Live data from Hacker News

I built Timeframe, our family e-paper dashboard

hawksley.org

361–370 of 385 posts

Re: I built Timeframe, our family e-paper dashboard

#361
post #330

Earlier quoted context omitted.

I love how every chain in this thread starts off defending the practical utility of all this but ends with defending "learning" and "hobbies" when someone points out that you don't need to spend $2000 and 100 hours to know when a washing machine cycle ends.

Its disingenuous to claim that OP spent that much time and money to know when the washing cycle was complete. That's one of several different things the screen can do in addition to everything else it can display.

It's amazing how enticing dominos of chiming in with snark can feel.

It's also possible that the sentence struck a nerve because it's a pretty simple lens and test.

All the scrolling is free labour for tech/social media companies. Other folks seem to use it more as a platform to create, publish or be more mindful of their interactions compared to passive consumption and reaction.

Family schedules can be a unique and valuable problem to solve, namely how much more valuable time becomes, as well as how much a little bit of optimizing can give back.

Re: I built Timeframe, our family e-paper dashboard

#362
The insight that a blank status area means "the house is healthy" is the best part of this whole project imo. Most smart home dashboards try to show you everything all the time and you just end up tuning it all out. This is basically the opposite approach and it makes way more sense for something you glance at 50 times a day.

I tried something similar with a Kindle a few years back for just weather + calendar and ran into the same jailbreak maintenance hell. Ended up giving up. The Visionect displays look great but $1000+ per screen is brutal. Curious if the author has looked at the Waveshare e-paper panels driven by an ESP32, they're like $40-80 for a 7.5" screen and you can do partial refreshes. Obviously way smaller than the Boox but might work as a cheaper bedroom/mudroom option for people who want to build something like this without spending $3k.

Re: I built Timeframe, our family e-paper dashboard

#363

Earlier quoted context omitted.

Not anymore - that was months ago when he did all that labor, it is no a sunk-cost. We can debate if it was worth the investment though.

Literally every person I know who is obsessed with home automation has spend thousands of bucks and hundreds if not more hours on automating things that take a second or cost two bucks for a paper version that doesn't use electricity The article is about a 2000$ eink display that shows a calendar and the weather. Your phone does that for free and you don't need to walk to the hallway every time. This is basically ant…

The sunk cost and dollar store arguments assume the goal is information retrieval at minimum cost.

The relationship with technology you're designing around your family is worth considering too.

Looking at a shared hallway screen that shows a shared calendar which doesn't exist to pull anyone into a feed doesn't make a worse phone, it's solving a different level of shared understanding entirely.

The assumption that the phone is a neutral free tool tends to come most naturally to people who haven't yet thought about who designed the defaults and why. Free at point of use isn't the same as free. Someone optimized very hard for your continued presence.

Re: I built Timeframe, our family e-paper dashboard

#364
post #362

The insight that a blank status area means "the house is healthy" is the best part of this whole project imo. Most smart home dashboards try to show you everything all the time and you just end up tuning it all out. This is basically the opposite approach and it makes way more sense for something you glance at 50 times a day. I tried something similar with a Kindle a few years back for just weather + calendar and ran…

Same experience, the job of the display is to help manage attention, instead of information delivery.

If we think about paper calendars hung on a wall, and updated sporatically we know what's there is likely good information, only that it might not be up to date.

If a calendar can be calm by default, surface what's changed or newly relevant, and fade when it resolves. The next level could be understanding who's attention something needs, and when, in a personalized way.

Re: I built Timeframe, our family e-paper dashboard

#366
This is such an impressive tool. As others have noted, the price of entry is high, but it's incredible what it pulls off.

I've been doing a cheaper version using a Waveshare 7.5 inch screen, a Raspberry Pi and a 3d printed case coupled with Inkycal: https://github.com/aceinnolab/Inkycal. This works well for my needs, but seeing what else can be done maybe it's time for an upgrade.

Re: I built Timeframe, our family e-paper dashboard

#367
post #170

Earlier quoted context omitted.

I did a similar thing with a regular backlit computer screen. It automatically shuts off after 30 seconds of inactivity. I added a $3 webcam, and use openCV to detect motion. If three consecutive frames (sampled 0.5s apart) are each sufficiently difficult from the previous one, it attaches a virtual USB mouse, then moves it one pixel. This wakes up the display whenever you walk past, then puts it back to sleep again…

You can probably just use a cheap motion sensor instead of the webcam if you wanted to. There so many now

Found a few aliexpress sellers offering LD2410C's, but all cost 30% more than the webcam I used.

Re: I built Timeframe, our family e-paper dashboard

#368
post #310

Earlier quoted context omitted.

I would like to suggest a third alternative which is to set an alarm for the maximum time it will take. The HA option is more fun though for sure.

But why? Why add a manual, less accurate step instead of just using HA and solving it once? Steaming takes 20 minutes, a quick wash is ~1h, a boil wash more like 4h. Why do I want only my phone to have a notification? Why do I want it to override other settings and go off at a set time rather than when I choose to interact (as a notification would)? You can absolutely solve this in other ways, but adding an automatio…

I'm not saying don't use HA - you do you. I get the alure of home automation and I'm glad you're finding it useful. All I'm saying is that the argument that 'the phone timer might be wrong therefore use HA' is a false dichotomy. You know what kind of wash you've selected and you know how long it takes roughly - you wrote it in your comment :-) It takes a few seconds to set a timer for that plus 10% for variability.

> ... adding an automation into HA ... took a few minutes max

... plus the hours it takes to setup and maintain HA and the machine its running on and the wireless hub and the dishwasher... etc. I bet you won't save back the total time spent across the whole lifetime of the system. That's not the only measure of success and that's fine.

Re: I built Timeframe, our family e-paper dashboard

#369

Earlier quoted context omitted.

Even with no beeps you put washing on > you go get it when its done. It doesnt matter if it sits in the washing machine an extra 10m or an hour.

For me it's not the washing machine, it's the dryer. The time remaining reported by the dryer when you start the cycle has almost no relation to how long it will actually take. Sometimes I go down to the basement after an hour (the dryer says 45m when you start it), and it still says 30m remaining. It's not the end of the world of course, but it is annoying, and it's the sort of annoyance technology can solve pretty…

I got a 4-pack of zigbee power plugs that report usage, and I have a home assistant automation that goes ding (or whatever) when the washer or dryer had been using electricity for at least a few minutes and then stops using electricity.

Re: I built Timeframe, our family e-paper dashboard

#370

Earlier quoted context omitted.

I incorporated OpenMeteo into a project recently and got frustrated with their aggressive rate limiting. If in the US, weather dot gov has an excellent, free API. Or, OpenWeatherAPI which works internationally and has support for more things that weather dot gov does not. OpenWeatherAPI will also synthetically provide weather data based on their models if there is missing station data

> I incorporated OpenMeteo into a project recently and got frustrated with their aggressive rate limiting Which one? They seem to do 600 calls / min, 5.000 calls / hour, 10.000 calls / day, 300.000 calls / month, how many times do you need to look up the weather for personal use? Fine, maybe you want 3 different locations, you can still call each of those sufficiently with those rate limits, no?

I have to sample multiple lat/lon across the world all at once
Post reply on HN