On 20 Jun 2016, at 15:38, Scott Jenson <email@example.com> wrote:I'm hoping that having a required set, tied to a versioning mechanism, will allow for manufacturers to still enable their optional settings, so companies can be encouraged to experiment and push the existing version. However, by having a fixed required set, it makes it much more likely that devices within that category will interoperate.ScottOn Mon, Jun 20, 2016 at 1:48 AM, Dave Raggett <firstname.lastname@example.org> wrote:On 19 Jun 2016, at 21:11, Scott Jenson <email@example.com> wrote:On Thu, Jun 16, 2016 at 4:14 AM, Dave Raggett <firstname.lastname@example.org> wrote:Essentially, to integrate to any existing IoT device, we just need the metadata describing the data and interaction models exposed to applications, the protocols and communication patterns that can be used, and descriptions of the kinds of things are involved and the relationships between them.There is quite a bit packed into your sentence. As a UX designer, I'd like to make sure that we appreciate that Apple is doing much more than just "metadata description and discovery":
- A *limited* set of device categories
- Clear and *limited* functional schema for each category
- A fairly limited app to find and control these devicesI'm not disagreeing with your comment, this is a "yes and" type of reply. If we truly expect these devices to work in in the home, we have to go beyond just the metadata description and discovery. We need to appreciate interoperability comes from hard decisions: by having a strong, if limited set of devices and functions. I realize this is a more business strategy comment than an engineering one. However, I'm fearful we're just going to recreate another "32 Bluetooth profiles" mess all over again.Is there any proposal to have the concept of a required base set of functionality? There can also be an optional set but taking a hard stand on required, as Apple as done, goes a long way in providing interoperability.
ScottHi Scott,Thanks for replying. You touch on an interesting challenge, how to foster convergence on a common set of models. A joint proposal from say Apple and Google could be a big help with that. A related challenge is how to support an agile process for standardizing models, with a clear progression for increasing maturity from experimental, to commercially deployed, to massively deployed. I am thinking of lightweight semantics along the lines of schema.org, and a easy way for people to browse for vocabularies (to encourage re-use), to register their support, link to implementations and so forth.p.s. In respect to the application paradigm, there seems to be general support for things with properties, actions and events, based upon the success of event driving programming.