Am I missing out on Node Red

Hey ... I guess size really does matter?
:woman_shrugging:

2 Likes

I tried Node Red 6 months or so ago. I found it nearly impossible to use, due to documentation. And when it came time to troubleshoot why it did not work, same thing, sparse or no documentation.

I was able to install it, but it seemed like it was missing something. I assume I installed what I needed to? I could never figure out how to even get a basic set of devices, or anything resembling anything that looked like a rule. I did get a bunch of pretty rectangles that had some weird sounding functions though!

Home Assistant, Node Red, MQTT, and other supposedly wonderful things are just out of the reach of those that aren't programmers or familiar with programming languages. Like many things Home Automation, if you don't understand it, good luck. Almost nobody will take the time to either dumb it down step by step for the rest of us, or fix it so that it makes sense in plain English.
:angry: :face_with_symbols_over_mouth:

I had the exact same thing with MQTT, I could never get even a peep from either device, they just sat there and looked cool. I know there is something missing, but every piece of documentation says crap like "Enter Secret Client" and you don't know how, were, or what that means, and there is zero further explanation anywhere. "Create a broker" LOL wut?

:exploding_head:

Even Hubitat is terrible for having the documentation and the text in the UI written for programmers rather than average or beginning users. I don't need paragraphs of explanation, but things like "Additional Switches to Turn Off When Turned Off" (direct quote from Motion Lighting) are ambiguous at best, and confusing for new users or if you don't muck around in ML everyday.

5 Likes

I'm not going to deny that, but things with node red are getting better now. There's an interest within the group, so responses are better. I've even gotten great responses for questions I've asked and accompanied that with examples into the forum. You should try it again. There's some great documents in the community now about what to load and how some things work. Once you get a visual. It gets better, albeit it still beats the heck out of me.

This is very true! You can share or use automations made on a completely different platform, swap out a few of the platform specific nodes with the equivalent ones made for your platform, and you're done. This opens up a new world of automation potential.

2 Likes

Actually, Node Red is hated by most true programmers because for the majority of your automations you don't need to know a single line of code. It's designed to be visual.

I had to watch a few youtube videos and it took me a good day or 2 to wrap my brain around it. But I challenge anybody to fully learn groovy in a day or 2 of playing around after only watching a couple youtube videos. :rofl:

My biggest advice to anybody trying it is to stick it out. There's tons of examples and youtube videos on how to do everything from the very basic to the most complex. There's even a thread on here with over 3k posts, I'm pretty sure it is the largest thread on this board (I think it's time to create a node-red subforum here!!)

2 Likes

It was brought up, but declined because the admins felt like it would look like something that HE supported. although I agree that a subforum would be really nice.

I get it. Unless they created a new category section for "Third Party Applications" with a disclaimer of "Hubitat assumes no responsibility for any actions caused by the use of these applications". With a 3k post thread, the need is clearly there.

1 Like

There’s a tag for node-red, if that helps.

1 Like

That is just insane. How is anyone supposed to find anything in there?

1 Like

I find if I post for help there. I rarely get an answer because that thread has progressed way past the beginner. And It's nearly impossible to search it because it's hard to know what you're really searching for.

2 Likes

Also a good point. I was following it from the beginning, making it marginally easier to find things.

There’s loads of gems in there - just hard to find them.

1 Like

Same, I was following from the beginning and jumped in around post 400. I've had to re-read that thread at least twice - but only when it was <1000 posts.

2 Likes

Well, you can search within a topic when you're in it but that doesn't help if you don't even know the lingo around what you are looking for. :roll_eyes:

Their community is a huge disappointment coming from this one but there are people who will help... when they get around to it.

1 Like

Do you mean Hubitat/Node-RED? or Just the Node-RED community? If it is the latter, then I agree with you.

If you mean the former - post your questions in the giant node-red thread. It will be answered. Might take a day or two, but it will be answered.

2 Likes

Sorry I wasn't clear about that.

1 Like

Having had to train many people to use node-red in my personal life and at work, I've found that if you read the node-red docs AND work the tutorial examples you usually have a good basis for moving forward.

Where I see people have trouble (the most) is when they read none of the docs, worked none of the tutorials, and tried to learn it by importing complex flows and "figure it out".

Now, that is a generality - and not a specific comment about any one person.

2 Likes

I'll agree with that, at least for getting started.

Yeah, just the basics. You won't be an expert by the tutorials alone. But at least coming out of that you will know how to use debug nodes to troubleshoot, where the online info/docs for a node are located, had an overview of the "flow" order of execution, and have covered most basic node types.

When engineers at work come to me and say "my function/switch/change/whatever node isn't working the way I expect". 1st thing I ask is - what does the msg look like coming into or out of the node? If they say "how can I tell?" I know they didn't read the docs or do the tutorials.

So then I give them homework to do and tell them to come back tomorrow if they still need help. (I'm getting soft in my old age, and sometimes I'll at least give them a hint that they need to use the debug node)

1 Like

:fu: this

It seems to me that RM is the weakest aspect of HE from a userbility and a performance perspective. Having come from Stringify, with its drag/drop UI, RM is extremely clunky. And I've had many rules get corrupted over the last year. Also there is no way to share rules. Yes, it's ultimately quite powerful but its complicated and slow to learn and deploy. And the experiences here and in other posts suggest it's become something that the current hub struggles to run. It's also restricting further development, with some suggested enhancements not being considered because RM is already too complicated (according to the product's author and authority on the subject). I was surprised to see no additional power in the C7 although we have been assured that cpu is not the issue. It seems to me RM needs a complete rewrite.

Given the above and the various positive comments about NR, including the recommendation from @aaiyar who I trust a lot, l've decided to take up the challenge and follow @april.brandt down this rabbit hole. Which is really a bit of a shame for HE because I honestly wish this wasn't necessary.

2 Likes

On the other hand, "If it works..." something something "don't" something something "fix it." :wink:

1 Like