My biggest issue with this is that Google wants to inject a piece of hardware into my home that acts as the gateway to my data. A device through which all :
> Incoming data runs through
> Internal data runs through
> Outgoing data runs through
And then wants to be vague about what data they will be pumping to their back-end for analytics/data mining.
I have had a long standing proposal for home automation and I'm curious to know, in this era of insecurity and people vacuuming up all your data, how many of the tech minded people here would be interested in a device which ensures your 'home automation' data stays in your home. This can be quite clearly achieved in hardware and by having an out-of-band oversight controller that literally will not allow certain data to exit the physical domain of your home...
Cloud nonsense? It's called an application and a home automation application doesn't need to run in someone's cloud...
Whose interested? I think it's about time this cloud foolishness for the sake of monetizing someone else's data in an insecure manner come to an end. Everyone loves to rant about 'disruption'... I feel its about time this data monetization cloud bananza be disrupted.
Oh, come on. Just connect your Google router to your Google fiber connection and connect to it with your smartphone or laptop running a Google operating system and Google browser. Visit your Google home page (using Google's DNS servers, of course) to read your Google Mail, or perhaps catch up on the news with Google News, or use Google+ to see what your friends are up to, or get a little work done on Google Docs. Should you do some Google searches and end up on some non-Google sites, don't worry - you're still safe under the watchful eye of Google AdSense and Google Analytics. What have you got to be so paranoid about?
Don't use Google Fiber. Well, shucks - there's not a lot of 1,000 megabit ISPs in my area.
Don't use a Chromebook. Okay. What virus scanner should I install, then? And where should I store my data for backup? And didn't we hear that Microsoft tried to have Windows 10 phone home every damn keystroke or something?
Google browser. Fair enough, Firefox is decent.
Google DNS. Fair enough, unless I have Google Fiber. Then it's kinda weird to not pair them.
Google Mail. Something that looks like GMail has all of the problems of Google Mail. Is Microsoft really a better solution in your mind?
Google News. Come now, the problem of monopoly in news existed long before Google News did. I'd rather people also use Google News than JUST CNN, MSNBC, or shudder, the other one.
Google+. Right, because Facebook is so much better at respecting users.
Google Docs. Again, is Microsoft really better? Or Microsoft + Dropbox?
Google Search. What, I should use Bing?
AdSense and Analytics. Not really my choice, is it?
I genuinely ask this question. What do you think the common person could or should do, that would be better for them?
What, the Apple computer, Apple router, Apple browser, Apple Mail, and Siri?
You have pointed out many of the reasons why this is indeed a hard problem. That's why some of us have been trying to warn about these dangers for almost 20 years. The problem was significantly easier to fight 10 (or even 5) years ago, but everybody - including the technically knowledgeable people that should have known better - decided that shiny features were more important than paying attention to the larger picture and defending their future freedom.
> What do you think the common person could or should do, that would be better for them?
They should not use any service that goes against their long-term interests. While having a replacement is nice, this might require making a sacrifice. The lack of an alternative doesn't justify supporting the bad option[1].
Do you think that this problem is going to get any easier as time goes on? The cost of leaving Google is only going to increase, so it might be a good idea to find a way to pay that cost sooner rather than later.
At least by using a non-Google "free" mail service, file storage service, etc, you're distributing your mineable information among competing companies instead of handing it all over to a single entity. Or, you can do as I do and just find a reputable service you can just pay and not worry about it. (I pay Zoho $24 a year to host my mail.)
> What, the Apple computer, Apple router, Apple browser, Apple Mail, and Siri?
The difference between Apple's services and Google's equivalents is that Apple makes its money by selling you the boxes that the services run on - it doesn't directly profit on the services themselves. For Google, the service is the product, so if they can't make you pay for the service, then it has all incentive to data mine it for profit.
I fail to see how copying my data to more companies reduces my attack surface.
> all incentive to data mine it for profit.
