The Vergecast - Google I/O and Microsoft Build

Episode Date: May 11, 2018

So much happened this week in the world of The Verge. Both Google and Microsoft hosted their annual developer conferences and announced a whole bunch of stuff, so Nilay, Paul, Natt, and Dieter gather to break it down and give you the highlights. And this wouldn’t be an episode of The Vergecast without the segment Paul does every week, “Kick flip the kickstand script.” It’s a big one, so listen to the whole episode to get everything you need for this massive week in tech news. 01:28 - 10 Biggest announcements from Google I/O 03:45 - The selfishness of Google Duplex 21:09 - Android P 26:41 - At I/O Google showed its willingness to change and shape our lives 34:33 - JBL’s Android-powered soundbar does a lot of things right 38:43 - Volvo’s native Google integration is the next level for Android Auto 44:06 - Six new Google Assistant voices, including John Legend 47:49 - Nadella’s Microsoft 56:51 - What is edge computing? 1:08:16 - Paul’s weekly segment “Kick flip the kickstand script” 1:10:38 - Net neutrality, mergers, AT&T, and Michael Cohen: what we know so far Learn more about your ad choices. Visit podcastchoices.com/adchoices

This episode of Vurchcast is brought to you by IBM. By 2050, the world population will reach nearly 10 billion, and food production will need to grow by 70%. What if artificial intelligence could help? Farmers are already using to help increase crop yields. Watson and the IBM Cloud provide access to weather data and analyze satellite imagery to help them monitor soil moisture levels and reduce water waste. So as the population grows, more food can be put on tables. Let's put smart to work. Find out how at IBM.com slash smart. Hello and welcome to the Vergecast, the flagship podcast. about parenting, babies, you know, the normal stuff that I talk about all the time. No, it's the flagship technology podcast in the world, I think.

I'm Nelai. I'm joined, as always, by Paul Miller. Hello, Paul. Hello. Deeter Bone is here. Hey, Deeter. You're back. Hi, I'm back.

Hi, everybody. And Nat Garen is joining us. Hi. Nat, you and Deeter have had a huge week because it's Google I.O. You've been out in San Francisco at I.O. And then Dieter, you also, it was also Microsoft Build, and you interviewed Sachin Adela, the CEO of Microsoft. You're just doing all kinds of stuff. Yeah, last week was a crazy week.

This week is also a crazy week. I unfortunately didn't go to build itself, but we can talk about that later. It was kind of a sleepy, sleepy sort of build, I think. Yeah, I don't think it running around the same time as I.O. helped. Yeah, that, that overlap was rough. Yeah. Well, massive week in news. There's so much going on.

We should start with I.O. I do want to say build is quiet, but Microsoft has been quiet lately in the sense that they're moving away from sort of their consumer-facing Microsoft existence to this very enterprise focused. We're going to build, I know, cloud services for people. We'll talk about that a minute. But in a normal world where they had a whole stage themselves, it still would have been quiet. But next to what Google just did at I.O., which might be the biggest, most interesting developer conference we've seen from anyone in years, it's tough. But let's start with I.O. and unpack that a little bit. They announced a ton of stuff. Deider Natt, you want to walk us through it?

I mean, they announced so much stuff. I don't think they just put up a blog post today. This is Thursday of 100 things we announced at Google I.O. And I don't think they padded those numbers at all. It's crazy. So they, of course, announced everything that's coming to Android P. They had just a ton of, you know, wild AI things, many of which launched right away. So, I don't know, Google Maps now is augmented reality and, like, you can point it at stuff and see it. You can just hit tab and Gmail to auto-complete all your emails. They announced release dates for their smart displays. They have six new voices for the Google Assistant.

Plus John Legend. Plus John Legend. They're very relevant news for me. They added more editing to Google Photos. They relaunched Google News as a whole new news app that's going to have stories chosen for you by the algorithm, which sounds quite a lot like the Facebook News feed. But without your friends in it, I'm sure that will never cause problems ever. They added some capability to Google Lens so that it's able to do OCR properly.

and you can copy and paste from your picture. And then, of course, this thing called Google Duplex, which is a robot voice that sounds so much like a human that it ended up creeping the hell out of everybody everywhere. It's like Bart Simpson mode for Google Assistant. Now your assistant could do prank calls on hair salons. Let's just start with Duplex. It's by far the thing that got, I mean, I want to talk about Android.

There's so much stuff I want to talk about. But we got to start with Duplex. It's the thing that broke out at I.O. The beta, it's not even a beta. It's a demo. It's the prototype demo that launched 10,000 takes, right? Because it's not shipping. It can only do.

Actually, I'm going to actually you there. I talked to someone, a person in charge on background a little bit. And it's not shipping, but they're not going to be testing it just inside, like, Google's labs with a bunch of people, like, watching it happen. they do intend this year in like the coming months quote unquote to have it work for real consumers out in the real world. But as a product or as a thing that if, you know, you happen to know Sundar, he's going to give you a download code? That is an excellent question. But they really do think that they can get it out and have it being really useful by real people this year. Okay.

So if you don't know what Duplex is really quick. Yeah. Sundar got on stage. You can go listen to the demos online. There's a whole page form. But sooner I got on stage during the demo and he said, we've developed the technology for assistant to actually make phone calls on your behalf. And then he demoed it calling a hair salon with ums and Oz.

And, you know, it got through some misunderstandings. And it just had a conversation with this person at the hair salon. And it made the appointment and hung up. And that was that. And it's a roe. having a conversation with a person where the person doesn't know that they're talking to a robot. Google has been very clear so far that it only works in limited domains as they're calling them. So I think what are the three things that can do?

Can schedule a hair appointment? Oh, her appointments, restaurant reservations and store hours. And then for the store hours thing in particular, they're planning on doing this thing where when holidays are coming up, they'll call a bunch of stores that they don't know the holiday hours for and then ask once and then update all of Google's dataset with the answer. So that's crazy. So it's a limited, it's just insane. It's very cool.

I don't think there's anyone who doesn't think just its most basic level. This is insanely cool. Then I think in a modern context, when every piece of technology that's ever been developed has been used for evil instantly, and we're having this reckoning about what technology should and shouldn't do and how it should be used, the obvious problem. with having robots call people on your behalf, people are identifying them and asking what we're going to do about them before the product has even come to consumers,

whether or not Google's actually given to consumers this year, as the reader you're saying. Do you talk to anybody at Google about that? Because having robots call poor, and you wrote a great piece about this, calling service industry workers who are already kind of overworked, like having wealthy people offload human interaction to robots that people who own stores just have to deal with.

That will increase the volume and increases their work. There's really big questions about they necessarily are going to have to record the calls. It's illegal to record a call in many states unless you have consent. I mean, there's just like ethical issues every which way you look. Did you talk about that with anybody, Dieter? Yeah. So my understanding, although we're double checking this because it was a really quick meeting, is they're not going to be doing this in every state because not all states have.

Some states require two-party consent. Some states only require one-party consent, and so therefore they could record the calls of that informing. There's also this whole question of how can they get this robot to talk to people such that they'll actually talk to the robot? Because Google does believe that they need to inform the people they're talking to that they're speaking to, you know, an AI instead of to a real person. They think they need to do that. But as soon as you hear that you're talking to a robot, what are the chances you're going to be? going to continue to talk to that robot. Why did they put ums in it if they're going to tell people it's a robot?

It just feels more natural. I don't know. For the same reason that when you call Delta, it makes fake typing noises at you in the phone tree. Yeah. There's nothing more insulting than the idea that I will be pacified into waiting for a slow computer than like, like what is that? I hate that.

But if you're talking to me and I go, uh, and then I say something, you will, you will, unconsciously and naturally make accommodations in what you say back to me based on the tone of my voice and how many ums and a ayes I've got, you'll wait a second longer, you'll maybe try and speak more clearly the next time you say it to me, you'll be more

forgiving when you need to repeat yourself. So it's ums as a UX. Yeah, basically. I think that there's a lot, like ums and oz they don't just make you sound human. They actually serve real linguistic communication functions when you're

talking to somebody and they are trying to build that into the way that this thing talks, not just to make it sound like a human, but also because they do genuinely like communicate something when you hear them. Yeah, because I think like most times you call a phone number and you can hear a robot and it's like, please say pharmacy for the pharmacy department, please say this for the other department. You like cut it off immediately. But I think the ums and odds make people a little more patient and realize that like, okay, this person's processing and information.

