Last week I was at the Home Assistant community meetup in Cologne. Packed room, good food and drinks, and the kind of conversations that only happen when you put a hundred smart home nerds in one place.

At some point I ran into T-Flow. He was wearing a DOOM t-shirt. (You can actually spot it in his video about the meetup.)

That's where this started.

We got talking about the classic internet tradition: if something has a screen and a processor, someone will eventually make DOOM run on it. Calculators. ATMs. Thermostats. Even a LEGO brick. There's an entire subreddit dedicated to it.

And then one of us said:

"But… can Home Assistant run DOOM?"

We laughed. Then we paused.

Because technically… there was no real reason it couldn't.

I made a note on my phone, and when I got home, I opened VS Code and typed the following to GitHub Copilot:

I have a wild idea :D

I want to build Doom (the game) as a card for a Home Assistant dashboard. I want the original doom game to be playable inside Home Assistant that way.

This needs to be distributed as a custom component (so don't build it into Home Assistant directly), instead, let's build a repo inside the doom folder.

The card needs to be configurable from the UI (no fiddling with YAML) and will be distributed through HACS. We can use https://github.com/piitaya/lovelace-mushroom as a repository that is a good example on how to set it up.

I want it to run the original doom game, not a recreation.

That was it.

No detailed architecture spec. No scaffolding. Just intent.

Two hours later

Less than two hours later, I had a working Home Assistant custom integration.

You install it via HACS. You add the integration. You drop the card onto your dashboard. You click "Play"...

And you're running the original DOOM.EXE inside Home Assistant.

Not a video. Not a stream. Not a recreation.

The real shareware episode of DOOM (1993), running locally in your browser through js-dos, which compiles DOSBox to WebAssembly.

It's absurd.

And it works.

If it runs in Home Assistant, it should behave like Home Assistant

Embedding DOOM in a dashboard card would already be funny.

But that's not how Home Assistant works.

Home Assistant revolves around devices, entities and their state. Lights, switches, sensors: everything participates in the same model. Automations react to changes in state.

So if DOOM runs inside Home Assistant, it shouldn't just be a gimmick.

It should behave like a proper integration.

The custom component exposes a full set of entities under a single DOOM device.

There's a binary sensor that tracks whether someone is actively playing. The game sends a heartbeat signal while the game runs. If the heartbeat stops for more than three seconds (say, when you close the page), the sensor flips to off.

The states?

On → Ripping and tearing 🔥 and Off → Dormant 💤

Yes, I enjoyed adding that. 😂

It also tracks who is playing, how long the current session has been running, total accumulated play time across all sessions, how many sessions have been played, and when the last session ended. All persisted across restarts.

And then there's an enum sensor that rotates once per day, picking from a pool of 46 DOOM facts, development trivia, lore snippets, and classic quotes.

Oh, and there's an easter egg. Type iddqd anywhere in the Home Assistant interface (just start typing, no input field needed) and a DOOM dialog pops up, letting you play immediately. No dashboard card required. God mode activated.

Is any of this useful?

Not really. 🙃

But once DOOM becomes part of the state machine, it becomes automatable.

You can:

• Turn your office lights hellish red when DOOM starts.
• Restore the previous lighting scene when it stops.
• Display the daily DOOM fact on a wall tablet.
• Announce who just started a DOOM session.
• Track your household's total demon-slaying hours.
• Trigger a playful notification.

Completely unnecessary.

Technically clean.

Very Home Assistant.

I didn't write it

Here's the part that made me stop and think.

I didn't write the backend. I didn't write the frontend card. I didn't write the packaging. I didn't write the translations (64! languages). I didn't write the README.

AI did. ✨

I used Visual Studio Code with GitHub Copilot (Claude Opus 4.6 under the hood). As part of the GitHub Stars program, I get access to Copilot, which made this experiment possible, and I'm genuinely grateful for that. 🙏

From that initial "wild idea" prompt, it scaffolded:

• The integration structure
• The config flow
• The UI-configurable card
• The HACS-ready packaging
• The heartbeat mechanism
• The player tracking and session sensors
• The quote sensor
• The iddqd easter egg
• The documentation

I iterated. I reviewed everything carefully. I tested it. I fixed edge cases.

But I didn't sit down and write it line by line. As a matter of fact, I didn't write a single line at all.

That's just impressive.

Why this worked

Here's the thing that surprised me most: this wasn't luck.

I've spent years helping shape Home Assistant's architecture. Strong conventions, a predictable integration structure, well-defined patterns, consistent extension points. That structure exists to help human contributors be productive.

Turns out, it helps AI too.

If Home Assistant were inconsistent or loosely defined, this would have been chaos. Instead, the architecture made it possible to describe intent and have working software emerge.

Good architecture doesn't just help humans anymore. It now amplifies machines as well. As someone who cares deeply about open-source structure and long-term maintainability, that's a fascinating thing to witness.

What this is, and what it isn't

Let's keep this grounded.

This is a silly project.

It started with a t-shirt at a meetup. It is not a roadmap item. It's not a strategic direction. But it does show something important.

When you build privacy-first, open, flexible systems, you enable experimentation. You lower the barrier to trying ideas that would otherwise feel like too much work.

And experimentation leads to learning.

Sometimes serious learning.

And sometimes… demon-slaying learning.

So… can Home Assistant run DOOM?

Yes! 🙌

It can...

I've made everything available on GitHub, so you can actually install DOOM yourself. Or just go to HACS and search for the DOOM integration directly in there.

More importantly, it shows what becomes possible when strong open-source architecture meets modern AI tooling.

And sometimes, it lets you automate your lights while fighting demons on Mars.

../Frenck

We got the keys 48 hours ago. The house is full of boxes, our stuff is everywhere, and I have absolutely no idea how half the smart home equipment already installed actually works.

Perfect. 😓

This is the first video in my new series where I'm documenting the entire journey of building out a Home Assistant smart home from scratch. No polished "here's my finished setup" content. Just the real process: the decisions, the discoveries, the mystery devices, and every spectacular failure along the way.

What am I working with?

The previous owner left behind some interesting gear. There's a full Philips Hue outdoor lighting setup in the garden (with a single remote controlling everything), 40 solar panels on the roof with a SolarEdge inverter I need to claim ownership of, and a jacuzzi with a Gecko control system that I'm determined to monitor through Home Assistant.

But here's where it gets fun: I found two thermostats in the living room controlling... something. The floor heating has a wireless Watts remote that I have no idea which room it's assigned to. There's a heat pump, multiple air conditioning units with missing remotes, and a mysterious "Hunter HC" box in the electrical closet and one that has a Bluetooth logo. 🤷‍♂️

Oh, and the networking situation? The electrician is coming Friday because the fusebox area is, in my professional technical opinion, a huge mess. 🙈

The real journey starts now

I'm tackling this the way anyone would: by walking around 😄 But... with a camera, pointing at things, and saying "I need to figure out what this does". Watch the video above to see the full tour, including all the mystery devices, the chaos of moving boxes, and my initial thoughts on what to automate first.

What would you tackle first in this setup? Drop your thoughts in the YouTube comments. I'm genuinely curious what priorities you'd set with this much smart home potential and this many unknowns.

CES is coming up in January, so I'll be bringing back inspiration and ideas. But for now, it's time to unpack, set up my (Wi-Fi) network, and start figuring out how all these systems actually work.

../Frenck

I can't sleep.

It's not stress, exactly. It's not anxiety. It's just... my brain won't shut up about the new place.

We're moving mid-December. The house is bought, the old place is sold (and yes, I took every single smart device with me), and now I'm sitting here with all this smart home gear packed in boxes and an new empty house waiting for me that I can't even touch yet.

And my brain is going absolutely wild with possibilities.

The exciting part

This is a fresh start. A blank canvas. Well, not completely blank, there's already some equipment installed that (I'll talk about in another post). But in terms of my smart home setup? I get to rethink everything.

At the old place, things kind of evolved organically over time. I added a sensor here, automated a light there, gradually built up the system as I learned what worked and what didn't. It was great, but it was also... chaotic. There was no grand plan. No intentional design. Just iteration after iteration after iteration.

Now I have the chance to do it right from the start. To think about the whole system before I install a single device. To plan the network infrastructure, to choose the right hardware for each job, to build something cohesive instead of accumulated.

And that's incredibly exciting 🤗

It's also completely overwhelming 😧

Where do I even start?

Seriously, where do you start when you're building a smart home from scratch if you are into smart homes already?

