Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[JimH]

We're thinking about expanding our scope to Home Automation (aka IoT).  We'd like to hear what you think and what you've done in this area.

Thanks,

Jim

P.S. If you have something you feel is lacking in MC, please use another thread. 

[Al ex]

Just thought about it some days ago. Liked the idea to control my HiFi, TV, Htpc and living room lamps with a Harmony RC (I have). Discovered that this Phillips Hue is prohibitive expensive, not enough added value for me for the price.

I personally would rather be interested in additional features for organizing my A/V collection like sharing the info (not files) with friends.

[jjazdk]

Home Automation is a BIG subject, as in REALLY BIG...

From where I see it, what is needed is a Home Automation base/program/app/whatever that ties together all the different devices/apps.

In other words, I don't see JRiver as the "Home Automation Facilitator" but rather as one of the many possible devices/apps connected to the "Home Automation Base".

Making a "Home Automation Base" that will connect all the different clients/apps/devices together is no simple task and will require extensive and frequent updates, to be even remotely capable of supporting this wildly growing industry that at the moment seems to branch in way to many directions.

I would prefer if JRiver focused on what they are good at, and keep the focus on MC (new features and fixing bugs).

But if anything, I think the "JRiver Home Automation Division" should be separate from MC, and try to get buy-in from other leading brands in the industry.

Note: Presently I am using GhostEvent for my limited Home Automation, and I am not fond of the needed Python programming :-D

Idea (that popped up while writing this post) : Perhaps the JRiver Id could be a Home Automation Base, connecting everything else, but what about redundancy when it for somereason breaks down?

[CountryBumkin]

I would prefer if JRiver focused on what they are good at, and keep the focus on MC (new features and fixing bugs).

I would really like to see BD menus and 3D come next. I don't care about or use home automation.

[mpg732]

I use iRule as part of my home automation.  JRiver is controllable with iRule.

[ssands]

Home automation is massive - there are so many devices that need to be monitored. I have a small Control-4 setup at home. (I don't particularly recommend it, BTW). Comcast (my TV, Phone, ISP) made a "minor" change to their set top box. It was not announced. It broke my Control-4 automation - not horribly, but it was very annoying. My A/V guy had to remote in and update a script. What a pain.

Focusing on adding/improving the A/V experience and squashing bugs would be my preference.

[_mulder]

I think that home automation is a big thing and you`d lose focus on the force of MC (spread the energy too much and have even less improvement of MC).

[BillT]

Yep, it's massive and full of proprietary, undocumented, incompatible interfaces.

I use Homeseer, produced by a small software company with many similarities to JRiver. Their program attempts to be as universal as possible, and AFAIK is one of the best commercial offerings. Unfortunately, any useful HA system needs lots of technical involvement, either from a keen DIYer or an expensive professional in order to ensure that all the interfaces work correctly and that the automation rules operate correctly.

It aims at a continuously moving target. E.g. Z-wave is the interface that Homeseer concentrates on. The Z-wave spec seems to be in constant flux; Z-wave modules don't always adhere to the standards and old modules have different capabilities to new ones. It seems to be a constant battle to try to keep the Z-wave plugin bug free and compatible with as many modules as possible.

Another example; they wrote a JRiver plugin so Homeseer could control JRiver. Unfortunately, JRiver evolved so fast they couldn't keep up and abandoned the project.

User interface is another issue. Homeseer is a pretty sound rules engine, but it's interface is lousy. Designing good interfaces is another difficult problem, one which I'm afraid JRiver hasn't been amazingly good at!

The easy market of using a phone to switch lights on and off etc is likely to be picked off by Apple or Google. The market for proper automation is tiny and difficult to design for and implement.

For the keen amateur there are free alternatives which are pretty capable like Domoticz and Openhab.

[BradC]

Please give us the ability to run scripts from jriver (for controlling lights, amps, projectors etc)

Ideally there would be an option to run scripts as part of the zone editor (eg change a zone, run a script)

[JimH]

No feature requests, please.

[flac.rules]

I would really LOVE a bigger focus on home automation, and depending on the scope of the program would easily pay 3-5 times more for the licence if it is fully featured.

