MyServer or CQC Anyone?

[simplextech]

5 minutes ago, Michel Kohanim said:

You are welcome to revert to anything you want alas I still haven't figured out why you cannot just develop a CQC driver for ISY?

I can do this.  This is not a problem for me.

I started this thread to look at and document findings of ways to extend my environment beyond the limits of the ISY and even Polisy which currently is just basic Polyglot.  I have system requirements that the current UD ecosystem does not cover currently.  Can/will it in the future?  Perhaps... but when?  5-10 years from now?  This whole thread somewhat went sideways because there is a great opportunity at hand for a CQC + ISY/Polisy merge/partnering but I think the larger picture is lost in translation and so I will as typical go and do my own thing.  Others will wander off and use Home Assistant (many already do) many already use HomeSeer on top of ISY because they do more than Polyglot can do and will do anytime soon.

 

[Michel Kohanim]

@simplextech,

I appreciate the feedback and all your support and help. This said, however, I take major issue with your repeated statements of this flavor:

30 minutes ago, simplextech said:

even Polisy which currently is just basic Polyglot

What do you know about Polisy? How do you know it's basic Polyglot? Do you know about all the package management, networking services, serial port/USB abstraction layers, and 2 years worth of code for Polisy that's 100% outside of Polyglot? Do you know about our abstraction layer for everything and anything within Polisy to make migration of ISY and other things easier (btw, ISY migration has started)?

It's really one thing to say you want a powerful hardware and something like CQC right now, which I do appreciate even though I might not see the ROI/business case. It's also 100% OK to disagree with our design methodology and choice of hardware/OS. But, it's completely another to disseminate, at best ill-informed and at worst completely false statements on a public UD forum relating to UD products. We have been completely open with all our design decisions (good and bad) and thus I am shocked to hear such statements.

 

With kind regards,
Michel

[simplextech]

5 minutes ago, Michel Kohanim said:

@simplextech,

I appreciate the feedback and all your support and help. This said, however, I take major issue with your repeated statements of this flavor:

What do you know about Polisy? How do you know it's basic Polyglot? Do you know about all the package management, networking services, serial port/USB abstraction layers, and 2 years worth of code for Polisy that's 100% outside of Polyglot? Do you know about our abstraction layer for everything and anything within Polisy to make migration of ISY and other things easier (btw, ISY migration has started)?

It's really one thing to say you want a powerful hardware and something like CQC right now, which I do appreciate even though I might not see the ROI/business case. It's also 100% OK to disagree with our design methodology and choice of hardware/OS. But, it's completely another to disseminate, at best ill-informed and at worst completely false statements on a public UD forum relating to UD products. We have been completely open with all our design decisions (good and bad) and thus I am shocked to hear such statements.

 

With kind regards,
Michel

Calm down I didn't call your baby ugly.

Here's the thing.... what you know versus what everyone see's.  It's called perception.

You and I both know that the internals of Polisy are awesome and has huge potential.  The average every other user out there's perception from what they can SEE today is only Polyglot....

What *I* need and what the average ISY user need are probably vastly different and I know this.  I didn't start this thread to become a pissing match between platforms I was actually hoping to get some input from others that I know would have some experience with them and some insights good bad to layer on top of the ISY.  I know many here run HomeSeer on top of the ISY and only use ISY for device management and others do the same with Home Assistant.  I actually use the full ISY stack and Polyglot without anything else on top of it and I've hit a wall where my needs aren't met neither are my wants for expanding the system.  So I originally wanted to layer on top, not replace.

[larryllix]

When I hear about other systems using ISY as a hub just to interface with an Insteon PLM, it comes to mind why the two logic systems? Why the extra cost? Why wouldn't say CQC just write drivers for the PLM?

The Wink always comes to mind here. I bought two, with their seven different protocol interfaces to everything else that was going to eliminate everything else,, the first thing I ran into was the cool video demo setup videos that show you where the linking button was on the Insteon Hub, and the linking button on the Philips Hue hub and the linking button, and the linking button..... you get the picture.
Now you can buy a hub that can interface with all other hubs....wow!  another hub...just what I always wanted.  ? :puke.