Do you start with energy monitoring? Get solar production dialed in, understand power consumption, build automations around when to run high-draw appliances?

Or do you start with climate control? It's going to be December. It's winter here. Making sure heating is automated properly seems pretty important when you're moving in the middle of cold season.

Maybe lights are the right first step? They're usually the easiest wins. Quick to install, immediate value, you feel productive. But are they actually the foundation, or are they just the fun part that makes you feel like you're making progress?

And what about security? Locks, cameras, door sensors, window sensors, alarm system. That feels important. That feels like something you'd want sorted early. But which systems? Which brands? Which approaches?

I'm genuinely stuck on where to prioritize.

The hardware spiral

Then there's the hardware question. And this is where I really start losing sleep.

Smart locks: I've never had them before. My old place just had regular keys. But now I'm thinking... wouldn't it be great to have codes for family, temporary codes for guests, automations that lock up when we leave? But which ones? There are so many options. Z-Wave, Zigbee, WiFi, Matter, proprietary protocols (kidding, I don't want anything proprietary of course). Some have keypads, some have fingerprint readers, some integrate with existing deadbolts, some replace everything. And since we're all on Apple in our household, Apple HomeKey support would be amazing. But does that limit my options too much? I have no idea which direction to go.

Network infrastructure: Everyone seems to love talking about PoE (yes... Trevor, P. O. E.). Power over Ethernet for access points, for cameras, for sensors. It sounds great in theory. But does it actually make sense for a residential setup? What cabling is already there? Do I need to run more cables? Do I just stick with Wi-Fi for most things and call it good enough? And what about the core network, managed switches, VLANs, proper segmentation? How deep do I go?

Wall tablets: I see people with these beautiful wall-mounted tablets showing dashboards and controls. They look amazing. But are they actually useful or are they just pretty? Which tablets? How do you power them? How do you prevent them from being dated in two years?

Door and window sensors: Every door? Every window? Just the ground floor? Just the exterior doors? Zigbee, Z-Wave, or Matter? How many do I actually need before I'm just collecting data for the sake of collecting data?

Alarm system: There isn't one installed currently, so I get to pick from scratch. Do I go with a traditional alarm panel that integrates with Home Assistant? Do I build something entirely with Home Assistant and sensors? What about professional monitoring? Are there alarm panels that work well with open systems, or am I looking at proprietary nightmares? What actually makes sense for real security versus just having fun with automation?

Cameras: Okay, I already know I like Reolink. I've used their stuff before and it's solid. But they have SO MANY models. Do I want PoE cameras? Again, where are the cables heading? Battery powered? Wired but not PoE? PTZ or fixed? Indoor, outdoor, doorbell? 2K, 4K, 5MP, 8MP? I already have their NVR. And how many cameras is the right number? Because I could easily see myself going overboard here.

I'm not even scratching the surface and I'm already dizzy.

The questions I can't answer yet

The worst part is all the questions I can't answer until I'm actually in the house.

Where are the existing electrical outlets? Where would wall tablets actually make sense? Where are the blind spots for cameras? Where does the network equipment even go? Is there a proper place for a tiny server rack or am I improvising with a closet?

What's the Wi-Fi coverage like? Do I need more access points? Can I hardwire them or am I running everything wireless in a mesh?

Where are the light switches? Are they in places that make sense for smart switches, or am I better off with smart bulbs? Or a mix of both?

I don't know. I can't know. Not until I have those keys and can actually walk through the space and start making decisions based on reality instead of speculation.

But that doesn't stop my brain from spinning through every possible scenario at 11pm.

What I actually need

Here's what I'm hoping for from the community: recommendations.

Not just "buy this brand" recommendations, but "here's how I approached this problem" recommendations.

Where did you start when you built or rebuilt your smart home? What was the first thing you automated and why?

What hardware has genuinely made a difference for you? Not the stuff that's cool in theory, but the stuff you use every day and couldn't imagine living without now.

What did you overbuy that you regret? What seemed like a good idea but turned out to be more trouble than it was worth?

What infrastructure decisions set you up for success? Did you run ethernet? Did you plan for PoE? Did you go heavy on Wi-Fi? What would you do differently?

Smart locks, anyone have strong opinions? What actually works well and integrates cleanly with Home Assistant? Bonus points if it supports Apple HomeKey. Any manufacturer/model is MUST try?

Camera recommendations for someone who likes Reolink but is overwhelmed by options? What models have you actually deployed and been happy with?

Wall tablets, worth it or just decorative? If you have them, which tablets and how did you handle power and mounting?

Alarm systems, what are my options when starting from scratch? Traditional panel that integrates? Pure Home Assistant? Something else entirely?

I'm genuinely asking. I want to know what actually works in practice, not just what sounds good on paper.

Reality check

I know I need to calm down. I know that I'm not going to have everything perfect on day one. I know that the smart home will evolve over time just like it did at the old place, and that's fine.

But having the chance to think about it intentionally before I start installing things? That feels valuable. Even if it's keeping me up at night.

Mid-December. That's when I get the keys. That's when this all becomes real and I have to actually make decisions instead of just spiraling through possibilities.

Until then, I'm going to keep losing sleep and asking questions.

Thanks for letting me spiral a bit here. I'm sure once I'm actually in the house and can make concrete progress, I'll sleep better.

Until then, happy automating, and Choo Choo! 🚂

../Frenck

📺 Rather watch a video version of this blog? Click here to view the video on my YouTube channel.

I opened an architecture proposal to rename "add-ons" to "apps" in Home Assistant. Before you think this is just bike-shedding about terminology, let me explain why this seemingly small change actually matters.

The confusion problem

Here's something I keep noticing. New users start using Home Assistant and when they want to add something for the first time, they often end up trying to add add-ons because it sounds like something that adds to Home Assistant. And while that's true, it's often not what they're looking for. They're actually looking for integrations.

This signals something important: the name doesn't really tell you what it is and what you need.

When you're new to Home Assistant, you open the interface for the first time with a light bulb you want to add. You see add-ons and integrations in the settings. Add-on sounds pretty much like the thing you want, right? You want to add something. Sounds logical. The names really don't tell you what they actually do.

Add-ons are software applications that run alongside Home Assistant. They're not Home Assistant itself. They run next to it. You can start them up and use them, and they can provide additional services to Home Assistant. Good examples include AdGuard Home, ESPHome, and MQTT Mosquitto.

Integrations integrate a device or service into Home Assistant. If you have a light bulb, you integrate that with Home Assistant and you get capabilities to control and monitor it through automations, dashboards, and so on.

So that's the difference. One is integrating something into Home Assistant, while add-ons run software alongside Home Assistant.

The AdGuard example makes it even more confusing

The two can connect to each other, and this makes it even more confusing. Take AdGuard Home, for example. You can find AdGuard Home in both the add-on store and in the integrations list.

That's because Home Assistant can run AdGuard alongside itself as the actual ad filtering system for your house. But Home Assistant can also integrate with that add-on you're running to get statistics from it, automate with it, and control it.

So there are two parts to AdGuard Home: the integration and the add-on. It's super confusing for new people.

The solution: apps

Let's just rename add-ons to apps. Because that feels way more natural. They're software applications running alongside Home Assistant. It's just more clear in terms of a mental model.

Apps run. You install apps onto something and you run them. Put yourself in the shoes of a user that just starts. I think it makes more sense that you integrate something and you can run apps too.

Think about your phone. You install apps on it. You know exactly what that means. An app is something that you run, you start it up, you do something with it. Your phone has a runtime capability of starting up those apps. Add-ons are kind of the same. They're different pieces of software that run on them.

So AdGuard has an app, ESPHome has an app, and they just run alongside Home Assistant.

Does this name change solve all the problems we can think off? No. A big step forward? Yes!

No functional changes

The proposal is very clear: it doesn't change a bit. It only renames it. Nothing is breaking. We're literally just changing the label in the UI. Call it apps everywhere, and Bob's your uncle.

The alternatives we considered

In the GitHub discussion, we've been discussing alternatives, and I think it's fair to list them out here.

Widgets came up, but I don't think that makes sense. A widget feels more like a UI element, not something that would be a software application running alongside Home Assistant. I actually think widgets would be something we should consider using in the future for dashboards. Dashboards full of widgets makes sense. But it's out of scope for this one.

Extensions also came up, a little bit like Google Chrome has. But I would argue that browser extensions are more like add-ons in the traditional sense. They extend functionality. They're more like our integrations, basically. Not suitable.