MC is the main program for user interaction on my computers, so having it being the "hub" for controlling other stuff is very useful

MC already has a command-system in place in conjunction with the remote part of the programs, just adding conditions (example: video is running) and the ability to run MCC-commands there would be a good first step. I am not trying to make a feature request per se, but I think this automation has 2 tiers:

Tier 1: MC being better to control what happens on the computer itself (where the scope can be from small, as the example above, to very large, like a full girder-style functionality, I think MC can be a lot better on this regard with some pretty small changes)

Tier 2: Controlling the hardware. I think this is more difficult, and also one has too choose what to support, there are quite a few protocols out there. This would be a dream come true for me, if the program supports my system, but quite a lot of work to implement i suspect

I have written some words about it here.
http://yabb.jriver.com/interact/index.php?topic=64305.0

Per this date I have a full home automation system with some 1-wire and Enocean, and Mainly Eib/Knx (Somewhat of an industry standard in Europe for non-radio-based systems, and also open as far as i know), it is controlled by a server with software called Loxone.

[willmcd]

I too think this is a great idea. I'd be interested in a little more detail on what JRiver is thinking so I could provide more useful input.

Like BillT, I use Homeseer at home, it talks to lights, stereos, TVs, heating, alarm. I just haven't had much success getting Homeseer to work with MC. There is a DLNA plug-in, but I haven't had much luck and haven't put the time in to get it working properly. Homeseer is another small organisation with a philosophy not too far away from JRiver and a large user community.

I'd like to be able to do a few things, for example...

1 Use Gizmo as my primary remote in the Home Theater. e.g Gizmo turns on the TV, receiver, maybe dims the lights etc, when a movie is selected.
2 Get MC to stream the radio into the bathroom in the morning when I get up, e.g. using the PIR to trigger MC at certain times of the day
3 Display a message on the home theatre screen if the doorbell rings etc...

My suggestion to JRiver is first to concentrate on getting it to be a great "thing" in the IoT. Once that works well, then perhaps MC could start controlling other appliances directly. The place to start to get MC into the market properly is to build some high quality plug-ins for the main systems so that they can work well together communicating in both directions.

This is an area that Sonos does reasonably well, which in my view has established them as the default system to work with Home Automation systems. Media Center with a low cost playback client machine, working really well with the major home automation controllers/software would be a compelling option with more functionality (video anyone?) and lower cost than Sonos or other multi-room audio systems.

[eezetee]

I'd like to ask a question to the question.

Do you think you've covered the home media center market enough to branch out? I have a couple of home automation systems, including one now owned by a very large company and they are a kludge of endpoints/devices are various manufacturers.  The minute you jump into that, you'll leave this product behind thus loosing your base.

A spinoff perhaps but keep the funding separate.

[JimH]

I'd like to ask a question to the question.

Do you think you've covered the home media center market enough to branch out? I have a couple of home automation systems, including one now owned by a very large company and they are a kludge of endpoints/devices are various manufacturers.  The minute you jump into that, you'll leave this product behind thus loosing your base.

A spinoff perhaps but keep the funding separate.

We don't intend to abandon JRiver Media Center, and I know there is more to do, but I believe home automation is a logical extension of what we currently do, and that many parts of the current application will apply.

If you're not happy with that, I can understand, but we had similar push back when we first ventured into video from our music beginnings, and it was clearly the right thing to do.

Remote control is key.  A single remote that can give you control of all your devices.  That's been my dream for a decade, at least.

I'm interested in opinions, but I would not like to see this thread turn into arguments for and against.  It's enough to say, "not for me" or "it's interesting".

[ssands]

Remote control is key.  A single remote that can give you control of all your devices.  That's been my dream for a decade, at least.

I'm interested in opinions, but I would not like to see this thread turn into arguments for and against.  It's enough to say, "not for me" or "it's interesting".

As I mentioned before, this seems like a competitor to Control4. I really like the idea (and generally, the execution, of Control4). It's a single remote and it can do all that you mentioned.

