| Re: [m2m-iwg] DeviceHive |
|
A few other observations: -
Kind of a “one size fits all” command approach that leaves the payloads pretty much a black box – needs to be more specific command types/formats/associated semantics -
I like some of the higher level object semantics and the device typing, although it is incomplete and doesn’t really let you discover the device’s capabilities (another “black box” blob for that) -
Similar to the approach we’ve taken at TW where the “pipe” and “wire format” could be just about anything (JSON or binary over HTTP, web sockets, raw sockets, serial/BT), so I like that part ;-) -
Device registration doesn’t require any authorization, which is a pretty big security hole -
The concept of a “network” is an interesting organizational concept which I also agree with – enables faux multi-tenancy and bulk permissions to be managed more easily. We call them “ThingSpaces” but
same basic idea. -
I like some of the QoS stuff built into the command structure to allow lifetime for commands. Should be more structure/specifics to the command status/flags though That’s a quick brain dump if it helps. From: m2m-iwg-bounces@xxxxxxxxxxx [mailto:m2m-iwg-bounces@xxxxxxxxxxx]
On Behalf Of Rick Bullotta Just took a look. Seem another protocol missing a metamodel and concrete data/command formats. Server is .NET only for now. From:
m2m-iwg-bounces@xxxxxxxxxxx [mailto:m2m-iwg-bounces@xxxxxxxxxxx]
On Behalf Of Ian Skerrett Does anyone have experience with DeviceHive?
http://www.devicehive.com/ |