 Daily Tech News show is made possible by its listeners. Thanks to all of you, including Johnny Hernandez, Hi Tech Oki, and Logan Larson. Coming up on DTNS inside the Amazon Style Store, eBay's move to live shopping streams. A lot of streaming ads happen when your TV is off and we solve the disinformation problem. You're welcome. This is the Daily Tech News for Thursday, June 16th, 2022 in Los Angeles. I'm Tom Merritt. And from Studio Redwood, I'm Sarah Lane. From Austin, Texas, I'm Justin Robert Young. And I'm the show's producer, Roger Chang. Yes, I can't wait for us to just finally solve that nagging problem, but hold your horses or your cars or scooters or whatever vehicle you use to get around. Let's start with a few tech things you should know. Samsung released a new Samsung Wallet App in the UK, the US, and parts of Europe, which will handle payment cards, also passwords, and integrate with the smart things to store digital home and car keys, while it takes place of the existing pay and pass apps, while it also integrates with a Samsung blockchain wallet app to let users check values on their dwindling crypto assets. Oh, so sad. Microsoft announced Microsoft Defender for individuals as part of the Microsoft 365 subscription for Windows, of course, Mac OS, iOS and Android. This provides a dashboard with the security status of connected devices and runs its own malware protection engine on Android and Mac OS. Microsoft plans to add identity theft protection and other consumer-focused features to Defender going forward. The markup looked at the websites of Newsweek's top 100 hospitals in the US and found that 33 of them included the Metapixel, which is used to track data for Facebook's advertising system. Pixels are small one-by-one images that log the IP address of an individual loading the page. IP addresses are somewhat big indicators of a person's location, but they can also send other referral data when they do load, and the markup found instances of the pixel packaging in medications, also descriptions of symptoms, doctor's names, search terms, and conditions selected from a drop-down menu. The markup wasn't able to determine how personally identifiable the information was or how Facebook might use it or already be using it. Discord launched its auto-mod autonomous moderation tool available on all community servers. It scans for select words, phrases, and malicious links with automated responses like blocking messages and alerting admins. Elon Musk's The Boring Company has received approval from the city of Las Vegas, Nevada to bring its underground transportation system called The Vegas Loop to city limits. The Vegas Loop was originally a 29-mile tunnel network connecting 51 stations. Now the network is said to expand, at least granted the ability to expand, to 34 miles and have a total of 55 stations, including ones that will serve the Harry Reid International Airport and the Allegiant Stadium. Tunneling could begin in 2023 according to boring company president Steve Davis. You really will be able to never see the sky in Las Vegas from the minute you land it. That was always my grape with Vegas. Too much sky, not enough indoors. Oh man, I had to actually step outside. I only want the fake sky that's in the mall in the Venetian. Come on. I lost $300 at the tables in the Hyperloop. What a bummer. All right, let's talk about what's going on with shopping online. Let's do it. So if you can believe it, it's really been a year since Andy Jassy took over from Jeff Bezos, a CEO of Amazon, How Time Flies. The Wall Street Journal has an article looking at that first year and noting the moves by Amazon to cut back on the spending that accelerated during the pandemic lockdowns. Amazon is certainly not alone in a company that has had to scale back on that. One aspect of that is shutting down physical stores like Amazon's four star and bookstores, but also opening new ones such as Amazon Style. That's a clothing store. Now, Tom, we know you went to one yesterday. Not everybody is near enough to go to physically hang out at an Amazon Style store, so how to go? Yeah, so the Americana in Glendale has one of the Amazon Style stores. They shut down the four star store there and opened up Amazon Style. It's interesting. The idea, if you remember us talking about it before when they launched it is there's one of everything on the rack. You find the thing you like. You scan the QR code and then you say, send that item in my size and the color I want to either a dressing room that I'll go to and try it on or to a pickup location. So I found a purple flannel shirt that I liked and I decided, you know what? I'm not even gonna try it on. I'm just gonna send it to the pickup station. So I went over to the pickup station and picked it up. I had to wait like five to 10 minutes from the time that I scanned it. So I actually had a little seating area that I sat at and waited. Then I saw it come out and go on the rack and I walked up and I was