1 00:00:04,400 --> 00:00:07,800 Speaker 1: Welcome to Tech Stuff, a production from I Heart Radio. 2 00:00:12,080 --> 00:00:14,760 Speaker 1: Hey there, and welcome to tech Stuff. I'm your host, 3 00:00:14,960 --> 00:00:18,160 Speaker 1: Jonathan Strickland. I'm an executive producer with I Heart Radio 4 00:00:18,200 --> 00:00:20,600 Speaker 1: and I love all things tech. And this is the 5 00:00:20,680 --> 00:00:26,200 Speaker 1: tech news for Thursday, May six one. For once, I'm 6 00:00:26,200 --> 00:00:29,880 Speaker 1: actually recording this on the day it's going out, because 7 00:00:30,400 --> 00:00:34,680 Speaker 1: if you listen to yesterday's episode or rerun episode, you 8 00:00:34,720 --> 00:00:37,680 Speaker 1: know that I was seriously under the weather with some 9 00:00:37,760 --> 00:00:41,720 Speaker 1: food poisoning, and um, yeah, recording was just not in 10 00:00:41,840 --> 00:00:45,000 Speaker 1: the cards for me yesterday. So I appreciate your patients. 11 00:00:45,400 --> 00:00:49,440 Speaker 1: I apologize for the delay. Let's get on with the news. 12 00:00:49,440 --> 00:00:52,000 Speaker 1: We've got a lot to talk about. One of the 13 00:00:52,040 --> 00:00:55,480 Speaker 1: big tech related stories that broke this week was about 14 00:00:55,560 --> 00:00:59,520 Speaker 1: how the Facebook oversight board upheld the company's decision to 15 00:00:59,520 --> 00:01:02,840 Speaker 1: ban the account of Donald Trump, former President of the 16 00:01:02,920 --> 00:01:07,800 Speaker 1: United States. For those blissfully unaware of all this, I'll 17 00:01:07,840 --> 00:01:13,040 Speaker 1: try to sum up on January six one, though honestly 18 00:01:13,080 --> 00:01:16,680 Speaker 1: it feels like it was a lifetime ago now, insurrectionists 19 00:01:16,720 --> 00:01:20,080 Speaker 1: stormed the US capital in an attempt to overthrow the 20 00:01:20,200 --> 00:01:24,000 Speaker 1: U S election. Trump's behavior and words served as a 21 00:01:24,040 --> 00:01:27,520 Speaker 1: motivator for this group, and Trump was often walking a 22 00:01:27,760 --> 00:01:32,720 Speaker 1: very thin tight rope of plausible deniability while simultaneously encouraging 23 00:01:32,760 --> 00:01:36,319 Speaker 1: his base, some of whom clearly intended to harm members 24 00:01:36,360 --> 00:01:40,440 Speaker 1: of Congress, and so Facebook made the decision to suspend 25 00:01:40,520 --> 00:01:43,399 Speaker 1: Trump's account to reduce the platforms he could use to 26 00:01:43,400 --> 00:01:47,480 Speaker 1: stir up anger and incite potential violence. The ban has 27 00:01:47,480 --> 00:01:50,720 Speaker 1: stayed in place since then. Facebook caught a lot of 28 00:01:50,720 --> 00:01:54,840 Speaker 1: flak from some groups for instituting this ban, though the 29 00:01:54,880 --> 00:01:58,080 Speaker 1: company stated that it was in response to Trump violating 30 00:01:58,200 --> 00:02:02,000 Speaker 1: terms of service on Facebook's platform. Twitter likewise had a 31 00:02:02,080 --> 00:02:04,800 Speaker 1: lot of anger directed towards it for a similar ban. 32 00:02:05,440 --> 00:02:08,480 Speaker 1: In an effort to distance itself from the decision and 33 00:02:08,520 --> 00:02:12,600 Speaker 1: to kind of take the heat off of itself, Facebook 34 00:02:12,600 --> 00:02:16,560 Speaker 1: formed and oversight board intended to review incidents like Trump's 35 00:02:16,600 --> 00:02:20,440 Speaker 1: ban and to determine whether the company acted appropriately or not. 36 00:02:21,400 --> 00:02:23,840 Speaker 1: But now Facebook is going to have to hitch up 37 00:02:23,840 --> 00:02:27,200 Speaker 1: its breeches and make another decision. The board came back 38 00:02:27,200 --> 00:02:30,480 Speaker 1: and said, well, here's your problem. You were justified and 39 00:02:30,600 --> 00:02:33,079 Speaker 1: banning Trump for Violet in terms of service, But while 40 00:02:33,160 --> 00:02:37,959 Speaker 1: y'all did was this fuzzy, indefinite ban, which is an 41 00:02:37,960 --> 00:02:40,200 Speaker 1: issue you're all gonna have to either lift that ban, 42 00:02:40,840 --> 00:02:43,960 Speaker 1: make the band permanent, or decide exactly how long the 43 00:02:44,000 --> 00:02:48,240 Speaker 1: band can last. I don't know why the oversight board 44 00:02:48,240 --> 