What is wrong with CQC?

[Dean Roddey]

This is Dean's CQC thread and by the way you plugging your Kickstarter Hub in his thread was not cool.

 
Not a biggie...

 

[ChrisCicc]

Not a biggie...

 
Sorry Dean, really wasn't trying to plug! I'm going to create my own dedicated thread tomorrow, CastleOS members have 24 hours early access so I probably shouldn't have even mentioned it. I really only mentioned it because I thought it was relevant - what you have considered doing is what we are doing today for the reasons suggested... I really respect you as a programmer and wish you the best!!!

 

[Neurorad]

Interestingly, Crestron has begun a Crestron Connected hardware approach. All Crestron Connected hardware can be discovered automatically, allowing for automatic discovery and programming. Maybe CQC could take advantage of this? I think initially it's directed to commercial installs, but in a few years it might catch on globally. I believe C4 has something similar, but I really don't have the faith in C4 that I have for Crestron.

I've always thought of CQC as the DIY Crestron equivalent. Any automation is possible, with CQC.

 

[ano]

Pricing is a tough one, but the idea that someone else brought up to charge for each driver is a good one.  In my opinion, again just my opinion, is that drivers are the weak part of CQC. Very important for users but sometimes neglected. I like the Homeseer model of how third-parties could sell drivers, but maybe with a few changes. If an author stops supporting a driver, I think another person should have the option of coming in, supporting the driver, and getting money for new sales. Dean could sell CQC like he loves, and others would be motivated to keep drivers current.
 
Everyone needs a few drivers, and if those drivers are bad, then the greatest CQC in the world makes no difference. For me it was HAI and UBP, and bugs in those really killed it all for me. The driver issue is a problem, again, in my opinion.

 

[pete_c]

Here over the years of using Homeseer 1-2 purchased many Homeseer sources / 3rd party sourced plugins.
 
One sale in the 2000's for a package upgrade of Homeseer 2 to Homeseer 2 Pro included all of the Homeseer Plug ins.
 
I purchased that even though I had already acquired most of the plug ins a la carte.
 
Its been a few years now that one Homeseer 3rd party author passed away.   
 
His plugins were very popular.  Originally his widow was collecting the funds for the plug in with no support. 
 
The source of the plug ins was put up for sale and there were no takers. 
 
BTW - after about a year that he had died I disabled all of the home automation per widows request.
 
I was involved and mostly folks just wanted the source for free.   There are plug in authors today still using the first generation of software for their automation and only developing and not really automating on the newest generation of software.
 
Earlier mentioned one of my "drives" in purchasing Homeseer 3 was Linux.  Another drive has been the use of plugins on remote devices talking to the mother ship.  Typically here have tested a windows plug in in Linux and remote use of said plugin.
 
I still prefer the serial / USB wire to the mothership.  Over the years have played with the virtual over the network serial / USB wire (for VMs) and network connected automation devices.  The PITA stuff though relates to having to insure that the on the switch/POE automation device is always powered.  Gets much worse when wireless is involved.  There are always many assumptions made relating to using the home network, network wireless / POE stuff; wireless related to the magic that it's always connected no matter what.  Today many automation yut's assume much when its a cloud application or wireless application talking to their phone/tablet. Typically as long as the widget (eye candy) is there; automation is up and running fine in their eyes.
 
Many many years ago the most popular software (when using modems) was Compuserve, People Link, and AOL stuff.  While mostly passing text via modem; the graphics and eye candy were based in the software running on an OS on the PC and that was what sold it and made it popular (in the desktop PC days). 
 
I understand software is software. 
 
Here too you see the "regression" to simple firmware (software) doing simple things on the Elk / HAI combo panels; well and they just work without nary a thought about them.
 
Differences though relate to the kernel/ROM/RAM (whatever it is - EFI or '86 today), OS, programming language utilized which puts in those pieces that can fail sometimes.  Then too the kernel/ROM/RAM back then was just few bytes of whatever and  today it's gigabytes.  I played a bit with machine language and compilers and thrilled at the time to increment one number in the middle of a screen. 
 
I always suggest to test it out the use of those additional layers of topology from the hardware piece to the controlling piece of the software.  IE: connect a networked automation device, disconnect it randomly and see if the connection abends the software or eventually abends the mothership. 

 

[wuench]

Pricing is a difficult thing. You have to remember that we aren't just dealing with the DIY market. It becomes very difficult to come up with a pricing scheme that will work for both individual users and pros. You can't afford to sell pro systems for very little money since that's where the actual money is, but you can't sell to individuals for less that you sell to the pros (because their customers will complain that they can buy it for less.) We use to have a pro price and a DIY price (which was much lower.) But got endless complaints about that, because pros expect to get things for less than individuals, because they are buying 'wholesale'.
 
