I don’t mean to sound snarky or passive aggressive on this at all, but I’ve spent the last couple days playing around with the Rules API… and it blows my mind that there is not a web based rules engine.
SmartThings could pull one developer off a project and have one working in a week or two, not to mention anyone else in this community that is web development savvy. It just seems like such a simple interface to develop and with the Rules API already in place… why hasn’t this been done? Am I missing something?
I’m ok with “closed Beta”… beta is closer than alpha and its on the smartthings subdomain, so it would give a pretty good indication of something fully supported by ST is “coming soon”.
They promised one early on, but then shut down the project, so I’m not sure it would just be an easy one week fix or I think they would’ve done it already.
Many community members are using the third-party SharpTools product which does have a nice visual interface. It’s web-based so will work and pretty much any browser. They have a free tier for simple rules and a $30/year subscription if you need more advanced features. It’s very popular and already works with the new architecture, so might be worth a look.
And here’s the link to their rules engine feature.
I certainly agree with you, though, I am surprised that smartthings doesn’t have its own official equivalent for rules. I just don’t think it’s only a one week project.
I may be mistaken, but wasn’t “routine creator“ the project that was publicly announced in 2021, and then abandoned later? Or am I confusing it with something else? Are you in the beta?
I have recently converted to SharpTools (from WebCore) which does seem to be a good replacement, but I am attracted to the local execution (at least promise of) of the Rules API. I could probably recreate almost every automation I have in the Rules API, but manually typing out and forming the JSON is clunky and slow. I would be really nice to just log into a webpage, and the Rules API loads up a drop down box with all of my device names and gives me the ability to apply the if/else/then statements with the applicable capabilites and device states.
The Rules API is the hard part of all of this, just need to layer a user interface on top of it.
I believe that on the whole I would be seen as a champion of the Rules API in the community, but there has always been a bit of a niggle about it and that niggle has grown with time. I think the best way I can express that niggle, and I’ll apologise in advance for implied use of a vulgar expression common in British speech, is that I could imagine being in a discussion about the project where someone suddenly says “Hang on, that’s ******** isn’t it?” and there is an Emperor’s New Clothes moment as the whole thing collapses.
I can’t quite put my finger on what it is. Perhaps it is that if you are going to run in the cloud there quickly comes a point where the required complexity means putting the effort into an easy to use wrapper for a common programming language makes more sense. Similarly when it comes to running on the hub you start looking at some increasingly app-like Edge drivers and pondering. I find myself wondering if actually the Rules API has pretty much reached the limits of where it makes sense and just needs parity of what is available to internal and external developers.
