1 00:00:01,080 --> 00:00:05,679 How'd you like to listen to dot NetRocks with no ads? Easy? Become 2 00:00:05,679 --> 00:00:09,839 a patron For just five dollars a month, you get access to a private 3 00:00:10,000 --> 00:00:14,439 RSS feed where all the shows have no ads. Twenty dollars a month will 4 00:00:14,439 --> 00:00:19,039 get you that and a special dot NetRocks patron mug. Sign up now at 5 00:00:19,039 --> 00:00:23,800 Patreon dot dot NetRocks dot com. Hey, Carl and Richard here with your 6 00:00:23,920 --> 00:00:28,800 twenty twenty four NDC schedule. Will be at as many NDC conferences as possible 7 00:00:28,800 --> 00:00:32,679 this year, and you should consider attending no matter what. The Copenhagen Developers 8 00:00:32,719 --> 00:00:37,359 Festival happens August twenty sixth through the thirtieth. Early bird discount ends April twenty 9 00:00:37,399 --> 00:00:44,840 sixth. Tickets at Cphdevfest dot com. Ndcporto is happening October fourteenth through the 10 00:00:44,880 --> 00:00:51,840 eighteenth. The early bird discount ends June fourteenth. Tickets at Ndcporto dot com. 11 00:00:51,880 --> 00:01:07,920 And we'll see you there, we hope. Hey, welcome back to 12 00:01:07,000 --> 00:01:11,799 dot net rocks. I'm Carl Franklin and I'm Richard Campbell. We are past 13 00:01:11,840 --> 00:01:17,640 the nineteen hundred mark here post build yeah yeah, we're also this is like 14 00:01:17,680 --> 00:01:21,560 the one studio show. We're doing in between a whole bunch of in person 15 00:01:21,599 --> 00:01:26,040 shows, right right, So it's gonna be fun. Yeah, we're going 16 00:01:26,120 --> 00:01:33,719 to be at NDC Oslo very soon. And yeah, really good feedback from 17 00:01:34,239 --> 00:01:38,519 show nineteen hundred was Scott hansomen people really love that we had some fun there, 18 00:01:38,599 --> 00:01:41,640 you know. And that was your idea too, to yeah, a 19 00:01:41,719 --> 00:01:46,120 surprise him with the with the quiz the clips and he got one hundred percent. 20 00:01:46,159 --> 00:01:49,840 It's ridiculous, right, I know that's sky. All right, let's 21 00:01:49,079 --> 00:01:59,680 roll the crazy music. All right, man? What you got? Well 22 00:01:59,719 --> 00:02:01,840 for bed, I know a framework today I don't have anything about the framework, 23 00:02:01,959 --> 00:02:07,519 but okay, but on my other show Security this week, we found 24 00:02:07,039 --> 00:02:15,280 a hack of vulnerability in a TP link gaming router nice that exposes users to 25 00:02:15,360 --> 00:02:19,879 remote code attacks. If you look at this thing, I'm thinking. The 26 00:02:19,919 --> 00:02:23,759 first thing I thought was, what's the difference between a gaming router and a 27 00:02:23,159 --> 00:02:29,000 line of business router? And the gaming router has more blinky lights and it 28 00:02:29,039 --> 00:02:32,319 costs more, yeah, and cool colors and things too. But yeah, 29 00:02:32,360 --> 00:02:38,039 it also prioritizes with QoS, It prioritizes the game packets, and they tend 30 00:02:38,039 --> 00:02:42,080 to use as many channels as they can you pay a premium for these things? 31 00:02:42,159 --> 00:02:45,560 Yeah, exactly. But it also kind of looks like it could take 32 00:02:45,599 --> 00:02:49,199 off, you know, like it's a like a drone of some kind. 33 00:02:49,439 --> 00:02:53,400 Yeah, the idea that your router is a statement piece for your coffee table. 34 00:02:53,400 --> 00:02:57,960 I don't know. It's like nine prongs sticking up out of it with 35 00:02:58,080 --> 00:03:01,080 red stripes on it. Yeah, they're in tennis. Yeah yeah, all 36 00:03:01,159 --> 00:03:07,120 right. So people who know security cvees, which are you know, threats 37 00:03:07,199 --> 00:03:15,879 or reports of problems, know that there's a CVSS score with them from one 38 00:03:15,919 --> 00:03:19,759 to ten, right, and ten is oh my god, this is horrible. 39 00:03:20,039 --> 00:03:23,840 It's the worst possible thing, right, but it only that score is 40 00:03:23,919 --> 00:03:29,840 only if you have been attacked. How bad it is opposed to this router? 41 00:03:29,919 --> 00:03:32,919 I presume, well, yeah, if somebody has attacked the router and 42 00:03:35,039 --> 00:03:38,560 hacked it, right, it's bad, like remote code execution is bad. 43 00:03:39,039 --> 00:03:45,919 But we at security this week have started coming up with how likely are we 44 00:03:46,120 --> 00:03:52,360 to be exposed to this number? So it's called like a contagion factor or 45 00:03:52,360 --> 00:03:55,319 something like that, which is not part of this number. So we went 46 00:03:55,400 --> 00:04:01,159 through the problems and first of all, you have to be rebooting. It 47 00:04:01,199 --> 00:04:04,599 has to have just booted up. So if your power goes off, your 48 00:04:04,639 --> 00:04:08,360 power comes back on, or you turn it off and turn it back on 49 00:04:09,159 --> 00:04:14,719 right in that boot up sequence, that's where you're vulnerable. And then certain 50 00:04:14,759 --> 00:04:18,759 ports have to be open right right, and then you're vulnerable to this attack. 51 00:04:18,800 --> 00:04:24,199 And my question was, well, does the attack have to happen from 52 00:04:24,199 --> 00:04:30,079 inside the firewall or also outside? And the answer is outside. It's exposed 53 00:04:30,120 --> 00:04:34,680 to the Internet, which makes you vulnerable in those seconds when it's booting up. 54 00:04:35,199 --> 00:04:40,279 And if somebody knows about that and sees it and connects to it and 55 00:04:40,360 --> 00:04:46,000 sends you the special blow up commands, they will get remote code access. 56 00:04:46,360 --> 00:04:49,439 Right, It's pretty scary. Yeah, it just means that. And you've 57 00:04:49,439 --> 00:04:54,199 got to know there's some script kiddies running this, running that call continuously, 58 00:04:54,319 --> 00:04:57,639 just walking back and forth. It's IP addresses, right, Yeah, that's 59 00:04:57,720 --> 00:05:00,600 right. So anyway, this is a kind of a big deal, and 60 00:05:00,639 --> 00:05:03,240 I wanted to share it with people. So, yeah, you know what, 61 00:05:03,319 --> 00:05:09,720 patch your routers, patch everything regularly, patch all the things on run 62 00:05:09,720 --> 00:05:12,800 as a little while back into a show with Sammy Laho is one of these 63 00:05:12,920 --> 00:05:15,079 one of the security folks. It's scarce it's not out of me on a 64 00:05:15,079 --> 00:05:19,800 regular basis. And and he talked about the fact that with multi factor authentication 65 00:05:20,000 --> 00:05:26,399 and you know, current security password technology, phishing is stopped being the primary 66 00:05:26,399 --> 00:05:30,920 exploit PLATH. Unpatched gear is now the primary exploit PLATH. So we had 67 00:05:30,959 --> 00:05:36,040 a whole conversation about you know, most IT folks want to test before they 68 00:05:36,120 --> 00:05:40,600 up they update, and so often it takes a week or two before you 69 00:05:40,639 --> 00:05:44,519 get updates onto your machines. It's like, that's now a higher risk than 70 00:05:44,560 --> 00:05:48,000 the risk of an outage by applying the patch immediately and then figuring out if 71 00:05:48,000 --> 00:05:51,720 it's a problem. Yeah, and I know that we're going to be talking 72 00:05:51,759 --> 00:05:58,319 to Veronica about AI. But AI is now in the game of hacking, 73 00:05:58,519 --> 00:06:02,079 you know, searching for vulnerable as soon as they're reported and immediately going out 74 00:06:02,120 --> 00:06:08,120 and trying to exploit them, generate and exploit, so generated exploit. It 75 00:06:08,199 --> 00:06:13,439 speaks to how we've got to release this