Yes, and it has all incentive to protect my privacy, because if they screw that up, they lose all of my business, and all of everyone else's business, too.
> I fail to see how copying my data to more companies reduces my attack surface.
I'm talking about your privacy and mine-ability, not your attack surface. (Though it does help with that too, in a way - a bigger chance someone will get some of your data, perhaps, but a smaller chance they'll get all of it.)
No. That does not help minimize my attack surface. You're confusing surface with depth. And unfortunately, someone doesn't need to steal very much of my data in order to screw me over royally. If my Social Security number is held by more companies, that does not help me.
>Yes, and it has all incentive to protect my privacy, because if they screw that up, they lose all of my business, and all of everyone else's business, too.
There have been innumerable data breaches and we haven't seen those companies go bankrupt.
Firefox may be a decent browser, but unfortunatly Mozilla has given in when it comes to adopting data kraken integration and has embedded software that can make Firefox en par with Chrome et al in the future.
No more good guys, I am afraid. Mozilla profited greatly from being spread by hackers initially, but now they no longer qualify for being supported.
No source here, I am on a crappy notebook, on the move.
And still, the HN crowd uses Gmail a lot. Here's the most technical crowd on the planet and people are obsessed with Gmail as if it was impossible to get email to work by any other means.
Furthermore, I will never understand why people would use Gmail, if there's IMAP and hundreds of email clients to choose from.
Same is true for Google Reader, that apparently caused a lot of buzz when it went away. Never understood this.
The only Google product I use is Google Maps, and Google Search occasionally. Replaced the latter with DuckDuckGo mostly and works well most of the time.
I would gladly use email client with IMAP server but desktop mail clients are really still in the 90ties.
Do you have a good to alternative Gmail/Inbox which has Inbox like archiving, and easy cleanup of the ... inbox, hiding emails for some period of time, pinning emails?
As for the hosting of emails (assuming I don't want to have 10GB dedicated to emails on my machine), do you have anything that gives me 10GB of mail and has the ease of use of Inbox on all the devices (desktop and mobile)?
Disregarding the (slow) advancements of E-Mail encryption (DANE), and the fact that's desireable that no one company has your data. The big point is that it's fair to assume that everybody has access to your mails while in transit, but using gmail Google has access to all your mails at rest.
"Ditching GMail changes nothing about your privacy situation."
Actually, there is quite a bit to be gained by running your own mail server - especially in relation to other people on your mail server.
For instance, everyone at rsync.net logs into (al)pine over SSH. So yes, if you email rsync.net and we converse, that is like the postcard - every hop it goes through can see it.
But no piece of internal rsync.net email has ever traversed a network of any kind. Internal rsync.net emails are just local copy operations from one mailspool to another.
The same could be true of your company ... or your family.
Agreed. Even if Gmail is superior in some regards (I don't know - is it? There must be a reason so many people use it), as a tech savvy person I consider it a matter of pride (and privacy of course) to run my own mailserver with project like Sovereign https://github.com/sovereign/sovereign
I have been thinking of doing this for a while. Thanks for sharing https://github.com/sovereign/sovereign. Is that all one needs (whatever if mentioned on the README)?
This is a common belief but it hasn't been an issue in my experience of running my email server (with Sovereign) for the last two years. I run about a dozen email addresses over a handful of domains for myself and some friends and colleagues.
Sovereign includes instructions/configuration to run an upstanding email server citizen, including SPL, TKIP, MTA encryption, etc. Remote mail servers seem to respect it at least as much as gmail.
It also runs a rbl-check script once a day to notify you if your IP ends up on a blacklist. In two years running on both DigitalOcean, I've had no issues. Even gmail routinely gets added to RBLs from time to time.
I use Mail on OS X and it is good enough for my purpose. Could certainly be a bit more sophisticated, but most of the time, it works. I like to have a real email client for my mails and not use a web app.
There might be dedicated search engines for emails though, if search is very important.
