Hi guys!
Okay, before I start I need to admit that I am no developer, programmer or have any knowledge in json, java or all the fancy names. I get by by copying code and to some extent learn along the way. But it is quite often testing my patience and is very time consuming.
When my interest for home automation got more of an addiction, I threw my money on a Homey home controller. A little ball that handles all protocols with a really nice UI. A bit limited if you ask the hardcore home automation people, but terrific entry product. I have switched more and more to Home Assistant, but still miss some of the nice features of Homey, foremost the flow (automotation ) engine.
I know this request is huge, but why not. It looks a lot like Node Red but is far far simpler to use for newbies but also more experienced users, which also provides a better overview of complex automations. It does NOT require any programming knowledge, the interface knows your devices and depending on what trigger/action you chose will be able to select the relevant devices.
The idea is that this might work in the same way as the automation creator looks like today, where you can toggle between yaml view and the existing. This way, you can create your automations natively in HA (and not in Node Red), and you can toggle how the automation should be presented.
Here is an example of when to charge your Volkswagen EV
Implementing this feature would have a lot of benefits, but obviously HA would attract more beginners, but it would also be benefical for also experienced users to get better overview over complex automations, but also help with troubleshooting.
Thanks for your feedback. But I am not interested in Node Red, I propose to have this automation engine in Home Assistant natively. Just as an easy option to view the automation in visual way just like being able to view it as a yaml file as you can today. Node Red is really good for some people, but if you have no knowledge about some coding, even on a really low level, Node Red is a bit too much.
I use Node Red and have it installed, but it is not convenient to use for me.
So you want something that works almost exactly like Node Red, but it must not be Node Red. Hmm. Why reinvent the wheel, especially for something as complex as that ?
Okay, can you create an automation in HA, and then go to Red Node and see that in a flow instead? Or why not, the other way around, create a flow in Node Red and pop in to HA and see it in the ordinary automation view? Or toggle to yaml view?
I hear what you say, it is similiar to Node Red, but not the same thing. You might not see the benefit, but I do. I know more people close to me have expressed the same wish. We might just be 10 people out of the thousand and thousands of HA users, but I guess I am free to put a request here?
Youāre definitely free to submit Feature Requests.
Community members are free to comment on the proposed Feature Request.
Most Feature Requests are never implemented.
Those that are implemented are usually done by a volunteer software developer. So the Feature Request must interest them sufficiently in order for them to invest their time and effort. Plus they must gain the approval of Home Assistantās development team.
Home Assistantās development team must ultimately approve inclusion of the new feature because they will be assuming responsibility for it.
A wireflow type of ui for automations in Home Assistant would be a great addition imo too. I am a more a graphical oriented person rather than a code oriented person.
Think it can also help to make the total number of set automations of a user to go down and give overview. Sometimes I have to look trough all my automations to determine which one I have to edit. Which also might have to do with the way I set up my automations. But usually when I create an automation I am glad it is working and donāt take the time to rewrite them so I have fewer.
I know Node Red and am using it (not for my automations), but still coding is often needed and I think the UI is too difficult for the not so technically skilled user.
Absolutely lobby for it thatās great but to set proper expectationsā¦
You are basically saying you want node redā¦ Without the complexity of Node Red. Ok i can dig it.
A visual UI wire frame editor that allows ease of creation. Ok ill buy that. Butā¦
The problem stems from the fact that HA is NOT. Hive, or Homey, or SmartThings or any of the standard consumer facing products marketed at the general home user. The complexity comes from HA itself. Build your UI tool. By the time youāre done supporting all of the things you need to support to make this workā¦ Multiple types of nodes. All of the various interfaces for interaction types (i mean cāmon just look at the number of options under helper types aloneā¦)
I used to use WebCoRE in SmartThings. It was light years beyond what ST could so natively and as such it was WAY more complex. And as such there were users that didnāt want to use it because it was ācomplicatedā (hint webcore is still way less capable than the HA builtin editor and to get feature parity it would need to get an order of magnitude MORE complex). In STs rewrite (they basically ripped out the core and built a new one based on Lua) they built a new graphic wire frame editor that looks like NR but more simple and a rules engine. Guess how much either are used? Only very advanced users use the rules engine and the simple editor is barely capable of doing light automations and itās already as complicated as NR.
