As per announcement I guess the current set of Smart Apps will cease to function? Am I right? Which means I heavily use “Smart Lighting” for example, “Webcore”, “Lock User Management”, etc. will they cease to run / operate / visible in the mobile app (iPhone)?
If so : are there any alternatives?
On similar lines I have DTH for many of the products since by default SmartThings/Samsung do not support them. What happens to those drivers?
There is a new plugin version of Smart Lighting. ST will automatically migrate the old version to the new one before the year ends.
There is Smart Lock Guest Access you can look at. You can create users and set codes.
For webcore… consider SharpTools and Routines/Scenes/Rules API.
If there is not a stock Edge Driver or community developed on, you can ask one of the developers about possibly creating one.
yes, but almost all the substitutes don’t behave exactly the same.
For example, the new Smart Light supports only one-way synchronization.
For alternatives to groovy SmartApps, see the following discussion:
Replace Groovy with Automations—what’s your plan?
If you are using any Rboy SmartApps, he has been working with SmartThings on a transition plan, but there aren’t a lot of details. There is a transition FAQ on his site:
If you mean what was called “mirror behavior” in the old Smart Lighting, it was only one-way there, too.
If you want two devices to mirror each other, you need two routines in either the old or new Smart Lighting/lighting.
I read the announcement and it says local processing will occur soon instead of cloud. A quick side question: will my hub (with battery backup) support it hardware wise or I need to get a new hub too?
It’s going to be “more local processing,” but not “all local processing.”
The platform still expects that your hub will be actively connected to the SmartThings cloud most of the time. Indeed, the SmartThings app will only work when you have an active internet connection, even if your phone and your hub are on the same local WiFi. (They didn’t have to design it that way, but they did.)
Anyway, the good news is that you don’t need a new hub. 
Well – I don’t know. I am reading about some other product which is all local and consequent to this migration and other “connected” issues (like privacy / what happens if no internet / outage) many are migrating to this alternative platform? I ain’t sure and need to go over these links to see what effort it’ll be from my end to continue be in SmartThings vs effort it’ll take to migrate to this “all local” platform.
Although I observe many of my current devices are having Edge support and probably will work locally, I am hesitant to continue with Samsung as even in the new platform it seems they are mandating a real-time and continuous connection to their cloud. In my naive but particularly alarmed mind - question is why?
If they are so much into marketing and even evolving itself with significant resources and money into this whole re-architecture of running local then it begs me to question their rationale why they still want consumers to retain a connection to their cloud.
I am more and more gravitating towards Hubitat now. It seems for me the effort involved will be comparable so why not do it once?
Finally, Hubitat seems to have support for equivalent of webcore, something I use heavily. I am 200% certain with Samsung even in their newest version will not have the complexity which webcore provides - ever. Additionally, I have seen Samsung’s automated approach on arming / disarming to be pathetic and unreliable and don’t see they have made any serious investments to rectify this.
The only thing that’s preventing me for adopting Hubitat is I am not sure of that company’s future, both in general and finance wise. I have too many of these tech companies come and go.
At least with SmartThings, its backed by Samsung.
I purchased a hubitat hub about a year ago and have it standing by. If the ST a.g. (after groovy) starts to look less functional than ST b.e. (before edge), I’ll be putting my effort into migrating to hubitat rather than wrestling with ST. I already feel like my smarthome peaked about 18 months ago and some of my integrations have slowly stopped working over time. (Not always completely the fault of ST.)
My only slightly educated guess:
Moving drivers and the IDE of of the cloud is presumably saving them a significant amount of money in hosting costs for those functions. It possibly came down to re-architecting or charging every hub user a subscription fee.
But they haven’t re-architected how the app works. Yet.
You went into SmartThings when it needed a continuous Internet connection to perform all but the most trivial things. Seems to me it will be, if anything, less dependent on the internet when Groovy is fully retired.
Respectfully, that’s definitely not what SmartThings employees have said.
Remember that over 90% of SmartThings customers don’t have a hub: they have a Samsung smart appliance or television, all of which do need a cloud connection. So the fact that the app does, too, is essentially irrelevant to that customer base.
And SmartThings employees have said in just the last few months, since the end of Groovy was announced for December, that the architecture assumes the hub will have a near continuous Internet connection. It’s not designed to run without it.
Right now a lot of the rules still run in the cloud.
They put a major effort into creating a REST API for the new architecture, but that itself is part of the cloud, not on the hub.
And don’t forget their entirely new “directly connected“ device class, which is for Wi-Fi devices, connecting to the smartthings cloud.
And their Matter integration will be one way in, so they don’t have that pressure to be local.
There will be less need for the cloud for Zigbee and zwave devices, but it’s still largely a cloud-based system. I don’t see anything to indicate that that’s going to change.
Maybe one day once appliances have matter compatibility and everybody else is running local There will be enough market pressure to get Samsung to reconsider. But I expect that day is a long way off.
JMO, of course.
Okay - we can banter over this but let’s get into real meat of the issue. I checked what the notice says - option 1, then 2, then 3 and then 4, etc. What I gathered is this: I need to identify what I have. For example I have 10 of open/close made by manufacturer = X, 5 by manufacturer = Y, etc. Then I have motion sensors, again group by manufacturer. So on and so forth.
Then I have to go this link: Quick Browse Lists for Edge Drivers - Things That Are Smart Wiki
Identify if supported. Then for each go to that thread, read thru what’s required and do it.
Rinse - repeat.
Did I get this right?
Another Q Q : How to determine if a device is using Edge drivers or not? For example - I unlinked KASA (TP Link) stuff, made sure I have installed the Edge drivers and then re-linked Kasa. They are now all showing up as “placeholders” in the old IDE. So how do I know its using Edge?
Placeholder indicates it is either a directly connected device using an Edge Driver or cloud to cloud integration using the new architecture.
Devices that are directly connected using an Edge Driver would also be indicated in the ST app, by opening the device, tap on the 3 dots and having Driver appear in the pull-down menu. Cloud to cloud integrations would not have that Driver option under the 3 dots.
If you relinked Kasa by add device > by brand > kasa… it would be using the official integration on the new architecture. If you installed a community developed edge driver, you would need to choose add device > scan to get that edge driver.
If you relinked Kasa by add device > by brand > kasa
Yep - what I did.
But: 3 dots don’t give me “Driver” option.
I think that is what he suggested. If you do it by brand you’ll get the official integration on the ‘new’ architecture which is, presumably, a C2C integration using the ST-Schema.
If you want a community Edge driver to be used you’ll presumably need to use Scan Nearby instead.
How can “scan nearby” work for Kasa??
