Jump to content

There is a train leaving and we're not on it. (Matter and Homekit)


Recommended Posts

Posted (edited)

This is both a request for a strategic investment and a bit of a venting of frustration. 

Home automation has been a hobby of mine for 20+ years now. I've used everything. I think all of us know it's an investment in hardware and time. My recent life has mostly been trying to keep my system relatively modern while allowing all of my investments to talk to one another under one rule base. What I am finding is that it's not really worth it to me anymore. My most recent change was to move to Ecobee thermostats away from an RS485 system controlled by 50+ rules in my ELKM1. Holey cow, the ecobee is excellent. It's like using an iphone after having been using a cd player for 15 years. My point is that there are some very nice hardware and software coming out, but I am tired of being trapped in an ecosystem that fails to understand its customers needs - and failing to grow to add more customer segments.    There have been many ecosystems over time that have attempted to take seed, but none that have thrived. You have to ask yourself why that is the case. 

For a long time my ISY was this hope. It is my Zwave + Elk and later + SmartSpeaker (though super clunky) hope. I am starting to realize, now having bought a Polisy, setup home assistant, and homebridge, and played with it for a few weeks, that I'm not on the right train. The Universal Devices train is not converging with where smart homes are going. As much as you might hate apple and google, they are going to own (or decide) the ecosystem. I am actually happy they will win, because from what I've seen - it JUST WORKS and is democratized for anyone to use, including my wife and kids. 2021 and 2022 are going to be inflection points for hardware moving to new standards to access the explosion of users from non technical market segments. 

I am an engineer and a geek at heart. I can make my ISY do what I need it to do with enough Pi's and ductape. But it's no longer worth it to me. For example, I was hoping my Polisy would be able to connect to my Aqualink RS and tell me when my chemical tanks were empty - there is a node for iAqualink. BUT you know what, the node is a toy. It does nothing useful. It's because the developer did not attempt to work with Jandy to get access to an API, instead it's a hack unsanctioned by the manufacturer. Most other nodes I have running are the same. 

I can't help but wishing that Universal would attempt to converge with where automation is going and focus on being a bridge to this new world. I think they could do an amazing job working with HomeKit. They are very good developers and they have excellent customer support, not to mention they have fostered a passionate community.  But alas, the options that connect to homekit are rubblish (I'm sorry to say this, as I know there are passionate individual developers out there who are trying hard). The Homekit optinos are not embedded-system reliable like I have been used to (like ELK and ISY), and it's a lot of work just to make a single device useful.

I'd like to see a way to put devices (either directly, or via a virtual wrapper) into homekit natively from ISY. I don't want to be forking github projects to fix bugs or revitalize something that has gone dead (like the ISY homebridge connector).

I see Universal's goal to create an ecosystem that allows developers to make money. That's interesting, but there are not enough of us geeks to make that enough of a financial proposition for developers in that ecosystem such that they create long lasting and robust solutions. They need to get more users, and I think the only way that happens is if Universal becomes the bridge from the old world to the new world. There are COUNTLESS reddit conversations about getting elk or zwave into homekit (as an example) and zero mention of ISY. 

So, to repeat my ask. I would like to see Universal become a bridge to HomeKit (and Matter) as a priority and native ability. I don't expect that to happen. So I will be gradually migrating to devices that support that standard, and eventually, like I've done by removing my ELKs control over my thermostats, remove my need for my ISY to manage my doorlocks, lights, and sensors.

If you read that entire rant, thank-you (and I'm sorry for the rant). 

Edited by jlegault
  • Like 2
Posted (edited)
8 hours ago, jlegault said:

This is both a request for a strategic investment and a bit of a venting of frustration. 

Home automation has been a hobby of mine for 20+ years now. I've used everything. I think all of us know it's an investment in hardware and time. My recent life has mostly been trying to keep my system relatively modern while allowing all of my investments to talk to one another under one rule base. What I am finding is that it's not really worth it to me anymore. My most recent change was to move to Ecobee thermostats away from an RS485 system controlled by 50+ rules in my ELKM1. Holey cow, the ecobee is excellent. It's like using an iphone after having been using a cd player for 15 years. My point is that there are some very nice hardware and software coming out, but I am tired of being trapped in an ecosystem that fails to understand its customers needs - and failing to grow to add more customer segments.    There have been many ecosystems over time that have attempted to take seed, but none that have thrived. You have to ask yourself why that is the case. 

For a long time my ISY was this hope. It is my Zwave + Elk and later + SmartSpeaker (though super clunky) hope. I am starting to realize, now having bought a Polisy, setup home assistant, and homebridge, and played with it for a few weeks, that I'm not on the right train. The Universal Devices train is not converging with where smart homes are going. As much as you might hate apple and google, they are going to own (or decide) the ecosystem. I am actually happy they will win, because from what I've seen - it JUST WORKS and is democratized for anyone to use, including my wife and kids. 2021 and 2022 are going to be inflection points for hardware moving to new standards to access the explosion of users from non technical market segments. 

I am an engineer and a geek at heart. I can make my ISY do what I need it to do with enough Pi's and ductape. But it's no longer worth it to me. For example, I was hoping my Polisy would be able to connect to my Aqualink RS and tell me when my chemical tanks were empty - there is a node for iAqualink. BUT you know what, the node is a toy. It does nothing useful. It's because the developer did not attempt to work with Jandy to get access to an API, instead it's a hack unsanctioned by the manufacturer. Most other nodes I have running are the same. 