They're trying to convey something to me. let me finish, let me let them finish their thought. Yeah, there's, I think it might be from the book, Microsurfs by Douglas Copeland, but there's this famous description of Bill Gates that he doesn't say, um, when he talks, and it's the most unsettling thing. You, like, you just don't know how he feels because he never says, um or ah. Now, look, I think all that's fine. I think the idea that computers should interact with humans in a more natural way is good, fine.

but the question of should we outsource some of these tasks where on the one end there's a human being doing the work. And on the other end, there's a robot on behalf of the human being doing the work. It's kind of like what you really want is for your local business to just have like a scheduling API and for Google Assistant to talk to the API instead of you tell Google Assistant, which then tells Duplex, which then talks to a person, which then uses like their own scheduling software. Like, that's really, there's something squishy in there. The optimistic case, and I hate defending Google on this because they could have just addressed this. They could have, they should have known how much this would stand out to everybody as a demo. But the optimistic case, this is a reasonable step towards that future where it is an API. It's reasonable to expect in the future robots will talk to robots and solve our scheduling problems for us.

Maybe step one is that we start using robots. on the consumer end. But, you know, obviously Google, like Dieter mentioned, is also, you know, doing this in the sense of, like, we'll call once for holiday hours and then update everybody with that information, which in that case will reduce the load of phone calls. Nat, can you talk a little bit about your piece that you wrote about? So actually, funny story, Nat, I'm driving. Nat texts me. Google's robot voice reads out her text to me, and she says, hey, I'm just working on this duplex piece about service workers. And then I had to call her and be like, what the hell does Duplex have to do with Chrome browser technology?

It makes no sense. Service workers are... And I was like, what are you talking about? Service workers are an open standard. They've also been implemented in Safari and Firefox. I know, I know. Not at all what I'm getting at. No, the whole thesis to what I wrote about is the fact that I just couldn't help walk away from the demo feeling like Google,

used a bunch of human workers as they're like something to experiment their new technology on without letting them know, which felt really uncomfortable to me. And then the whole idea of outsourcing technology in the way that, you know, like online ordering or like just using the internet to make our lives more convenient has widened the gap between how people who are tech privilege interact with people who are, you know, underpaid workers who are working around the clock to serve your food or deliver your items and all these other gig economy stuff. Yeah, so I guess the feeling that I got walked away out of Google Duplex is just that this technology

at a glance, it's clearly very early. But the demo they gave was like, use Duplex when you're too busy because you're hanging out with your kids and not something that, you know, like they write in their blog post that it's going to help with accessibility and language barriers, which is fine, that's great, but they didn't really lead with that in the keynote. They talked about how, like, you can spend your time with your kids and, like, have Google Assistant do all your chores and errands for you. But I think that we've seen in the past with services like Seamless where people are using

Seameless to put in insane special orders to restaurants, so much so that Seamless turned it into an ad campaign, which is super obnoxious and crazy. And, yeah, like, I just think that the wider technology makes people who are using them. The farther these people get to not speaking to the people who are servicing them, the more we, as a society, just, like, have this gap between the rich and the poor. And I don't know, I just worry about that. And, yeah, like, the solve for that would be so that Google implements this on the business side so that businesses also have a robot that's handling this transaction

and just telling the business owners on, you know, here are your reservations for whatever, and you don't have to bother with them. Or the people who are talking to really rude customers on the phone don't have to talk to rude customers on the phone. That's all fine, but I don't know. Like, none of this was brought up in the keynote. Yeah, I don't know. Like, I, you know, I used to work as a waitress.

It gives me a lot of, like, has taught me a lot about how to be empathetic to the people who are working really hard to do it. And I just feel like it's, I just feel very uncomfortable about the whole idea of, like, people abusing this type of technology and kind of erasing people who work really hard. Yeah, that's a real, it's like such a deep, meaningful concern. And I think it's only buttressed by the idea that probably eventually you have to go get your haircut. Like they can schedule the haircut for you, but you, like, you as a person have to go.

And ideally you are polite and kind, the people who are working for you in that moment. And you're like, that is the root of. at every stage of this transaction, like, you're not scheduling an appointment to have a robot cut your hair. You're scheduling an appointment to have a human being cut your hair. And maybe that first interaction where you're scheduling the appointment is where that relationship begins. Like, I don't think that anyone has ever really thought that through because we haven't had to. Yeah, I mean, like, again, it's so early that I have no idea what this is going to look like if it ever comes out. I just wanted to, like, the first people I always think about when it comes

in new tech like this are the people who are often very invisible to tech and clearly the people who are cutting your hair and serving your food. Not often the people who are using said technologies we saw this week. Yeah. I think the big question for me is something you brought up, which is, okay, what if the restaurant deploys technology too? And now you have a robot talking to a robot, that might be the most inefficient thing that could possibly happen, right? Like if I get Google assistant to navigate the Delta phone tree for me, something has gone horribly wrong. I mean, I also don't think that this technology could ever scale to the point where, at least not in the next couple of years, where big corporations are now outsourcing their call centers

to robots, which is another issue on its own. But I think that, like, at launch, it seems like they're targeting small businesses that don't have, like, an open table booking thing that you have to call them on the phone in the first place. which is why it's so troubling to me. But yeah, I don't know that in the next few years this would ever scale to the point where, like, Delta is now going to employ these realistic-sounding robots to change your flight or deal when your flight gets delayed or canceled. Yeah, we'll see.

I mean, look, rich people have always had assistance, right? I mean, the very rich among us have traditionally not made these phone calls anyway. Now we're making that, like, more mass market. I think that will, that definitely changes things in a serious way. But what's amazing to me about this is that I don't think anyone is disputing whether or not the technology is powerful or interesting. The entire conversation is, how is it deployed, who is it going to affect? How will it affect those people? Do the negatives clearly outweigh the positives?

The one party consent stuff that Dieter brought up in those states, just to be clear to everyone, just for the computer to process what it's hearing, it necessarily has to make a recording. Right? There's no, it's not they're recording it so they can play it back for you later. I don't know who in the world wants to listen to a robot make phone calls later on its behalf. It's literally they have to, they have to capture the audio into a file so that, you know, a tensor flow unit somewhere can figure out what the hell is happening and then generate a response. There's no way to do this without capturing and recording that audio.

and making a copy. And if you don't change the laws, then this will never go out to every state. And I have no idea how to write that law. Just none, right? Yeah. Well, I've been talking to Sarah John about that. She's writing a piece.

And, like, one of the things is there are different laws for, like, data in storage and data at rest. And at what point does this data become, does your voice data become data at rest? It's very complicated. And, like, I'm, I am sure that the laws are not, as are currently composed, anywhere near being able to understand what to do with this situation. Well, yeah, but that's like, oh, man, this is such a rabbit hole. It's like, there's, like, copyright law, right?

So, like, early internet law, we had to decide as a country, as a policy, that if I request an image from the verge.com, that the copies are made on every router along the way and in my computer's memory. were not copies that were actionable, right? They're transient copies in memory. That's like we had to, there were lawsuits about that shit. Like big, meaningful, I got taught them in law school lawsuits about that stuff. And obviously, it makes no sense to treat those as actionable copies that create liability. That has nothing to do. And we'll have to do that with this stuff.

But that has nothing to do with wiretapping laws, right? Or call recording laws. Those are like a different, that's a whole different thing. So you've got these two entire different bodies of law that are both implicated by this technology that Google is like, well, here it is. Like everybody figured out now. And I honestly don't know how to write two-party consent state law that's like both parties need to consent to having this call recorded unless one of those parties is a robot. Like that doesn't make any sense. And like it's just like bizarre to me because you can't even get to the consent.