See, if you use search a lot and Google has the superior product for your use case, that's totally fine. But I guess most people use Gmail out of habit and not because it is truly superior to other email clients.
I'm with you here - I also find OS X's Mail plenty "good enough," including for search. And it works great for connecting to Google Mail accounts via IMAP…
Searching is a solved problem if you use OfflineIMAP to mirror your mail store locally, then run mu or notmuch on top of it. Honestly, it's faster for me to search my local mail store than using some web interface. Plus, i can do it in the train when i have no internet connection. I happily pay a few euros a month for an email server which provides me with IMAP.
I tried Bing Maps the other day and was surprised how much more usable it is. It loads faster, moves faster, and actually has a nicer interface because it uses semitransparent text to show names of areas.
Just finished reading 'The Circle' by Dave Eggers - think Google's monopoly and totalitarianism come true - and this just further reiterates how Google are become so pervasive and invasive in monetising off everyone's data and lack of privacy awareness.
I'll go back to dial-up rather than have all my activity centralised in the Google ether.
The NSA should stop the charade and just buy Google and Facebook already. They already have us by the balls, lets at least make it official and make the world embrace the heroic and always rightful eye of the US government.
I'm glad that people found this to be a point of significance. I've worked in the commercial networking hardware industry for some years and shelved several consumer ideas due to a lack of interest in security/privacy.
I have watched over many years as this 'cloud' software bonanza has eclipsed the tech industry. It went away from its original intent a long time ago and now is used as the holy-grail method of :
> ensuring (rent) is always paid for service(s) people should OWN
> vacuuming up and selling people's data.
I have several projects on my plate. However, I have a solid one for securing people's data when it comes to the IoT movement (in hardware)... It revolves around the same hardware/technology that the industry uses to scrape and funnel your data except the home user will now have it at their disposal as well.
Users should have 100% control of their data. If someone wants to 'monetize' it or use it to 'improve' their platform, the user should be rightfully paid for the opportunity to do so.
Disruption? Yeah, I think it's time. The foolishness has gotten far too long in the tooth.
.. And the backwards thing about it is: there are simple and straight forward ways to secure data (in hardware). It's only because everyone in the industry wants to ship your data all over the place and data-mine the snot out of it that things are as insecure as they are.
The very (loop-hole) or (door) that is used to funnel/mine/monetize/cloudify data is the very one used as an attack vector. Complicating the crap out of things at that point defeats the whole purpose.... Get rid of the door all together and the home is the least of all places where you should be exposing yourself.
Therein lies the 'product' ... I'm working on several projects so I wasn't shy to put the general idea of this one out there... It's one of many things I feel passionately are going to be changed by disruptive paradigm shifts ahead.
That being said, you come up w/ the 'solution' and you'll have your 'product'... I already have a general architecture/approach outlined for myself. Create one and, from the looks of the interest here, you'll have an audience to buy it.
Also, with the scale and depth of hacks occurring around the world, you'll soon have an even broader audience.
If you get funded, look me up ;). I'll be looking for work in the coming months =P.
I'm wondering if / how the identity problem, blockchain proof of physical ownership and ...
Well, firstly applications should stop thinking they are CoW - data is something that gets passed in and operated on.
Secondly data operations should be side effect free - and distributed, so that we can ask for processing- in many ways I would expect not to share my data but to accept payment in return for some slice of processing someone wants
This may work ok, and generally the things we want (social networks) will be fine with local data sharing - facebooks walled garden would have to be replaced with open protocol (ie AOL -> SMTP) and no idea how to achieve that.
So im not setting the world on fire here. But I think the shape of future is clear - just getting there ...
That is why i don't buy into this hype what so ever. Anything that has 'cloud' on it when it doesn't need to I avoid.
I'd rather set up my own router with pf-sense and a industrial WiFi access point.
I really love the concept of complete home automation however the data needs to stay in the home and companies need to ask for permission to mine it.