00:02:51,240 Speaker 1: has the same accent that I do naturally, but it's 45 00:02:51,280 --> 00:02:53,040 Speaker 1: just that's how it is in my head. So, in 46 00:02:53,080 --> 00:02:56,760 Speaker 1: other words, Facebook can't just have a band of indeterminate 47 00:02:56,840 --> 00:03:00,240 Speaker 1: length on Trump. The parameters need to be defined. Mind. 48 00:03:00,480 --> 00:03:02,920 Speaker 1: The board even pointed out that Facebook was trying to 49 00:03:02,960 --> 00:03:06,840 Speaker 1: pass the buck here. Now clearly the company would prefer 50 00:03:07,000 --> 00:03:10,640 Speaker 1: that some other entity swoop in and make this decision 51 00:03:11,080 --> 00:03:16,160 Speaker 1: on behalf of Facebook that way Facebook and say, hey, folks, listen, 52 00:03:16,760 --> 00:03:19,760 Speaker 1: we know, we totally know we're with you, but we 53 00:03:20,000 --> 00:03:24,000 Speaker 1: had to go this way because so and so told 54 00:03:24,080 --> 00:03:28,840 Speaker 1: us to. Except in this case, the so and so said, no, dice, Mr, 55 00:03:28,880 --> 00:03:31,600 Speaker 1: you made this bed, you lie in it, or rather 56 00:03:31,840 --> 00:03:36,920 Speaker 1: they said quote. The board declines Facebook's request and insists 57 00:03:37,080 --> 00:03:41,440 Speaker 1: that Facebook apply and justify a defined penalty end quote. 58 00:03:41,920 --> 00:03:46,320 Speaker 1: So yeah, I was paraphrasing. Meanwhile, there are plenty of 59 00:03:46,360 --> 00:03:49,200 Speaker 1: critics who say that the oversight board itself is just 60 00:03:49,240 --> 00:03:52,520 Speaker 1: a bad approach, and there should be a truly independent 61 00:03:52,600 --> 00:03:56,040 Speaker 1: review board that's in place to look at issues like 62 00:03:56,160 --> 00:03:59,160 Speaker 1: Trump's ban. I think this is a case where Facebook 63 00:03:59,200 --> 00:04:02,440 Speaker 1: would be really torn about the idea of regulation. On 64 00:04:02,480 --> 00:04:05,040 Speaker 1: the one hand, regulation would take some of the heat 65 00:04:05,160 --> 00:04:08,000 Speaker 1: off the company when things like this pop up. On 66 00:04:08,040 --> 00:04:11,320 Speaker 1: the other hand, it could force you know, the company 67 00:04:11,360 --> 00:04:15,400 Speaker 1: to act more frequently in cases that involve influential users 68 00:04:15,920 --> 00:04:20,040 Speaker 1: and other entities on the platform when they say controversial 69 00:04:20,560 --> 00:04:24,400 Speaker 1: or perhaps you know, stuff that outright violates terms of service. 70 00:04:24,600 --> 00:04:29,480 Speaker 1: As I've said many times, Facebook's revenue comes from advertising, 71 00:04:29,600 --> 00:04:33,560 Speaker 1: and it makes more money the longer people are actively 72 00:04:33,720 --> 00:04:36,960 Speaker 1: on the platform, so it wants to keep people there 73 00:04:37,000 --> 00:04:40,040 Speaker 1: as much as possible. That means there's a real incentive 74 00:04:40,040 --> 00:04:43,920 Speaker 1: for Facebook to allow controversial, influential people to be active 75 00:04:43,960 --> 00:04:47,719 Speaker 1: on Facebook because they drive a lot of engagement, and 76 00:04:47,839 --> 00:04:52,040 Speaker 1: engagement means more ads get served, and that means Facebook 77 00:04:52,160 --> 00:04:55,039 Speaker 1: makes more money. So I imagine the company is not 78 00:04:55,279 --> 00:04:59,360 Speaker 1: super jazzed about being more proactive in dealing with influential 79 00:04:59,400 --> 00:05:05,120 Speaker 1: people who violate terms of service. Moving on, mining frequently 80 00:05:05,200 --> 00:05:08,600 Speaker 1: has a negative environmental impact, and for that reason, some 81 00:05:08,720 --> 00:05:12,040 Speaker 1: regions have put in a lot of restrictions around mining. 