This story was first published on Sept. 16, 2022 in my weekly newsletter.
After participating in a panel on Matter at Silicon Labs’ Works With event on Tuesday and conversing with attendees throughout the event, I think the initial response to Matter is going to be tepid at best and downright aggravated at worst.
One person I spoke with, who declined to be named for this article, said they had the feeling that, given the delays and the constant influx of new members to the Connectivity Standards Alliance (CSA), which led to ballooning features requests, the CSA decided to just push something — anything — out. While I imagine the CSA’s thinking was probably along the lines of “Don’t let perfection be the enemy of the good,” based on my conversations, it may be that the 1.0 version of the standard may not be that good.
Part of the reason is because getting light bulbs to interoperate isn’t really the goal of people who want to buy into the smart home. Rather, they want the home to actually be intelligent, and that level of features will require several elements that Matter isn’t going to solve at the start.
So without any further ado, here are the five ways that Matter won’t work as expected. (For more on the panel, and a kinder view on Matter, check out Kevin’s story from Wednesday.)
1. Consumers will still need a lot of apps: One of the initial promises of Matter was that consumers would be able to add a device — like Amazon’s Echo — to their smart home controller but wouldn’t have to download a special app for every outlet or light switch they bring into the home. But at launch, and likely for a couple of years as the standard gets more robust, consumers will still need apps for anything beyond the basics, including installation. Even my panelists realized that this was the case.
2. Consumers won’t be able to control video cameras, appliances, video doorbells, or robotic vacuums: Those devices aren’t supported in the initial version of the spec, and won’t be for a while. How long, exactly? It’s not clear, but think version 2.0 as opposed to 1.2. Other specialty features of smart plugs, such as electricity monitoring, won’t initially be supported. Nor will lighting recipes. That is why you’ll need to download that device-specific app.
3. It will be hard to add a bunch of existing devices to more than one controller — or even one controller: Matter is going to make it easy to add a new light bulb to the smart home, but if a user tries to bring, say, more than 50 in one go, there are going to be issues, according to Jim Kitchen of Comcast. Comcast plans to update its gateways with Matter, and the existing devices on homeowners’ networks will continue to work. Mostly this will happen by keeping those devices on their existing Zigbee networks and simply doing any Matter communication in the gateway.
That said, I suspect someone like me, with a hodgepodge of devices and systems, is going to find the transition rough. And it will likely continue to be rough if I plan to run Google and SmartThings as my controllers. Such a multi-controller situation will lead to fights between the two that will need to be resolved. The good news is that the Matter working group is aware of the situation and Mark Tekippe from Samsung said the working group is continuing to work on ways to easily diagnose any issues.
4. Users will struggle with complicated routines: When it comes out, Matter is going to have the basics. And given all the concerns about getting devices from more than 500 member companies to play nicely across the millions of possible permutations, it’s likely that anything more than the basics will break. Again, while this situation will get better over time with subsequent versions of the standard, it’s still a bit of a punch to the face for folks who were hoping Matter would solve the interoperability challenges throughout the entire smart home.
5. Developers will still have to build for multiple ecosystems: One of the original pitches for Matter was that developers would no longer have to support multiple Works With-style programs for the big ecosystems. And indeed, Matter-certified devices will work with HomeKit, Amazon Alexa, SmartThings, Google, etc. Except again, those are only the basics, and as we’ve seen with Google and Amazon both embracing more in-depth experiences, developers will need to make sure they keep up to date with Google’s Home Device SDK and intelligence clusters or Amazon’s Ambient Home Development Kit. During the panel, Kevin Po of Google said those features are additional and that some companies will not choose to use them, but my hunch is that since consumers are jazzed more about truly smart homes as opposed to programming a million routines, most device makers will need to hop on board.
I’m still bullish on Matter, but it’s clear that the vision from 2019 when the protocol was first announced has shrunk in scope and in the meantime, the CSA has been caught out by the delays in its launch. What worries me is that consumers will find the experience of Matter devices at launch to be not at all worth the fanfare that preceded it. And they will likely be right.
A few more..
1) Matter is way too focused on “Alexa turn the light on” and they are ignoring the use cases of people using buttons and switches to control things.
2) There is no “replace device” button for when a device dies and you buy an identical one to replace it. You have to manually rebuild everything over all of the multi-admins.
3) Scene support got dropped from the Matter 1.0 release. You can build scenes inside a Homepod using Apple code (not Matter) but now the Homepod is single point of failure.
4) Thread meshes are great until they don’t work, and then they are almost impossible to debug unless you have a PhD in computer networking. Pray that yours does not stop working. Wifi doesn’t have the fancy self-healing, but it is much simpler to debug — you can either see the AP or your can’t.
