I do. And I recognize the hypocrisy of that. I think the difference for me is that my phone isn't a permanent fixture in my house - it's not explicitly a surveillance device. Echo/Home are explicitly audio surveillance devices, though ostensibly benevolent in their design. I can stick my phone in a Faraday bag easily enough if I want to.

I guess what it comes down to is that while I recognize that I sacrifice a measure of potential for privacy with my phone, I judge the benefits to be worth it. I can't say the same for Echo/Home, and it somehow feels qualitatively different in that phones aren't designed to be always-on microphones constantly processing audio to send it off to a server somewhere else for analysis.

To be honest, I trust that Google and Amazon are probably acting in my best interests with these products, for the most part. I don't think it's all an evil conspiracy to get microphones and cameras into my inner sanctum or anything. I believe it's very possible for a product like this to be mutually beneficial for provider and consumer. And hey, if a state actor wants to surveil me, it's going to happen regardless of these devices being in my house. But as we've learned just how fragile security is, it seems foolish to me to rush off to install black-box monitoring devices all over and just hope that they do what we think they are doing. I love the idea of these things, but they are such potentially powerful windows into our lives that I think that we have to be really sure the value they provide is worth the security risk they entail.

Its not that I don't think that Google and Amazon are probably acting in my best interests now. They probably are. Also not that I think that the government couldn't spy on me if they wanted to. They could.

The scenario that worries me is that years from now, Google and Amazon may be coerced into turning over what I said yesterday to a corrupt government fishing through the past for word-crime. And boy-howdy, it turns out they recorded everything and retained it forever. Surprise!

>The scenario that worries me is that years from now, Google and Amazon may be coerced into turning over what I said yesterday

Amazon has said that they're only recording what happens after the hotword is triggered. That's why they can't do custom hotwords - listening for "hey alexa" is built right into the hardware and it doesn't do anything until it hears that. Unless they're flat-out lying, they can't provide any information about what you said yesterday to anybody because they don't have that data.

My 2012 Moto X has a trained keyword (and in fact a dedicated low-power specialized processor to listen for it - one of the things I like about that phone). No "OK Google" for me; it's "Listen please", in Russian. Never tripped by mistake.

If we believe Snowden then Google and Amazon as well as any technology company that collects user data are enabling spying on you for many years now.

> I think the difference for me is that my phone isn't a permanent fixture in my house - it's not explicitly a surveillance device.

For most people, its a permanent fixture on their person: people these days leave their home more than they leave their phone.

> Echo/Home are explicitly audio surveillance devices, though ostensibly benevolent in their design.

Echo/Home are no more explicitly audio surveillance devices than cell phones are.

> I can't say the same for Echo/Home, and it somehow feels qualitatively different in that phones aren't designed to be always-on microphones constantly processing audio to send it off to a server somewhere else for analysis.

Modern phones (considering the hardware and software stack together) are, in fact, designed to do just that, both Siri and Google Now voice recognition works (though it is, IIRC, disabled-by-default in the latter case) from the lock screen. Its just as always-on as Echo/Home.

One key difference is power consumption, and to a lesser extent, data consumption. It is not practical for your mobile phone to continually spy on everything you say, due to battery issues; however, it can listen for a keyword to wake up to, and then do the more high-powered voice recognition. So the scope is limited, or you will just have a dead phone in your pocket (which, of course, will also not be listening to you).

With something like Echo / Home, the always-on power source makes it technologically feasible for it to always be listening to you as well, without needing to wake up to a keyword. Now, they may as a feature only start processing based on a keyword, but you don't have the "comfort" of knowing that it is technological limitation rather than a software decision that may or may not be in your control.

Cell phones aren't allowed into classified environments in US government buildings because they can be remotely turned into listening devices without the owner's knowledge. There's been guidance on this for a long time. The cell phone in your pocket is definitely a surveillance device.

Let's be real. You don't keep your phone in a Faraday bag and you probably never will.

Even if you do, it can still record audio, and transmit it when you take it out

Ever tried recording audio through a closed bag, Faraday or otherwise? I doubt it would be of much use...

The carrier needs to remotely tell the device to enter into a special mode, so a faraday bag would be an effectively method to block this.

Unless the device is programmed to automatically enter into that mode in the absence of any signal on the assumption that if it is in a shielded location it is likely interesting.

Until the next time you take it out of the bag to actually use it.

Nope, I dont. And I agree with @cheald. It's a big shift to have a "press to query" as opposed to a vague keyword-based "always available".

I know some technophiles coworkers who have an Echo, and I couldn't for the life of me trust Amazon, Google (or even Apple) with potentially the always-on stream of audio from my house.

I really wish these various voice offerings would offer push-to-talk as an option. It would solve a whole bunch of problems: It would activate reliably when you want and almost never accidentally, you don't need to use awkward code phrases, and it can trigger results as soon as you release the button instead of having to wait for silence. It would also allow for better privacy by making a phone in which the microphone is hardware controlled by the button.

It just seems so obvious to me that I don't know why no one has done it. I guess needing a new button makes it harder to sell, but this seems like a great use case for smartwatches.

I think that's what some folks hope the Apple Watch and/or Android Wear can do - be conduits (like ST:TNG Communicators) to cloud supercomputers.

I've ditched the smartphone. And, for the most part, the phone.

I'm not at all happy with the trade-off any more.

I've got a tablet, both cameras taped over. Killing the mics would be a Good Thing.

Sigh.

Honest people do not go offline, only people with something to hide do. Honest people have all microphones accessible and all cameras untaped. I am afraid you have been flagged by NSA because of your suspicious behavior pattern.

Tablet connected to a network?