Containers is technically accurate. Under the hood, add-ons are Docker containers managed by Home Assistant with specific APIs. There's a little bit of layering on top of them and between them. In their essential form, they are containers.

However, if I think about my dad starting to use Home Assistant, telling him "hey, you can install more containers on this thing," I don't think that will fly. Containers is too technical for most users.

Why this matters: the Discord story

This is about reducing frustration for newcomers. When someone installs Home Assistant and immediately gets confused and frustrated about terminology, that's not good.

Here's how bad this confusion has gotten: a few months ago on Discord, we actually gave up on having separate channels for add-ons and integrations. We just merged them and slapped them together because people didn't really understand the difference. That's really problematic. That is something we need to fix.

And that's what this proposal is about.

Your input matters

This is still an open discussion. The architecture proposal is discussion #1287 on GitHub, and I'd love to hear your thoughts.

Do you think "apps" works? Should we keep "add-ons"? Is there a better alternative? Jump into the conversation and let's figure out the best path forward together.

../Frenck

I also recorded a vlog about this proposal on my way to GitHub Universe 2025. You can watch it here if you prefer video format. Fair warning: it's my first vlog, so there's a bit of awkwardness as I share my thoughts this way!

📺 Rather watch a video version of this blog? Click here to view the video on my YouTube channel.

I need to be honest with you.

A lot of you keep asking when I'm going back to streaming. I see the comments. I see the Discord messages. "When's the next stream, Frenck?", "Miss the live coding sessions!", "Time to stream homie!"

And look, I get it. I miss it too. 😢

But it's not coming back. At least not anytime soon. Could it return someday in some form? Maybe. But not in the way it was, and definitely not right now.

And I think that's actually okay. Let me explain.

What happened to streaming

I used to stream regularly. Live coding and building Home Assistant integrations and add-ons, debugging weird issues, tinkering with my own smart home, just hanging out with the community, and talking through problems in real-time.

I loved it. 🥰

The direct connection. The spontaneity. The ability to show the messy, real process of building software; not the polished, edited version. Being able to answer your questions immediately and riff on ideas together.

But here's the reality I had to face: I'm the Lead of Home Assistant. I'm helping build the Open Home Foundation. I have a family, 2 kids! And streaming consistently requires 6-9 hours a week of scheduled, focused time when I have to be "on". Like really hyped and ON.

Honestly, I was heading straight for burnout. 🥵

Something had to give. And for a while, I just... stopped. Figured I'd come back to it when things calmed down. Except things don't calm down when you're building the world's biggest open source smart home platform. They get busier.

So I had a choice: Keep feeling guilty about not streaming, or admit it wasn't working and find a better way.

The thing I actually loved

Here's what I realized: What I actually loved about streaming wasn't the "being live for 3-4 hours" part.

It was showing you the real stuff. Taking you behind the scenes. Explaining why things work the way they do, from my perspective. Teaching without all the polish and marketing BS.

And here's the thing: I don't need to be live to do that.

What if I could take you WITH me instead of making you and me show up at a scheduled time? What if you could see what I see when I'm at GitHub Universe, or at CES 2026 in January, or building out a smart home at my new place?

What if I could show you the behind-the-scenes of Home Assistant decisions, the architecture choices, the controversial stuff, the "why the hell did they do it that way" moments? But do it on my schedule and yours, and with MY perspective on them?

That's when I realized: I don't need to stream. I need to vlog. 🤳

Enter: The DJI Osmo Pocket 3

I got this little camera. You'll see it a lot. It's small enough that I can record anywhere, anytime, without a big production setup.

Walking through a conference explaining what I'm seeing? Done.

Sitting in my car sharing my thoughts on a controversial decision? Easy.

Building out a room in my new house and explaining MY integration choices? Or even better, showing my failures? Perfect.

It's the spontaneity I loved about streaming, but without requiring you and me to be there at exactly 8 PM on a Tuesday and Friday.

What's actually coming

So here's what I'm doing:

Vlogs. Raw, unscripted, taking you places. GitHub Universe: I'm taking you with me to share what I see. CES 2026 in January? You're coming. A Home Assistant community meetup? You'll be there. My perspective on all of it.

The smart home build. I'm moving to a new house in December, and it's got ridiculous smart home potential. I'm documenting the build from day one. Device choices I make. Integrations I set up. Of course, every time something breaks spectacularly (and it will). You're seeing all of it.

Behind the scenes of Home Assistant. What it's actually like making decisions that affect millions of users, at least from where I sit. The debates. The trade-offs. The 2am realizations. The mistakes. Not the sanitized PR version, my version. My perspective as someone deeply involved.

But here's the thing I'm most excited about...

Ask Frenck: Unfiltered

I'm launching a series. Every episode, I'm taking ONE question, a meaty, interesting, maybe controversial question, and I'm giving you 10-15 minutes of MY brutally honest answer.

This is NOT rapid-fire Q&A where I answer 20 surface-level questions, like question about my taste of music...?! 🕺

I'm talking about questions like:

• "Why doesn't Home Assistant have RBAC support yet?"
• "What's the actual difference between add-ons and integrations, and why is it so confusing?"
• "Why did Home Assistant deprecate older 32-bits architectures? Doesn't that collide with the sustainability values of the Open Home Foundation?"
• "Why do backward-incompatible (also known as breaking changes) keep happening?"
• "Is Home Assistant too complicated for beginners and becoming too limited for the advanced users?"

The hard questions. The ones that deserve real explanation. The ones you've been arguing about on Reddit at 2am. 😬

I'm going to pick one every episode and explain it. Completely. Honestly. No marketing speak. No corporate BS. No beating around the bush.

Why? Because I'm the one making a lot of these decisions, or at least deeply involved in them. And I think you deserve to know WHY things are the way they are. The constraints we work under. The trade-offs we make. The things we tried that failed.

The real version.

Why this matters

Look, there are a lot of Home Assistant YouTube channels. Most of them do tutorials. "Here's how to set up X" or "Here's my automation for Y", and that's great, we need those.

But I'm not trying to be another tutorial channel.

I can show you the WHY behind the decisions. I can explain the architecture. I can be honest about the mistakes. I can give you the insider perspective that only someone in my position can give.

And I think that's actually more valuable than another "How to install HACS!" video. I rather dive into the "Why do we even need install HACS ourselves and why isn't this just built-in into Home Assistant in the first place?".

What I need from you

Drop your questions. In the comments on this post, on the YouTube video, on X/Twitter, on Mastodon, on Bluesky, on Instagram, wherever.

Ask me the hard questions. The technical ones. The controversial ones. The ones that matter.

• Why doesn't Home Assistant do [thing you wish it did]?
• How would you approach [the thing you'd like to automate]?
• Why did we decide [decision that frustrated you]?
• What's really happening with [topic everyone's confused about]?
• What's your opinion on [controversial thing]?

Simple questions are fine too. But I'm hunting for the meaty ones. The ones that deserve some time of real explanation. The ones that make you go "I've always wondered about that."

Summing it all up

Streaming isn't coming back. Not right now, not in the near future. Could it return someday in some different form? Maybe. But that's not what I'm focused on.

What's coming instead is me taking you with me. On the journey. Behind the scenes. Into the decisions. With all the honesty and mess and authenticity that I loved about streaming, but in a way that's actually sustainable.

First episode of the "Ask Frenck: Unfiltered" I will drop soon. I'm heading out for GitHub Universe the end of October, and around Christmas, I hope to start my brand new smart home journey in my new house.

If any of that sounds interesting, stick around. Subscribe on YouTube if you haven't. Sign up for the newsletter if you want the behind-the-scenes stuff in your inbox.

And most importantly: Ask me the hard questions.

Let's do this.

Choo Choo! 🚂

../Frenck

After four years of community requests and 1,650+ reactions on issue #1182, GitHub finally added YAML anchor support on September 18, 2025. Now, about three weeks later, I have mixed feelings.

I'll be honest: when I saw the announcement, I was excited. Finally, after years of begging (I think I've raised a couple of dozen times over the years in any fitting occasion I could), GitHub was going to let us write maintainable workflows. Then I read the details and felt genuinely disappointed. The implementation has a massive limitation that makes the entire feature feel half-baked: merge keys aren't supported. 🙁

Let me be clear about what this means. GitHub announced "YAML anchors" with language about "better conformance with the YAML spec". But they deliberately excluded the one YAML feature that makes anchors actually useful for composing configuration. After four years, they shipped something that technically adds anchors while missing the entire point of why we wanted them.

