Hacker News Reader

Comment by donor20 | Hacker News Reader

This article is pretty interesting.

- Supposedly Amazon has operated things like AWS at a loss for long periods of time (I highly doubt this - AWS has never been cheap).

"And it's through this mass adoption of Alexa devices that consumers (and the army of human beings Amazon hires to listen to consumers) can collectively train Amazon’s voice recognition system, which is then monetized through Amazon’s primary source of profits: Amazon Web Services."

There is no way Amazon has echo / alexa out there just so it can develop a language recognition model. They could just record their customer service reps talking to customers, train it on the millions of movies and shows with subtitles on amazon prime or train using audio book narrations. And the profits from their transcription service are not likely very high relative to other services they offer.

Whoever becomes the device that people want in their home, the one they turn to to coordinate their heating and cooling, arming their security system, getting the weather and traffic is going to be in an incredible position.

Google is also chasing this market, and apple to a degree. Google has an incredible voice recognition model already. They are chasing this market because it is THE most powerful and personal way to have a connection to the consumer. If you don't understand WHY these companies want apple siri, google assistant or amazon alexa everywhere you don't see the future. These companies will OWN consumer loyalty. A child growing up in a amazon echo / alexa integrated house may become a lifelong brand consumer including when they buy their house and setup their home automation etc.

The other thing, despite all the claims about "amazon microphones" - people will PAY for these - by the truckload. If I could get a space already pre-integrated with home automation etc? Very appealing. And normal people do call these things echo devices or alexa.

These companies are trying to deliver things to consumers that generate high brand loyalty. It seems to be working. I noticed this when browsing for cars - EVERY salesperson emphasizes that their cars supports "apple" carplay etc. My guess - 50%+ of even CAR buyers want to have a connection to their brand loyal assistant.

View on HN

> My guess - 50%+ of even CAR buyers want to have a connection to their brand loyal assistant.

For me it's that I don't want to be beholden to the car manufacturers schedule of updates or choice of features. The only thing that really dated on my previous car and tempted me to look at buying a new one was things like lack of bluetooth audio and no big screen for maps.

As an example, it tends to cost extra (at least it did historically) for maps in cars, I've also heard of having to pay annual fees for updates to those maps and even when the maps are "upto date", their information is still not as good as say Google Maps on your phone. So you land up with these expensive luxury cars, with huge infotainment screens, and then people still using a smart phone with its little screen on the side with its maps app simply because it's substantially more usable than what the car provides.

Additionally, each car manafacturer has their own version of media players and capabilities all with varying amount of features. They all insist on figuratively re-inventing the wheel, and landing up with one which is still way worse than my rapidly evolving phone and its apps.

Car infotainment systems these days should be little more than a largely dumb interface to my smart phone so that for the 10-20 years I own the car, I get regular updates on its utility as my smart phone and its apps improve.

My parents recently bought a Peugeot 2008 with a very crappy infotainment system but it doesn't matter really because it supports Apple Carplay so now it has onboard Spotify, Waze and Google Maps.

I agree completely. All I need are manual stereo controls and a nice spot in the middle of the dash to mount my phone without blocking air vents or windshield.

> Car infotainment systems these days should be little more than a largely dumb interface to my smart phone so that for the 10-20 years I own the car, I get regular updates on its utility as my smart phone and its apps improve.

That doesn't align with car makers incentives and business plans. Right now, all the talk is about mobility services (which don't make a profit for anyone currently) and chasing Tesla. They need something to replace spare parts and servicing, which is where all the profit is. With COVID, the EU commission fines on CO2, and the beginnings of EVs taking over (=spare parts and servicing will be reduced to almost nothing), they need a future profit generator. Except that doesn't line up with what consumers want.

That doesn't align with car makers incentives

If does if lack of CapPlay/Android Auto prevents sales.

I won't buy a car that doesn't have them. I don't want to install more apps just for my car. I don't want to learn new systems when the native integration works just fine. I certainly don't want to pay for updated navigation when I get it "free" from Apple or Google.

For a while, that meant no Toyota/Lexus (it appears they've since caved and have native phone integration).

This is the worst thing about my 2017 Highlander. No idea why Toyota thought that infotainment system could compete - it’s pretty terrible.

I’ve seen videos of people installing CarPlay compatible head units while maintaining all of the original car features, but it kind of spooks me. I’ll have to build up the courage to make the switch.

If you're going to have the car for 10+ years, why in the world would it be a problem to learn the system in it?

a) many people dont' keep cars that long b) the system is probably poorly designed and implemented

You have no argument here.

the average loan for a new vehicle is 3-5 years. Is it your belief that you're likely to default on that loan, therefore why would you learn the system in your vehicle?

The average ownership length for a car is roughly 6 years. Taking the time to learn a system that you'll have for anywhere from 3-6 years is completely reasonable.

This is the very definition of #firstworldproblems.

I don't think it's that hard to understand. People want cars to be compatible with their phone system because it's easier and it works better.

You certainly can learn the lousy interface in your car and there is some value in that, but it isn't hard to understand why the sales & marketing team is emphasizing that you don't have to (whether or not true in practice).

Yep, I'm pretty sure that's the reason, they're having to be dragged kicking and screaming to give consumers what they're asking for.

Every time I took my car in for a service and would look at cars on the showroom floor, I would ask about Apple CarPlay options and tell the sales person how I have no inclination to buy a new car until the current one becomes problematic or the new ones have the option of me not having to rely on their not great UI for the lifetime of the car.

Porsche did something I thought was pretty smart. They introduced a line of radios with Car-Play functionality for their cars going back to the 1960s.

https://www.porsche.com/international/accessoriesandservice/...