You can't have the robot in a two-party consent state. You couldn't have Duplex call you and say, hi, I'm a robot. Do you consent to have this call recorded? Because in order for it to hear and understand your consent, it has to record you. You could say press one. It's not listening until you press one. Yeah, but it has to say. But that's like nonsense.

Like, press one to have your call recorded by this robot. On the other hand, legal robocalls exist today. That's true. I think that is like the legal stuff is like, ah, but it's not so, like, so fraught that I don't think that it will be a huge problem. I think the bigger question is like the ethics of the thing and the ethics of informing somebody that they're speaking to a robot. And then if so, what's the likelihood that that person will be willing to do so and like actually want to talk to that robot when they call you, not when you call it. Yeah. And then you combine it with the deep fake stuff where you can like record a minute of audio of somebody's voice.

and now it actually is Obama talking to you to ask for money. Like, who, crazy. All right, let's talk about Android. So that was one down, 99 things to go. Good job, everybody. I think we're really on pace.

Oh, man. Halfway through the show, we're finally to Android. So, Deider, here's... It was almost to the end of the keynote of Google's keynote before they got to Android. It took a while. I was like sitting there waiting and waiting and waiting.

So, I don't know. Google split up everything into into three themes. So which one do you want to tackle first? Like they put it into general intelligence AI stuff. They put it into digital well-being. And then they had, you know,

the sort of, they called it simplicity, which is silly, but it's like, it's the gestures. It's the gestures. Let's talk about the gestures. I think we've done enough AI will doom us for one segment of the show. Okay. I will say one thing about the AI.

The idea of app slices is really interesting. Google's been trying to get apps to work outside their inside the app for years, and this seems like the really the best cut at it in a long time. So that's cool. So the gestures are there's a home button. You tap it to go home. You swipe up to go to an overview, which depending on whether or not you have a proper Google phone, will show you a Google search bar and predictive search apps.

And then you can slide the home button over to jam through a bunch of recent apps, and the back button sometimes shows up and sometimes isn't there. and to me the reaction to it immediately after, because it came out right away on beta on a bunch of different phones and everyone installed it was like, oh, this kind of sucks. I think it's jankier on the beta than some of the other builds that are out there. And in fact, I don't know this for a fact, but it feels jankier on the beta that I have on my own pixel

than it did when I was using it at Google's campus the week before. But they are mixing metaphors. There's still a tap to go home, but then a swipe to do other stuff. and it's a lot like the iPhone 10, except that Google has one more surface than the iPhone does. The iPhone has a home screen, a most recent app screen,

a multitasking screen, and a notification tray slash lock screen. They also have, you know, we could keep going to the other screens, but whatever. Google has one more. They have a home screen, plus they have an app drawer.

So that extra thing that they've got to contend with is still there. And so it necessarily means that they have to have their bottom bar gesture thing has to do at least one more thing. And in Google's case, it has to do two more things because they didn't go all the way towards trying to get rid of the back button. It's still there because too many apps still have a drawer on the left for the swipe. Wait, when you use the back button? The back button is there when you have an app that needs a back button. So basically every app lets you go back.

And so the back button appears. Not in the multitasking view. No, it disappears in the home screen and in the multitasking view. It only appears in apps. Right. Okay. So why are they doing this? That's like what I don't understand.

Like, I sort of get why Apple did it with the iPhone, right? Like, they got to get rid of the button. They want to be all screen. Like the hardware drove the software design. Why is Google doing this? That's a really good question. I think it's a trend. They just want to be trendy.

I think that's one thing. I think that they genuinely don't like the way that multitasking. has worked in their quote unquote overview screen before. They really did want to show multiple things. And not for nothing, but this also gives them another place to put their Google search bar. And then you do a Google search. And then maybe you might accidentally see an ad at some point in your future. So I think there's that.

Oh, no, I'm not kidding. I think they want to build to getting rid of the back button. And then they'll be able to build to not needing a bar down there at all because we'll all just learn how the gestures work. I think that's like the long term plan. But to get there, they have to do this half step. That's my best, like, most charitable read. The really, like, cold-hearted read, not cold-hearted, but like clear-eyed read is they just, you know, they saw the iPhone edit. They saw people liked it, so they're trying to do it too.

This is just a brief tangent. We don't have to go down the whole path of the fact that ChromeOS is getting Linux apps, and I'm very excited. But on ChromeOS, the Capslock button is search. So you hit search, and now I want to launch the terminal all the time to use Linux. Well, when you search on a pixel book, it's not like Spotlight. So I type in T-E-R. And the first result is Hungarian ter, which is English, is the word space in English. And then below that, I have the terminal app.

Like, they're conflating Google search with the things I actually want to do in my app. Maybe it works better on Android, especially with these like actions and stuff. slices and stuff like that. Yeah, over time they, if you were to keep typing turn, keep selecting terminal, over time it would become the first result. That's like their first answer probably. On Android, they sort of split stuff out. So if you search for an app, the apps appear at the bottom, which is closer to your thumb. And so you can like type TWI and then tap the Twitter icon.

And then above it at the top are, you know, a bunch of stuff you might search for that starts with TWI. So it's an interesting thing that they're doing with it. But so there's like stuff you might, you'll probably want to tap at the bottom instead of at the top because that's where your thumbs already are next to the keyboard. So that that is like instead of, you usually think it was like you want to just arrow down and hit enter. But because it's a tapping touch interface, you actually, they put the most important stuff at the bottom. Right. Yeah. I mean, Paul, the dream is that you will have no files.

You will only have what Google Surfaces is you on the web. You have no files, no applications. You will just have the cloud. And everything is an endpoint to the cloud. And all the buttons are search buttons. No, dude. It's all about the edge. The intelligent edge.

The intelligent edge. So, Deeter, so that is the gestures, which is like one big thing. And then there's actually more Android stuff in Android P, new features, ideas, than any recent Android I can think of. Because there's a gestures. There's a new material design. Yeah. which is like a wild. And then there's all the time well spent stuff.

Let's do time well spent and then let's talk about the design part. What's going on with these like stop using your phone features? So the main thing is there's a dashboard which shows you just a stupid amount of data. It's how long you spent in every single app every day for the past week and what time you use that app. So you can you can just see what you did today. How many notifications you got? How many notifications you got from each app? how long you had each app open, you know, blah, blah, blah.

And then you can tap on an app or slide over and, you know, pick Facebook. And you can slide over and say, okay, so on Monday I used Facebook for half an hour, but most of that time was at night. And then on Friday, I used Facebook for three hours. And most of it was in the morning because I was goofing off at work. And I really got to stop doing that. And so I'm going to limit myself to only use Facebook for an hour a day or 20 minutes a day or whatever. So you could just, like, lose yourself in this insane quantified self-worth. of all the time you're spending and all the different apps on your phone.

That's one. Two, you can limit the apps so that you can only have a certain number of minutes per day per app. And when you hit that time limit, it just turns gray and it won't open. And it pops up a little dialogue box saying, sorry, you said you didn't want to use this, so I'm not going to let you open. I thought it just turned gray. It won't let you open it. It turns gray, but it won't let you open it.

