Insteon acquired and servers coming back up

[RPerrault]

12 minutes ago, upstatemike said:

5 Products like Lutron, Insteon, Hue, Broadlink Fastcon. and Yolink LoRa, will do a bare minimum Matter integration to get the logo but will otherwise ignore it as it provides less advantages to consumers than the native versions of those products.

the standard - i suppose - could let a lutron device still speak lutron

"An old argument for using proprietary messaging is that products need differentiating capabilities that are not necessarily part of a universal standard. But how many different ways do we need to set a temperature or turn on a light? The correct answer is “one.” However, some products have unique features that Matter does not define. In these cases, manufacturers are free to define proprietary messages because IP networks do not restrict message content. So, functions common to many products such as transmitting well-known commands and data types and setting up new devices on a network work predictably across all Matter devices without restricting manufacturers from using non-interoperable product-differentiating messages where appropriate."

 

[nowandthen]

23 hours ago, Dub aka WHaas said:

Snip...

I'm cautiously putting this out there cause I have pieced this together from multiple reads and is my understanding.  There are also some very passionate opinions about Matter.....fingers cross this doesn't turn into a thread hi-jack! hahah

I think it's time to put this Matter to rest.

Edited August 7, 2022 by nowandthen

[asbril]

21 minutes ago, nowandthen said:

I think it's time to put this Matter to rest.

What is the Matter with you ? 😁

[Dub]

2 hours ago, upstatemike said:

So my take away from the above is:

1 I probably misused the term protocol when I should just say platform to identify a particular technology.

2 While it is possible for a routed technology to avoid the popcorn effect it would take some attention from the manufacturer to prioritize that and none have published any intent of addressing it. Insteon being a broadcast technology avoids the issue automatically.

3 Popcorn effect really really really sucks and we need better channels to communicate to manufacturers that fixing this is important. I suggest any device sold on Amazon deserves a question about smooth group actions to deliberately emabarrass any product manufacturer who did not make an effort to provide this in their offering.

4 Most of the benefits of Matter will only be realized when end point devices speak Matter natively (so plan to replace everything you have). The kluge of having a hub do translation is just a way to get by in the near term.

5 Products like Lutron, Insteon, Hue, Broadlink Fastcon. and Yolink LoRa, will do a bare minimum Matter integration to get the logo but will otherwise ignore it as it provides less advantages to consumers than the native versions of those products.

 

 

Thats where I am with all the info presented.  

I did read something else that Matter over Thread will have the ability for devices to communicate with each other even with the hub down.  Hopefully this isn't like Z-Wave offers direct device to device communication(very inconsistent across manufacturers).

@nowandthenand @asbril

 I can't resist....

If you don't mind, it doesn't matter!!!!🤣

Loose attempt to get this back on Insteon.  Would love to see a poll of who thinks we will see newly manufactured devices from Insteon(not some leftovers from the old company, but the new orders to manufacturing) or Matter released to the general public first with devices for sale!!!

[upstatemike]

2 hours ago, RPerrault said:

multicasting?  

Only works if the manufacturers implement it. If Z-Wave used it I could probably live with that platform; but they don't and nobody else has stepped up to say "yes we know that this is important so we will make sure multicast control of groups is part of the most basic command set that will be required to be supported across all Matter products and manufacturers"... unless they did and I just missed it somehow.

[upstatemike]

2 hours ago, RPerrault said:

the standard - i suppose - could let a lutron device still speak lutron

"An old argument for using proprietary messaging is that products need differentiating capabilities that are not necessarily part of a universal standard. But how many different ways do we need to set a temperature or turn on a light? The correct answer is “one.” However, some products have unique features that Matter does not define. In these cases, manufacturers are free to define proprietary messages because IP networks do not restrict message content. So, functions common to many products such as transmitting well-known commands and data types and setting up new devices on a network work predictably across all Matter devices without restricting manufacturers from using non-interoperable product-differentiating messages where appropriate."

 

So a couple of things here:

1 I don't care if there is only 1 way to control light as long as "control" includes ramp rate, color or color temperature, behaviour in a scene, status reporting that the command was recieved and executed, etc. Who is deciding what "control" means and what needs to be included as part of the universal standard?

2 As many folks have noted on other threads, automation is not just about remote control but also about triggering things from current state, state change, manual vs programatic changes, multiple button presses and long vs short button presses. etc. Where is my assurance that all of that is part of the universal standard?