5) There is no way to back up your Matter system and restore it. For example if you are making a bunch of changes and want to roll back.
Another big one…
Matter is going to be the first widely deployed IPv6 app in the home. This has the potential to expose a lot of untested and broken IPv6 support in home routers. For example Ive already encountered routers with broken IPv6 multicast support while testing Matter devices.
Matter is not supplying tools to test whether your network correctly functions with IPv6. Google/Apple/Amazon need to coordinate on building a test app that a home owner can run to verify that IPv6 is correctly working. This app has nothing to do with Matter, it is simply verifying that IPv6 works correctly in your house. If it doesn’t, you many need to upgrade router firmware or even replace your routers if no fixes are available.
In my case I had a ten year old Cisco-Linksys dual band router in AP mode where IPv6 multicast simply did not function, the packets went into the router and never came out again. My router was updated to the last firmware release from six years ago. Once I was able to identify the broken AP I bought a new one to replace it and Matter groups started working as expected.
A typical homeowner is never going to figure this out without some type of testing tool.
Excellent point. This is definitely a potential area of customer confusion and unhappiness. 😥
Surely the lowest hanging fruit are commercial, not resi, buildings? Nothing I’ve read about Matter would lead me to conclude that it’s suitable for typical commercial lighting/building automation applications, while qualified Bluetooth mesh evidently is.
Very good article!
I would add another disappointment: “supports matter” still allows for too many different variations.
For example, 3 different popular systems with 3 very different outcomes from a consumer’s point of view.
1) Philips Hue Bridge. Matter “bridge,” but the Hue app is not a matter “controller.” So you will be able to add the Hue bridge to many other platforms, pretty much like you can now with Alexa and HomeKit and SmartThings, but you will NOT be able to add a thermostat with the Matter logo to the Hue app.
2) Samsung SmartThings hub. The app is a Matter “controller,” but the hub is not a Matter “bridge.” You should be able to add pretty much any device with the Matter logo to the SmartThings app, but you will NOT be able to add the SmartThings hub to the Apple Home app or the IKEA Dirigera app.
3) ikea Dirigera. I haven’t been able to 100% confirm this yet since the device is still in ore-release, but it looks like this app will be a Matter controller and this hub may be a Matter bridge. If so, it would be full two way support. It can bring Matter-compatible devices into its own app and you can add its hub to other Matter controller apps. But again still waiting on details.
——-
So…3 popular hubs with the matter logo. One is one-way out. One is one-way in. And one is full two-way support.
Having to do that detailed research is not what most consumers expect from Matter.
Always a great article and points from Stacey and other commentators.
I think a lot of people are expecting far more from Matter than was promised. Matter seems like the halloween horse costume. There’s two different people trying to control the costume. There’s a device protocol, and an interoperability standard.
I agree, we will still have our various apps, but these apps should be able to connect the end devices to different matter controller fabrics.
1. Hue App allows the hue bridge to join connected devices via Matter to other systems. The biggest issue here i think will be the existing walls, ie, will the Hue bridge expose even third party devices to Matter Controllers including Apple? Right now, Home kit won’t acknowledge IKEA, or Cree bulbs for example that are paired to a Hue Bridge. Amazon, google, Smart things and others will. Will Hue/matter expose the Lutron Aurora Dimmers to Matter controllers?
2. Smart things is unfortunately taking its own walled garden cue from Apple. While it’s happily bridged devices for years, it no longer wishes to bridge to Matter. So Smart things users will likely see no real change
3. Ikea Dirigible is replacing the tradri gateway. Being the new ikea gateway for zigbee devices, will it be the Matter Bridge continuing to expose Zigbee products to other systems. Will it continue to support generic devices.
4. Cameras, vaccums, and other unsupported devices in the standard will continue to require a reasonable effort to read the box before you buy.
I expect Matter to provide the status quo with bit more interoperability for some time. We’ll get all our devices in our preferred ecosystems, and we’ll continue to build our appropriate scenes in the platforms we want to trigger them. If we want to say “Hey Siri, Go Packers” or “Alexa Go Packers” we’ll have to program each fabric how to set the colors on our many devices. Alexa will be able to control the Eve lights though and Apple will be able to control the Cree and Seedan bulbs.
Matter won’t eliminate everything but it should clean up some of the mess so we can party like its 1999!
