Why remove GPIO-integration and support? BAD BAD!!! The great idea with running Hassio on RPI, is among other things the possibilites with gpio-ports. I have just ordered BMP280 and DHT22 for using with Hassio and my RPI4, and my plan was to move my last home automation things from Domotics to Hassio. Now I consider about this. Please do not remove GPIO-support!!!
As has been said many times, you can install it as a custom component
It is still possible to connect in example BMP280 to i2c (gpio) and DS18B20 to gpio and get the data into Hassio?
Having alarm done using GPIOs i would call âhomeâs securityâ. That home made device is defo not âsecurityâ device and it is not reliable. Get some cheap alarm motherboard and connect it via serial port - way higher security level and GPIO problem solved at same time.
âallowing information to make educated development decisions is treacherousâ
That comment is such a typical and disappointing cop-out response. I donât use GPIO on my RPi, but this is just another an a growing line of such âdeprecationsâ for convenience that has begun to concern me about my past (and continued) investment in HA, including Nabu Casa and the salaries being paid there. I get having to prioritize, given limited resources, which is why I support Nabu Casa, and have been thinking of additional contributions, but actions like this give me pause. I use probably 0.8% of the integrations in HA. I look at the list above, and I see scores of functionality that I will NEVER use, yet a lot of effort is going into developing and improving those efforts.
Perhaps they are pet projects of the developers, which is fine, given that this is OSS, but some of these recent decisions seem âoffâ, like the continued trend to remove YAML support in lieu of the UI, for example. I donât use the UI for any automations, scripts, helpers, or anything else, because I use comments A LOT to keep track of things, and the UI deletes them. So, until that changes, I will continue to use YAML.
I love Home Assistant and contribute when I can both on discord/forum and financially. Iâm just concerned that decisions are being made that will change the things that have been so attractive about HA. We donât want to dumb down HA in a rush to appeal to the mindless masses who can barely set up their WiFi passwords. Just my 2 cents, if it matters.
According to a previous comment, your option is âjust donât upgradeâ. Problem solved. Welcome to being among the 0.8%.
So, who makes these decisions on which integrations may get core support and which are relegated to the âunmaintained-and-soone-to-be-deprecatedâ pile? When are WE going to have a vote? I may be primarily an âend userâ, but I do work with vendors, promote HA, test new products on HA, and provide test data and feedback to vendors so they increase their support for HA in their product lines.
I also share my growing experience with others on the forums whenever I can. I also actively provide feedback to support improving integrations when I find issues. I also support financially. There are a LOT of ways people contribute to an OSS project beyond cutting code.
Very true. I get the feeling that different projects value different inputs.
I have never contributed any code to an open source project, beyond a line or two here and there. Nevertheless I became a team member on a very large project, purely by helping, and then moderating on their forums, and generally being useful. Hell that entitled me to free travel and accommodation to their annual team summit on the other side of the world!
In contrast, I have done the same here, the very odd bit of documentation and being very active on the forum? But never any acknowledgement, praise or even an invitation to moderate on the forum. Not that I am complaining, just pointing out a difference in emphasis.
I know, I see you all over the forums, and you have helped me on more than one occasion. Much appreciated.
The RP_GPio isnât maintained by anyone. as mentioned previously, if it breaks it doesnât work anymore.
A lot of integrations require specific hardware to be used, we cant expect thos who maintain HA to purchase hardware just to keep it running
In the case of RP_gpio, there are a number of options given if you wish to continue to use it, they are not blocking its use, just that it wont be in the core installation. Someone in this case has already forked and added it to HACS,. so that people can install using that or manually
Then perhaps more effort could be expended in planning for a smoother transition path for current HA users (no matter how small the percentage) when such decisions are made. Clearly, hearing about it in a Breaking Change isnât working too well for many folks. NOTE: We donât all read the developer GitHub discussions (where, I was recently told, all these âdiscussions and decisionsâ take place), nor do I frequent the Dev forums on Discord to try to get a glimpse of what planning I need to do. The last thing the principals want to do is alienate their worldwide loyal user base. I think thatâs all most of us are looking for. Weâve seen too many of these now that some of us are beginning to get a little âuncomfortableâ. Food for thought.
The thing is, it isnât removed in this release just deprecated, in the case of rp gpio itâs removed in 4 months.
Iâve been around for just over three years and I feel I can respond to at least a few of your questions. However, my replies merely represent opinions based on personal observations. Others may have different responses.
