Wink gets a poke in the eye

NeverDie said:
  After all, if there is going to be an IoT (admittedly, a big assumption), what's it all going to connect with? 
 
What do you mean?  And I don't think that Wink belongs in what we call the IoT.
 
The Internet of Things is nothing more than Internet Connected devices that connect via APIs.  It is IFTTT, or Works with Nest - the entire point of it is that there is no hub necessary.  The devices can communicate directly over their API.
 
The best example of this is Harmony Hub integrating with Hue, which can integrate with Nest and Dropcam, which can connect with August.  All of this stuff can happen without a hub.
 
A lot of what is driving the cloud idea is companies wanting to collect monthly fees for making it work. It also think it makes accessing the data easier (not necessarily better) than having to deal with dynamic IP addresses, holes in firewalls, etc. But the primary reason is the monthly revenue. I don't want my stuff on the cloud.
 
JimS said:
A lot of what is driving the cloud idea is companies wanting to collect monthly fees for making it work. It also think it makes accessing the data easier (not necessarily better) than having to deal with dynamic IP addresses, holes in firewalls, etc. But the primary reason is the monthly revenue. I don't want my stuff on the cloud.
Most of the hubs out have no monthly fees. The Wink doesn't. There are a few based this way, but many aren't.
 
jkmonroe said:
What do you mean?  And I don't think that Wink belongs in what we call the IoT.
 
The Internet of Things is nothing more than Internet Connected devices that connect via APIs.  It is IFTTT, or Works with Nest - the entire point of it is that there is no hub necessary.  The devices can communicate directly over their API.
 
The best example of this is Harmony Hub integrating with Hue, which can integrate with Nest and Dropcam, which can connect with August.  All of this stuff can happen without a hub.
I'm not understanding what you're trying to say.  Are you asserting that  communication is a non-issue because everything could, at least potentially, communicate via WiFi?  Or are you asserting that some form of ad hoc networking would obviate the need for a hub?  Saying that devices communicate using API's seems to assume the communication issues are already solved.
 
HomeSeer's approach is simple. We offer locally managed intelligent controllers that leverage a cloud service (MyHomeSeer-Remote) for remote access only. This protects user information by storing it on the controller (in the home) and insulates the home automation and access functions against internet outages.  If our cloud service is interrupted, local access (via home network) still works. Beyond that, users can also employ a dedicated DDNS setup as a backup or if they prefer not to use our cloud service.  Any company can produce an update that's a lemon; we know! But.. devices like the Wink hub (and many other new "smart hubs") that rely completely on the cloud (for access and for automation) are particularly vulnerable to outages and firmware problems. 
 
Our HomeSeer IFTTT channel extends our reach to products and services we don't have native drivers for. However, this level of integration is also vulnerable to cloud problems. For this reason, native drivers should be used for "mission critical" functions and IFTTT should be used less important functions.
 
Apparently this Wink thing is a certificate management failure.
 
People are working around it by setting DNS settings such that their devices are pointing at servers which have had their clocks rolled back.
 
Had this happen to me with an early e-commerce site. Cost the vendor a great deal of money, as it went down over the Thanksgiving weekend.
 
NeverDie said:
I'm not understanding what you're trying to say.  Are you asserting that  communication is a non-issue because everything could, at least potentially, communicate via WiFi?  Or are you asserting that some form of ad hoc networking would obviate the need for a hub?  Saying that devices communicate using API's seems to assume the communication issues are already solved.
 
I'm saying that there is no benefit in thinking about it in terms of the device you have in your home.  Your device connects to its own API - in the IoT, the hosted API's talk amongst themselves - from Google Servers to AWS to Azure, or wherever the devices API resides.
 
The Internet of Things is nothing more than the API's talking to each other in the cloud, and not directly with your device.  This is what you get with Works with Nest, IFTTT, and what I assume is going to be HomeKit.  If you lose connectivity to your device in your home, anything that you've configured via Works with Nest is still going to attempt to operate, and when your device comes back online, stuff will start working again.  
 
Wink is trying to break that by placing itself in your home and bridging devices to the cloud.  It goes a step farther and connects those various IoT device APIs to the Wink API directly, which may turn out to be their fatal flaw.  
 