And when the dialogue box pops up saying you can't open it, there's no button on the dialogue box. Say, just kidding, I really do want to open it. You have to go back to the dashboard and change your time limit in order to reopen your app. I love it. Okay. This is like the Amazon like kids free time or whatever their product is called. It's like literally you're teaching your kids how often they can be spending their time on a device and now we have just made an adult version of it. I want this, but I want to give the keys to like friends and family. Right? I will aggressively No honestly

One thing I would love to do If my friends used Android phones more Is to just take their phones away Put that on so that when we go out to dinner They can stop Instagramming their food And taking stories about the fact that we're hanging out And just like talk to me Can I just a phone notices

That you're around other phones of people you like All the phones shut off Whoa That would be great No that would be awesome Everyone to just opt into like This is dinner and then everyone says yes, we're going to dinner

and it can be tied to like Google Maps or you know, Maps is going to have that thing where you all agree on some business to or a restaurant that you guys can go-to together. And then when you all show up, your phone graze out and then you guys are forced to actually have dinner together. That would be great. And if there's like an emergency and you got to turn on the phones,

you all have to like do a dance, you have to like a gesture that's a dance move. Like I don't have to like shake their hands in the air. Can I, can I, I'm really excited about all this, this well-being type of. stuff. But can I also point out that this is a developer conference. What does Google do at developer conferences? It tells developers, here's new, cool ways to make apps. You know how Google has all this data about when you're using apps and how much you're using apps and what time

you're using apps? Guess who else? Google likes telling that information to you. To developers. Why does it tell that to develop, like a literal example Google used on stage at I. It was like, hey, you noticed that a user opened your game and then quit after like a short amount of time. You can use that to optimize your game difficulty so that people use your game more. Yeah. Like Google is turning around and like not selling, kind of selling all of this data to developers to make their apps stickier. Yeah, the crashlytics is now being integrated with Google Analytics. So developers have a single dashboard to see and they'll know which people are more like.

to spend a couple dollars on upgrades and two dots, Nat. I will never. And then they'll be more likely to try and offer you purchase opportunities. And like Paul said, if they see somebody dumping out at a level a bunch, the next time they open the game, they can make that level easier on them. So they're giving developers a ton of usage data on how people use their apps. And then they're just now starting to give users usage data on how they're using their apps. Wait, but is that in conflict?

Just to play, I mean, I think it is, but to play devil's advocate, is that actually in conflict? Right? You're letting the people who sell the products get more data about how to make their products better, which is not necessarily a bad thing. And you're letting the people who consume their products have more data about how they're consuming it so that presumably they're adults in, you know, live a life of moderation and good jobs. judgment. Again, a bit of a stretch. I feel like this is a leveling of the playing field. Now you get the data that developers have been using to, in a negative light, manipulate you to use their apps more.

No, but you're not getting that data. The goal, I guess, is quality over quantity, right? It's like if you only limit yourself to 30 minutes or whatever, then you're going to use those 30 minutes wisely on whatever app you're opening. I mean, that's the hope. I mean, if I picked up my phone and it said you have 30 minutes, use it wisely, every time I picked up my phone, that would be great. But it's not what it's doing.

Like, Paul, you're not getting that data. You're not, when you quit the game, it doesn't say, hey, we notice that you always quit games just when they get hard. We're telling developers to make the games easier. Right? It's just saying, this is how much time you're spending. It's letting you set your own limits, which is great. Yeah.

But it's almost like on the user side, there's this recognition that we're terrible at setting limits for ourselves. And then on the developer side, it's, here's some more information. that will help you make this stuff more addictive. That's why I think it's in conflict. I just, you know, the ideal is that everybody is an adult that makes smart decisions. I just think we know that's not true, right? At all.

Like that game's slidey on the iPhone, like, I'm like, yeah, you can have three more dollars. Like, here you go. Like, I need more jewels. Like, it's ridiculous. On the other hand, Google Play this morning pushed me an extra dollar so that I can spend money in games and I'm like, no. Wow. Not doing it.

Can't get me. A whole dollar. I pay for nothing. They will not get me. They can play ads for me, though. That's fine. Oh, that's awful. Yeah, Hulu has started watching the second season of Legion.

FX Now does a thing where they're like, if you watch this one interactive ad at the beginning, we'll cut out all the rest of the ads, and we'll just do one ad break per time, and there's 10,000 ads. So I'm like, sure, I'll do it. The interactive ad is just a video for a BMW, and at the end it just says click here. So that's fine. But then it's like I'm not interacting with anything. And then it plays the same ad seven more times to the course of the show for salad dressing.

Throughout all of this, I was like, I would have spent an hour doing a 360 degree turnaround of your new BMW, if not for this horrible experience that you've made me suffer through. Anyhow, that's ads, whatever. Look, I think this time well spent stuff, Casey wrote a great piece. Tristan Harris is the guy who has been pushing time well spent. He's the one who is smart for an addiction. Like, he's the one who's pushing that narrative. he was on stage of code. He's all over the New York Times.

Casey showed he published his slide deck that he published at Google when he worked at Google years ago. My favorite part, you should go look at it. My favorite part about this is that he mocked up a Verge article that was like Google has a team now. That's the best. I love it. So go read Casey's piece. It's just interesting that this has been building for quite some time. And now we're in the moment where, hey, maybe we're not so good at regulating our behavior on phones.

we need the technology to help us do it, is coming into the forefront. Okay, so that's, I think we've covered three out of 100 things. I want to talk about one more thing just to troll you. Are you aware of what this JBL Google Assistant soundbar is? Yeah, motherfucker. And you called it an HDMI switcher and not a receiver. I'm so confused. It's a receiver.

It's like a, it's like both. How is it a receiver? Yeah. Well, it's a receiver. It's got a bunch of inputs. It routes audio, right? It routes video. it's a receiver.

Okay, I guess... I'll sign it's a receiver, not an HTML switcher. To me, it's a glorified HTML switch, but whatever. Sure. The thing that's interesting about it is it overlays the Google Assistant on top of those other dumb HTML inputs, so you always have it

available. And what's cool about that is that means you can always use it to jump with a deep link directly to some show in a proper app that you want to watch. So do you remember the original Google TV was this? The Xbox 1 is this thing.

They've looped all the way back around. So the original Google TV, you plugged everything into it. You plugged your cable box into it into your TV. It did HTML pass-through. Xbox 1 does HTML pass-through. This just has more inputs. And they would overlay Android TV on top of whatever. The Xbox 1 can overlay the Xbox interface over whatever.

So I think this is really smart. I think putting the intelligence into the sound bar is I think you're going to see every company start to do it. The big question for me is whether they can watch and interact with the TV content, right? So if you plug your cable box into the JBL soundbar and that into the TV, are they going to be able to say, oh, we know you're watching X. When you're done with it, do you want to watch this related YouTube video that unpacks all the Westworld theories? Right. That is the dream.

I just don't know if they're going to get there. And that also, by the way, tons and tons of data privacy issues there because then Google's watching everything that you watch, which no one likes. Right. Or we see that you're playing Xbox and we know you're having a really hard time getting past

this level of Far Cry 5. Here are some YouTube playthroughs that will show you this. Like, can they do that? Because they could do that. A, it's really powerful and interesting, but B, it's also terrifyingly creepy. Yeah, I think the answer is, Definitely not.

This is like a pretty like standard issue Android TV set top box that just happens to be crammed into a soundbar. Yeah. No, I think it's really cool. And I love the idea of being able to call up assistant wherever you are. You have a Vizio TV that has Alexa, right? Or do you have the one of the older ones? I don't know, man. I'm so angry at that Vizio TV that I've refused to think about it or do anything special.

I bought a Harmony remote, the one with the hub that, you know, is Bluetooth and whatever. and I hate it because it thinks the TV is from 1995 and so it goes through all these careful time-delayed steps when it would just be faster for me to pick up the app remote, hit a button, and have CEC turn everything on. So yeah, I'm full in on, I'm all in on H-DMI, CEC. It's a perfect technology. The only thing will make my life better is an IR repeater.

Oh, God. No, but like, there are lots of TVs that have a system. There are Sony TV with Google, runs Android and has a stick. You plug the Sammore into it. what answers you? Unknown, right? Like, you got to figure

like, who is this thing for? I do think this is a lot smarter than sort of the Sonos play. I think Sonos is, you know, there's an event coming up. The product's been leaked. It sounds like a new play bar. I think that's being made for smart TVs.

So if you have like a Samsung smart TV and you have a Sonos rig, you can like plug this into it, you get better sound and it's a Sonos. It doesn't run any of the smart TV smarts on it, right? Yeah. This is the smart TV stuff. So this is like you have a much older TV that doesn't have any smart features or, you know, it's old and it's running him slow.

So you upgrade your soundbar, you get free Android TV. As always the big question is like, where does the processor live? Do you want it to be modular? Do you want it all to be built in? I'm into this thing. But it's a receiver, not HGMI switcher. I'm just letting that right now at the end. Now, was there anything super cool about the Android Auto thing that you looked at?