-
I canât recall anything thatâs ever been done based on a vote. This project is run as a benevolent dictatorship and thatâs a term coined by one of its inner circle (and based on all evidence, a statement of fact).
-
Unless you submit a PR, whatever other ways you believe you are helping the project is not recognized by the core team. It doesnât matter if you spend every waking hour of your life helping users here, you donât get any added influence with the decision-makers, or recognition. Itâs a meritocracy based on code production.
-
Whatever money you pay to Nabu Casa simply keeps the lights on at Nabu Casa. Its employees continue to enhance Home Assistant and provide you with remote-access services. It doesnât buy you any influence like, perhaps, a traditional customer would.
-
The few decisions that have ever been amended (I can only think of two) were the result of tremendous pushback from the community. When they were amended, frankly, it was a bit of a surprise to see it actually happen; most decisions are final.
Anyway, thatâs how it looks from my chair. It may sound cynical but I prefer to describe it as being pragmatic.
If you think four months ahead is not smooth enough and uncomfortable then please tell how to make it smoother and more comfortable.
I highly doubt it will be noted since I believe most find it more than fine.
But Iâm genuinely curious what you find appropriate
I agree; 4-months advance notice of deprecation is more than adequate. However, you and I know that even if they provided a full yearâs notice, some affected users would still complain.
These release threads are magnets for posts that complain about how some Breaking Change or another will cause the demise of Home Assistant. Yet the user-base continues to grow and so does the community forumâs membership.
We may not all agree with every decision but, on the whole, the project continues to grow and prosper.
A lot of products change over time.
There are options. The gpio as example is very well supported in third party. The existing integration is being moved to a custom in hacs. It is not even being deleted. You just have to change how you use it.
This is open source. FREE software. If things are no longer maintained they have to eventually be deprecated. That is life, that is the way everything goes.
I had a 1979 Toyota Corolla. It could not be viably or economically maintained. I had to deprecate it and then remove it. Software is no different.
I donât code. I could say things in Home Assistant have changed since 7 years ago when I started and the scripts I had 7 years ago no longer work. The thing is, there are much better, faster and more efficient ways of doing the same things now that took lots of YAML back then.
Deprecation is not a bad thing. People just donât like change.
Thatâs actually a common term and fact when talking about open-source projects at large.
Nowhere in any charts has it been mentioned that open-source implied any kind of democracy
Quite a big update for me, migration of Unifi Protect from custom component to core, Android TV and WebOS moving to UI along with all the normal changes. Was a little bit worried so made sure I had a backup and crossed my fingers. Incredibly pleased to say the upgrade was seamless. Great work to the Development team and everyone involved, thank you!
Time doesnât equal âsmoothâ, a clear path to an installation that doesnât fail IS.
TBH, Iâve never had an issue with any update. Iâve never really been affected by a breaking change either, even in this release. But I do empathize with people who put their trust in a product and feel theyâve had the rug pulled from under them. Letâs not be like Nest.
@koying nowhere in any charts has it been mentioned that open-source implied any kind of democracy
So, what youâre saying that OSS is no better than commercial SW in a box sitting on a shelf then? If I contribute to a developer or developers, does that not make a difference? Do all HA developers feel this way? I want to know because I plan (or planned) to start sponsoring a few. If we get what we get and will like it is the theme here, then that would be good to know. I could be happy to just sit back and get what I get and not participate or contribute.
@brendan Deprecation is not a bad thing. People just donât like change.
I never said it was a bad thing, nor did I say people donât like change. Nobody did. There is a LOT to celebrate here. Thatâs all change. People just donât like change that negatively affects their HA installation, and they voice that concern. That can be either taken as constructive criticism or complaining.
@123 These release threads are magnets for posts that complain about how some Breaking Change or another will cause the demise of Home Assistant.
Thatâs a bit dramatic. and IIRC, not the first time youâve posted that sentiment. I donât think anybody has claimed that HA is doomed because of a breaking change. Itâs pretty clear that HA is growing, and Iâm out there promoting it every chance I get to users and vendors alike, helping other users, and sharing what Iâm doing in HA. And just writing off what people voice because you believe that âsome affected users would still complainâ isnât productive.
Having said that, Iâm extremely excited about this and future releases of HA. Just in the two years Iâve been using it, HA has matured and grown at an incredible pace. I donât believe any other ecosystem even compares!