Amazon Alexa Comes to Pebble Core | Hacker News Reader

Amazon Alexa Comes to Pebble Core

blog.getpebble.com

187 pointsyurisagalov109 comments

I'm an (unhappy) Apple Watch owner, but this is one thing which has gotten me to back the new Pebble on kickstarter -- Alexa is the most useful audio platform I've encountered so far.

You won't be dissapointed, although you may want to get an android phone. The benefits outweigh the costs.

I thought Pebble Core didn't need a phone? Or am I mistaken?

It doesn't. The Pebble Core is a completely standalone computer with GSM/WiFi/Bluetooth.

The OP's post is ambiguous - replacing an Apple Watch with a Pebble Core doesn't make much sense, but replacing it with a Pebble watch doesn't make sense in this context either as it doesn't have Alexa.

My thought was the OP has an Apple Watch, thus obviously an iPhone. If they want to buy into the Pebble ecosystem, it'd be a better experience with an Android phone, not to mention you get amazing features like an external battery and an actual choice in software.

The Pebble Core seems like a very different proposition to a Pebble Watch though - it runs Android itself and doesn't depend on a phone. So much more viable if you have an iPhone, though obviously a totally different form factor.

Not sure what you mean by "external battery" - both Android and iPhone have external batteries available.

I assume he meant a removable internal battery. I'd rather have some other way to hard-disable the phone, because integrated un-replaceable battery is a big win for sealing/size/durability, but I admit that's one of the advantages of the android world. The biggest advantage of Android is being able to run third-party ROMs, override core functionality, etc., though.

Yup, hence my "actual choice in software" bit. Generally it seems all you do with iPhones is lease them from Apple.

>The biggest advantage of Android is being able to run third-party ROMs, override core functionality, etc., though.

The average person has no interest any of that t

That's why smartphones keep sucking more and more with time. Building products for the lowest common denominator get you shit, shallow products that don't use even a fraction of their potential.

>The biggest advantage of Android is being able to run third-party ROMs, override core functionality, etc., though.

The average person has 0 interest any of that.

>> "a totally different form factor."

THIS. Core's basically a smartphone the size of a an iPod shuffle where the interface will not need a screen, and where you're expected to interact with it via voice or the 2 minimal programmable buttons (which will probably be context aware and provide even more functionality depending on what app you're using). This is why I find the whole device fascinating! The iPhone removed the keyboard, the core is removing the screen!

I'm Apple Watch/iPhone. I'm probably anti-watch now, thanks to shitty experience with the AW -- I won't be replacing it with another watch of any kind. However, I'm super interested in the Core for many non-watch applications.

I also hate android for many reasons and am unlikely to move away from iPhone except to something entirely new/better than either, which doesn't particularly appear to be in the pipeline. In the next couple years I want to get rid of all uncontrolled radios on my person, though, so I'll probably be switching to a de-radioed iPhone or iPad or iPod Touch or something.

Why don't you like the AW?

What's to "hate" about android? When is the last time you used one? Android has closed the Gao with iPhone, and surpassed it in many ways.

What does android phone have to do with it?

It's my understanding that the Android app that interacts with Pebble (the watch) is more robust and better supported.

I don't think "better supported" is the most apt description. As I understand it Pebble puts similar level of resources into both the Android and iOS support but there's simply more restrictions on iOS which means fewer features compared to Android.

From my reading of this, Alexa has not been announced for the new Pebble watches, only for the Pebble Core (the GPS/3G/IoT module)

And it won't. Alexa CAN NOT run on Pebble watches. Alexa needs a proper mini-PC like a smartphone or raspberry pi to run. Let's not forget Pebble watches are just mini-programmable controllers that run a modified version of FreeRTOS and have a nice compact screen that fits on your wrist, and a mic.

TL;DR: you need a smartphone to run Alexa.

Well sure... but Pebble also isn't running Dragon Naturally speaking directly on your watch. It hits a "cloud" API with your audio and gets results back.

Why couldn't they do this exact same thing with Alexa? And from what I read you need to provide your own SIM to get Alexa working, why would that be the case if you were required to use a phone?

The Pebble Core takes a SIM card

The Pebble Core is actually a very small screenless Android-running machine. It has more in common with your phone than your Pebble (assuming Android / Pebble owner ;)).

> TL;DR: you need a smartphone to run Alexa.

Or Pebble Core :). Which seems to be intended to let you leave your smartphone at home when you go outside.