Good Points. I think the most significant change that smartthings hub users will probably see is the ability to have a local connection to a lot more Wi-Fi devices that right now are a cloud to cloud integration. For example, a Meross smart plug has a local integration to Apple HomeKit, but requires the Internet for a cloud to cloud integration to smartthings. Smartthings is writing edge drivers for Matter devices that will run locally on the smarthings/Aeotec , so that will be a nice option if it actually gets delivered.
indeed. I think of the few iHome plugs I bought way back to see how a smart plug using wifi would work. (Should have stuck with Zwave and zigbee) Now they only work with Homekit and no longer with anything else.
I wonder if we would be able to control Matter Wifi with simple IP links. Bookmark https://officelightsIPv6address/on to turn the lights on from our windows desktop. Maybe Windows 23H2 will give us a Matter controller built in that we can struggle to join our devices too.
Honestly the smart home landscape is way too fragmented, in my opinion, to make me want to spend any more money on this stuff at this moment in time. And I say this as an enthusiast who has Insteon in every junction box, a network DVR camera system, smart locks, smart garage doors…the works. What you basically have is a situation where aside from advanced networking and tech savvy, out of the box these devices are completely dependent on their individual maker’s cloud services to work at all. And at times I have found support very lacking. Starting in 2017, I spent a lot of money building a smart home that was working amazing at one point.
Then support started to drop. My expensive Johnson Controls Glas thermostat suddenly was no longer “supported” by JC. So no more smart integrations for a great piece of hardware. Then my irrigation system, Rainmachine, suddenly began requiring a subscription if I want to tell Alexa to water the grass. Then Insteon’s cloud services suddenly disappeared for a few months with no notice or warning…and then came back, for now, with a subscription. I’m starting to realize a smart home isn’t that smart when it’s brains are entirely someone else’s computer. So for right now, I have created my own “Matter” on a Linux PC. It runs HomeSeer, a piece of software which gives me back total control of my Insteon system, completely cloud independent. It has local support for controlling thermostats, door locks, and everything else. And HomeSeer’s one integration with Alexa makes all those devices available through that one piece of brain software running on a computer I control. It also hosts a local RTSP bridge that lets me stream my DVR cameras directly to my Echo Show screens for a seamless front end integration. Bottom line is this: a house full of devices each dependent on the cloud services of someone else’s server…may not be all that future proof and can always be bricked or require a subscription the moment the manufacturer has a couple of bad quarterly earnings.
There is another thing to add here: Matter does also not require controllers to support all device types, or at least a minimal set of major ones. For example, there is no guarantee that a particular eco system supports bridge devices at all. Or even lights or thermostats, or sensors, or blinds, … so the promise that the Matter logo is sufficient to determine interoperability is moot. This is because while servers are thoroughly tested, controllers are not. For them it’s sufficient to support anything, not everything.
Clearly, Thread is not ready for mass deployment. Getting it to work reliably has proven to be a nightmare even for experts, let alone the average consumer. No idea how it scales to 50, 100 or 200 lights in a multicast group. Something our Zigbee stacks do very well. Also, self-powered energy-harvesting maintenance-free products are not going to be feasible with Thread in the foreseeable future.
But Matter does a great job in providing a universal protocol, the new lingua franca of IoT. And we see it as a great extension of Zigbee to the IP world. It complements Zigbee but certainly does not replace it. No doubt, Zigbee remains much more power-efficient and scalable than WiFi or Thread. And for audio/video and other higher band-width applications it is going to bring real advantages – once it supports these.
Before Insteon, there was Z-wave, a standard that then and still today is independent of any one manufacturer’s app. I’m sorry you got tricked into buying into it after X-10 was embraced by many different manufacturers including Insteon’s founders. I’m sorry you didn’t discover Z-wave back then which worked on HomeSeer back in 2001 and was already supported by a number of brands.
Yes the industry is “fragmented” amongst the Wifi embracing segments. But consider if Matter can slap a new industry standard protocol on top of all those wifi devices akin to Zwave and Zigbee where the devices no longer need a particular app to function.
Just today I was messing around with my iphone with iOS 16.1 and saw the new Matter Accessories page in general settings. I took my little ESP32 board from HomeAssistant’s Matter workshop and paired it easily with Apple Home. My first Matter device. I wonder if I can get some beta firmware with Matter support for my Echo’s and AlphabetGoogleNest Hub now.
Now I hope Home-Seer will adopt Matter and We can all turn full circle and automate like its 2002 again. I think I still have my HomeSeer v1 cd floating around in the basement.
Basically I feel it all comes down, once the technical base is running, to the marketing vision each company has. Amazon and Google have to get it right, Hue never has and starts by saying they’re only upgrading a hub, and no Thread. Ikea are outsiders and could give the surprise while people like Tuya haven’t said anything.