On that note anybody else remember the day when you could plug in an interface adapter in your toy PC that could connect to a bus where more modules could plug into that bus with a few more ribbon cables that could connect to an adapter that could connect to a printer and the whole thing needed spaghetti sauce by the time it was done. Next came hardware connector contact problems where every tech had a white plastic pencil eraser and isopropyl alcohol, that cleaned contacts every week in order to avoid those HDD read/write errors and software crashes? (Just before IBM developed 9 bit wide memory with parity bits for every byte?) 

I think Polisy is trying to prevent history from repeating itself, here. I hope!!   

[lilyoyo1]

I think a partnership would be great as it would give the best of both worlds. For what it does; I feel the isy is second to none. It's not designed to be a full scale pro grade home automation system so it doesn't have all the bells and whistles that those systems provide. DIY systems that do try that generally struggle due to various reasons. This is why I always say homeseer is a jack of all trades but a master of none. It tries to do too much and in the end struggles overall. 

I've had the touch screens in the walls and realized after the 2nd month they were a waste of money. Sure it looks good and is impressive at first sight but once that wears off, you realize there are more efficient ways of doing things. 

With all that said, there is a growing contingent (or maybe they're just complaining more/louder) of users that want that type of set up. They want the glitz and glamour that a Crestron, C4, and savant has. I get that. To be honest, with nodeservers allowing A/V to be easily integrated, it will probably come down to that at some point. 

Polisy is the next evolution of ISY. Some may not see the future in it now but when it to gets where I believe it will go, there will need to be a some type of interface. Since A/V can be integrated, many will want/desire/need something to interface with. If I were an app user, I wouldn't want to have to open up mobilinc to turn on the lights and then close that app to go to my Sonos to turn on my music. Because Polisy takes the isy to the next level, it needs something to showcase it's ststus. I do think CQC could be that if something could get done between both companies.

A partnership would allow both party's to push their own agendas separately with minimal exposure while making everyone on the Isy side happy. People such as myself that are happy to keep the isy the way it is, the app folks get their fancy interface, those who want a little of everything, and UDI can continue the growing the isy as they envision it to be. Sure there will be those complaints about the cost but you can't please everyone. 

Maybe CQC is reserved for Polisy holders (raise the cost accordingly) while standard ISY users can add it as a purchased module. We all have ideas. At some point we need to put those ideas into motion. Simplextech seems willing to do his part and I'm sure other developers would be willing to chip in. 

[Dean Roddey]

3 hours ago, Michel Kohanim said:

@simplextech,

I appreciate the feedback and all your support and help. This said, however, I take major issue with your repeated statements of this flavor:

What do you know about Polisy? How do you know it's basic Polyglot? Do you know about all the package management, networking services, serial port/USB abstraction layers, and 2 years worth of code for Polisy that's 100% outside of Polyglot? Do you know about our abstraction layer for everything and anything within Polisy to make migration of ISY and other things easier (btw, ISY migration has started)?

It's really one thing to say you want a powerful hardware and something like CQC right now, which I do appreciate even though I might not see the ROI/business case. It's also 100% OK to disagree with our design methodology and choice of hardware/OS. But, it's completely another to disseminate, at best ill-informed and at worst completely false statements on a public UD forum relating to UD products. We have been completely open with all our design decisions (good and bad) and thus I am shocked to hear such statements.

 

With kind regards,
Michel

Hey Micheal, it's Dean from Charmed Quark. I'll admit that I share your skepticism. SimpleTech has been very enthusiastic, but I just think that the market has been changed by the big players getting into the game and providing 'good enough for the hoi-polloi' type systems really cheaply. So now products like ours are squeezed even more tightly on the low end. Things like OpenHAB and the like, I dismiss since the folks who use those things are the free beer crowd and ain't gonna spend any money. So we are left with a very small sliver between those lower end systems and the great gulf that lies between the low and high ends. And we are squeezed into a similarly thin sliver on the other side of that gulf as well (our system is more than powerful enough for professional installation.)

I would also though point out that if you want to get to a place like CQC it's a very, very big undertaking. CQC is a million lines of code. The problem with a product like this is that there is literally no end to what it could do, hence there's no end to the things that people want it to do. So it just ends up encompassing more and more and more, because every new customer wants something else. Of course a problem then becomes that you are increasing the complexity of the product non-linearly for each incremental increase in revenues. That's not sustainable. At some point you need to increase the revenues non-linearly, or exponentially.