But I can't wait to use it as a companion. Google Now generally sucks, and the only way I found to make it usable is to hack it with Tasker. I don't mind carrying Core all the time, along with the smartphone.

(I just need to figure out how to keep it alive for the whole day with the 3G always on.)

The Pebble watch can communicate with a smartphone, and uses voice-to-text for SMS responses. I imagine a similar implementation could be used for Alexa. The bigger hurtle may be the lack of speakers on the watch.

I have a Bluetooth headset bought specifically for the purpose of playing with voice assistants. Pebble Core will support Bluetooth, so I look forward to pairing them this way.

Core charges in two ways: power connector that attaches to the 3.5mm headphone jack, or via the attachable plate that allows wireless charging. Thus, you'll probably be able to attach a small external battery pack to make it run all day.

That said, I'm more interested in figuring out how to get offline speech recognition running on Core with something like PocketSphinx and only relying on the 3g connection when needed.

All this has got me thinking about what the smallest Android boards/devices out there are and how I could get a Raspberry Pi Zero connected to a cellular network. Need to do more research on this.

To build on that idea, you could use PocketSphinx for (battery-draining) always-on listening for a keyword, then sending the actual recording to a server for transcribing.

Alexa could be running on the smartphone paired to your Pebble, though. It already does voice recognition that way.

If Apple sees it is gaining some traction then you know what is going to happen - Siri coming to Apple watch

Edit - Sorry, I thought Siri is not on apple watch

Siri is on the Apple Watch. https://www.apple.com/watch/more-to-love/

However, unlike Pebble Core, it requires your phone to be nearby for Internet connectivity.

My theory on Pebble Core is it's a way for Pebble to get experience building products with cellular radios. In another generation or two I wouldn't be surprised if Pebble watches could connect directly to the internet.

isn't Siri already on Apple watch?

Why Apple Watch sucks:

- ugly design - overpriced - no clear killer app

Why smart watches are inevitable:

- payments - swiping through subways, doors, hotels, airports - effortless exchanges between individuals (replacing business cards and cash transactions) - absorbing the fitness market Fitbit etc

Recently I've been getting very annoyed at Apple's terrible software quality. I've also had an opportunity to use Android, and I've found it a far _inferior_ experience (unfortunately).

I want to leave Apple but it's still the best option, even after a year of tanking quality.

Conclusion: we need a new Apple.

I was actually doubting Pebble would survive given the last product (Pebble Round) and the way they were marketing themselves (like a kid's toy) BUT then they pulled these new products out of nowhere and now I'm left with my jaw on the floor. And not just because of how quick they recovered and improved on the last product flaws, but also at the GENIUS and potential of what Core could be. Have to hand it to Eric and the rest of the Pebble team. They're doing great work.

Also, how cool is it that Pebble stems from that Steve Jobs story about him polishing pebbles? https://www.youtube.com/watch?v=K-Yv-UdsmSo

Curious, what do you see as the major flaws of the Pebble Round? My brother bought me one for my birthday a few months ago, and it's been fantastic. For me it's the perfect size (no larger than a normal watch), amount of functionality (get/dismiss phone notifications, control music, track steps) and battery life (days).

IMHO, the HUGE bezzle. It detracts in all manner of way to the appeal and of course, surface visibility of the device.

Fair aesthetic criticism. I don't find it to be an issue or distraction for my uses.

So, let me get this straight.

You're unhappy with the two dominant players in consumer electronics. Given the vast number of things we need/expect software for, neither of the two dominant players satisfy your requirements.

And you're conclusion is that we need a third, new dominant, player that will also try to satisfy your vast, disperse needs?

See, I reach the opposition conclusion. We need less "this only works with that" type of devices/services. We need more interoperability. We need to be able to choose the best watch / phone / calendar service / office suite without worrying about the other things.

I wasn't really referring to the type of ecosystem Apple had, more that we need a new company who can design brilliant products in the way that Old Apple (Jobs Apple?) repeatedly did.

If you get a new Apple to create your perfect smartwatch, what is that killer app that they should create? That's an odd & very unspecific feature to ask for.

Sorry if it wasn't entirely clear, but the second list were the killer apps I was referring to.

> swiping through subways, doors, hotels, airports

This is not going to happen. We already had a technology that was meant to be this - RFID. But instead of one RFID that you tie to all devices and services, the lack of standards caused your wallet and key ring to be crammed with unique tags for each utility. And the tech industry now is even more wall graded and alienated to standards then 15 years ago.