information more carefully too, Yes, 76 00:06:13,600 --> 00:06:15,199 like more time to patch. But yeah, it's an interesting time, isn't 77 00:06:15,199 --> 00:06:17,800 It's an interesting time. It's you know, it's cat and mouse. But 78 00:06:18,319 --> 00:06:21,160 the sooner you patch, the safe you are. Well, it's also we've 79 00:06:21,199 --> 00:06:26,959 got there are professionals now, like these folks are not hacking for fun. 80 00:06:27,120 --> 00:06:30,680 They've hacked for money, and then so make sure they don't make any money, 81 00:06:30,879 --> 00:06:34,279 right, all right, who's talking to us today? Richard grabbed a 82 00:06:34,319 --> 00:06:38,959 comment off of show eighteen seventy four, which we do backup November twenty three 83 00:06:39,079 --> 00:06:44,560 with our friend vishwas talking about his applications of large language models. Vishwats is 84 00:06:44,560 --> 00:06:46,160 always good for that sort of thing, right, Yeah, when you know 85 00:06:46,199 --> 00:06:50,759 you think about We recorded that show early November twenty three, so he already 86 00:06:50,759 --> 00:06:54,560 had code out in the field. And that's what I love about vishwas it's 87 00:06:54,560 --> 00:06:59,160 like he calls me when he has something real to show. He's not speculating. 88 00:06:59,240 --> 00:07:00,279 It's like, let me talking about this project we have out in the 89 00:07:00,319 --> 00:07:08,199 field. And this was an app that he that he was working with where 90 00:07:08,199 --> 00:07:13,360 the large language model was understanding texts coming in from various sources. It was 91 00:07:13,399 --> 00:07:18,079 almost a kind of send him an analyzer, but more sophisticated by far. 92 00:07:18,879 --> 00:07:23,680 And we've got a hilarious comment here that I had to read from a zirene. 93 00:07:23,800 --> 00:07:27,800 I could presume it's an acronym or a pseudonym. Just a few months 94 00:07:27,800 --> 00:07:31,199 ago. And actually it's just quoting me. All he says is you don't 95 00:07:31,240 --> 00:07:38,560 ask a stochastic parrot for facts. That's not what it's for. Yeah, 96 00:07:38,639 --> 00:07:42,600 that's pretty good stochastic parent. Pretty easy for me to agree with myself. 97 00:07:44,800 --> 00:07:46,120 So siron, thank you so much for your comment, and a copy of 98 00:07:46,160 --> 00:07:48,680 music code By is on its way to un If you'd like a copy of 99 00:07:48,759 --> 00:07:51,639 music code By, I read a comment on the website at dot rocks dot 100 00:07:51,639 --> 00:07:55,160 com or on the facebooks. We publish every show there, and if you 101 00:07:55,199 --> 00:07:57,079 comment there and I read on the show, we'll send you a copy of 102 00:07:57,120 --> 00:08:00,360 music code By. And also you can follow us on Twitter. We've been 103 00:08:00,360 --> 00:08:03,399 on their ore ex Twitter as I'm calling you now, and we've been there 104 00:08:03,439 --> 00:08:07,560 forever. But the cool kids are hanging out on masthadon. I'm at Carl 105 00:08:07,600 --> 00:08:11,399 Franklin at tech hub dot social, and I'm Rich Campbell at Macedon dot social. 106 00:08:11,519 --> 00:08:15,279 Yeah, send us a two. Those are other ways that you could 107 00:08:15,759 --> 00:08:20,120 get yourself a free copy of music to code buy. All right, well, 108 00:08:20,199 --> 00:08:24,920 let's bring on our esteemed guest software engineer by day, an AI enthusiasts 109 00:08:24,920 --> 00:08:33,000 by night, Veronica Kolesnikova has extensive experience in full stack development using c sharp, 110 00:08:33,080 --> 00:08:39,360 dot Net, Java and Typescript, Azure and AWS clouds. Public speaker, 111 00:08:39,600 --> 00:08:45,360 hackathon volunteer, co organizer of Boston Azure user group, and mentor. 112 00:08:46,080 --> 00:08:52,279 She's curious about mobile and augmented reality Veronical's a master's degree in information technology. 113 00:08:52,440 --> 00:08:56,000 She also likes dancing, traveling, and practicing arial yoga in her free time. 114 00:08:56,759 --> 00:09:03,840 Arial yoga suspended from a helicopter, right, of course. If I 115 00:09:03,879 --> 00:09:07,960 had not heard of aerial yoga for forever and like in the past two months, 116 00:09:07,240 --> 00:09:11,360 I've thought, you're like the third person, like what's going on? 117 00:09:11,960 --> 00:09:15,480 Like if it's as if yoga wasn't hard enough. Now you're hanging in the 118 00:09:15,519 --> 00:09:20,559 air while you do it, and it's it hot or cold aeriel defense. 119 00:09:22,759 --> 00:09:24,759 Yeah, if I practices in Florida, then yeah, it's still hot, 120 00:09:24,919 --> 00:09:31,399 but here in Massachusetts it's sometimes cold. Yeah, it's just starting to get 121 00:09:31,440 --> 00:09:35,159 hot now. Yeah, So what's on your mind in the world of AI? 122 00:09:35,279 --> 00:09:41,080 Forroonica, everything is changing so fast every day I feel like I'm behind, 123 00:09:41,360 --> 00:09:48,080 although I'm reading the news and trying to connect to people online, but 124 00:09:48,320 --> 00:09:54,120 still I feel like I'm always behind. There is something new happening all the 125 00:09:54,200 --> 00:09:58,440 time. Yep, I feel the same way. It's not a bad thing, 126 00:09:58,559 --> 00:10:03,879 you know, like really we don't get sprints like this all that often, 127 00:10:03,639 --> 00:10:07,600 right with a new technology. I was explaining you to some high school 128 00:10:07,639 --> 00:10:13,879 students last week or earlier this week even I said, Look, the scientists 129 00:10:13,960 --> 00:10:16,600 come up with a new material, and then the engineers take the new material 130 00:10:16,679 --> 00:10:20,279 and they try stuff with it. What you're seeing right now is the science 131 00:10:20,320 --> 00:10:24,720 has been done. General of AI is where it's going to be, and 132 00:10:24,799 --> 00:10:26,879 the engineers you're running with it now, and we're just going to make some 133 00:10:26,960 --> 00:10:31,639 stuff and some of that stuff will be bad and some of it will be 134 00:10:31,679 --> 00:10:41,399 pretty useful. Yeah. A lot happened actually recently with Jenny AI and large 135 00:10:41,440 --> 00:10:46,039 language models, and now I feel like there is another trend coming with small 136 00:10:46,120 --> 00:10:50,480 language models. Yeah, how do you define the difference between its large language 137 00:10:50,480 --> 00:10:56,840 and all a small language model? The science? Can you put numbers around 138 00:10:56,879 --> 00:11:01,000 that? Is that like a number of parameters? Is it actually like the 139 00:11:01,039 --> 00:11:05,200 physical data size? Yeah? The data size is the major factor. There. 140 00:11:07,879 --> 00:11:13,799 Lots of those small language models you can actually put to a mobile device. 141 00:11:13,120 --> 00:11:16,639 So if you're created a mobile application it can run on the device, 142 00:11:18,200 --> 00:11:20,519 which is really cool. That seems to be the goal. Is I want 143 00:11:20,559 --> 00:11:26,919 to run this on a phone? That's debatable. Yeah, want is the 144 00:11:26,960 --> 00:11:28,600 struggle word, Like, I don't know that I want to run this on 145 00:11:28,639 --> 00:11:31,240 the phone, But it seems to be what they're offering us. Isn't the 146 00:11:31,279 --> 00:11:35,919 context of a small language model usually smaller and more focused on a particular area. 147 00:11:37,440 --> 00:11:39,240 Is that the story? Like, you wouldn't ask a small language model 148 00:11:39,799 --> 00:11:43,159 the current price of a foot long tuna subway, it would be right, 149 00:11:43,960 --> 00:11:50,360 Yes, yes, exactly. So large language models they were trained on more 150 00:11:50,480 --> 00:11:58,480 data. That's why they're large, and they're more they know more about stuff, 151 00:11:58,639 --> 00:12:03,399 and then small language model us they're more focused on specific areas. Yeah. 152 00:12:03,679 --> 00:12:05,679 Well, I don't think i'd ask any of them what the price of 153 00:12:05,679 --> 00:12:09,600 a sub sandwhich was because that's a fact that they're not good at facts. 