I can't help but wishing that Universal would attempt to converge with where automation is going and focus on being a bridge to this new world. I think they could do an amazing job working with HomeKit. They are very good developers and they have excellent customer support, not to mention they have fostered a passionate community.  But alas, the options that connect to homekit are rubblish (I'm sorry to say this, as I know there are passionate individual developers out there who are trying hard). The Homekit optinos are not embedded-system reliable like I have been used to (like ELK and ISY), and it's a lot of work just to make a single device useful.

I'd like to see a way to put devices (either directly, or via a virtual wrapper) into homekit natively from ISY. I don't want to be forking github projects to fix bugs or revitalize something that has gone dead (like the ISY homebridge connector).

I see Universal's goal to create an ecosystem that allows developers to make money. That's interesting, but there are not enough of us geeks to make that enough of a financial proposition for developers in that ecosystem such that they create long lasting and robust solutions. They need to get more users, and I think the only way that happens is if Universal becomes the bridge from the old world to the new world. There are COUNTLESS reddit conversations about getting elk or zwave into homekit (as an example) and zero mention of ISY. 

So, to repeat my ask. I would like to see Universal become a bridge to HomeKit (and Matter) as a priority and native ability. I don't expect that to happen. So I will be gradually migrating to devices that support that standard, and eventually, like I've done by removing my ELKs control over my thermostats, remove my need for my ISY to manage my doorlocks, lights, and sensors.

If you read that entire rant, thank-you (and I'm sorry for the rant). 

I've debated on whether I should post a response to this or not. So you're complaining about UDI not dumbing down their system in order to integrate with lessor systems? If Homekit, google, Amazon, or Matter matters (pun intended), why do you need the Isy? 

The Isy can already be used with Amazon and Google. Maybe a few people will benefit from Homekit (but judging by posts not enough to justify the costs). Outside of Homekit, what is it that you can do with matter that has you so frustrated? I won't wait for a response cause the answer is nothing at all since Matter is irrelevant right now with it being delayed until 2022.

With all the promises of integrated systems over the years, why would UDI jump at the very beginning? Look at all the changes homekit has been through and still only popular with a select number of people! They would've wasted so much time and money on something that a small percentage really cares about. The same goes for matter. If they focused on it right now, where would they be? Other more pressing and customer desired needs put on the back burner for something that's been pushed off for a minimum of 6 months away...and potentially longer. 

I think UDI does understand their customer base and realize they will NOT please everyone. As humans, our needs, wants, and desires change quickly. They'd go broke and out of business if they tried to keep up with our whims..... especially with trying to keep up with every latest and greatest system that comes to market. UDI customers want automation and full control over their system. Something you won't get with these other things. They will probably support more than the isy does but not at the level the Isy does. For UDI customers, this is what matters most.  

Unfortunately for you, it sounds like your needs, wants, and desires, no longer align with UDI which is ok. We outgrow and move on from things all the time whether they're phones, cars, houses, or people. 

Edited by lilyoyo1
  • Like 5
Posted
I've debated on whether I should post a response to this or not. So you're complaining about UDI not dumbing down their system in order to integrate with lessor systems? If Homekit, google, Amazon, or Matter matters (pun intended), why do you need the Isy? 
The Isy can already be used with Amazon and Google. Maybe a few people will benefit from Homekit (but judging by posts not enough to justify the costs). Outside of Homekit, what is it that you can do with matter that has you so frustrated? I won't wait for a response cause the answer is nothing at all since Matter is irrelevant right now with it being delayed until 2022.
With all the promises of integrated systems over the years, why would UDI jump at the very beginning? Look at all the changes homekit has been through and still only popular with a select number of people! They would've wasted so much time and money on something that a small percentage really cares about. The same goes for matter. If they focused on it right now, where would they be? Other more pressing and customer desired needs put on the back burner for something that's been pushed off for a minimum of 6 months away...and potentially longer. 
I think UDI does understand their customer base and realize they will NOT please everyone. As humans, our needs, wants, and desires change quickly. They'd go broke and out of business if they tried to keep up with our whims..... especially with trying to keep up with every latest and greatest system that comes to market. UDI customers want automation and full control over their system. Something you won't get with these other things. They will probably support more than the isy does but not at the level the Isy does. For UDI customers, this is what matters most.  
Unfortunately for you, it sounds like your needs, wants, and desires, no longer align with UDI which is ok. We outgrow and move on from things all the time whether they're phones, cars, houses, or people. 

Absolutely hate agreeing with you!

I’ll add, the major cornerstone of the ISY Series Controller is Local First vs Cloud First.

As Cloud First products and services began rolling out and the public at large started adopting the same. UDI slowly if *begrudgingly* rolled out their integration for IFTTT, Amazon Echo, Google, etc.

Unlike a lot of other companies the foundation of the ISY Series Controller for local first hasn’t changed. It has simply allowed more flexibility and integration of using cloud first services via ISY Portal / Polyglot Cloud.

Given they are a small but agile company they need to focus their limited resources on products / services that balance profits, support, and overall costs.

The world has always had people who blindly follow X vs Y brands. All the while losing any common sense and having the ability to stand back and be objective!

Apple has spent decades honing their brand while convincing the less than bright iSheep everything they make or sell is top shelf?!?

The reality is this company (Apple) is like Google, Microsoft, name whoever . . .

Profit before people . . .

One only needs to search for every failed product and services they have released. Look up the millions of computer systems that have known design flaws - yet continued to sell! Only because Americans love to sue, Apple was found guilty hundreds of times for selling faulty hardware!

If it wasn’t sh^tty hardware it was sticking it to the consumer to ring out yet another wooden nickel!

Think early battery failures after a iOS update. Think 4 generations of Mac Books with defective wiring, video, and LCD issues. Think bendy phones, and exploding batteries (well not on the same order as Samsh^t battery explosions)

Think stealing IP from other companies and found liable and guilty by the courts and international trade commission.