82 00:05:12,520 --> 00:05:15,880 Speaker 1: But what if the mining was virtual, Well, the state 83 00:05:16,000 --> 00:05:18,920 Speaker 1: legislature of New York is considering a bill that would 84 00:05:18,920 --> 00:05:22,160 Speaker 1: put a three year ban on bitcoin mining in the 85 00:05:22,200 --> 00:05:26,200 Speaker 1: state while the state conducts an environmental study looking at 86 00:05:26,240 --> 00:05:29,640 Speaker 1: the impact of bitcoin mining. So let's do a quick 87 00:05:29,680 --> 00:05:34,520 Speaker 1: refresher on what bitcoin mining is. So it's all about 88 00:05:34,560 --> 00:05:39,279 Speaker 1: a proof of work approach to blockchain, and I'll dive 89 00:05:39,320 --> 00:05:42,039 Speaker 1: in just a little bit. There is a network of 90 00:05:42,080 --> 00:05:46,680 Speaker 1: computers that are connected into the bitcoin system. This system 91 00:05:46,880 --> 00:05:50,520 Speaker 1: keeps track of every single transaction that's made with bitcoin 92 00:05:50,920 --> 00:05:54,360 Speaker 1: all over the world, and every computer that's connected to 93 00:05:54,440 --> 00:05:59,000 Speaker 1: this system can see the shared ledger of all transactions. 94 00:05:59,360 --> 00:06:03,320 Speaker 1: These transactions get packed into blocks, and then once a 95 00:06:03,360 --> 00:06:07,720 Speaker 1: block of transactions gets validated by this system, that block 96 00:06:08,000 --> 00:06:12,720 Speaker 1: joins a chain of past transaction blocks that stretches all 97 00:06:12,760 --> 00:06:16,320 Speaker 1: the way back to the very origins of bitcoin. Each 98 00:06:16,360 --> 00:06:20,159 Speaker 1: block that follows builds upon the blocks that were behind. 99 00:06:20,560 --> 00:06:22,480 Speaker 1: That also means there's no way to go in and 100 00:06:22,520 --> 00:06:25,320 Speaker 1: alter a past block, which would mean that you know 101 00:06:25,360 --> 00:06:29,560 Speaker 1: someone's potentially trying to erase or duplicate a transaction. You 102 00:06:29,600 --> 00:06:31,560 Speaker 1: can't do that because if you make a change to 103 00:06:31,600 --> 00:06:35,760 Speaker 1: a past block, it affects every single block that followed. 104 00:06:35,800 --> 00:06:38,839 Speaker 1: It further down the chain up to the most recent one. 105 00:06:39,120 --> 00:06:42,160 Speaker 1: That means everyone would notice the change because everyone can 106 00:06:42,200 --> 00:06:47,000 Speaker 1: see the entire chain of transactions. The process of validating 107 00:06:47,000 --> 00:06:53,200 Speaker 1: these transactions includes mining, and essentially mining involves very fast 108 00:06:53,240 --> 00:06:58,160 Speaker 1: computers trying to guess an extremely large number. Now, how 109 00:06:58,200 --> 00:07:01,760 Speaker 1: difficult that number is to guess depends upon the amount 110 00:07:01,839 --> 00:07:06,640 Speaker 1: of processing power that's being thrown at the bitcoin mining system. 111 00:07:06,680 --> 00:07:10,280 Speaker 1: The goal is to stick to a ten minute cycle 112 00:07:10,440 --> 00:07:14,240 Speaker 1: as closely as possible, and so the system actually dynamically 113 00:07:14,400 --> 00:07:20,520 Speaker 1: shifts the difficulty of this problem. This guessing problem based 114 00:07:20,520 --> 00:07:23,080 Speaker 1: on how much processing power is being thrown at the 115 00:07:23,120 --> 00:07:26,600 Speaker 1: system at that time. The computer that guess is correctly 116 00:07:27,000 --> 00:07:30,160 Speaker 1: gets rewarded a certain number of bitcoins six point two 117 00:07:30,240 --> 00:07:32,760 Speaker 1: five right now, and each bitcoin is worth more than 118 00:07:32,760 --> 00:07:36,120 Speaker 1: fifty dollars apiece. And since there is a chance to 119 00:07:36,200 --> 00:07:40,760 Speaker 1: mind bitcoins every ten minutes, and each successful mining operation 120 00:07:40,880 --> 00:07:43,560 Speaker 1: nets you more than six bitcoins, it can be really 121 00:07:43,600 --> 00:07:47,360 Speaker 1: profitable to own the computer that makes the right guess. 122 00:07:47,960 --> 00:07:52,760 Speaker 1: But these guesses require computers that are ridiculously powerful and fast. 123 00:07:53,200 --> 00:07:56,680 Speaker 1: Bitcoin miners will set up networks of computer systems just 124 00:07:57,160 --> 00:08:01,800 Speaker 1: dedicated to this, using multiple GPUs. GPUs work really well 125 00:08:01,840 --> 00:08:04,880 Speaker 1: because they lean heavily on parallel processing, so they are 126 00:08:04,880 --> 00:08:08,240 Speaker 1: particularly well suited for bitcoin mining. And there is a 127 00:08:08,320 --> 00:08:11,880 Speaker 1: huge incentive to build out these really powerful computer systems 128 00:08:11,920 --> 00:08:15,080 Speaker 1: to do this, which is expensive, but obviously the payoff 129 00:08:15,160 --> 00:08:18,080 Speaker 1: is huge. So if you make the investment and it works, 130 00:08:18,600 --> 00:08:21,000 Speaker 1: then it pays for itself very very quickly. I mean 131 00:08:21,000 --> 00:08:23,560 Speaker 1: every ten minutes you have another chance to net another 132 00:08:23,960 --> 00:08:28,400 Speaker 1: six plus bitcoins. However, running these things requires a lot 133 00:08:28,680 --> 00:08:30,880 Speaker 1: of electricity, and this is where we get to the 134 00:08:30,960 --> 00:08:34,360 Speaker 1: environmental impact. If the source of electricity is a power 135 00:08:34,400 --> 00:08:38,120 Speaker 1: plant that's powered by fossil fuels, the mining operation is 136 00:08:38,160 --> 00:08:41,600 Speaker 1: contributing to carbon emissions and other types of pollution and 137 00:08:41,640 --> 00:08:45,320 Speaker 1: climate change. In upstate New York, bitcoin miners have been 138 00:08:45,400 --> 00:08:49,080 Speaker 1: using converted fossil fuel power stations to set up shop. 139 00:08:49,440 --> 00:08:52,480 Speaker 1: The price of electricity is really low there. That helps 140 00:08:52,559 --> 00:08:55,920 Speaker 1: cap the costs of operation because running these systems means 141 00:08:56,040 --> 00:08:58,640 Speaker 1: you're not just using up electricity, You've gotta pay for 142 00:08:58,679 --> 00:09:02,040 Speaker 1: that electricity too, So you want to have the lowest 143 00:09:02,080 --> 00:09:05,679 Speaker 1: amount of cost per electrical unit in order for you 144 00:09:05,760 --> 00:09:09,319 Speaker 1: to be able to run the business profitably. The weather 145 00:09:09,360 --> 00:09:11,559 Speaker 1: and Upstate New York also plays a factor. It's a 146 00:09:11,600 --> 00:09:14,160 Speaker 1: little cooler than in other parts of the country. That 147 00:09:14,240 --> 00:09:18,000 Speaker 1: reduces the costs of keeping all those processors cool enough 148 00:09:18,040 --> 00:09:21,559 Speaker 1: so that they don't overheat. Now the state is potentially 149 00:09:21,640 --> 00:09:24,560 Speaker 1: banning mining operations for three years in order to make 150 00:09:24,600 --> 00:09:28,839 Speaker 1: sure that the environmental impact isn't more harmful than good. 151 00:09:29,120 --> 00:09:31,840 Speaker 1: And honestly, I think this points at how the proof 152 00:09:31,920 --> 00:09:35,720 Speaker 1: of work approach, that's what I described earlier, it's flawed. 153 00:09:36,080 --> 00:09:39,840 Speaker 1: If the cryptocurrency becomes the center of speculation like Bitcoin did, 154 00:09:40,240 --> 00:09:43,840 Speaker 1: and it has an incredibly inflated value, then you'll see 155 00:09:43,840 --> 00:09:46,079 Speaker 1: the sort of rush to corner of the market on it, 156 00:09:46,120 --> 00:09:50,480 Speaker 1: and you get these unintended consequences. There are other methods 157 00:09:50,520 --> 00:09:54,560 Speaker 1: that blockchains can use besides proof of work, but they 158 00:09:54,600 --> 00:09:57,000 Speaker 1: too can have their own drawbacks. I'll have to do 159 00:09:57,040 --> 00:09:59,280 Speaker 1: a full episode on it in the future to kind 160 00:09:59,320 --> 00:10:03,720 Speaker 1: of just gcribe the different methodologies and the pros and 161 00:10:03,840 --> 00:10:08,199 Speaker 1: cons of each one. And now for our segment called 162 00:10:08,400 --> 00:10:12,920 Speaker 1: News that Continues to Baffle, Jonathan IBM announced it has 163 00:10:12,960 --> 00:10:17,160 Speaker 1: created a silicon chip that has two nanometer architecture, meaning 164 00:10:17,200 --> 00:10:20,320 Speaker 1: the components on this chip measure just two nanometers in 165 00:10:20,360 --> 00:10:23,839 Speaker 1: size on an individual basis. Uh. The nanometer, by the way, 166 00:10:23,920 --> 00:10:27,360 Speaker 1: is one billionth of a meter. Recently, Apple introduced the 167 00:10:27,400 --> 00:10:30,280 Speaker 1: in one chip and its new Imax. These have a 168 00:10:30,360 --> 00:10:34,280 Speaker 1: five nanometer architecture. Other companies are relying on chips that 169 00:10:34,320 --> 00:10:39,479 Speaker 1: have a seven nanometer architecture. Intel is still using architecture 170 00:10:39,559 --> 00:10:42,880 Speaker 1: that's got ten or fourteen nanometer components on it. So 171 00:10:42,920 --> 00:10:46,080 Speaker 1: what's the big deal with all this. Well, smaller components 172 00:10:46,240 --> 00:10:49,679 Speaker 1: means that potentially you can cram more of those components 173 00:10:49,679 --> 00:10:52,000 Speaker 1: onto a single chip. So if you've got a chip 174 00:10:52,040 --> 00:10:56,520 Speaker 1: that measures I don't know, a centimeter per side, well, 175 00:10:56,600 --> 00:10:59,880 Speaker 1: then you could fit way more to nanometer scale com 176 00:11:00,000 --> 00:11:04,640 Speaker 1: opponents on that than you could with fourteen nanometer scale components. 