1 00:00:01,120 --> 00:00:05,679 How'd you like to listen to dot NetRocks with no ads? Easy? Become 2 00:00:05,719 --> 00:00:09,880 a patron For just five dollars a month, you get access to a private 3 00:00:10,039 --> 00:00:14,439 RSS feed where all the shows have no ads. Twenty dollars a month will 4 00:00:14,480 --> 00:00:19,039 get you that and a special dot NetRocks patron mug. Sign up now at 5 00:00:19,079 --> 00:00:24,160 Patreon dot dot NetRocks dot com. Hey Carl and Richard here with your twenty 6 00:00:24,199 --> 00:00:29,359 twenty four NDC schedule. We'll be at as many NDC conferences as possible this 7 00:00:29,440 --> 00:00:33,600 year, and you should consider attending no matter what. Ndcoslo is happening in 8 00:00:33,679 --> 00:00:37,880 June tenth through the fourteenth. Get your tickets at ndcoslo dot com. The 9 00:00:37,920 --> 00:00:44,000 Copenhagen Developers Festival happens August twenty sixth through the thirtieth. Early bird discount ends 10 00:00:44,079 --> 00:00:51,719 April twenty sixth. Tickets at Cphdevfest dot com. Ndcporto is happening October fourteenth 11 00:00:51,759 --> 00:00:56,880 through the eighteenth. The early bird discount ends June fourteenth. Tickets at Ndcporto 12 00:00:57,079 --> 00:01:12,200 dot com. We'll see you there, we hope. Hey, guess what 13 00:01:12,560 --> 00:01:17,519 it's dot net rocks. I'm Carl Franklin and I'm Richard Camp Richard, where 14 00:01:17,519 --> 00:01:19,439 are you in the world Now? I am in Las Vegas where I'm at 15 00:01:19,480 --> 00:01:26,359 the Microsoft Fabric conference. This is their new data analytics stack. And so 16 00:01:26,760 --> 00:01:30,040 we're putting on a show with me in four thousand of my closest friends. 17 00:01:30,200 --> 00:01:37,040 You know what's crazy is that Kelly has a friend who we don't know what 18 00:01:37,120 --> 00:01:40,079 she does, but we know she's in it. And she said she was 19 00:01:40,120 --> 00:01:42,840 going to Vegas this week to go to some Fabric conference and I had no 20 00:01:42,920 --> 00:01:47,480 idea what it was. And now I know she's hanging out with me. 21 00:01:48,159 --> 00:01:51,200 She might be and we'll talk afterwards. You can hook up with her, 22 00:01:51,719 --> 00:01:55,280 you know what I mean. I'm not saying we up. I'm just saying 23 00:01:55,599 --> 00:02:00,439 maybe made for coffee or tea. Because you're a brit you know. We're 24 00:02:00,439 --> 00:02:02,359 going to be back here in May for dev Intersection. Yeah. It's gonna 25 00:02:02,359 --> 00:02:05,680 be fantastic. And you're signed up for that one. Yeah, I'm really 26 00:02:05,719 --> 00:02:07,599 looking forward to it. I'm going to be doing a Blazer workshop, I 27 00:02:07,639 --> 00:02:12,680 think mm hmmm, yeah. I think it's called Carl Franklin's Blazer Workshop. 28 00:02:12,960 --> 00:02:16,080 I believe it is. So therefore I'm the only one that can teach it. 29 00:02:16,639 --> 00:02:20,919 I guess it's your workshop. That'd be dev intersection dot Com. Yeah, 30 00:02:20,960 --> 00:02:23,280 I hope we'll see you there. Restrictions apply, void, we're prohibited 31 00:02:23,280 --> 00:02:27,319 by law. Okay, order now get a free set against your knives. 32 00:02:29,080 --> 00:02:32,199 See it's five o'clock here, I have wine. Michelle is in Australia. 33 00:02:32,240 --> 00:02:36,240 It's eight o'clock in the morning. She just had brecky eight o'clock in the 34 00:02:36,240 --> 00:02:39,800 morning. Tomorrow. She's in the future. Yeah, it's great love being 35 00:02:39,800 --> 00:02:43,759 in the future. I wasn't going to have a bowl of seal, but 36 00:02:43,800 --> 00:02:46,199 then I sat here and I was starving, so I was like, I 37 00:02:46,280 --> 00:02:49,680 bet it eight alluhs, I'm not gonna last. So your podcast, you're 38 00:02:49,719 --> 00:02:52,759 not gonna make it. We make a long show. We need to shine. 