Philips reverses decision to close the Hue Platform
developers.meethue.com
developers.meethue.com
They had the solution available to them from day one. Since they can clearly identify third-party bulbs, they could have simply presented a warning along the lines of "We've detected you're using bulbs that are not certified by Philips. For best results, we recommend using only certified bulbs (link to purchase here) and cannot guarantee a quality experience with the bulbs you've purchased. Click "OK" to continue."
https://developer.apple.com/library/ios/documentation/Networ...
Would they? Wouldn't they blame their bulb first? It's like complaining to your PC manufacturer that a program you downloaded doesn't work. Most people would blame the program and look for another one. I'm sure there are some who have it backwards, and they probably will be calling support, but I can't imagine there's enough of them for a company to actually care about damage to the brand those people may be doing.
This happens far, far more often than you would think. That line from Men in Black -- "A person is smart. People are dumb, panicky animals." -- rings more true every time I work with a client who handles front-line tech support.
I do have to wonder how much of a difference it would make right now, though. Hue is presumably still in the early adopter stage (especially if we're talking about people who are adding non-Hue Zigbee devices to the setup), where you'd hope this kind of confusion would be less common.
Edit: this is one of the primary reasons Apple exercises so much control over third party integration on iOS. They get the blame for everything.
What remains is a mix of "dumb, panicky animals" and people for whom calling is the best option (they may be in a hurry; or for many older people, phone call is the method they're used to and one they trust). It's something you have to deal with, it's for them that you have support in the first place. But for a company like Philips, I can't imagine those support calls make any difference brand-wise. Most companies I know could drop their private-customer help lines altogether, having them respond with busy tone all the time, and nobody would likely notice.
However since I can't back up my original claim I'll leave it at that.
I think it's a smart move and reasonable compromise, similar to Microsofts WHQL program.
But I used a different analogy on purpose. You bought a Hue Bridge - presumably, with at least one Hue bulb (do they even sell Bridges separately?). So you know it works. Then you buy a noname cheap bulb, plug it in, and it - the cheap bulb - doesn't work. The Bridge still works, the Hue bulb also works, it's just the new, different one that doesn't. I find it hard to believe there are many people who can't connect the dots.
It doesn't surprise me at all. The key to remember is that people are not rational creatures. They work off of emotions and then apply logic to try to rationalize their emotions.
lol, bro ... have you worked with humans like, ever?
Usually $20 for the Philips vs $15 for the GE Link at 800 vs 750 Lumen :)
I only have the colored Hue lights and the white GE Link so far. I might buy a Lux just to see how it works.
It took especially long since I had rewritten most of System.IO to support 32K paths in .NET so I assumed that I had subtly fucked something up, it was only when the SQL Server started crashing that I even considered drivers.
Have you met "most people"? They would blame PC manufacturer indeed.
Just remember Vista and how suddenly all kinds of software stopped working. Surprise: most of the time, this software used undocumented features and sometimes even bugs present in previous Windows systems.
I remember Vista (ironically, I'm probably the only person I know who didn't have a problem with it and, besides the slow file copying, geneally liked it) - but there, in a way, some blame towards MS was justified. They were releasing a next iteration of their operating system to the already mature software market that's built on interoperability. Microsoft itself had a strong tradition of caring about backward compatibility. With Vista, they failed to provide an OS compatibile with existing application ecosystem, even if it's third party developers who didn't follow the specs.
Here, on the other hand, the "bulb" ecosystem is only starting, and Hue is the reference platform.
Just do an XP fork for VM to run all the legacy stuff on and start fixing stuff already, damn!
Oh yeah. I can just see "most people" powering up VMs to run their outdated broken software that they depend on.
Why I'm sarcastic? Because the situation is completely broken, there's no right solution. All that's left is to ridicule all sides of the argument, because everyone who thinks that there is a way to make it right is wrong.
The core question is how much isolation optimizes future development while maintaining the ability to run older applications.
You are a typical consumer. You buy a HUE system because it has the best reviews. Then you realize you need lightbulbs as well. You go on amazon, look at your options. You can get the Phillips ones, but they're too expensive, so you buy the other ones.
You pair the bulbs. It's complicated, but you just follow a guide on the internet word-for-word to get through it.
A few months later, your system starts acting wonky. Regardless of what's actually causing the problem, the reality is that you now have to go to the trouble of figuring out what is wrong with your system and then fixing it.
Your experience with the Phillips Hue has been soured.
> A few months later, your system starts acting wonky.
What if "your system starts acting wonky" means "Only the non-Phillips bulbs stop behaving correctly."?