> and no big screen for maps

One of the reasons I haven't bought a new car is because I pretty actively don't want a touch screen in my car. It just screams unsafe to me, and all the research I've seen coming out recently agrees with me.

I don't want to have to stare at my console just to change the radio channel, to turn up the A/C, etc. It's insane to me that people think this is a pro instead of a negative.

It’s a pro the same way web apps are a pro: cost to manufacture & deliver.

You don't have to stare at the screen in a Tesla to change the radio or turn up the A/C, those controls are on the steering wheel.

Here's what I actually said, with added emphasis.

> I don't want to have to stare at my console just to change the radio channel, to turn up the A/C, ETC.

https://en.wikipedia.org/wiki/Et_cetera

> et cetera is a Latin expression that is used in English to mean "and other similar things", or "and so forth".

I'm sure this will come across as harsh as it's meant to be, but you went after a strawman so fair is fair.

My actual point is that if you're doing ANYTHING while driving that involves staring at a touch screen to ensure you touch the correct part of it, often multiple times as you hunt through menu's, then it's less safe than not having the touch screen.

This point is not invalidated because you can change the volume via your steering wheel.

The fact that you're using the actual radio at all puts you in the minority of new car buyers. Pretty much everyone I know under 30 just plays music or podcasts off their phone (which is an even tinier touchscreen). I don't see that trend reversing anytime soon.

I agree that controls for things like AC on a touch screen are a terrible idea.

I use Waze all the time as it shows speed limit and warns of road blocks by cops (which are notoriously corrupt where I live).

But for a lot of cars in my experience, the screen is only used for infotainment in which case I would love to have Waze to show on it. I always plot my route before starting to drive so I have no need to touch the screen during the trip.

Fortunately push-to-talk and volume control buttons on steering wheels is very common these days, so I don't need to touch my screen to skip tracks (or switch between pre-programmed radio stations), change volume or initiate a hands free phone call. Siri activated by the push-to-talk button is also good enough to use to message back to people you're driving and will get back to them later, or tell your spouse you're on way to them now.

If interaction with the big screen were purely passive then I wouldn't have an issue with it. The problem is that it's a touch screen and is, by definition, not passive.

For the way I use the screens, to all intents and purposes they are passive while I'm driving. I don't have a need (or want) to touch them at all while the car is moving. I set up my map route before I pull off, often on the phone before I get in the car, then that's it for the rest of the trip.

Controls for everything else like AC, volume, switching radio station/track is not needing to be done with the screen, it's all through classic buttons/knobs or steering wheel controls. I would not buy a car which required the screen to be touched for these things.

I don't see anything objectively more or less dangerous about operating a radio in this way, the only real difference is that instead of your radio frequency being visible through 7 segment displays, it's now on a high resolution display.

> I don't see anything objectively more or less dangerous about operating a radio in this way

In my current vehicle I can change stations without taking my eyes off the road, I don't feel as if I'd be able to do that if it were on a touch screen due to the lack of tactile feedback.

Also, I've rented a few vehicles and I've never seen one that didn't have things like A/C on the touch screen, which is part of why I've formed the opinion I have. OTOH, what I've rented are minivans, where A/C is a bit more complex since it will often have a "driver/passenger" dichotomy as well as a "front/back"dichotomy.

There are brands that have not lost their mind. Many new cars do keep physical controls / buttons / dials for A/C and other stuff out of the touch screen. If you want to dial radio when having navigation on the screen, it might be a different story, though.

Not to be a shill but you should look into Mazdas. They feel similar to you and have removed touch screens, and the since you don't need to reach it they then moved the infotainment screen further back and higher up so you don't have to take your eyes off the road as much.

I still don't want Waze on my car's screen because it sits low in the dash. I intentionally mount my phone such that it's high up and I can still see the road while looking at my phone.

It's absolutely less safe to look down and away from the windshield in order to navigate.

It's a fine balance between restricting the view slightly with a smartphone holder vs keeping both driving and navigation in the same approximate line of sight.

But it's better than looking down and to my right to navigate (or change radio stations, or adjust climate control, etc etc)

I have a car from 2017. it has a modestly large touchscreen with android auto / apple carplay support, but it's old enough that most of the basic car functionality (lights, A/C, seat heaters) is still implemented in hard buttons. this is the sweet spot, imo. if desired, I set up navigation and queue some music while parked. there's no need to interact with the touchscreen while the car is in motion. I do find it much nicer to have maps on an 8" infotainment screen than on a tiny phone screen with a mount that's suction-cupped onto some part of the interior. it's not so great to be worth buying a whole new car over though. as long as you have bluetooth in the head unit, a phone in a mount with a charger is still pretty good.

yeah, if it were passive only, I'd 100% be ok with it. Being able to slap the map up on the larger screen is a definite win.

I just don't want to be interacting with it in a touch screen way while driving.

While their end goal is pretty clear(brand loyalty, profiling).. i doubt that it will come to pass with our current tech.

Lets face it - IT professionals do not trust such systems, nor use them - they are inferior to any type of input mode at the moment.

Most normal people do not care about them, nor use them.

There is vocal but relatively niche group of tech enthusiasts who are really into it, but they are already a brand loyalists.

On a side note - the whole concept of brand loyalty is completely idiotic from consumer PoV. There is no merit for the customer in it.

My kitchen Alexa has an excellent user interface for kitchen timers and casual music listening IMO.

For me the killer app for voice assistants is in the kitchen 100%. Being able to juggle a bunch of timers, control the AC and fans, dictate to multiple lists, play music, control the TV, and respond to texts while my hands are covered in food is easily worth the like $40 the thing cost.