like, oh, that's my purple shirt. They're like, oh, great, can we see your app? And so I opened the app, they scan the app. I'm like, okay, great. And then they go back. Honestly, I honestly, they looked around for it. I'm like, it's the only thing on the rack. Like it's not gonna be hard to find. Then they gave it to me and said, all right, you just gotta go check out now. So then I had to carry it over to the checkout stand and buy it. And I bought it in the app that I had just used to scan it. And then I had to scan that app to prove that I paid for it before they handed it over to me. But yeah, it's a nice purple shirt, very, very comfy. I think for that use case that I used, way more complicated than it needed to be. Well, yeah, cause it seems like a more complicated version of the treasure truck, which some folks have interacted with, but that was an Amazon thing that would bring one item to various different locations in major cities. And because one of them was very close to my house, I bought a bunch of stuff off there because every once in a while, there'd be an air fryer for cheap. But I would buy it right there on the app like you would any other product and then I would just pick it up. So the fact that this is two stages, as opposed to one is interesting cause they've already had a successful one stage version of this. Yeah. And in fact, I think that's what this is meant for is I wanna just try it on. I don't wanna buy it yet. So I go in the app and say, put it in a dressing room, I'll be there soon. And then you show up, try it on and buy it, right? Or I think you can also buy it and just pick it up and not have to check out. I did the normal thing you do in shopping, which is browse around, find something and then not try it on and hope it fits. And it's not really meant for that. Yeah. Yeah. And what you described, Tom, is like there's one step too many for something like that, you know, to sort of sit there and wait for somebody to take something off the rack that you could have just taken to the register yourself, seems a little counterintuitive. But what if you were shopping for 20 shirts? I don't know how often somebody's gonna be doing that, but there'd probably a lot of reasons that you would. And then I think that would make a lot more sense of, okay, I'm trying on a bunch of things, colors aren't right, fits not right, help me just kind of organize everything. So I'm not schlepping, you know, all the things that I decide to finally buy, you know, to the register myself. So it does seem like it has a good use case, but not necessarily for everyday use. I don't want to get anyone in trouble, but I did ask like, hey, how's it going since you opened? And one of the employees like, kinda slow. I'm not sure anyone knows we're here. And it was, there were other people in there. It wasn't totally empty. It actually felt like a normal crowd, but it wasn't like a new store had just opened kind of crowd. There were ads all over the Americana for the store, but I'll be honest, I didn't realize they were for the store and I was going there. They just looked like Amazon clothing ads. I'm like, oh yeah, Amazon sells clothes. Like they didn't make me think, oh, there's an actual location I should go to in this mall. Man, if they had a location targeted mailing list. Yeah, right. Once you were in the store, did the store remind you of any other retail stores you've seen before or did it feel really different? It felt like a Zara, but with only one of each thing on the rack instead of five. Okay, yeah. So it sounds like it's optimized for people who are trying on a bunch of stuff. Exactly. This browsing and optimized for browsing stuff for that, I would say traditionally, women's clothes tend to be a little bit more Byzantine in their descriptions and listings and everything. So trying things on is more of a necessity than men's clothes, which is a little bit more forgiving. You need to know a few numbers and a few letters and you're pretty much able to dress yourself, even if you get it a little wrong. So I'd be interested to see how it goes. But as for there being nobody there, again, I made a joke, but holy smokes. Let the world know Amazon. They are. It's just that people are like, yeah, Amazon. I know what Amazon is. Like they don't, it doesn't even click that that's what, when you see that Amazon logo, that's what you think. Anyway, another trend that's dominating in China and other Asian marks is live stream shopping. Amazon has some of that. They haven't gained as much traction on it though. Twitter, YouTube, TikTok, Facebook, Pinterest, they all have their own efforts, but there's a new one. Indeed there is Tom. eBay is getting into the game with eBay live rolling out in beta this month. You might say, well, what is it? The first event will be a selection of rare trading cards from Bleaker Trading in New York. To note, these are actual cards. They're not NFTs. This is physical stuff. That starts at 3 