Growing up, I knew of many, many people who would only stick with brand-name things because of quality or compatibility concerns. I would expect that many affected users would be able to recognize that their non-Phillips bulbs were acting up and their Phillips bulbs were not.
If I had a dime for every time I heard someone say "This _____ing computer!" every time a program crashed...
"If Bill Gates had a nickel for every time Windows crashed... wait a second, he does!"
Considering how many people get mad at individuals who are named Isis, as if they have any relation to the IS, I wouldn't doubt some of them blaming Philips.
What has to be remembered is that the types of people who read Hacker News would understand, in more detail, what might be causing the issue and know troubleshooting is all part of the process. I bet printer companies get many calls a day from people who bought third-party cartridges (sometimes without realising) and complaining that their stupid printer isn't working and that it must be the printer's fault.
The vast majority of consumers who would walk into an Apple Store or Best Buy to purchase something like this, they just think of it as one big ecosystem. If it doesn't work with a bulb they bought off the internet, they will simply assume the product, as a whole, is terrible.
The first question they asked on the phone was whether I'd used a 3rd-party cartridge. I said yes.
The second question was where I'd like the free replacement printer delivered (now with added WiFi, and a full set of cartridges). Painful for Dell, but I'll buy from them again.
I remember discussing with a website support guy who would get calls about printers not working, screen flickering and claims about some other unrelated websites that he didn't even know existed.
There is of course tweaks that can be made to a product to reduce the amount of support calls, but misplaced blame should be par for the course whatever you do.
Not disputing that some printers just refuse to work with third-party carts, but it's not the case with most printers I've used.
People calling tech support are a minority of users and do not represent the full market. It's really strange that smart people here are actually proposing making the product worse due to some loud incompetent individuals.
Selection bias. The people coming to the genius bar obviously think there's something wrong with the computer.
We tech-savvy people have a better understanding of things like compatibility, the drawbacks of third-party/unofficial accessories etc. But try explaining that to someone who has no interest or experience in this, and they'll just not understand - or care.
If Philips had stuck to their decision, it would've meant that, for anyone contact support, they could handle the entirety of the problem. This avoids the boomeranging of "well it might be your bulb, go speak to the people who made it".
I don't know many people who use home automation, lighting systems like this, but of those that do - all but one is not someone I'd consider tech-savvy.
I have no strong feelings one way or the other about this decision. Honestly, if I ever used this, I'd probably stick to first party products. As I get older, I find that I just don't care or want to spend the time fiddling with compatibility issues. If I'm in the market for automated light bulbs, chances are I'll have no problem stretching my budget to first party bulbs.
Steve Jobs' early instinct to deliver a walled garden experience to consumers was correct. My guess is we'll see a widening bifurcation of the market, where consumers go ever-deeper into a walled garden, while professionals/commercial purchasers pay much higher price points for more accessible, manipulable gear. Once the complexity of some commercial gear reaches a certain threshold of support costs however, I expect that gear will head towards black box, closed designs. Maturing the software we write so it tames the complexity and makes managing it fun/ego-engaging in some manner could go a long ways towards pushing back that trend (if it emerges in the way I think it will).
Factually I think we should forbid selling devices we are forbidden to touch and avoid the whole issue.
As semiconductor fabrication processes get smaller, die sizes larger, and operating frequencies higher, I think we will see the hardware become more opaque. Already the test equipment to hack on cutting edge workstation- and server-class processors is spendy; just a 2GHz 10GS/s oscilloscope is 5 figures USD $ even used and just-calibrated. As manufacturers head towards greater integration on the die to differentiate themselves in a furiously- and constantly-commoditizing industry, we'll likely see more binary blobs at first, then market segmentation hits those of us who want open/powerful gear with a double-whammy. We enjoy the fruits of commodified hardware components now, but I'm not sanguine that state of affairs will last indefinitely due to the aforementioned integration trends.
As everyone noticed that personalized gear enjoys outsized margins, we're seeing ever-increased pressure to further integrate and miniaturize into smaller packaging; repairability and accessibility to the hardware not invited at all to the design party. I really wish this specific trend would reverse.
I suspect that the PC as we know it will morph into a greater walled-garden device in the next several decades, and if we want open/powerful gear we will find ourselves in a tiny market segment. At best, paying very high prices for the small volumes catered to, and at worst, in a technology ghetto.
For now, I agree with you, but the long-term constraints are not promising to me further openness of the hardware.
There was a phenomenon I called 'first blamer advantage', whoever the customer called first whether it be the NIC maker, Microsoft, their ISP would blame their problems on someone else and the customer would think whoever was blamed was at fault.