What I don't like is that I cannot program it. It has to be done by a certified person (or you buy an expensive license, etc.). I would really like Control4 if I had the option of programming it without being "certified" by Control4. Open access to a clear, robust scripting language (and debugger?) would allow me to tinker and revise until I had the control I wanted and the ability to add capabilities without having to call the tech and pay more dollars.
Also, it gets really tricky when your cable box has a minor upgrade. Mine did and I wasn't even aware of it. All of a sudden the right arrow in the menus stopped working, but the left arrow worked just fine. My cable box remote continued to work properly, but the right arrow on Control4 remote didn't work for the cable box/dvr, yet it worked on other devices.

This would likely cause a lot of churn on your support forums as people blamed JRiver for a bug in the latest upgrade.

Compound this will all the other connected devices (TV, BluRay, Amp/receiver, etc.) that are all getting somewhat regular software upgrades and you face a potential exponential nightmare of running down broken script issues.

I'm sure you are aware of this. I suppose I would be a very cautious, potential customer. I wouldn't want to purchase a new remote and software until I was really sure the product was solid and stable. I guess I'd take a wait and see attitude, but watch closely.

[DoubtingThomas]

.

[wburkett]

I don't care for the idea unless it is a completely separate product.  I would prefer that MC be very good at a few things than be mediocre at a lot of things.  (IMO, that's where iTunes went evolutionary-ly astray: trying to do too much within one App/UI.)

[syndromeofadown]

Porting to Linux and the JRiver id are both genius ideas. To be able to connect a 100 dollar device to any TV or stereo to make it JRiver compatible is amazing. As the hardware and the Linux version improve, it will only get better. It is a practical, cheap, awesome system. I don’t feel the same about IoT.

IoT has a lot of possibilities when used for infrastructure, healthcare facilities, farming, retail, and industrial plant sites. These things already have sensors in place that are already being monitored and controlled remotely so there is nothing revolutionary happening at the moment.

IoT has great possibilities for the physically and mentally handicapped.

IoT in the home feels like a big pile of marketing. I can see some people wanting to dim their lights and close their blinds with their remote when sitting down to watch movie, but that’s about it. I am happy to use my finger to twist the knob that is already on my wall. I am happy to walk to my window and pull the string that is already there. Americans, and Canadians, are in debt. A lot of people are just trying not to lose their houses, and a lot don’t even have houses anymore. Upgrading appliances to be connected to the internet cannot be of the slightest concern to the majority of people.

I just bought a new fridge. It is very high quality, expensive, and made in the USA. It doesn’t connect to the internet. In researching, the thought of IoT didn’t even come up. It is a feature that would have deterred a purchase. I would be interested in a fridge with a computer diagnostic system that automatically ordered a part that was about a fail so I could replace it with the help of the supplied video instructions when it arrived in the mail. In reality the fridge of the future will constantly send Kraft advertisements to the owner because Kraft has a partnership with the manufacturer. When the fridge breaks, it will be from planned obsolescence. The suggested repairman will be from a company in partnership with the manufacturer.

Maybe the predicted boom is due to the simple fact that manufacturers are going to shove a computer into everything. Partly for marketing, and partly for data mining. I suspect a lot of proprietary systems. “Use your Samsung phone to connect to you Samsung coffee maker”. Maybe Apple will buy out Kitchen Aid and Maytag like they did with Beats.

IoT could be great, but it won’t be. IoT will be about the bottom line, just another thing to shove down consumer’s throats. It’s 3D blu-ray all over. Something corporations want people to want.

I am curious to hear something more positive than my thoughts on the subject, and also what Jim has in mind.

[RD James]

I would prefer to see JRiver focus on automating and streamlining things within Media Center itself first.
There are so many things which could be automated that are currently a multi-step process.

[JimH]

Please don't make MC feature requests in this thread.

[mattkhan]

I use an RTi remote and have a similar point of view to ssands

It works really well when it is setup correctly but it takes a load of effort to make it really work slickly, the equipment is *really* expensive for what it is and you can't program it yourself. This last point is a blessing in disguise in some respects because I've seen the faffing around required to get the programming to just work with a variety of equipment, each of which has its own randomly idiosyncratic behaviour. I have neither the time nor patience to do this myself and I don't see how an abstraction layer is going to help when the underlying devices are the things that don't behave.