I'm not super impressed with any of the assistants "smart" features but in terms of being able to shout canned commands at it in a noisy environment it's top notch.

In this case "brand loyalty" is just the chrysalis of vendor lock. At first you use it because it's convenient, then the model changes and your digital ecosystem is held hostage. Microsoft did this with my generation and Office, Google is capturing the next generation with Docs and Chromebooks in every school. Amazon microphones endeavor to be brand-synonymous with smart homes of the future. The more they can integrate with your Amazon account, phone, tablet, Smart-TV, HVAC, security, grocery list, etc., the more "brand loyal" you will be.

>Most normal people do not care about them, nor use them.

>There is vocal but relatively niche group of tech enthusiasts who are really into it, but they are already a brand loyalists.

Are you serious? My 97 year old grandpa even asked me to set up an Alexa for him, so he could shout at her to check the weather. My most tech illiterate friends LOVE those devices, because they're so easy to use.

Yep.. and if a privacy-positive company (say, apple) were to create a voice calling device (Facebook portal, echo show), I’d buy it for the grandparents in a heartbeat. The current system involves a nearby grandchild driving an hour there and back every time we want to have a family video chat to get their computers back into the state we told them to never modify.

Have they perhaps figured that part out? And suddenly the computer is "broken"? Count on me to do that when the time comes...

Do you mean to say my grandparents are purposely breaking the computer for more face time with the grandkids? I could see it, though this has only really been a necessity post-COVID, when it’d really be much better for them to not have a lot of physical interactions with those outside their local friend pod.

iPad, child mode, turn off every other app from Facetime and Siri, and turn off the app store: https://support.apple.com/en-us/HT201304#allow-apps

There is a beautiful essay by Nassim Nicholas Taleb Nassim Nicholas Taleb titled, "The Most Intolerant Wins: The Dictatorship of the Small Minority" that bears reading. Vocal minorities do change society.

https://medium.com/incerto/the-most-intolerant-wins-the-dict...

The ability to say, "Google, text my wife and tell her I will be home after 8", is truly invaluable while driving.

It would be indeed a great feature if that was something only shared between you and your wife.

Alas you get a lot of extras to go with it: An algorithm decides that you then must be hungry so ads for Arby's (they really can't rely on word of mouth) keep popping up in your navigation.

After three such messages an algorithm decides your wife might be in the market for a divorce lawyer, because in 50%+ percent of the cases this might be a sign of marital disagreement. Your wife's newsfeeds change just a little because she is now deemed pro-divorce with messages about huge settlements and divorce horror stories, because those keep people engaged.

Safe driving...

Actually, I've never seen an ad in my navigation (Google Maps).

Really? Just the other day it told me to turn left at the Starbucks.

The system knows about some buildings as landmarks. I don't think I would categorize that as advertising; "turn left at the Starbucks" is an extremely human way to give directions. I haven't thought about it as advertising when my parents tell people to turn left at the gas station on the corner; There's a gas station on the corner. It's an easy landmark.

Two nights ago, I had Maps tell me "turn right after the McDonald's" because there was a McDonald's at the corner.

I guess, but the street had a name. And a Burger King, for that matter.

What are the map coordinates for us to look at and see ourselves?

None of that matches reality, but it makes for some fun speculative fiction.

I've never seen ads in Google Maps, are you using another app?

I think you're sort of... following a slippery slope argument past what is reasonable expectable.

Try looking up Noma, Copenhagen the best restaurant in the world (presumably). It then shows locations of other restaurants in the area. Aren't those ads?

As what is reasonably expected (an this is old): https://www.forbes.com/sites/kashmirhill/2012/02/16/how-targ...

It is not a slippery slope argument, because they have that data and can use it as they see fit. I just suggested 2 ways in which Google can make more money with information people shared with them. Why is it reasonable to expect them not to do it?

Im sorry, I thought we were talking about Google Maps navigation, not the desktop app.

on the desktop app, I don't see ads, but I see an option to "Show nearby restaurants". If I click that, the map zooms out and shows me nearby restaurants, not ads. It's all informational, not selling me something. In my work computer, I don't have uBlock Origin installed.

I see a lot of ads on Google Maps while navigating on the map as markers, while searching for something specific like a restaurant and if I make a route to go somewhere most of the time I get uber ads injected as an option I should take instead of using public transport.

IT professional here. I use a google home and trust it. I’m not a tech enthusiast.

Professional is a rather broad term. Do you work in ITSec or at least some related field?

dollars to donuts, app developer or db admin

PhD in CS from a top five program, specializing in mobile security. My day job is similar.

Why is it so hard to believe that people can be different from you?

so you work in infosec? or I'm right about your day job. Education to a Phd Level implies some serious specialization. You say "mobile", but those best practices move pretty quick and yesterday's degree doesn't do much to confer today's knowledge.

I know everyone has a different comfort level, but if you have an infosec background you have to admit you're giving away alot of information and potential control. Maybe you're ok with that.

> so you work in infosec?

Yes. My daily work is on vuln detection.

It's true that the field has moved since I graduated. But I still attend conferences and engage with academics for recruiting purposes.

I work specifically on mobile security. I have a PhD in CS, specializing in mobile security.

On the supply side, with a vested interest in people accepting these sort of devices?

You just keep looking for reasons why I must be ignorant or lying. I have no interest in trying to sell smart speakers.

What other information do you want?

The first line of this comment is your opinion, which can't be false, as it's just an opinion, but the other four lines are all completely false.

