No, I don’t think so. At the very least it is not something I will not work on as it is really just a stop-gap to avoid adding this to your cards. Which I can understand can be tedious for updating an existing large config, but should really not be a big issue going forward, in my opinion.
1 Like
It was in the beta for a week and no complaints were raised during that time. A PR open for a week and a week in beta is, in my opinion, far from being rushed 
I also can’t keep my ear to all the feedback on the forums. There is a LOT going on here 
I focus mostly on what is happening on GitHub.
Gotcha, so no plans for color state options? It would make it sooo easy to set up how state colors behaves for different entities. For example, using Custom UI (which is buggy as hell but somehow still works) my global config looks like this:
"switch.*":
templates:
icon_color: if (state === 'on') return '#fdd22f';
"binary_sensor.*occupancy*":
icon: mdi:home
templates:
icon_color: if (state === 'on') return '#FF1C8C';
"binary_sensor.*window":
device_class: window
templates:
icon_color: if (state === 'on') return '#5a5a5a';
This is easy as I can just name my entities a certain way and they will be colored accordingly.
Not by me 
I’m just one dev though and a selfish one who usually makes thing that I want 
I could see an icon_color option being added to cards, but I doubt it would be done the custom UI way, i.e. globally. But if icon_color becomes a thing, a custom solution to set those by templates becomes much easier.
I actually proposed icon_color previously, but don’t recall why I closed it…https://github.com/home-assistant/home-assistant-polymer/pull/3956
Ian, (I presume) that’s because most of the beta users are the very same people who propose changes or make PRs. I personally only update when .2 comes out as I cannot afford disrupting my setup. (I also open issues and sometimes make PRs, but not as great as yours 
)
And still, I think now you have at least 2 FRs on your hands related to the lack of functionality in various types of cards making them difficult to use after the change. We can call it whatever we want but it could have been less stressful, and it’s not only about time - more importantly, it’s taking into account users’ opinion.
If you meant that 2 weeks was enough for everyone to get familiar with that PR… I don’t know who reads all these PRs before they become part of HA. I got a link to it from the release blog when we discusses “why could it happen at all?”
but did you try to gather it? like a poll or something? if no, then yes, you can’t
I believe you and that’s great, really!
However,
it’s ok but do you really think that your discussion with @SeanM and @balloob could for sure take into account needs and use cases of say, 50% of HA users registered on this forum alone (50% of roughly 50k If I remember it right). Or 10%?
I’m happy either way as I can see HA progressing. But its progress ultimately relies upon its user base, which in turn is directly influenced by relationship between users and developers, by the way developers respond to their needs and issues - just because users are here if the product satisfies their needs.
Beta users will stay for sure as that’s their hobby, what about the majority? Everyone (both developers and users) should understand that.
1 Like
I can understand frustrations on all of that, but at the end of the day, I’m just a guy who likes poking around in a project in his free time. I don’t really want to be more than that.
I mean, we try not to break things, we try to make informed choices, we try to make things better. But asking me or anyone really to talk to 50k users, to do case studies and user polls and filter through all that to make a change to icon coloring…that’s not something I’m going to do. But I’m not a core dev, I’m just a guy, trying to have some fun and make cool things
3 Likes
Got it, thanks
2 All: is everything clear now?
I’m not entirely convinced that “no news is good news” during testing. No complaints may also be a result of:
- superficial testing
- exclusively focusing on pass/fail
In contrast, within a day of the feature’s release, members of this (large) community forum were reporting on its impact (reported in this thread and others and in GitHub).
I feel the community is a resource that ought to be leveraged earlier in the development process.
1 Like
Except that he is creating code, not just consuming it and, therefore, his point is also very important.
This sounds reasonable at first, but it ends up just opening a can of worms. All communities would (rightfully) want equal treatment. This forum has 55k users but the Home Assistant Reddit community is even larger. So it’d make sense to discuss it there as well, right?
Now you’re already up to 3 different locations, why not make it 4 or 5 and poll Facebook and Twitter too? Where do you make a cutoff that doesn’t make certain sub-communities feel left out? And now if you only participate in the forums (as several people here have indicated), you’d only be seeing a small fraction of the entire discussion. It quickly spirals out of hand, and places a very large burden on the contributor too.
GitHub is specialized for this exact purpose, the frontend repo has a feature request template and a bug report template. Things can be labeled and organized into roadmaps, easily cross-referenced and searched, etc. It’s far more efficient for all involved if the pre-release discussion happened there.
