For years now I have wanted a “dashboard” somewhere central in my house. I think the inspiration comes from my childhood when my Mom would keep a paper calendar with family plans. At some point during my childhood, probably when I was around 8 or 9 years old I started checking the calendar at least a couple times a day. I was the oldest of 6 and I was becoming more and more aware that if I could answer my own questions about what was going on and not interrupt my Mom, everything worked a little bit better.

My own kids are getting to that age range, and I thought they might appreciate something similar. Of course, we live in a digital age with all the advantages and pitfalls that brings, so there are some extra decisions to make. My own kids are getting to that age range, and I thought they might appreciate something similar. Of course, we live in a digital age with all the advantages and pitfalls that brings, so there are some extra decisions to make.

Options

Several ideas for what to use for a home dashboard have been kicking around in my head for quite a while. Here’s a summary of the ones I’ve given the most thought to.

An iPad in a Picture Frame

The most obvious solution to this problem is just putting an iPad in a picture frame. And if I had had an old iPad lying around I might have just done it.

Pros

• There already is lots of software
• I already own a bunch of Apple devices and I’ve gotten pretty good at managing them
• We have the option for interactive applications, like a “morning checklist” for the kids, or videos.

Cons

• iPads are expensive
• iPads need to be charged regularly or plugged in all the time
• I have a strong preference for my kids not interacting with more touchscreens (and me for that matter)

Skylight

Once you start searching around for a digital home calendar, you pretty quickly find Skylight. I know some people who make pretty good use of these, and it seems like they work pretty well. After a bit of research, these were the pros and cons.

Pros

• There are days where I really would like fewer devices to manage
• The Skylight hardware looks really nice (so does their software)

Cons

• Skylight is expensive
• Skylight needs to be plugged in
• Touch screens
• As far as I can tell you’re not just buying the hardware you also will need to buy a subscription (but at least they’re upfront about that)
• While I’m sure the software Skylight provides is quite good, it seems like it’s a pretty closed ecosystem

Raspberry Pi Powered Digital Sign

The hacker/builder in me really likes this idea. I’ve set up a lot of useful projects with Raspberry Pis and I’m familiar with the hardware, and I saw recently there are several options and it really seems like a thriving ecosystem of hardware and software.

Pros

• Lots of options for hardware and software
• I could do just about anything I want with it

Cons

• Plugged in
• Sounds like a lot of work

Building my own E-Ink Display

I can’t remember where I first came across it, but I think I got this idea from the Inky Frame project. Even though I didn’t go this route I’d still like to buy the kit because it looks like a lot of fun.

Pros

• It looked like a fun kit to put together
• Battery powered (not sure how often I’d have to charge it)

Cons

• Sounds like a lot of work
  • Not sure how much of my own software I’d have to roll
  • Making it look nice would be extra work

TRMNL

While waffling back and forth between all these options, a friend of mine told me about TRMNL. When I found the project they had already shipped all their Kickstarter units and were starting to sell to anyone.

I was drawn to this for a few reasons.

• It’s a nice looking device (not as nice as the Skylight, but still nicer than if I tried to build something myself)
• They seem like they are building an open source ecosystem where you can either build your own hardware or software or both.
• The way TRMNLs work is very simple and elegant
• No touch screen.
• Battery powered.
• The devices can go a long time between charges
• The batteries are replaceable
• No touch screen

The downsides are kinda what you’d expect with an e-ink device:

• Slow refresh rate (it’s not going to show you anything in real time)
• No video
• No color

Hardware

The TRMNL itself

There’s not a lot to say about the TRMNL itself. It’s pretty simple. The kickstand was really nice when I was doing my initial testing.

The battery has a surprisingly long life.

I’m not used to working with e-ink screens so it took me a little while to get used to whatever was on the screen staying there when I turned it off.

Setting up TRMNLs is pretty straightforward, you just have to do the same 2-ish minute song and dance to hook up most WiFi IoT devices. The main downside to this (as I see it) is that if you ever need to change your WiFi network name or password, you’ve got some work to do. I’d recommend having a dedicated WiFi SSID for your IoT devices so it will be easier to make other changes to your network and your IoT devices will keep working.

3D Printing a Magnetic Fridge Mount

Shortly after I got my TRMNL the founder of the company, Ryan, sent me a pretty standard email asking me for feedback. He said in the email that I could just respond to the email if I wanted to instead of doing the online survey. I looked up this Ryan guy because the friend who had told me about TRMNL had implied he was an interesting guy.