p.m. Eastern on June 22nd. Highlights will include a 1995 Top's finest Michael Jordan M1 red bordered refractor PSA9 Mint. And a 1998 Kobe Bryant Skybox, that's jam PSA10. If you know, you know. Shoppers can access the live event in the eBay app or website. Also interact with the sellers in the chat to ask some questions and make sure it's what you want. And then of course, buy the things. Yeah, well, best of luck to eBay on conquering live shopping in the United States when it's huge in other parts of the world. What do you think that is? I don't know. I think, yeah, I think people think QVC when they hear it and then don't try it. And there was not that history elsewhere. Or you think QVC salted the earth. Yeah, and the people who like QVC go to QVC. So they don't go try the new thing. I think it's because it is fairly pervasive in our culture. Like there's a lot of sponsored stuff. There's a huge market for people demoing things and trying things out. And that's ultimately what you get from this kind of stuff because we're making informed consumer decisions. Yeah, let us know what you think. If you've got a theory on, especially if you're someone who's in one of the markets like China or Korea, where this is huge, like why do you think it's not catching on over here in North America? Feedback and daily discussion. There's probably more than a few folks saying it has caught on. And I'm one of the people that is super excited about it. Yeah, but by numbers, it's not even close to what it is. Not even close. Yeah, I mean, my answer is always, well, only so many people kind of want to be collectors. But again, that's still a large number of people. Well, moving on to ad companies. We all love them. We often hear about them targeting you based on a huge amount of personal data that they can access based on your behavior and your history. We also occasionally joke about how this seems to end up with you seeing ads on pretty much every website or social network for a product that you just bought and therefore are no longer shopping for. For example, I bought the linen sheets already. Why do you keep showing me the ads for the linen sheets? Truly a thing I go through every day of my life, although sometimes I think about getting more. But the Wall Street Journal has another story about ineffective ads, ads that just don't work at all. So Justin, what is the story here? So glad you asked, Sarah. A study by ad buying company Group M and ad measurement firm iSpot.tv found that about 17% of ad shown on television connected to a streaming device, Roku, Apple TV, Chromecast, et cetera, are playing while your screen is off. It happens most often while viewers are watching free streaming channels like Pluto TV and is estimated to result in a couple billion dollars worth of ads unseen by human eyes. Sometimes when you turn off the TV, the device gets the message through the HDMI and turns itself off. But not always. This pretty much doesn't happen if you use smart TV app since the app stops running when the TV turns off. Yeah, it sounds like we're gonna start seeing a lot of work to detect if your television is truly off, even if the screen is off and has stopped playing ads because that is important information for advertisers. And you'll definitely see more attempts by apps to put up one of those. Are you still watching NetflixX prompts after a certain amount of time has passed without you taking any action in an app? Yeah. When I first heard this story, I was like, really? I mean, I turn off my TV, you know, with the voice commanded pretty much every night. Does this apply to me? And I don't really think it does with most of the apps that I use. But this is really good intel for anybody who cares about this sort of thing. Yeah, if you're watching Netflix, this doesn't apply, right? Netflix doesn't have ads yet, might someday. And if you have autoplay on a Netflix, it could apply. But you've probably been trained to remember like, oh, crap, I forgot to close Netflix last time. And it just kept playing. And now I have to go back and reset my play. Pluto TV, though, there's no downside. If you're just watching that grid and you're watching that Golden Girls channel and then you turn off the TV, it just keeps going. And it doesn't affect your viewing history, but it is playing ads and those ads get counted, but you never see them. And apparently that's a problem that is 17% of ads shown on television's large and streaming ads are getting bigger and bigger, not smaller and smaller. And advertisers everywhere like, la, la, la, la. Don't hear it. No, no, don't wanna know. It has been a problem since the beginning of display ads from the very first time that somebody sold a spot of their cave for which a local business could paint their logo on it. We have had the problem of trying to track exactly how many people are going to see this thing. Newspapers, count circulation, despite the fact that there is no guarantee that they're going to see your