There are two approaches to building something that broad. One is that you can bring in more and more and more third party black boxes and try to duct tape them together, which is what most folks would do. CQC's approach is the other way around, which is bespoke code from the ground up. That provides it with a level of cohesion and stability and quality that's very hard to achieve in something so large and complex. It doesn't even use the C++ standard libraries, which are pretty crap. It literally is from the ground (well lower level OS APIs) up.

And the thing is, it's not just a product it's really a full platform (not meaning CQC itself, though it is as well) but the underlying code could support the infrastructure to become the data processing, media management, automation, telephony, etc... hub of the home. But that's something that I don't have the resources to make happen. And, even if I did, I again share your doubts as to how that could be monetized in a sufficiently large way to justify what it would take.

So, to reiterate in a slightly less coherent fashion, it's hard to really appreciate how much functionality something like CQC encompasses until you dig into it. It's pretty vast. For you guys to match that would be a pretty epic undertaking. For instance, media management is a big part of it these days, which goes way beyond device control. And just the touch screen UI system and the underlying supporting code is a huge undertaking.

Anyhoo, I'm not sure what the point of all that was. I guess it's to say, that I both share your doubts about how such an endeavor could work at a business level, but at the same time I can't imagine how you could create something that could compete with it in any business practical time scale (assuming you even care to.)

* Oh, and to address the Linux thing, one thing about it is that the effort is not linear to the size of the product. CQC is built on my CIDLib general purpose platform which is highly portable. It has a 'virtual kernel' layer that abstracts the OS services, and everything above that works in terms of my own interfaces. It's not going to be easy of course since I'm not a Linux guy per se. But it's not one of those 'hacks all through the code' type of ports. It's a 'get a Linux driver for the virtual kernel' type of ports. And CIDLib used to be supported on Linux long ago, I just let it go since it wasn't required for my needs. So it's been designed for portability from day one. My porting efforts are related to getting that old support caught up with CIDLib/CQC as it is now and the language and Linux OS changes since then. Still, having said that, it's of course quite possible I'll die before it's done because I can only put so much time into it.

If anyone cares, CIDLib is open sourced, here:

https://github.com/DeanRoddey/CIDLib

 

[Michel Kohanim]

4 hours ago, simplextech said:

Calm down I didn't call your baby ugly.

Ugh ... it has nothing to do with ugliness or perceptions. It has to do with disseminating false statements for a product.

With kind regards,
Michel

[lilyoyo1]

50 minutes ago, Dean Roddey said:

Hey Micheal, it's Dean from Charmed Quark. I'll admit that I share your skepticism. SimpleTech has been very enthusiastic, but I just think that the market has been changed by the big players getting into the game and providing 'good enough for the hoi-polloi' type systems really cheaply. So now products like ours are squeezed even more tightly on the low end. Things like OpenHAB and the like, I dismiss since the folks who use those things are the free beer crowd and ain't gonna spend any money. So we are left with a very small sliver between those lower end systems and the great gulf that lies between the low and high ends. And we are squeezed into a similarly thin sliver on the other side of that gulf as well (our system is more than powerful enough for professional installation.)

I would also though point out that if you want to get to a place like CQC it's a very, very big undertaking. CQC is a million lines of code. The problem with a product like this is that there is literally no end to what it could do, hence there's no end to the things that people want it to do. So it just ends up encompassing more and more and more, because every new customer wants something else. Of course a problem then becomes that you are increasing the complexity of the product non-linearly for each incremental increase in revenues. That's not sustainable. At some point you need to increase the revenues non-linearly, or exponentially.

There are two approaches to building something that broad. One is that you can bring in more and more and more third party black boxes and try to duct tape them together, which is what most folks would do. CQC's approach is the other way around, which is bespoke code from the ground up. That provides it with a level of cohesion and stability and quality that's very hard to achieve in something so large and complex. It doesn't even use the C++ standard libraries, which are pretty crap. It literally is from the ground (well lower level OS APIs) up.

And the thing is, it's not just a product it's really a full platform (not meaning CQC itself, though it is as well) but the underlying code could support the infrastructure to become the data processing, media management, automation, telephony, etc... hub of the home. But that's something that I don't have the resources to make happen. And, even if I did, I again share your doubts as to how that could be monetized in a sufficiently large way to justify what it would take.

So, to reiterate in a slightly less coherent fashion, it's hard to really appreciate how much functionality something like CQC encompasses until you dig into it. It's pretty vast. For you guys to match that would be a pretty epic undertaking. For instance, media management is a big part of it these days, which goes way beyond device control. And just the touch screen UI system and the underlying supporting code is a huge undertaking.