39 00:02:52,840 --> 00:02:55,319 Yeah, exactly. You know, our hardcore listeners know that whenever we 40 00:02:55,400 --> 00:03:00,680 do shows later in the day like now, that we tend to get little 41 00:03:00,719 --> 00:03:04,240 punchy. So like, you know, I'll sell a glass wine. You 42 00:03:04,240 --> 00:03:07,360 don't believe Michelle such a serious person. So right, I don't know how 43 00:03:07,360 --> 00:03:13,159 this is gonna work. That's serious. All right, Well let's get on 44 00:03:13,240 --> 00:03:23,400 with better no framework. All right, dude, what do you got? 45 00:03:23,159 --> 00:03:27,840 I got something that's interesting and a little scary. I like it. So 46 00:03:28,599 --> 00:03:32,840 this is a paper that came out from Microsoft on March thirteenth, This being 47 00:03:32,879 --> 00:03:37,120 Show eighteen ninety two. You can go to one eight nine to two dot 48 00:03:37,240 --> 00:03:46,000 pwoppwop dot E. It's a Microsoft paper entitled autodev Automated AI Driven Development, 49 00:03:46,599 --> 00:03:53,000 and I'll just take the most interesting in sixtync paragraph from it. We present 50 00:03:53,120 --> 00:04:00,840 autodev, a fully automated AI driven software development framework designed for autonomous planning and 51 00:04:00,919 --> 00:04:10,479 execution of intricate software engineering tasks. Autodev enables users to define complex software engineering 52 00:04:10,520 --> 00:04:16,079 objectives, which are assigned to Autodev's autonomous AI agents to achieve. These AI 53 00:04:16,240 --> 00:04:23,160 agents can perform diverse operations on a codebase, including file editing, retrieval, 54 00:04:23,240 --> 00:04:29,199 build processes, execution, testing, and get operations. They also have access 55 00:04:29,240 --> 00:04:33,000 to files, compiler output, build and testing logs, static analysis tools, 56 00:04:33,040 --> 00:04:39,480 and more. This enables the AI agents to execute tasks in a fully automated 57 00:04:39,560 --> 00:04:44,800 manner with a comprehensive understanding of the contextual information required. You know, Richard, 58 00:04:46,000 --> 00:04:53,560 anytime I hear AI and fully automated in a sentence or a paragraph, 59 00:04:53,839 --> 00:05:00,319 I get the willies well, that's just your crap detector going off, gone 60 00:05:00,319 --> 00:05:04,399 what sure? Yeah, well, I mean fully automated kind of means that 61 00:05:05,000 --> 00:05:11,439 it has, you know, it has unfettered access to things. It's building 62 00:05:11,519 --> 00:05:14,319 software. Right, Well, maybe it will, maybe it won't. I 63 00:05:14,319 --> 00:05:16,759 think I think a person wrote that line and hasn't made it true at all. 64 00:05:16,920 --> 00:05:20,360 It's a paper, not a product. Yeah, but eventually, eventually 65 00:05:20,399 --> 00:05:24,120 this is going to happen, right, I mean, And the whole idea 66 00:05:24,160 --> 00:05:29,839 is that software developers may, in fact, in the not so near distant 67 00:05:29,959 --> 00:05:35,199 or distant future, become more supervisors than developers. Well, one would argue 68 00:05:35,199 --> 00:05:40,439 they already are. They mostly supervise the utilizational libraries and glue them together. 69 00:05:41,839 --> 00:05:45,519 Boy, you're just a curmudgeon, aren't you. You're like the AI curmudgeon, 70 00:05:45,959 --> 00:05:49,040 got it'll never work. I'm just more of a Hey, you know 71 00:05:49,079 --> 00:05:54,920 what, I've seen this play a bunch of times. Someone still has to 72 00:05:54,920 --> 00:05:59,040 do requirements gathering, some still has to describe the problem space, someone still 73 00:05:59,040 --> 00:06:01,959 has to validate it. You know. The code writing was the simplest part. 74 00:06:02,240 --> 00:06:06,279 Yeah, oh did I mention this? It only writes Python lovely. 75 00:06:10,199 --> 00:06:13,240 Well, that means it's all c under the hood anyway, So there we 76 00:06:13,319 --> 00:06:17,600 are. Yeah, no, no offense to our Python programming listeners, because 77 00:06:17,639 --> 00:06:21,600 Python is probably the most popular language in the world right now. It's yeah, 78 00:06:21,600 --> 00:06:30,040 absolutely, and it's awesome. Yeah, what is the first javascripts? 79 00:06:30,040 --> 00:06:32,199 You're probably right? Yeah, yeah, I thought you were gonna say, 80 00:06:32,240 --> 00:06:35,920 Jason, and I was going to be sad because it's not a language. 