ad on page three of the sports section. There have been problems with television ads and you are only trying to guess based on a flawed Nielsen system exactly how many people are doing it. Same thing with magazines, same thing with radio, same thing with every other version of this. These kinds of ads are by definition scattershot. So the fact that we even have a quantifiable look at something that we now understand is total waste is actually a huge want to leap forward for this method of advertising. There was a reason why targeted ads that came through on Facebook and Google immediately overturned the industry. Adding that level of granular analytics to advertising was and remains today in the year of our Lord, 2022, a game changer. Yeah, I think the only difference is with a display ad, you had to survey people to find out if they saw the ad in the newspaper. With web delivered ads, internet delivered ads, you know they got delivered and you know where they got delivered. And so the fact that somebody might not have seen them is something you think you can solve with an HDMI sensor, you know, like with a prompt. It's like, well, maybe we don't have to settle for the problems we had with the display ads. But that's so well said, Justin. That's really the only difference here. Yeah, we just know. And that's, you couldn't guarantee that when you bought an ad in a popular magazine that it was ever going to be seen by the amount of number. I mean, because you're buying an ad based on circulation. That's how these things used to work. Let's say magazines, for example. So you're saying, okay, the ad rate is set based on the fact that there are X number of subscribers. There's X number of people that we distribute this to. They sell it at X number of stands across the country. You're paying for that, despite the fact that almost certainly not everybody that it has a circulation for is going to read the magazine, let alone see your ad. But that's just how it is. You are building in- But that's not how it is. That's not how it's been on the web since the 90s. Yes. And so people got used to the idea of like, yeah, but we don't want to go back to those days. Well, I mean, even then though, right? So let's look at web display ads, which ran on impressions. And as somebody who ran a blog. But they got better and they got better and they got better. And now we're very sure that those ads we deliver at least get seen. They get served. Whether or not they get seen is a different story. As we are seeing in today's story, yeah. Especially as they are refreshing, right? Yeah, yeah. So this is the problem. But if you're in this game, you understand that you're buying in bulk. You understand that there's going to be a lot of ways to hear and what you're hoping- They don't want to go back to that though. They don't want to go back to that. They're there now. There's no choice. It would be kind of cool as like Pluto TV or any other network where ads are being served to somebody with their TV off. Like configure out how the AI robot within the app can see and absorb the ad and then serve content to the human when the TV comes back on later. Let's get Lemoine on that. I bet he can make that happen. Sure. What do you, either human or AI listening to us, want to hear us talk about on the show? One way to let us know is in our subreddit. You can submit stories and vote on them at dailytechnewshow.reddit.com. The specter of misinformation on the internet and disinformation, particularly for this story, has loomed large for almost a decade now and governments and tech platforms keep making stabs at fighting it. In the latest move, the European Commission updated its code of practice on disinformation to now cover manipulative behaviors that include deep fake videos, bots, demonetizing accounts that spread misinformation. It requires platforms to give users tools to recognize and flag disinformation. 33 companies voluntarily signed on to these revised rules. They include Meta, Google, TikTok and Twitter. Companies have six months to show they are in compliance with the new rules and can be fined up to 6% of global revenue for any violations. The code was first introduced in 2018. And for whatever reason, it's arguable it has not had much effect on the spread of disinformation. Which leads us to a protocol story about the effectiveness of banning political ads. A narrow effort to prevent misinformation related specifically to elections. Twitter's policy, for instance, was to stop taking ads from political candidates, parties, government officials or PACs, political action committees, Spotify, LinkedIn, TikTok and Pinterest all banned political ads in a similar fashion. Nonprofits in smaller political organizations complained that the policies penalized them more than big corporations. Protocol uses the example of Exxon being able to run an ad about their view on climate change while a climate advocacy PAC could not run a counter ad. So what effect have