154 00:12:09,759 --> 00:12:13,679 Yeah, they're not, are they? Yeah, you have to use RAG 155 00:12:13,720 --> 00:12:18,320 for that. Yeah, I'm sorry. We've got to be accurate in police 156 00:12:18,440 --> 00:12:22,879 RAG is which we will augmented generation. There you go. It's such an 157 00:12:24,080 --> 00:12:30,480 as. Yeah, the terms we have to live with now or did you 158 00:12:30,519 --> 00:12:35,519 did you see any of the keynotes or the sessions from build about Microsoft announcements 159 00:12:35,240 --> 00:12:41,639 around their AI stuff. Yeah, I was following online. Unfortunately I couldn't 160 00:12:41,679 --> 00:12:46,919 make it to Seattle this time around, but they released a lot of stuff. 161 00:12:48,960 --> 00:12:54,759 I'm glad that they're focusing not only on being up to date and up 162 00:12:54,799 --> 00:12:58,679 to speed, they're partnering a lot with open Ai, and I think it 163 00:12:58,759 --> 00:13:07,720 took them basically a week to include GPT four oh into their offerings as well 164 00:13:07,799 --> 00:13:15,480 for as open ai. But also they're focusing on responsible AI, which is 165 00:13:15,639 --> 00:13:18,399 very important. That impressed me. I mean, I don't know how responsible 166 00:13:18,480 --> 00:13:24,559 or secure it is, but the fact that they have this security layer between 167 00:13:24,600 --> 00:13:31,759 you and the models and the outputs and inputs makes me feel asleep a little 168 00:13:31,759 --> 00:13:35,039 bit better at night. Yeah. Yeah, And I think there was a 169 00:13:35,039 --> 00:13:39,360 really issue way that they presented it with the AI studio, where you know, 170 00:13:39,399 --> 00:13:41,799 one hand, it was like we're here to protect you, but it's 171 00:13:41,840 --> 00:13:46,240 also there's also a sense of we're here to also protect people from you too. 172 00:13:46,480 --> 00:13:50,759 Write like I don't know how a bad actor could use Microsoft's AI tools 173 00:13:52,000 --> 00:13:56,480 and not have the tool essentially stop them saying this is not an appropriate use, 174 00:13:56,000 --> 00:14:01,120 and that that to me, you know, relieve me, because we've 175 00:14:01,120 --> 00:14:03,519 been you know, we talk about open AI and this whole idea of an 176 00:14:03,559 --> 00:14:05,360 open model. It's like, do we really want these tools out in the 177 00:14:05,399 --> 00:14:09,559 while where anybody can do anything with them? Yeah, same as open source 178 00:14:09,600 --> 00:14:18,320 projects. So everybody here is probably old enough to remember the heyday of SQL 179 00:14:18,360 --> 00:14:22,720 injection attacks, which, by the way, are still the number one vulnerability 180 00:14:24,039 --> 00:14:30,000 in software. SQL injection attacks number one. So if you think about what 181 00:14:30,039 --> 00:14:33,919 a SQL injection attack is, you know, you have a field on a 182 00:14:33,919 --> 00:14:37,720 web form that you're filling in, and because you're smart and you know what 183 00:14:37,759 --> 00:14:43,799 the database is, you can embed SQL commands right in that data. And 184 00:14:45,000 --> 00:14:52,759 assuming that the back end code just concatenates all those inputs together without sanitizing them 185 00:14:52,840 --> 00:14:58,080 or without checking them, and sends it as a query to the database, 186 00:14:58,720 --> 00:15:03,480 therein lies the problem. Right. So think about in the context of AI. 187 00:15:03,840 --> 00:15:09,480 You know, the even the APIs like the open AI APIs take a 188 00:15:09,519 --> 00:15:15,840 sentence, it's text, right, So now we have prompt injection attacks. 189 00:15:16,320 --> 00:15:22,320 It's like a whole new level of threat. So they have to be a 190 00:15:22,440 --> 00:15:31,039 very diligent about sanitizing and discovering ways that people are trying to either hack into 191 00:15:31,080 --> 00:15:37,799 the system remote code execution or or you know, injecting and getting into the 192 00:15:37,840 --> 00:15:41,879 data model or something like that, or just asking stupid questions or dangerous questions. 193 00:15:43,480 --> 00:15:46,879 So it's really good that they have this this filter. Yes, people 194 00:15:48,000 --> 00:15:56,159 come in with creative ways to break the system, and now Microsoft is that 195 00:15:56,279 --> 00:16:03,519 in the jail jail breake detection, which is very good. But you know, 196 00:16:03,639 --> 00:16:11,360 it's always cat and mouse. Who is more creative, people who created 197 00:16:11,399 --> 00:16:18,440 those jail breaks or people who are trying to catch them. Yeah, and 198 00:16:18,799 --> 00:16:22,919 yeah, I tend to bat on the white hats just because there's a lot 199 00:16:22,960 --> 00:16:25,399 of them. When they're pretty skilled and they don't have to hide. They 200 00:16:25,399 --> 00:16:29,159 get to share each other's information and collaborate with each other, where the bad 201 00:16:29,159 --> 00:16:30,879 guys have to stay in the shadows. All right, Well enough doom and 202 00:16:30,879 --> 00:16:37,120 gloom. Let's talk about some positive things in the changes in open AI, 203 00:16:37,279 --> 00:16:40,559 maybe in particular or AI in general. What do you see out there? 204 00:16:40,799 --> 00:16:47,519 Yeah, so they just released that new model GPT four oh, which looks 205 00:16:47,559 --> 00:16:56,440 like it can do everything. It can analyze visual content, speech and text 206 00:16:56,759 --> 00:17:03,519 all from one place based. I remember the days when we had to use 207 00:17:03,559 --> 00:17:08,559 different APIs in order to process all kinds of information. But now it's not 208 00:17:08,599 --> 00:17:15,599 the case, which is very convenient. Those days were like a few months 209 00:17:15,599 --> 00:17:18,559 ago. And this is what they mean when they say multimodal, right, 210 00:17:18,720 --> 00:17:23,039 that it doesn't matter whether it's text or images or it can pandle it. 211 00:17:23,200 --> 00:17:29,559 Yes, okay, how are you? I mean four oh seems niche also 212 00:17:29,680 --> 00:17:32,720 in the sense that it's kind of a client. I don't know how it 213 00:17:32,799 --> 00:17:37,279 embeds in my apps. Yeah, you can create all kinds of applications with 214 00:17:37,359 --> 00:17:44,079 it, all kinds of assistance or I don't know, support tool. I 215 00:17:44,160 --> 00:17:52,000 know some people they are struggling with creation of graphs or maybe processing information from 216 00:17:52,039 --> 00:17:59,480 graphs, and by just connecting GPT four oh to application like that, it's 217 00:17:59,680 --> 00:18:03,559 easy for people to maybe read the graphs, maybe create new ones based on 218 00:18:03,640 --> 00:18:10,400 the data that they have. That's the use case that I was thinking about 219 00:18:10,480 --> 00:18:12,759 right away, because I remember some of flower and we are talking about like 220 00:18:14,079 --> 00:18:18,039 the Excel style graphs like graphs and overload a term these days too. Yeah, 221 00:18:18,279 --> 00:18:22,440 yes, interesting, So take up high chart. By the way, 222 00:18:22,519 --> 00:18:27,039 never use a pie chart. You send it to four oh and ask it 223 00:18:27,119 --> 00:18:30,440 to analyze that, Like, what do you get from this, what would 224 00:18:30,480 --> 00:18:33,519 be a better way to represent this data? Because that's a great problem. 