Like this: http://imgur.com/qI504GV
 
jkmonroe said:
I'm saying that there is no benefit in thinking about it in terms of the device you have in your home.  Your device connects to its own API - in the IoT, the hosted API's talk amongst themselves - from Google Servers to AWS to Azure, or wherever the devices API resides.
 
The Internet of Things is nothing more than the API's talking to each other in the cloud, and not directly with your device.  This is what you get with Works with Nest, IFTTT, and what I assume is going to be HomeKit.  If you lose connectivity to your device in your home, anything that you've configured via Works with Nest is still going to attempt to operate, and when your device comes back online, stuff will start working again.  
 
Wink is trying to break that by placing itself in your home and bridging devices to the cloud.  It goes a step farther and connects those various IoT device APIs to the Wink API directly, which may turn out to be their fatal flaw.  
 
Like this: http://imgur.com/qI504GV
Thanks for the explanation and for sharing the picture.  That's an interesting approach,  To date I've been imagining that IoT happens through a lower level framework like MQTT, which a few of the upstart home automation efforts are trying to leverage.  Aside from that, and now the paradigm you're referring to, I hadn't heard much as to how IoT might be actually accomplished.  Whenever I tried to look into it, the specifics always seemed to get glossed over with a lot of handwaving, and so I thought it was all very much still in a state of flux, waiting to get pinned down.
 
Yeah.  Obviously the space is still changing, but going forward I think this will emerge as the ideal way to interact.  Although the US is also at a slight disadvantage because our infrastructure is so latent (thanks Comcast).  For events which are not immediately noticed, like a closure triggering a light, the latency can be easily dealt with, but it really falls off the rails for immediate requirements.  
 
This is why I think Wink is so intriguing - it is cheap, it has radios for popular protocols, and it has an open API.  So you can get the benefits of behind the scenes IoT and you can bridge your local RF devices to become IoT capable.
 
And if none of that is for you, just root the Wink and hook it up to OpenHAB and get a $49 device that has ClearConnect, Zigbee, Z-Wave, and Kidde radios for your automating needs
 
:)
 
Samsung and the SmartThings hub is listening to feedback. Then next version out will completely run locally, and it will even have a battery backup should the power go out. Give these companies a break, most of these hubs have only been out less than a year. Sure there are problems, but I wouldn't write them off so quickly just yet. 
 
Just look at some pricing. I just bought a Leviton Zigbee Interface Module which is basically a Zigbee radio with a serial connect. Cost $200 
A Wink, with Clear Connect, Zigbee, Z-Wave, Bluetooth, and Kiddie radios costs $50.
 
I'd be curious as to whether you can deploy multiple winks and have the devices associated with each interact via your cloud account.  That would overcome some zwave's reliability problems.  For example, I'm presently using four z-trollers networked over ethernet, and it was a huge improvement.
 
You can do that with multiple Homeseer Z-Net Z-Wave + devices POE powered and sticking them anywhere.  You can also modify the antenna on the GPIO board in the Z-Net device.  Personally here I installed my one Z-Troller on the second floor of my home and talked fine to three floors and peripheral edges of property (berms all have 120VAC outlets).  Also did the same with the Leviton / HAI Z-Wave PIM (except it was in the basement between conduit adjacent to the fuse panel) Tested Zigbee to work fine on a metal rack in the basement using same scenario mentioned for Z-Wave. 
 
Z-Wave + is better than Z-Wave. 
 
z-stuff.jpg
 
Personally playing with one Homeseer Z-Wave + device and I like it better than the Z-Troller. 
 
That is me and my personal opinion though. 
 
Best to see for your self; helps adjust for the abundant flow of rhetoric sometimes.   (put that pragmatic hat on for bit)
 
Long term (mañana).....
 
I believe that the Homeseer Z-Net device will be a automation multiwireless "doo what".
 
RF is RF is RF.  Might as well just have everything wireless in one box anyways; it will be easier to manage.
 
This is a copy and paste from Sigma Designs dot com.
 
Some of the new features of Z-Wave Plus are:
  1. Greater Minimum RF Range requirements for all products
  2. Better defined expected behavior for more consistent user experience
  3. Better defined technical capabilities enabling more reliable and easier to set up networks
  4. Maintains backwards compatibility with previous generations of z-wave products
Now for a mix....a copy and paste from the Securifi site...
 
