet of Things adds a great many devices to the Internet, a
number of basic interoperability issues arise:

- Discoverability

    - How do the devices find the controllers or peer networks they need?

    - How do the controllers or peer networks find and register new devices?

   - How are devices identified?

    - Is it necessary to have anonymous / pseudonymous devices, and
how does that work?

- Authentication and access control / trust

   - How do we decide to accept new devices?

    - What do we allow new devices to do?

   - Do we trust the data produced by new sensors?

    - How do we prevent attacks?

- Capability / function

   - How do devices communicate what they do and what their limitations are?

    - How do controllers restrict devices -- limit them to a subset of
their capabilities?

   - How do devices interact when their capabilities are related /
complementary?

That's just the tip of the iceberg, of course, but we have to start a
discussion of the models for answering those questions, on which we
can build the necessary protocols to manage networks of IoT devices
and to have the devices interoperate.

For a long time, I've considered a basic scenario:
I have an early meeting tomorrow at an office I don't usually work in.
The meeting is on my calendar, along with its location.
My alarm clock automatically sets itself to wake me earlier.
My coffee maker starts the coffee earlier.
My car's navigation system gets the location and automatically sets my route.
...and so on...

The issues above fit directly into that scenario:
How do my calendar, alarm clock, coffee maker, and nav systen get tied
in together?  What are the security and privacy issues of that?  How
is everything coordinated?  How do I, the human user, fit into the
control system?  How do I keep track of the devices involved?  How do
we make sure it's really my calendar resetting my clock and nav
system?

-- 
Barry Leiba