I am not optimistic that this problem can be solved without the scale of a samsung (or google or apple etc) behind it.

[JimH]

I am not optimistic that this problem can be solved without the scale of a samsung (or google or apple etc) behind it.

Maybe,  but I think it's fair to say that JRiver has had some success competing with media software from Apple and Microsoft, and a number of other big companies like Sony and Nokia haven't.  It takes more than money.

[mwillems]

I have neither the time nor patience to do this myself and I don't see how an abstraction layer is going to help when the underlying devices are the things that don't behave.

I think that's really the heart of the matter.  By analogy look at upnp/dlna.  There's a spec, but every device is different and many implementations just don't play nice out of the box.  For that reason there's a huge volume of support requests here that relate to DLNA, and sometimes the answer is just "the device can't be made to work the way you want."  I think the IoT is going to be that times 100.

My own experience of DLNA has involved experimentation folowed (in most cases) by giving up and just finding a way to run JRiver on the endpoint because JRiver works nicely with itself.  In Iot, though, that steam-valve won't be there (unless the Id expands in scope), so the user experience really will just be at the mercy of unreliable vendors. 

I would buy a JRiver smart thermostat (for example) because I would have faith that it would work, but I'm not sure I'd be interested in another control surface for other devices unless it "just worked" everytime, and that's a tall order when you're dealing with random hardware.

[JimH]

In general terms, what you're talking about are the idiosyncrasies of devices.  We know this subject well. 

What happens now, is that some smart person figures out what they are and gives advice to others.

What I think we can do is "codify" this special knowledge, so that it can be shared as a package of special instructions for setup.

For example, if you learn what the settings are that work best for a given device, you will save them to our cloud server.  Another person who has the device will be able to search for settings files for the device and download them to their copy of OneRemote (the phone app).

This is similar to what we do for cover art and for CD lookup.  User contributed data is made available to other users.

A step beyond would be to impose standards on device makers.

[mattkhan]

What I think we can do is "codify" this special knowledge, so that it can be shared as a package of special instructions for setup.

does this mean the macro functionality is in or out of scope for you at this point? I suppose I'm asking whether you're looking to be a (or "the") broker between the automation system itself (control4, rti, irule etc) and the devices or that you would be the automation system?

[JimH]

does this mean the macro functionality is in or out of scope for you at this point? I suppose I'm asking whether you're looking to be a (or "the") broker between the automation system itself (control4, rti, irule etc) and the devices or that you would be the automation system?

I don't know exactly where this will lead.  I wrote this a few months ago and it still seems close:

http://yabb.jriver.com/interact/index.php?topic=99054.0

Our starting point looks like it will be support of a Z-wave dongle.  Hopefully that will be possible to support on Linux, so we could use an Id-like device as a hub.

We're also working on an Android phone app called OneRemote.  It allows you to set up any general configuration you want, so you  could organize a push button menu by function or geography or ...  It has Categories (for organizing the buttons), Devices, Links, and Apps.  You can save the tree structure to a file and load it in OneRemote on another Android.  You can save a branch so one person could save their device(s) and make it available to another.  We're close to having a cloud server for it, similar to what we do for Doctor Who.

Larger image

[eezetee]

I don't know exactly where this will lead.  I wrote this a few months ago and it still seems close:
...

I have 40 z-wave devices and a few zigbee's using the Smarthings Hub. Solid solution at $100 for the hub and roughly $30 for the devices.  However, competition is suffocating margins for these hardware supplies. Ecolink sells a $20 window/door monitor that works a treat.  That said, I know JRiver took on Microsoft and is better

Wish you the best but i'd suggest keeping it as a plugin/option for those that want it and please don't detract from the excellent product (and future development) that is JRiver.  You've got a solid product here.

[JimH]

Thanks.  Could you name 2 or 3 Z-wave devices that you like?