225 00:18:33,200 --> 00:18:37,839 Yeah, I'm sure you can do that. And I know that some of 226 00:18:37,160 --> 00:18:42,160 our project managers they have to create all kinds of graphs pretty much every day, 227 00:18:42,680 --> 00:18:48,400 and they have to process a lot of data. But now they have 228 00:18:48,559 --> 00:18:53,799 that really good support from GPT four. Oh interesting. Yeah, I've been 229 00:18:53,839 --> 00:18:57,480 looking at these tools for a while. Now is good summarizers, you know, 230 00:18:57,759 --> 00:19:00,319 and you know, it's funny to think in terms of graphs are supposed 231 00:19:00,359 --> 00:19:03,000 to make it easier to read data. But I'm in the same situation as 232 00:19:03,000 --> 00:19:07,960 a lot of those managers. I get twenty pages of visualizations of information where 233 00:19:07,960 --> 00:19:11,759 I'm like, how do I summarize, you know, get this down to 234 00:19:11,839 --> 00:19:15,839 one thing, you know, to be able to push it through a tool 235 00:19:15,920 --> 00:19:18,839 to say what's the one page I should look at, or you know, 236 00:19:18,000 --> 00:19:22,279 what's the sum of all of that and still have it being meaningful. That's 237 00:19:22,319 --> 00:19:27,759 an interesting problem because I'd have never had a tool that could actually take a 238 00:19:27,759 --> 00:19:32,960 bunch of graphs and give you a single description. Now, have you taken 239 00:19:32,960 --> 00:19:34,599 this out for a spin? Like, have you tried doing that? I 240 00:19:34,640 --> 00:19:38,759 tried to play with it a little bit. Unfortunately, didn't have enough time 241 00:19:40,319 --> 00:19:48,319 to actually test it really well, but small scenarios something that they showed at 242 00:19:48,279 --> 00:19:56,720 the presentation that open Ai did a couple of weeks ago that worked. Look 243 00:19:56,839 --> 00:20:03,680 like it worked. I'm sure there are some downsides too, like I saw 244 00:20:03,720 --> 00:20:10,920 people complaining online that it's a little slower than GPT four and GPT four vision. 245 00:20:11,440 --> 00:20:17,079 But you know, there are always ups and downs. It also was 246 00:20:17,119 --> 00:20:18,759 brand new, so everybody's playing with it at the same time. I'm sure 247 00:20:18,759 --> 00:20:22,559 it's under a lot of load. Yeah, so I took it for a 248 00:20:22,599 --> 00:20:27,200 spin. Of course. I have chatchapt four oh in a browser, and 249 00:20:29,279 --> 00:20:33,359 I don't see any speech or anything, you know. I know, I've 250 00:20:33,480 --> 00:20:38,920 used my voice to communicate with chat GPT four on my iPhone version before four 251 00:20:40,000 --> 00:20:45,240 oh came out, and I could even and you can even attach pictures and 252 00:20:45,000 --> 00:20:52,480 upload them. So I don't really know. And I don't see any like 253 00:20:52,279 --> 00:20:59,559 speech things or anything different in the interface on the web anyway for chatchipt four 254 00:20:59,640 --> 00:21:03,720 oh than I did for four So how does that work? Yeah, so 255 00:21:03,799 --> 00:21:10,400 I think for end users the difference is maybe unnoticeable, as you said, 256 00:21:11,079 --> 00:21:18,559 but behind the scene, I think that they just had to convert different inputs 257 00:21:18,599 --> 00:21:25,440 into texts and then pass it to the model. So the API has endpoints 258 00:21:25,440 --> 00:21:30,319 now that accept streams of audio and video, et cetera. Yeah, so 259 00:21:30,400 --> 00:21:34,720 in theory it should be faster because we are eliminating that conversion part of it, 260 00:21:36,039 --> 00:21:38,200 right, Okay, Yeah, I'll have to check my phone version. 261 00:21:38,279 --> 00:21:41,240 But I didn't see anything in that either that was any different. Yeah. 262 00:21:41,319 --> 00:21:45,720 I guess this is about digging into the API enough to see how the different 263 00:21:45,759 --> 00:21:49,160 sources could be fed to it and and then what it will return. I 264 00:21:49,240 --> 00:21:56,400 mean, I've always had it return as checks right, like summarize this additional 265 00:21:56,440 --> 00:22:00,480 information, but I've never tried to have it returns like draw me a better 266 00:22:00,559 --> 00:22:06,920 graph like that, because that also seems very precise for a tool that tends 267 00:22:06,920 --> 00:22:11,400 its outputs don't tend to be that precise. So on the in the mobile 268 00:22:11,599 --> 00:22:15,079 app, which I updated so it supports four Oh, now you have you 269 00:22:15,119 --> 00:22:19,000 know a camera, you have you know, select a file, you have 270 00:22:19,079 --> 00:22:23,960 select an image, you have a microphone, and you have headphones, which 271 00:22:25,039 --> 00:22:30,880 I'm thinking is going to be you know, speech output. So so that's 272 00:22:30,920 --> 00:22:36,640 good. The web version, however, no different. And I couldn't find 273 00:22:36,640 --> 00:22:40,240 anything in the settings to enable it either. So then, yeah, did 274 00:22:40,240 --> 00:22:44,240 my homework limitation on the web Pageah, we'll see. I even asked chat 275 00:22:44,279 --> 00:22:51,680 GPT, right, how can I use speech to interact with chat GPT four? 276 00:22:51,720 --> 00:22:55,720 Oh? And it says to use speech to interact with GPT four, 277 00:22:55,799 --> 00:23:00,880 Oh, you need to utilize voice input with voice recognition or voice input on 278 00:23:00,920 --> 00:23:06,880 web browsers like Google Chrome or Speech to Text APIs right, So it really 279 00:23:06,960 --> 00:23:11,440 doesn't offer any anything at the at the app level. It sure sounds like 280 00:23:11,559 --> 00:23:15,559 Microsoft documentation too, right, Like, yeah, it's an answer but not 281 00:23:15,599 --> 00:23:22,200 actually used, not very helpful. Maybe I'm missing something. If I am, 282 00:23:22,240 --> 00:23:26,680 somebody email me please let me know how this works. I yeah, 283 00:23:27,000 --> 00:23:32,839 obviously all the demos I've seen, it's got that her vibe you know, 284 00:23:32,920 --> 00:23:37,279 the movie and even to the point where Scarlett Johanson I believe, is suing 285 00:23:37,359 --> 00:23:38,960 and I'm like, ah, why are we doing this? Why are we 286 00:23:40,000 --> 00:23:42,240 going down this path? I know it's so stupid. I'm trying to get 287 00:23:42,240 --> 00:23:47,400 this out of science people not think about this as science fiction and worry about 288 00:23:47,400 --> 00:23:51,519 the fact not helping. Yeah, they're really not helping. Actually. Sam 289 00:23:51,559 --> 00:23:57,119 Altman posted on x Twitter, saying her a couple of days out they release 290 00:23:57,480 --> 00:24:03,720 Yeah, you know it's a copyrighted term there, Sam, right, like 291 00:24:03,799 --> 00:24:08,920 what are you doing? Yeah? Yeah, it's not Scarlet's voice, but 292 00:24:10,079 --> 00:24:15,160 it's someone who sounds like Scarlet's voice. Yeah, it's very very close. 