The most famous trying to steal the round face clock in their iOS and thinking a company who created the same for more than 100 years was just going to let it go?!?

This is Crapple, and their insistence in controlling every thing in the market! HomeKit like the many other companies such as Sony have this innate ability to push out a so called standard in hopes of controlling the same!

Think Sony Beta, MD, every stupid proprietary plug or connector they ever came out with just to screw the general public. Think Apple, stupid aszz 99999 pin connector for iPod later to push stupid once again with lightning, thunder port etc.

Only to come full circle to use and follow standards used by EVERYONE.

Think USB Type C . . .

Crapple decided and thought we’ll release a speaker for five times amount and the iSheep will gobble it up. Guess them dropping both versions do to poor sales only to try another 3rd iteration of sh^t?!?

I mean if one looks at the comical and stupid concept of a F keys on their Mac Pro?!? Oh geeeese, that was killed off might quick eh? Then again it only took them 3 generation to fix those cheap and unreliable butterfly keyboards!

Then again this is the same company that convinced the entire world to adopt a product that has no method to upgrade the memory, hard drive, video, battery on everything!

HomeKit is so great - Not

Like all of their wares it was created for the less than bright population. I mean does anyone remember Steve Jobs insistence that wifi / cellular problems were caused by every customer holding the freaking device wrong??!?

The sheer hubris and stupidity of that guy!

So no, HomeKit isn’t the market leader Apple wants it to be. Companies integrate with it in hopes of driving sales not because it’s so awesome. If that wasn’t true how come Amazon / Google continues to kick Apple Siri in terms of adoption, use, and shear integration & accuracy?!?
  • Like 3
Posted

@jlegault,

I completely understand. We have had the same discussions internally and it's one of the main reasons why most of our revenue comes from utility contracts and not home automation. I am not sure whether or not you have also noticed that most home automation hubs are slowly disappearing simply because of simple command/control solutions provided by HomeKit, Alexa, and Google Home. And, except for Zigbee/Z-Wave/UPB/X10/INSTEON, you don't really need a local hub. X10 is dead, UPB can be ignored, INSTEON is dying, and Z-Wave is already promoting ZIP (IP based Z-Wave) and Zigbee has become Matter. So, in the brave new world of the future - and excluding solutions for the rich and the famous - if all you seek is command/control and simple rules, then I don't see any point for an automation device including polisy. 

With the aforementioned out of the way, we have spent so much time trying to figure out how to support HomeKit. It's just way too expensive (>$100K and increasing by the day) with an unknown ROI. As mentioned above, most things are becoming IP and most things will support HomeKit. So, what's the point? 

Anyway, I am sorry that you feel this way but, unfortunately, that's the reality we live in now.

With kind regards,
Michel

  • Like 2
  • Thanks 1
Posted

As someone with 50+ switches and another 20 in the cart while they're on sale, and just bought the last policy online, I question my judgment on this. I feel like I'm too far in now to bail and these are expensive ways to buy out some more time and keep my current set up functional.  I like my local control and rules. I bought a rain machine recently. that is my favorite thing right now. I have a pi with Homebridge, but its messy.

For Micheal's comment on simple rules, I hope UD stays in the game and finds a way to keep their cult followers(myself included) pacified. I didn't wake up planning to stick another 20 devices on my network while I simultaneously wonder what will happen if the new usb plm I'm buying dies?  I know there are more like me out there. If there is a way to tap into our interest and still be viable financially then I hope you can do it.  I hope the pendulum swings in our favor. I know I'm invested in this. I know I don't like the insecurity of  wifi let alone cloud based operations or the lag time.

Sorry for rambling. just wondering if I should be spending the money on something with a more secure future. Do I need a usb plm for Polisy as soon as my serial one dies?

Posted

@noobmuch,

Although I can never make future promises, I can say that we are financially secure and will continue enhancing Polisy and making it easier to average user at the same time. HomeKit is just something else and, at the moment, we just cannot justify the extra cost. 

You can use either the serial port or the USB. But, if you think your PLM is going to die, it's best to get 2413U while they last. No one know when they will be sunset as well.

With kind regards,
Michel

  • Like 4
Posted

Micheal,

 

no matter what happens thanks for your service, timely and actually useful responses and all of the other enthusiasts who make this a great little community.

the app is a welcome addition and negates the need for HomeKit integration imho. I look forward to using policy to program my switches to gain more control over some receivers without beating my head into the ground with network resources etc.

  • Like 1
Posted
2 hours ago, Michel Kohanim said:

But, if you think your PLM is going to die, it's best to get 2413U while they last. No one know when they will be sunset as well.

Michel,

I don't think it is a matter of if someone's serial PLM will die but when.  I've only had one fail but others have had worse luck.  I do have one 2413S in reserve and wish I could find another for less than the current $150 used price on ebay.

Should I buy an ISY for my spare parts inventory?  I think you have said previously UDI intends to continue making them for an extended period of time as a parallel control platform.  Hopefully I remember that correctly and the company position hasn't changed.

It is good advice for people with a signficant investment to pick up a fairly inexpensive 2413U to buy a little insurance for their investment.  It might not be needed, serial PLM's may be back, but if one has invested with a bunch of time, devices and logic, it makes sense to put a USB modem on their shelf just in case Smartlabs leaves them in a tough spot.   

Posted
31 minutes ago, DAlter01 said:

Michel,

I don't think it is a matter of if someone's serial PLM will die but when.  I've only had one fail but others have had worse luck.  I do have one 2413S in reserve and wish I could find another for less than the current $150 used price on ebay.

