Simple problem simple solution, Replace smart lighting with RM, or Add the missing functionality that RM provides. Problem solved, There would be no state variable.
1 Like
It goes without saying that these are just personal opinions; ā¦
I disagree with your assessment and conclusion. While the ācontained scope app paradigmā that I imply and infer SmartThings to be based on and vested in may not be optimal, I think it was a creative and innovative choice (definitely not āstupidā), and the concept has held up pretty well overall.
ST engineers are genuinely impressed with how far and diversely the Community has stretched, nay, carried, the model; beyond their vision by far!
And that perhaps actually validates the paradigm, ironically.
But SmartThingsās official apps and UX is more likely to grow incrementally rather than radically diverge from the original vision and current implementation. And I think this is not a fatal path.
1 Like
I agree, partially. In the beginning yes. But as you said growth must occur and growth occurs in all directions. It finds the mostt most productive path and flourishes there. A tree starts as an acorn. The tree is nothing like the acorn.
To inhibit change is to inhibit growth.
And growth of the company is not just in the tech, but in the vision. The vision is the driving force.
1 Like
The reply you received is typical of the reason I no longer open support tickets.
2 Likes
Why is it that you pretend to be the spokesperson for what ST intended? Do you work for ST? Are you authorized to speak for ST? This is BS.
This is a company with ācore principlesā? Oh, like things should work? That kind of principle? Or apps should have blue buttons or green buttons? Thatās core, itās coming from Samsung.
2 Likes
Chill, Bruce 
ā¦ I donāt really know āwhyā I write what I write; and I donāt purposefully pretend anything (in fact, I try to emphasize frequently that what I post is speculation, hypothetical, and/or just humble opinion; but really, thatās a given by now).
I guess I just enjoy exploring ideas and philosophies, even with limited information. I also feel that I have a lot of knowledge based on my long term exposure to the company and product, and come to conclusions (or assumptions) that I like to share or bounce off the Community ā¦ even if there is disagreement.
And BTW: My avatar / profile summary explicitly states Iām not a SmartThings employee.
I hope that helps eliminate any ambiguity.
Pet peeve, those who present themselves as masters who are but apprentices. I will chill nowā¦ 
6 Likes
Hereās another lost ruleā¦
The first two are pics are of an identical good rule.
The second two are the destroyed rule.
General question regarding triggers in RM (not related to the current ST issues in particular)ā¦
Say I want to create a simple trigger ā at a certain time, turn off a switch if itās on. Simple enough.
Trigger: at a certain time
Condition: switch on
Action: turn switch off
However, there is no harm in sending āoffā to a switch thatās already off, so this could be simplified to simply trigger and action. Is one preferable to the other? If so, why?
I wouldnāt bother with checking to see if itās on, except in special circumstances where it actually might matter.
Yea, just turn it off. If itās off already, nothing happens.
1 Like
Hey Bruce,
I would like to review a previous topic/issue I am having again. When I create a āRuleā and set it to notify me via text w/device name, I get the device name, which is perfect
When I create a Conditional Trigger and have it send me a text w/device name, it actually sends me the name of my hub and not the device name.
Here is the CT
Here is the text msg, notice the bottom one, itās the CT and it shows my hub name. The one above it is the Rule and it shows the device name
What am I missing?
Rick
Yea, I know. Hereās the deal. The device that gets sent has to send an event to the Rule so the Rule can grab its name to send to you. In a Rule, that works as expected. In a Conditional Trigger it doesnāt work because the rule isnāt even listening to the devices in its Conditions. It looks at them when itās triggered.
Iām not sure what, if anything to do about this. I know exactly what youāre dealing with because I had the same problem. So I ended up sort of redesigning how I did things so that there was a Rule where the device could be captured, and Triggers or Conditional Triggers that messed with that Ruleās private Boolean. It isnāt pretty.
Iāll continue to think about this problem, but it doesnāt have immediately obvious solutions.
1 Like
Not sure if someone got to tell you @bravenel but I was just on the developer call (Iām not a developer though) and Tim mentioned that engineering believes they identified the state issue and are working on a resolutionā¦
3 Likes
That would be awesome if they can come up with a resolution.
I have a bunch of rules to rebuild that arenāt worth rebuilding if theyāll get neutered again tomorrow.
Iām down to about 6 working rules right now.
Well that makes about 5 or 6 big ones theyāre working onā¦
2 Likes
I opened a ticket with support on this issue as well (207997)
It isnāt an RM-only issue. Iām losing simulated switches because they donāt know their off/on state.
Love rule machine, but @bravenel, I was wondering if you could add support for HTTP get requests like alexa helper has? I would love to use rule machine to chain together HTTP GETS in order to control web-based devices.
1 Like
Seems to me that this would be best implemented by writing a custom device type to handle the HTTP interaction, and then using rule machine to interact with that custom device. Putting things like this into Rule Machine would cross the abstraction layer intended to exist between devices and smartapps which is part of what makes ST so powerful.
1 Like
I opened case: 207655 with Support and they replied acknowledging the issue (headless/zombie Rules) and are working on a fix. Advised Iāll be contacted once resolved.