I have nothing against data mining it can be incredibly helpful but i want full control over weather i choose to share it or charge a small fee for my data.
The whole idea of home automation (and most of IoT) over cloud is absurd. The devices in your home should not communicate by sending packets around the entire globe. The data should never leave your internal network unless explicitly meant to (web interfaces are cool and all, and you want to have remote access).
I'm actually somewhat surprised people don't seem how wrong the current model is, but I suspect it's a mix of cloud being The Hot Sexy Thing and being paid not to understand this (via business models that rely on monetizing users through cloud solution; as some of my hardware startup friends told me once after talking with the investors, hardware won't be making money, the cloud platform will).
So yes, I would be very willing to help reverse this and make local network communication the default for home automation and relevant technologies.
Google's model of how it wants computing to work is fundamentally flawed and now it's infected Microsoft. Really the only shelter is Linux, when will the ignorant masses wake up and realize this?
"Please note that some features may not function with certain privacy settings turned off, and some information (such as the association of your Google Account to your OnHub) is stored by Google even if all privacy controls are turned off"
I am sticking with my Asus AC66U which has more features and is also cheaper than this Google device
http://amzn.com/B008ABOJKS
All software has security vulnerabilities, but organisations have vastly different attitudes toward fixing vulnerabilities as they are identified, and potentially different capabilities when it comes to fixing vulnerabilities in a timely manner.
But Google will fix the security flaws automatically.
You have to compare this device to the typical home router, which is NEVER updated.
And even if you do run an Open Source firmware, you have to make sure it's kept up to date.
Please don't bundle Android and the rest of Google software. They are two different teams with two different philosophies, and one team doesn't like to have it's reputation hit by the other teams faults.
Remember Android wasn't originally designed by Google, and many of it's security design decisions and culture were inherited.
Unfortunately when it comes to data and cloud companies these days, that's a little like saying you'd trust Blofeld more than Dr No because he's a more competent villain.
In case HN is wondering about the reception: it's great by default, but I had some hardware mounted in inaccessible places, which could still have issues with reception.
So I replaced the middle antenna with a WL-ANT-157 from Asus, and wow. It's the best home wifi experience I've ever had in this price range.
Hmm, that "pretty clear about exactly what date they're collecting and why" seems a bit too similar to what the proposed Australian data retention laws ask for, which the community usually refers to it as "vague and overreaching". They make a vague note about not collecting the content of network traffic (but you can infer almost everything important from source, destination, time, protocol, etc anyway) and the table on the page is just described as "examples" of what it collects, and is never stated to be the full list. Given that Google is in the data collection business, and has a licence to update the device (and what it collects) automatically, I would assume that within a short timespan it will be collecting everything it can get away with.
Even what they do mention is enough to start inferring things about your personal life:
With "historical data consumption" they can determine who is in the house and when. With the number and make of connected devices, they can take a pretty good estimate of family size, annual income, how many of your household are working, etc.. (Though they probably first care how many Apple devices there are connecting... maybe we need to send you some more Samsung adverts)
And sure it strips the URLs from the logs -- but between Google DNS and Google Analytics being on much of the web, they can piece back together every site you visit anyway and now your router has a Google account they can tie it right back to the router in your home.
And of course it's a $199 router with a license agreement that says Google can stop it working at any time they like (clause 5c).
Proposed data retention laws? They are in fact actual laws, but they just haven't agreed on how much money the telco's and ISP's are going to get for implementing it.
Good link, thanks for sharing it. I gotta say, though, when I saw that wall of text my first thought was that this is going to be a tough sell for some of us. Thinking about it for half a second, I have to say that I would gladly prefer my router to be as dumb as possible. Maybe I just don't understand, at a gut level, how a cloud-connected router will actually improve my life?
OnHub seems to have a bunch of features which the cloud connection enables that are helpful, but not vital. For some people, having those features isn't worth the potential for additional complexity, and potential vector for privacy or security issues, that having a smarter router entails (or the increased device cost). For some people, it will be totally worth it. I personally think there's value in having options for both kinds of device on the market.