293 00:24:15,359 --> 00:24:18,720 And then again it's like, why would you do that? Scar Joe is 294 00:24:18,799 --> 00:24:22,920 kind of famous for being willing to take on anyone. She took on Disney 295 00:24:22,920 --> 00:24:27,920 for crying out loud and one yeah, and then still gets to work there, 296 00:24:29,319 --> 00:24:33,640 Like who does that? Anyway? I Yeah, I wish they would 297 00:24:33,640 --> 00:24:37,440 stay away from the science fiction parts, just because I think it's bad for 298 00:24:38,319 --> 00:24:44,559 the for people to get caught up in the non facts. That mean, 299 00:24:44,559 --> 00:24:47,839 it's good for fundraising just because Okay, if you want to deceive investors, 300 00:24:47,880 --> 00:24:52,240 great, but trying to deceive regular people it's not not appropriate. Hey, 301 00:24:52,240 --> 00:25:00,920 we should break for a moment for this very important message. Hey there, 302 00:25:00,200 --> 00:25:04,720 this is Jeff Fritz, the purple blazer guy from Microsoft, letting you in 303 00:25:04,799 --> 00:25:08,720 on a little secret about my friend Carl Franklin. You know, the guy 304 00:25:08,720 --> 00:25:14,279 who started dot net Rocks, the first podcast about dot net, way back 305 00:25:14,359 --> 00:25:18,640 in two thousand and two. The guy who's been teaching Blazer on YouTube since 306 00:25:18,680 --> 00:25:22,759 twenty twenty. Yeah, that Carl Franklin. Well, Carl has joined up 307 00:25:22,799 --> 00:25:29,279 with Codina Castle to teach a week long hands on Blazer class this September. 308 00:25:29,519 --> 00:25:34,480 And get this, it's at a castle slash villa in Tuscany. It's sort 309 00:25:34,519 --> 00:25:40,599 of a luxury vacation with Blazer learning built in. He's calling it the Blazer 310 00:25:40,799 --> 00:25:45,480 master Class. You'll learn Blazer from the ground up, finishing the week with 311 00:25:45,559 --> 00:25:49,960 the ability to build and deploy Blazer applications. Now, Carl did this class 312 00:25:51,039 --> 00:25:55,079 last year and it was a huge success. You can watch the awesome video 313 00:25:55,160 --> 00:25:57,960 from twenty twenty three on the website, which I'll share in just a few 314 00:25:59,000 --> 00:26:03,920 seconds. Since the training happens for only four hours in the morning over six 315 00:26:03,039 --> 00:26:07,279 days, you can bring your significant other with you and you should. This 316 00:26:07,359 --> 00:26:11,359 part of Italy is absolutely beautiful. There's so much to see and do, 317 00:26:12,000 --> 00:26:18,359 and Larry and Marco from Code in a Castle are organizing daily activities both at 318 00:26:18,400 --> 00:26:22,240 the castle and in the area. The castle is in the Marema, a 319 00:26:22,319 --> 00:26:27,599 less tourist y region of Tuscany, offering both classic Tuscan hill country as well 320 00:26:27,640 --> 00:26:34,279 as easy access to the Etruscan Riviera with sublime local food, wine and olive 321 00:26:34,279 --> 00:26:40,200 oil around every corner. Breakfast is included every day. There will be two 322 00:26:40,240 --> 00:26:44,759 communal dinners at the castle, bookending the experience, and most other meals in 323 00:26:44,799 --> 00:26:49,240 all activities are included. And did I mention you'll learn Blazer in person from 324 00:26:49,279 --> 00:26:55,279 Carl Franklin. Space is limited and for good reason. This is a quality 325 00:26:55,319 --> 00:27:00,079 training experience in a beautiful setting. Go to Code in a Castle dot com 326 00:27:00,279 --> 00:27:08,319 slash Blazer twenty twenty four that's b la zo R two zero two four. 327 00:27:08,920 --> 00:27:15,640 To take advantage of this amazing opportunity to join Carl in Tuscany for an unforgettable 328 00:27:15,680 --> 00:27:30,960 week of ladocha vita while advancing your programming skills in this important new technology. 329 00:27:32,799 --> 00:27:37,759 You know it's common for business application to contain fifteen percent repetitive code just because 330 00:27:37,799 --> 00:27:44,440 of metaprogramming limitations in the C sharp language. Why write boilerplate manually when a 331 00:27:44,480 --> 00:27:48,519 machine could generate it for you? Enter Metalama, the code generation and Verification 332 00:27:48,680 --> 00:27:53,599 toolkit for C Sharp. Their C Sharp to c sharp template language is simply 333 00:27:53,720 --> 00:28:00,880 amazing. Logging caching memento observable if it's repetitive of Metalama can automate it. 334 00:28:02,279 --> 00:28:07,119 Visit metalama dot net today and learn to automate your code patterns with their free 335 00:28:07,240 --> 00:28:11,640 edition. Remember it's Metalama with one L M E T A L A M 336 00:28:11,680 --> 00:28:17,640 A dot net. Now we're back, it's dot net rocks. I'm Richard 337 00:28:17,640 --> 00:28:21,720 Campbell. That's Carl franken Hey talking to our friend Veronica a bit about the 338 00:28:21,759 --> 00:28:26,319 new GPT four O and that's a oh, is an omni zero? Because 339 00:28:26,359 --> 00:28:32,720 why would that be confusing? And just a reminder, if you don't want 340 00:28:32,720 --> 00:28:36,200 to hear any ads, you can be a patron for five bucks a month. 341 00:28:36,240 --> 00:28:41,400 You can get an ad free feed. Go to Patreon dot dot nenrocks 342 00:28:41,480 --> 00:28:45,960 dot com. All right, awesome, where we we were talking about the 343 00:28:45,000 --> 00:28:49,279 things that you think are interesting that are coming up here and that you've played 344 00:28:49,279 --> 00:28:56,200 around with. Is there something that you have not yet seen that holds a 345 00:28:56,279 --> 00:29:02,319 promise that that you think is really really valuable for people, not just developers. 346 00:29:02,720 --> 00:29:08,960 That's a good question. I feel like those companies are so creative Open 347 00:29:10,000 --> 00:29:15,359 Ai and then Microsoft joined them, Google Italy as trying to catch up. 348 00:29:17,519 --> 00:29:22,200 Now we'll on mask with x Ai, which I heard a lot about but 349 00:29:22,319 --> 00:29:27,039 I haven't trusted it or I haven't touched it. Actually. Yeah, so 350 00:29:27,680 --> 00:29:37,519 there are new opportunities every day and new functionality. I feel like they're improving 351 00:29:37,039 --> 00:29:45,640 of something like copilot for developers, for example. It is pretty good now, 352 00:29:45,960 --> 00:29:53,160 but there are still a lot of opportunities for improving it and making it 353 00:29:53,240 --> 00:30:00,839 better. I'm always laughing with my colleagues. We got access to COPI a 354 00:30:00,880 --> 00:30:07,039 couple of weeks ago, maybe a month or so ago, and I'm always 355 00:30:07,480 --> 00:30:18,519 fascinating how it tries to fill out like text. So you type in text 356 00:30:18,880 --> 00:30:23,680 name of your variable and it's trying to feel it for you, or a 357 00:30:23,920 --> 00:30:30,279 name of a LaMDA function for example, or connection string, and screen fills 358 00:30:30,359 --> 00:30:36,039 up with gray letters. Yeah yeah, but I'm like, is it useful? 359 00:30:36,119 --> 00:30:40,599 I don't want it to fill out everything. Apparently you can turn that 360 00:30:40,680 --> 00:30:42,920 feature off if you don't want it constantly suggesting things. If you want to 361 00:30:42,960 --> 00:30:47,519 just read your comments and generate code, you can do that too. I 362 00:30:47,599 --> 00:30:53,279 find that chat chat GPT I use when I need like more lines of code 363 00:30:53,359 --> 00:31:00,039 to generate than co pilot is going to do very easily. Copilot's really good, 364 00:31:00,039 --> 00:31:04,079 I think for small algorithms, small blocks of code, but if I 365 00:31:04,160 --> 00:31:10,279 need it to if I need some serious generation alice chat GPT. So but 366 00:31:10,559 --> 00:31:15,079 between those two, it you know, really helps a lot. I saved 367 00:31:15,160 --> 00:31:22,039 hours just in the last day implementing some stuff that was really really uh you 368 00:31:22,079 --> 00:31:27,119 know, case heavy, like switch heavy. You know, like there's there's 369 00:31:27,200 --> 00:31:32,240 this I don't want to go into details, but there's this class with like 370 00:31:32,279 --> 00:31:37,480 four hundred fields in it that represent every piece of a CSV file. And 371 00:31:37,920 --> 00:31:40,799 you know that's the way they wanted it coded. And so if I need 372 00:31:40,799 --> 00:31:47,160 to do anything with those with all of those things, it it's a lot 373 00:31:47,200 --> 00:31:52,000 of a lot of repetitive code. And chat GPT made short shrift to that. 374 00:31:52,599 --> 00:31:57,720 It was great. How you tried the co pilot chat co pilot chat 375 00:32:00,119 --> 00:32:04,759 have co pilot chat. Oh so yeah, so that's right from Visual Studio. 