Well, so the news with Android Auto is that last year, VOLBOS said that it would be working with Google to pre-install Android Auto and just make their infatement system an Android-based system. This year they have it for demo, and the cool thing about it is that you don't have to have a phone to be able to use it. So whether or not use an iPhone or an Android, you can use the same Android auto system in the car, and it'll look like a Volvo skin because they have their own version running on Android P. So the cool thing, Yeah, like the fun thing there is that you can have it adjusted to your profile. So when you get in the car, you pick the driver.

Like, you pick your own profile. The seats adjust to your height. All your favorite locations are preloaded to the Google Maps on the screen. You can adjust to your favorite stations and all that stuff, which is really cool. One thing that is kind of odd to me is like there is a play store that you can download stuff from, which it's neat, but I don't know the kind of stuff they'll allow on it. They said that it's going to be a bunch of media stuff so you can listen to podcast music. And there's some messaging stuff and there's RCS so you can get the chat protocol stuff sent to your phone or your screen.

So, yeah, that's all cool and fine. But, yeah, I don't know. It just looks like a giant Android tablet. And I feel like even though it looks really, I mean that in a positive way. How could you mean that in a positive way? I mean, because it's not like the Tesla version where it's like a massive iPad Pro with way too much. much information. Their version is like four menu bars and it gives you the top thing is maps. The second thing is media. The third thing is messaging or phone and the last thing

is like calendar. Fine. There's also the second screen that's the one behind the wheel that gives you like a Google Map navigation thing if you have it turned on. I think it's pretty neat. It's not going to be available until next year, the 2020 models, which means that you can't order it until late next year. But yeah, that's kind of the thing that's new at Android Auto. I don't know. Like, you hate Android Auto. I hate it on my car. It's super crashy on the Pixel 2XL and my Honda Civic.

It's just the worst. That's partially because my Honda Civic is running on Android 6.1. Oh, right. So because it's running on Android P, which apparently assistant can make appointments for you at some point, Volbo thinks that in the future, the car, because it's running on Android P, will, like, recognize when the car needs maintenance or some other issue. And in Google Assistant, it might be like, hey, your car is due for me. maintenance, would you like me to make this appointment? And you can just say yes while you're driving

and it'll handle that for you. So there's that component and we can go back to the whole ethics problem, whatever. But yeah, like that is a thing that Google and Volvo thinks that you can now do from the car, which I guess it's cool and kind of weird and creepy, but cool. And it's Volvo doing it. They will presumably install software at their dealers. So you push the button and it like does that over software, right? It's not. I highly doubt Volvo is going to have their own cars call anyone but Volvo dealer for service. Yeah, yeah, for sure. I mean, absolutely. I think that's definitely true. Yeah, but like the demo version that they had for us didn't, it was absolutely like a very half-paked version. You're like half the time it wasn't very responsive when I touched it.

So like it didn't really get to do that much with it. No, that's just Android Auto. Hey, man, Android Auto worked great in my Ford Escape. I'm just saying it hasn't given me any problems, but I just don't know why they won't make an Android auto app for the iPhone. There's one, you can run Android Auto on an Android phone. Just you push the button and it shows up and it's the display. Like, something to run that on an iPhone. I'll be happy. Like, I run Google Maps on my iPhone in the car all the time anyway.

It seems like a missed opportunity to get everybody who is already using Google Maps in their car on their phone to this other kind of Google experience. just do that. I don't need this. Anyway, whatever. Well, how else will they sell you new cars if you can get it on your phone? Yeah, it's true. There's like there's a, there's a, the stats a couple of years old now,

but your point is correct that more people buying cars were looking at CarPlay Android Auto than any other feature is the driver of purchase decisions. I don't know if that set still holds, but it was definitely true a couple years ago. All that said, what if I want to buy some like cool old Camero? Like, I just want to run Android Auto on my phone.

Well, then you go to Volvo, and then you rip their Android tablet out of the center of their car. Just pull it out. And then attach it somehow to your old fancy car. Yeah. The producer of the Cirquebringer show, Creighton, he was telling me that he wanted to buy a Labaron. And I was telling you, like, don't do that. Like, that's a horrible idea. Buy an old Camaro instead.

So that just sent me down. Like, I got nothing to do. The kid sleeps on me for, like, hours a day. I'm just looking at Camaro's son. Oh, no. Oh, no. Very nice life. It's really bad.

I wasn't looking at a laborans. I promise you that. All right. Four out of 100 Google things, I think we've covered the waterfront. Anything, Deere, anything else from Aya that really stands out? You want to talk about material design for two minutes? Just what's happening with it? No, we've talked about Google for an hour.

They've made it so you can customize material design to your preferences. The end. So Google Assistant this week, they announced that they will have now six new voices, and it's male, female, different accents. It's kind of like Siri when Siri had a bunch of different. like new voices. Kind of cool. And then they did the gimmicky thing

with John Legend, which is really fun. But apparently none of these voices have names. They're going to be colored, which is weird. The plan is to give, instead of, they don't want to give the voice a name, because they give the voice a name, then it'll be gendered,

and they really like that Google Assistant isn't gendered. But if you're going to give it names, if you're going to give it different voices, you've got to give them names. Instead of like voice one, voice two, voice three, apparently the plan is eventually to give each name a color. So there'll be the orange voice and the blue voice and the yellow voice and the red voice.

The red voice. The green voice. They're going to start with Google's primary colors. But eventually they're going to run out of colors. And there are going to run to colors that should not be named for people's voices. Yeah. Yeah, because I don't think anyone feels good about having a white, black, or brown voice. Or just the ethics around that is really uncomfortable.

Really uncomfortable. Confident Google. of all of the pitfalls that you should avoid, that one seems very obvious. Although yellow is part of Google's colors. Yeah. So I don't know how they're not going to navigate around that one. This episode, Virtress also brought to you by HBO Silicon Valley. As you might know, Kara Swisher from Recode has been on Silicon Valley a number of times.

And this season, they're doing kind of a crazy ad thing. Russ Hanuman, the venture capitalist character from Silicon Valley, is on RICOD decode as the character, Russ Hanamon. So listen to Kara, interview Russ Hanneman. This is going to be why I'll check it out. Today's show is brought to you by HBO, and today in the red chair is Russ Hanneman. He's one of Silicon Valley's most notorious investors, and he's recently emerged as an aggressive player in the cryptocurrency market. Welcome to the podcast, Russ. Thank you, Kara, and you're welcome, by the way.

For what exactly? What I mean for what? I basically invented the podcast. For what? You invented the podcast? I put radio on the internet. That sounds like a fucking podcast to me. Not that I'm making shit off it.

That actually brings me to my first question. The standard internet funding and sales models have served you pretty well over the years. But now you're jumping feet first into ICOs. Why? Kara, this town is filled with assholes getting rich off crypto by doing Jack. The Winkle Voss twins put in some loose change five years ago. Now they're Bitcoin billionaires. So yeah, I'll buy a ticket for that fucking ride. You don't feel like you've already missed getting in on the ground floor?

If I could change the past, I wouldn't have a kid at home right now snorting up my ADHD meds. I can only focus on the few years. future. HODL, bitches. So I'm hearing you already have taken 36 companies to ICO. How have you fared so far? Well, you know, I'll be honest, Kara. It's been down, you know? It's been up. It's been mostly down.

You know, 35 of them have, you know, fucking tanked. 35 out of 36, what happened? I mean, this is the game, all right? First, it's the SEC. Then it's one of your founders running away with your cash. Then it's a bunch of fucking hackers deciding that instead of edging in their basements that afternoon, they're going to come after your blockchain. Then one of your CEOs dies, like a pussy.

Anyway, listen. I'd rather focus on my successes. My success. One of them worked. And what was your ROI on the one that worked? Radio on Internet? No, return on investment. Return on investment.