Of course it's going to be a tough sell for some people. We could launch a potato and it'd be a tough sell for some people, let alone launching a WiFi router. :) At the end of the day you either trust the information made available by Google or you don't. If there was less text there, plenty of people would be suspicious that something was being hidden (heck - people are suspicious even with the current amount of text)!
It's okay to decide not to adopt the device in question, and it's fine to weigh how much you trust Google against what benefits you might get from any given device. For the average tech-savvy user, perhaps there's a lower value prop, perhaps there isn't. Some people like having absolute control, other people like someone managing things for them.
I personally think the real answer is not to have a central device, but to spread it out into a peer2peer 'fog'.
There's been some movement in industrial applications via stuff like https://twitter.com/FilamentHQ which leverages hardware accelerated ECC and telehash, using blockchain based DNS systems and other decentralized methods of command and control.
I think a similar direction needs to be taken for consumer hardware and data. The cloud needs to become just another peer.
I've been working with a bunch of interesting ideas on trying to get a distributed IoT network setup using programmable blockchains and Eris industries stack as part of an internship with them.
In P2P networks you'll need reliable peers with sufficient capacity. Making a home server into an always connected peer could solve plenty of issues, such as with routing and hosting.
I would be very interested in helping write some code or contributing money towards making improvements to lock down phone home and other unwanted behavior. I think an excellent starting point is to see if you couldn't take a Mikrotik or OpenWRT/Tomato router and bake a lot of this extra filtering and functionality in. For some of the things that Google advertises like speed checks or choosing an optimal channel or whatever, should be do-able in open source firmware, so I think it would be hugely beneficial, at least for me (and probably many others).
This is exactly why I found the Google branding a bit weird. The smart guys over there surely must know that a lot of us would be suspicious of this device. Why didn't they brand it under Nest, or a new Google subsidiary that would take the initial suspicion-edge off? I think it's a good router at a decent price, but no way I'm putting that inside my house!
> The smart guys over there surely must know that a lot of us would be suspicious of this device. Why didn't they brand it under Nest, or a new Google subsidiary that would take the initial suspicion-edge off?
Why? Most consumers aren't in the ultra-suspicious-because-it-says-Google camp, and most of those who are in that camp are likely to take using a different branding for a project from the same ultimate corporate parent as evidence that not only is the project trying to violate privacy, its also trying to be extra sneaky about it.
I'm not buying it precisely because it is Google branded.
Unlike most of the comments here it isn't a privacy issue for me, I totally respect the privacy argument but personally just don't care enough to make decisions based on it... for me, the issue is that when I think of Google and hardware I think of the Nexus Q, Google TV, etc. Google suddenly (and relatively quickly) drops projects like this on a fairly regular basis and when the whole thing is all "cloud-this-cloud-that" dropping support basically means you've got a conversation-starting paperweight.
It's unlikely Google drops projects any more frequently than startups fail. I have (or have had) a lot of paperweights manufactured by failed startups, so I know it happens. I'm sad when startups fail, but they usually failed for the right reasons. It doesn't taint my overall perception of the startup concept.
If you think the Nexus Q was a fantastic product and that Google made the wrong decision to kill it, that's one thing -- you're saying they have bad taste, or bad product sense, or an inverted sense of quality vs. crap. I wouldn't agree with that assessment, though I admit it's an valid, internally consistent opinion.
But it's more likely you never owned a Nexus Q and are just using it as an example of how projects at Google get killed. Sure, Google kills projects. Just as startups fail. That's no more astute an observation than saying that sometimes it's sunny and sometimes it rains. You wouldn't expect Google to keep funding a stalled or not-quite-thriving project any more than you'd expect investors to keep plowing money into a startup that can't find product-market fit. Sure, the opposite outcome sometimes happens. But generally it doesn't, and that's OK.