This is surprising. GitHub had years to get this right. Other CI/CD platforms (GitLab, Bitbucket, CircleCI) implemented full YAML anchor support including merge keys. It's not a technical mystery. The YAML 1.2 specification exists. Yet GitHub chose to ship a partial implementation that forces you into the exact same architectural constraints you already had.

I'm writing this because the announcement felt misleading, and nobody's clearly explaining what you actually get versus what the community asked for. Let me show you both the wins and the significant limitations.

What GitHub actually shipped

The changelog announcement describes "better conformance with the YAML specification." In practice, you get basic YAML 1.2.2 anchors (marked with &) and aliases (marked with *). Define something once, reference it multiple times. Standard stuff.

Here's the critical limitation buried in the documentation: merge keys (<<:) don't work. At all. They're not mentioned anywhere in GitHub's docs. They just silently fail with syntax errors if you try to use them. 🙈

If you've used GitLab CI/CD or Bitbucket Pipelines, this is shocking. Merge keys are what make anchors uniquely useful. They let you create base configurations and extend them with overrides. That pattern is standard in those platforms. GitHub deliberately chose not to implement it.

Without merge keys, you can't do something like this this:

# This does NOT work in GitHub Actions
job1: &defaults
  runs-on: ubuntu-latest
  timeout-minutes: 30
  retries: 3

job2:
  <<: *defaults  # ❌ Syntax error
  timeout-minutes: 60  # Would override timeout if merge keys worked

You're stuck with exact duplication. Want to vary one field? You duplicate the entire block or find another solution. This isn't "better conformance with the YAML specification." This is selective compliance that avoids the hard parts.

GitHub's parser is clearly custom-built rather than using standard YAML libraries that include merge key support. They made an architectural decision to exclude a feature that the community specifically wanted and that other platforms consider essential.

What you can actually do

Despite my disappointment, let me be fair: there are real use cases where anchors eliminate duplication. Just... fewer than you'd expect.

Environment variables across jobs

The most straightforward win. Define your production environment once:

jobs:
  deploy:
    environment: production
    env: &prod_env
      NODE_ENV: production
      API_URL: https://api.example.com
      LOG_LEVEL: warn
      DATABASE_URL: ${{ secrets.DATABASE_URL }}
    steps:
      - run: ./deploy.sh
  
  smoke-test:
    environment: production
    env: *prod_env
    steps:
      - run: ./smoke-test.sh

Simple. That said... this is a situation where merge keys would be needed pretty quickly; but for this example I guess it works.

Path filters

One of the earliest documented production uses comes from developer Aleksa Cukovic:

on:
  push:
    paths: &paths
      - "website/**.[tj]sx?"
      - "website/**.css"
      - "website/**.astro"
      - "website/public/**"
      - "website/package.json"
      - "website/pnpm-lock.yaml"
      - ".github/workflows/deploy.yml"
    branches:
      - main
  pull_request:
    paths: *paths
    branches:
      - main

Maintaining identical path filters across push and pull_request triggers was annoying. Add a new file pattern once, both triggers update automatically.

Service containers

Database configurations get complex fast:

jobs:
  integration-test:
    runs-on: ubuntu-latest
    services:
      db: &postgres
        image: postgres:15
        env:
          POSTGRES_PASSWORD: postgres
          POSTGRES_DB: test_db
        options: >-
          --health-cmd pg_isready
          --health-interval 10s
          --health-timeout 5s
          --health-retries 5
        ports:
          - 5432:5432
    steps:
      - run: npm run test:integration
  
  e2e-test:
    runs-on: ubuntu-latest
    services:
      db: *postgres
    steps:
      - run: npm run test:e2e

Multiple test jobs need identical Postgres setup. The service definition is complex enough (health checks, ports, environment) that duplication invites errors.

Home Assistant actually migrated

Despite my reservations, we did refactor Home Assistant's CI workflows. Pull request #152586 reduced our workflow configuration from 504 lines to 206 lines. That's a meaningful improvement.

The PR description acknowledges the limitation directly: "Github finally added support for YAML anchors in workflow files. They don't support key-merge / overrides but it's at least a start".

The migration worked because we had straightforward duplication. But we hit the merge key limitation immediately. We have base test configurations with environment-specific overrides. Without merge keys, we had to choose: accept some duplication or restructure jobs entirely. We chose to accept duplication for those cases. The alternative was making the workflow harder to understand just to work around GitHub's limitation.

Here's the thing: that 298-line reduction (504 to 206) sounds impressive. But with full anchor support including merge keys, we could have cut even more. There are patterns we couldn't consolidate because GitHub's implementation doesn't let us compose configurations.

The security controversy nobody's addressing

Security peeps are not happy about anchors. The problem is analysis tools.

Static analysis tools like actionlint, zizmor, and poutine map workflow configuration to source locations. When something goes wrong, they tell you "line 47 has a permission issue." Anchors break this. One anchor definition at line 10 might be referenced at lines 50, 150, and 300. When a security issue appears, which line is the problem?

One developer put it bluntly: this "blows a massive hole in the larger open source ecosystem's ability to analyze GitHub Actions for correctness and security."

The concern is valid. Anchors create non-locality. Configuration defined in one place executes in many places. Code review gets harder. Junior developers might reference anchors without understanding what they're inheriting. A permissive permissions block in an anchored definition affects every reference.

Here's my perspective on this: the security concerns are real, but they're solvable. GitLab and Bitbucket have had anchors for years. Security tools adapted. The analysis complexity is a trade-off, not a showstopper.

What frustrates me is that GitHub shipped anchors without merge keys, gaining all the security analysis complexity while missing the compositional benefits that would justify that complexity. We get the costs without the full benefits.

Use anchors for simple, obvious duplication. Avoid them for security-critical configuration like permissions blocks or secret handling. The explicitness of duplication aids security reviews. This is pragmatic advice for today, but it's advice we shouldn't need if GitHub had implemented the feature properly.

Why this feels disappointing

I maintain Home Assistant. We have complex CI/CD needs. I've used GitLab CI/CD before. I know what full YAML anchor support looks like. I know what you can accomplish with merge keys.

When Issue #1182 stayed open for four years with 1,650+ reactions, I assumed GitHub was working on a complete implementation. Maybe they were solving technical challenges. Maybe they were considering the right abstractions.

Then they shipped this. Basic anchors with aliases. No merge keys. No explanation of why. No roadmap for adding them later. Just "better conformance with the YAML specification" that's actually selective conformance that avoids the useful parts.

It's surprising that after all these years, GitHub still can't (or won't) comply with what the YAML specification offers and what competing platforms have implemented. The technical capability exists. Other platforms proved it's doable. GitHub chose not to do it.

The announcement felt misleading. "YAML anchors" implies you're getting YAML anchor functionality. What you actually get is YAML anchor syntax for a subset of use cases. That's a significant difference that the announcement glossed over.

Anchors versus alternatives

You already have other options for eliminating duplication. How do anchors compare?

Reusable workflows let you share entire workflows across repositories. Perfect for organization-wide deployment patterns. Overkill for eliminating duplicate environment variables in one workflow. You also can't add steps before or after a reusable workflow call.

Composite actions bundle step sequences into reusable actions stored in .github/actions/. These work cross-repository and provide proper encapsulation with inputs and outputs. However, they require separate action.yml files, can't specify runners, and add invocation overhead. For simple within-file duplication, composite actions feel heavyweight.

YAML anchors eliminate duplication within one file with zero setup overhead. No separate files, no input definitions, no shell specifications. They work for any YAML node: environment variables, steps, entire jobs.

The architecture forces a choice:

• Reusable workflows: Multiple jobs, specify runners, but cannot add steps before/after
• Composite actions: Bundle steps, work cross-repo, allow steps before/after, but cannot specify runners
• YAML anchors: Within-file only, no parameterization, no composition (no merge keys), but zero setup overhead

Most mature workflows combine all three. Use workflow-level env blocks for truly global variables, anchors for within-file duplication, composite actions for reusable step patterns across projects, and reusable workflows for complete deployment patterns.

The problem is that without merge keys, anchors can't fill the compositional gap. You're still choosing between duplication and heavyweight abstractions for any case where you need base configuration with overrides.

Should you refactor your workflows?

Depends on your pain points and expectations.