[eezetee]

Hi JimH,

Note: I'm in no way affiliated with these companies, it's just what I use at my place.  

The EcoLink Door sensor is a is sub $30 item and works great. Open/Close only.
http://www.amazon.com/Ecolink-Z-Wave-Window-Sensor-DWZWAVE2-ECO/dp/B00OJMD2FA/ref=sr_1_2?ie=UTF8&qid=1451365003&sr=8-2&keywords=ecolink+door

The Smarthings Multi does Temperate, Open Close, Vibration and orientation -
http://www.amazon.com/Samsung-SmartThings-F-SS-MULT-001-Multipurpose-Sensor/dp/B0118RQW3W/ref=sr_1_1?ie=UTF8&qid=1451365103&sr=8-1&keywords=smartthings+multi

Aeon Labs Mini Mote - For setting status such as away/home or for triggering a light (4 buttons)
http://www.amazon.com/Aeon-Labs-DSA03202-v1-Minimote/dp/B00KU7ERAW/ref=sr_1_12?ie=UTF8&qid=1451365079&sr=8-12&keywords=Aeon+Labs

As well as a few power plugs and such.


Would be cool if once a particular status is set (Away mode) and motion is detected by Sensor A,  trigger JRiver to play a dog barking.

In fact, I've been working on my own tool to do that except I can press buttons and have various sounds trigger but not linked to JRiver at all

[JimH]

Thank you.

[dallasjustice]

It's a good idea.

I was an early adopter of Irule. Based on my experience I strongly believe there is a big gap JRiver could fill in the home automation market. The reason I mention Irule is that it's one of the best open platform home automation systems. It's like Crestron for DIYers. However, there is a huge short coming to Irule which no other manufacturer has resolved.  

IMO, tablet-only home automation isn't good enough. Tablets are bright, run out of battery power fast, are slow to use and require the user to take his/her eyes off the screen for too long.

Somebody needs to build a home automation system like Irule but designed for a customized hard button remote. Maybe a remote with both hard buttons and a small touch screen.  A small touchscreen could be nice for two way feedback communication over tcp/ip.(eg. Volume control setting, zone and even program guide data)

There is the NEEO. But that looks like perennial vaporware. I wouldn't buy into that. I believe JRiver could build a better mousetrap. Of course, an excellent product would also bring more folks into JRiver anyway but wouldn't require that JRiver be used. Global Cache products could be used to control devices over serial or IP.

Michael.

[brett]

I got into X10 years ago but it wasn't worth the trouble (at the time). Now I have ADT Pulse, which is a Z-Wave / Netgear based system. I've installed numerous in-wall Z-Wave switches, a couple of cameras, a front door deadbolt, garage door opener, etc. I even have the inside and outside Christmas lights controlled by Z-Wave modules. The system has been pretty reliable, but every once in a while a module goes-off line.

I also have a Nest thermostat, which can be controlled by its own app, the ADT Pulse app and my Logitech Harmony Ultimate remote (talk about redundancy).

I would not be adverse from stepping away from ADT Pulse, but the replacement would have to be Z-Wave compatible due to my investment in Z-Wave technology.

[JimH]

IMO, tablet-only home automation isn't good enough. Tablets are bright, run out of battery power fast, are slow to use and require the user to take his/her eyes off the screen for too long.

Somebody needs to build a home automation system like Irule but designed for a customized hard button remote. Maybe a remote with both hard buttons and a small touch screen.  A small touchscreen could be nice for two way feedback communication over tcp/ip.(eg. Volume control setting, zone and even program guide data)

I think we can leverage our current tools to be able to run on many platforms.  The same remotes should also work (phone, tablet, MC remote, etc.).

[rudyrednose]

I think what Home Automation needs before it really start to fly is standardization.  The situation is untenable if a IoT lightbulb will not talk to rest of the system because it is not the right “brand”.  Can you imagine the world if you could not count on standardized power delivery (120/240V AC) and you would have to match you devices to household power that could be 57V, 85V, 120V, 148V, 220V, 304V (and different from door to door) ?

It needs roles and APIs formalized.

