With hubigrphs no longer being actively supported, is there an alternative available that does something similar? Not that its anything i NEED, but more just idle curiosity.
influxdb v1.8 + grafana running on my 24/7 NAS but any 24/7 computer will do.
What happened to hubigraph developer?
Apparently he left the platform and is no longer supporting it.
Wow, I really feel for him. What platform is he moving on too? Not Home Assistant surely?
This is an important post for folks to read. It clarifies some unspoken truths that lots of old timers accept as being the terms of ownership, "that's life in Hubitat-land" in other words.
OK, fine, but from a new USER perspective -
How many more examples are we going to see of highly useful applications that industrious Community Developers introduce, and achieve solid Proof of Concept/Benefit with, that never get integrated into the system so that said developer is stuck being an unpaid volunteer as system upgrades and user input cause impossible expectations. And the worst outcome is that HE users become reliant on a helpful product without a long term sponsor.
But oh, if you never upgrade it will always work is the line I hear. Like that's what we all want to do.
As said above influxdb with grafana will do it all. You can run it all on a raspberry pi as well.
+1 from me too.
I think some of the disconnect with the original developer was he wanted to build some kick-■■■ extensive visualisations. which i would not consider the best use of a Hubitat hub/platform (except as a status/data source). Hence @djashones solution.
i - and i expect a good percentage of folks - just want a basic chart or 2 without having to build/maintain/power a separate device.
i wish that app was still supported...'cause i'm not aware of an equivalent.
It's a shame, but hubigraphs still works and is on the github
I try to keep my hub as standard, only having drivers for unsupported devices which I wish HE would adopt and I would be happy for them to for the ones I've authored
But out of the box HE is lacking a few key things
Hubigraph to display above
For me these are key tools to monitor the performance of the hub.
I don't look at them every day but now and again to make sure everything is running smoothly
And Chromecast is still in beta so Chromecast helper
This is an awesome idea! I was working on a project I was calling "HubiChart" that I wasn't targeting for Dashboard integration immediately (just a way to view device history; was going to try to figure that out later if people asked) and was using chart.js instead of Google Charts--bascially a way for me to finally get rid of Home Assistant, which I was keeping around more or less only to (ab)use its simple built-in history graphs/charts. My problem is understanding the chart.js API. I might just give up and start using yours instead.
reading through the Hubigraphs post, i spotted this from a certain ubergenius app developer... sound familiar @bertabcd1234 ?
now that hubigraphs has reached the end of the road, i don't suppose you spent more time on your graphing project?
Nope, I gave up when HubiGraphs became popular...and never got the hang of ChartsJS enough to really get my project going any farther. Hubitat itself is also a bit different now, with the default event history being significantly less than it used to be, and my intent was always to read that rather than have the app subscribe to device events (I think this is part of the reason HubiGraphs created the Long Term Storage app). So I'd have to re-think some things if I did.
I do have some history graphs, but I just use Home Assistant for those instead. I don't really have any automations set up there; its default visualization capabilities just happen to do everything I need out of box, and I can do minimal customization to get things displayed a bit better if I want. Hubitat integration is easy with Maker API on the Hubitat side and an unofficial component on the HASS side (HACS makes it easy to install and update it, but it's not technically required--sort of like HPM). Of course, this is intended to be full-blown home automation "hub," but there's no reason to use it for that if you don't want to (I much prefer Hubitat, particularly for automations). Just this part works for me!
Probably almost all of them. And there are many IP, commercial, and legal reasons for that.
I'm really not sure what answer you want/expect?
Hubitat obviously wants to continue sales, so will keep improving the product in ways that make commercial sense to them. They've already said they will be improving dashboards in the future, for example.
But thinking that they will "buy out" community apps and integrate them into the product is likely never going to happen (outside of very small/select cases).
Clause #4 of TOS..
- Customer Intellectual Property. Customer shall be permitted to create its own derivative software from the Software for its own personal use (the “Customer Software”). “Customer Software” means any and all software that the Customer writes, designs and develops to supplement Customer’s use of the Software that does not include the Software, including Apps and Drivers installed on a hub in source code form. Hubitat does not claim any ownership over Customer Software. The foregoing notwithstanding, the Customer acknowledges that the Customer may not assign, sublease, distribute, share or otherwise make the Customer Software available to third parties for commercial purposes; provided, however, that Customer may distribute and share Customer Software for non-commercial purposes for no compensation. Hubitat shall have no backup, improvement, maintenance, support or other obligations whatsoever with respect to the Customer Software, and Hubitat shall not be responsible for Customer’s ongoing ability to use any Customer Software upon the termination of this Agreement.