177 00:11:05,160 --> 00:11:09,559 Speaker 1: And more components potentially means more powerful and more power 178 00:11:09,679 --> 00:11:13,959 Speaker 1: efficient chips. Now, it's not quite as simple as shove 179 00:11:14,040 --> 00:11:17,000 Speaker 1: more transistors on there. Uh. The way you lay out 180 00:11:17,040 --> 00:11:20,000 Speaker 1: the transistors matters a lot. In fact, it matters so 181 00:11:20,080 --> 00:11:24,079 Speaker 1: much that Intel, while having those larger components than its competitors, 182 00:11:24,640 --> 00:11:27,040 Speaker 1: is really able to pack those components in at a 183 00:11:27,080 --> 00:11:30,640 Speaker 1: density that allows for a lot of power output. But 184 00:11:30,679 --> 00:11:33,679 Speaker 1: another big element, as I said here, is efficiency, as 185 00:11:33,679 --> 00:11:37,360 Speaker 1: in power efficiency. IBM says these chips will allow for 186 00:11:37,480 --> 00:11:41,800 Speaker 1: greater power efficiency, which means longer battery life. But the 187 00:11:41,800 --> 00:11:43,920 Speaker 1: thing that blows my mind is that when you get 188 00:11:44,000 --> 00:11:46,719 Speaker 1: down to this super tiny scale you have to deal 189 00:11:46,760 --> 00:11:50,160 Speaker 1: with some fundamentally puzzling problems. You're getting down to the 190 00:11:50,240 --> 00:11:54,560 Speaker 1: level of quantum effects, which we don't experience in our 191 00:11:54,640 --> 00:11:57,719 Speaker 1: day to day lives, right, you just don't. You've got 192 00:11:57,720 --> 00:12:00,480 Speaker 1: to get down super super tiny before you start to 193 00:12:00,760 --> 00:12:04,440 Speaker 1: observe quantum effects. And one of those quantum effects is 194 00:12:04,440 --> 00:12:08,080 Speaker 1: called tunneling. And this is a doozy all right, I'm 195 00:12:08,120 --> 00:12:09,959 Speaker 1: going to give you a very high level view of this. 196 00:12:10,200 --> 00:12:12,920 Speaker 1: So circuits, when you really get down to it, our 197 00:12:13,040 --> 00:12:18,880 Speaker 1: channels for electricity, and electricity generally speaking, is the flow 198 00:12:18,920 --> 00:12:21,800 Speaker 1: of electrons. Well, it turns out that we really can't 199 00:12:21,840 --> 00:12:25,480 Speaker 1: say exactly where an electron is at any given moment. 200 00:12:25,640 --> 00:12:29,320 Speaker 1: It's more like we can have a general idea of 201 00:12:29,360 --> 00:12:31,920 Speaker 1: where that electron might be. You can think of it 202 00:12:31,960 --> 00:12:36,280 Speaker 1: as a hazy field, and the electron could be anywhere 203 00:12:36,280 --> 00:12:39,600 Speaker 1: in that field. There's certain probabilities that it's at any 204 00:12:39,800 --> 00:12:44,480 Speaker 1: one position within that general field. All right, So let's 205 00:12:44,480 --> 00:12:48,720 Speaker 1: say we've got this field and it's moving down the hallway. Now, 206 00:12:48,720 --> 00:12:51,000 Speaker 1: the walls on either side of this hallway are pretty thick, 207 00:12:51,320 --> 00:12:55,840 Speaker 1: so the field can't overlap into the space beyond those walls, 208 00:12:56,120 --> 00:12:58,840 Speaker 1: so the electron can only be somewhere within the field 209 00:12:59,040 --> 00:13:02,280 Speaker 1: that's inside hall But then the field gets to the 210 00:13:02,520 --> 00:13:05,280 Speaker 1: end of the hallway and there's a closed door at 211 00:13:05,280 --> 00:13:08,080 Speaker 1: the end. This door is not as thick as the 212 00:13:08,080 --> 00:13:10,880 Speaker 1: walls in the hallway, and so once the field gets 213 00:13:10,880 --> 00:13:13,480 Speaker 1: close enough, some of the field actually overlaps to the 214 00:13:13,520 --> 00:13:17,400 