Our tiered system is an attempt to deal with that, by creating a more layered system, in which DIYers can get in for a reasonable price, while the larger systems that pros tend to use can still bring in enough revenues (for us and the pros) to be practical.
 
Ultimately, no matter what system you come up with, someone will think its stupid or a rip off or some such thing. But this is the best we can think of so far.

 
Maybe one answer is to rebrand either the DIY or Integrator product so that the integration customers don't confuse the two.

 

[fcwilt]

Been pondering the "what's wrong" thing a bit - a few other things have come to mind.
 
- The install not being your typical Windows install. You've mentioned in the past some reasons why but I haven't found them compelling. The odd install is "off putting". If tens of thousands of other applications can use a more typical Windows install then why not CQC?
 
- Too many parts. Now I've only worked with 5 other "home automation" systems but they all have a single "point of entry" - one application that you run to configure all aspects of the system. Now that's not to say behind the scenes it's not divided into modules of some sort but from the end users point of view it is one application - and if you have to login you only have to do it once.
 
- Too many inconsistencies. For example some windows have a "close" button in the upper right corner of the window, some don't. Common buttons like Cancel or OK do not have a consistent location.

 

[samgreco]

Maybe one answer is to rebrand either the DIY or Integrator product so that the integration customers don't confuse the two.

I think that this a great answer to the "problem" of price perception in the DIY community.  In fact, there is the model where a package has a free open sourced version and a "premium", pay for version.  Look at Zimbra mail server for an example.  I happen to use the OS version.  But there are features only available in the premium version.
 
I have been on the fence for ages on CQC.  I KNOW how good it is.  I also know that there isn't anyone better than Dean in supporting their product.  But for me, it has always been a financial issue.  The last 10 years have been very tight financially for us and we are approaching retirement.  I pruchased all of my hardware on the cheap.  eBay, closeouts, used.  I don't think that I have more than $3k in hardware in my house and I have UPB lighting control, Xantech multizone, amps, etc.  It doesn't hurt that my job when I started was as a manufacturer's rep for A/V products.
 
But every time I am frustrated by the software side of things and am about to pull the trigger, I realize the money is "better spent" on other, non-tech stuff.  So I stay with Premise.  Yes, it's really old, but it still works, and works really well for my needs.  It's just a bit ugly.  And there are certainly new technologies that it doesn't support.  But I don't need that yet.
 
If money were not an issue, I would buy CQC in a minute.  And I know that the entry level is only a couple hundred bucks, but I need more drivers than that.  So for me, it's the $499 package at the minimum.

 

[Dean Roddey]

Auto-sensing of hardware is something that's definitely on the list. Though, as I mentioned earlier, that's not always going to help. In many case the device doesn't provide enough information to communicate with it without getting more info from the user, and often for very good reason (i.e. security info required to access the device.) So, even if the device is automatically located, it's not necessarily going to be automatically useful. It may still require configuration by the user.
 
On the two separate brands, I'm not so sure that would work. It's not about confusion, but about the fact that the internet means that customers can find out anything they want. If they find that they can buy something for X and the installer wants Y, then they start complaining about it. Installers hate that. I'm sure that they could pretty easily find out that CQC and DIYQC is the same thing for different prices.
 
On the drivers front... well, that's sort of a no win situation. It's already  been complained about a lot in this thread that there's too many things to have to decide on and it's not monolithic enough, and adding drivers to that list would make it a lot worse. And, frankly, allowing people to make money from drives just doesn't make it any more likely that they will remain viable. In fact it probably makes it less so, because then the drivers can be considered the property of the person who created them, and they have no obligation to make it available if they stop supporting it.
 
Actually I think that CQC does considerably better than most on this front by having them all be free and open source (well except for a small number that we have to do in C++ for some reason.) And having them all be free encourages people to finance the development of drivers, because they themselves are benefitting from others having do so in the past.

 

[Dean Roddey]

If money were not an issue, I would buy CQC in a minute.  And I know that the entry level is only a couple hundred bucks, but I need more drivers than that.  So for me, it's the $499 package at the minimum.

 
What all gear do you have? If it's just a few extra drivers, you can buy them separately without moving up to the next tier.

 

[samgreco]

Well, this is a good example of some of the "complaints" about the pricing model being confusing.  I am unsure of how many drivers and clients I need.  I just went through the site again and I am still unsure.  I know I can tell you here exactly what I have and what I'd like to do and you could tell me.  But I can see where a casual user might just go away.
 
So my question is why do you have so many tiers?  Do support cost go up along with the amount of drivers and clients?  Would you get a lot more sales if you had a simpler model with more included?
 