these ads actually had? Well, it's hard to say since none of the platforms have opened up their data on the policies or shared any of their own research on the matter. However, former director of Facebook's public policy team and professor at UNC Matt Perot and his colleague Scott Bobwood Brennan did some research on publicly available data. They claim that Facebook's temporary ban on political ads that started immediately before and then ended immediately after the 2020 election had minimal impact on the spread of misinformation. It also limited the reach of smaller organizations and limited political fundraising efforts on Facebook. Political ads as a result have become more expansive and placed on programmatic ad systems that have less transparency. So we have less of an idea of who's buying the ads and where they're going, but this is one study and it's far from what is needed to fully understand the impact. My big drum that I've been banging for years is you have to study the effects of these things to understand if you're even solving a problem or just taking a wild guess that something might be wrong and then doing a thing to make yourself feel better and having no effect, which I'll be honest kinda looks like what has been happening. Well, and especially when we enter into the realm of politics considering how mainstream it has gotten and how much tech companies have taken it upon himself as being the entire arbiters of all communication, some elements rightly others wrongly. It means that you also mix in did the wrong political message get out and how do we affect that? How culpable are we to affecting it? Tom, I wanna ask you this question, you and Sarah, this question, what is the line between disinformation and misinformation? Because I think a lot of people can be confused about this including us, cause we went back and forth with that definition even in the copy here. Yeah. I mean, I guess in a very broadly, I would say disinformation is straight up not correct. It is just, it is. Give me an example. If I said, I am a canine, I'm a dog. You are, that would be disinformation. Well, okay, now I'm telling you. That could almost be interpreted, like I was born a canine and that's what I am. You'd say, well, that's not true. But if misinformation would be, well, you know, I have some dogs upstairs and they are really good at math. You know, it may be like, is that true? In my dog's case, it isn't. Love you Otis, but you're not good at math. So that's kinda how I identify the two separately. So that would be a thing where somebody would be like, well, cognitively, your animals could not be that good at math compared to a human. And you'd be like, well, I've owned a lot of dogs and this one seems to know that I always give three treats and when I hold back after the second, they're still looking at me. And so I consider that good at math for a dog. And so there's wiggle room. Yeah, yeah. Sure, maybe the wiggle room is the differentiator here. Sarah's definition fits what I've seen most people define this as, which is disinformation is like, absolutely it's wrong. Misinformation is sort of misleading and is a little bit less solid. But honestly, I don't think if you're doing a proper study, I don't know that you even need to define that what you're looking for is the effects. What you're looking for is, did this policy lead to a certain effect on the election? Did people feel like they were well informed and made choices that they meant? And I think that's where this gets so wibbly wobbly is sometimes people really just wanna prove that the candidate they didn't like shouldn't have won. And that's not a good study. A good study says, was the electorate well informed about why they chose something? And you can design those studies. I know that sounds really wibbly wobbly the way I said it, but there's a long history of being able to design studies to be like, let's measure if a policy had an effect and what that effect was without presupposing what the values should be and what people should have done. Let's just see if it had an effect. You could learn something just from doing that. Yes, I think even beyond that, we have to understand what is the best version of our society, right? What do we want to incentivize and what do we wanna stop? And that is an element that I do think it's important to categorize misinformation and disinformation because misinformation I think is very, very, very hard to get untangled from just a absent-minded or well-intentioned untruth for which happens in society and a malicious element of it. And it gets even more dicey when we enter the realm of politics, which is the art of misinformation. You are trying to shade the reality as much as you humanly can to twist it so your candidate looks excellent and your opposition looks terrible. And that's what politics is and always has been. It's why whenever we get into these conversations, it's so unfortunate to me that politics tends to be