And this thread already has a good example of the current system working fine, @TazUk raised a valid issue using the proper avenue and @iantrich is adding support for it. Ian also improved the icon coloring for alarm control panel yesterday as well (here) based on feedback. Feedback is always welcome, just try to raise these issues in the proper spot for them.
2 Likes
Fair enough. I’ll monitor future frontend developments on GitHub (and chime in if I feel I have something worthwhile to add).
I wish the backend repo also had a Feature Request template. The current template explicitly states that Issues are for bug reporting, not feature requests.
This forum has a Feature Requests sub-section but, as everyone knows, that’s where ideas go to die (i.e. very few are ever implemented).
FWIW, I’ve commented on some PRs in the backend repo but, typically, only if the PR affects me directly. Overall I would say most developers are open to suggestions.
2 Likes
I agree, that’s a very accurate description of the Feature Request forum.
Most of those ideas (ie new integration support) require the purchase of expensive hardware, learning or reverse engineering an API, hours and hours of coding, etc… So it’s somewhat understandable that very few of those requests ever get implemented. I assume that’s the main reason feature requests aren’t allowed on the backend repo, they’d be just as dead/ignored over there too, with the additional downside of making it harder to find bug reports etc.
But feature requests are very much encouraged on the Frontend repo. A big focus right now is on making Home Assistant easier to use, and there’s now a full-time employee (Bram) focused on the frontend, so it’s been getting a lot of attention lately. Any ideas, critiques, feedback of any kind is appreciated there
My point was “he’s not the one who gives green lights” and he clearly is not in the position to address this completely non-developer issue.
I never underestimated his opinion.
I’d discuss that. Is this forum a part of HA ecosystem? I mean, it looks like part of https://www.home-assistant.io and people can access it from the main page of HA website.
There is no links to Reddit group there, is it? And I think we don’t need it - I believe it’s clear that if you want to know about something about a product/have an issue with it/want to be heard you head to the website and do it there. So I’d call this forum “an official HA forum”. Does anyone need more than one? I’d say no. There can be numerous “unofficial” ones but there is no need for more than one.
And on this premise your reasoning is not correct. HA authorities have the right channel and plenty of opportunities to gather user’s feedback. Here.
How about the majority, mere users. They are often far from programming at all and not that proficient in finding ways to that “holy Graal”, let alone the fact there is no clear link from HA’s front webpage to its Github repository.
Yeah, it is possible to break a leg and then heal it (but it won’t be as new anymore). We’re suggesting here some measures how to prevent breakages because it’s less painful (if it’s your leg, of course).
I believe there are different types of feedback and HA needs both. You’re talking about reporting issues. We’re asking for a feedback about upcoming breaking/serious changes as it can easily eliminate a lot of after-break issues. You just cannot say it’s the same, can you?
I know it can slow down the development process a bit but it depends on how it’s organised.
It seems like it’s down to @balloob/HA management (as there should be some), not to contributors. They need to integrate this bit into the development process and make sure the good balance is achieved.
I hope they can and will do it.
FR graveyard, very grim place.
I don’t think the majority of them are for new integrations support. I have a feeling it’s core enhancements (like ability to use variables in templates) so I have to disagree here.
Way to nowhere, honestly. FRs should be collected and regularly and carefully considered/re-evaluated (as HA evolves) by a panel and then prioritised. Some of them will be rejected, but not all.
That’s great. But in the last few days I saw at least two issues related to this state_color change closed swiftly with a short “it’s by design” comment. No further explanation given. So what do we have out of that feedback? Nothing. Would the person who opened such an issue do it again? Probably not as it’s not the greatest experience.
I know people are different and there are rude users and kind devs but it’s wrong to threat all users as an unnecessary annoyance when they bring you a new issue and ask for help. (I’m not talking about you @SeanM, you’re great!)
It’s not going to change so if you want an input use discord/github otherwise you’re just talking to yourself.
2 Likes
people vote with their legs.
Sorry, but IMHO your posts are too lengthy and repetitive.
No problem, I’ll leave the party.
Please don’t feel offended, it was not my purpose. Sorry again. I believe you want something better as well as many others here. IMHO, the conversation became polemic and not constructive.
I’m leaving the topic anyway as I cannot justify spending my time here - there is nobody to convince. People who can change the situation are unreachable.
As for my receptiveness and lengthly posts - I tried to show devs the situation from users’ point of view and some of them didn’t seem to appreciate pretty obvious things so I had to explain further or show their own contradictory statements. It’s never too easy.
Anyway, I really need to do something else.