Yes, if: You have obvious duplication (environment variables repeated across 3+ jobs, identical step sequences, repeated service containers) and you're okay with the "exact duplication only" constraint. The feature is stable despite being new. GitHub enabled it automatically for all repositories. It won't break existing workflows.

No, if: You expected merge key support and need compositional patterns. You'll be disappointed, like I was. Also skip it if your workflows are small and duplication isn't painful. Adding anchors to a 100-line workflow with minimal repetition makes it harder to read, not easier.

Summing it all up

GitHub Actions now supports YAML anchors for basic duplication elimination. You can anchor environment variables, step sequences, service containers, and path filters. This eliminates maintenance burden for simple repeated configuration.

You cannot use merge keys to compose configurations with overrides. This is a significant limitation compared to other CI/CD platforms that implemented full YAML anchor support years ago. It constrains what you can accomplish and forces architectural decisions about when duplication is acceptable versus when you need composite actions or reusable workflows.

At the time of writing, this feature is just a few weeks old. Real world examples are scarce. Valid concerns have been raised about analysis complexity. Early adopters like us at Home Assistant are still figuring out best practices.

Here's my honest take: use anchors for obvious, simple duplication. Skip them for complex compositional needs or security-critical configuration. The benefits are real but modest. The limitations are frustrating.

And honestly? After four years of waiting, I expected better. I expected GitHub to implement YAML anchors properly, with merge keys, like every other major CI/CD platform. Instead, they shipped something that technically adds anchors while deliberately excluding the functionality that makes them transformative.

The announcement said "better conformance with the YAML specification." That phrase bothers me. YAML anchors exist in the spec alongside merge keys. GitHub implemented half of it and called it conformance. That's not conformance. That's selective implementation.

Maybe GitHub will add merge keys in a future update. Maybe there are technical reasons they haven't shared. Maybe they fundamentally disagree with the community about what's useful. I don't know.

What I do know is that after four years and 1,650+ reactions asking for this feature, the community deserved either full implementation or honest communication about why that's not happening.

../Frenck

People on the internet sometimes seem to have an idea of what my connected smart home looks like. And honestly? They're all wrong. 😄 Which is odd, considering I've shared my setup on multiple occasions–through multiple live streams, podcasts, and blog posts. Very recently, I came across a Reddit comment that stuck with me:

“They don’t even use Home Assistant”.

That Reddit comment has been resonating in my head for weeks. Is this person right? Am I even allowed to say I have a smart home? Do I practice what I preach?

I guess it's time for some confessions about the state of my smart home. 😅

The unusual start of my smart home journey

My Home Assistant journey was, I guess, anything but typical. If you are using Home Assistant today, chances are you started with another system–Google Home, Amazon Alexa, Apple Home, OpenHAB, Homey, Domoticz, or maybe something self-built– before looking for more flexibility, power, or local control. You are what we call an "outgrower". In that case, you have outgrown your previous smart home system and have been looking to do more.

Here’s my first confession: I was never an outgrower.

When I first stumbled upon Home Assistant, I didn’t know anything about smart homes. I didn’t have any smart devices (at least not that I was really aware of); I wasn't even thinking about automating my home at all.

I was just bored to death. 🙈

I'm not even kidding. I wanted to contribute to an open-source project–something where I could apply my development skills and give back to the community. That's how I found Home Assistant. I came across some issues I could solve, and while doing that, I discovered the amazing community 🫶 that surrounded it. It pulled me in deeper and deeper.

And yes, it is true: For the first year of contributing to Home Assistant, I didn’t even use it myself. Sure enough, I had test setups for development but no actual smart bulbs or sensors in my home. Who even wants that? 🥴

No worries–the saying is real–once you start going down the home automation rabbit hole, it's hard to stop. Eventually, I set up my own "production" instance alongside my dev and test setups. And slowly, more and more connected smart devices started to sneak into my home.

The ever-growing pile of devices in my smart home

Haha, yeah, so here we go. People assume that, as the lead engineer of Home Assistant, I must have the most extravagant 🤩 and insane smart home setup in the world…

My second confession: Yeah, well... no. 🤣

The reality? I have hundreds of smart devices and sensors in my home. But I have no idea which ones are actually connected. And if they are connected, I'm not even sure which of my many Home Assistant instances they're hooked up to. 😬

I basically run four production instances of Home Assistant, all originally meant to become the ultimate smart home setup. Each runs on different hardware: a Home Assistant Blue, a Home Assistant Yellow, a Home Assistant Green, and one running on a Beelink Mini S12 Pro Mini PC.

To make my mess even more complete, I have a massive mix of devices:

• Zigbee, Z-Wave, and Matter devices over Wi-Fi and Thread. 🛜
• Tons of Wi-Fi devices running ESPHome, WLED, or other (custom) firmware.
• A ridiculous number of cloud-based devices (😫 yeah, I know... a topic for another day).
• And I don't think I own more than five devices from the same brand.

Why? Because I love experimenting—testing the latest smart home tech, exploring what’s out there, and staying ahead in the industry. I want to keep fueling my ideas with innovation, helping me push Home Assistant forward. There is so much interesting stuff! 😵‍💫

Imagine throwing a huge pile of smart home gadgets into your house, barely setting them up, and calling it a day. Congratulations–you now have my smart home. 😓

If you’re curious about the smart devices I actually use, here’s a list of my favorites:

Franck Nijhof | Frenck’s Amazon Page - Found in my own smart home

Shop recommended products from Franck Nijhof | Frenck on www.amazon.com. Learn more about Franck Nijhof | Frenck’s favorite products.

Frenck's Amazon Page

Automations? Kinda...

Given the above, you can imagine that controlling devices in my home is... complicated. 😅 Some devices, I just use the old-fashioned way–yep, I actually press the physical buttons of these smart devices. 🙈

So for the third confession: Some parts are automated, but... badly! 😰

Heck, some automations work only partially. For example, I have an automation to turn off some devices when I leave home. But the kitchen lights? They stubbornly stay on. Why? I have no idea. It is probably a simple fix, but I haven't fixed it yet.. because things just happen (or don't). 🤷

That said, I do love writing automations–especially the old-school way, directly in YAML. At one point, that kitchen light automation reached over 700 lines of YAML 🙈 Now you know why it's broken 🤣

If I had to count, I’d guess I have maybe 15 working automations across all my half-baked instances. That's it.

My dashboards are a disaster

Dashboards are one of Home Assistant's most powerful features. With the new drag-and-drop capabilities, they're easier and more flexible than ever. But my dashboards?

Fourth confession: They're a total mess. Easily the worst part of my setup.

”Can you share some screenshots of your dashboards, Frenck?"
“Show us what you’ve got!”

I hear this all the time. And the truth is, technically, I do share them–almost every month. When I write the Home Assistant release notes, most of the screenshots come from one of my own production instances.

Yes, this means I regularly break my own home setup just so I can provide good examples for the release notes. As a matter of fact, because of it, my dashboards are so bad that if I turn on the "Coffee machine" switch, it actually powers on my 3D Printer. 🤦‍♂️

Home Approval Factor

I don't live alone. Mrs Frenck (also known as the Ninja 🥷) and the kids have to deal with this madness too.

How do they manage? Honestly... I have no idea 😓

Confession five: I've never asked them... so that approval factor is non-existent, I guess... 🙃

Here are a few examples:

• Many wall switches do not work.
• Mrs. Frenck has the Home Assistant Companion app to control one of the instances–at least, the one with the most devices in it.
• The kids? They just ask us to turn things on and off. Which is really ridiculous.

But over time, I guess they got used to it. 😓 They accept that things don't always work–or at least, they stopped complaining about it.

The fact that they’ve just "accepted" things don’t always work—that stuff is flaky, experimental, or sometimes just plain stupid—is probably even more worrying. 😩

So, what does your smart home actually look like?

I guess my smart home is best described as a real-life, constantly evolving experiment–a playground where I test devices, reproduce bugs for the community, stage demos, and create release note screenshots.

I guess I do have a "smart home". But not in the way people imagine. And yet... I want to fix it.

Confession six: I want to have the most extravagant, highly optimized smart home setup in the world. 😎

Don't we all? 🙃

Every time I set aside a free Sunday to work on my own setup, it takes about 30 minutes before I realize there's something I could fix or improve for everyone. And suddenly, instead of fixing my own mess, I'm contributing to Home Assistant again.

But for my family's sake, I need to sit down and do it right. They've put up with this disaster for too many years.

So, to Mrs. Frenck and my kids:

I'm sorry! 🙏 I'll fix it!