Should I buy an ISY for my spare parts inventory?  I think you have said previously UDI intends to continue making them for an extended period of time as a parallel control platform.  Hopefully I remember that correctly and the company position hasn't changed.

It is good advice for people with a signficant investment to pick up a fairly inexpensive 2413U to buy a little insurance for their investment.  It might not be needed, serial PLM's may be back, but if one has invested with a bunch of time, devices and logic, it makes sense to put a USB modem on their shelf just in case Smartlabs leaves them in a tough spot.   

Insteon has zero intentions of discontinuing the PLM. As with most companies, due to the chip shortage, they've had to make difficult decisions when it comes to their product line in order to meet demand ( still can't do that).  I'm an ISY user myself and it sucks (though i have more than enough backups), but reality is they aren't the only one going through this. One only has to look at new cars, posts, etc. to see how this effects everyone. Under normal circumstances, we wouldn't even be talking about this but this isn't normal times. The PLM is still in production but as of now it's on hold. It'll be back eventually and at full stock. Unfortunately we're just in trying times

  • Like 1
  • Sad 1
Posted
1 minute ago, lilyoyo1 said:

Insteon has zero intentions of discontinuing the PLM. As with most companies, due to the chip shortage, they've had to make difficult decisions when it comes to their product line in order to meet demand ( still can't do that).  I'm an ISY user myself and it sucks (though i have more than enough backups), but reality is they aren't the only one going through this. One only has to look at new cars, posts, etc. to see how this effects everyone. Under normal circumstances, we wouldn't even be talking about this but this isn't normal times. The PLM is still in production but as of now it's on hold. It'll be back eventually and at full stock. Unfortunately we're just in trying times

Considering your "inside" tknowledge I'll take your word for it.  I had a 2413U in my Smarthome cart at 30% off when I got your post.  Thanks for saving me $56++.  I will, however, probably pick up another 2413S if/when they come back in stock.  

  • Like 1
Posted
On 8/27/2021 at 7:20 PM, DAlter01 said:

Should I buy an ISY for my spare parts inventory?  I think you have said previously UDI intends to continue making them for an extended period of time as a parallel control platform.  Hopefully I remember that correctly and the company position hasn't changed.

@DAlter01,

ISY994 will eventually be sunset and replaced by Polisy. Not so sure you would want to keep a spare ISY since a) defect rates/malfunction even after 10 years is less than 0.01% and, out of those, 95% are either SD Card or Power supply. 4% due to firestorm. b) we will eventually stop supporting ISY994 firmware when Polisy is robust enough. The main issue is that I do not know the ETA. 

With kind regards,
Michel

  • Like 6
Posted

While I don't share the OP's concern about Homekit (why would anybody sink money into that walled garden?) I am interested in the parts of the discussin that suggest IP will eventually rise to be the ruling IoT protocol. I am having some trouble picuring that future worls and would am curious to know more about how folks think that would play out. For example:

What is the purpose of ZIP? What does it do that Z-Wave does not (apart from add congestion to a LAN?)

If IoT is going to live entirely in the TCP/IP world then home routers are going to need to be beefed up a lot... like faster than many current small commercial routers. Is there any evidence that this is happening in the home router marketplace? I haven't seen anything even approaching what would need to happen there.

Wi-Fi frequencies are high to enhance data throughput while solid IoT protocols use lower frequencies to enhance penetration through obstacles and building materials. Isn't using Wi-Fi for IoT going to lead to substandard IoT products because, you know, physics. 

And how practical is Wi-Fi for battery devices?

And what about 2.4GHz congestion? If it already a problem (and it is) how is dumping another whole class of devices into the mix going to work? Is the FCC going to open up more 2.4GHz bandwidth to support IoT expansion? Are IoT devices going to start using 5GHz for even poorer penetration and shorter battery life?

And speakin of expansion what about address space? If the new stand jumps say from 255 per subnet to 1024 is that going to work well in a home environment? Is there no practical limit to how large you should make a broadcast domain?

And what benefits does Matter bring really. I have seen vaguely worded benefits attributed to this protocol but nothing very specific that would make me say "I need that".

I understand this new train is leaving the station but I can't for the life of me see that it is going anywhere.

Posted
2 hours ago, Michel Kohanim said:

@DAlter01,

ISY994 will eventually be sunset and replaced by Polisy. Not so sure you would want to keep a spare ISY since a) defect rates/malfunction even after 10 years is less than 0.01% and, out of those, 95% are either SD Card or Power supply. 4% due to firestorm. b) we will eventually stop supporting ISY994 firmware when Polisy is robust enough. The main issue is that I do not know the ETA. 

With kind regards,
Michel

@Michel Kohanim,

That is an amazingly low failure rate on the ISY.  Your firm is to be congratulated for that.  Considering that low failure rate and since the Polisy will follow behind the ISY994 as its replacement in the near/middle future, I agree that having a spare ISY994 isn't necessary for my one install.   Thank you for some clarity on the future.  I do see why you suggest picking up a 2413U should be considered.

Posted
17 minutes ago, upstatemike said:

While I don't share the OP's concern about Homekit (why would anybody sink money into that walled garden?) I am interested in the parts of the discussin that suggest IP will eventually rise to be the ruling IoT protocol. I am having some trouble picuring that future worls and would am curious to know more about how folks think that would play out. For example:

What is the purpose of ZIP? What does it do that Z-Wave does not (apart from add congestion to a LAN?)

If IoT is going to live entirely in the TCP/IP world then home routers are going to need to be beefed up a lot... like faster than many current small commercial routers. Is there any evidence that this is happening in the home router marketplace? I haven't seen anything even approaching what would need to happen there.