Frequently, yes. Though not 3GL/4GL mobile networks.

I'm aware of the risk and hypocrisy. I'm also far from happy with it (I increasingly dislike Android, Samsung, and with varying levels, Google).

But given the balance of something I can power down, that isn't always on me, and the portability and cost benefits over an alternative laptop, its a far-less-than-perfect satisficing solution for now.

I'm thinking my nexus will be my last Android device, not decided yet what I want to replace it with but dumb phone is looking likely, I'm either at home, the office or somewhere I can use my laptop so don't really have the need for a tablet, I might been able to google the odd thing now and again when out and about but that's a once or twice a week thing for me.

I've come to resent carrying a phone all the time, I rarely answer it and resent that "I tried ringing but you didn't answer!" whine you get as though it's beyond incredulous that you would ever not answer the phone.

What I like about the tablet is the size, the battery life, the switching between landscape (generally: composing w/ keyboard) and portrait (generally: reading long-form content) modes. A 10" form factor plus bluetooth keyboard is sufficient for some Real Work (though that's hugely crippled by system and software limitations).

As a comms device, on a sufficiently fast WiFi/broadband connection, it's also useful. I've not done that often, but it's been done, and I was impressed.

The OS is a mess in many particulars.

It handles external storage poorly. It doesn't multi-taske well. Multi-window mode is specific to application. The default shell userland is unforgiveably anemic (Termux, one of a few available add-ons is much better). Lack of comprehensive privacy controls, or insight, is a huge issue.

The app "ecosystem" is pathologically broken. I've written on this at G+, my feeling is it needs to be scrapped.

Samsung's own tweaks are also pretty bad. The fact that the device is entirely locked down from a bootloader perspective (it was a bit of a panic on-the-road buy) has become beyond annoying. Inability to remove certain stock apps is a major concern.

Google's "Play" store is beyond annoying.

I'm watching for news of Firefox OS and Ubuntu's mobile offerings. I'm not particularly hopeful. A small laptop _might_ be a suitable replacement, if I can park Linux on it (without systemd).

Mostly it's a very effective advertisement against itself.

And more generally: the tech world very, very, very badly needs to find a non-advertising business model.

Have you tried Bq Aquaris M10 Ubuntu tablet [0]?

I've got one and although it is not ready for the general public yet, it's a little gem for us free software fans.

[0] http://www.bq.com/uk/aquaris-m10-ubuntu-edition

No, I haven't, though that looks interesting. Quick thoughts:

16 GB storage is far, far, far too little. That should be upgradeable. The problem with microSD expansion is vfat. That's crippled on both capacity and permissions. I'd also prefer encryptable removeable media.

Speakers. I want my devices to NOT have them. I'll attach via Bluetooth or a wired headset/mic combo. I'd also prefer cameras be shuttered. That's both for protection (lenses get scratched) and privacy.

Other than that, yes, it's headed the way I'd like things to go.

Erm, meant mics not speakers.

Input devices should be clearly and unambiguously disableable, or external.

The external storage thing is an accident of history, with perhaps some MAFIAA meddling.

When the first Android device shipped, all internal storage was dedicated to OS and apps. But OEMs wanted to advertise that they had X gigs onboard. And thus they started partitioning internal storage, and mounting one part as if it was removable memory.

Problem is that apps could not go on removable memory, so people bought a device only to find that a few apps was enough to max out that side of the partition.

And thus Google came up with "move to SD".

But then that got into trouble with USB data transfers, as the "SD" had to be unmounted on the Android side before it could be mounted by the PC over the cable.

So somewhere around 3.x Google adopted MTP, and union mounts to emulate a FAT partitioned SD (for legacy reasons).

But at the same time they introduced new access permissions, but reserved write permission to actual removable storage for system level apps. Thus you could use your PC to transfer music and such onto a SD, but you could not officially do so from within Android. Various OEMs, Samsung for example, modified the permissions file so that their devices retained the old behavior.

Since then Google has in fits and starts modified the MTP support so that it can also be used by internal apps to write to removable storage without having to deal with permissions. At the same time they have tried to lock down access to the "SD" even further as various dumb app devs have used it to store sensitive user data for any other app to see.

Thing is that going via the MTP subsystem invites DRM, as the apps talk to a database that then talk to the FS. And that database can be incomplete, or the subsystem can outright lie to the apps. And it so happens that Google adopted MTP along the same time as they introduced Play Music and Play Movies.

Then again, the consumer UI/UX designers have long been tilting the windmills over the whole files and folders thing. For some reason they see it as a failure of UX, and thus something that needs to be at least hidden from the user, if not outright abolished. See numerous attempts at replacing files and folders with a grand database of metadata, that will "magically" pull out everything you have stored related to that company project or aunt Tillie as needed.

I don't know about Android (though I assume it works the same way), but iPhone's "Hey Siri" is done entirely on-device, so until you activate Siri, nothing you say is sent to a server.

People keep bringing this up and it's silly. It can just record what you say and send it later.

It makes literally no sense for the Hey Siri functionality to do that. This is just a very slightly altered version of "it's an internet-enabled device with a microphone, it could be hacked and told to turn on its microphone remotely", which is a legitimate, if fairly paranoid, complaint that applies to all such devices, even fully open-source and open-hardware ones.

Though is that auditable? Siri's not Free Software AFAIK.

You can at least audit the network traffic. And since people have been able to hook their own software into Siri, you probably somehow can break open its encryption and look at the content. (Unless I misremember how that worked)

Which is why I keep mine in a lead box at home.

an interesting side effect is that, until you open the box, the phone is in a superposition of both ringing and not ringing at the same time.

I probably don't -- my phone is often in another room when home.