I told him in the email that I was exploring wall mounting options. Sure enough, he got back to me and said his wife had recently finished some designs for some 3D prints and it turned out one of the designs was for a fridge mount. Mounting the dashboard on the fridge was perfect, and an idea I had effectively dismissed because I had spent so much time thinking about options that would need to be plugged in all the time.

Another very convenient fact was that I recently got a 3D printer at home and just happened to have some white filament that matched the color of the TRMNL cases I had.

Software

Using TRMNL’s Software

You can figure your TRMNL using their cloud infrastructure and setting up all my TRMNLs was a smooth process.

One thing to keep in mind, if you want to make your own TRMNLs there’s an additional one time “developer” fee of $20 (at least there was when I was setting mine up) per device. But I don’t think I had to pay for all 4 of mine, because on 2 of them, I’ve just stuck with plugins TRMNL provides.

I do get a little confused sometimes when I’m trying to find a screen that has some setting for a plugin I want to tweak and it seems like it’s had some downtime but none of this is exactly critical so it hasn’t been a big deal.

The process of making your own TRMNL plugin (app) is quite good. They have a great design system and you can get your data into the terminal system either by polling an API or via webhooks.

One cool thing to keep in mind, if you’re looking for something “quick and dirty” they have a plugin that just takes a screenshot. I didn’t actually try it, but if you already have a dashboard-y thing I’d imagine just providing a URL to a website or image would be a pretty fast way to something up on the TRMNL, see if it’s useful. Then you can carefully craft your API and design the plugin to make it look better.

Weather Station

I made a plugin for displaying weather data from a weather station I installed in my backyard. The TRMNL plugin was the easy part (although I do need to take another pass on the design, see if I can get it looking a little bit nicer).

The hard part (which was still not very hard) was setting up the backend. I bought the weather station because it has the ability to send its weather data wherever I wanted it to go (not just some “proprietary cloud”, although that is also an option). When I actually got into it though it was a little bit limited. I basically could point the weather station (actually the base station) at a URL I had 2 choices of how it would send the data. I could also set the interval at which the data would be sent. This is a perfect job for an LLM, just a simple webserver with a database backend. You could whip up your own, or you could take a look at mine. Even if you do make your own looking at mine might make things easier.

I had assumed my weather station would keep at least some historical data on it, after all it did have enough computing power to run a simple web interface. That turned out not to be true, but I think there are some weather stations made by the same company that will log weather data to an SD card.

Growing Degree Days

The reason I have a weather station in my backyard is because I’m working on another project having to do with tracking Growing Degree days and I wanted to see how different my backyard weather station’s data is from the nearest official weather station at MSP (Minneapolis-St. Paul International Airport).

There’s still more work to be done on this project, but the nice thing was, I already had a very easy to use API for getting growing degree day data. The one issue was, it was not packaged up nicely for a TRMNL plugin to just use (as is). I do some work in the browser on Degree Day Dashboard and the one thing that’s very limited when building TRMNL plugins is the extra logic you can do. I personally think that’s great. It should stay simple.

So I had to add an extra piece to my process to generate the JSON with just the data I wanted for the Degree Day TRMNL plugin. So far it’s been working pretty well. I do need to make it look better though, and I am hoping soon to submit it as one of their “official” plugin “recipes”.

My Conclusions so Far

I’ve had my current iteration of my home dashboard running for a couple of months and it has been great. The whole family has been using it to find information we need. There have already been a couple cases where I have failed to update the family calendar and I’ve been called out on it by my kids.

My plugins still need refinement and the TRMNL plugin ecosystem seems to be growing rapidly. Hopefully I’ll be able to carve out some time to work on my own plugins and get caught up on what my new options are. One thing I’d like to explore is showing different things at different times of the day.

It also looks like their self-hosted server software for managing what’s on the plugins is also getting more mature. I realize the value is marginal, but I really like the idea of my Internet being down but still being able to see the local weather on my TRMNL thanks to my backyard weather station (and the pieces to connect the TRMNL and weather station) all not requiring the Internet.

Another motivation for self-hosting is that there have been some hiccups with the TRMNL’s cloud service. More than once some of my TRMNLs have gone into an error state because they weren’t able to connect to the cloud service. It’s always sorted itself out though.

This project has been a lot of fun so far, and who knows, it might even save me some time, in the long run.