376 00:32:04,839 --> 00:32:07,640 You can use that, right, Yeah, I have not and VS 377 00:32:07,759 --> 00:32:12,480 code works there. Yeah. What's better about that than say, using chat 378 00:32:12,519 --> 00:32:15,599 GPT. It's just convenient. You don't need to jump back and forth. 379 00:32:16,720 --> 00:32:23,519 You can also reference your code that you've already written or specific files you wanted 380 00:32:23,519 --> 00:32:29,799 to reference as well. Yeah, Okay, there's so many GitHub co pilot 381 00:32:29,839 --> 00:32:36,359 products now because there's there's also co pilot workspaces, not to be confused with 382 00:32:36,480 --> 00:32:39,400 workspaces, because why would anyone get that confused. Yeah, but I just 383 00:32:39,480 --> 00:32:45,039 got my my beta access to it, and I haven't had time to really 384 00:32:45,079 --> 00:32:47,480 take this out for a spin. But boy, talk about an interesting collaboration 385 00:32:47,640 --> 00:32:54,359 tool. It's just to have more information available to you and to think project 386 00:32:54,359 --> 00:32:58,960 wise. Like I'm with you, Carl, little chunks of code, right, 387 00:32:59,039 --> 00:33:02,640 like that seems to work fine. I don't think about discussing with the 388 00:33:02,680 --> 00:33:07,200 tool the architecture of a project. No, but you do need to you 389 00:33:07,279 --> 00:33:10,119 know, I find that with chatchipt anyway. You need to give it just 390 00:33:10,240 --> 00:33:15,039 enough information so that it can, you know, do stuff. So if 391 00:33:15,119 --> 00:33:22,039 I have something that requires, you know, four hundred cases in a switch, 392 00:33:22,200 --> 00:33:25,039 right, I do the first ten and then there's an obvious pattern there 393 00:33:25,319 --> 00:33:29,599 and I'll say here fill in the rest of these, so very internal, 394 00:33:29,880 --> 00:33:31,920 right, let the intern do Yeah, yeah, yeah, all right, 395 00:33:31,960 --> 00:33:37,400 I mean I respect that that's useful. It saves time, less typos maybe 396 00:33:37,480 --> 00:33:40,440 right, No, it worked perfectly, did exactly what I asked it to 397 00:33:40,480 --> 00:33:45,759 do. Yeah. Well, anyway, I saw an example with GPT four 398 00:33:45,799 --> 00:33:52,920 oh, when they just provided the diagram and then ask it to create code 399 00:33:53,960 --> 00:34:00,960 for that application. It wasn't like super complex application, but based on what 400 00:34:01,359 --> 00:34:07,960 they posted online it worked. Was it like a UML dihighground? Yeah? 401 00:34:08,039 --> 00:34:15,280 I think it was just simple A couple of boxes connected together and then flow 402 00:34:15,400 --> 00:34:19,840 chart. I saw that. I saw a Twitter post about that maybe a 403 00:34:19,880 --> 00:34:25,639 few months ago, but I never saw it in action, never tried it. 404 00:34:25,679 --> 00:34:30,920 Interesting idea because I remember seeing that early in early days of Azure they 405 00:34:30,960 --> 00:34:36,000 had sort of a diagrammer you could use in the portal for you know, 406 00:34:36,079 --> 00:34:39,920 laying out an app and then it would generate like the network calls and and 407 00:34:40,039 --> 00:34:46,280 vms and things like that. Remember Windows Workflow didn't write code like it's interesting 408 00:34:46,320 --> 00:34:51,760 to actually, you know, build that flow or draw out the flow and 409 00:34:51,760 --> 00:34:53,280 have it generate the code, because and I've seen it go the other way 410 00:34:53,320 --> 00:34:57,760 where it's like show me a flow chart of this code, you know, 411 00:34:57,840 --> 00:35:00,239 which I think is super useful too, because I hate trying flow charts and 412 00:35:00,280 --> 00:35:01,960 often I end up writing a bunch of code. It's like, what have 413 00:35:02,000 --> 00:35:06,559 I done? The fact that it could tell you that that should be a 414 00:35:06,559 --> 00:35:14,159 tool in visual studio. Do you remember Windows Workflow foundation briefly I'm trying to 415 00:35:14,199 --> 00:35:17,519 forget, but it seemed awfully cumbersome without you know, it was a one 416 00:35:17,519 --> 00:35:21,840 point oh product. It came out, We're dot net three. It's probably 417 00:35:21,880 --> 00:35:25,880 part of the fallout of Longhorn for Crimelud, and I like the idea. 418 00:35:25,920 --> 00:35:30,559 They were trying to make a generalized workflow engine. It's just a that's a 419 00:35:30,559 --> 00:35:34,599 really hard problem be. It was tainted with the disaster that was what was 420 00:35:34,679 --> 00:35:37,840 Longhorn. Yeah, yeah, we're still getting over it. But now apparently, 421 00:35:37,920 --> 00:35:42,400 you know, maybe you could use a tool like visio to create a 422 00:35:42,480 --> 00:35:45,679 diagram which includes classes and variables and stuff and just say here, build this, 423 00:35:46,440 --> 00:35:49,920 yeah, and build a framework for you. That's what I want. 424 00:35:49,960 --> 00:35:52,559 Really, I don't want the whole application. I just want it do. 425 00:35:52,679 --> 00:35:59,239 I just want help getting started with stuff that's gonna be several types, you 426 00:35:59,239 --> 00:36:05,440 know, several class and lists and things, and you know, implement a 427 00:36:05,480 --> 00:36:10,360 custom list that overrides the you know, the methods for adding into leading and 428 00:36:10,440 --> 00:36:15,159 updating and all that stuff. You know, things like that, stuff that 429 00:36:15,199 --> 00:36:21,039 I really don't want to write. I like your word, Richard the intern 430 00:36:21,880 --> 00:36:24,960 Yeah, it's like, here, write this code. Then, yeah, 431 00:36:25,239 --> 00:36:28,840 do you see these tools starting to take lead on projects. Veronica, Like, 432 00:36:28,880 --> 00:36:30,599 this is really a thing, because you know, we've had We've had 433 00:36:30,599 --> 00:36:34,280 folks talk that way too. It's like, when are we the intern and 434 00:36:34,360 --> 00:36:37,840 the tool is now you know leading? Yeah, I don't see right now, 435 00:36:39,039 --> 00:36:44,480 who knows in a couple of years. Oh, I've read too many 436 00:36:44,480 --> 00:36:46,840 books that say, listen, there's no determinism in this software generated I is 437 00:36:46,880 --> 00:36:51,920 good some stuff, but that's not it. I mean architecture as checklist. 