Anyhoo, I'm not sure what the point of all that was. I guess it's to say, that I both share your doubts about how such an endeavor could work at a business level, but at the same time I can't imagine how you could create something that could compete with it in any business practical time scale (assuming you even care to.)

* Oh, and to address the Linux thing, one thing about it is that the effort is not linear to the size of the product. CQC is built on my CIDLib general purpose platform which is highly portable. It has a 'virtual kernel' layer that abstracts the OS services, and everything above that works in terms of my own interfaces. It's not going to be easy of course since I'm not a Linux guy per se. But it's not one of those 'hacks all through the code' type of ports. It's a 'get a Linux driver for the virtual kernel' type of ports. And CIDLib used to be supported on Linux long ago, I just let it go since it wasn't required for my needs. So it's been designed for portability from day one. My porting efforts are related to getting that old support caught up with CIDLib/CQC as it is now and the language and Linux OS changes since then. Still, having said that, it's of course quite possible I'll die before it's done because I can only put so much time into it.

If anyone cares, CIDLib is open sourced, here:

https://github.com/DeanRoddey/CIDLib

 

Thank you for coming here and opening up to us. From where I'm seated, it sounds like the Isy and CQC would be a match made in heaven if both joined forces. 

With both companies stuck between the cheapskates and the big boys, a partnership could produce something that could benefit both sides. While I'm not a fan of app control, I'd buy into it just to support the endeavor. While many people on here probably wouldn't want to pay the cost of things, I think there are quite a few here that would do the same simply out of loyalty. 

[simplextech]

7 minutes ago, Michel Kohanim said:

Ugh ... it has nothing to do with ugliness or perceptions. It has to do with disseminating false statements for a product.

With kind regards,
Michel

Then edit the post to re-word it or delete it.  There was no intent to make a false statement about any product or potential capabilities.

[Dean Roddey]

1 minute ago, lilyoyo1 said:

Thank you for coming here and opening up to us. From where I'm seated, it sounds like the Isy and CQC would be a match made in heaven if both joined forces. 

With both companies stuck between the cheapskates and the big boys, a partnership could produce something that could benefit both sides. While I'm not a fan of app control, I'd buy into it just to support the endeavor. While many people on here probably wouldn't want to pay the cost of things, I think there are quite a few here that would do the same simply out of loyalty. 

The thing is, if it doesn't pay of very significantly it's not worth doing. That's always the problem. Making a few tens of thousands is useless. Making a few hundreds of thousands is useless in business terms, and it would be a huge stretch to picture even that happening. It would have to be more like some millions (on an ongoing basis, not just one time) to really be of use, and that's a far huger stretch.

A big problem with the type of markets our product would appeal to is that they are quite limited. So you make something, and you start selling a nice number of units, then it just drops to a comparative trickle because that's as big as the market is. But, now you have ongoing obligations that are the size of that original burst, but ongoing revenues to support those obligations (leaving aside moving things forward) that are vastly smaller.

[Michel Kohanim]

@Dean Roddey,

Welcome to our forum and thank you for sharing your invaluable thoughts. Not for a second do I doubt the amount of complexity, # of lines of code, and the ingenuity of CQC/CIDLib design and management. And, as a geek, I do very much appreciate it.

ISY/Polisy were never designed to be anything like CQC. ISY is an event based automation and energy management framework. And, the majority of our revenue comes from energy management/utility related projects and not HA. And, there's absolutely nothing preventing anyone to make a layer on top and do whatever they wish. But, pigeonholing ISY/UD into X, I cannot accept.

With regard to CQC partnership, I am not skeptical simply because I do not have enough financial/market information to be able to ascertain the ROI. Do we have access to CQC customers? What percentage of our existing customers would want something like CQC? How much are they willing to pay? Assuming we do anything, what does the topology look like? Multiple boxes? If not, can we run Polisy and CQC[Lite] at the same time (resource wise)? If so, what does it take to port CQC to FreeBSD? And ...

With kind regards,
Michel

[kewashi]

On 1/22/2020 at 2:11 PM, simplextech said:

Why do you think I pulled @kewashito the party?