81 00:06:35,920 --> 00:06:42,399 It's for that. He well, anyway, that is when I have it's 82 00:06:43,639 --> 00:06:48,360 it's certainly we're certainly creeping to this. And and I remember having this conversation 83 00:06:48,439 --> 00:06:53,040 with you on dot net rocks in the two thousands Richard, yep. And 84 00:06:53,120 --> 00:06:55,519 do you remember. I got to look up what show it was and maybe 85 00:06:55,560 --> 00:06:59,079 an alert listener can remind us. But I said, you know, there's 86 00:06:59,120 --> 00:07:02,079 going to come a day where a business person can walk up to a machine 87 00:07:02,160 --> 00:07:06,920 and have a conversation with it, you know, either typing or English or 88 00:07:08,160 --> 00:07:12,560 other language, and tell it what the requirements are and it would spit out 89 00:07:12,600 --> 00:07:19,759 a fully ready to go qualified application. And you said that will never happen. 90 00:07:19,959 --> 00:07:25,199 Yep. And it still hasn't. Yeah, it still hasn't happened yet. 91 00:07:25,360 --> 00:07:28,720 No, but but you know, but we've seen we've seen papers like 92 00:07:28,720 --> 00:07:31,839 this before the actually making it happen something else entirely, and I would point 93 00:07:31,839 --> 00:07:35,680 you a copilot and power platform. Yeah, although you find the average business 94 00:07:35,680 --> 00:07:40,680 person cannot describe a business process adequately to explain it to another human, much 95 00:07:40,720 --> 00:07:44,519 less to a piece of suf. There is that problem. Yes, yeah, 96 00:07:44,560 --> 00:07:47,279 Actually breaking down workflows is hard. It takes skill, and it takes 97 00:07:47,399 --> 00:07:53,199 time. It does, and it also takes some kind of uh, some 98 00:07:53,439 --> 00:07:58,279 kind of thing that we don't think machines have yet. Software machines don't have 99 00:07:58,439 --> 00:08:01,160 anything. It's just software. Yeah, programs. I guess. All right, 100 00:08:01,199 --> 00:08:07,319 Well enough of the philosophical bs. Let's hear who's talking to us today. 101 00:08:07,399 --> 00:08:09,519 Richard Gravity comment tub show eighteen fourteen, the one we did with one 102 00:08:09,600 --> 00:08:15,720 Michelle Mannering, you know, back when she was still Michelle Manoring at NDC 103 00:08:15,920 --> 00:08:20,399 Oslo. We did it in person talking about this product called get hub Copilot. 104 00:08:20,399 --> 00:08:22,560 It's the thing that's going to replace all developers. As I recall, 105 00:08:22,759 --> 00:08:31,639 absolutely and Trudje and said, I've been kind of slow with gethub Copilot and 106 00:08:31,680 --> 00:08:35,120 midtally this is now a year ago, but I started last week. This 107 00:08:35,240 --> 00:08:39,639 week I got repaid in full. I cannot understand why I hadn't tried this 108 00:08:39,679 --> 00:08:43,440 earlier. The suggestions from visual Studio good, but this is brilliant. My 109 00:08:43,559 --> 00:08:48,240 current project is a CE sharp Blazer application using Entity framework for creating and seeding 110 00:08:48,279 --> 00:08:54,120 test data to a database with repeating patterns. After the first three or four 111 00:08:54,240 --> 00:08:58,080 entities, get hub Copilot got the grip and wrote almost entirely correct classes in 112 00:08:58,279 --> 00:09:03,080 just a few time. That's because it was trained on Blazer train repos. 113 00:09:03,120 --> 00:09:05,679 Come on, there you go, that's what it's all about. Yeah, 114 00:09:05,720 --> 00:09:09,399 it's all about me. However, the difference is between the eneities made it 115 00:09:09,480 --> 00:09:13,159 hard to just copy the whole class and change them the names Copilot did that 116 00:09:13,200 --> 00:09:16,360 the first few times. By for it embraced pattern and created the unique code 117 00:09:16,360 --> 00:09:20,639 I needed. I guess the result was about ninety percent. It required some 118 00:09:20,799 --> 00:09:22,840 fixes, but all in all, I save so much time on the single 119 00:09:22,840 --> 00:09:26,840 project that I guess get hub Copilot has counted as free for the next few 120 00:09:26,919 --> 00:09:33,120 years for me. I'm seriously impressed. You