Usually we'd blame the NIC company, and then offer to send a tech out to fix the problem we blamed on them as a courtesy which made the customers accept the ridiculously long time it took to get a tech out.
Yes.
>Wouldn't they blame their bulb first?
No.
>It's like complaining to your PC manufacturer that a program you downloaded doesn't work.
Yes, and people do that all the time.
>Most people would blame the program and look for another one.
Bro, do you even tech support?
(I speak from experience. I was able to track a surprising number of bug reports in my apps to internet filters, kernel extensions, etc. It's so common that it's the one of the first questions I ask customers when they report a bug. And it's why sandobexed apps are so successful -- all of the sudden you can install apps without worrying that they'll break your computer)
I gotta disagree. I tend to find "normal" people will blame the manufacturer or microsoft whenever anything goes wrong regardless of the actual cause.
We really need to stop compromising our tech just because some loud minority refuses to understand how things work.
I don't disagree with you -- a subset of customers are going to be angry regardless of warnings, they only go so far. But the issue at hand is how best to deal with the problem they have publicly stated was the reason for this decision: Some subset of third-party bulbs do not work ideally in their ecosystem. Option #1 was to warn customers of this problem and let them decide. There are several other options that were available to them. The farthest down that list, and worst option they had was to disable the bulbs. I would expect that last option to be used only for bulbs that caused fires or other safety issues. That wasn't the case (or I would have expected them to have stated that directly since it would have been something a lot of people would have understood). I don't think Philips realized how major of a feature this is, particularly to early adopters.
http://www.geek.com/wp-content/uploads/2014/01/Hue.png
See those warning marks on the middle screen? I'm not sure what they do, but this is where I'd put them, along with a small text "this bulb is not supported by Philips and may not work properly", and people wouldn't miss it. It would be a constant reminder for them that if something breaks with one of those bulbs, it's their fault.
[1] Customers are people with an important relationship. Consumers are things with a behavior, e.g. my toaster consumes electric power. When people are seen as consumers companies do dumb things...like communicate with utter bullshit.
Turns out people don't care that much.
And all can search with Google!
The Philips Hue line is pretty clearly a premium product. What they're selling is a high-quality experience.
Warning boxes are easily clicked and forgotten. Or if they're clicked by somebody you've hired to install a system, you'll never even know. If people have irregular, unsatisfying experiences, they will associate it with the Philips-branded interface they're clicking on, not the off-brand lightbulbs whose labels are tucked away in ceiling fixtures.
Personally, I congratulate them for starting their system open. The most brand-friendly and wallet-friendly solution they could have chosen was to close it from day 1. But they started open and stayed open until it became a problem for them. Closing it suddenly and cutting people off was a mistake, but a certification program seems like a reasonable outcome.
I can argue that the value of their brand is being eroded when the company is openly hostile to interoperability with other products, and of course customers will have shitty experience because this systems is stand-alone and can't be connected to anything or used in any way except for one specific case.
And most companies usually pay per call/minute to their customer care contractors.
The average Philips Hue customer is an early adopter who was willing to spend a large amount of money on a next-generation lighting system. This customer is an early adopter.
The question is not whether they should allow third-party bulbs, they already allowed third-party bulbs. The question is whether, after reaping the benefits of creating a product with compatibility built in, if it is a good business move to remove that feature via a software update.
People flipped out when their paid-for Kindle copy of 1984 was removed from their e-reader due to legal issues that were out of Amazon's control. And those customers were refunded. They chose to break the lighting of their customers due to user experience issues.
That move got the hardware hackers attention, and the PS3 copy protection was cracked in 3 weeks.
Hubris, meet Nemisis.
Every single comment that conflates Philips Hue users with the average barely-literate technophobe who blames random companies for everything that goes wrong with their computer is completely wrong.
It's awful and frankly its why my company -- Twist -- is not offering any third party integration support.
Until the industry can focus on quality over number of integrations and interoperability, the only customers buying these products will be DIY'ers and professional installers.
I'm sure that third-party products were causing problems, however, wholesale blocking of them via software update is a terrible solution. They, literally, turned out the lights on their customers. Meanwhile, I'd be willing to bet support costs immediately spiked -- people call support when things don't work and they just pushed out a solution that increased rather than decreased that.
Unfortunately, I think they've bruised their reputation quite a bit with this move. It's now delayed my purchase of such a product until I am convinced that they have a solid third-party certification program in place (with very low licensing fees) or (even better) a guarantee with the product that they won't try this again when the market is more mature and they have the option of ignoring complaining customers.
Their competitors could see a rise in sales by taking advantage of this blunder and committing to open protocols. I haven't looked at the landscape in this category, yet, and had just assumed I'd be buying the Philips Hue eventually, but they've motivated me to do more research.