3 If manufacturers implement proprietry features without following a framework or standard then what prevents another manufacturer from implementing the same feature in a different proprietary way? At what point does Matter declare that function to be common enough that it should be part of the universal standard and how will thay walk back all that proprietary junk and force it to conform to the newly updated universal standard?

[upstatemike]

2 hours ago, nowandthen said:

I think it's time to put this Matter to rest.

Happy to go back to talking about the latest communication from Inseon instead... If only there was one.

[RPerrault]

8 minutes ago, upstatemike said:

Only works if the manufacturers implement it. If Z-Wave used it I could probably live with that platform; but they don't and nobody else has stepped up to say "yes we know that this is important so we will make sure multicast control of groups is part of the most basic command set that will be required to be supported across all Matter products and manufacturers"... unless they did and I just missed it somehow.

my guess is that if it were in 'matter', we would not know yet

i am not saying its the only way to do it - i am saying that it might be one way it could be implemented - instead of declaring it impossible

we do know wifi supports it - don't know about the matter application - both would be necessary to implement it

[upstatemike]

2 minutes ago, RPerrault said:

my guess is that if it were in 'matter', we would not know yet

Which itself is a concern. We are into double overtime now for the published standard and we still only have vague marketing descriptions about what it will and will not include. There should be a Matter spokesman who can get down into the weeds about this stuff by this time but all we get are smoke and mirrors.

[RPerrault]

8 minutes ago, upstatemike said:

2 As many folks have noted on other threads, automation is not just about remote control but also about triggering things from current state, state change, manual vs programatic changes, multiple button presses and long vs short button presses. etc. Where is my assurance that all of that is part of the universal standard?

doesn't have to be - they allow for that

"...manufacturers are free to define proprietary messages because IP networks do not restrict message content. So, functions common to many products such as transmitting well-known commands and data types and setting up new devices on a network work predictably across all Matter devices without restricting manufacturers from using non-interoperable product-differentiating messages where appropriate."

9 minutes ago, upstatemike said:

3 If manufacturers implement proprietry features without following a framework or standard then what prevents another manufacturer from implementing the same feature in a different proprietary way? At what point does Matter declare that function to be common enough that it should be part of the universal standard and how will thay walk back all that proprietary junk and force it to conform to the newly updated universal standard?

it IS in the 'framework'

if the capability is later adopted by matter - it would not remove the ability to adopt the new command or keep doing it the way they always were before the adoption

[upstatemike]

2 minutes ago, RPerrault said:

it IS in the 'framework'

if the capability is later adopted by matter - it would not remove the ability to adopt the new command or keep doing it the way they always were before the adoption

So if you buy a device it might follow the current Matter standard or it might work using legacy proprietary messages (buyer beware). Don't we already have that with Z-Wave and Zigbee so called standards? What was the point of Matter again?

[RPerrault]

2 minutes ago, upstatemike said:

Which itself is a concern. We are into double overtime now for the published standard and we still only have vague marketing descriptions about what it will and will not include. There should be a Matter spokesman who can get down into the weeds about this stuff by this time but all we get are smoke and mirrors.

they may not have it all defined - i have not looked, but seen little info on the commands that will be in matter

i think they chose a good path allowing proprietary commands - ge matter devices probably won't need to be concerned with setting a status light to magenta - if lutron is, they can still do it without violating the matter standard - the larger the command set, the more resources needed

we do now see there is a chip for the networking - people here were claiming the impossibility of that working

matter is speaking to the people that 'matter' - companies that can build and adopt the technology - matter is not selling to end users - they have to sell manufacturers

[upstatemike]

2 minutes ago, RPerrault said:

 matter is not selling to end users - they have to sell manufacturers

Who for all their market research still seem unable to grasp what users actually want from these products.

I'll quit bashing Matter for now and wait until there are some actual products available to evaluate. 

I still might change my username to "antiMatter" though.

[RPerrault]

3 minutes ago, upstatemike said:

So if you buy a device it might follow the current Matter standard or it might work using legacy proprietary messages (buyer beware). Don't we already have that with Z-Wave and Zigbee so called standards? What was the point of Matter again?

do you care which it uses?

no - z-wave cannot deliver a packet to anything but z-wave devices

in theory, i suppose insteon could wrap its existing commands in a ''matter compliant' packet and never use any matter commands 

[lilyoyo1]

I don't think we'll see a unified standard per se from the early generation devices. Over time, things could potentially get better as (if) lessons are learned and applied.