1. IT professionals absolutely do use such systems. In droves. It's true that we are more likely to be wary of them, but a large number of IT pros I know and work with do use Alexa/Siri/etc. for home automation and all kinds of other things. And no: smart speakers are not "inferior to any type of input mode", at least not in a wide variety of circumstances. I can turn my air conditioner on or adjust its temperature while lying in bed, without getting up, and without even waking up all the way. Try that with another "type of input mode". I can turn on lights while I am stumbling to the bathroom in the middle of the night, but I can easily use only 10% brightness so as to not wake up my wife. Try that with another input mode. I could go on for pages, but I trust the point is made.

2. There are at least 300 million smart speakers out there already, and the number is growing exponentially. Amazon alone already had 100 million sold by the beginning of 2019. I don't know the exact number right now, but it's a very big number.

3. See #2. 300 million units is not a "niche".

4. Brand loyalty, in fact, gives substantial benefits to the consumer, contrary to your assertion. Greater integration of devices and ease of use is the most obvious one.

> 4. Brand loyalty, in fact, gives substantial benefits to the consumer, contrary to your assertion. Greater integration of devices and ease of use is the most obvious one.

No, it's the other way around. I don't get updates on my 3 year old phone because I'm loyal to Apple, I'm loyal to Apple because I get updates on my 3 year old phone.

Dare I say it can be both? Some things are "virtuous cycles." You get updates, and in your experience those updates add value, which increases your expectation of value should you stick with this phone and possibly replace it with another of the same brand.

That IS brand loyalty, generated by delivering value, not just fancy advertising. And conversely, the experience customers have with their phone gives them certain expectations should they buy the matching AirPods rather than Bose or even B&O.

When it turns out that the AirPods have special integration with their iPhone, it sets up positive expectations for buying an iPad. And on and on.

The value creates brand loyalty, and the brand loyalty leads consumers to buy more products from the same brand, which increases their value because of the integration within the ecosystem.

When the brand is actually delivering the value, it's a virtuous cycle: value -> brand loyalty -> more value -> more brand loyalty.

Apple products are a bit special in that the integration has always been a huge part of their value.

And that goes back to Macintosh 128K.

Obviously, Apple stole mice and bitmapped graphics and GUIs from Xerox Parc. But a huge part of their pitch for MacOS in the early days was integration between apps on the platform.

Consistency of UX meant that there was lower friction for adopting new apps. Integration between apps such as being able to copy and paste text between apps from different vendors was a revelation by the consumer microcomputer standards of the day.

In an ideal marketplace the best product is selected on it's merit.

Brand loyalty, vendor lock-in, financial leverage, etc. are all attempts to interfere with free market, to prevent a consumer moving to a competitor.

There is also no reason you can't have integration between different vendors, in fact thats what standards are for. Apple, etc. are notorious for refusing to integrate to protect their advantage.

The have the right to do that to some, limited, extent, but to advocate it as good for the consumer is just self-delusion.

> In an ideal marketplace the best product is selected on it's merit.

> Brand loyalty, vendor lock-in, financial leverage, etc. are all attempts to interfere with free market, to prevent a consumer moving to a competitor

As I understand it, the only way to prevent vendor lock-in is regulation. Regulation is considered anathema to a “free market”. Subsequently, I’ve come to expect that a free market _guarantees_ that vendors will attempt to lock their customers in.

I suspect your “ideal market” is not merely an idealised fantasy but also self-contradictory and thus impossible.

I’d love it if you could explain otherwise because it’ll probably help me understand basic economics better.

My take is this:

Ideal market is certainly a fantasy, we have an entire branch of behavioural economics dedicated to irrational behaviour.

On regulation, general consensus is that at least some regulation is needed to enable fair market to exist at all.

In most countries in a state of disorder you get market manipulation, Pump&dump schemes, monopolies, cartels, financial pyramids, snake oil salesmen, etc. That's not conductive to economic growth.

Obviously it is hotly disputed where is the line between ensuring fair competition or, on the contrary, stifling the market.

Cosider car repair - is it the manufacturer's right to decide who they sell parts to, or are they leveraging a position in market A (selling cars) to secure an unfair advantage in market B (car repair)

Without regulation you don't have a market at all, you have war, where the best and cheapest way to beat your competitors is to murder them. Even the Mafia found it was better to establish a council to work out beefs and reduce inter-family violence. Market law goes back to the Middle Ages in Western Europe, where it was international in scope and enforced by private courts paid for by the merchants themselves.

We don't have actual war yet. But we do have a world of duopolies, in which two (or three or four) control almost all of the business between them. That's not good for customers or suppliers or retailers; it's only good for the rent-seekers who own those companies.

Unfortunately, the U.S. has for several administrations now basically not enforced the antitrust laws unless a short-term rise in prices is visible. Long-time rises in prices, as well as knock-on effects like reduction in quality, are simply ignored.

I would never turn my lights on while trying to use the restroom at night. I explicitly don't want to turn them on because it wakes me up fully. As for your A/C suggestion, that's what thermostats are for. Those have been automated for years now.

I agree with the poster you're responding to. The only true automation that hasn't been done in the home are those that are unimportant. I certainly don't need Alexa to open my garage door, regulate the temperature of my house, and so forth. If I really really want to be able to turn lights on and off without walking over to the light switch, I've had options for that for over 20 years now.

Between 9pm and 9am, when I press my bathroom light button, it turns on at 10%. Not something I thought I needed, but now its nice having a light switch that knows what time it is.

oh wow, that's a really good idea.