Some think Google is valuable because it takes more risks than companies its size. The implication of your opinion is the opposite -- that Google should be more risk-averse (not starting this router project because a router is a crazy thing to build), or innovate more slowly (launching it later than today because it's not ready), or ignore market feedback longer than a startup would (damn the torpedoes, it sucks and nobody wants it, but let's keep its team on a death march). Is that how you'd run Google if you were its CEO?
"It's unlikely Google drops projects any more frequently than startups fail."
I don't disagree, which is why I also don't spend money on startup consumer goods that have any sort of requirement on "the cloud" (if the company dying makes the product virtually worthless, count me out) and also why I virtually never back tech kickstarter-style projects.
My opinion on google is that they are extremely bipolar (or at least give the external impression of being so) when it comes to experimental projects, they seem to go through periods where they are open to trying new things and then (very quickly) to periods of retraction where things that aren't ad focused are left to wither and die or just killed outright. I don't want my money caught up in their mood swings unless the value proposition is amazing, and in this case it really isn't.
The difference is that a non-cloud-reliant device can be sold to me and then abandoned by the company that made it some time in the future -- yet it still remains a usable device. Now whether the OnHub falls into that space or not I can't really say for sure; in this instance I actually doubt it does (my guess is it would remain a useful router even if the cloud-connected features were dropped for some reason).
It's also the frustrating potential that the product will be 98% awesome but with niggling problems that don't seem to get fixed as it's abandoned quickly; ie- my recent Ask HN submission about Android TV.
The yellow Google Search Appliance for enterprises is about a decade old. You can still buy them today.
The Nexus One was branded Google. I wouldn't say it's supported by Google anymore because it doesn't receive firmware updates. Google first released it in early 2010.
The Chromebook CR-48 was first released in December 2010, but it wasn't a Google product (unbranded). It still receives software updates today.
BTW the new router isn't a Google product. TP-Link makes it. Google just controls the software, the way Microsoft updates Windows on third-party PCs.
The ratio of people who think about the implications vs the general public who consume rather than question is weighted in favour of big name companies.
There should simply be more coursework on privacy, classroom discussion on information monopoly, corporatocracy, ect... To level the field, it takes Education. (and I'm not saying the router is evil)
"I feel its about time this data monetization cloud bonanza be disrupted."
Ironically, having users controlling their own routers could be the best chance to do that. That is, toss out the crappy consumer routers and instead embrace the router as a user-controlled computer that sits between the user's devices (including IoT devices) and the ISP connection (e.g., modem). This router could run an open source source OS and be programmed _by the user_ to do all sorts of useful things, such as block ads, block tracking, perhaps even create private overlay networks among family and friends, protected from spam.
In a world where the Internet user can have some respite from radio, TV and other advertising, Google (=slave to advertisers) should not be selling routers. Should they come to dominate the market, the respite will come to an end. Users probably will not even know what happened. No one pays attention to routers as computers. They just want a strong WiFi signal.
I don't think there will be a significant market for your idea. One of the most rapidly adopted technologies of the 20th century is Color Television [1] and it still commands 36% of our attention [2].
Consumers have made it crystal clear that they are willing to trade their time and data for lower prices and more convenience. Watch the commercial on that landing page. It couldn't be targeted less at technical people.
False choice. It's not impossible to make something that tastes good and is also good for you. Some people find that Apple products are desirable, and incidentally, Apple provides a relatively high level of privacy.
But you can never expect Google to provide such a product as long as it makes money off your data. Maybe it needs to be under a different letter of the Alphabet.
Well I don't think there is anything inherently wrong with Google. So suddenly people have started to become suspicious about what data a router is able to collect, while in fact various routers which are well capable of executing malicious applications have been around forever? This is just a trend and Google is not some inherently evil guy that acts outside of the norm. Ditch Google? Fine. The next alternative you find is equally, if not more, likely to perform some data collection. Of course you can go full Stallman and ensure your data are truly guarded by your own in every possible sense. But the heavy price that comes with it, we all have seen.
