So this is interesting… Phillips pushed out a firmware update for both the V1 bridge and the V2 bridge which now locks out any bulbs other than those that are “friends of hue.” That means that GE, Osram, and Cree bulbs can no longer be attached to a Hue bridge. Nothing to do with SmartThings, this is a Phillips decision that they don’t want to answer customer support questions about devices that they have not certified compatibility with.
Some of this has to do with the GE link problem of their frequently showing as not available to the Hue bridge. Some of it apparently has to do with some other issues with Cree firmware. And the fact that some of the osram bulbs are color tunable which the Hues are not. Phillips wants their customers to have a “consistent customer experience.” Which is entirely their right when it comes to the bridge.
People complaining that they aren’t meeting the ZLL standard don’t understand the ZLL standard applies to the Hue bulbs. The bridge is a different type of device. Zigbee and zwave have always accepted that coordinator/controllers might have unique features as part of market differentiation.
Anyway, lots of complaints and concerns over on the Hue developer forums. And ST discussion in the following topic:
But I wanted to start this topic as a look at this as part of a trend, validated by HomeKit, towards separate ecosystems for IOT. I think we’re going to see more and more of this. The current “try everything and see what works” approach to compatibility is frustrating for consumers. The protocol certifications become a minimum, but don’t really tell you which features of which devices will be available with any particular controller.
In the last two weeks two different SmartThings employees have advised prospective customers to use only zigbee devices. Obviously, that’s not a company position, but I will not be at all surprised if a future V3 SmartThings is weighted heavily towards thread and zigbee devices.
I do expect IOT consumer sales in the low-end to fragment just like the mobile phone industry. People will choose an ecosystem and then stay within it. Like android versus Apple versus windows phone.
As I mentioned before, my expectation is that there will be HomeKit/Insteon versus works with Nest versus Weave Thread at a minimum. And probably several others as well. I do think it likely that Phillips Hue will work with all of these, but they’re going to limit the devices they bring with them to a very predictable, very stable, set with common features.
Indeed, I think it likely that all of the candidates with the exception of the weave thread group will have a limited set of devices and features. And if that will be OK with most consumers as long as the typical use cases are met.
We’ll have to see. But in terms of a future of the IOT industry, I think the Philips move is huge. The third-party certification organizations simply haven’t done a good enough job of defining a minimum set of required features for compatibility.
Personally, I think they needed to have defined More detail in the device classes, the way it’s done in other engineering areas. But that’s my background and my bias. I don’t think “multilevel switch” as a single category is sufficient.
I think if they really wanted interbrand compatibility, they needed to define probably seven or eight different types of wall switches, each supporting a different group of command sets/clusters. So consumers could know that their controller would support all of the features of an A, B or C switch, but not a D, E, or F switch. But the alliances didn’t do that. And consumers end up buying an even installing devices that don’t work the way they expect them to.
So now we’re seeing market pull in the opposite direction. Again with HomeKit as the big validator. And consumers accepting a very limited choice of devices, but knowing that those devices will work when they bring them home.
Philips’ willingness to step back from supporting other brands probably won’t hurt their sales much, particularly now that the hue white is priced so competitively. But I think it does say much about the importance of reliability in this market segment. We’ll see. 
@johnr @Sticks18 @juano2310 @twack @mitchp @tyler @infofiend @bravenel @pstuart @workmonk