For the most part, having lights you can only talk to is kind of annoying. Having lights only on a dumb switch gets kind of annoying. Having smart lights on a dumb switch is kind of annoying. But having smart switches or smart lights with coordinated smart switches is great. Consistency and simplicity is key. You can get the best of both worlds. Being able to turn of 3, 5, 10 switches at once. Being able to change brightness with your voice is something I use way more often than turning the lights on or off with my voice. Turning off the closet or bathroom light once already laying in bed. Alexa and Google exist to let you mix and match brands. You can have 6 brands of lights, best in breed for each purpose or cost point (cheaper lights is lesser rooms) and nobody besides the person maintaining the system needs to even know.

The software and wifi connecting has sort of been trash, but the WeMo Smart Dimmer and WeMo Three Way are GREAT pieces of hardware as far as being a LIGHT SWITCH is concerned. It's the apple of switch design. A single button for on or off. Slide your finger up the track for brightness. Long press to control other switches on the network (glitchier.)

https://www.belkin.com/us/p/P-F7C059/ https://www.belkin.com/us/p/P-WLS0403/

The Philips ecosystem has a wide range of accessories, including smart dimmer knobs to control smart lights. https://www.lutron.com/en-US/products/pages/standalonecontro... LIFX now has a switch, but I've never actually seen it in stock so it may be vaporware. https://www.lifx.com/products/lifx-switch

It can be annoying when it glitches out, but overall its a positive change. Don't knock it until youve really given it a serious try.

How many millions are sitting in closets or junked? At least one over here. Occasionally my daughter plugged it in as a speaker, but rarely and not in the last year.

Wow! Who are these people using these microphones? I think I know only one person with one and he is an early adopter tech geek who typically buys every gadget. And he just uses it to set timers. I’m shocked to learn that millions of these things have been sold. They don’t solve a single problem I have, but instead introduce more problems.

With open standards i have no problem integrating anything cross-brands.

The only really hostile brands are preventing it to be honest(apple comes to the mind and so does nintendo).

Brand loyalty mattered in the past, where it was the only reliable way to judge new products in relation to each other. It only works to the merit of consumer if you were already subject to vendor lock in.

There are plenty of alternate input modes that work like that too - back in the 90s i had lightswitch that could change intensity of the light dynamically - no need to even utter a word so even less risk of waking your wife up. you can add module to AC and control it via smartphone/webpage, without risk of voice recognition getting it wrong. you can get lights that automatically turn on to 10% brightness when someone walks etc.

not to mention various privacy concerns for your guests - you did agree for amazon to collect the data, but did they agree too?

Speakers sold is kinda a bad metric - active users/usage would be much more interesting - as i wonder how many of them ended up in attic, or thrown somewhere because they were useless.

> Highly disagree

I think what people actually uses them for is very different than what tech and marketing think its for, but there is a massive market and everyday people love it to play music, call another room in a big house, control the light, check what's on tv tonight and if it's raining tomorrow.

Having my 64 years old mom emptying her fridge and then randomly saying "Alexa, what can I cook with VEGETABLE ? [answer] Okay, remind me tomorrow morning" was something very opening, as was "alexa, where is my PRODUCT" to hear about when its delivered.

Especially since I - a tech user - was a first buyer of many of these techs but mostly don't use them.

My mom doesn't know how to use tech. She makes a face when the TV remote has button that are not numbers and arrows, she still doesn't understand nor care about what the 3 buttons at the bottom of her android phone do (home / back / ...), but she has fully integrated alexa into parts her daily routine, yet completly left it out of others that it could handle.

The last time I saw something that eye opening about tech for non tech user was chromecast/firetv.

So yes, I totally understand why those companies want that spot, because whoever gets my mom queries gets her ad profile, gets her delivery orders, her music subscription, dictates what compatible brands she will use, etc ...

> whoever gets my mom queries gets her ad profile

This got me thinking —- are folks who can’t/won’t/rarely operate computers but are heavy users of voice control living in a post-ad world? Where would personalizes ads even be displayed? All the streaming services don’t have ads (at the right price point), and sure the buying products through voice devices could push particular brands, but amazon already does that with “amazon’s choice” products and I know many people who will only buy those, so nothing is really changing there. Similar for recommended content on streaming services.

Alexa, what butter should I order for this recipe?

Alexa, buy me the shoes that Kim Kardashian has on her latest instagram pic, size 7.

Amazon will control that gateway and can sell those answers to the highest paying advertiser. That's even better than paying for the #1 position on google search.

How in the heck are butter suggestions going to be a personalized ad? That’s the same as “amazons choice”, or super market branded butter. Same for the shoes, that’s even more not-an-ad, as you’ve specifically requested a make/model. Personalized ads work by getting you to buy something you’ve been considering or getting you to consider something you haven’t heard about, in both of these cases they aren’t doing that.

It's not advertising per se, it's having the first place on that query on the list of amazon products. If you search on amazon you have sponsorised products at the top, products that aren't at the organic top with amazon's ranking system but pay to be above it.

Right, that’s “Amazon’s choice”/similar. Which isn’t dependent on tracking/personalization, it’s the same as super markets getting paid to put particular products in better locations. The ads I’m talking about are when I’m on a random website and I see sidebar ads for products totally unrelated to that page, but related to my interests. Voice-only folks are post-ad in that sense.

And the original comment was specifically about an “ad profile”, which only matters for personalized ads. Without personalized ads, the ad profile is meaningless, so getting one via voice devices is not all that valuable. (Again, for voice-only folks)

That's not amazon's choice (which is still there too), that's sponsored content, and marked as such with a little "sponsored" under it.

I just searched "shoes" on amazon and made a screen capture but you can't post on imgur without an ad blocker apparently.

