SmartThings Publication of Apps to Run Locally

This topic was not started by me.


Thanks for the constructive feedback @JonTuxIvy

We are doing this. :smile: We speak with a group of about 40 developers on a daily basis. Give them the sneak peaks into what we’re doing. :smile:

Exactly, and we told you we will be after you asked us if we would. :smile:

1 Like

Wow… “not a priority” (per Developer Advocacy…) really?!?

We’ve not tried to have the same discussion for SmartTiles (i.e., “Is there any possibility that SmartTiles can be published for local Hub V2 execution?”), because it’s essentially a “client” that uses that SmartThings Cloud as a web-server. (We are, however, working on general submission and publication).

But Rule Machine? That’s just a tremendous enhancement of something like … Smart Lighting. In other words, the perfect candidate for local execution.

Sure, as a whole SmartThings needs to (must?) have a lot of #1 priority issues to address, but when it comes to the SmartApp publication channel, since Rule Machine can literally replace dozens or more other SmartApps, it seems like it would be a highly efficient, high value – and thus, high priority – project.


And we agree which is why we have slated @bravenel’s app for the next sprint, after we work with you Terry to get the SmartTiles endpoint published (and 6 other endpoints).

Local execution is another beast we will tackle later.


Thanks for the update, Tim!

To be 100% clear to the you and the Community, though, (if my vote carries any weight):

If you can confirm that SmartTiles V5.x existing installations will continue to execute after the OAuth changes have been deployed (i.e., we acknowledge that new installations will not be possible) … then approval and publication of the new SmartTiles V6 endpoint can voluntarily take a back-seat to @bravenel’s Rule Machine SmartApp.


Take this with the mountain of respect intended:

Rule machine is awesomely powerful. I agree it should go local if possible, but that would be a challenge I am certain, so these comments are in the vein of cloud based app:

I think it replacing a dozen other apps is true, but its not for everyone and confusing for others as I observe the threads.

There is room for all types of apps, and moreso I think I want to see some clarity of the priority for ST review queues in a way so that we understand the protocols and even help vote community desires (like uservoice type of thing)…not sure really what it would look like for how to prioritize, but Rule Machine #1 right, cause its awesome (to me)? Or is it a smaller, specific apps, less impactful to the system, and easier to vet its pro/con ratio. I guess I have a hard time with the right approach as until now, most apps approved are simple use cases for retail users. I think ST sometimes gets stuck in that vortex.

All told, straddling this line of supporting high end community developer group while making it simple for retail users is a hell of a high wire act.

I would like to see Rule Machine expedited, and put in an Advanced category perhaps?

Thanks Terry, I’m not sure the other 6 people would agree with that, though.

Rest assured, Bruce is next though. :slight_smile:

And we want more types of apps, complicated, non-complicated, big, small… The community will push the crap down and let the cream rise.

We are ramping up our certification again after going through and creating a true process for it. The certification team has taken this over and they are now treating the review queue similarly to the device review queue. Every app will get it’s review period. The queue is large, but we are beginning to work through that.


Good to hear, Tim, thanks. I understand the mountain y’all face.


That’s not a typo, is it? Is not to,be read ‘next spring’ loool. Glad to hear you you guys are seeing the light at the end of the tunnel!!!

1 Like

Duh, there is not a single automation that Smart Lighting can do that Rule Machine can’t do :grinning:; of course the opposite is not in the least bit true. @alex said that lighting has to run locally, which I agree with. But Smart Lighting is not anywhere near being a robust lighting automation application. Alas, RM is stuck in the cloud.

I just finished replacing all but 4 of my SmartApps with Rule Machine. I still have things like Echo and Harmony integrations, and Button Controller+. All of the remaining automations are handled by RM, which wasn’t true until today, due to last night’s update. I finally realized that I could add the features that I need to RM, to get my needs met. Voila: mode based motion controlled lighting, with button and switch triggers, time and mode restrictions, and disablement options.