So you build the basic visual ui - itās simple. But now it needs more features to support HA. More features more dev, more controls. Some things only work on event triggers or service calls so you HAVE to support those. Basic templates will be a part of any editor (count the threads asking for template support on basic cards and automations.)Now itās getting more complex. Andā¦ Suddenly you haveā¦ Node Red.
Homeassistant is a fantastic product but with its flexibility comes complexity. We all want it to be easier to adopt but unfortunately thereās some things you just wonāt be able to change. Here that thing is HA is complex and to write an automation editor for it that is reasonably capable itās complex too, and will start at a baseline of complexity you have to assume will be there.
Im not saying that its not a lofty goal. But you really need to take a step back and wonder if what you want is even POSSIBLE because you want āless complicatedā
I agree with you on that it has to doable and that it should give added value if done. This is just a feature request I am supporting from my perspective as a user. If there are good reasons to not to choose to spend time and effort in implementing this, thatās all good.
As I read your reply, I looked up a tutorial video by Homey on youtube. The flows in that video seem pretty complex, but the interface seems way more accessible to me than Node Red. The language used is more understandable at first look.
The video:
When I got to know Home Assistant, I quite early added Node Red, but didnāt understand it at first sight while I am fairly computer savvy. You have to quite understand the nomenclature and kind of objects there are within Home Assistant to work with Node Red. And read tutorials and/or watch youtube videos.
So the learning curve of Node Red is more steep then the current native Automations UI, imo.
When a user sees the current Automations UI of HA, he/she will probably easily get something working. I donāt think thatās the case for Node Red for someone that hasnāt seen Node Red ever before. Maybe that will always be the case with wireflow like UI, maybe it is also the way Node Red currently is presented to the user. Although Node Red is an awesome piece of software though.
I donāt agree on this. I do think I am looking at it more out of the perspective of the general user. I think Home Assistant could evolve into software that serves both the general user and the more advanced user. And also think that is the goal of the founders and most contributing developers.
At the moment already a lot can be achieved via the UI. A lot of widely available consumer products are easily added to Home Assistant and itās data can be used in HA. If a more advanced user want to do more complex stuff or use more advanced type of products or fringe products, that usualy can be achieved via YAML or with templating. The templating is easily accessible via the UI when needed and the more general user can leave that alone (altough it is to expect that stuff done within templating can be more and more done via the UI). Such design choices seem to serve both groups very well.
If Home Assistant would be more appealing to general users, altough ofcourse it already is very appealing, the whole project would benefit.
I am a big Node Red user as 95% of all my automations occur in Node Red. Yes it took a few weeks to get the hang of it but IMO having a visual flow of my automations just makes sense. I am pretty sure Homey got the idea from Node-Red and just decided to try to make it palatable to the general consumer. Which is great but lets all be realā¦
WE ARE NOT THE GENERAL HOME AUTOMATON CONSUMER
Home Assistant is power user mode, the majority of us who use it probably use it for that reason, because we can basically pretty much do whatever the heck we want. Node Red works perfect in that sense as the amount of access it has to pretty much everything in HA is insane. For example, I use it to change the LED lights on my scene controllers to status me on things around the house. It is things like this which a more consumer grade solution would probably not even mess with. That said I am all for it, just do not know how much traction it would get.
What do you consider is the general Home Automation user? Do you consider the Homey users genaral users? Or are those power users as well?
Do you think Home Assistant is so much more difficult to set up than a random brand smart lighting system?
I think it is possible to serve both the power user and the general user in Home Assistant. I canāt see why not. Yes, the general user is maybe satisfied with a some easier automations and the power user would like more advanced stuff. I think both can be served.
I canāt speak for the people at NabuCasa that determine the direction the project is going. But when also general users are appealed, the more people use Home Assistant, the more manufacturers will be open to make or help make possible integrations with Home Assitant. And also the funds for the project will rise. Which ultimately gives more improvements to the software, backend and frontend.
Sure, but does it need to be like that? Why canāt we have that, and at the same time also include ānon-power usersā that are just starting to dip there toes in this space?
I think the narrative provided by Paulus, Frenck and the team is to make HA more easier to start with, for instances the new additions to the Automation part.