I was classifying all search result ordering manipulation as "amazons choice/similar", as they're all equivalent to a supermarket placing some products in desirable locations, and they all can not really make use of an "ad profile", as once the person is actively looking to buy a particular product, most of the work of traditional advertising has already happened.

This means the "ad profile" of folks who do not use screens and only interact with voice is relatively worthless: the best you can do is push them towards buying your brand when they're already looking for your product-class, and at that point the "ad profile" adds relatively little to the experience, probably just an idea of how much money you're willing to spend. Given Amazon doesn't currently adjust search results based on a prediction of how much you'll spend [1], this information is probably not that valuable.

Compare this to the value-add of traditional ad profiles: mine would say that I'm interested in, lets say, high end audio products, outdoor equipment, and dog products, which means anywhere I go online or on mobile apps I constantly get a barrage of ads for various products in these categories. If I see one I'm interested in, I'm likely to buy it right then and there. There's no voice-only equivalent of that, unless people start asking "tell me something I should buy" (Asking "tell me an item in {category} I should buy" is skipping over the ad profile)

All in all, this means folks without screens are post-ad: they will only be shown products they're already looking for (in adjusted order, sure), rather then being shown products they had no idea about but might be interested in. This is bad news for advertising firms and new product developers, though they probably weren't getting much money from these folks in the first place.

[1]: I only buy high-end audio products, I've bought many high-end audio products, and if I search "bookshelf speaker", I'll get cheap trash for the top results, telling me that Amazon is not interested in targeted whale-hunting, for whatever reason.

Those have specific answers, why would advertisers be involved at all? That would be asking Amazon to specifically bring in middle-men they don't need to go through.

"Alexa, what's the most healthy butter" would be a better example of something advertisers might pay for.

But it's not the same as advertisement today.

If I search for shoes on Amazon and I see an advertisement for a different brand amongst the search results, that's fine. I'll ignore it.

If I ask someone what shoes Kardashian was wearing and they give me a single answer that's incorrect, that's an entirely different thing.

I don't buy this at all. This modern incarnation of The Clapper has wide appeal. Everyone wants to turn things on or off. No one wants to get up.

I've been driving the same vehicle for 16 years and the only parts I've had to replace are the starter (< 3 months ago) and the battery (3 times in 16 years). You damn well better believe I have some brand loyalty.

As others stated, you're welcome to your opinions, but those aren't facts.

In our household, my wife loves our Google Home's- and uses them constantly. Playing music, broadcasting messages, adding things to grocery list, and following recipes are among her main uses. I tend to use it too for playing music and the grocery list.

In addition to all the other people saying how this is wrong, particularly for older people and people who are better at speaking and listening than reading and writing, there are several other use cases where they are a superior input mode.

As a detractor of these devices and someone who will never own one unless it does all the voice recognition locally, I found this to my surprise and regret.

First is when your hands are full. If you're in the kitchen, or working with tools, you can get a unit conversion, look up some fact, or change the music.

Second is as an input device to a TV. Particularly when there are a number of different applications you could open. Fidgeting around with a remote is difficult, and even if you had a keyboard it might be on the other side of the room from where you are. (Part of the problem being solved here is integration between apps and input devices, not just the voice input.)

Despite their advantages, the current situation is as if you could only buy a keyboard from 2 or 3 companies, and you were only allowed to type things that they allowed on it. And all your keystrokes were sent to their servers to be interpreted first.

AWS is not a loss leader. It's very profitable. $16.3bn in the last 12 months for AMZN as a whole. Most (if not >100%) of the profit is from AWS.

https://finance.yahoo.com/quote/AMZN/financials

All it takes is one major "trust incident" and accompanying press coverage to destroy "brand loyalty". There are still too many folks who see these for what they are: remotely-controlled microphones. That is why Amazon is asking for the wording of the article to be changed. Otherwise why would they care, if the manifest destiny you describe is true. More likely, no one can be sure yet whose normative behaviour will prevail: those welcoming the microphones or those rejecting them. It is what it is: an internet-connected computer with the mic always on. Eventually the Jetsons novelty wears off. Imagine every product with an embedded computer connecting to the internet without any user intervention and having a built-in mic always listening, purporting to be "voice-operated".

One can only speculate as to what sort of incident would threaten brand loyalty; no one knows. All we know is there is some history of opposition to mass surveillance.

>All it takes is one major "trust incident" and accompanying press coverage to destroy "brand loyalty".

What event could meet that rise to that level that already hasn't happened?

This reminded me; People seem way too comfortable / in the dark when it comes to voice-cloning, which, with the added data of general speech-patterns, is likely to become a major issue wrt political/legal issues.

see: https://github.com/CorentinJ/Real-Time-Voice-Cloning (think it made its rounds here on HN before)

Conceptually, even though it currently is a market-strategy where the benefits are as you point out. In the longer term this could be argued as an uncomfortable seat of power to whoever sits on the user-data. (without bringing up too much of that issue here... might be time to generate obfuscation of voice similarly to the obfuscation of google-searching which I've seen on here earlier=)

> ”Supposedly Amazon has operated things like AWS at a loss for long periods of time”

The article is implying that devices like the Echo are loss-leaders which drive profits elsewhere at Amazon.

As stated in the article, AWS made $11.3 billion profit on $48 billion revenue in the last fiscal year. AWS has been profitable since 2015, and is now the majority of Amazon’s profits.

>operated things like AWS at a loss for long periods

They are spending billions of dollars to expand into new markets and improve existing offerings. They are definitely spending more than they make. Of course, it would be ludicrous to suggest AWS is not insanely lucrative. But, technically correct is the best kind of correct, right?

>There is no way Amazon has echo / alexa out there just so it can develop a language recognition model.