The other thing that differs between ZigBee and Z-Wave is the transmission range. We’ve generally found ZigBee to offer better range than Z-Wave, regardless of the higher frequency which should technically mean shorter range. Z-Wave is in general rated at 30m/100 ft vs 100m/330 ft for ZigBee, both in an open space. Both technologies allow for range extenders to be used. Most, but not all, mains powered Z-Wave and ZigBee devices can act as a signal repeater. Z-Wave has a limitation of four hops, whereas ZigBee can go as far as 30 hops, although the more hops, the longer the latency, or delay, from the signal to go back and forth between the Almond+ and the sensor/device. This is something to keep in mind if you have a large home, or want to be able to reach sensors outside of your home.
 
pete_c said:
Z-Wave + is better than Z-Wave. 
 
attachicon.gif
z-stuff.jpg
 
Personally playing with one Homeseer Z-Wave + device and I like it better than the Z-Troller. 
 
That is me and my personal opinion though. 
 
Best to see for your self; helps adjust for the abundant flow of rhetoric sometimes.   (put that pragmatic hat on for bit)
Thanks for the tip.  In your experience with it so far, what do you like the most about the zwave+ device versus the z-troller?
 
ano said:
I just bought a Leviton Zigbee Interface Module which is basically a Zigbee radio with a serial connect. Cost $200 A Wink, with Clear Connect, Zigbee, Z-Wave, Bluetooth, and Kiddie radios costs $50.
 
What is the Leviton Zigbee Interface Module? Is that basically the Zigbee equivalent of a VRCOP? If so, that could be a very useful thing. Does it come with the HA profile on it or something like that?
 
BTW, I would just say that price is not everything. Leviton isn't perfect, but I'm guessing their product is likely more solidly built and engineered. You do tend to get what you pay for. RA2 is not cheap, but it works without drama.
 
In your experience with it so far, what do you like the most about the zwave+ device versus the z-troller?
 
It is faster. (everything related to Homeseer/Z-Wave+)
 
That is all right now as I am playing with it.  Made a gizmo using multiple power strips for quick add/Include and remove/exclude nodes. 
 
I went to speedmode remove/exclude nodes though using the Leviton RF-Vizia device and trying to break it.  I did not so far.
 
Personally started to and then didn't migrate from the Z-Troller.  Started from scratch.  Low on the WAF a bit as my home office floor is covered with Z-Wave devices except for a path to the door right now.  I started by filling up one bathroom and that was really low on the WAF.   I have done the remove/add thing now a few times; erasing everything initially (my preference).
 
I also keep moving Z-Wave + from one place (catxx wall plate) in the house to another. (because I can and nothing I am testing right now is "production"?)
 
My preference though is the Leviton/HAI way with the RF remote uploading to the serial Z-Wave PIM; well I got used to it and its faster for me. (speedmode wise).  I don't see the granular details unless I look; well and I don't utilize reading glasses but have issues with the LCD display being a bit small.  Almond + / Almond 2015 LCD is a bit bigger but not portable. (no comments on my Vera stuff cuz never spent much time with it).
 
What is the Leviton Zigbee Interface Module? Is that basically the Zigbee equivalent of a VRCOP?
 
No.
 
ZIM.jpg
 
It is a serially connected LCD display device combo / Zigbee radio device.  It is called a ZIM (Zigbee Interface Module).  You can put it on your coffee table if you want. 
 
The Intel Atom based OpenPeak tabletop utility devices (2009) were sort of like this except that the original ones have an HD 8" capacitance touchscreen, web surfing, games, movies, Zigbee (electric meter + thermostat stuff), Gb, Wireless, Bluetooth, DECT radio and a SIM slot interactive connected to the cloud kiosk like touchscreen device running a Linux flash based OS sort of similiar to the Chumby's (neato bedside clock radio) but on steroids. (ran EFI booting OS on mmc, ZIF pata or USB)
 
I have not read about any issues with the ZIM here on the forum; well maybe one relating to the Zigbee Omnistat (polling?). 
 
Back
Top