know where get hub coopilot really 121 00:09:33,159 --> 00:09:37,159 really works well for me is if I have things that repeat, like I've 122 00:09:37,200 --> 00:09:43,039 got five or six lines that you know, I'm working with data in a 123 00:09:43,039 --> 00:09:46,879 grid and there's controls in each cell, right, and it understands that I 124 00:09:48,000 --> 00:09:52,080 have the what the data and the headers are in the grid, and I 125 00:09:52,200 --> 00:09:56,720 make one of these blocks of six lines and then it just starts spitting out 126 00:09:56,759 --> 00:09:58,720 the rest of them and it's like tabin or tabin or tabin or taber. 127 00:10:00,320 --> 00:10:03,600 What I don't like about co pilot and Michelle will probably jump in here and 128 00:10:05,440 --> 00:10:07,679 either agree or disagree with me. But what I don't like about copilot is 129 00:10:07,720 --> 00:10:13,960 when it tries to read my mind and it's almost like like, okay, 130 00:10:13,039 --> 00:10:18,840 here's a story. I have a customer in the studio who wants me to 131 00:10:18,759 --> 00:10:26,120 work on mastering this solo piano CD right, and whenever I say, you 132 00:10:26,120 --> 00:10:28,039 know what I'm gonna do, I'm gonna and he goes, where are you 133 00:10:28,039 --> 00:10:31,000 going to turn down the volume there? Or are you going to bring up 134 00:10:31,000 --> 00:10:33,919 to five K there? Or are you going to be able to And it 135 00:10:33,960 --> 00:10:35,679 goes on like four or five times, and then I just look at him 136 00:10:35,720 --> 00:10:39,360 like, will you shut up and let me say what I'm going to do? 137 00:10:41,200 --> 00:10:46,320 This would go faster. I have two answers, see you that go 138 00:10:46,320 --> 00:10:50,360 ahead. So the first one is get off a copilot can't actually rad your 139 00:10:50,360 --> 00:10:54,879 mind as feels like it might be able to read your mind. All it's 140 00:10:54,919 --> 00:11:00,600 doing is taking context from the file you have opened. Any other open files 141 00:11:01,320 --> 00:11:03,919 sounds like you use visual studios if you have other open tabs. It's taking 142 00:11:03,919 --> 00:11:09,919 the context and that so sometimes client exactly So sometimes it does feel like you're 143 00:11:09,960 --> 00:11:13,159 like, oh, it is trying to read my mind. You're like, 144 00:11:13,480 --> 00:11:16,320 just just wait, like I am going to so a couple of things. 145 00:11:16,320 --> 00:11:20,080 You know, you can just ignore it and keep typing over it, or 146 00:11:20,159 --> 00:11:24,399 you can turn in line suggestions off and mostly use to get up copilot chat 147 00:11:24,440 --> 00:11:30,240 picture. Yeah. What I like is if if you make a comment about 148 00:11:30,320 --> 00:11:33,240 exactly what you want to do, and then it's pretty really really good about 149 00:11:33,639 --> 00:11:39,440 you know, coming like Richard was saying, if you're able to articulate very 150 00:11:39,519 --> 00:11:43,039 clearly what you want, you're going to get a response that is much more 151 00:11:43,279 --> 00:11:48,399 or much closer to what you actually need as opposed to it just actually trying 152 00:11:48,399 --> 00:11:52,120 to read your mind. Because as much as it feels like all these ais 153 00:11:52,159 --> 00:11:54,639 can read our mind, they really can't. As I mentioned before, it's 154 00:11:54,679 --> 00:11:58,240 a bunch of software. It can't read your mind. It can only take 155 00:11:58,279 --> 00:12:03,120 in the context you're providing it, and it's data training set that it has 156 00:12:03,159 --> 00:12:07,399 been already trained on. It's a good suggestion to turn off that inline stuff, 157 00:12:07,480 --> 00:12:13,600 especially when it starts spitting out one hundred online of code solutions and you 158 00:12:13,720 --> 00:12:20,720 finally your screen fills up with gray bloody blah they God help you if your 159 00:12:20,720 --> 00:12:26,759 present, you need to go. It's a great technical term. Great blare 160 00:12:26,039 --> 00:12:31,320 to use that. I like that. Let's say that today Hey trying. 