Google famously did it with goog411, Amazon's claim here is the best kind of lie: it's not technically a lie

>These companies will OWN consumer loyalty.

Im locked into alexa. and now, i'm digging my way out. I fell into the trap!

I locked myself in and committed to it. The best time to invest in AMZN was 10 years ago, the next best time is now.

You too can profit off mass surveillance.

The majority of Amazon profit (not revenue) is from AWS. And AWS started because they already had the servers for holiday surge capacity. Was extremely lucrative, so they invested more in it. Maybe in the first years there was a loss there. They didn't start separating the numbers until 2012, and in 2015 their margins were ~25% on AWS [1]. So sure they are spending billions, but they are not spending more than they make on AWS.

Perhaps Alexa / Echo are not directly profitable as a device, but they make money from people buying stuff on Amazon.com with it, and gain that brand loyalty from it. Which has definite value. But there are estimates it costs somewhere around ~$30 to produce one [2]. So at ~$60, its very profitable. And when they sell them for $30 for a special or something, its still in the break even area. A couple sales and you are in the black.

[1] https://www.wired.com/2015/10/get-used-to-amazon-being-a-pro... [2] https://voicebot.ai/2018/02/15/apple-homepod-costs-216-manuf...

I think they meant that AWS is hugely profitable according to accounting profit, but is cash flow negative due to the large amount of capital spend. When you buy assets, that uses cash but doesn't reduce accounting profit.

> And AWS started because they already had the servers for holiday surge capacity.

That doesn't seem very likely. There's a limit to how much server work you need to handle one online shopping website.

What is the limit? It seems reasonable to me that it might change based on number of users, frequency of transactions, amount of user engagement, etc. After all, non-shopping websites have to scale with these metrics.

That is what I had read or heard at some point. The majority of their business was in the holiday season, and needed several multiples of servers to handle it. So in say May, 80% of the machines were sitting basically idle. So why not rent them out.

Unfortunately I appear to have been mislead by a myth: https://www.networkworld.com/article/2891297/the-myth-about-...

So, thanks for calling that out. Would have gone on believing it otherwise.

>AWS started because they already had the servers for holiday surge capacity.

That's a myth. Source - worked at AWS for four years, but also common sense. If you rent out all your surge capacity to customers then what happens when the holiday season comes? You kick all your customers off and shut down their business for a while? Not a business model for happy customers and not what happened in real life.

> Im locked into alexa. and now, i'm digging my way out. I fell into the trap!

I'm curious, how do you get locked into alexa?

i bought a ton of them and then hooked some zigbee light switches up. My own fault.

The only device I would allow in my house is a module that translate language. Locally within the unit.

So that it can be bundled with my controller that would make use of the translation/parsed voice command.

I do not trust Amazon/Google/CorpXYZ with running a microphone in my house.

Yet if you have a smartphone you trust the manufacturer with running a microphone everywhere you go.

Try putting a lav mic into your pocket or bag, and then listen how much non-muffled audio you get and how much you can understand from it.

Phones are in similar position. Assistant devices are not, they are usually in a fixed location with a unobstructed way to the environment.

I agree but to be fair my phone spends a lot of time laying on a table near me.

Placing your phone in your pocket to record audio discreetly works great, unless you're moving a lot. When you're sitting or standing still, the audio is almost perfectly clear.

It is worth noting that human voice is not the only audio you might want to collect from an open mic. Ultrasonic beacons are not uncommon.

By that logic I can only live in Faraday's cage or not complain at all??

You can complain about banking system and be proponent of bitcoin while having a bank account and paying for goods with fiat currency.

What's life without a smartphone, tablet or laptop like these days? Are landlines still usable?

Ah yes, the good old:

"Complains about pollusion, but still breathes the air"

That's a good one. I need to remember it.

If they had said "I don't like allowing CORPXYZ to run a microphone in my house" or "I have an issue with ... " that would be a reasonable objection.

Instead, this is "I don't allow anything to pollute from my property, but you're welcome to join me at my nightly bonfire".

I've started not taking my phone anywhere with me unless I think I may need it for something (grocery shopping, for example).

mobile phones are not nearly as necessary as people seem to think, especially during covid.

Satya Nadella made a parallel observation that operating systems have moved off of particular hardware (e.g. running Windows 95 on your PC) and into more of a consistent interface for interacting with all the devices in our lives, thus Microsoft Office's transition to a cloud service (and his general push for Microsoft to become more serious about the cloud).

Even though I don't care a bit about brand loyalty, I can definitely see the appeal of every device (PC, phone, car, TV, eReader, watch, whatever) all sharing the same abstract "operating system" interface, sharing your preferences, schedule, etc without the user having to even think about syncing this information.

There's obviously a lot of appeal to a corporation to have this data as well, but even those of us who are opposed to the corporate abuse of data have to acknowledge this is the future. I'm optimistic that an open-source self-hosted solution can accomplish the same thing, but I'm not sure one actually exists yet.

> Satya Nadella made a parallel observation that operating systems have moved off of particular hardware (e.g. running Windows 95 on your PC) and into more of a consistent interface for interacting with all the devices in our lives

They haven't. But Microsoft (and perhaps Google?) would like for this to happen - with their kind of OS, user interface and cloud "services".

> have to acknowledge this is the future.

Well, absolutist monarchy used to be the future, in the middle ages. Never too early to get started on bringing the future after the future.

> Even though I don't care a bit about brand loyalty, I can definitely see the appeal of every device (PC, phone, car, TV, eReader, watch, whatever) all sharing the same abstract "operating system" interface, sharing your preferences, schedule, etc without the user having to even think about syncing this information.