438 00:36:52,199 --> 00:36:54,519 Sure, it's like, hey, I want to lay out an architecture, 439 00:36:54,559 --> 00:36:58,440 and now the tool is going to keep your mindy you to fill in these 440 00:36:58,480 --> 00:37:00,360 bits like have you done this part? And have you done that part. 441 00:37:01,079 --> 00:37:07,599 That's what they're talking about with workspaces is like now that the LM is the 442 00:37:07,760 --> 00:37:13,039 nagging project manager saying you said you'd get it done by this time, and 443 00:37:13,079 --> 00:37:16,000 you said you do these pieces like it's making sure all the boxes are checked, 444 00:37:16,280 --> 00:37:21,000 which I got un useful because I don't mind yelling at the automated PM. 445 00:37:21,519 --> 00:37:24,760 Well that's when you feel like the tool. That's Oh, I'm the 446 00:37:24,880 --> 00:37:35,639 tool. This code created by a tool. It still seems like a partner 447 00:37:35,840 --> 00:37:39,679 for the most part. But we're only a year into this for crying out 448 00:37:39,760 --> 00:37:47,119 loud. Yeah, but it's really helpful. I remember it took me a 449 00:37:47,159 --> 00:37:52,159 lot of time to actually search for something online and get answers, so maybe 450 00:37:52,519 --> 00:38:00,719 I obviously don't remember all the functions ever or like libraries. Yeah, so 451 00:38:00,199 --> 00:38:05,960 it was taking me a long long time to find something specific, but now 452 00:38:06,039 --> 00:38:14,239 it's getting really easy with al Alam and Jenny Ai, whether using GPT or 453 00:38:14,679 --> 00:38:21,760 using GILB copilot for example. I like what Microsoft is promising with you know, 454 00:38:21,800 --> 00:38:24,599 your databases and all the stuff that's in your one drive and all the 455 00:38:24,639 --> 00:38:32,400 stuff that's in M three sixty five. It just becomes you know, vectorized, 456 00:38:34,079 --> 00:38:37,119 and so you can use natural language just to find things in your email 457 00:38:37,159 --> 00:38:43,639 for example, or on teams, just stuff that happened that you don't have 458 00:38:43,679 --> 00:38:49,559 to go searching for. Sort of like the search idea has just taken over 459 00:38:49,639 --> 00:38:54,360 AI in terms of you know what Microsoft is doing anyway, they but we 460 00:38:54,440 --> 00:39:00,280 can't stop. We can't leave here today without bringing the elephant in the room, 461 00:39:00,320 --> 00:39:08,800 which is the R word recall. So something that Microsoft announced that would 462 00:39:08,800 --> 00:39:15,039 be coming to Windows eleven soon that's going to take pictures of your screen and 463 00:39:15,159 --> 00:39:20,239 keep them somewhere. I don't know, maybe in the cloud and then let 464 00:39:20,320 --> 00:39:24,519 you and when you do these semantic searches about things that you were doing on 465 00:39:24,519 --> 00:39:29,440 a particular date and time, when was I talking about that? Blah blah 466 00:39:29,480 --> 00:39:32,280 blah, Like it's going to use that to figure things out. But it's 467 00:39:32,320 --> 00:39:37,559 also you know, the the privacy nuts are going crazy over this. Yeah, 468 00:39:37,639 --> 00:39:44,000 except these tools existed for decades, Like there's nothing there's been there's been 469 00:39:44,039 --> 00:39:47,719 a widget for taking screenshots at regular intervals for like twenty years. Yeah. 470 00:39:47,800 --> 00:39:52,079 I don't know why this is different that people are flipping about it. Well 471 00:39:52,199 --> 00:39:53,039 is it? You know, if you don't like it, don't turn it 472 00:39:53,079 --> 00:39:58,239 on. I was under the impression it's going to be on by default, 473 00:39:59,320 --> 00:40:02,360 which is different, and now you have to turn it off. No, 474 00:40:02,760 --> 00:40:08,039 no, no, no, okay. You can also search through your history. 475 00:40:09,119 --> 00:40:15,039 So you were taking screenshots, you still had to go and manually search 476 00:40:15,119 --> 00:40:20,920 for information or we were back to it, and now it's all automatic. 477 00:40:21,519 --> 00:40:24,320 It's going to be factorized, right, Yeah, I mean that's the big 478 00:40:24,320 --> 00:40:29,679 difference now is we have really great tools for looking through all those screenshots to 479 00:40:29,719 --> 00:40:32,719 actually find what you were doing. Yeah, and it is I don't know. 480 00:40:34,159 --> 00:40:38,360 I was confused about the amount of data they're going to store. Yeah, 481 00:40:38,679 --> 00:40:43,519 okay, I just found I mean, how many screenshots? How much 482 00:40:43,519 --> 00:40:45,400 space do all those screenshots take? Yeah? Yeah, it's going to use 483 00:40:45,480 --> 00:40:52,800 local storage, so that that at least is good. You know, it's 484 00:40:52,880 --> 00:40:55,760 not sending stuff up to the cloud where other people can mine it for information 485 00:40:57,199 --> 00:41:00,079 very easier. Local storage is going to be filled up in a couple of 486 00:41:00,199 --> 00:41:05,079 days. Yeah, I don't know the answer. That's a good question. 487 00:41:05,960 --> 00:41:10,320 When or when do you let it purge, clean itself up to I mean, 488 00:41:10,599 --> 00:41:17,360 there's tiers here, there's recall for your own use. We you know, 489 00:41:19,039 --> 00:41:21,159 as soon as you go down the M three sixty five path. And 490 00:41:21,159 --> 00:41:25,000 we talk about this on run as all the time. It's like, now 491 00:41:25,000 --> 00:41:29,960 you're in the corporate surveillance side of things, which there's some legitimate parts to 492 00:41:30,039 --> 00:41:34,159 that, Like you know, I've spent enough time working with email administrators where 493 00:41:34,159 --> 00:41:38,320 it's like, listen, if you harass someone via company email, the company's 494 00:41:38,400 --> 00:41:43,360 liable. So the company has an obligation to be aware of what's going on 495 00:41:43,440 --> 00:41:45,400 inside of email. That does mean read everybody's email, but it does mean 496 00:41:45,440 --> 00:41:51,920 there are tools watching for harassment that then can lift it up to the point 497 00:41:51,920 --> 00:41:53,760 to say, hey, somebody else should be able to see this and do 498 00:41:53,840 --> 00:42:00,880 something about it. And so I'll recall as a tool around activity inside the 499 00:42:00,880 --> 00:42:04,559 company. That's interesting, But you would hope all this stuff is pretty tightly 500 00:42:04,599 --> 00:42:08,119 locked down. But you know what, I like having a dashcam on my 501 00:42:08,239 --> 00:42:14,360 car, you know, because then I have an incontribusible copy of what happened 502 00:42:14,360 --> 00:42:16,960 in my car, how fast I was going, when when that guy did 503 00:42:17,000 --> 00:42:20,639 cut in front of me, like all of those sorts of things. I 504 00:42:20,719 --> 00:42:23,800 kind of like this dash cam for my computer too. What was it I 505 00:42:23,960 --> 00:42:27,760 was doing? Yeah, when that when I had that problem. If I 506 00:42:27,800 --> 00:42:31,800 have storage space, I'll use it. What I don't want is a dashcam 507 00:42:31,920 --> 00:42:37,639 that's sending my data to somewhere. Yeah, And olers don't know why you 508 00:42:37,639 --> 00:42:39,960 would think that, seeing how Microsoft is supposed to be the privacy guys. 509 00:42:40,199 --> 00:42:44,880 Yeah, well it's not the case. So I was concerned about that, 510 00:42:44,920 --> 00:42:46,400 but you're right, it's not the case. And it does say pre rex 511 00:42:46,480 --> 00:42:52,920 are a co pilot plus PC, of which I think there's like none right 512 00:42:52,960 --> 00:42:57,360 now for and there's only well there's only beta versions or alpha versions. Yeah, 513 00:42:57,440 --> 00:43:00,199 no, they it's a particular machine. These are once they announced at 514 00:43:00,239 --> 00:43:06,239 build, right, you can pre order for June, so by the time 515 00:43:06,280 --> 00:43:07,280 this show is out, in theory, you're gonna be a get one. 516 00:43:07,280 --> 00:43:12,039 But it's got a co pilot key, which is really important. But it 517 00:43:12,079 --> 00:43:15,440 has something else too, right, is there some sort of chippy stuff built 518 00:43:15,440 --> 00:43:21,039 in for AI that isn't on a copilot plus PC? Well, the ones 519 00:43:21,079 --> 00:43:25,239 that they were announcing it build were either Windows on ARM Snapdragon yeah right, 520 00:43:25,559 --> 00:43:30,960 Snapdragon. And it's got a pretty seriously powerful MP unit like