Is it necessary to have the content proxied through your API server? If the company doesn't work out, it would be a shame to have this device stop working, even though it's fully capable of reaching the internal URL I would be hosting my content on.
Yes, that could be changed, but it's a bit of a hassle to implement with the bluetooth and and storing state on an embedded device. So I haven't implemented it yet and I can't promise if and when it's coming.
If I am not hit by a metaphoric truck, I give my best to make sure not to disappoint my customers. My plan is to keep the backend running for 10 years after I sell the last display.
But of course that's just a plan und I understand that you may not like the odds. That's totally reasonable.
No offense to you, but this comment sounds like one of those customers that are never going to use your thing, even if you implement all that.
Find the people who want to use a product despite all the lacking features, because the one thing it does is exactly what they need, and build on that.
They have customizable dashboards. What I'm really looking for is something I can use as a sort of command center in a few places in my house. It allows me to control the things I care about, shows my upcoming schedule, and whatever else is relevant. By my work desk, the #1 thing I care about is upcoming meetings. Other places, I care about other things.
Critically, though, I don't want it to glow. I want it either e-ink or purely reflective LCD.
This is probably a few devices. Ideally, there'd be small / cheap ones with mechanical switches and a tiny screen to replace my light switches, and big ones for my living room, work room, etc.
I think the trick would be to start with something, so one doesn't have an infinite engineering task, and so there are ready frameworks for modularity. Either Home Assistant or dash would be decent starting points, depending on which direction this took.
But yeah, now that I mention it, I have a vision in my head for what I want (which I would buy). Retrokludged features without the same vision would probably not get us there.
So I'll cut this back to what everyone else is saying: Calendar flexibility. This needs to sync with home, work, and other calendars, so at the very least, Outlook.
Yes, but, as you say, reimplementing Home Assistant wouldn't really be that valuable. What you might want is a (very dumb) e-ink display with a touch panel, which can maybe display an image and register touches. Then, that can interface with Home Assistant to make touch panels. That would be much more manageable as a project, and very useful. Bonus points for physical buttons.
My experience is that 90% of the work on these projects is integration: getting the toolchain working, and all the pieces talking. If there were an easy-to-use digitizer and a documented toolchain, I'd buy a LilyGo T5 right now.
Without that, I suspect I'll never get around to making one work, and I'm not buying another half-started project.
As a footnote, 50% of the remaining work is on the stupid stuff like cases and power supplies. I'm glad you did that. But if you ever sell this as a device, consider provide a variety of cases for different use-cases (desk stand, wall mount, metal box with room for other stuff, etc.) and your sales will double.
As a footnote, I like when devices are compatible with something standard, like a Micro:bit or a Circuit Playground. That guarantees a basic working toolchain for at least getting code on, as well as a support ecosystem.
Home assistant is still a dream to come true in terms of where I want to go with the product. And home-assistant.io may be a good starting point.
I'll save that link ...
Regarding calendar: You can already connect as many google calendars and as many .ics sources as you want. If you can create an .ics link from outlook that's supported. (At least on iOS, the Android update is still being worked on. Did you know about Jetpack Compose? I didn't...)
Of course, Outlook support would be nice. It's on the roadmap. But as always, I don't want to promise things that haven't happened yet.
There isn't a single thing I made more than 15 years ago in my house which:
(a) required an operating system
(b) still works
It all bitrots to hell. For comparison, everything I've made on a PIC, Atmel, ESP, etc. continues working.
FPGAs have a more mixed track record; the devices work, but they're unmaintainable (the dev toolchains require e.g. Windows NT and some activation server which no longer exists).
Things I've made out of wood, metal, and similar materials work too, as do PCBs I designed with analog.
My point is I want a piece of hardware and not a computer.