Yeah, I know. It's 300M, all right? That's a million. And it's on some thumb drive in the middle of a landfill. My boys are on it, though. You ever lose a drive with a ton of crypto on it?

No, Russ. I haven't. Yeah, you have. Uh, no. But thanks for coming on the show, and good luck with that thumb drive. Watch new episodes of Silicon Valley Sundays at 10pm on HBO.

We were saying at the beginning of the show that Google obviously had this huge event. It kind of overshadowed what's happening at Build. But Build was us this week. Big announcements for Microsoft at the scale that Microsoft is not making big announcements. But, Deeter, you interviewed such Nadella, and we were talking a lot about how

it seems like it's very much Nadella's Microsoft. It has been, obviously, for several years. And Windows is becoming a much smaller piece of that puzzle. What was your takeaway from talking to him? The main thing is that it is, Windows is a smaller piece of the puzzle, but that they're going to be trying to work harder to make sure Windows will work with more devices. But they're much more interested in you thinking about how their AI and Azure stuff will work in little IoT devices. So it's the Azure Cloud Edge, if you want Paul to get into explaining what Edge computing is.

I did. Paul, I loved your piece. I want to hear more about that. Go through it all at first here. So, I mean, the news from Build is they're making this thing that will have your phone sort of appear inside Windows, so you can do a bunch of phone stuff directly inside Windows. They are going to have Cortana and Alexa talk to each other a little bit more. And so I want to tell this story about how Windows is the most open operating system. And it's going to try and work with everything. But I can't quite tell that story because Cortana still, you know, it'll talk to Alexa, but it won't talk to Chrome.

That's a sure bet. So that's a pain. They're going to have timelines so that you. You can pick up and drop off tasks from your phone to your computer. They're also going to be giving a bigger cut to developers, which is cool. I wish everybody would do that. They're going to mess around with tabs. There's a bunch of other stuff.

But the thing that I mostly talked to Nadella about was, you know, got him to like basically say, yes, we're the Windows company. We're still, Windows is still a thing, which was nice to hear that he isn't going to shut it down. But then they're really trying to tell a different story about what they are. And what they are is not relevant to consumers for the most part. Like the Windows stuff still is. But, you know, whether or not a drone is able to look for a dent in an oil pipeline is not a story for me. It's cool that I could pick up an off-the-shelf drone and have it stream video to my laptop. And the laptop can use machine learning to look for the dent in the oil pipeline.

Or you could just build Microsoft software directly into a big industrial. drone and then it can do it. All that's very cool and all they're, you know, they're doing just fine as a company and they're making a ton of money, but they're doing it in a way that is completely different from those other companies. And so I came away from it basically like, I'm not angry at Microsoft anymore. I've spent many years being angry at Microsoft, especially about Windows phone. You just missed it, guys. You just screwed it up. You had all this potential and you just wasted it. And I'm not angry anymore. It's like, yep, there's that gap between what you wish Microsoft was and what they are.

And it's fine. They're just more like IBM now. It's fine. That's how I feel about Microsoft right now. He also talked a lot about, you know, the ethics of AI, and he, like, is saying the right things about it. But I have more mixed feelings about that having come out of both Microsoft and I.O. But that's maybe too long of a story right now.

He also said privacy is a human right, which I thought was really interesting. Yeah. Okay. So, no, I'm going to get into it. So he said we feel a great responsibility for users. protecting users privacy. Sundar has also, Sundararararar at Google has also said we feel a great responsibility. And I'm tired of tech CEO saying that they feel responsibility because you know what? I don't care how you feel. I care what you're actually legitimately responsible for yourself.

There's a big difference between feeling responsible and being responsible. And until there are actual consequences to you screwing up your AI and doing something bad to the world, then feeling responsible is just puffery. It just, it's a nice thing for you to say and I believe you that you actually do feel bad if something happens and you are taking some actions to try and prevent it from happening. But unless there's teeth, it's not a real responsibility. Like, if I don't turn in my draft and I, you know, whatever, then, like, Nilai yells at me and like there's, like, actual consequences to me not doing my job. So if you say that you are taking on responsibility, I want to know what that actually

means because right now what it mostly means is you're telling me that you feel responsible. And I just don't care. I really noticed this. I wrote just sort of a like a McSweeney's piece about. if Google could make a Paul. But what I was thinking about was this idea that I have a way that might be inaccurate, but I have like a narrative in my head how I describe to people how Apple tries to keep your data private. I have a narrative in my head of how Microsoft or Google are trying to do that.

And Google's products especially are very reliant on knowing everything you're doing at all times. They know what restaurants you go to, what you like. They now know everything you ever point your camera at. Opening your camera app is a Google search. I don't have a narrative that can say, well, yeah, Google does do that, but here's how they have it completely sandboxed from any bad actors. Yeah, and then it's harder even still with Microsoft, right? Like the question for Microsoft, if they're going to be,

and Nadella's conception of Microsoft is, We make tools so that other people can make tools that help you do stuff, right? Which is a great conception of Microsoft. Like, don't get me wrong. That's like a powerful way to think about your company. But if he feels this great responsibility, then what he actually has to do is limit the tools that the other people use so that they can't do bad things. Yeah, and he wouldn't go that far when I talk to him. Yeah, that's a weird place for a toolmaker to be.

Yeah, I specifically asked, you know, if you're making all this AI available to your developers to build stuff on, Do you actually have to limit what they can do? And he was a little bit muddy about it. He's like, well, we want to provide guidelines. And we think that people should think about good ethical use of AI in the way that we currently think about good usable use of user interface UI. So that's really smart. I really like that a lot.

But that's a good way to think about it. But the question is this comes back to responsibility. If I'm a bad actor and I take Microsoft's AI and I use it to do something awful, Is that my fault as a developer, or is it Microsoft's fault for providing me the, like, AI foundation to do that awful thing? And if you're going to say that you feel responsible, you need to actually, like, that you need to take on the responsibility for that bad actor that used your tools. Or, well, it's, I mean, this gets real shaky. Like, are they a gunmaker or are they, like, if a robot shoots a gun. Right. If a robot shoots a gun, do you blame the person who programmed the robot, the person who provided the programming last?

language for the robot or the person who made the gun. You blame everyone. That's what I. That's what I, listen to the Vergecast and I blame everyone. No, the line about good UI is crazy. And it's crazy for this reason, which is everything looks pretty now, but we are surrounded by dark patterns everywhere we look. And those dark patterns, like, they trick you into buying things in games.

They make the levels of the games easier, right? when you're about to quit. Right. They, they, there's fake news, like, we are surrounded by people using the elements of good UI. Even,

right. Even companies that we like, brands that we like, apps, developers that we like, to do not nefarious things, but to make us pick settings we wouldn't otherwise pick. Or Facebook is like,

upload your contacts. You can pick yes or later. Like, right? Like, all over the place, there's good UI that's deployed in icky ways. Right, but I think his point is that we don't need, like, we know what good ethical UI is.

We also know what good and bad just from a usability perspective is because there's a ton of research and thinking and experience of having built user interfaces, especially visual user interspaces. With AI, we don't have enough of that research or that experience building stuff for it and on it. And I think his point is that we have to at least get to a place where we actually know what the hell we're talking about when it comes to building stuff. an AI. And I don't think Nadella thinks that as a industry and as a tech community were there yet. I do think that's saying, well, let's figure that out first and then we'll get the, you know, then we'll do it. It's like, that's a little bit of a cop-out. I think that you have to do both in parallel. Well, I think this leads, I want Paul to talk about edge computing, but I think this leads into it because one of the realities of this is that the processing power you need to do

some of this powerful AI stuff right now is not particularly local. You need to do it on Microsoft Data Center, you know, on Azure or AWS or Google's Cloud, like wherever you're going to do it. You know, Microsoft C is a bad actor. They can just shut it off. We all need to, like, when are these companies going to start shutting the stuff off? That's a conversation that we've been having, yes or no, but it points to a world where everything is much more managed. And Paul, you're, you have this, you talked about the edge and a piece here out this week, but at the end of it, you're like, I'm not even sure I understand how I feel about having managed client devices. Right.