Apple has been doing this for something like 10-20 years now.

Wasn't AWS the only profitable part of Amazon, and was propping up their retail section so they could spend money on customer acquisition? Pretty sure I remember reading about this strategy -- they basically used it as a way to make it possible to undercut competitors and do things like taking a loss on shipping without basically going broke.

Yes and almost exactly what the article claims. Only it goes a step further to imply that Amazon is loosing money on other products to encourage AWS adoption. Not sure how the other comment interprets it as "AWS is loosing money".

If people are willing to install amazon microphone in their home. Why can't we call them amazon microphones? Why does amazon try to hide that under a meaningless word like "device"?

I think it's a reasonable distinction.

The Echo is much more than "a microphone". Yes, it has a microphone, and I don't think Amazon is trying to pretend it doesn't. But a smart speaker is obviously not just a passive listening device, as implied by "Amazon microphone".

When will the truth be palatable? When we normalize it. By calling the device an Amazon microphone, we are telling the truth that this device is a listening device, not just a speaker. This name has less sugar coating, so if it is still too bitter, just call it what you want.

By that logic, your phone should be called a "microphone".

Ignore all the additional stuff it can do, it has a microphone, therefore it's OBVIOUSLY a passive listening device, ergo "microphone" /s

Well it is quite more.

It also tracks your position via satellites.

I don't think they are, and Amazon might not think so either. I think people would be less willing to install these devices if they were more often called Amazon microphones.

Just like a Ring is a spycam in a doorbell. I feel shit passing these things when just walking the streets in my city, and look the other direction so the facial recognition that may or may not be there, does not catch me. But hey, that's the curse of being privacy-aware. I dread the day that I find Amazon or Google or Apple super-microphones running in a friend's home, and I try to avoid controversial subjects in our conversation.. just in case.

People who buy Rings know 100% that they're security cameras with a doorbell feature. They compete with other security cameras, not doorbells.

Then, when you look the other direction, the Ring on the house across the street gets you anyway.

Security cameras have been watching everything we do for decades but the one network vibe of these doorbells is kind of dystopian.

> There is no way Amazon has echo / alexa out there just so it can develop a language recognition model. They could just record their customer service reps talking to customers, train it on the millions of movies and shows with subtitles on amazon prime or train using audio book narrations

Training their recognition model on a range of pre-recorded, staged audio sources would likely result in it being very good at generating subtitles - but not much else. As for the telephone, again, it's relatively different (the customer reps will likely be using a script in some cases, and are going to be putting a lot of effort into being comprehensible). And neither would effectively simulate the situations (for example, a noisy family home) an Alexa will likely be used in.

On your point where you said "Supposedly Amazon has operated things like AWS at a loss for long periods of time" I agree AWS hasn't been losing money for a long time.

https://www.businessinsider.com/amazon-earnings-aws-amazon-w...

https://www.forbes.com/sites/petercohan/2020/01/06/how-much-...

edit:Formatting

> If I could get a space already pre-integrated with home automation etc? Very appealing. And normal people do call these things echo devices or alexa.

I've spoken with several architects and builders who are telling me more people want a "smart house" when they buy/design/build their next house. Builders are now creating model homes with nearly everything connected (alexa, nest, home apps, etc) and leaving some options open to their buyers.

This is very much something that is already here.

A child growing up in a "amazon echo / alexa integrated house may become a lifelong brand consumer including when they buy their house and setup their home automation etc."

Yes, and etc. What a distopian vision, they should make a Black Mirror episode based on this, or perhaps they already did.

Edit: there is an interview with Charlie Brooker where he says that ideas that are not good enough for the series they sell to Samsung.

> EVERY salesperson emphasizes that their cars supports "apple" carplay etc.

This is a completely different market and there are different motivations for people being so loyal to Android Auto and CarPlay. They're the first infotainment interfaces that do what they should have done from the beginning, just be a nice porcelain / glorified screen to my phone.

> My guess - 50%+ of even CAR buyers want to have a connection to their brand loyal assistant.

I don't think this is what drives it. Most car companies are just bad at (software) interface and inconsistent between makes and models; carplay etc. has the dual improvement of a) isolating you from these changes somewhat and b) usually giving a better UX

TLDR; Alexa buy batteries.

I’m never going tie my money to a voice activated anything. The last thing I need is to have a prankster on YouTube say “Alexa buy 1000 rolls of toilet paper” at March 2020 prices.

Sure, but the people who do buy things via Alexa buy way more stuff from Amazon. Just owning an echo suggests you’re one of Amazon’s most lucrative customers:

https://www.fool.com/investing/2018/01/13/look-how-much-amaz...

That article only says that Prime members spend more than non-Prime, and that Alexa users are more likely to be Prime members. It then uses the high sales numbers of Alexa-enabled devices to conclude that spending on Amazon is going to increase.

The problem is this:

> In a press release earlier this month, Amazon revealed the Echo Dot was one of the two "best-selling products purchased by U.S. Prime members from any manufacturer in any category across all of Amazon" (the other was the Fire TV Stick).

The month in question was December. All those Echos and Fire Sticks were being bought to give as gifts. How many of those get used a dozen times before the new owner loses interest and it gets left unplugged in a drawer? I'll be generous and say it could be as much as 55% since the sales were for Prime members and holiday gifts are frequently exchanged among family, so the Echo will end up in the same Prime household. But why should I believe that giving an Echo to a non-Prime member will increase their engagement with Amazon at all?

Well, who knows. Maybe soon it will have voiceprint and the delivery automatically goes to the prankster's home address. /s

Reply on news.ycombinator.com