Swiping through subways is already happening. You can pay your fares with Apple Watch on the tube (London subway). See https://tfl.gov.uk/fares-and-payments/contactless/other-meth...

Also pebble has a documented interface that people can build things that physically connect to the watch and add new features. Using this is Pagaré: https://www.pagare.me/

> Pagaré is the simplest, fastest and most secure way to add contactless payments to your Pebble smartwatch.

Things are converging around various standards now, which is good, although further progress on that kind of thing would be great. Whenever something needs to tie some information to me, it'd be nice if I could just present any device/card that conforms to a particular standard for NFC and register that (hotel entry, train tickets around the rest of the UK, work pass card, all loyalty cards, gym entry...)

I know, but that's just using it as a debit card, not an Oyster card.

Using it as a proper travel pass is far more interesting, because that will be an instance of using a watch as a custom access card, which will lead to the popularisation of the idea for everything from workplace security to hotels to boarding passes. It would be a gateway app in terms of normal people seeing the utility of a smart watch.

Microsoft introduced the tablet PC in 2001, nine years before the first iPad. As with Apple's watch now, the experience wasn't worth the cost. Clues about what the future will look like are already all around us and I think this is one of them.

If you have a phone on you, a screen and direct manipulation will always win out. There's too high a perceived psychological barrier compared to being able to 'do it yourself' with your hands. The echo works because you don't have any device to manipulate, and often you're using it while shouting at it or are busy with your hands in the kitchen (and even then the majority of its current usage is timers and music -- the easier stuff that works decently well with voice).

If the watch itself has data (cellular or otherwise) and you don't have your phone, that's when there's some advantage to being able to do this. However it's unclear when that's better than having a phone -- if its about going on a run then dedicated fitness trackers might be all you need compared to a watch that has to solve all problems with a terrible and tiny UI.

There are many reasons smartwatches (not just apple's) have been doing poorly on the market -- it's a demand from industry rather than consumers, and the advantages have not been clearly communicated at all (the most personal device ever? compare that to the iPhone's you can email/web anywhere anytime -- clear value).

Well, don't forget that the article talks about the Pebble Core, which is the screen-less processing module on the new Pebble Kickstarter. There is no direct manipulation (other than the two buttons on the device) or feedback. It is intended for development or things like run tracking.

If you are running, it may be considerably easier (and safer) to use a voice interface like Alexa than it would be to use a screen. Which is probably why they focused on Alexa for the Core and not the watches.

Slightly OT question: Would the Pebble 2 be useful as a privacy-respecting health tracker?

I am looking for a device for long-term heart-rate tracking (additional included gimmicks appreciated). But I am worried about things like: Does the health app upload to cloud services? If so, can it be turned off? Is the Bluetooth synchronization protocol documented to support open-source app implementations?

Pebble engineer here

> Does the health app upload to cloud services? If so, can it be turned off?

It does by default, but you can disable it. I'm actually working on making those options more user friendly :).

> Is the Bluetooth synchronization protocol documented to support open-source app implementations?

Not officially, but we've published open source implementations of at least part of the protocol (see libpebble). Several unofficial implementations followed (e.g. a Blackberry app, among others).

Blackberry, eh? Good choice! How's the Psion Revo support coming along?

No bluetooth on the Psion Revo :(

Our psion support only goes up to Series 5, sorry.

Health data per default uploaded, are you honestly nuts?

They're not nuts. It's the standard default everywhere.

Remind me, who else does allow to disable that upload?

Even worse if everyone does it. Sometimes I wish that some wearable companies get slapped in EU courts for violating the now-defunct safe harbour with personal health data.

I wish that too, but in the meantime, if everyone is doing something wrong, then let's give some praise to those who are doing it less wrong - this way we can create an incentive for others to follow.

The built-in apps seem to be heavily cloud-based. You can turn them off but you can't use them in local-only mode. However there are third-party apps that can collect data on the watch and store it on your phone. I wouldn't be surprised if there are a few that keep data on the watch itself.

Edit: Sibling says you can disable uploading, but I just looked through all the menus on my Android app and can't see how.

After an exchange here (and fra answering that very question as well) the Pebble support team explicitly wrote this in a mail:

So Usage Logs send step count, and sleep count, to Pebble's servers. Usage Logs are also necessary for Voice responses. The voice is sent to our voice service, transcribed and sent back (this voice data is not stored at any point).

