barrygordon

Couldn't sleep so . . . First off, the code is unable to retrieve the version of the Bridge. That indicates that something has changed in the bridge. I will need to take a look at the BSW site. That is not fatal however. The code then tries to get the list of harmony devices from the emulator. From what I can tell it looks like the format that is returned by the emulator to a getDevices request has changed. The emulator returns a JSON string as it did before. After parsing the JSON object my code expects to see a set of un-named javascript objects making up an array. The objects now (according to the log) seem to be named with the name "device". I should be able to kludge up something based upon the log file you sent me and I will try and do that when I get back. It will be interesting and I like puzzles. Still think you should try and get away from the emulator but I understand your issue. The performance with the ISY connected home skill is very good, at least it is in my home.

leaving town for a few days. I will look at this when I get back next week.

What is the version of my program that you are running. I tried it with 5.0.66 but I no longer have any BWS emulators running on my RPi's. When it tries to get the BWS version it hangs for me since I have BWS emulators and I didn't bother to put in any timeout code. It should not be dependent on the version number, it should get a response. the version number checkwas for a change having to do with the complexity of the JSON returned from the emulator. What are you trying to do to have the harmony control the ISY? I know how to control an ISY over TCP/IP but I have no idea what the Harmnoy puts out. Does a Harmony put out TCP using WiFi? Is the a reference protocol document? I just re-vamped my iPad systems to use an RPi based proxy to do all IR work and some other things. The Proxy will take commands from the iPads or from the ISY networking resources and either control any number of Global Cache devices or other TCP based systems. I use the Proxy to control my pool with voice. I have put a lot of things into network resources so I can use voice control. I just installed an electric fireplace that can be controlled by IR . I built the correct network resource to tell the proxy to turn it on or off under control of a simple program. Works like a charm.

I assume that the reason you are still using the bridge is due to the Harmony capability. I have completely switched to the ISY connected home skill and am totally satisfied. Could you possibly use both? The bridge to deal with the harmony, and the ISY connected home skill to deal with the ISY? I don;'t see where amazon would care. The bridge looks like a Hue system, but the ISY connected home looks like itself. so I am not sure there would even be an issue. Amazon should handle multiple smart things. I have not been keeping up with what BWSystems has done with the bridge so I have no idea why the configuration system no longer works with it.

Right now all of my Echos and dots (2 of each) are working superbly. Response is as close to immediate as I believe possible with cloud based processes. Even the delay for the "okay" is minimal.

if you have done what jerlands suggested and it still doesn't work try the following: Try a different name. Make a program that turns on and off the appliance-linc and have the echo control the program. I had a lot of failures and issues with appliance-linc's and controlled outlets (prior to echo) and stopped using them. I now just place a micro-linc in the box with a normal receptacle and split the receptacle to two individual units by breaking the connection between the two hot screws.

It seems Amazon has cleaned up their act. Now if they will only leave it alone (fat chance). Names of music (e.g. Waterfall) now tuns on my pool waterfall as it should as in "Alexa turn on the waterfall". The action is completed in less than a second after I finish speaking the command. The "OK" comes within a second after the command action. Very nice, as it should be. I suspect that part of the improvement is in the UDI smart home skill. Kudos to all involved. My latest project involves removing a gas fireplace (in Florida?) and installing an electric LED unit. I am not planning on using the built in heaters. The unit has a remote which was easily reverse engineered (i have all the right tools) to determine the protocol (NEC 256) and the device and function codes. When it all gets done Alexa will control the unit naturally.

The issue with the Waterfall is not at the back-end (Autelis), but rather in the interpretation of the command . I also use Autelis to control my Pool/SPA. However I also use Prime music and Tunein radio. There is a song in Prime music named Waterfall and that is what Alexis tries to play when I say turn on the waterfall. Turn on the pool waterfall seems to be okay. I have a similar problem with the CNN channel on my TV. Turn on CNN seems to get interpreted as to play the Tunein radio feed for CNN. I have begun a conversation with Amazon regarding the use of secondary reserved words such as "Play" for music and not allow turn on to start music in the Selected Home Skill, similarly for radio stations. The first Barry

We need many to write in on feedback regarding the names of songs and the names of devices. The use of play vs t he use of turn on and turn off.

My pool has a waterfall feature. It was working fine with commands "Alexa, turn on|off the waterfall" it no longer responds to Alexa, turn on the waterfall, Alex choosing instead to play the song waterfall by Chris connelly. Amazon has to straighten out their verbs and restrict them to avoid this. They need to publish a list of action words and what they do. IMHO turn on / Turn off should not play music, but the action word "Play" should be used to play things such as music, videos, etc, perhaps with a modifier like music, movie, slideshow, video, etc. perhaps some day there will be a song named kitchen lights which should really highlight the issue. I have contacted Amazon on this but the squeaky wheel . . .