Wi-Fi frequencies are high to enhance data throughput while solid IoT protocols use lower frequencies to enhance penetration through obstacles and building materials. Isn't using Wi-Fi for IoT going to lead to substandard IoT products because, you know, physics. 

And how practical is Wi-Fi for battery devices?

And what about 2.4GHz congestion? If it already a problem (and it is) how is dumping another whole class of devices into the mix going to work? Is the FCC going to open up more 2.4GHz bandwidth to support IoT expansion? Are IoT devices going to start using 5GHz for even poorer penetration and shorter battery life?

And speakin of expansion what about address space? If the new stand jumps say from 255 per subnet to 1024 is that going to work well in a home environment? Is there no practical limit to how large you should make a broadcast domain?

And what benefits does Matter bring really. I have seen vaguely worded benefits attributed to this protocol but nothing very specific that would make me say "I need that".

I understand this new train is leaving the station but I can't for the life of me see that it is going anywhere.

I think you are looking at things the wrong way. Michel Is correct in what he says about IP and what's happening out there in this field.

If you notice he also mentions control a few times in reference to those specific solutions. Why? Because that's what the masses want. Overall, they aren't doing what we do. Whether you're doing this for money or a hobby, most people are not interested in whole home solutions. They simply want to turn a couple of lights on and off or say look what I can do. 

Because of what those people end up doing in their homes, bandwidth isn't necessarily an issue. Home routers will never be enough no matter what anyone does. The same things that plagues smartphone batteries is what causes problems for routers. No matter how big you make it, people will demand more from it making progress seem at a standstill. 

Today's routers for the most part are beasts compared to 10 years ago. When you look at the cheapest routers now compared to high end routers from 2011, there's no comparison. Today's routers would run circles around high end from back then.

Why are they so limited now then!? It's because we have 4-8k TV's, laptops, high powered smartphones, and tablets connected to them. Nevermind 4k Firestick, apple tvs, Riki's, and Google cast devices. That's even before you get to always connected voice assistants with screens, and connected smart clocks, fans, air purifiers, and everything else. 

No matter what gets put into routers and network equipment, device makers will simply add more to their products that will take advantage of that increased power.

I say this alot but we are the minority. Yes, hobbyists, installers, geeks, etc. started this whole craze but we're the minority now. There will be systems like the Isy, homeseer, HA, and others for those with more discerning needs. For most people however, the market is catering to them

Posted

The frustrating thing is that every incremental improvement, whether it is routers, FCC allocating bandwidth, protocol scalability, or whatever, the new thing on the horizon always looks like it is meant to address the problems of today instead of the problems that are projected to exist when the thing finally becomes available. The main reason I am so dismissive of Matter is because I expect it to be too little too late and therefore obsolete upon arrival.

I know you feel people don't really want to do that much with automation yet there are plenty of requests for faster response times from IFTTT, and/or logic in Alexa routines, or a larger family of products from their chosen technology platform so they can stick a handy keypad next to the door to shut off all the lights on their way out or stick smart bulb in that old pull chain ficture that can't be wired with a switch.

I actually agree with the fact that the average person does not want to deal with complex hubs or controllers to mix technologies because no one protocol does everything but those same people have a pretty low tolerance when they can't do basic lighting control because the Internet is down. I just don't see any way to solve this using IP unless a simple way is found to make IP stuff work locally for most basic operations and I don't see that happening without addressing some of the conerns I raised above.

I will also note that I live in Upstate New York with few Internet options. Even with two ISPs and a dual WAN router I would never ever ever consider having web based automation as my primary means of control. The idea that you can do everything with just Alexa and a few other web services is not just a problem philosophically for me but rather something I have tried and proven to be unreliable through personal experience. This is why I struggle to see a pure IP based future or any realistic path to one.

Posted (edited)
52 minutes ago, upstatemike said:

The frustrating thing is that every incremental improvement, whether it is routers, FCC allocating bandwidth, protocol scalability, or whatever, the new thing on the horizon always looks like it is meant to address the problems of today instead of the problems that are projected to exist when the thing finally becomes available. The main reason I am so dismissive of Matter is because I expect it to be too little too late and therefore obsolete upon arrival.

I know you feel people don't really want to do that much with automation yet there are plenty of requests for faster response times from IFTTT, and/or logic in Alexa routines, or a larger family of products from their chosen technology platform so they can stick a handy keypad next to the door to shut off all the lights on their way out or stick smart bulb in that old pull chain ficture that can't be wired with a switch.

I actually agree with the fact that the average person does not want to deal with complex hubs or controllers to mix technologies because no one protocol does everything but those same people have a pretty low tolerance when they can't do basic lighting control because the Internet is down. I just don't see any way to solve this using IP unless a simple way is found to make IP stuff work locally for most basic operations and I don't see that happening without addressing some of the conerns I raised above.

I will also note that I live in Upstate New York with few Internet options. Even with two ISPs and a dual WAN router I would never ever ever consider having web based automation as my primary means of control. The idea that you can do everything with just Alexa and a few other web services is not just a problem philosophically for me but rather something I have tried and proven to be unreliable through personal experience. This is why I struggle to see a pure IP based future or any realistic path to one.

As I always say, forum members are not the avg person these companies are catering to. You trying and failing is because you want more than what those systems can provide. I'm with you, i can't do those systems myself. However, I know plenty of people running their homes with them with great success and are happy as can be with them. It's hard to fathom something that doesn't exist in your world. 

IFTTT exists for a subsection of general users and the way it works makes automations easy for those people. More advanced users may use it for some stuff but that's on a limited basis compared to general users. 