You can use the Pebble Health app with the Usage Logs switch off and your data will not be sent to Pebble's servers. You can also sync that data to either Google Fit or Apple Health if you choose while still keeping the Usage Logs switched off.

In other words: 'Usage logs' seems to be the all-in-one switch that sends the data to the cloud, everything Peble Health related should work if you disable that. Ignoring the limitation mentioned: You won't be able to use voice replies.

So I'd say that the Pebble is a really decent choice for a privacy concerned person.

On that note: I certainly hope that Alexa can be disabled on the Pebble Core..

> On that note: I certainly hope that Alexa can be disabled on the Pebble Core..

It should be. a) Pebble Core is described as a pure Android device (i.e. easily rootable, no permanent preloaded crap), and b) they answered on Reddit that Alexa requires you to authenticate with your Amazon account, which you don't have to do - and then Alexa won't be enabled.

I would not advise to use or buy pebble if you're concerned about privacy. As a company, they don't seem to be particularly interested in it.

For example: I was investigating some time ago (might have been updated) how to register a new pin in the timeline from the phone and apparently the way to do it was to send the contents of it to Pebble servers. Assuming their app uses the same API, it means every single calendar item you see in your timeline is actually sent (and possibly logged) in their servers.

I raised this point to their support and in pebble forums and got no answer.

I assume this is because there may not be sufficient storage on the watch itself, nor the phone, and you don't want to drop potentially important data!

My understanding is that all the data (apps, notifications, etc.) on the watch effectively sits in a cache, which pulls from the phone which is just a secondary cache of all your stuff in their cloud. This removes storage as a limitation. Then it just becomes a matter of ensuring your most pertinent data is in the right caches.

I agree however that it would be nice for Pebble to explain how they protect your personal data in their cloud. Hopefully they don't mine/sell it.

Sure, but it should be possible to only keep the data on your phone, and not let it leak all the way to the pebble cloud. It's a bit of a strange decision to force people to use their cloud :|.

Yeah. Personally I don't care that much about what they're doing with the data on their cloud, but it seems weird that my watchapp (or a companion app) needs to talk to the Internet at large to push pins onto the watch timeline. Doubly weird because I hoped to use it for pushing data generated by my phone, and there's no reason this shouldn't work completely off-line.

On their pebbledev Slack, some of the engineers talked about how the timeline system works a while back. While end-user apps go through Pebble's servers for timeline posts, the local Pebble app on the phone can also add timeline entries; your calendar data stays on the phone and the feed to the watch is a merge of the feed from Pebble's cloud and the local data.

You're right, the public timeline API does require hitting our servers.

However, the calendar pins we push from our mobile apps do not use that public API and thus do not - as far as I know - hit the cloud.

Any plans on changing that? I mean, there doesn't seem to be anything in Timeline that inherently requires routing stuff through your server. That's the one thing that always perplexed me - you could do almost anything to Pebble without touching the outside world, and here there's that one thing that suddenly requires the cloud. If I may ask, what is the reason you built it this way?

You can put things in the timeline from an external place, so I can push something to many timelines (e.g. sports scores) using pebbles API. That definitely requires going through pebbles servers.

Makes sense, but it still shouldn't preclude my companion app from pushing pins to my watch's timeline completely off-line. The sane model seems to be a timeline stored locally on your Pebble app that can be synced to the Pebble cloud. What I understand from the current model is that Pebble app is just a dumb cache.

I wonder if Amazon has ever considered buying Pebble.

It seems like they would want Alexa on every platform, so owning a platform (other than the Echo, which they needed to launch it) may not be advantageous. Though you might say the same for the Kindle, and that didn't stop them from making their own.

You'd think so, but Amazon doesn't think that way. They don't sell chromecasts or Apple TV's in their store, when they'd make far more money selling them and getting buyers to subscribe to Amazon Video...

Is there a confirmed analysis of this? Just curious because I'm sure Amazon probably has an army of excel analyst warlocks (and chief warlocks) + a board to answer to...

I was a happy user of the Pebble Time. However, I recently decided to switch to the Fitbit Blaze to participate in some physical activity challenges with friends and family. So far the experience with the Fitbit Blaze has been bad.

* Music control intergration with Android is flaky and rarely work * Face of the watch is hard to see outdoors * Minimal Android phone notification integrations / settings * Watch face selection is minimal * No third party app market

My Pebble felt like a very well crafted device, with solid software backing it. It wasn't perfect but when compared to the Fitbit Blaze it is far superior.