Seriously the potential that once existed with CQC being able to pull in various systems ranging from automation up through A/V and communications (serial/ip/mqtt) into a cohesive structure and a full touchscreen UI builder.  Powerful... complex but powerful.  With MyServer more on a commercial direction and if CQC does just finally die... the DIY market is left only with HomeSeer and they're changing direction to cater to the lowest common denominator of consumer users to compete with "hubs".

Hi everyone! I am new to the ISY ecosystem and still eagerly waiting for my equipment to arrive. In the meantime let me tell you all what I have and my plans. I authored a dashboard app called HousePanrl that is DIY oriented that is local first and cloud capable. It is customizable via CSS and a built in GUI. I am presently porting it over from PHP to Node.js so it will be easy to offer to this community. I am still thinking through the business model. At present it is donation-ware on the SmartThings amd Hubitat platforms but that won’t likely carry over here - we will see. HousePanel has many very cool features including customized graphics, user colors, floor plan mode support, and more. The Node.js porting is going well and is proving to be far superior to the PHP version - I should have started this way. Anyway stay tuned as more will come as I make progress and test the port when my box arrives. You can read more about HousePanel at www.housepanel.net

 

cheers

[simplextech]

Playing around a little last night and this morning.  A couple hours of work and using stock icons.  I'm not an artist and the scale is six ways from Sunday being wrong but it's a start.

[kewashi]

Really nice start... is this just a graphical mockup or a working example using HP on Hubitat?

[simplextech]

2 hours ago, kewashi said:

Really nice start... is this just a graphical mockup or a working example using HP on Hubitat?

It's mostly functional.  The fans show status on/off but I haven't added any control.  The camera icons load a camera wall page with the 4 cameras.  The bulb icons turn on/off lights the red halo's appear when motion detectors are active the door lines for sensor turn red when a door is open.  All functional... not pretty but functional. 

[Dean Roddey]

What I generally do is use a vector graphics program (like Inkscape) to draw a floor plan type image that fits the overall look, and then drop stuff on that. Here's a video on how you can create one (in this case to integrate it into the pre-fab ones that can be generated based configuration.) A simple floor plan works pretty well and can be made attractive without too much effort. This one is just line based, then I copied it and turned the copy into a drop shadow.

 

[Samazar]

Hello fellow members,

I am new to this forum so please bear with me and be a little patient.

I decided to become a member of this forum because I came across this thread and realised how knowledgeable members are in this thread, some of you sound like full stack developers with years of experience. It is such people that have a deep insight into automation technology and systems.

We a small startup and specialise in AV solutions, we are small but focus on good service and a professional approach with the aim of providing quality and value. I am very particular about our products and partners because we don't have budgets for advertising and most of our business comes from good experiences and referrals and that is how we are gradually growing.

I would like to include and offer home automation services too and have been doing some research. The general feel is that Control4 is a good all rounder system in terms of functionality, ease of use and value. An installer friend of mine says it is a bit limited and that I should look into Elan or RTI. However, I know C4 updated their new OS to version 3 and hence, C4 may have improved a lot since.

Crestron seems to be very reputable but people say it is expensive. I read that the new Crestron Home OS3 is excellent, powerful and easy to use. Spoke to them just out of interest and found them to be very approachable and friendly. Refreshing in this day and age.

Let me explain my needs:-

1. I am not a programmer but studied computer science and basic programming in high school so I am capable of learning an automation scripting language.

2. Most of my clients want a simple and reliable home automation system and don't mind spending a little more for quality.

3. Things that most clients will need automated: lighting (including someone-at-home simulation), security (Cams)/NVR, music. Locks and shades will be rare.

4. A single easy to use remote control that integrates into the automation system for dedicated home theatres and non-dedicated entertainment rooms.

 

I am of the opinion that C4 may be ideal but love Crestron's clean and intuitive interfaces and reputable brand, I also like what I've read about Crestron Home OS3.

I've come across various other solutions including Allonis and Home Assistant but I don't feel confident about them owing to the lack of professional and immediate support and my lack of deep software development / troubleshooting ability. At the end of the day it's not only about profit, it's about building an exceptional reputation and minimising service calls.

 

Any input from highly knowleadeable fellow members will be appreciated because I've been struggling to make the right decision for some time now.

[simplextech]

Welcome to the forum @Samazar.  There are several developers here as well as dealers for various platforms (RTI, C4, ISY) and installers do handle ranges from DIY and up...

I'd suggest you start a new thread so things don't get convoluted in this thread.  Then ask away your questions.