I started developing a self hosted home automation framework for my senior project last year. Been on a bit of a hiatus, partially because I don't have my own place currently, but its here: https://github.com/terramod. Central web application that can run "apps" on your house, with raspi nodes that connect the hardware around the house.
I too am a bit surprised to hear tech minded people get excited about products that basically exist to collect their data.
From what I can see though, the hardware looks decent... Personally I'm curious for when someone does a teardown, to make sure there are no hardcoded callbacks to Google servers that shouldn't be.
If not, this wouldn't be so bad (for $200) running DDWRT/OpenWRT?
[edit] Also, does anyone know if the base firmware will be open source similar to Chromium? If so, one could theoretically still take advantage of their security updates (which would honestly be leaps and bounds above current consumer-grade hardware) with open source tooling.
I agree with you, but is this specific to only home automation? Shouldn't the same paranoia / concern (depending on who you're talking to) apply to critical company data (documents, source code, etc.) as well as personal health information (biometrics, etc.)?
It isn't just specific to home automation. The same paranoia should apply to all data and will once people understand the gaping holes in existing cloud centric/data funneling architectures. In the meantime, a new approach to the problem is worth looking into. As the speed of technological innovation increases more and more, so will the disruptive paradigm shifts.. and this cloud model is long overdue for one.
Yeah, but by giving more data you let software to personalize better. So by reducing your data exposure you are acting as techno-Luddite stiffing progress.
I don't think we will suddenly abandon the cloud. If anything, we will centralize even more information. My biggest philosophical debate is what will happen to people who deliberately protect their information. Surely they are meatier targets by various adversaries, but the thing that bugs me most is the fact that you will become irrelevant. Lack of data will impede even some most basic services and probably you won't be trusted because Big Cloud doesn't have your profile.
> Yeah, but by giving more data you let software to personalize better. So by reducing your data exposure you are acting as techno-Luddite stiffing progress.
Is this sarcasm? I guess I fail to understand how Google now knowing all the details of my life impedes "progress". It shouldn't stop them from developing new technology; why does the "cloud" need to figure out how to "personalize" things more for me anyways, outside of the obvious answer to serve up better targeted ads (which I'd rather make go away by any means necessary)?
Google's core is still about serving you information and possibly some knowledge. To get better insights they need more information. It is as simple as that. MS entered the same field with Windows 10. They need more data to give better Cortana experience. If it's not all the data, Siri, Now & Cortana would suck. Now we are witnessing them going mainstream.
Merging cloud tech (automatic redundancy, intelligence) with P2P tech and giving the user full control is what is needed.
Secure smart home servers that act as rendezvous servers for your end-to-end encrypted connections so you have smooth links and network switching, routing, file hosting for your privately shared media (using tech like Tahoe-LAFS), anonymization (why run multiple Tor/I2P clients in a network when you can have one trusted device do it?), simple access control so you can easily determine to gets to connect to your home automation hardware, etc...
Worried about Security then look @ this wolflink routers. They are far better than this onhub and it can be controlled directly from mobile and the highlight of this router is that it comes with parental control https://www.wolflinks.com/
> Incoming data runs through
> Internal data runs through
> Outgoing data runs through
And then wants to be vague about what data they will be pumping to their back-end for analytics/data mining.
I have had a long standing proposal for home automation and I'm curious to know, in this era of insecurity and people vacuuming up all your data, how many of the tech minded people here would be interested in a device which ensures your 'home automation' data stays in your home. This can be quite clearly achieved in hardware and by having an out-of-band oversight controller that literally will not allow certain data to exit the physical domain of your home...
Cloud nonsense? It's called an application and a home automation application doesn't need to run in someone's cloud...
Whose interested? I think it's about time this cloud foolishness for the sake of monetizing someone else's data in an insecure manner come to an end. Everyone loves to rant about 'disruption'... I feel its about time this data monetization cloud bananza be disrupted.