161 00:12:31,399 --> 00:12:33,159 Of course, you kicked off some great conversations. Thanks so much for your 162 00:12:33,159 --> 00:12:37,080 comment. A copy of music Cobey is on its way to un if you'd 163 00:12:37,080 --> 00:12:39,399 like a copy of music Codbey. I read a comment on the website at 164 00:12:39,440 --> 00:12:41,440 dot NetRocks dot com or on the facebooks. We publish every show there, 165 00:12:41,600 --> 00:12:45,799 and if you comment on the show and I read it, we'll send you 166 00:12:45,840 --> 00:12:48,039 copy music Goby. And you can definitely follow us on Twitter if you like. 167 00:12:48,159 --> 00:12:50,480 We've been on Twitter for a long time or ex or whatever the hell 168 00:12:50,519 --> 00:12:54,200 they call it these days. But the cool kids are hanging out. I'm 169 00:12:54,240 --> 00:12:58,440 Master Don, I'm at Carl Franklin at tech head Social, and I'm riche 170 00:12:58,440 --> 00:13:01,480 Campbell at masadon do Social send us a tuote, and of course, if 171 00:13:01,519 --> 00:13:05,159 you're looking for more ways to get in touch with us or Meme particularly, 172 00:13:05,159 --> 00:13:09,919 you can go to Carl Franklin dot com. Okay, let's bring in Michelle. 173 00:13:09,919 --> 00:13:16,600 That person who was interjecting her brilliance there Michelle mish Manner's Duke no longer 174 00:13:16,639 --> 00:13:22,639 mannering. Congratulations on your marriage much Yeah. Michelle Duke is a multi talented 175 00:13:22,639 --> 00:13:28,679 personality in the tech and gaming communities. As a developer advocate, she gets 176 00:13:28,679 --> 00:13:33,600 to create awesome experiences and engage with the vibrant GitHub developer community. Michelle has 177 00:13:33,600 --> 00:13:37,919 spoken at over two hundred and fifty events on topics like AI, the future 178 00:13:37,919 --> 00:13:43,240 of work, communication, teamwork, and has given technical demos. She is 179 00:13:43,240 --> 00:13:48,960 a respected leader in the hackathon community, having one organized and mentored over one 180 00:13:50,039 --> 00:13:54,559 hundred hackathons. Michelle has founded several tech companies, including an AI company, 181 00:13:56,000 --> 00:14:00,399 an e scooter business, and as a result sits at the forefront of Alburn 182 00:14:00,559 --> 00:14:05,039 science, tech, esports and startup scenes in her spare time. In air 183 00:14:05,120 --> 00:14:11,120 quotes, Michelle is a streamer journalist and is always working on something exciting. 184 00:14:11,279 --> 00:14:18,799 Catcher at an event or streaming on Twitch or on dynam Rocks yeah. I 185 00:14:18,840 --> 00:14:24,360 actually haven't streamed for a little while. My computer broke recently, so I'm 186 00:14:24,399 --> 00:14:28,440 reduced to doing not gaming and streaming at the same times. It can no 187 00:14:28,480 --> 00:14:35,200 longer handle that. It's really hard on a machine to do to run OBS 188 00:14:35,240 --> 00:14:37,519 and the game and you know, and all the filters and stuff. Yeah, 189 00:14:37,600 --> 00:14:43,080 so I'm usually running OBS, the game a chat feature, which is 190 00:14:43,080 --> 00:14:46,720 a JavaScript program that is scraping on my chat, my Twitch chat, so 191 00:14:46,759 --> 00:14:50,360 I get to keep that and recording, and it's just it's got to the 192 00:14:50,360 --> 00:14:54,120 point where I'm like, how old is my computer? Now? Okay, 193 00:14:54,159 --> 00:14:56,879 it's five years old. I definitely need a new once, so I haven't 194 00:14:56,080 --> 00:15:01,840 Why do I smell smoke? The benefit the benefit putting your tea on the 195 00:15:03,080 --> 00:15:05,559 gun the CPU? Right? Yeah. No, It's one of those things 196 00:15:05,559 --> 00:15:09,120 where like you turn the computer on and it just like loudly beeps at you 197 00:15:09,240 --> 00:15:13,360 because there's something wrong and won't turn on, and you're like, now, 198 00:15:13,399 --> 00:15:18,120 I really, Or when you're trying to write a presentation for someone and you 199 00:15:18,240 --> 00:15:24,000 realize it takes off full twenty four hours to render the demos within that presentation 200 00:15:24,240 --> 00:15:26,600 that are like a couple of minutes long, and then