Unless I missed it, I don’t think @adorobis ever made such a claim.
There is a lot more agreement overall than disagreement, but also a lot of talking past each other over technicalities.
I think maybe you read more into the statement than intended, maybe I’m not reading too little into it.
I remarked on another thread about the weird development dichotomy right now. Chasing consumer market windmills but still at the Agile alter of ultra short release cycles. The two don’t mix well. I don’t think they mix well even for a “pro-sumer” target.
It should be very doable, yes. Bit at which cost?
I am concerned less about revenue and more the many bugs that need fixing and will not be fixed because resources are dedicated to this new project.
Voice recognition is many things, but new or visionary it is not. It is old, well established and used for years in many far more complex systems. So it is nothing flashy or new.
And I question if a feature that is knwoingly implemented in the most basic fashion will be worth the hassle given the limited interest of people and the limited applications it will support. Especially compared to existing solutions.
But anyway, we will see and we will have to hope that HA does not turn into one of those projects that tries to constantly add new stuff purely for the joy of adding something new whilst ignoring the many severe bugs that actually matter.
P.S.: I understand the thrill of doing something new versus the hassle of fixing the old. But unfortunately, ignoring problems and instead playing only with new stuff results in poor products/results with no long term future (well, quality future… there will always be the masses like Apple users that do not care about quality/technology and only care about marketing).
I doubt it is that. More likely trying to differentiate their product and broaden their appeal. A shiny new bullet point feature to shout about can bring in more new users than a laundry list of little improvements. They have enough funds to hire another developer, there is a calculation about what will bring in the best ROI.
HA is Nabu Casa’s product at this point much more than it is a community project. There isn’t anything necessarily wrong with that. I get annoyed every time I see post about some criticism being to harsh for developers volunteering their free time. There certainly is a lot of volunteer development, but most issues where that line is thrown out in the forums concern nabu casa maintained code.
That said, I commend nabu casa in keeping HA fully open source. So far they have avoided the pitfall and annoyance of closed commercial add on modules and such. I pray they keep it that way.
If they think this will grow the product and related revenue, then so be it, but I have my doubts.
Some of the recent “little” changes and updates, or more precisely how those changes were handled, are more concerning to me than chasing voice.
Based on what I’ve seen in the past four years, severe bugs are fixed promptly. So the good news is that nothing is allowed to cripple the system for (too) long.
In contrast, little bugs can persist for a long time. One bug I reported about 2 years ago was never fixed (re-enabling a disabled entity shouldn’t require a restart but, in practice, it’s needed, despite an on-screen message claiming otherwise). There was some initial discussion but then it was abandoned. Another bug I reported about 2 months ago (contrary to the documentation, expand fails to work as a filter when used in Template entities), has yet to receive a reply.
One could chalk that up to insufficient human resources and/or a severe bug triage protocol. Either way, it means some bugs don’t get fixed promptly, or ever.
To be clear, not a complaint, just a personal observation.
I’m old enough to remember the first Alexa. It was magical. It was sci-fi become reality. You could just say the magic word and it would hear you! To start with, it wasn’t great at understanding what was asked if it, but the fact that could just say the world and and 99% of the time it would hear you, wherever you are in the room, that was the magical thing. Finally, it was as if God was listening! It was like being in the deck of Star Strek Voyager! Young and old, everyone was compelled. I had voice control on my computer in 1999, but this was the future!
Voice assistants are far more than voice recognition. Voice assistants are about the experience, the ease of use that speaks to our inherent laziness. Voice recognition is part of that, sure. A small part of it. But let me tell you what is huge part of it, 90% of it: Beam forming microphone arrays. These clever microphone arrangements and their algorithms running on DSP chips ensure that, wherever you are in the room, even turning your back to the device, with music blasting, it will hear you!
As far as I know, there aren’t any open source solutions in that domain. Do we’re stuck with a point source model. That means standing a few feet from a microphone, talking straight into it. Kinda breaks the magic.
But maybe someone far cleverer than me had already thought about that and come up with a solution? I’m all ears
It’s like reporting little thefts to the police where you know they will not doing anything about it.
I see a lot of little bugs which probably a lot of people see. Never get reported. Everyone finds a work-around and don’t get bothered about it because it’s not worth the hassle of reporting it… Nobody is accountable anyway. It’s Open Source
Ever read your terms of service? If you have a iPhone, Alexa or Ring device, or use Firefox or Chrome, you probably did give your permission for this assumed conspiracy nonsense.
I think you have a right to complain if you feel like it
And I would call the entire dashboard (incl. the cards) one big, severe bug
Maybe manpower should go into things like that.
And my personal observation: new features are released that have all kinds of bugs or missing “basic” functionality/features and then those are not fixed or it takes ages so people (or at least I) lose interest and abandon that shiny new (but pretty useless) feature.
Energy dashboard being one of the flashy new things that are… well… not very good due to the way it is implemented and the small bugs it still has.
Firefox should really not be in that list.
The consents you give to use Firefox is much less and nowhere near the tracking done by mobile phones or ad-driven browsers.
I think it would be a great addition to HA in general if you could assign devices to people - not only device_trackers, but also all other devices in the same way as you can assign them to an specific area…
No, when I bought my iPad those trackers didn’t exist. Those added terms are forced on me because my bought hardware is rendered utterly useless until I accept it and there is no way out. It was my point exactly that there’s no escape but revert to the digital stoneage. Apple is just more hypocritical about it.
