They could certainly replace Routines with RM. Routines are pointless.
They do provide you with a button that you can press to make things happen (which of course you could do with RM and a virtual button/switch) but other than that, yeah.
Actually, that could be carried over quite easily.
Routines are just a handful of actions, with a very limited āAutomatically performā function. Pressing the button runs the actions.
I agree. I was looking at the routines tab a couple of days ago and thought, this is stupid. I can do all of this with RM and it works!
So are you hinting at buttons being implemented???
No, because there really isnāt any way to do that.
I have a virtual momentary button that I build into my rules as a āfire when I want it toā trigger. I use it on things like my wake up ruleā¦ If I get up early or ST fails to turn off the trigger then I can fire it.
I have a bit of a snag Iām hoping RM can solve. Iām running a routine that will run when motion is inactive at point A (condition) and starts at point B (trigger).
However the distance between the two isnāt far and the sensor at point A is still reporting motion in its blind state by the time anyone passes by point B, so it fails to run.
Any ideas how I can delay the evaluation of the condition rather than the action (which I think itās all I can do currently)?
BTW Iāve tweaked the sensor to reduce the blind state but as itās in a busy area Iām worried about battery drain and itās still not quite enough anyway.
TIA
Newest way? Really? Itās not like he just developed it yesterdayā¦
I donāt know how you can algorithmically overcome the lag problem. Even looking at it after the fact, is basically too late, isnāt it?
You can use one rule to launch the evaluation of another rule, and have the first one delay that action. The second one evaluates and takes action accordingly.
November 2015 can still be considered new.
Bruce,
Is there any way to set a rule/trigger such that it only can run once in any X minute period? Basically, I am trying to set something up for an elderly person with a motion sensor in bedroom and bathroom. So the rule is motion in bathroom within 2 minutes of motion in bedroom and if not it would set a trigger. The problem I have is that when the person comes back to the bedroom (after going to the bathroom), the trigger goes off because there is motion (again in the bedroom but no motion in the bathroom).
So what I need is an āadditional optionā for the rule/trigger not to be tested more than once in any given [X] minutes or the rule test to stay fixed for a period of [x] minutes or have the ability to have a trigger be any motion in the LAST [x] minutes.
Is any of this possible?
It would be possible with a Virtual Switch.
Use motion detector being active in the bedroom and a Virtual Switch for the bathroom. Tie the motion detector being active in the bathroom to setting the Virtual Switch on. Being a switch, it will stay on.
Set a rule to turn the Virtual Switch off after 10-15 minutes, whatever time period will work best so it resets itself.
The only potential issue I can see here is not having any sort of notice that they made it back to the bedroom safely, although I guess that could be done by setting up a rule with a trigger of bedroom motion and a condition of the Virtual Switch being on.
If my internet connection drops will a rule still work?
Iām pretty sure there was no motion in the house and my wife was out so it should have been set to away but I donāt see any activity in my notifications that the rule happened.
Cheers,
Matthew
No. Rule Machine only runs in the cloud, as do all SmartApps except Smart Lighting.
OK, that explains it. Is there a way to follow how a rule has fired.
Currently if I look at recent for a device it shows when it open/closed, fired, etc.
Cheers,
Matthew
You can see them in Live Logging in the IDE, they log the events they receive and the outcomes.
Thank you!!
Been looking for this and will try it out! THX!
Bruce
Thanks for all the help, Iāll dig into that later with my next test.
Cheers,
Matthew