Yes the portal appears down. my.isy.io refuses to connect. The admin console (after a reboot) appears very sluggish but no busy messages, I will assume it is busy with the non-responding portal. Systems that interface to the ISY (e.g. my iPads) are working but very sluggish 4-5 seconds to do something such as turn on/off a light switches work fine. Alexa can not find any devices (logical as portal is down) I suggest that when the portal is discovered to be down that an email be sent to every registered portal user apprising them of the fact with an estimate of restoration time.

The ability to identify an Echo in the processing code is one of the top 5 requests by skill developers. One of the others is the ability to handle notifications. Notifications would allow the Echo to speak some message on request as either a code initiated item or as a result of some action like a mailbox getting mail. I suspect that someday both of these will be implemented.

a cold start or reboot resets all the firmware in the ISY to a known state. As with many software controlled devices this serves to clean up issues, at least for a while. As an example lets say a bug in the ISY connected home code/ISY code causes the ISY to get into an illogical state due to some garbage transmission. A reboot will put the ISY back to a well known state

I have a proxy server as Mwester describes that I use for many non-ISY endpoints that is triggered by ISY network resources. I generally use it for things I can see happen such as my SPA coming on or the pool lights, etc. One of the issues with handling a Sonos like device well is feedback. At the current time the Excho can not speed notifications sent by a user program so the only audio response the Echo would make is Okay doing it this way through the connecetd home. I assume you would like to know what is playing and that would take a skill written to handle the Sonos.

I agree, it was not the thunderstorm. When I rebooted my ISY the first time it was very sluggish and system Busy" just stayed up. I reset the switch and again rebooted the ISY. This time the ISY came up properly and when I tried the Echo it was fine.

Does Sonos obey bluetooth IR commands of the control type. This is commonly referred to as AVRCP The Audio/Video Remote Control Profile is a Bluetooth profile that allows Bluetooth devices to control media playback on remote devices. It is typically used with A2DP devices for next/previous track selection and pause/play functions. The DOT handles AVRCP. I am not sure if the original echos do, but they probably will in some future update.

Woke up this morning after a late night and found the Echo's not working. Got the "Sorry the device XXXX is not responding. Please check its network connection and power supply." message on both echo's. Tried a reboot and that did not clear it up, in fact the ISY was having a hard time starting up. The ISY was acting very sluggish with "System Busy" being displayed. I rebooted the network switch that the ISY is connected to and then rebooted the ISY. Gave it a few minutes and started the Admin console. The ISY came up normally and was its usual responsive self. Tried the Echo's and they are now working as they were before. Bottom line: I suspect the network switch got hosed up during some thunderstorms last night even though it is sitting on a UPS as are all the electronics in the house. There is nothing foolproof in the world of software driven devices, stuff happens. As along as it easily recoverable I am okay. Not happy, but okay. For information purposes; I have 99 connected devices and all voice operations start with the two Echos which deal with the ISY. I have a Proxy that runs on a RPi that handles special network commands from the ISY that are triggered by the Echo via a set of simple programs. For example "Alexa, turn on the Kitchen T V" will cause the ISY to send a simple "DOIR House,1,ITV,Power" command to the proxy which then sends the appropriate timing strings to a Global Cache named "House" using the first port of that GC device. The "power" entry in the timing tables for the device ITV (Insignia TV) will be used. The Proxy handles my Pool/SPA control system and my Home Theater in a similar manner but not via IR but rather TCP. The Proxy also handles the Skill I started to develop, but stopped as the Work by ISY to implement the connected home now handles about 98% of my needs. One of these days I will get back to finishing the Skill (Sarah). If you are interested in the Proxy drop me an eMail and if there is enough interest I will write up a detailed User document and make it available. BEWARE: To use it for other than IR, you need to be able to deal with javascript and using TCP or whatever your end point device needs.

Yes, you are quite correct re older vs. newer devices. My TV sets (other than the one in the theater) are all "older" and have IR control as opposed to IP. Also the set top boxes (other than the TiVo's) use IR control. I had a need to do IR control and as time passed i have migrated to different types of "Remote Control systems". I have always used a Proxy for IR so I didn't have to put timing information in each control system. This latest effort just migrated the proxy from running on a PC to running on an RPi. I just saw an announcement re the Raspberry Pi 3, a 1.25 Ghz processor with 4 cores. I will probably pick up one of those just to run Kodi. At $35 it can't be beat. Being a heavy Javascript user I wanted to get some experience with Node.js. The combination of Visual Studio 2015 (free) running on a Win 10 platform for development and testing with production running on a headless RPI exactly met my needs. The proxy as it exists now runs synchronously for its setup phase where it locates the Global Caches and loads the IR timing tables and macro data into memory. It then switches to asynchronous operation for all the TCP/IP work so its CPU load is very low. Was an interesting learning experience. I once again learned that Google is my best friend.