you're like that's not 201 00:15:26,720 --> 00:15:31,600 right, sounds broken. It's like going on your computer on its way until 202 00:15:31,639 --> 00:15:35,279 then. Yeah, so it's ordered, it's ordered. It's actually one of 203 00:15:35,360 --> 00:15:39,279 those like fancy custom build ones tourn. I've got my graphics and branding on 204 00:15:39,320 --> 00:15:43,399 it, and it looks it looks really awesome, like we just came back 205 00:15:43,399 --> 00:15:48,879 from our honeymoon from Japan as well, so it's all Japanese theme and oh 206 00:15:48,960 --> 00:15:52,320 I see, Yeah, timing matters. You were very Japanese centric when you're 207 00:15:52,320 --> 00:15:56,679 giving a design, So tell me it's fancy on the inside as well. 208 00:15:56,519 --> 00:16:02,799 Yeah, yeah, X five, forty nine ninety like something crazy. It 209 00:16:02,840 --> 00:16:06,879 didn't quite go for the forty ninety. They're a little bit pricey still. 210 00:16:07,000 --> 00:16:10,919 But yeah, we're up there. We're in like the forty eighty series. 211 00:16:10,919 --> 00:16:12,399 We're in, soyeah, you go, that's right up there. Yeah, 212 00:16:12,480 --> 00:16:18,159 exactly, thank you. I built an EPC during the pandemic and it's an 213 00:16:18,200 --> 00:16:25,720 I nine and my fan is just about as deep as high as the case 214 00:16:25,799 --> 00:16:30,279 is deep, and it's just one big looks like a radiator on a seventy 215 00:16:30,320 --> 00:16:33,879 four corvette. You know. That's literally what they are, though, like 216 00:16:34,360 --> 00:16:38,360 the radiators. So my one that had my old here was like this big 217 00:16:38,879 --> 00:16:44,360 sitting which is about thirty centimeters big, sitting on top of a tiny little 218 00:16:44,440 --> 00:16:48,200 like seaping chip that's like a couple of centimeters wide. And you're like, 219 00:16:48,679 --> 00:16:53,919 why, but I've never had a more quiet computer in my life. When 220 00:16:53,960 --> 00:16:56,320 I built the old studio, Richard, do you remember, I put a 221 00:16:56,320 --> 00:17:00,960 hole in the wall so I could put the computer behind the waller room in 222 00:17:00,000 --> 00:17:03,640 a different line I don't have to do. Yeah, yeah, I got 223 00:17:03,680 --> 00:17:07,799 water cooling, water cooling, Yeah, o sid, I got water cooling 224 00:17:07,839 --> 00:17:11,519 in the new one. It's that's how we know. And I also did 225 00:17:11,559 --> 00:17:15,880 it like with one of those things where you're just like, those are unnecessary 226 00:17:15,920 --> 00:17:19,200 amount of turns you've put in the water cooling and it's awful. Looks like 227 00:17:21,039 --> 00:17:26,000 oh show, Yeah I got rid of the phase change. Coolers are good 228 00:17:26,079 --> 00:17:29,720 enough. They're quiet enough. Use the big fans, the one forty millimeters, 229 00:17:30,720 --> 00:17:33,960 the fourteenth centimeter ones are quiet, like you can get a machine right 230 00:17:33,000 --> 00:17:36,559 down. Yeah, yeah, I mean, no reason to go hard. 231 00:17:36,680 --> 00:17:40,000 I have water cooling, but then I also have ten fans in the computer 232 00:17:40,079 --> 00:17:45,240 as well, so I mean there's very nothing succeeds like excess, right, 233 00:17:45,440 --> 00:17:51,519 that's my line fully JFA everything like it's going to also use some photos when 234 00:17:51,559 --> 00:17:56,440 it's done. It looks pretty cool. So, speaking of good co pilot 235 00:17:56,519 --> 00:18:00,279 stuff, are you really going to adopt the grit? What did I say, 236 00:18:00,480 --> 00:18:07,960 Gray bloody Blash, the Great Bloody Blas. I use GitHub copilot a 237 00:18:07,000 --> 00:18:12,519 lot. I do. I do leave the inline suggestions on because I like 238 00:18:12,599 --> 00:18:18,519 it, but I also use the chart feature quite a lot. It's just 239 00:18:18,480 --> 00:18:23,119 the things you can do these days are just incredible. Like we heard from 240 00:18:23,160 --> 00:18:29,119 the quote that Richard read out that it's saving lots of time. I've literally 241 00:18:29,160 --> 00:18:33,000 just read like four or five articles just this morning on various companies and