With the amount of homes I've been in over the years, i haven't been in many where a person wished for keypad anywhere. Most are good with Alexa or Google doing the work for them via voice. That's not too say people can't and don't grow being what those systems offer. But then they are no longer that type of consumer. Even with myself, i have an away button but the house will go into away mode on it's own if completely empty long enough. Hitting a button or using voice only makes it happen immediately. I've found those wanting keypads generally have larger systems and bought me in because they were in over their heads. 

People have low tolerance early on. They'll either abandon it completely or learn to live with the short comings. Overall though, if their internet is down that much, they're more angry over lack of tv and surfing than they are about their lights. 

Incremental updates happen because r&d is expensive and people are finicky. Do you build for tomorrow, charge accordingly, only to watch people buy your competitors products because it's half the price of yours. Not only that, by the time tomorrow comes, your product is still obsolete. Look at Ethernet cables. If i put fiber optic throughout your home today, how long will it be before you can take advantage of what it offers? Maybe 15-20 years from now (if then)!? In 20 years, where will internet be? What's going to be better than today's fiber optic standard? Yes, your house is future proofed in many ways but at the same time, by the time that future gets here, it's obsolete. This shows with high end routers. A 500 dollar router from 5 years ago still works great today. But how does it fare compared to today's 500 dollar router?

Technology moves forward every day. Developing for tomorrow today is a fools errand when it comes to product development. It's one thing to have a long term plan and work towards that. It's another to do it all at once

Edited by lilyoyo1
  • Like 3
Posted

I am currently running about 45 HA devices on WiFi. I really like it's quick response and reliability.

But... you have to have the routers and infrastructure to support it. I don't use hardly any cloud services with my system except for Alexa voice stuff via ISY Portal, another cloud service.

I use a WiFi6 mesh system but in conjunction with lilyoyo1's comments, the system and requirements will always keep growing for users like us. The latest band after WiFi6 now is WiFi6E, for more bandwidth and speed, but the current router support is running about $1500 per AP. It's like keeping up with the Jones' but more like keeping ahead of the Jones'....for this year. :(

Posted

I don't use WiFi unless I have no other option (Amazon Echos, Lifx Bulbs, Nest Thermostats, etc) but somehow nearly half the 200 reserved IP addresses on my LAN are wireless. What would happen if I converted all my Insteon and Z-Wave and Hue devices to WiFi? Actually nothing would be happening because my router would melt and my 10 Access Points would explode. 

Clearly you could never automate a home entirely using WiFi with current technology. So we have one standard for the average user and then you have to throw it all away and go to something else if you want to scale up even a little bit. Still waiting for the one true standard to come along that works for DIY and Power User and High End Integrator alike.

Posted
10 hours ago, upstatemike said:

 

Clearly you could never automate a home entirely using WiFi with current technology. So we have one standard for the average user and then you have to throw it all away and go to something else if you want to scale up even a little bit. Still waiting for the one true standard to come along that works for DIY and Power User and High End Integrator alike.

I don't think alot of avg users are really looking to take things to scale. They're looking at solutions for specific tasks such as a bedroom or livingroom lamp. Maybe even both. There may be those looking for larger solutions which means they'll probably do more research and go with scalable solutions such as insteon or zwave. Those with more money than time will generally pay someone from the start regardless of solution. 

Unfortunately, I don't think you will ever see a one size fits all approach to automation. The needs are too vastly different. Nowhere in life do you see it that way. If you think about it, hobbyists are tinkerer's. They want to dig deep into what they are into to get the most out of their setup. The avg person doesn't want to spend a lot of time researching, reading, etc. They just want stuff to work. Those differences alone makes it hard to come up with a singular system that keeps both crowds happy. 

An installer doing high end homes doesn't want either situations. After taking the time to learn the ins and out of a system, why would an installer want to make minimum profit on a job that can carry on for weeks and months; if not longer? Would their clients really want to walk into their local best buy or home Depot and see the same devices on sale for 29.99 that you just sold him for 250 a piece? Even if they don't see them in stores, do you really think an installer would want them to be at a friend's house who did their home for 2k that he charged 25k for the same stuff. 

Even with that, as an installer, there isn't much of a markup on low end devices to warrant the time and effort that goes into a high end home. The math simply doesn't add up. Let's say an insteon or zwave switch retails for 50 bucks. Add a 20% markup on that number and now youve made 10 bucks. If you have a dealer account, you'll make maybe 30-40 bucks per device (if you markup from retail cost). Multiply that by 100 devices and that 4k you've made. Using polisy and marking that up, you may be able to get away with making an extra 125-200 bucks on that. Once you add on design, installation, programming, and admin fees, the total job might make you 10-12k.

Compare that to a high end system where the retail price varies online but generally runs 200-250 per switch. Yet, depending on supplier, you'll pay maybe half that. Depending on system, Controller costs can go into the thousands of dollars. Just on hardware alone, you've made the same as a low cost installation in profit alone before even including other charges such as installation, programming, and other fees. 

The way people are, they will search online to see how much products costs. This information is readily available with low cost systems which greatly limits you to how much you're able to charge which limits your income on the flip side. These reasons and more adds to why one common system for each type wouldn't work

  • Like 1
Posted

I'm keenly interested in the new Matter devices and protocols. However, on reading through this forum, I see that few people realize that Matter devices don't run on WiFi. Instead, they have their own thread based mesh network which accesses your LAN through the thread network's border router. Much of the thread traffic might not appear on your LAN. If it does appear on your LAN, it might only be one group of matter devices in one thread subnet talking to another group of matter devices in another thread subnet using communication between the 2 thread subnetworks using their border routers.