The IR_Proxy handles the Macros. To properly do Macros (a sequence of sent commands) with the Global Cache you need to use the feedback that the Global Cache supplies when an IR command has been fully sent to the end component. This provides the minimum pacing between commands in a a macro sequence. The ISY at the current time cannot do that so you need to somehow introduce delays which are either too long or a best guess. The IR_Proxy sends a command to the GC and then waits for the response before it picks up the next command in the queue. The Macro design also allow for inserting additional timing (delay) between commands if needed by the end component. Hope that helps.

I have a large library of all my IR controlled devices plus many I do not own. Using the IRgen program it is just drag and drop to build the necessary timing files for a Global Cache. Building these files (called irp files) is not hard but you do need the protocol details of the remote unit that generates the IR. This information is available on the web for many files. IRgen will also accept Pronto Hex format and generate the Global Cache timing files. It includes a full test capability to any GC unit it can reach, and the ability to generate macros. I find it much easier to type in something like "DOIR House,1,STB,Select" then "sendir,<connection code>,<frequency>,<repeat>,<offset>,<on 1>,<off 1>,<on 2>,<off 2> . . . I have excellent reliability on all the PC's in my home. I have started using "headless" Raspberry Pi's (no keyboard, no mouse, just power and internet connections) for things I need 24/7 and have been having perfect up-times. They just do not go down. I am doing all my RPi work in Node.js using Visual Studio 2015 on a Win 10 machine for development and test. I am confident that the IR_Proxy on a RPi will run 24/7 with no issues once all of the "bugs" are out. It is very stable now and I have tested it fairly heavily. On my Win development system I run WINSCP and tiny VNC to deal with the RPi's and have no issues doing anything I need to from the comfort of my home office. I am currently triggering commands via Echo. Hope that answers your questions.

I have just finished an interesting application for the ISY. It is in two parts, one part IR_Gen runs on a Windows machine and builds the correct timing files for a Global Cache unit. The input is the protocol information for the device in a notation known as .irp, or the pronto hex codes for the device. It also has the capability to build Macros from the various timing files that have been constructed. The second part, IR_Proxy runs on a Raspberry Pi and acts as a proxy between the ISY (or any other device that can issue raw TCP commands), and any number of Global cache IR units you have on the LAN. IR_Proxy uses the timing and macros files produced by IR_Gen. IR proxy accepts a simple English language text string to Identify the Global Cache unit, the port on that unit, the device file containing the IR timing information and the specific command from that file to be transmitted. As an example lets say I have a Scientific Atlanta set top box I have chosen to name STB as a device on a Global Cache unit named 'House' connected on IR port 2. Then the ISY can send the TCP string "House 2,STB,Guide" to turn on the guide. The string "Macro, House,2,NBC" will send the NBC macro out on port 2 of the House Global Device changing to that station. I am just starting the documentation phase for the system and should be done this week. I am interested to know how many ISY users would be interested in a copy of this system. Installation on a RPi is not difficult but not trivial either. I will try to supply enough information for a non-linux soul to be successful.

If I understand what you are asking search this thread for "math" the emulator allows you to do simple math on the value like doubling or halving it.

Yes, I am using different sequences of commands at different times so a scene or a fixed program does me no good.

My only issue is with How Amazon implemented the connected Home. If you wish to give several commands to Alexa in a row, then the delay before the OK is a RPITA. If Amazon could close out the session by default a lot quicker and eliminate the "OK" perhaps multiple commands would not be such a PITA. The performance for a single command from end of speech to action taking place is quite good, almost perfect. In my house it is less than 2 Seconds, and often less than 1.

I have noted a significant improvement in response time between when I finish speaking and the action takes place. Good Job!! Unfortunately the time it takes for the "OK" to came out is way too long, but I do not believe that is your problem but rather Amazons and related to the number of network interactions needed. If Amazon were to provide an option in Alexa app to short circuit the "OK" and have Alexa ready for another command as soon as the action has been sent that would help. Perhaps even the inclusion of "And Then" at the end of a spoken command to make Alexa quickly ready for another command. I know that is more like a skill as opposed to the CoHo, but . . .