So, I’m kinda new to SmartThings. Been on a couple developer calls. Wrote a few Smart Apps. Definitely liking SmartThings and it will be a core piece of my Home Automation AI project.
I’m trying to plan out the Big Picture and have a road map for my home, office and car over the next few years. I have a load of questions and I though instead of emailing Ben directly I’d post them here in case anyone else could benefit from the answers and hopefully some of the Gurus have the answers. Thanks in advance for helping me.
Can you have more than one location on your profile?
Can you have more than one hub (probably 1 per location) on your profile?
Can proximity devices like the keyfob, cellphone, etc. be registered on multiple hubs so it knows if I am in my House, Office or Car?
Are you planning to open source any of the hardware or share the schematics if we wanted to design our own sensors, fobs, etc?
Is there a way via the API to get the locations, hubs and devices? I’ve seen examples for switches, locks, etc. I want to be able to access all the info on my account.
Are there any plans for 3G on the hub? I know this was part of the original device. Is this being looked into as a backup?
If we have suggestions for software and/or hardware changes where is the best place to post or send them so they get to the SmartThings team?
Yes. You can add locations on your account. These can be “hubless” locations or locations with a hub.
Yes. You can have more than one hub. You can even have multiple hubs per location. (most homes would not need this) Hoping @urman or one of the other developers will jump in and set me straight on what is publicly available now versus coming.
Yes.
Yes.
Letting someone else tackle this. We do allow access to many things in the API but we are discreet about it. For example you cannot just say “give me a list of all devices on a location” as we feel this could lead to privacy and security issues. Again, devs weigh in.
We had 3G on our first prototype hubs but the cost was pretty high (added $70 per unit) and then we had to figure out service for it. We wanted to provide the hub without service fees and so dropped it. We think the strategy going forward with this would be to add a cellular card to the hub - in between or connected to the router. Future version of the hub will have a USB slot for such expansions.
These forums (and the next forums) are a good place. We’ll have a more official place on the new forum.
#1 yes. You can have hub and hub-less locations. Try creating new location, follow the steps and add maybe a mobile device. Then you can get notification when the mobile device is near/at that location. It’s handy to know when wife is at the store.
As always… you guys rock! Thank you for the quick responses. I guess 2pm on a Friday is the best time to post questions
#5 is probably my most important question of the moment so hopefully the other Devs can weigh in.
Since I have an already written Home Automation software in .NET that does a lot of stuff outside of controlling devices this is the reason I would like to query or import my Locations, Hubs, Devices, etc. and allow the my software to use this information to call the endpoints in my SmartApps and control what they need to. I’m trying to keep the SmartApp for my automation system generic and allow my system to just include SmartThings into my already existing Devices and Hubs schema. I have other things I control that are also Devices and Hubs. Since I have voice, speech recognition and facial and object recognition in my system, it only makes sense to just call the API and control the devices. I don’t necessarily need the ability to Add, Update, Delete Locations, Hubs, Devices from my system. I am fine setting up devices through the SmartThings IDE site.
You can’t do what you’re looking to do using the SmartApp Web Services method where you create a SmartApp that exposes its own endpoints. This is due to the security model that SmartApps run in - they are only given access to the devices that the user explicitly selects at install-time, and don’t really understand the concept of locations or hubs.
To do what you want to do, you’ll need access to the Core API, which is what we use for our client applications. We haven’t opened this API up for developer access yet, but we’re in the process of documenting the API and building the infrastructure for developers to request API keys that are limited in scope to their own accounts. This means that you’ll have full access to the entire API that can do anything that our clients can do, but only to your own account.
@Dan Lieberman That is awesome information and pretty much what I expected as far as your design and security model.
I am glad to hear you are making plans for access to the core API. This is exactly what I will need. I’ll just have to learn to be patient… so many idea swirling in my head.