Matter devices have IPv6 addresses. There are only about 4 billion possible IPv4 addresses, which is only enough to provide 1  IP address for half the population on the planet. However, IPv6 provides about 32000 IP addresses for every human. This means that Matter devices can communicate through your router to the WAN without network address translation. They need to be extremely secure, and the Matter spec allows the firmware to be updated remotely and automatically as soon as a security exploit is discovered.

Matter is currently being implemented only for devices and situations with low data rates. Zigbee is now part of Matter, so it will turn lights on and off. Matter can control door locks, thermostats and control TV's. It is being used to synchronize Google's Nest Audio devices and in this situation is works the same way Sonos does when Sonos synchronizes its speakers using SonosNet.

Matter is also in networking equipment such as mesh routers and switches. If you are watching video from the cloud in several locations in your home, on a normal LAN all the packets for all your video streams would be visible everywhere on your LAN. Matter solves this problem. I believe it can manage your mesh routers and switches to utilize your LAN efficiently so that packets are sent where they are needed and not everywhere on your LAN.

 Matter devices have standardized APIs that will be supported by about 100 companies that support Matter. The list of companies includes Google, Apple, Samsung and many others. People can go to Best Buy and just look for the Matter logo on a product and know that it will work with almost everything.

Now to my situation. I have 2 Ethernet LANs with mesh routers. On these I have about 120 devices including 13 cameras, 50 Google devices, over 20 Sonos speakers and much more. So I am really happy to have my other 250 devices such as light switches, HVAC, irrigation, etc on INSTEON. I couldn't add these 250 devices to my LANs that already have 120 devices.

However, INSTEON has its problems, such as keeping RF noise out of my electrical wiring. Ultimately, I would like to replace all my INSTEON devices with Matter devices. Moreover, Matter will support door locks, which I can currently not controlling with my ISY.

I've thought about how the ISY could work with Matter and here are 3 possibilities.

(1) The ISY supports Zigbee, and Zigbee is now part of Matter. The ISY has a Zigbee radio so it could communicate with Matter devices using their IPv6 addresses. This would almost certainly require a firmware update in the ISY's Zigbee radio. I don't use Zigbee and don't know much about it, so I'm wondering if the Zigbee radio chip can be updated in this way.

(2) People who use Home Assistant and Open HAB are very keen on connecting to Matter devices. Suppose they succeed in getting software running on their Linux systems that can talk to Matter devices using their IPv6 addresses. Then it should be possible to put the same software on the POLISY and connect the matter devices to the virtual node server framework. Then a Matter node with your Matter devices should appear on the ISY.

(3) We (or rather I) don't know if interacting with Matter devices using the IPv6 address has all the capabilities that would be available if the POLISY could talk to Matter devices directly using Thread.  I think there are going to be devices with a thread radio that could be used to build a POLISY. Maybe there will be a RaspPi with a tread radio. I think people using HA or OpenHAB will probably start using these before we will. However, whatever software is developed, it should run on the POLISY and we ISY fans should be able to interact with Matter devices.

In my case, as INSTEON slowly dies, I'd like to swap all my INSTEON devices for Matter devices and continue to run my smart home with my ISY.

The questions I'm most interested in having an answer to in this forum are (1) is UDI thinking about whether or how the Zigbee radio can work with Matter, and (2) are any of you POLISY developers interested in getting the Matter software running on the POLISY.

Posted (edited)
6 hours ago, bruceyeg said:

I'm keenly interested in the new Matter devices and protocols. However, on reading through this forum, I see that few people realize that Matter devices don't run on WiFi. Instead, they have their own thread based mesh network which accesses your LAN through the thread network's border router. Much of the thread traffic might not appear on your LAN. If it does appear on your LAN, it might only be one group of matter devices in one thread subnet talking to another group of matter devices in another thread subnet using communication between the 2 thread subnetworks using their border routers.

Matter devices have IPv6 addresses. There are only about 4 billion possible IPv4 addresses, which is only enough to provide 1  IP address for half the population on the planet. However, IPv6 provides about 32000 IP addresses for every human. This means that Matter devices can communicate through your router to the WAN without network address translation. They need to be extremely secure, and the Matter spec allows the firmware to be updated remotely and automatically as soon as a security exploit is discovered.

Matter is currently being implemented only for devices and situations with low data rates. Zigbee is now part of Matter, so it will turn lights on and off. Matter can control door locks, thermostats and control TV's. It is being used to synchronize Google's Nest Audio devices and in this situation is works the same way Sonos does when Sonos synchronizes its speakers using SonosNet.

Matter is also in networking equipment such as mesh routers and switches. If you are watching video from the cloud in several locations in your home, on a normal LAN all the packets for all your video streams would be visible everywhere on your LAN. Matter solves this problem. I believe it can manage your mesh routers and switches to utilize your LAN efficiently so that packets are sent where they are needed and not everywhere on your LAN.

 Matter devices have standardized APIs that will be supported by about 100 companies that support Matter. The list of companies includes Google, Apple, Samsung and many others. People can go to Best Buy and just look for the Matter logo on a product and know that it will work with almost everything.

Now to my situation. I have 2 Ethernet LANs with mesh routers. On these I have about 120 devices including 13 cameras, 50 Google devices, over 20 Sonos speakers and much more. So I am really happy to have my other 250 devices such as light switches, HVAC, irrigation, etc on INSTEON. I couldn't add these 250 devices to my LANs that already have 120 devices.

However, INSTEON has its problems, such as keeping RF noise out of my electrical wiring. Ultimately, I would like to replace all my INSTEON devices with Matter devices. Moreover, Matter will support door locks, which I can currently not controlling with my ISY.