Get into that a little bit. Well, I think a good word to unpack is the word security. So, so, so take, so take, take the Xbox. Take the Latin root. Take the Latin root for secure. It comes from, I have no idea. Okay. So take the Xbox.

Yeah. Please. Take the Xbox. Can you, Neelai. Can you run, can you run random.exe files on an Xbox? I do not believe I can, no. Well, congratulations. You are secure. So one of Microsoft's hot new products, you know, for the cloud and one of its very IBM-E new things it's doing is called Azure Sphere,

which is kind of a fun, exciting thing in the world because Microsoft has made its own version of Linux. And it's like a secure stack from the microcontroller to the software. to the software that developers can use to develop for it, to the cloud services, and they're basing a lot of their security practices on the Xbox. And if you think of what the Xbox is, it's something that is very hard to hack, but it also is very disempowering for users.

And it kind of makes sense. You probably don't want to run random.exe files on your toaster, and you don't want your toaster to join a botnet and be hacked. And so edge computing really needs to be managed in order to stay secure or possibly right now it looks like the solution to securing IoT devices is to manage them very centrally. And that's something that Microsoft's pushing. But the other thing about edge computing, I mean, I broke down like a bunch of different reasons why edge computing is exciting. But one thing the edge computing enables is it's like big data 2.0. Like right now, companies have big data that they do their machine learning on and they, like, you know, do inferences about their customers and it's a big competitive advantage and they sell it to people and all that sort of stuff.

But you're kind of limited by how much data you can get into the cloud and then how much money you can pay to process. So, like, Facebook was able to process 3.5 billion photos from Instagram because Facebook has a lot of money to do that. But one thing the edge computing is going to allow is a company will be able to design a machine learning algorithm, like take like lens, for instance. You're running machine learning very rapidly on the phone, identifying objects, but all the meaningful data still ends up in the cloud. So now Google, Microsoft, Facebook, companies like that who are embracing these sort of more edge paradigms by, offloading some of their big computing work to edge devices, they get to do even more big data.

They get to process maybe 100 times more, like, video footage than they could have ever done. So, okay, let me give you some bad metaphors. It's like SETI at home, but instead of using my processing power to help find space aliens, I'm using my processing power to help give yet more data to Google and Microsoft such that they can use it against me for advertising later? Absolutely. That's a good one. Okay. Edge computing is the best.

The thing is, like, I mean, I'm playing with the pixel book right now because it's, you know, they added Linux. It's a very managed device. Same with iOS. You know, you get a lot of upsides from having a company that just runs everything for you. But at some point, one, that company maybe knows more about you than you'd like them to know. And two, you want to install a data EXE that you trust that you might not be allowed to. Like the thing is, is that back to the word security.

By the way, Seikura, which is without care, used often beginning in the 1530s as without care, comma, dreading no evil. Dreading no evil. Well, you know the evil. Dreading no evil is good, but without care is complete opposite because all people care about now as security. The primary, as far as I can tell, the evil. that is really dreaded by these companies is their users. No.

Wow. I mean, as far as the biggest security vulnerability of your computer is you. You are the person who puts the least thought into keeping your computer secure. Well, that's why all these future computing platforms are managed. Yeah. Wait, can I just offer you that the, so Paul, I completely agree with you. I thought you meant something else.

I thought you meant to like, you're going to run your own software instead of our subscription software. But no, you are absolutely right. Let me just offer you this example. Max is here. We have a lot of family coming to visit us. Those people have laptops. They're using the laptops. I'm noticing a lot of tape on laptops. Do you know I'm not noticing a lot of on the webcams? You know I'm not noticing a lot of pass codes on phones. Really? We have had more family members come through here to just like play with the baby who have tape over their webcams but no passcode on their phone. And I,

like, it short circuits my brain. I'm like, how are you, how are you thinking of the FBI's on your, but you don't, how do I, and Becky's like,

just leave it alone. Like, hold the baby. That is like, you do want a bunch of these client devices managed, because the people will screw up and click on the bad link and put malware on their computer. It's just,

it's inevitable. The question is, there's another class of users. or there's the kid who's learning how to use a computer. And they, in order to innovate, they also need to be able to, like, run their own code outside of a managed system. You need to be able to break things. Yeah. To build.

And it's unclear as we get into more and more of these managed devices, how you will be given the permission to go ahead and break stuff. Right? It's also unclear. If you have a bunch of hardware that relies on a connection and a cloud service to function, because that's where a lot of it's going. What happens when Microsoft decides that, you know what, Xbox Live? Like, we just lost. We're shutting it off.

Right. Right. Like, Google does that every week. Like, how are we going to contend with that? Because we're buying this hardware and it's in our lives. I don't know the answer to that at all. Right. And I just have a visceral internal reaction to ever having to ask a company for permission to do things with products that I've bought.

Yeah. I mean, I'll say like, this is ridiculous. Like we had to buy a baby monitor. There are lots of beautiful, fancy baby monitor startup set there. This is true. And I just refused. Like, we have the simplest, it's a video one, but it's like it has a crappy piece of, it's like, it's, I wouldn't even call it a tablet. It's just like a remote RF video monitor. It's the same one Joanna Stern has. She was like, you should just buy that one. It doesn't have Wi-Fi. Because I don't want to buy some. I don't want to buy some. kickstarted baby monitor and the company fails. And then the cloud service it relies on to send video to my phone disappears. And I've got a $500 piece of junk. And like every single product like that, I now think, well, okay, I'm buying this expensive piece of hardware. If it's cloud service goes away, I'm doomed. And as you move, Microsoft's entire future is enabling people to build more and more of those products.

Which is good because we'll get cool stuff, but bad because the hardware is, starting to become completely commodity and replaceable in a way that I think is kind of scary. Right. The optimistic vision of the future is Microsoft solves these problems well enough that like the company that you bought your baby monitor from, your cloud baby monitor can go out of business, but somehow the the fabric of intelligent cloud and intelligent edge is, is well enough established and defined that you can like move your service over to something else and you can keep on, you know, gaining that service. But the history is that Xbox Live just goes down and you can't play.

Yeah. Well, like another example, Amazon owns Ring, the doorbell company. Like, and Google has a doorbell. I cannot imagine telling anyone to buy, recommending anyone buy a video doorbell that isn't from one of those companies. Because the reality is they will probably stay in business. whereas some startup doorbell probably won't, right? And like, that is a huge competitive advantage.

But at the same time, it's like, ah, I, A, first of all, does anybody need one of these products? The answer is probably no. But B, like, how can anyone start to compete once everything is managed into a cloud? And all we're depending on is longevity of services. Who knows the answer to this question? All right. That was dark.

Microsoft's great. Microsoft's great. They're great. And I think what they're trying to do with Windows, like they're trying to make it integrate with a phone better, they're doing that app called your phone. And they, it's funny,

Tom interviewed Joe Bell Fiore about basically that app. And the first thing they said was we really wanted to work with iMessage. We haven't talked to Apple yet. Fingers wrong. It's like, oh,

you poor, oh, you poor man. I just wish all Microsoft needs to do for me right now is if they just had one, person who pretended like Windows was important to them. Like, you have a lot of people who pretend that improving Windows is an important thing for Microsoft, but it seems like as a company, they're treating it as such a small,

insignificant, eventually dying market. It's just, it's just, imagine a startup that made Windows, you know, this startup company is like, look how large of a market it is. We're so excited. we're going to deliver tons of interesting features to our core users. Instead, like, Microsoft sees Windows as this thing that, like, they can leverage into cloud things. And I don't know. Also, never want to hear the words of Microsoft 365 ever again.

Yeah. Sorry for your luck, Paul. That's never going to stop. Paul, I'm going to say something to you now. And then we're going to have an ad. But I'm just going to say this to you. This is the year of Linux on the desktop. Yes.

As you know, this episode, Vertras brought to you by IBM. We live in a world that's creating AI-enabled everything, a world with more IoT devices than people. Today, technology has never been smarter, but smart only matters when you put it to work where it matters. When we put smart to work, we can help save the species, increase crop yields, and make progress. Not just for a few of us, but for all of us. So let's get to it. Let's put smart to work. Find out how at IBM.com slash smart.