I get that people make mistakes, but their original move showed that Philips has essentially NO understanding of their market, and that they are willing to casually engage in extraordinary hostility towards their customers. This isn't a winning combination.
At least in the future they'll be able to stick to "if it's not certified by us..." for customer support, which was likely the original impetus (along with a desire to cut off cheap alternatives to their devices).
I'm not mad at this at all.
I am. Because it is a way to extort money out of other ZigBee participants.
No, it is not. They have two options - have their own standard or support other ZigBee devices. This has not been a problem with another similar standards like KNX.
but I imagine supporting cheaper third-party devices is becoming a headache
It is not about supporting some cheap third-party devices but other ZigBee devices that Philips claims to support.
Also, want to point out that philips is not "extorting money" from other companies - they can choose to enter the "friends of" program to make the general consumer feel better, or choose not to, at their discretion.
The class of user that realizes any zigbee bulb will work with a hue hub is already technically proficient, and will more often than not be willing to take on any associated "risks."
Oh wait it can't (if it can because your wiring is crap and the protection devices are not working you have much bigger problems)
And of course it's not a software issue
Of course it's not. You weren't talking about one, you asked how a LED lamp can burn your house down. It can.
Personally, I can't see how someone could legitimately blame Philips for a Hue-connected-but-unsupported third-party bulb starting a fire, but I don't doubt some will try.
Funny thing though, but that AC/DC converter is manipulated by firmware activated by a wifi protocol. If the fire only starts when the converter is activated into its highest conversion rate in a particular sequence by certain commands sent across that wifi protocol and those commands are being chosen by a user of an app on a mobile device two rooms away, is that a software issue? It's certainly a gray area.
If you want an analogy, then it is similar to the case where MS Office would refuse to open ODF documents from another office suites and then after uproar would only support documents from friends of Microsoft.
With this as the background, it's surprising to see a large crowd defending the equivalent of Ford-branded gasoline.
Pure hardware is a lot easier to make compatible consistently than hardware+software.
https://www.mef.net/certification/equipment-certification-ov...
Switch vendors are also notorious for locking out SFP modules from other manufacturers for no reason, which is particularly annoying because there are so many different types applicable to specific situations (e.g. DWDM) and your switch manufacturer doesn't even necessarily make the ones you need.
(Disclaimer: I've been out of the networking game for 4 years)
They have proven they can't be trusted with this sort of power, and that is a one way trip. You don't come back from that, you don't get back off my list.
This decision was made by someone in marketing. Phillips engineers (in this division) were ambivalent because it meant less verification and testing (yea!) but also means they have a less capable product. It got approved because someone (likely a director somewhere) put together a market strategy that showed they could make x dollars in the next 2 years doing this.
Public backlash was bigger than expected. VP gets involved. Decision is changed.
However, I would say that a company (big or small) that actually changes their strategic marketing decision based on a user complaints _is_ rare. And this shows that Phillips has some good mechanisms in place to respond. To me, they come out a bit ahead at the end of this.
People in Marketing and Product tend to be way more in-tune with customers and don't make boneheaded decisions like this. Another easy tell: the company was shocked by the reaction of users. That meant the company wasn't aware of user impressions of the decision. That also meant Product/Marketing teams weren't involved in the decision.
When the "what does it cost us to test this compatibility" calculus comes out as more expensive than "what is the cost of the backlash to our company," you realize that Engineering divisions without enough resources are driving this type of decision, 99 times out of 100.
I disagree. Almost every shitty (customer-wise) decision I've seen a company doing was made by Marketing / Sales people. It's them who figure out that they can get more money by doing something repulsive.
This here seems to me like someone was calculating how much more they can earn by pushing manufacturers down the "Friends of Hue" line, but underestimated just how many people use non-FoH bulbs (or just how much nerds care about interoperability). Probably not someone from marketing, definitely not someone from engineering.
> When the "what does it cost us to test this compatibility" calculus comes out as more expensive than "what is the cost of the backlash to our company," you realize that Engineering divisions without enough resources are driving this type of decision, 99 times out of 100.
Why would they care? They're not testing compatibility with Chinese knockoffs anyway, that's why they use ZigBee, and that's why they have a "Friends of Hue" program.
I disagree. I don't see engineers coming up with or getting on-board with such a clearly anti-open-standards decision. Much more likely, what I think happened is that this decision was forced on engineering from above, morale hit all-time lows, there was much grumbling and consternation as they implemented this anti-feature that they clearly didn't believe in, then they rejoiced at the huge public outcry when the change was pushed, and are now celebrating that those assholes up in management had to reverse course with a heavy dose of "I told you so".