forty five tops, 521 00:43:31,000 --> 00:43:37,559 which is the current measurement system, which is questionable but okay, questionable 522 00:43:38,760 --> 00:43:46,079 meaning questionable just because like measuring the measuring the computational potential of a neural processor 523 00:43:46,320 --> 00:43:52,199 is an interesting problem. Like those are very academic benchmarks in their early stage 524 00:43:52,199 --> 00:43:55,639 benchmarks for new technology, So you know, how will we be sure that 525 00:43:55,639 --> 00:44:00,199 that number is meaningful in terms of actual use? Are you gonna order or 526 00:44:00,239 --> 00:44:09,199 buy a one of these Snapdragon PCs Veronica? I wasn't planning to the relatively 527 00:44:09,280 --> 00:44:14,480 reasonably priced. I think that the mid range unit was like less than two 528 00:44:14,639 --> 00:44:19,679 k USD And somebody asked me if we're going to do a dot at rocks 529 00:44:19,760 --> 00:44:22,639 on Windows on ARM. I'm like, that's a pretty short show. Yeah, 530 00:44:22,239 --> 00:44:27,360 sure, yes, you know, yes, yeah, we done like 531 00:44:27,400 --> 00:44:30,599 that. I don't even know if you'll need to recompile. I suspect the 532 00:44:30,679 --> 00:44:34,760 gitteral just go oh it's armed, let's go. Yeah, I think so 533 00:44:34,880 --> 00:44:37,480 too. That's funny, right, like, okay, yeah, not really 534 00:44:37,519 --> 00:44:42,599 think they the CLR team solved this long ago. I want to see how 535 00:44:42,679 --> 00:44:46,320 successful it's going to be. It's not all Microsoft products, especially on hardware 536 00:44:46,480 --> 00:44:52,159 site, were successful, that's right. Well, they do have a bunch 537 00:44:52,199 --> 00:44:54,760 of third party vendors. The asis is and Lenovo's and so forth are supposed 538 00:44:54,760 --> 00:45:01,559 to release machines too, and the Snapdragon ships that is in pressive truly, 539 00:45:02,400 --> 00:45:07,559 that's pretty cool. But I'm with you. You know, generally speaking, 540 00:45:07,599 --> 00:45:12,599 the surface units are the most expensive versions of this. Like, i'd be 541 00:45:12,639 --> 00:45:16,000 interesting to see what an Asis or any of the other vendors actually can make 542 00:45:16,039 --> 00:45:21,840 from that gear, and you know how regional price would be. Battery life 543 00:45:21,880 --> 00:45:22,719 should be epic, right, I mean, that's the whole thing with the 544 00:45:22,840 --> 00:45:27,400 ARM chip set is you should be able not even need to charge every day 545 00:45:27,440 --> 00:45:30,039 anymore, right, which that would be nice so Veronica, what's in your 546 00:45:30,159 --> 00:45:36,119 inbox? What's coming up next for you? So? I am planning to 547 00:45:36,360 --> 00:45:45,159 actually play with all the stuff that was just announced Build, especially on responsible 548 00:45:45,199 --> 00:45:50,920 AI side. There are lots of new tools and it is really interesting. 549 00:45:52,039 --> 00:45:58,000 They showed them I built. Looks like they're really cool, working, really 550 00:45:58,000 --> 00:46:05,239 great. I want to see it for my self. Speaking wise, I 551 00:46:05,320 --> 00:46:09,480 am going to Kansas City in a couple of weeks. That's the show we're 552 00:46:09,519 --> 00:46:14,079 going to both be at. Yeah. Yeah, I won't be there, 553 00:46:14,599 --> 00:46:16,039 but we could probably do something remote, Richard, if you want to. 554 00:46:16,519 --> 00:46:22,920 That's fine. We got We're gonna have so many shows between Build and and 555 00:46:22,920 --> 00:46:27,440 and OSLO, and I think we're going to be well into August, so 556 00:46:27,679 --> 00:46:30,800 I wouldn't too far ahead. Well that's really great. What's the talk you're 557 00:46:30,800 --> 00:46:35,440 doing in Kansas City, Fornica. So in Kansas City, I will be 558 00:46:35,480 --> 00:46:40,079 talking about responsibily I thanks to Microsoft, I probably have to change half of 559 00:46:40,119 --> 00:46:49,159 my talk, Yeah, because I before the talk was mostly focused on the 560 00:46:49,280 --> 00:46:54,840 responsibility. I inside the machine learning studio that they have and they still have 561 00:46:54,960 --> 00:47:02,159 it, the dashboard, then the report, but now I feel like they're 562 00:47:02,199 --> 00:47:08,280 moving more towards general of AI analysis and all those content safety tools. So 563 00:47:08,480 --> 00:47:13,079 I do want to include more of it into my talk. And is that 564 00:47:13,159 --> 00:47:15,639 part of as your AI studio. Where's a lot of those safety tools living? 565 00:47:16,239 --> 00:47:20,559 Yeah, so I know they have a separate portal, but I prefer 566 00:47:20,760 --> 00:47:24,920 to access through them as your REI studio. Yeah, which makes sense. 567 00:47:24,960 --> 00:47:29,599 It's a really great place to work. Yeah, and certainly get to you 568 00:47:29,639 --> 00:47:32,800 know, you get a huge dashboard of choices, like how many models would 569 00:47:32,800 --> 00:47:36,800 you like to play with? Because there's way too many. Yeah. Yeah, 570 00:47:36,840 --> 00:47:40,239 I'm looking forward to using that studio as well, just to see what 571 00:47:40,440 --> 00:47:45,360 the heck you can do in it. The demos we're impressive. Yeah, 572 00:47:45,679 --> 00:47:51,280 with that security approach where it's going to keep you from accidentally making a bad 573 00:47:51,360 --> 00:47:57,760 thing as well or putting yourself in danger or inappropriate responses like I don't know, 574 00:47:58,039 --> 00:48:00,960 I got to think the average dev is pretty free. Doubt about incorporating 575 00:48:00,960 --> 00:48:04,519 this as software when you when you have all these terrible stories of how badly 576 00:48:04,559 --> 00:48:07,639 it can go off the rails, right right, You know, I came 577 00:48:07,679 --> 00:48:10,079 for tech support. I got told to get a divorce. We need more 578 00:48:10,159 --> 00:48:16,320 positive stories. Stories are boring, Carl, No, they're not they're amazing. 579 00:48:17,119 --> 00:48:21,280 I mean I told one today. You know, it's just but it's 580 00:48:21,360 --> 00:48:23,199 just a day in the life, you know. How about you, listeners, 581 00:48:23,239 --> 00:48:28,719 Do you have any positive stories of interactions with AI or negative ones for 582 00:48:28,760 --> 00:48:30,400 that matter, but we want to hear the good news. Let us know. 583 00:48:31,199 --> 00:48:34,760 You know how to get in touch with this well. Veronica, thanks 584 00:48:34,840 --> 00:48:37,320 very much for spending the hour with us. It's been a pleasure talking to 585 00:48:37,320 --> 00:48:40,599 you. Thank you so much. You're welcome. Yeah, see in Kansas 586 00:48:40,800 --> 00:49:00,920 and we'll talk to you, dear listener next time on dot net rocks. 587 00:49:06,119 --> 00:49:09,760 Dot net Rocks is brought to you by Franklin's Net and produced by Pop Studios, 588 00:49:10,159 --> 00:49:15,079 a full service audio, video and post production facility located physically in New 589 00:49:15,079 --> 00:49:22,079 London, Connecticut, and of course in the cloud online at pwop dot com. 590 00:49:22,280 --> 00:49:24,159 Visit our website at d O T N E T R O c k 591 00:49:24,400 --> 00:49:30,599 S dot com for RSS feeds, downloads, mobile apps, comments, and 592 00:49:30,679 --> 00:49:35,360 access to the full archives going back to show number one, recorded in September 593 00:49:35,400 --> 00:49:37,920 two thousand and two. And make sure you check out our sponsors. They 594 00:49:38,000 --> 00:49:42,880 keep us in business. Now go write some code. See you next time. 595 00:49:43,800 --> 00:49:45,639 You got jam Vans