Most likely matter will not be able to integrate disparate devices in such a way that they operate in unison. It May be possible with those devices that use thread directly but not when mixed with other technologies such as insteon, zwave, ZigBee, or clear connect. Everyone has invested too much into their chosen ecosystems to simply abandon their technology. Even with the Isy - commands are sent out to the individual protocols one at a time vs simultaneously. It's fast but still distinct separation. 

@upstatemikeBecause of unnecessary complications with automation, I doubt Matter will accomplish that any more than what's currently possibly with Alexa/google. Automation vs control is part of our world, not necessarily theirs. Having a controller such as Polisy or eisy can deliver the automation that matter probably lacks. Most people think what we do is cool but want simplicity. They feel it's too complex. In a sense, we'd get the best of both worlds since we matter has the control while UDI has the automation. 

The fact that they don't talk about these types of details leads me to believe that automation and  simultaneous communication across platforms will not be present. 

@Dub aka WHaas Early on, I think older devices will be released. Depending on timeframe for new devices, this could be their doom. If new customers were buying insteon like that, they wouldn't be in this position now. Couple that with all those they've already lost means a smaller customer base to to work with. With their recent history- i just don't see newbies flocking to use their technology. 

[apostolakisl]

It seems to me that UD already has all of this handled.  The have nodes that work with anybody who has an API and someone willing to write a node for it.  They have standard definitions for various node parameters (like on or off) and then an ability to add however many unique node definitions for the odd ball device that does something no one ever thought of.  The only place I find that UDI comes up short is integration with google and Amazon.  But the whole yelling at my google home or Amazaon Alexa kind of lost its appeal a long time ago.  The "ok google" or "Alexa" thing just doesn't work.  I have my phone, my google home, my wife's phone, my kids phone, etc.  Seems like every time I say "ok google" either none of them or all of them respond, either way it screws up.  I find myself whispering "ok google" into the microphone of my phone to get my phone only to respond, then my phone doesn't respond and my google home that is 50 feet away somehow hears me and responds, but then doesn't hear me when I tell it what I want.  OK google is good for one thing, setting a timer for whatever you just put in the oven.  Alexa is good for the shipping list.  I love my Harmony Hub with its ISY node server, I don't have to look at it, I don't have to worry about background noise, I just push a button that my finger knows well and it does what I have programmed the button to do, and it does it instantly, and without any need for confirmation.  It is way faster than I could ever speak a command and it works every time, no worries about my screaming kids messing it up.  I don't get any popcorn affect, it just happens as if the button in my hand was a hardwired light switch, or tv power button, or whatever.  And all the stuff I just want to happen at a certain time or when I open a door or a window, or whatever, it just happens as programmed.  For the life of me, I can't figure out what any of this matter is going to do better than I already have.  Until I can say "ok google: rub my feet, bring me a beer, clean the house, and do my laundry", and have it actually do those things, I won't be any better off than I am right now.

[RPerrault]

2 minutes ago, upstatemike said:

Who for all their market research still seem unable to grasp what users actually want from these products.

I'll quit bashing Matter for now and wait until there are some actual products available to evaluate. 

I still might change my username to "antiMatter" though.

i understand - change is scary

you do not know what matter will support - their market research skipped you so they did none?

matter might clean your toilet - or have overcome popcorn - or have the magenta command

the argument that is completely inane is 'who needs just another standard' - why would anyone object to yet another standard since matter is so crappy - ignore matter - no one has to adopt it - keep guessing what insteon will do - and hoping

it sounds like a good concept - a huge part of the battle is the network component - people saying it can't be done - well it HAS been done

not attacking - i enjoy the thought exercise and thanks for challenging me - i like speculating on how challenges might be overcome - the ones we can envision

 

[Dub]

32 minutes ago, lilyoyo1 said:

Early on, I think older devices will be released. Depending on timeframe for new devices, this could be their doom. If new customers were buying insteon like that, they wouldn't be in this position now. Couple that with all those they've already lost means a smaller customer base to to work with. With their recent history- i just don't see newbies flocking to use their technology. 

@lilyoyo1I agree with this.  It sounded like from the Podcast interview some things were being worked on like local control by the old Insteon.  However,  I don't see any way the new Insteon could have placed manufacturing ordering for anything but the old products.  Nokia has said the licensing deal was dead and if that was back on and hardware available or being manufactured for it we would probably be to hold to create some excitement.  If I remember correctly you actually have some Nokia gear.  I don't see newbies going there either.  The burden is definitely on the new Insteon to find a way to make this work.  We have messaged about the installer base and I think that is critical, because when all my switches in my house have been replaced and all my other devices are in place, I'm not generating new revenue.