Source: I am an engineer at a big company and have seen this scenario play out many times internally.
There is zero reason for an engineer to have "anti-sentiment" to an open platform because it's moot. Most likely, an engineer would be completely indifferent to a decision like this.
They reversed the decision because of the user outcry, they noticed a lot of users would rather risk stuff not working than not get them outright. This is a major step and I like that a lot. Being able to change your course to what users want instead of what you think is the most profitable should be applauded.
A ZigBee network seemed like the right idea.
/s
They mad a public about face, admitting that their decision was not the right one for their audience, and changed it. That's not easy to do for most people, let alone companies.
The tone is very much "we did nothing wrong." I don't expect more, but I was hoping for it.
That's perfectly valid. They felt that it was a minority doing it (which is probably true) so this would be a non-issue. It wasn't, so they reversed it.
Hat down to whoever made this happen over there! The world is better when things are open.
Do they really believe it is a small number of customers that use non-Philips light bulbs? I mean, good for them in reversing the decision, but the damage is already done (check out Amazon reviews for one) and it should have been easily foreseen.
Can't say I disagree with the idea. Everyone wants a system like this to be open on principle, but in reality it won't really get used that much. This stuff is still too hard for the average consumer.
This is mainly because the Hue bulbs are quite pricey and I've had a couple of them semi-fail (some colors in the spectrum stop working due to the blue LEDs crapping out). I've been keeping an eye out for compatible bulbs that are a bit more affordable since Hue prices haven't dropped at all and I'm more likely to drop $20-30 on a fancy bulb than another $60 when they can still fail.
Granted, I've already got the hub so they have my money. Their risk is that I only buy third party bulbs instead of Hue bulbs from now on. Still, the next time I go to buy more lamps to expand my setup, if there are less expensive options that are reasonably equivalent, I'll buy them instead anyway. If Hue lamps become more affordable, I'll stick to first-party by default. But if Hues are still $60 each and some other platform starts selling good RGB LED bulbs for $15-25 each (and Hue has locked out third party bulbs), I'll just drop Hue in general and cut my losses.
Not a single Hue-using person I know, outside of HN, has any clue about ZigBee or interoperability or any of the stuff that a small fraction of Hue users care about. None of them. If it wasn't on the placard in Best Buy next to the bulbs, it wasn't even a passing thought.
I am pleased they reversed their decision. I don't actually care about interoperability, even though I am a technically minded user deep into home automation and relevant protocols. But I still recognize the significance of a big company offering up a mea culpa when they get heat from a vocal group of customers. Not many have the courage to do so.
Let's not diminish that they responded to customer backlash by manufacturing other reasons to hate them, or by generalizing. They actually did what the users wanted.
Who writes these things, and why do their supervisors allow them to keep working there?!?
[0] http://www.mcsweeneys.net/articles/an-interactive-guide-to-a...
3/5 stars: http://www.amazon.com/Philips-455303-White-Starter-Generatio...
4/5 stars (previously 4.5/5): http://www.amazon.com/Philips-456210-Ambiance-Starter-Genera...
It's like connecting to your office chat with an IRC client because you figured out that's what they are using under the hood. Why would you scream bloody murder when one day your IRC client stops being compatible with it? They never advertised this to begin with!
You can't exactly demand functionality that you were never sold.
They are free to mutilate that standard as they see fit for their own product, and since they didn't make it into a selling point, there is no reason for them to expect compatibility.
Likewise, "mutilating" a standard to leverage standard technology in a not-really-standard product isn't the same as deliberate artificial incompatibility for purely commercial reasons.
That phrase was up there before the DRM, during the DRM, and now after the DRM. So yes, they were advertising the functionality.
They openly tested and added 3rd party bulbs to be controllable by the system in the past: https://twitter.com/tweethue/status/567368692452048896
They were advertising it and committing open actions to support 3rd party bulbs. The community interpreted their actions as shit rightfully.
In order to do it properly, there should be standards that major providers agree upon making integration much easier and predictable. That takes plenty of time.
Then you probably need some walled garden to control the experience. Approved apps, approved 3rd party providers, etc. If some crappy app is released, regular users won't blame the developer but the platform, as it was discussed in great details in other threads. We need to get out of the HN bubble. Seriously. We forget that a computer is a device to watch porn and browse facebook and that's about it for A LOT of people. Chances are, it will cause a wave of anger in communities such as this one (where there's a strong sentiment for open systems).
This work has to be done be a number of large providers (read: long processes) and followed by startups popping up and disappearing now and then. This stuff always takes time.
They got many people very pissed off and probably never buying or building products with their chips again.