I am really looking forward to coming back to Pebble Time 2 at this point.

Alexa is already in the app store. What's the story?

Thank you! People all caught up in the buzz.

Granted, I'll participate in the Kickstarter, but this isn't truly news once you realize Core will be running Android. I don't think that is common knowledge.

Does the Alexa app actually let you use voice to ask questions and issue commands, or is it just an app to configure your Alexa devices?

Oh. It's a companion app. I didn't check closely enough. I only checked to see if there was an Alexa app.

When I read the announcement the thing I really wondered was: If Amazon has an Android port for Alexa, what's the motivation for making an exclusive deal with Pebble? Why is that preferable to offering it on the app store to everyone?

Dumb question maybe but why not bring this to other pebbles? There's a mic and you have a network connection via smartphone.

My first guess would be that it was just easier to right it directly for the Core as it's running Android.

Alexa only has an audio output at this time. We'd love to bring Alexa to all Pebble watches! Tell amazon to add the api :)

The API doesn't provide access to cards?

Anyway, even if it didn't, the watch could still just display a speaker/microphone image and a graphic indicating audio playback.

Honestly, this is my personal hold-up on buying a smart watch. I'm pretty set on getting one with Alexa built-in.

Would you not be able to proxy that audio through Nuance and display text on the Pebbble screen?

Core runs Android, while Pebble watches don't.

That said, it would be a great partnership for Amazon to connect with the Pebble app on people's smartphones to allow for this. As another poster has suggested, Amazon could even buy Pebble at some point. While as a Pebble owner I'd be sad to see them lose their independence, some solid financial backing wouldn't be a bad thing.

Ah ha, that's the context I was missing.

I remember reading the article about Pebble and how smartwatch growth had not panned out.

With a capable voice interface though, I can see the possibility of that magic combination / killer app. I wonder if the Pebble Core will be as competent as the Echo in filtering voice. Is this going to be a passive-listening device, or do you press a button?

I'd assume push-button, since passive listening would drain the battery pretty fast.

Come to think of it, there's no battery-powered passive listening solution currently. Hmm...

Sorry for the late response, but I apparently left it open over the weekend.

I believe that the Core has no inbuilt microphone, instead relying on a bluetooth headset / 3.5mm jack for sound in/out.

How about "Amazon Alexa comes to countries outside the US"? That would be nice.

Because having an always-on microphone in your house wasn't enough, let's now put one on your wrist.

Pretty sure the core isn't always on.

On top of that, the Core is probably less capable than your cell phone - which could already be listening to you 24/7. You'll trust Facebook, Microsoft, Google, and Apple in your pocket everyday but not Amazon? Get real, if someone wanted to spy on you, they wouldn't wait for you to buy a core.

This is specific to the Pebble Core, which is not a smartwatch and isn't worn on the wrist.

You don't have to if you don't want to.

Better leave that Internet-connected microphone that's in your pocket at home in the freezer...

The Pebble Core doesn't have a built-in microphone or built-in speaker. You need to either plug in a headset with a microphone or use a Bluetooth Headset with microphone. If you used a wired headset with no mic, it can't listen to you.

Apple should be starting to get worried. They had a good lead in innovation but it looks like "everyone else" is catching up again. If apple isn't bleeding edge anymore, is it relevant?

to me it felt like Samsung had a much better product until Apple caught up with the iPhone 6, it feels like a long time since Apple was "bleeding edge".

They still just inch it IMHO - but fading fast

not a few Apple shills on here I see.

Just so ye know - PR, managing expectations and policing discussions boards aren't a good way to promote a high end devices company. Developing high end devices is, and useful services to go with them is "For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled." - a quote I'm sure Jobs was familiar with.

> "For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled."

If XXI century is proving anything it's that while Nature cannot be fooled, humans can, and rather trivially so. More and more businesses have discovered that spending money on marketing has much better ROI than spending it on making a useful, valuable product.

Want a quick proof in a tech sphere? Just look at the IoT market.

I don't think that's any proof - like you say it's "burgeoning" and complaints abound on security and reliability. I doubt it makes a huge amount of money at this stage.

I don't believe the 21st century has shown that at all - it wasn't marketing that made apple the behoemeth that it is ("it just works" was more than just a slogan). Look at the declining fortunes of IBM & HP??