I've got a hunch there are many Apps, maybe not the scale of Hubgraphs, that the initial developer would say, "yeah, have it...it's yours...I'm just happy you like it, will integrate it, and I won't have to support it anymore".
I really don't get why so many take this "no way" position on this topic instead of celebrating the ability to prototype good ideas for HE and then GIVE IT TO EM to integrate. EVEN IF THEY HAVE TO RE-WRITE IT....the benefit is in proofing the concept and seeing broad user acceptance.
That is likely the problem. I would bet they get a vision for what they want and then implement it in a monar to minimize support needs and maximise usage from new folks. I always hear how small the team is that they have. It could be very simply that they don't want to take on a dozen or so community applications and the responsibility of supporting them.
I also think there are allot of things the community has done that they simply just don't agree with. Lets just say i got a fairly firm response from someone on staff talking about wifi devices which amounted to shut up, they don't belong in a HA environment. The more I have used the platform I completely understand that response. Almost all of my issues that i have found have been related to Wifi devices. I still believe that is a very narrow minded answer though as the platform needs to be able to respond to what is coming to the market just as much as what is already here.. Allot of community apps help with that.
I honestly never understood the reasoning behind running Hubigraphs from the hub itself. It just seems like the wrong hardware for the job. I get that the desire is to keep things on one device, but we also have to keep in mind what the hardware is.
I also think they are constantly trying to add functionality without exposing themselves to much. I know in one of the last few releases they exposed some new libraries for use by developers.
I think what they have now is great, and the real solution to the power user would be a Hubitat OS as a VM or appliance OVA for running on Wintel platform. Then you could centralize the HA functions next to a more powerful database engine like InfluxDB, A supplementary HA platform like Home Assistant or Node-Red, and a Dashboard engin line Grafana. But this goes against two of their core values i believe and will likely never happen. Again i understand why. Especially when i can do all of that with a second box be it a PI, Wintel server, NAS with VM, or simply a PC. This is exactly what I do. I have Unraid with Node-red, Influxdb, Grafana, Motion Eye, Nginx, and other applications always running. Heck even a single raspberry pi could do it all if i wanted to set it up.
Agreed. I do graph some data acquired by the hub (temperature, humidity etc.), but use an external process to do it. I'd rather have my HE focus on controlling devices and running automations.
Well there are just some really good specialized tools for this. InfluxDB and Grafana really are designed for this kind of time series data and visualization. Trying to make something from scratch and not have issues on a powerful platform can be difficult enough. Doing it on a small shared arm platform and trying not to impact the more critical processes is a real pain even with proper tools to setup work load management. Storage concerns is also a big problem it doesn't take long for that database to grow to unreasonable sizes depending no how much you are capturing.
Back to the original ask if you want a alternative I would suggest using InfluxDB with Grafana on a Raspberry Pi if you don't have a Nas or always on computer otherwise.
I always suggest the 4GB model simply as the best bang for the buck now I believe. Get a Flirc case and a decent 5v3am power supply. Instead of Micro SD card get a usb connected SSD drive so that you don't have to worry about the SD Card failing. The Pi can boot from USB drives once it is current on firmware and such(you may need a 8GB micro sd card to update it the first time). Once that is done load Raspberry pi os and install InfluxDB1.x whatever is avaliable that isn't 2.x. Then use Grafana to visualize the data you want to display. Get the InfluxDB community app from the forums to load it. There are a few good tutorials for setting it up with ST and the same thing can be done for Hubitat except substitute Hubitat anywhere you see Smartthings.
The other option is Node-Red on the Pi as well which can open up a whole new world of automation as well. It has a dashboard pallet, Though i have never actually gotten it working.
With a Raspberry pi you can minimize your power consumption since it will sip a very small amount of CPU while running
I agree with HE idea of automation without the human intervention, I only have 1 permanent dashboard where thing are to variable but still in there the only buttons used is the tv power and blinds.
Like I said earlier they just need some back end tools.
I considered myself as just above normal consumer where as most people here are super users.
But this is still emerging technology, we are still early adopters, I only know 1 person who remotely has anything like my system
Speaking from experience, I've almost never found it faster/easier to take someone else's work and re-write it - unless they were following the same coding style guidelines I am using. So I disagree with that premise.
But anyway, Hubitat will do what they think is the best approach to development and their product I guess.
Our opinions are just idle thoughts/chatter in the end, spending other peoples money and support resources.