I am not sure what they plan to do with HUB v2. But, I was just thinking that many of us have local servers, or NAS boxes. Why not create a program that contacts the ST server and downloads all the updates to SmartApps etc, and runs the code and schedules locally? Then as developers we could run the code locally on our IDE’s and etc.
Plus then there would not need to be an additional device or hub unless you wanted to have the additional features.
This may have already been brought up, but I have not seen it yet.
Well, the problem with this is that it would require a change to the firmware in the Hub. Right now the hub is “hard wired” to phone home as soon as it’s plugged in. It isn’t designed to do anything other than communicate directly to the ST cloud and receive it’s instructions from there.
You’d need a new firmware that point to a different location. Further, ST would have to release a lot of info re: how SmartApps function and how everything ties together to allow individuals to install locally. There’s also the support aspect of managing the system on hardware that isn’t theirs. And there’s questions on whether this would be a full time server or only act as back up if the cloud went down (there’s advantages to either way in my opinion).
Of course, none of this is impossible, but it is complicated. It would stretch the resources of a team that I think is already pretty stretched. Beyond that, and perhaps most importantly, I believe that Hub 2.0 is suppose to do everything you listed above. While the price hasn’t been released I think they are aiming for less than $300 (at least).
So I guess it comes down to a question of saving some end users a few hundred bucks but taking on a large development project (new firmware, framework for a local server to operate the ST “os”, method for a local server to download SmartApps), and add new challenges to the support aspect.
In the grand scheme of things I think it would be smarter for ST to just focus on Hub 2.0. The potential lost sales in customers who walk away from ST because of lack of local server support is probably less than the costs incurred by developing and supporting this option.
Well, for me… Sadly, if they bring out a 300 dollar hub, that will not work for me. And honestly, I do not see why something like this cannot be done with a min of fuss.
I understand that they are stretched kinda thin. But, the primary driver is to keep hardware costs down, and to facilitate some local functions. Maybe I am asking too much, but it would seem that they could create a application that acts as a Hub essentially, and you just run ir as a service on a server.
1 Like
tgauchat
(ActionTiles.com co-founder Terry @ActionTiles; GitHub: @cosmicpuppy)
4
I strongly agree with the concept here, but I fear it is unlikely to become a part of SmartThings’s strategy.
Other vendors in the IoT space are considering how to make data storage and event processing either fully open or with some degree of flexibility. Spark Core, for example, has open-sourced their cloud, so you can run and support (and customize) your own instance for whatever reasons. So will Canopy.link.
ALYT is basing their hub on Android OS so automation / event apps are essentially equivalent to Android phone Apps.
Control4 and Crestron (and others?) have based their systems on an expensive in-home processor-controller for years now. Their business model, I believe, requires certified official installer/integrators to maintain the system. Only these VARs have permission to push major changes to “your” controller.
One of the threads here did mention other PC or Mac-(Mini?) based systems that are software driven.
But note:I have several colleagues that are 100% certain that companies like SmartThings, Nest, and others in the consumer IoT space are basing their business models NOT on the hardware or even the software, but rather on the data collected. Lifestyle data is an incredibly valuable resource… The entire basis for Facebook’s existence and certainly the reason Android/Google thrives.
Yes, I share your friends concerns about the lifestyle data. But, the floodgates have been opened, and I fear that we cannot shut them down. But, we can work with companies to determine what is allowed and what should not be allowed. I think that if companies do this, that they need to work toward paying their customers for that choice, or allowing us to opt out completely from data collection.
But, that is not what this thread is about. I really would like to see something as I describe come to pass, but I know that it is very hard to do this with the current model. Now if they allow for an upgrade path for prior users, that reduces the cost of the Hub v2. That might work for me. But, I really want to have more control over the local resources, and at the same time have the ability to write my own code as we do now.