It certainly doesn’t matter to me what category they put Rule Machine in, and I think you are right that RM is not the easiest app to master. ST has long been stuck with this view that only simpleton things would fly. But, we don’t really know, do we? RM certainly has proven that many users want and need more sophisticated tools. ST needs the Marketplace for SmartApps to have the same ability as in the Marketplace for Things to link to manuals. Rule Machine needs its documentation. Right now, they don’t have a way to provide that link.

BTW, RM is not more “impactful” to the system than “smaller, specific apps”. It just happens to be vastly more flexible. Each Rule tends to have a couple of subscriptions, and a tiny set of actions, just like the small specific apps.


Absolutely…forgot to even mention that deal. Absolutely.


Have you asked them??? :confused:

Seriously… I’m totally willing to “take one for the team”, here, put my full support behind @bravenel, and, well, frankly… “The needs of the many outweigh the needs of the few”.

While SmartTiles (for example) is appreciated for fulfilling a particular need, I would estimate that the other OAuth affected SmartApps have roughly the same degree of urgency … plus-or-minus …; namely, that as “good members of the Community”, they might be willing to tolerate a little bit of extra delay.

I can’t speak for them, however – but, as you are aware, I mentioned in another post, I feel pretty strongly that all folks that expect their SmartApp to be published ought to participate in the Community and thus be able to observe and weigh-in on discussions … like this particularly relevant one.


The caveat being that we announced the closure of the oAuth install method and delaying the publication of those SmartApps would have very direct impact on them. Given the nature of Rule Machine and its near endless possible configurations it will be orders of magnitude harder to QA than an app that exposes some rest endpoints. The bottom line is that we do care about developers and we want everyone to be published, it’s now a matter of getting the resources in place to make all of this happen and provide the best experience for everyone.



LOL!! How do you QA a Turing Machine? Wait, isn’t that the Halting Problem?

Good luck with that. :grinning:

The best form of QA has been and will continue to be usage, bug reporting and prompt bug fixing. Efforts spent to test are going to be wasteful, effort ill spent. But, knock yourselves out.

Odd that you guys would worry about testing a Community submission while allowing your own stuff to go out with obvious bugs found within hours by a handful of people in the Community. What, we get testing while you guys get a pass??


With due respect to the folks at SmartThings that have been given the responsibility of reviewing / testing / publishing Community SmartApps, and with appreciation to our Developer Advocates who are certainly doing their best to expedite the process and do it in as fair and efficient ways they can…

I’ve had the same feeling that @bravenel expressed.

But perhaps the particular set of us that are affected can have an offside discussion in this regard.


Alright we have gotten a little off track. Bruce, you have a private channel you can talk directly with Jody and I, let’s use that. Keep in mind we are going to certify your app, so not sure where the contention is coming from. :confused:

I’m specifically talking about the conversation around publication. I am all for the OP discussion.

Created a new discussion instead.

For those of you just showing up, we are going to put rule machine through the publication process. Rule Machine is slated for the next sprint.

1 Like

I appreciate the new Topic … it raises awareness and understanding of the publication process … and the complexities and limitations of Local Execution.

There are a few misconceptions regarding what is possible or not possible in regards to Hub V2’s “local execution” abilities. While it’s a given that for the indefinite future, Community SmartApps that *have not been passed through the submit, review, and publish" process are certainly not candidates for local processing, there are other constraints as well.

Some of these are obvious: If a SmartApp instance is linked to any non-local Device Type Handler.

But other constraints are not clear to me. As @bravenel points out, Rule Machine can fulfill the features of Smart Lighting. While that’s certainly not sufficient to assume RM could run locally, it is a starting frame of reference.

Can y’all help us understand the other constraints / criteria, perhaps?


I could understand if you are talking about QAing the intricacies related to allowing Rule to run locally. If the code would be made public to selected few, I am sure they can help out. But if you are talking about general functionality, that QA is done, vetted by so many users that love the app and have 50 plus rules set up. Unlike other officially QAd apps, Rule has far less bugs, and if there were any @bravenel and others helped out to correct them. This app, as many have said, is the resemblance of what the community used to be and it has given hope to many that ST can exist with no annoying bugs!