Someday.

../Frenck

In January 2025, I visited the Consumer Electronics Show (CES) in Las Vegas to view all the latest products and innovations in consumer technology coming in the upcoming years. As a full-time employee of Nabu Casa working on the Open Home Foundation's Home Assistant project, I was particularly interested in where the IoT landscape is heading for the smart home industry, both in terms of technologies and product innovations.

I was not alone at CES; besides me, four others from Nabu Casa joined, allowing us to explore all the floors of this enormous expo–Talking to as many companies as possible, strengthening existing relationships, and making potential new ones; as we seek to boost and push the Works with Home Assistant program, and finding new products and even whole product categories to integrate with Home Assistant.

Let me tell you a little bit about my views and experiences from visiting this expo.

Home Assistant is getting hotter 🔥

There's no surprise there, right? Home Assistant is showing up at many places around the whole CES expo; the logo is printed on many places 🤩 It was cool to spot the logo–it is a sign of how the industry is taking the Home Assistant open-source project more seriously. Even nicer, I've seen multiple Home Assistant Greens being on display!

A small little Home Assistant meetup was organized as well, which was awesome! Just look at the group picture below! Like... CES was cool, don't get me wrong, but just look at this amazing group of people in this picture! It was an honor to meet all of them 🙏

I want to specifically call out Justin (DrZzs). I've known him since the beginning of my Home Assistant journey many years ago; we even did some live streams together back in the day. It was a true honor 🙏 to meet you in person!

Also, Phil & Rohan, of the Home Assistant Podcast, we go back quite a long time as well. I've been a guest on their show a few times and even took over the show once. These guys are absolutely the nicest and rock even harder in real life. 👊

There are many more amazing people in this picture, like digiblur DIY, The Hook Up, Automated_House, CloudFree, people from Coulisse, and the lovely guys from Apollo Automation. 🤗

This was loads of fun! But let's dive into the actual CES expo...

AI is everywhere

Surprise! (not really) AI is everywhere; every company does its best to be able to slap on the AI 🪄 word on basically everything possible. No company wants to miss out on this hype, and even if they don't have a use case for using AI in their products, they will find something utterly useless to AI-ify so they can show they are aboard this hype train.

Personally, I'm not really triggered by all this AI stuff anymore. I honestly think I'm more bored and annoyed by it right now. Don't get me wrong, I do believe the generative AI era we are in right now is fascinating and can be used for a lot of cool things, but at this point, in my opinion, 95% of the AI-enabled features showing up in all products is pure useless overhyped crap. In a way, it might be fine? I guess the world needs to learn how to put gen AI to use helpfully by trying as much as possible, in as many ways and use cases as we can.

I'm most worried that almost all of this AI stuff out there isn't local and relies on cloud services. These services are expensive to keep up and maintain. Someone has to pay for it in the long run? Are we losing features when they pull the plug? Or do some products stop working? Do they pull the rug and force you into getting a subscription? Or worse... would they sell your data to pay for all of this? Food for thought. 🤔

The best AI "gimmick" I've seen at CES came from a small booth of the VideoLAN project. They are known for the (open source 🎉) VLC Media player, and showed how they can translate subtitles to any video using AI! Now, that is a practical application of AI, and it is very impressive!

Summing up my AI views: It is interesting and needs to find its place. Meanwhile, the market is completely flooded with utterly useless overhyped AI functionality crap that is mostly cloud-dependent and awaiting for the enshitification to happen.

The smart home extends with smart health

While every manufacturer is smacking AI on everything, another bigger theme is visible across the whole smart home industry: the movement from smart home to smart living.

No longer are we only looking at the most innovative light bulbs, vacuum cleaners, or locks, but also sensors one can install in one's toilet to track certain health indicators in your urine, smart scales, dozens of rings, fitness equipment, adult sex toys, and a lot more.

The smart home industry is no longer just about the smart home with our fancy lights, switches, and TVs; it is shifting quite strongly to care about the people who live in it.

This is not entirely new, though. You might have noticed this trend yourself over the last few years: sleep trackers have become the norm, smart watches measure a whole lot more nowadays, and for the smart home, more options have become available to measure air quality at home.

Personally, I think this makes a lot of sense, and I'm curious to see how all this smart health is going to connect up and be integrated with the smart home.

However, I'm also worried about how this hunger for recording medical health data would invade our personal privacy–even within the safe space of our own homes. I can perfectly imagine that not everybody would be OK with integrating smart health into a smart home that has multiple people living in it.

All hail the vacuum cleaners

It is not just the era of AI or smart health. Apparently, vacuum cleaners are the hot thing! I mean... I'm not sure why this is the case, but Oh my god, how many vacuum cleaners could be showcased at a single expo?

This year, some can jump over little bumps in your home, others have a robot arm to move socks out of the way, and another can carry your tablet or humidifier around the house. Okay, the jumping over small bumps thing might be practical; the others may be more of a gimmick IMHO. We are definitely at a point where vacuum cleaners become harder to innovate on.

You have probably seen most of them in all media coverage around CES, as these vacuum cleaners were definitely super prominent and visible. Our buddy Rob, The Hook Up, made quite a nice video that sums up all there is to know about the vacuum cleaners from CES 2025:

Meanwhile, my good old Roborock S5 Max is still going strong, which feels like a piece of ancient history after looking at all of these amazing vacuum cleaners at CES. It makes me wonder if spending a thousand on a new robot vacuum is worth it? 🤔

Does it Matter?

Matter! The “brand-new” smart home standard—though, at this point, it’s not so new anymore. Matter was launched with the promise of solving interoperability between smart home devices, allowing products from different manufacturers to work together seamlessly across platforms like Apple Home, Google Home, Amazon Alexa, and, of course, Home Assistant. Instead of worrying about whether a device supports Zigbee, Z-Wave, or Wi-Fi, Matter aims to be the universal standard that just works.

Matter was most definitely a highlight at CES. Many product manufacturers have added support for Matter or are working on adding support for it to their products, which is awesome for us consumers, considering the goal of Matter!

Sounds great, right? Well… in theory... yes.

I visited some of these Matter product manufacturers at their booths and asked them for their views on Matter and the challenges they have faced creating these products. Their responses were shockingly similar.

In almost all cases, Matter is added as an additional feature. This means that almost everyone keeps their existing apps and ecosystems working in tandem. Matter isn’t used as a standalone system; instead, it acts as an additional way for consumers to control their devices.

The first thing I thought was that this made tons of sense, as they don't want to risk breaking the experience of their existing ecosystems. But essentially, they could have just used Matter for this on their end as well. Why didn't they?  The most common response I heard was: “Matter doesn’t support all the features we want to offer.” That might be true, but Matter does support adding custom clusters for that use case. 🤷‍♂️ I think it is more a signal of how "new" Matter still is.

Consumers don't understand Matter yet. We all know and accepted Wi-Fi by now, but Matter? This is a huge problem, and it is not just a teaching problem but also a huge UX problem on all possible levels. Let's be honest: If you see a Matter logo printed on a box in a store, would you know if you can connect it to your home at this point? Most people probably answer this as: "I don't know." The fact that Matter can run over both Wi-Fi and Thread makes it even more confusing.

Thread? In case you don't know what it is: Thread is a low-powered mesh networking wireless radio protocol for efficient communication between smart devices. Unlike Wi-Fi, Thread creates a local, self-healing mesh network where devices can relay signals to one another without relying on a central hub. This makes it more reliable in theory, as there’s no single point of failure, and it’s energy-efficient—ideal for battery-powered devices like sensors and locks.

However, the reality is much messier. Thread requires a Thread Border Router to connect with the rest of the smart home, and different ecosystems (Apple, Google, Amazon, etc.) all have their own incompatible implementations. The result? Devices often struggle to join the right network, and compatibility issues between Apple, Google, and Amazon’s Thread implementations only add to the frustration.

This makes Thread the Achilles heel of Matter. Every single person I talked to at CES dislikes Thread. It causes a ton of support load, and consumers get stuck and get a shitty user experience. Matter over Thread products have by far the lowest first-time setup success rate for consumers for every single company I spoke to. End users don’t know better and blame Matter for the poor experience, not Thread—which ultimately reflects badly on Matter. Many manufacturers avoid Thread right now for that reason. Hopefully, Thread version 1.4 will help, as it promises to solve many of these issues.

Why Thread is Matter’s biggest problem right now

The fight for who will own your Thread mesh network is creating an even more fractured smart home.