Speaker 1: other side of the door. This means there is a 215 00:13:17,480 --> 00:13:21,200 Speaker 1: small probability that the electron is actually on the opposite 216 00:13:21,240 --> 00:13:25,079 Speaker 1: side of the doorway, and as long as something is possible, 217 00:13:25,440 --> 00:13:29,280 Speaker 1: it means that sometimes it actually happens. So that means 218 00:13:29,400 --> 00:13:33,320 Speaker 1: sometimes that electron is on the opposite side of the doorway, 219 00:13:33,400 --> 00:13:36,520 Speaker 1: as if it had tunneled through the door or if 220 00:13:36,559 --> 00:13:39,160 Speaker 1: the door had opened. Neither of those things actually happened. 221 00:13:39,160 --> 00:13:43,079 Speaker 1: The door remain closed, the electron did not physically tunnel 222 00:13:43,120 --> 00:13:45,960 Speaker 1: through the door. It's just that it was possible to 223 00:13:46,000 --> 00:13:49,360 Speaker 1: be on the other side, so sometimes it is. Now 224 00:13:49,400 --> 00:13:51,920 Speaker 1: this is really hard for us to grasp because nothing 225 00:13:51,960 --> 00:13:54,600 Speaker 1: in our experience behaves this way. But when you start 226 00:13:54,640 --> 00:13:58,400 Speaker 1: making chips that are down at this scale, you have 227 00:13:58,480 --> 00:14:02,160 Speaker 1: to factor these kind of things in with that design. 228 00:14:02,440 --> 00:14:05,600 Speaker 1: For a long time, physicists were leaning toward the end 229 00:14:05,640 --> 00:14:07,960 Speaker 1: of Moore's law, or at least the end of seeing 230 00:14:07,960 --> 00:14:13,120 Speaker 1: smaller components on chips because of these fundamental quantum effects. 231 00:14:13,160 --> 00:14:16,480 Speaker 1: So I always find it astounding when engineers figure out 232 00:14:16,600 --> 00:14:19,920 Speaker 1: ways to work around it. And to be clear, I 233 00:14:19,920 --> 00:14:22,680 Speaker 1: don't know what IBM has done in order to kind 234 00:14:22,720 --> 00:14:25,120 Speaker 1: of avoid these issues, because once you get down to 235 00:14:25,240 --> 00:14:29,600 Speaker 1: one to three nanometers, tunneling is something that I just 236 00:14:29,680 --> 00:14:32,600 Speaker 1: don't know how you avoid it. So I'm very curious 237 00:14:32,640 --> 00:14:37,440 Speaker 1: to learn more about this particular process. Twitter has created 238 00:14:37,480 --> 00:14:41,600 Speaker 1: a new feature that asks if you really seriously want 239 00:14:41,680 --> 00:14:44,320 Speaker 1: to send that mean spirited tweet out into the world. 240 00:14:44,840 --> 00:14:47,480 Speaker 1: The feature scans the words you use in a tweet, 241 00:14:47,760 --> 00:14:51,160 Speaker 1: and if those words are quote unquote mean, then Twitter 242 00:14:51,200 --> 00:14:53,680 Speaker 1: will send a prompt you asking if you really want 243 00:14:53,720 --> 00:14:56,760 Speaker 1: to send that tweet out, and you've got three options. 244 00:14:56,840 --> 00:14:59,320 Speaker 1: You can go ahead and tweet it, you can edit 245 00:14:59,360 --> 00:15:02,280 Speaker 1: the message, or you can just delete it. Now I 246 00:15:02,320 --> 00:15:04,880 Speaker 1: know that in the past I have shot off a 247 00:15:04,920 --> 00:15:07,880 Speaker 1: tweet in anger and then regretted it, and a lot 248 00:15:07,880 --> 00:15:11,000 Speaker 1: of other folks have done way worse than that. Sometimes 249 00:15:11,240 --> 00:15:14,960 Speaker 1: nothing comes of it and you just look a little petty, 250 00:15:15,080 --> 00:15:17,960 Speaker 1: or in my case, a lot petty. But other times, 251 00:15:18,080 --> 00:15:21,000 Speaker 1: sending off a mean spirited tweet brings along with it 252 00:15:21,120 --> 00:15:23,920 Speaker 1: some serious consequences that can have a huge impact in 253 00:15:23,920 --> 00:15:26,040 Speaker 1: a person's life. I'm sure you can all think of 254 00:15:26,080 --> 00:15:29,360 Speaker 1: a few cases in which someone tweeted something that they 255 00:15:29,440 --> 00:15:33,600 Speaker 1: either thought was justified or funny or whatever, and it 256 00:15:33,720 --> 00:15:38,160 Speaker 1: ended up causing them serious problems. People have lost jobs 257 00:15:38,200 --> 00:15:41,040 Speaker 1: over this stuff. So Twitter now has this feature to 258 00:15:41,080 --> 00:15:44,120 Speaker 1: help people take a moment to consider what they're saying 259 00:15:44,480 --> 00:15:47,400 Speaker 1: and how they're saying it before they commit to it. 260 00:15:47,600 --> 00:15:50,240 Speaker 1: And I like this. Twitter has always been a tool 261 00:15:50,320 --> 00:15:53,520 Speaker 1: that allows people to shoot off thoughts without any real delay, 262 00:15:53,600 --> 00:15:56,400 Speaker 1: and that's not always a good thing. I've joked in 263 00:15:56,400 --> 00:15:58,360 Speaker 1: the past that I kind of want to Twitter like 264 00:15:58,680 --> 00:16:02,560 Speaker 1: service where I can type and angry messages and hit send, 265 00:16:02,960 --> 00:16:06,840 Speaker 1: but the messages don't actually go anywhere. It's just a 266 00:16:06,880 --> 00:16:10,280 Speaker 1: way to vent without actually making things worse. This is 267 00:16:10,400 --> 00:16:13,560 Speaker 1: kind of like that, though, of course you do still 268 00:16:13,560 --> 00:16:15,640 Speaker 1: have the option to go through with it and tweet 269 00:16:15,640 --> 00:16:18,520 Speaker 1: out that awful thing you typed. I'm not sure how 270 00:16:18,560 --> 00:16:22,440 Speaker 1: extensive Twitter's filter is or what criteria it uses to 271 00:16:22,600 --> 00:16:26,120 Speaker 1: scan for quote unquote mean language, but the company has 272 00:16:26,160 --> 00:16:28,200 Speaker 1: been testing the feature out with a limited number of 273 00:16:28,280 --> 00:16:30,600 Speaker 1: users for more than a year and says that it 274 00:16:30,680 --> 00:16:34,560 Speaker 1: saw thirty four percent of users revising messages after receiving 275 00:16:34,560 --> 00:16:37,800 Speaker 1: a prompt, and that as a whole, they saw fewer 276 00:16:37,880 --> 00:16:41,960 Speaker 1: offensive messages in general, like people began to adjust the 277 00:16:42,000 --> 00:16:45,040 Speaker 1: way that they would word things after receiving prompts a 278 00:16:45,080 --> 00:16:47,720 Speaker 1: few times. I think that's a step in the right direction, 279 00:16:47,800 --> 00:16:50,760 Speaker 1: and it may mean in the future will really know 280 00:16:50,840 --> 00:16:53,320 Speaker 1: who the jerks are on Twitter. Those are the folks 281 00:16:53,320 --> 00:16:57,280 Speaker 1: who still send out mean spirited messages despite having this prompt. 282 00:16:57,920 --> 00:17:00,680 Speaker 1: They could also be bots, but at least know who 283 00:17:00,760 --> 00:17:03,000 Speaker 1: the jerks and the bots are, so I'm all for it. 284 00:17:04,040 --> 00:17:07,440 Speaker 1: One of the revelations out of the ongoing legal battles 285 00:17:07,480 --> 00:17:11,840 Speaker 1: between game developer Epic and Apple is that Microsoft has 286 00:17:11,920 --> 00:17:15,880 Speaker 1: never made a profit off their hardware sales for the Xbox. Now. 287 00:17:15,960 --> 00:17:18,719 Speaker 1: This doesn't come as a shock, but it is confirmation 288 00:17:18,720 --> 00:17:21,520 Speaker 1: of what was a widely held belief in the industry. 289 00:17:21,720 --> 00:17:25,520 Speaker 1: Lori Right, the VP of Business Development at Xbox, was 290 00:17:25,560 --> 00:17:29,080 Speaker 1: called as an witness for Epic, and the fact that 291 00:17:29,119 --> 00:17:32,800 Speaker 1: the company loses money on every Xbox sold is kind 292 00:17:32,800 --> 00:17:35,720 Speaker 1: of a crucial part to epics argument. So let's break 293 00:17:35,720 --> 00:17:39,480 Speaker 1: this down. So Microsoft sells the Xbox at cost or 294 00:17:39,600 --> 00:17:42,919 Speaker 1: at a loss, making no money on hardware sales. To 295 00:17:43,000 --> 00:17:45,960 Speaker 1: turn Xbox into a revenue generator, the company has to 296 00:17:46,000 --> 00:17:48,560 Speaker 1: make up for that loss in other ways, and one 297 00:17:48,560 --> 00:17:51,800 Speaker 1: way is to develop games and sell them for the console. 