Yeah individuals can be fooled, but en mass things revert to a nature state. "you can fool some of the people some of the time but you can't fool all of the people all of the time" as true in this century as the last

Anyone relying on "notifications" to define their wrist-worn device is in big trouble. I just bought a Garmin Fenix after debating to: wait and see what Apple has at WWDC, or wait for the Pebble to ship. I ended up with the Fenix, and commented to the wife: "notifications is the baseline, hell, just about everything in Garmin's lineup can do that, and that's not their main business."

So we see Pebble trying to define themselves with something other than just "you have mail!", only on your wrist. To me, Apple seems to be trying to do to much to not just be "notifications on your wrist", but not doing any of it all that well. I ended up with the Fenix because after two versions of Pebbles it seems my main use case for a wrist-worn device is fitness stuff, which Garmin does really well. Pebble looks to be trying in this next iteration. But do I wait on Pebble and take a chance they get fitness right, or do I get the watch from the company I've been buying from for years and have been happy with, now with wrist notifications? Apple's a non-starter, given that I have to bring a phone on runs (umm...no).

In summary, when buying a thing that goes on my wrist, notifications are a given. What else do you have to offer? And that differentiation is what will decide winners and losers, IMO.

>we see Pebble trying to define themselves with something other than just "you have mail!"

In fairness with the Pebble Core, we have just that. I love the idea and I can't think of anybody that's tried anything like it. A cross between a iPod shuffle, a wireless modem and a keyfob. I really like the concept as a personal area network focus besides my phone.

Oh, and as regards "just" being a notifications device - bear in mind Pebble were the first, it's simple, and they're cheap. Three areas the Apple Watch just can't compete on. They've gone all wrong - it wasn't glamour that made Apple - it was solid quality engineering-led product development.

Oh, I'm trying to be fair, and I see Pebble's branching out as a good thing for them. I just don't see them sticking around otherwise. But I wasn't going to pre-order (oh, sorry, "back") the next Pebble in the hopes that it can match Garmin by September. As I also told my wife, "getting notifications to a Bluetooth-enabled device is a (mostly) solved problem. Getting the fitness aspects right is a much tougher problem". Pebble has shown evidence of figuring things out after a few iterations, so they can pull it off.

Yes, Pebble's big selling point is price. But they also show their inexpensive price. My Pebble Time Steel is a scratch and scuff magnet. The bezel is huge. But you get what you pay for, and I still feel the Pebbles are a good value. Hell, the Fenix is a $600 device. Sure, it has sapphire glass and hardy exterior, alitimeter/compass/barometer, but still...$600. I can put the Fenix to good use and get my money out of it (big runner and hiker), but I'm guessing I'm not the mass market.

Apple, though, I dunno. Sure, they've sold bazillions of Apple Watches, but I'm not sure where they go from there to justify the price. Ours is an Apple household, but the Watch was never on the radar for us. No GPS in the device was the deal-breaker for me (see: fitness features), because I'm not going to carry a phone while running. Other aspects didn't work for me, but if the Watch were just a bit more money than the Pebble I might have gotten one. But it's a lot more money. Again, maybe I'm outside their intended demographic, and not well-qualified to say what the future holds for Apple's device, but I just didn't see the value.

Thanks for the info on the steel. Wasn't sure about it but durability is a big red line for me. I just have the original "white" pebble, which is pocketmoney price and arguably one of the nicer ones. I backed it on KS cause it was a new class of device. Same reason I'm backing the core. Disclaimer: I do get exercise but having a fitness tracker isn't a big deal for me and I'm quite happy to bring my phone most places - but am still on the smaller form factor devices.

Normally i dismiss most people who are wary of using a lot of services on AWS/GCE... due to concerns of lock in since I think the benefits of more rapid development outweigh the costs.

However, this seems like a terrible strategy. Are they really outsourcing all of the core service and voice integration to Amazon?

This totally reminds of windows. Microsoft built the core operating system which became the defacto OS to build programs on. Hardware manufacturers then had to compete with each other almost entirely on hardware specs and price since most programs would only run on windows giving microsoft an almost complete monopoly. Microsoft makes $$$ as monopoly, hardware manufacturers make hardly since they are all competing on specs and price.

Amazon is probably laughing their way to the bank on this one.

Are they really outsourcing all of the core service and voice integration to Amazon?

No. It's an Android device and the Echo integration is totally optional.

Reply on news.ycombinator.com