The VergeJennifer Pattison Tuohy

A concerning number of manufacturers create and release bridges/controllers instead of actual Matter end devices. This means that the actual device doesn't support Matter, but a bridge in between does.

For example, Philips Hue does this. Their Hue bridge can talk Matter to other ecosystems, and the bridge will talk to all the Philips Hue lights using Zigbee. Flic makes things even weirder, in my opinion. Their Flic Hubs are not a Matter bridge, but a controller. This means they try to sit in between as a mini-smart home system instead of exposing their Flic Buttons as devices using a bridge.

In the end, I think both harm the promise and vision of Matter. In the case of Philips Hue, they force you to use their bridge and ecosystem, even if you want to use Matter. You still need to create a Hue account and log in with it on your Hue bridge in order to use your Hue lights over Matter. That way, they keep you locked into their ecosystem.

Flic is probably working around Matter's feature limitations this way (their buttons can do a lot), but... a Matter network can only have five controllers (called fabrics)... If everyone takes this route, things will become problematic pretty quickly.

Other manufacturers I spoke with want to avoid certifying every single end device for Matter, as this adds to product costs, which can be prohibitive for smaller companies or start-ups.

While there are a lot of concerns around Matter, it is still relatively new, and even though Thread is a party pooper at this point–Matter definitely has momentum, and I hope that over time, this will help resolve its issues and allow it to truly deliver on its promise.

Energy consumption awareness hotness

Do you know what did matter at CES? Energy! ⚡️Making consumers aware of their energy consumption is no longer a nice concern–it's a key feature of every smart home. While some parts of the world (cough the US cough) are figuring out renewables 🌱 versus fossil fuels 🛢️, the smart home world is clearly moving towards energy efficiency, awareness, and sustainability.

More and more ecosystems focus on helping users track, manage, and reduce their energy usage—whether to lower their environmental footprint or simply save money on rising energy costs. Energy costs are rising worldwide, especially in Europe. 

Samsung is really not the only company. Preserving Mother Earth is a big item for many. From tons of smart plugs that provide real-time energy monitoring, to intelligent thermostats that include real-time power insights and energy-saving features everywhere.

CES wouldn’t be CES without concept cars (and similar ‘vehicles’), and unsurprisingly, they were almost all EVs. In fact, there might have been more EV chargers on display than vacuum cleaners! 😆 If you have an EV, you know it takes a large sum of your total energy usage. Having insights in that, and charging on smart moments is definitely something to save a lot of money on.

With that in mind, I expected to see a lot more solar tech. But aside from some inverters, I didn’t really notice a strong presence for it at CES—though maybe I just missed it… totally possible! What was visible was batteries! 🔋

Besides insights, batteries are showing everywhere. Companies are clearly heating up for a future where homes can store excess energy from solar panels or charge when electricity is cheap and use the stored power when rates are high. I'm not sure if this is an American thing or not, but large portable batteries (the ones with AC outlets on them) were literally everywhere.

I recently received a HomeWizard Plugin-In Battery. This one is specifically aimed at the Dutch market and is specifically created to store excess solar energy. The best thing about it? Just unpack it, plug it into a wall socket, done!

Some other things I've spotted and loved...

As I wrote in the introduction of this article, I've been searching for new things to integrate into Home Assistant. And honestly, I think there is enough choice when it comes to getting a light bulb to integrate into your smart home. But what else could we potentially integrate into our Smart Home? Here are some other small little gems I've spotted.

How about this smart bird feeder by Birdbuddy? 🦜 The camera captures pictures and videos, and the AI identifies the bird species to categorize the visit. I have one in my garden for a bit already, and it is loads of fun for the kids!

Next up is this smart indoor garden by Plantaform 🌱 It really caught my eye at the expo. This extremely nice futuristic-looking pod device is actually a mini garden to grow your own fresh herbs right at home. Just look how beautiful it is:

Unlike LETPOT, Plantaform can't integrate with Home Assistant yet, but I think we will see more of these indoor gardens in our smart homes in the future. It would fit perfectly, especially with notifications for things like giving water (which is why every plant basically dies in my house 🙃).

Talking about water, 💧Hydrific's Droplet, sure enough, looked interesting. This little device can be self-installed by clamping it onto your main water pipe, and it will tell you how much water you are using, and you can even figure out patterns using AI. More importantly, it is able to detect leaks and notify you of those! 🙏 It unfortunately doesn't work with Home Assistant yet; hopefully, that will change someday soon!

Alright, this sums up my report/view of CES 2025. I hope I've been able to deliver my slightly different view/perspective of this massive expo.

What do you expect to see in terms of innovation in the smart home world in the upcoming years?

../Frenck

Smart home tech keeps evolving, and HomeWizard is bringing something new to the table—a home battery that you can install yourself. No electrician, no complicated setup—just plug it in, and it works (allegedly).

As someone who loves testing smart home devices, I had to put this to the test! In my latest video, I unbox the HomeWizard Plug-In Battery, set it up, and check if it really integrates smoothly with Home Assistant.

Does it live up to the promise of being truly plug-and-play? Or is it just another overhyped gadget? 🤔

📺 Watch the full unboxing:

What do you think about self-install home batteries?

A few weeks ago, I became a GitHub Star, and today, a surprise package arrived from GitHub! 🤩 Naturally, I had to do an unboxing video—my first-ever attempt, so you can expect top-tier professional production skills (LOL 😂).

What’s inside? Let’s just say there are a LOT of Monas! 🐱

🎥 Watch the unboxing here:

PS: In case you wonder, after making the video recording, I basically lost some box contents to my daughter Lilly already... 😆

Yesterday, I wrote about my New Year's resolutions for 2025, to kick start this fresh new year. However, GitHub is sweetening the deal even more!

I'm now a GitHub Star! 🌟

So, what does that mean? Let me share a little snippet from their website:

The GitHub Stars program thanks GitHub’s most influential developers and gives them a platform to showcase their work, reach more people, and shape the future of GitHub.

I can't even start to express how incredibly honored, proud, happy, and super thankful I am! This recognition is most definitely a star ⭐ on all the open source work I have been doing since 2017. Most notably, of course, my work on the open source smart home software: Home Assistant.

Looking at the other 90-ish GitHub Stars in the program makes me feel quite small and humble. There are so many rock stars in this program that I look up to 😳. I'm honestly just some guy from the Netherlands writing some code behind his desk in his living room... 🫣 Imposter syndrome, as I have been told. 😅

I was at GitHub Universe 2024, where I met a lot of people, including many Hubbers (as GitHub employees call themselves) and existing GitHub Stars. Every single one of them have been so incredibly nice and like-minded; you all made me feel right at home. 🤗

Besides this nice-looking star, I will also have a voice in shaping GitHub's future, meaning early feature access and access to insight and brainstorming calls with their teams. Additionally, they will feature my profile and give me access to training resources. Lastly, we will help each other spread the word about open source all around the world. 🤝

To all who have nominated, selected, and accepted me for a GitHub Star, a sincere thank you! ❤️

../Frenck

Yeah, I know, these resolutions are boring as F*ck 🖕. Some might even say having a New Year resolution is key to making resolutions fail! (Yes, Missy, I'm referring to you) 🙃 Whatever everyone thinks, I'm going to try! 💪

Soon™, I'll publish a blog summarizing my 2024, as it was an extremely interesting and eventful year. 2025, however, isn't kicking off slowly either. I'm writing this while I'm actually on a plane back home from Las Vegas after visiting the CES 2025 expo (Consumer Electronics Show, the biggest expo on any consumer technology out there), which is an adventure on its own; I will write a blog on that soon™ as the smart home was very well represented this year.

Anyway, back to my New Year's resolutions.

1. Write my ass off

I love to write. ✍️ I actually write a lot. In fact, even today, I still write most of the Home Assistant release notes myself, and I write a lot in my journals. Heck, I even wrote many blog posts I've never published! 😭

I really don't understand why I never published those blogs, though; I've been thinking about "why". Maybe my website is not up to par? Maybe WordPress is blocking it? (To be honest, it is kind of slow and maintenance-heavy. Let's not get started on all the drama around it).

I also realize it is just a flimsy excuse 🙈 Yet, I've taken the time to migrate my website to the Ghost platform. Ghost is nice, simple, fast, and easy to write and work with, and I've opted to let them host it for me, so that excuse no longer flies now 😆

Welcome to my new website 🎉 And not just that, this change comes with the ability for me to write newsletters as well. 🗞️

Okay, so I'm going to do more writing, blogging, newsletters, spam socials, and whatnot. What do I want to achieve? If I look back to 2025 in 2026, when would I call this resolution achieved? I'm going to set the bar high on this one.