the realm we wanna fight on because it is by far the most tricky. Yeah, and I think that's why we're in such a garble about it and not making any progress on improving anything, whether it's political misinformation or disinformation or anything else because people wanna prove something different than what needs to be studied. What needs to be studied is, hey, they changed these policies. What effect did it have? Did it have any effect? Let's look at that. Then you all can decide what policies you wanna do if it had the effect you wanted, if you're trying to push a particular political agenda or if you're just trying to have a neutral playing field. Those are different conversations, but study the effects. We can't do anything until we actually find out like did these policies change anything at all? And we have vanishingly few studies that actually do that. Well, I'd like you all to behold something that I found very interesting this morning and that is the Flopatron. If you're not familiar, it's a somewhat ridiculously large PC hardware orchestra that plays music using only electric motors, no software. Polish engineer Paweł Zerodniak has been building various iterations of the instruments since back in 2011. So you may have heard about it. The first Flopatron was just a pair of floppy drives that played the Imperial March theme from Star Wars and quite well. However, the Flopatron 3.0 is capable of a full orchestra of PC peripherals. Specifically, and hold on to your butts here, 512 floppy disk drives, 16 hard drives and four flat bed scanners. Zerodniak explained in a blog post that the floppy disk drive wall is arranged into columns each handling a single note at a time. The number of drives involved varies the sound envelope with the floppies handling the low tones, the scanner section providing the higher pitches and the hard drives acting as the percussion section. And there is a video and it will blow your mind, honestly. This is the coolest thing I've seen in quite some time. Yeah, let's just listen to the beautiful music of the Flopatron, this is amazing. And you might say, I don't know, just sounds like MIDI music, right? Well, yeah, and it is, but maybe what has gone into creating this hardware wise is a real feat. I feel like it's better than MIDI. Yeah, salute to Pavel, man. Tour with this. Tour with this. You deserve it. Yeah, can you get that into a truck with a battery? Just take it around to festivals? That'd be, that is an amazing feat. Also, he should be playing the Met. He should be playing Sydney Opera House. Absolutely. Because of my art. Also, Pavel has been working on this for half my tech news career. We covered his first effort at Buzz Out Loud. No, wait a minute, I guess it was TNT if it was 2011, but still, crazy, crazy. Yeah, it's been some time. Yeah, I sent this to a friend who's into this sort of thing and he was like, this is insane. Do you know him? I was like, no, I don't know him. I could try to find his information and connect the two of you, but. Email the show, Pavel. That's right. We'd love to hear from you and we love your music. We sure do. We also love all the feedback that you give to us. If you have questions, comments, or anything to add on anything that we talk about on the show or anything we might talk about in a future show, we would love to get that feedback. Feedback at dailytechnewshow.com is a good way to send us your feedback. Thanks to Justin Marbury Young for being with us today. Justin, I know you're a busy man, so what's new? I have a brand new panel political podcast called We're Not Wrong featuring myself, Jen Briney of the Congressional Dish Podcast and Andrew Heaton of the Political Orphanage. This week we discussed the January 6th hearings, the bipartisan gun control deal and the case of Happy the Elephant, an elephant at the Bronx Zoo that had a case filed for it for habeas corpus. So it would be freed from the zoo. All that discussion, it is the most fun thing that I've done in a while. I love talking with those guys and I think if you enjoy the banter on this show and you're interested in politics, you're gonna love it. We're Not Wrong available everywhere that you get your podcasts. Excellent. Also wanna extend a special thanks to Mike Beatty, one of our top lifetime supporters for DTNS. Mike, we see you and we raise you a glass of thanks. Thank you for all the years of support. Yeah, could be you next, just stick around. Yeah, exactly. There's a longer version of the show called Good Day Internet. Many of you already know about it, but if you don't, find out more at patreon.com.slashdtns. We roll right into it for 30 minutes after DTNS rolls up. But we are live on this show, 4 p.m. Eastern, 200 UTC, Monday through Friday. You can find out more at dailytechnewshow.com.slashlive. We're back doing it all again with Rob Dunwood and us up to soon. This show is part of the Frog Pants Network. Get more at frogpants.com. Club hopes you have enjoyed this program.