Paul. Yeah. Every week. My man. week. You do a segment. That's the same name. I never lose the tab, boy, I wrote down the name.

It's called Kickflip the Kickstand script. All right. Talk to me. Nintendo has solved the worst problem with this. No, they haven't. The two worst problems with the Nintendo Switch in one $20 product. It's a little, it's a little doc. So you put your switch in, it has a kickstand because the switch has a garbage kickstand.

It also allows you charge your switch while it's sitting on a table, which is the other problem because the USBC port is on the bottom. So everybody's happy now and the world is great. $20. Is the dock wired, I'm guessing? Yeah. So it literally just like a USBC pass through with a kickstand. It's not like you can really take it to your rooftop party and play with your friends like in the promo video because you have to plug it in. It does at least fold up pretty small.

So that's something. I mean, I think you could put it in this dock and not have the dock plugged in anything and it still work. I'm assuming. I see. Yeah. That also makes sense. God, I'm going to buy this thing and just leave it attached to my switch nonstop, aren't I? Yep.

Yes, you are. Do you think that when they were designing the switch, they like thought about this. They're like, in year three, we'll put out the exact same switch with a better kickstand and everyone will upgrade. Yeah, they absolutely did. Nintendo's not that evil. That's like a, that's a...

Have you seen their product release cycle for the DS? Yes, they are. It absolutely are. I don't think they mean to be, though. I don't know, man. I've seen what they've been doing with that 3DS and the 2DS and all that. and like they are ice cold about that stuff. This is true.

This is the company that just put out many versions of its 30-year-old consoles. The 3DS cycle reminds me of like somebody like naming like a word document like final V2. You know? Yeah. All right. We're out of time here. I just want to real quick talk about a little bit of policy stuff. it's been a heavy episode on actual tech products.

But this is true. We talk about neutrality on the show all the time. Obviously, the FCC had a vote. They rolled back the Obama-era net neutrality rules. But the Senate has forced a vote on restoring those rules through a little known procedure called Congressional Review. You can read about that all on the site. That is, so there's going to be a vote in the Senate. Now, obviously, to get it all the way done, there has to be a vote in the House and Trump has to sign the bill.

Those things seem less likely. But we're at least going to get to a point where I think what the Democrats in the Senate want to do is force every sitting member of Congress to vote yes or no on net neutrality, which I think is going to be very revealing and provide a lot of, I don't know, grist for this upcoming midterm cycle. So be aware of that. It's coming. Net neutrality is officially rolled back on the 11th. I'm sorry, on June 11th, not May 11th. June 11th. So there's a lot of net neutrality stuff. spiraling in this world. Again, there's no like policy debate to be had right at this point. If you've been listening to Veritas, we've had the net neutrality policy debate. I don't know, for a decade. Can I just say one thing about this?

Yeah. I obviously am the pro, the new rules, don't like the old rules. But I love that Congress has realized it has oversight over, over regulatory. This is a very exciting development. to me, even if I lose on the specific policy. Yeah. Look, I think it's very instructive that we as a country are now saying, like, hey, our government should like do stuff.

Yeah. Like, we expect the government that we elected to like, specifically the elected people should do stuff. Yeah, checks and balance. Have you guys heard of those? Try it out. Like, that's happening. So that's good.

I, you know, I would be very surprised if this actually works is a gambit. I don't think that's anyone's point. I think the point is to make elected officials up or down and have that be part of the midterms. I think we've talked about this before. Net neutrality is a surprising engine for like young voter turnout. People are really into it. I take credit for that. I think just my work.

I'm totally kidding. People are into it. Like the internet's important to people. And I think this is like a really, this will be an interesting moment. So that's coming very soon. You're going to see sites like. like Etsy, Reddit, porn hub, others all like going red to tell people, like, go talk to your representatives about this stuff. So we'll see what happens there. Then sort of next to it,

Sarah Jong wrote a great timeline of, if you're following any of the Trump administration news, you know, that Michael Cohen has emerged. It's this cartoon of a hapless person. Yes. He's, he's in the middle of everything. He's Trump's lawyer. He's taken money. He's not apparently a great lawyer. He's doing all the stuff. But AT&T paid him some money, right when Trump took over his office, they claim this for consulting. Obviously, AT&T, they just wrapped up the antitrust trial about whether or not they should be able to buy Time Warner.

Obviously, net neutrality is rolled back. There is an incredible, just read Sarah's timeline. It points to nothing in one sense. And it points to everything in another sense, which is $600,000, I think, probably for most everyone who isn't a CEO of a major multi-billion dollar company is like a lot of money. But for AT&T, doling out $600,000 to Michael Cohen, just to have it done in case they needed it did not seem to be a thing that they thought of. Novartis paid, you know, something like over a million dollars to Michael Cohen for, you know,

in case we need consulting on pharmaceutical regulations. And their statement about it was, we had one unproductive meeting. I would like to offer anyone who listens to the Vergecast, one unproductive. productive meeting with me for a million dollars. Anytime you want, happy to do it. But swirling in the background of net neutrality of the 18-T time-water merger is whatever is going on with Michael Cohen. It's super interesting in the sense that Trump News is just sort of default interesting.

It hasn't amounted to any. So just read Sarah's timeline of this stuff because it's going to keep coming up over and over and over again. I don't think Michael Cohen got AT&T any favors, right? Like the Trump administration decided to block the Time Warner merger and they went to trial about it. So like I said, there's nothing explicit there. That's basically serious point. But on a larger level, the amount of money and corporate influence that's flying back and forth is everything.

Yeah. It's like there's no smoking gun. There's just like a giant like, I don't know, unused cannon. Just like getting rolled around and nobody hears it. But there it is. Well, it's just, I mean, it's the idea that this is how it works. I think just makes everybody squeamish. Right. And like, yeah, this isn't how you want it to, you don't want AT&T to be like, yeah, we've got $600,000 in a slush fund. Like, we'll kick it to this guy and hopefully that'll work. Right. Like, it's icky in its way. Anyhow, because by the way, there are no consumer groups that have that much money to just like throw with the problem. Right. Only the corporations do. And like you want that balance in there. But never fear because the verge will continue publishing forever. And we'll, we'll fight for the users. We're like Tron.

Yeah, we're like Tron. gone way over. Nat, thank you for joining us this week. Yeah, for sure. And I think that's it. We'll be back next week. There's other stuff to listen to. Why'd You push that button season two is going. This week's episode, Ashley and Caitlin talked to a bunch of people about push notifications. So of all the stuff we were talking about time well spent and using your phone better is interesting to you. Ashley and Caitlin talked to John Herman at the Times, who wrote a a whole piece about the red dots in your phone. They talked to the guy at the New York Times who actually writes and sends their breaking news push alerts, which is just incredible. And they talk to you

the CEO of a company called Swerve, which is a company that manages push notification targeting to users, which is not something I knew that was a business of this scale. So listen to Watch Switch Top Hot in this week's episode's great. The whole season's great. They're carrying on. And I also want to tell you listen to Polyons Quality Control, which is their review podcast. they're talking endlessly about video games, and this week, in particular, Infinity War, which I haven't seen yet, so don't tell me anything about it, but I'm sure they did a good job. You can also listen to Recode Decode with Kara Swisher. You can listen to Recode Media with Peter Kafka.

Virchcast fans, Josh Sepalti was the guest on Recode Media this week, so listen to that is actually a pretty fun interview. Josh remains Josh, maximum Josh, on that episode, so it's really good. And you can follow us everywhere. We're at Verge. Paul is at Future Paul Dieter's Backlon. I'm Reckless. Nat. We'll see you next week. That's it for the Vergecast. Rock and roll.

Promocode. This episode of Vergecast was brought to you by IBM. Technology today has never been smarter, but smart only matters when you put it to good use. Together, we can build a smarter future for all of us. Let's put smart to work. Find out how at IBM.com slash smart.