@lilyoyo1my comments on the poll were more tongue in cheek for a vote to see what people think what we will see first.....newly manufactured(not new products) from Insteon or Matter products for the general public.  Since Matter has been delayed several times already...who's to say Ken's slow boat from China won't get here with old design, but newly manufactured Insteon devices, before we can buy Matter devices.  

 

Edited August 8, 2022 by Dub aka WHaas
Spelling...

[RPerrault]

Just now, apostolakisl said:

It seems to me that UD already has all of this handled.

ud cannot deliver one packet

ud can present devices using insteon and z-wave to a user and allow our interaction to be something close to single pane

for packet delivery, ud needs a z-wave radio and a plm for insteon

what ud does is remarkable - and if you want to buy z-wave repeaters and more insteon to 'build out the network', go for it - it takes a ton of work to do what ud has done - but ud saw it could not expand the isy to include lutron (if lutron allowed it) and all the others - ud built a device with insteon (plm), z-wave (z-wave radio board) and elk - ud has those deliver the packets - node servers do not deliver a packet - they construct the packet and pass it to ud - ud passes it to whatever network interface it has that can deliver it

i got nervous when ud discussed the decision to move to the polisy modularity of just get out of the game - focus on energy - ud - if nothing else - could allow a graceful, gradual migration path for us

and nothing says ud would not add matter as just another node if they have a network interface - its not mutually exclusive

[lilyoyo1]

18 minutes ago, Dub aka WHaas said:

@lilyoyo1I agree with this.  It sounded like from the Podcast interview some things were being worked on like local control by the old Insteon.  However,  I don't see any way the new Insteon could have placed manufacturing ordering for anything but the old products.  Nokia has said the licensing deal was dead and if that was back on and hardware available or being manufactured for it we would probably be to hold to create some excitement.  If I remember correctly you actually have some Nokia gear.  I don't see newbies going there either.  The burden is definitely on the new Insteon to find a way to make this work.  We have messaged about the installer base and I think that is critical, because when all my switches in my house have been replaced and all my other devices are in place, I'm not generating new revenue.

@lilyoyo1my comments on the poll were more tongue in chek for a vote to see what people think what we will see first.....newly manufactured(not new products) from Insteon or Matter products for the general public.  Since Matter has been delayed several times already...who's to say Ken's slow boat from China won't get here with old design, but newly manufactured Insteon devices, before we can buy Matter devices.  

 

Matter won't be in the insteon protocol itself but part of the new hub which then talks to the devices. 

I know they were stocking up on the nokia stuff so they do have a starting point. It's just whether or not they are able to sell them and build upon that. I'd actually be willing to use the Nokia line. They're fantastic. In fact, in my test room, I'm still using them. By renaming them, they could at least have some separation to the original insteon company. Either way, they have a long road ahead. I think new stuff  gives them the best chance overall.

For clarity- when I say nokia- I'm only saying that to differentiate between the old and new stuff. Not the state of their deal. 

Edited August 8, 2022 by lilyoyo1

[RPerrault]

22 minutes ago, apostolakisl said:

The only place I find that UDI comes up short is integration with google and Amazon.

flat out wrong - ud was the first (or among the first) to closely integrate - meaning no 'tell insteon to...' formulaic crap - allowed a synonym so when echo heard damn lights it meant den lights - i control z-wave via ud integration - between echo and ud, 'turn off the lights' mean the den lights on the den echo - bedroom lights on the bedroom  echo

there might be better integrations but i doubt it

if you have to yell at your echo, you need one closer

if you think grabbing your phone to adjust your a/c at 3 am is good - go for it - i'll just say 'alexa, set the bedroom thermostat to 72 degrees'

 

[lilyoyo1]

23 minutes ago, RPerrault said:

flat out wrong - ud was the first (or among the first) to closely integrate - meaning no 'tell insteon to...' formulaic crap - allowed a synonym so when echo heard damn lights it meant den lights - i control z-wave via ud integration - between echo and ud, 'turn off the lights' mean the den lights on the den echo - bedroom lights on the bedroom  echo

there might be better integrations but i doubt it

if you have to yell at your echo, you need one closer

if you think grabbing your phone to adjust your a/c at 3 am is good - go for it - i'll just say 'alexa, set the bedroom thermostat to 72 degrees'

 

Depends on your definition of first. Smart things, insteon hub, revolv, Lowe's iris, homeseer, wemo, hue, nest, and indigo all supported Alexa before UDI. The having to tell (chosen device) was a limitation of early APIs and less a UDI (or any other company) issue. 