I've thought about how the ISY could work with Matter and here are 3 possibilities.

(1) The ISY supports Zigbee, and Zigbee is now part of Matter. The ISY has a Zigbee radio so it could communicate with Matter devices using their IPv6 addresses. This would almost certainly require a firmware update in the ISY's Zigbee radio. I don't use Zigbee and don't know much about it, so I'm wondering if the Zigbee radio chip can be updated in this way.

(2) People who use Home Assistant and Open HAB are very keen on connecting to Matter devices. Suppose they succeed in getting software running on their Linux systems that can talk to Matter devices using their IPv6 addresses. Then it should be possible to put the same software on the POLISY and connect the matter devices to the virtual node server framework. Then a Matter node with your Matter devices should appear on the ISY.

(3) We (or rather I) don't know if interacting with Matter devices using the IPv6 address has all the capabilities that would be available if the POLISY could talk to Matter devices directly using Thread.  I think there are going to be devices with a thread radio that could be used to build a POLISY. Maybe there will be a RaspPi with a tread radio. I think people using HA or OpenHAB will probably start using these before we will. However, whatever software is developed, it should run on the POLISY and we ISY fans should be able to interact with Matter devices.

In my case, as INSTEON slowly dies, I'd like to swap all my INSTEON devices for Matter devices and continue to run my smart home with my ISY.

The questions I'm most interested in having an answer to in this forum are (1) is UDI thinking about whether or how the Zigbee radio can work with Matter, and (2) are any of you POLISY developers interested in getting the Matter software running on the POLISY.

The Isy uses ZigBee for energy not ha which is a different ZigBee so no it could not use it's radio. Matter uses a combination of WiFi, Bluetooth, thread etc. Since the Isy/Polisy is connected to your network, it would just use that should they choose to support it. UDI has also started their position on matter.

I think most here do not care about matter at this stage. With it being pushed back until 2022 as well as the promises made over the years about unification, most are in a wait and see approach. If it works out, then great! I'm sure UDI would take a serious look at it at that point. But for now, especially with it delayed, i don't UDI or any developers are thinking about what they are going to do. 

The Isy has already been integrating products for years and at a level matter doesn't do it at. Most Isy users want full control over their systems, without cloud connections so there isn't as much of a rush from them either

Edited by lilyoyo1
Posted

@bruceyeg

It seems amazon routers may be looking toward the future with built in zigbee support. For HA things this may be the way to go, down the road.

But zigbee is only a low level protocol and was implemented many different ways with higher level protocols that are not compatible with each other. I have been told that standards have been created and are being followed over the last few years.

https://www.amazon.ca/Introducing-Amazon-eero-dual-band-extenders/dp/B086P34DPF/ref=sr_1_7?crid=1N7THHK4Q7070&dchild=1&keywords=wifi+router+with+zigbee&qid=1630412895&sprefix=Router+Zig%2Caps%2C241&sr=8-7

 

Posted
5 minutes ago, larryllix said:

@bruceyeg

It seems amazon routers may be looking toward the future with built in zigbee support. For HA things this may be the way to go, down the road.

But zigbee is only a low level protocol and was implemented many different ways with higher level protocols that are not compatible with each other. I have been told that standards have been created and are being followed over the last few years.

https://www.amazon.ca/Introducing-Amazon-eero-dual-band-extenders/dp/B086P34DPF/ref=sr_1_7?crid=1N7THHK4Q7070&dchild=1&keywords=wifi+router+with+zigbee&qid=1630412895&sprefix=Router+Zig%2Caps%2C241&sr=8-7

 

While there are tons of Zigbee battery devices and bulbs I have yet to see a Zigbee light switch or scene controller that I would put in my house (excluding proprietary C4 switches). Maybe these are coming soon? Plus Zigbee is 2.4GHz which can only compete with local Wi-Fi, not enhance it. 

Posted (edited)
6 minutes ago, upstatemike said:

While there are tons of Zigbee battery devices and bulbs I have yet to see a Zigbee light switch or scene controller that I would put in my house (excluding proprietary C4 switches). Maybe these are coming soon? Plus Zigbee is 2.4GHz which can only compete with local Wi-Fi, not enhance it. 

WiFi has many 2.4GHz channels and baby monitors, and some wireless phones all use the 2.4GHz bands. They don't compete. Many utilities use Zigbee in their metering to read your meters remotely. Speed is not a factor there though, as it would be for HA.

I haven't looked at any of the new standard Zigbee equipment yet, but I have always heard it is a much better protocol. Philips Hue users don't report problems that I have heard.

IIRC wikipedia had a good article on the zigbee standards coming out.

Edited by larryllix
Posted
3 minutes ago, upstatemike said:

Plus Zigbee is 2.4GHz which can only compete with local Wi-Fi, not enhance it. 

This is a common misconception.

Not all 2.4 GHz competes with WiFi.  For one thing, if it's a problem, the 2.4 GHz ISM band is fairly wide so one or the other can move out of the way, and another is that the 5, 5.8 and 6 GHz ISM bands are where speed and latency sensitive devices sit... but more importantly, ZigBee and WiFi (and Bluetooth) just look like a very slightly increased noise floor to each other due to using very different modulations.

Now, other WiFi, yes, that causes issues.  As does your microwave if the Faraday cage isn't perfect, as will a local amateur operator running a couple of hundred watts into a directional antenna... but these are all entirely different issues.

By and large, I would rather something run in 2.4 than the 700-900 MHz range inside of my house because of much better propagation and penetration characteristics, as well as the smaller, more sensitive antennas.  For HA devices, 2.4 ISM is the band you want to use. :)

Guest
This topic is now closed to further replies.

×
×
  • Create New...