What troubles me about Samsung not making an actual business out of SmartThings is that they must have a reason.
I just came home and all of the events in my arrival piston went off correctly except one. For some reason my thermostat didn’t update. It worked yesterday and it’ll work tomorrow, but it didn’t work this time. This happens far too frequently with SmartThings. Daily even, if you have a large enough install.
This is what troubles me about Samsung’s attitude towards SmartThings. If they could make a real business out of it, wouldn’t they? It seems to me that they know they can’t get a satisfactory level of reliability given the current platform. Not something they’re willing to stand behind as a product people pay money for anyway, and there’s little evidence that they’re working on revisions or a rewrite that are going to get them there.
So what are they doing? Posturing that they’re leaders in “IoT” and “HA” until the hype passes at which point they scrap ST and move on to the next thing? Buying an engineering team to work on other products while letting ST languish?
Meanwhile I’ve got 100+ devices and 50 pistons that I count on. I have a hard time seeing how this ends well if we don’t take control of our own destiny.
Why would they invest the money they did into buying out Smartthings if they didn’t have a pretty solid plan for it’s future? Given their vast technical resources, surely they were very familiar with what they were getting into and had some kind of long term plan for going forward… right?
It just doesn’t make sense that they would buy this platform, and we’d still be at the level we are at today with it. I think something else much deeper is going on in the bowels of Samsung, and I have a bad feeling about where it’s going to eventually leave us weekend HA hackers.
EDIT:
I would be willing to bet that they needed some small piece of ST’s technology or framework for a much larger, long-term business model, and figured it was cheaper to just dump whatever money into the buy-out that it took to get it. (Think Micro$oft and 80% of the Windows operating system )
Project x is probably being developed at Japan’s version of Area 51 while ST dies a slow semi-supported death to keep people from screaming too loudly before the announcement of SmartestThingsEver.
I could not agree more. The best example I can give that probably makes us feel this way is if you look at the innovation into the ST ecosystem. All innovation (webcore, actiontiles, and now konnected) are all being done outside of ST.
I came from a locally executed vera system was using pleg and some other tools for automation. It was a pain to manage and make modifications to. It did however have that sense of just working once everything was setup. However I can tell you that I did not have near the consolidation of devices I have now with ST. Not to mention the ease of getting them to talk to each other because of things like webcore. (I think we all agree on this point if we are here)
Second the things they do seem to be investing in are not working. Reviews on their latest science project with ST and ADT are not looking good. Apparently it is a ADT alarm system with it’s own sensors that do not talk to the ST ecosystem and vice versa the ST sensors do not talk to the ADT portion of the system. They just made it so that you can have both platforms in one device… (I have not touched one or experience with this device just going off reviews on bestbuy’s website)
I do agree though that if they were to offer a container for those of us that want to execute locally whether it be a docker image, or even a vm that say if you have 200 devices you need to have 2vcpus, 8GB memory, and 200GB of space to support your system and setup the hub to point to this container or vm first then have it proxy async my hub status out to their cloud I would jump on it in a heartbeat.
For those that want a more plug and play just beef up something like a intel nuc(not much larger than the hubv2) and package it as an appliance you add to your home network with the hub or add the dongle like they did with the shield and boom local resources (obviously this would have a device limit but how many devices do you think even a large ST user is running? more than 400?)
I am in the mindset at this point that we are going to wake up one day and be sol with ST. Until then I guess I will enjoy it while it works, and cross that bridge when it happens.
I have found ST to be pretty reliable over the two or so years I have been using it. Everything has an occasional glitch. I knew that when I started this adventure and accepted it as a possibility.
It would be really cool if we could change shards during an outage for our mission critical stuff - but I get it that would most likely be really difficult and expensive…
One thing I’d point out is that I’ve been in tech for years and its impossible to underestimate the stupidity of corporate M&A folks. They’re largely finance guys and don’t really understand the tech. They’d buy your rusty old delorean if you told them it was a time machine.
Everybody involved in the purchase got a pat on the back for “closing the deal” and moved on, leaving figuring out what the hell they actually bought to somebody else. Its entirely likely that samsung has no viable or well articulated plans for smartthings or for HA going forward. They’re busy selling phones and washing machines.
I’m really new to ST (December 2017) and even newer to WebCore (Yesterday). I think that you answered my question, but I wanted to make sure that I understood.
Are you saying that the SmartThings Apps do not run on the SmartThings hub and the hub really just communicates with the SmartThings cloud? So, no internet, no automation? Right?
Thanks for the reply. I’m still trying to get my head around all of this.
Is the main benefit to WebCore (vs Stringify/IFTTT), the flexibility or does it also respond faster because it is a SmartThings App (albeit a user-based App)?
Well, just to put an endcap on this thread, I got my Hubitat in the mail today. This is a $99 device that runs most ST type groovy code locally, including it appears, WebCoRE.
)