• At least 365 messages were published on my socials.
• At least 26 new blog articles were published.
• At least 26 newsletters were sent.

😅 Already sweating as I'm writing out these... Let's go! 💪 🚀

2. Contribute to open source

I'm a full-time open sourcerer, meaning I get to contribute code to open source projects all day. I work full-time as an engineering lead for Home Assistant.

In 2024, I made around 700 open-source contributions (in this case, code contributions via a pull request). This is a lot, but honestly, it has been the lowest number I've had in years. My peak was in 2021 when I made around 3600 pull requests. That was a crazy figure.

Okay, so I'd like to make more open-source contributions. When will I achieve this goal? I'd be very happy if I could make 1000 pull requests this year.

I've reset my little physical pull request counter, so let's go! 💪 🚀

3. A working smart home

"Hey Frenck, since you work on Home Assistant, your smart home must be crazy interesting, right?"

Yeah, it is crazy interesting because... NOTHING works! I'm not even kidding. This is not Home Assistant's fault; I am the one to blame for this.

Over the years, my home has become a test lab for Home Assistant. I constantly break stuff (on purpose), reproduce bugs, constantly try out new concepts, or rename stuff for a screenshot in the release notes. Heck, I run the development version of Home Assistant on my "main production" Home Assistant instance. Yet, an awful lot of things are being actively controlled through it.

PS: While writing this, I realized I could probably write a blog on the current state of my smart home... 🤔 Guess that will help with my first resolution 😄

The above screenshot says it all. The background was to test the recently added capability to upload backgrounds (spoiler... it has nothing to do with my office). "Meeting mode" is unavailable... because it actually controls the flycatcher, which I only have around in the summer. The "3D Printer" toggle is interesting as well, as touching those controls will not control my 3D printer but my coffee machine 😆

TL;DR: My smart home is a joke, and the people living in it are the victims of the connected home mess I've carefully crafted. 😓

The resolution, in this case, leaves not much room for guessing: FIX IT! 🛠️

I found it hard to define what a "success" for this resolution would be, as much of it is subjective. I can at least describe some things; for example, all devices should be connected, working, and well categorized. The energy dashboard should be neat, and a bunch of working automations must be contributing to the comfort of not just me but also my family living in our home.

The most challenging part for me would be dashboarding. I'm more of an automation type of person, but my family should be able to have a fallback they can understand and use. I should ensure they can operate every device from a sensible dashboard.

I'll take you along my journey and share my stories. Let's go! 💪 🚀

4. Home improvement and chores

As we say in Dutch: "Omdat je huis nooit af is"; I don't think there is an English variant of this expression, but it means that one's home is never finished. I'm sure you recognize that.

I have a lot of chores and things I still need to do. Install roof insulation and renovate the top-floor rooms for the kids. My son's old room can then be renovated to become our new storage/guest room, and finally, the previous storage/guest room will become my new office!

I started working from my living room in 2019 and still do. I was amazing while the kids were little, but them growing up now also means that me sitting in the living room isn't always the most practical. They stay up longer in the evenings, take home their friends to hang out with, and do more of these things. Especially when I have video calls or meetings, I don't want them to have to adjust to accommodate for me having these calls in silence.

So, for this year, let's get myself settled in my new office! Let's go! 💪 🚀

But wait... there is more!

I like these four resolutions actually. They fit me. Of course, I think I could improve or improve many more things (including the common ones you'll see everywhere). Some examples include:

• I could lose a few pounds and exercise more! Honestly, I can't wait for spring to arrive so I can do outdoor workouts on my road bike again 🚴‍♂️
• I want to try live-streaming again. Sure, I am live once a month when we do the Home Assistant release parties, but I used to stream as well, personally. I really loved doing that. Might I hop on a few times while reading out fixing my own smart home.
• I would like to appear on a few podcasts. I love doing those; there is something magical and unique about them. Did you know I had my own podcast many, many, MANY moons ago? 😆
• I'd like to visit a few expos and developer conferences and meet like-minded people! I'd also like to apply for a few calls for papers and speak at some.
• I've written a few integrations for Home Assistant this year, and I want all of them to rank as high as possible on the Integration Quality Scale. I also still have a few concepts lying around, and I'd like to revive them to see if they are still worthy of a shot.
• Improve work/life balance a bit. I'm a developer; it is my hobby and my full-time job. Working fully remote, doing what I love, dilutes the lines between work and hobby. I don't have too many issues with that, but I should always keep it in mind and constantly improve.
• More date nights! I just want to get out more with my fiancee for dinner, theater, etc. In general, always spend a little more time with your loved ones! ❤️

Let that last one sink in for a minute... and with that, all I've left to say is: Happy New Year! 🥂

What is your New Year's resolution?

../Frenck

We’re having amazing weather in the Netherlands 🇳🇱 right now! Even though it’s technically spring, it feels like summer, with temperatures hitting 25 degrees Celsius already in early May 😎. This is super rare for us Dutchies, so we’re really making the most out of it by spending lots of time in our garden!

However, I have noticed that my solar panels on the roof are looking quite dirty. They are covered with a layer of green grime mixed with sand from the Sahara that was carried over by recent weather conditions. It’s time to clean them up!

Please note that my home has a water filtration and softener system installed. This means that the water that comes out of my tap is clean and demineralized. If you don’t have a similar system, I wouldn’t recommend using regular tap water for this purpose. Instead, you could consider collecting and using rainwater.

Online resources suggest that dirty solar panels can cause a 5-20% loss of energy production, but the exact percentage can vary depending on the source. It is difficult to determine the exact amount of energy loss in my case, as environmental conditions can change during the cleaning process. The sun’s intensity changes, clouds move, and the earth keeps spinning, making comparing power changes before and after cleaning challenging.

What I do know: They look nice and shiny again 🤩

../Frenck

This blog is for those people whose daily job is IT-related (DevOps, developers, system administrators, etc) and who are using Home Assistant at home. If you fit that description, you just might suffer from what I call: The enterprise smart home syndrome.

When working in IT, you’ll often end up having to design, build, maintain, or at least work with, all kinds of enterprise solutions. Enterprise solutions that keep everything, in the company you work for, afloat.

High availability systems, ensuring everything is always online. Being able to scale to handle spikes in concurrency. Infrastructure as code to facilitate all the previous, and fully automated continuous delivery pipelines to deploy new code releases. All interconnected by state-of-the-art networks, with advanced routing between countless VLANs.

Our jobs are cool, we make all of this happen! It is exciting and awesome! It feels so nice to have all these things that just work! However, when it comes to enterprise practices, we do tend to take our work environment into our homes.

The enterprise smart home syndrome is the art
of overcomplicating your Home Assistant setup.

Franck Nijhof

Looking at the Home Assistant Community forum, Discord Chat, the Home Assistant subreddit, and many other places, I see people bring their work into their homes. Using Kubernetes, Terraform, Ansible, and many others to set up Home Assistant in their homes. Building clusters of servers, sometimes even whole racks!

Don’t get me wrong, that is fine and all, some have a hobby at doing just this, if not: why? Really… why?

These are all solutions that add complexity and have their own issues and problems. I can’t wait to read about an mDNS repeater issue across VLANs caused by an NoT network, or weird DNS issues in their home about every other day. And let’s say it is running perfectly and never fails: You’ll still have to maintain it all! But why?

It’s your smart home system, not some SpaceX mainframe to coordinate a Mars landing.

Not just that, all those extra’s, generally result in needing more devices and resources, wasting more energy; which is bad for your wallet 💰 and bad for the planet 🌱.

Anyways, now you know what the enterprise smart home syndrome is. The solution lies within a really old and famous saying, which holds really well in this situation:

Keep it simple, stupid!

../Frenck

I don’t write a lot on my blog (unfortunately actually, I would really wish I was better at this); however, yesterday I had to do some maintenance on the server running this website…

This made me realize; that I added Google Tag Manager & Google Analytics to this darn thing and haven’t looked at it, since… I don’t think I ever looked at it all 😬

And the above image might actually be true, maybe not. But should I care? More importantly, do I care? Considering I haven’t looked at it all, I guess it is just a sign of how much I don’t care if anybody reads this or not. But I should care about your privacy. This data ends up with Google after all!

Removed it.

../Frenck