298 00:17:52,240 --> 00:17:55,199 Speaker 1: Another way is to offer services through the console that 299 00:17:55,280 --> 00:17:58,680 Speaker 1: require a subscription h this is why the Xbox Game 300 00:17:58,800 --> 00:18:02,320 Speaker 1: Pass is really important and product for Microsoft. And another 301 00:18:02,359 --> 00:18:05,840 Speaker 1: way is to take a cut of in game transactions 302 00:18:05,880 --> 00:18:09,960 Speaker 1: for certain titles like Epics Fortnite. So when players make 303 00:18:10,000 --> 00:18:15,360 Speaker 1: an in game purchase in Fortnite, Xbox takes at cut. Well, 304 00:18:15,400 --> 00:18:18,960 Speaker 1: Apple does the same thing. And yet what started off 305 00:18:19,000 --> 00:18:22,880 Speaker 1: this whole legal battle was Epic trying to sidestep Apple 306 00:18:23,080 --> 00:18:26,119 Speaker 1: by offering users a work around to make in game 307 00:18:26,160 --> 00:18:31,120 Speaker 1: purchases on iOS versions of Fortnite without going through Apple's system, 308 00:18:31,400 --> 00:18:34,280 Speaker 1: which meant Apple would not get the thirty percent cut 309 00:18:34,280 --> 00:18:37,760 Speaker 1: of those transactions. So why did Epic try to sidestep 310 00:18:37,840 --> 00:18:41,479 Speaker 1: this with Apple but not with Microsoft. Well, according to Epic, 311 00:18:41,560 --> 00:18:45,000 Speaker 1: the big difference here is that Microsoft's revenue depends upon 312 00:18:45,080 --> 00:18:48,399 Speaker 1: these sorts of transactions because the company is not making 313 00:18:48,440 --> 00:18:53,400 Speaker 1: money off hardware, whereas Apple isn't taking a loss on hardware. 314 00:18:53,440 --> 00:18:56,600 Speaker 1: When Apple sells an iPhone, it does so at a profit, 315 00:18:57,000 --> 00:19:00,520 Speaker 1: and so argues Epic, Microsoft has a justific cation for 316 00:19:00,600 --> 00:19:04,760 Speaker 1: taking a cut, but Apple does not. Anyway, the interesting 317 00:19:04,800 --> 00:19:08,440 Speaker 1: thing here is getting confirmation that so far Microsoft has 318 00:19:08,520 --> 00:19:11,960 Speaker 1: never profited from the sale of Xbox consoles, though that 319 00:19:12,000 --> 00:19:16,119 Speaker 1: could eventually change with the latest generation of hardware. Also 320 00:19:16,280 --> 00:19:19,879 Speaker 1: this week, SpaceX had a successful test of a Starship 321 00:19:19,920 --> 00:19:23,600 Speaker 1: prototype vehicle. The starship is shaped like a rocket and 322 00:19:23,760 --> 00:19:28,200 Speaker 1: launches vertically from the ground. During descent, the starship maintains 323 00:19:28,240 --> 00:19:31,520 Speaker 1: a horizontal orientation for most of the journey back to 324 00:19:31,880 --> 00:19:35,199 Speaker 1: the surface, but engages its rocket engines to flip the 325 00:19:35,240 --> 00:19:39,399 Speaker 1: spacecraft back into a vertical position before its final descent 326 00:19:39,520 --> 00:19:42,080 Speaker 1: legs so it can land vertically on a landing pad. 327 00:19:42,640 --> 00:19:46,480 Speaker 1: SpaceX has held a few tests of this type of spacecraft, 328 00:19:46,720 --> 00:19:49,800 Speaker 1: but all the earlier tests ended with those test vehicles 329 00:19:49,840 --> 00:19:53,800 Speaker 1: being destroyed, or, as the New York Times tweeted, quote 330 00:19:54,119 --> 00:19:57,880 Speaker 1: SpaceX successfully landed a prototype of a spacecraft it hopes 331 00:19:57,960 --> 00:20:00,440 Speaker 1: to one day sent to the Moon and Mars. It's 332 00:20:00,440 --> 00:20:05,000 Speaker 1: the first time the vehicle didn't explode. End quote. I 333 00:20:05,000 --> 00:20:07,919 Speaker 1: had detect some shade in that tweet. I wonder if 334 00:20:07,960 --> 00:20:12,400 Speaker 1: the New York Times got that prompt before they sent it. HM. Well, 335 00:20:12,480 --> 00:20:16,400 Speaker 1: that wraps up the news for Thursday, May six one. 336 00:20:16,440 --> 00:20:19,800 Speaker 1: I hope you are all well. I am very much 337 00:20:19,800 --> 00:20:21,960 Speaker 1: well on the way to being fully minded, so I 338 00:20:22,000 --> 00:20:25,440 Speaker 1: appreciate that. If you have any suggestions for topics I 339 00:20:25,440 --> 00:20:28,560 Speaker 1: should cover on tech Stuff, send them my way on Twitter. 340 00:20:28,880 --> 00:20:31,919 Speaker 1: The handle for the show is text Stuff H s 341 00:20:32,200 --> 00:20:40,920 Speaker 1: W and I'll talk to you again really soon. Text 342 00:20:40,920 --> 00:20:44,399 Speaker 1: Stuff is an I Heart Radio production. For more podcasts 343 00:20:44,400 --> 00:20:47,160 Speaker 1: from my Heart Radio, visit the i Heart Radio app, 344 00:20:47,320 --> 00:20:50,439 Speaker 1: Apple Podcasts, or wherever you listen to your favorite shows.