how 242 00:18:33,039 --> 00:18:37,000 their productivity has gone through the roof of using it ub coke pilot. Even 243 00:18:37,759 --> 00:18:41,119 you mentioned we got married recently, so like getting good developer, when you 244 00:18:41,119 --> 00:18:45,759 get married, you turn to GitHub. So I use get up copilot to 245 00:18:45,799 --> 00:18:49,279 build our the website for our wedding, and instead of taking like two four 246 00:18:49,319 --> 00:18:52,480 weeks to build it, which probably what it would have taken me, it 247 00:18:52,519 --> 00:18:57,440 took like eight hours. And the fact that it was able to suggest to 248 00:18:57,559 --> 00:19:03,279 me like things I didn't really or probably wouldn't have quite thought of, like 249 00:19:03,400 --> 00:19:06,920 it. As I was going through it realized what I was doing, which 250 00:19:06,960 --> 00:19:11,079 is building a website for my wedding, and it said started suggesting things like, 251 00:19:11,359 --> 00:19:14,720 oh, would you like a countdown time on your website? How about 252 00:19:14,720 --> 00:19:18,799 a light box to show some lovely photos? The text version of Clippy, 253 00:19:18,920 --> 00:19:22,359 isn't it basically like when you think about like, you know, all the 254 00:19:22,400 --> 00:19:26,119 memes popping up is about you know Clippy and can I help you with that? 255 00:19:26,960 --> 00:19:30,200 That is lit get up copilot is doing with your cod is like I 256 00:19:30,279 --> 00:19:33,559 se you did agree? Is it a business grave or a personal grave? 257 00:19:33,920 --> 00:19:41,920 No, it's for you Clippy, for Clippy Clippi. Have we defined really 258 00:19:41,960 --> 00:19:47,200 a difference between because get how copilot chat came later? Right, the original 259 00:19:47,279 --> 00:19:51,519 co pilot they didn't have the chata No, So the original co pilot was 260 00:19:51,559 --> 00:19:56,559 the inline functionality. So the new getub co pilot has the two types of 261 00:19:56,559 --> 00:20:00,640 functionality. Has that inline line congestion, so as you're coding, it will 262 00:20:00,680 --> 00:20:06,359 show up that grade out text that you can accept or reject. And then 263 00:20:06,400 --> 00:20:08,480 there's the chat feature, which is a little bit more like some of the 264 00:20:08,519 --> 00:20:12,279 other ais that you may have interacted with, where you go in and you 265 00:20:12,599 --> 00:20:17,960 type out either a question or something that you want done, and a copilot 266 00:20:18,039 --> 00:20:21,960 can not just provide you with the code snippet for that, but a small 267 00:20:22,039 --> 00:20:26,799 explanation of what it does, and maybe if there's any elements within that code 268 00:20:26,799 --> 00:20:30,039 snippet that might need to be changed before it gets inserted into your code, 269 00:20:30,359 --> 00:20:34,680 and then you can just insert that straight into your code block and it's same 270 00:20:34,720 --> 00:20:38,079 as the inline code. It is built into the editor you already use. 271 00:20:38,240 --> 00:20:44,920 So the GitHub coropilot Chat for vas Studio came out the end of last year, 272 00:20:44,960 --> 00:20:48,880 so people have been able to use it since then. We just launched 273 00:20:48,960 --> 00:20:53,799 I believe, let me just double check get a propilot chat for jet Brains. 274 00:20:53,839 --> 00:20:57,160 It was on a wait list, I believe it is now it is 275 00:20:57,240 --> 00:21:03,240 now generally available as of this month, so like only like a couple of 276 00:21:03,240 --> 00:21:07,240 weeks ago. So now you can use it in Visual Studio and jet Brain's 277 00:21:07,279 --> 00:21:11,720 IDs as well. So you have that get ub coropilot chat funct now. 278 00:21:11,920 --> 00:21:17,519 So that's yes. Now I've used I've used chat GPT to help me with 279 00:21:17,640 --> 00:21:22,559 programming things, and I've used GitHub copilot chat and the biggest difference that I 280 00:21:22,599 --> 00:21:27,519 can see is that GitHub copilot Chat knows more about your application right off the 281 00:21:27,559 --> 00:21:30,599 bat. You don't have to explain yourself. You don't have to pace your 282 00:21:30,720 --> 00:21:37,000 entire project to chat GPT like you do with So the two main advantages of 283 00:21:37,079 --> 00:21:41,079 Get up Copilot chat over say something like chativity because a lot of people are 284 00:21:41,079 --> 00:21:44,319 saying, well, chat tip t is good. It does provide me with 285 00:21:44,400 --> 00:21:48,000 code suggestions. It's like, well, think about it. Chattivity is being 286 00:21:48,039 --> 00:21:51,880 trained on everything available on the Internet, and there's code on the internet. 