And I think the real question that this thread is asking is:   Why aren't more people buying CQC?
 
I hope that this perspective helps you.  I'd like to see you grow CQC further.
 
So here is my list.  What would the right purchase be?
 
Xantech ZPR68 with Extender (10 zones total)
SageTV (Server, one client and one extender)
Amazon FireTV
Pioneer Rcvr
Outlaw Audio Pre & Amp
Epson Projector
LG TV
HTPC
AM/FM Tuner
iTunes (I need to be able to run at least 2 streams simultaneously to different zones)
Volumio on a RaspBerry Pi for internet streams, although it could be something else, like Squeezebox.
UPB
Global Cache GC100 to control all of the IR stuff
Asterisk PBX
 
3 Tablets and 2 Phones.  One PC with a touchscreen currently.  How many clients is that?  Is it 6?  I am still unsure if each device is a client, or if they were all Android are they one?

 

[dgage]

To add to that, I have multiple tablets, PCs, and phones in my house. My kis have iPads. So to select music with Sonos, I select whichever device is closest. I've got a couple of older tablets too that are rarely used and likely will buy some of the cheap Windows tablets at some point. To me, the per device charge doesn't make a lot of sense now-a-days. If I have to pick a device based on whichever ones have the CQC client on it vs which device is closest, then I'd likely say CQCs days are numbered in my house. I've had that thought since before I even purchased CQC but figured it would work itself out...and I've got way more devices in the house now than I did 2 years ago.

 

[dgage]

 
I think that we finally have figured that out. Sometimes it can be tough to deal with the sporadic issues that come up once in a while. It looks like the issue with RA2 is that it doesn't by default send out async notifications when things change. The driver has to tell it to do that. But, it looks like if the RA2 repeater cycle due to a quick power outage, it can happen so quickly that the driver never actually loses the socket connection, but the repeater has in fact reset it self.
 
So we are just going to have to make the driver more paranoid basically. Normally it shouldn't be possible for such a thing to happen, but of course this is networking where you probably never should say never.
 
Definitely it would be a good thing to keep the RA2 repeater on a UPS to avoid any such kind of weirdness to begin with, but it should be possible to make the driver obsessive enough to deal with this unusual situation.

That's good to know Dean...thanks! And by the way, my RR2 Repeater is on a UPS although I think it may be going bad. Then again. I have a Linksys PAP2 VOIP box on the same UPS and I haven't seen any phone interruptions so I'm not sure. I'll replace that UPS when I rewire my equipment rack in a month or two.

 

[Deane Johnson]

If there was to be a CQC Home and a CQC Pro at different price levels, the Pro version would have to have some things in it the Home version didn't have.  They couldn't be identical or you'd have problems with your professional integrators.
 
On the surface, this seems like an interesting idea to explore.

 

[Dean Roddey]

The reason for the tiers is that, if we have fewer of them, then there's even bigger jumps between them in price, which makes it even harder to get what you want for as close a price as possible. As I said above, we can't have a situation where we sell commercial slash residential systems and the installer makes $25K and we make $200. That isn't feasible. We have to be able to charge survivable amounts for the software for professionals, and of course on top of that we have to give the pros discounts as well, so that they have room for margins. There isn't any on a $200 product.
 
And it makes no sense to sell a system that would be targeting a very large home for the same as one targeting a single machine in a home theater. There has to be capacity scaling or you either have one price that is probably too high for almost everyone, or one price that is so low that it's not worth selling the product. We could go a la carte, just buy however many clients or drivers you need and add more as you go, I guess. But the thing is the tiers have a considerable discount built in relative to just buying a la carte. We'd either lose that or have to introduce some (again complicated) scaling discount instead of just having some basic tiers with discounts built in (to which you can already add capacity as required.)
 
Since we can't sell the DIY product for less than the pro product, unless the DIY product is less capable anyway (in a way that is relevant to the pro), the only way to have some way to keep a lower price for the entry level is to tier the product. Ultimately a less capable DIY product would just be about the same as one of the lower tiers anyway, it would just be more work to make it seem like it's something different when it's really not.
 
On the what's a client and a driver, that should be covered in Try/Buy -> Information. If that's not clear enough let me know where it's not clear enough and I'll try to improve it. It's never going to be possible to just tell you exactly how many drivers you'll need because that's hardware dependent, so everyone's needs will be somewhat different. But, of course, as with a any business, if you have pre-sale questions, ask them on the forum and they will be answered. There's no way we are every going to make every possible detail of this type of product immediately apparent to you in a page or two, since everyone's needs may be different. That's what the forum is for, and all questions are answered quickly.