UDI did do a fantastic job of subsequent upgrades to each api version

Edited August 8, 2022 by lilyoyo1

[Dub]

9 hours ago, lilyoyo1 said:

By renaming them, they could at least have some separation to the original insteon company. Either way, they have a long road ahead. I think new stuff  gives them the best chance overall.

That’s an interesting thing to think about.  

 

9 hours ago, lilyoyo1 said:

For clarity- when I say nokia- I'm only saying that to differentiate between the old and new stuff. Not the state of their deal. 

I’m tracking.  That is also something else that I guess time will tell if Ken aka Insteon ordered the newly designed line(Nokia not Nokia) or the old stuff.  I think you mentioned your guess would be the old stuff.

Was there any new features in the “Nokia” (new line of products) that you can disclose without violating any confidentiality or NDA that might still be enforced(not sure how all that works or holds up  with all the business side changes).

[apostolakisl]

11 hours ago, RPerrault said:

flat out wrong - ud was the first (or among the first) to closely integrate - meaning no 'tell insteon to...' formulaic crap - allowed a synonym so when echo heard damn lights it meant den lights - i control z-wave via ud integration - between echo and ud, 'turn off the lights' mean the den lights on the den echo - bedroom lights on the bedroom  echo

there might be better integrations but i doubt it

if you have to yell at your echo, you need one closer

if you think grabbing your phone to adjust your a/c at 3 am is good - go for it - i'll just say 'alexa, set the bedroom thermostat to 72 degrees'

 

I have had to re-connect my UD to Google too many times to count.  To me, this is a failure.  It may be on Google, it maybe on UD, I don't know.  But if I were a professional installing this at a customers house, they would not be happy, nor would I.  I don't really have issues with my thermostat in the middle of the night.  I occasionally need to go to the bathroom, maybe I can ask Alexa to do that for me?

And I'm not blaming UD for Google issues.  It is on Google.  I have no way of stopping my phone from trying to do something I want my google home to do or vice versa.  I can't tell my phone to make a phone call because my google home will invariably try to make the call instead.  I have just given up on it.  Same with Alexa.  

Edited August 8, 2022 by apostolakisl

[apostolakisl]

12 hours ago, RPerrault said:

ud cannot deliver one packet

 

and nothing says ud would not add matter as just another node if they have a network interface - its not mutually exclusive

1) This is academic.  The user doesn't know, doesn't need to.  It will probably make the back-end easier, so it really is about simplifying integration for the manufacturer, not about improving the user experience.  The user interface can be made to work ~ identical regardless of what protocol you are connecting to.  I can't imagine that connecting a matter device will be any different than an Insteon device.  I'm sure you will click an "add device" button on your Matter User interface, then click a button on the device itself.  Then you will need to label it, categorize it and link it to whatever devices you want it to work in concert with.

2) And this effectively makes it another PLM or Z-wave module, a module that every device that uses Matter will have.  It really comes down to trying to get everyone on the same system so you only need 1 "dongle".  Z-wave and Insteon protocol both have their short-comings, and perhaps Matter will not have any.  We shall see, but not having any short-comings is unlikely. 

 

EDIT: Just recalling some of the comments Michel made when they were looking into acquiring Insteon assets.  One of the main things was "can Insteon protocol implement encryption".  In my opinion, this is really the issue with Insteon.  Otherwise, it checks all the boxes.  Low power, simple, latency less than human threshold of notice, excellent range/penetration.  It does suffer from power line noise but I am convinced that some minimal engineering where power line signal is decoupled from radio signal would fix that.  I suspect that encryption is not in the cards for Insteon.  I dont' know how z-wave could ever be fixed regarding the popcorn affect.  After trying a few z-wave devices, I uninstalled them.  I would use z-wave, perhaps for things where a secure protocol is needed and synchronization isn't an issue, like a door lock.  But my house has fancy door locks that can't be replaced.   I use Elk to control anything secure and was in a position to hardwire all those things during construction.    

IN SUMMARY:  I hope that Matter does take off.  Reason is that it will force the hand of most companies into published "API".  I put API in quotes because I suspect it won't be considered an API over the Matter protocol.  But the biggest issue I hear from node server development is that they get no support from the manufacturer re API and thus have to do a lot of reverse engineering and sometimes just can't get things to work.

Edited August 8, 2022 by apostolakisl