First and foremost is the core, a Database/Scheduler that is redundant (3 to 1 majority logic ?) and low power enough to work several weeks on batteries.  The system HAS to be operable even during a power or other HW failures (hence the robustness suggested).  A failure here is critical.

Then there are “Renderers”, devices that do something or sense something.  At minimum it is a switch (sense or drive), then it can be a thermometer or an actuator or a dimmer or the HVAC.  A very complex Renderer would be the bidi control aspect of a MC implementation and it’s home theater peripherals.  Any single device can fail here.

Then there are the “Controllers”. At minimum it is a single wall-mounted button.  In between all forms of RF remotes, then phone, tablets and other screen based interfaces.  Any single device can fail here.

As mentioned, APIs need to be defined at the Database-Renderer and the Database-Controller level.  The system also needs to be flexible enough to allow for custom pipes between specific Controllers and Renderers for unstandardized/unforeseen/unavailable functionality.

As much as I like gadgets, I will not jump big time into home automation as long as a forklift upgrade is needed every time I change house controller system.

Happy new year everyone !

[flac.rules]

EiB/KNX is a standard, supported by many companies. It is reasonably popular but it has the same problem many standards have, namely, not everyone uses it. (And the US can always be trusted to do things their own way no matter what )

[robydago]

I had my successful attempt with IoT in the past:
http://yabb.jriver.com/interact/index.php?topic=91293.msg627805#msg627805

Since then, my interest in IoT is faded away.

But I think and hope that adding IoT automation to MC will benefit the automation of managing media in general (i.e. more scriptable and complex actions) even when no IoT devices are involved.

[mhwlng]

I have automated as much as possible and have as few buttons as possible.

I have only one (web based) control screen for everything :
http://www.flickr.com/photos/mhwlng/11728700773/

for example:

• no snooze/on/off button for the alarm clock.
• no on/off button for the central heating.
• no on/off buttons for the lights (only sliders for the setpoint).

The system knows when I am sleeping or not and when I am at home or not (phone detection and motion detectors)
so it will switch the central heating, lights, tv, stereo, alarm clock etc. on/off automatically.

some trend data :
http://www.flickr.com/photos/mhwlng/11728439385/
http://www.flickr.com/photos/mhwlng/11729200356/

[jjazdk]

Very interesting! Unfortunately the pictures are not available anymore!?

I have automated as much as possible and have as few buttons as possible.

I have only one (web based) control screen for everything :
http://www.fluidr.com/photos/mhwlng/11728700773/

for example:

• no snooze/on/off button for the alarm clock.
• no on/off button for the central heating.
• no on/off buttons for the lights (only sliders for the setpoint).

The system knows when I am sleeping or not and when I am at home or not (phone detection and motion detectors)
so it will switch the central heating, lights, tv, stereo, alarm clock etc. on/off automatically.

some trend data :
http://www.fluidr.com/photos/mhwlng/11728439385/
http://www.fluidr.com/photos/mhwlng/11729200356/

[mhwlng]

Very interesting! Unfortunately the pictures are not available anymore!?

I think there is some problem with flickr API.

you can try the same url, but with flickriver.com or flickr.com instead of fluidr.com

[wolffe]

I've been working on a universal devices isy-994i plugin off and on for a bit.... so i think this is a great idea.

i use isy-994i and insteon, along with a global cache device, to control the power to my amplifiers, control the projector, raise and lower the projection screen and house lights, and have even managed a "jazz" button, where the lights dim a bit, the amps come on, and jriver starts playing jazz streamed from the internet broadcast of a local station.

[fooze]

Jim, sounds like you want to make a standard and become the killer IoT app? Do it. I won't touch IoT until it's matured, but I think it would be fun. The main thing I'd like to use it for is power distribution, turning wall sockets on and off and such, then maybe controlling appliances.

Please don't make MC feature requests in this thread.

I really hope you're hearing that some people don't care about automation and fear damage to MC as a result of this.

I am worried about the internet of things because I may offer a guest a cup of tea but -
"Oh no, can't make a cuppa, kettle's broke."
"Have you tried turning it off and on again?"