287 00:21:52,119 --> 00:21:56,799 Your think stack overflow. Even so chat tivit has that context behind it. 288 00:21:56,839 --> 00:22:02,559 But GitHub Copilot chat because it's built into your editor, it has all that 289 00:22:02,599 --> 00:22:07,400 context as I was kind of hinting at before, which is it has the 290 00:22:07,440 --> 00:22:11,920 context of it being not just trained on everything on the Internet, but having 291 00:22:11,000 --> 00:22:15,839 it trained specifically on code. So it's much more It understands code a lot 292 00:22:15,880 --> 00:22:19,799 better and is optimized better for code. So that's the first point in that 293 00:22:19,839 --> 00:22:25,279 it's training model is much more giared towards understanding code. And that secondly is 294 00:22:25,279 --> 00:22:30,519 it has so much more context. So if you're using visual Studio, any 295 00:22:30,559 --> 00:22:33,680 other things that you've written in the file that you've already got opened, that's 296 00:22:33,759 --> 00:22:37,720 context, we get up copilot, any other tabs that you have open, 297 00:22:37,759 --> 00:22:44,440 so other files that you have open either in one of your jet Brains ideas 298 00:22:44,599 --> 00:22:48,440 or in visual studio or vs code. If you have other files open in 299 00:22:48,519 --> 00:22:53,319 tabs, get up copart can see all of that. And that is all 300 00:22:53,359 --> 00:22:56,960 context, which is why when you're writing things, you might be like, 301 00:22:56,000 --> 00:23:00,119 how does it know what function of called something the file? How does it 302 00:23:00,279 --> 00:23:04,920 know what that class is called? It's because it has that context. And 303 00:23:04,960 --> 00:23:10,039 the other piece of context it has too, is your entire project structure. 304 00:23:10,440 --> 00:23:14,440 So while it can't see what's in each individual file, it can see your 305 00:23:14,440 --> 00:23:17,519 folder structure. So that means it can do things like, you know, 306 00:23:17,839 --> 00:23:21,240 if you're writing a website, say, for example, it can point and 307 00:23:21,319 --> 00:23:25,279 have the you know, the proper location pointer to your CSS file for example, 308 00:23:25,319 --> 00:23:27,599 because it knows where you've stored it, it knows what it's called, 309 00:23:27,680 --> 00:23:33,119 and all those kind of things. And because it then has that folder structure, 310 00:23:33,799 --> 00:23:37,079 it can then see the extensions of all the files, which means it 311 00:23:37,200 --> 00:23:40,799 knows what languages you're trying to write your project in. So it has a 312 00:23:40,839 --> 00:23:45,839 lot more context. And chat chat doing context chattivity has is its training datap 313 00:23:47,000 --> 00:23:48,720 just everything on the Internet and the prompt that you give it. That's it. 314 00:23:48,960 --> 00:23:52,319 Yeah. One of the real problems I had with chatchipt, and I'll 315 00:23:52,319 --> 00:23:57,119 tell you a little story just happened to me. I have this little tool 316 00:23:57,200 --> 00:24:02,759 that I used to mege all of our YouTube videos and by hours, I 317 00:24:02,799 --> 00:24:06,599 mean, you know, the dunt neet rocks. Every show is published to 318 00:24:06,640 --> 00:24:15,920 YouTube, and the speaker publishing system that we use doesn't set the flag that 319 00:24:17,000 --> 00:24:22,920 says it's not made for kids when they publish a show to YouTube, and 320 00:24:22,960 --> 00:24:25,759 so you have to go in. And the reason why that's important is because 321 00:24:25,799 --> 00:24:27,759 there are no comments. If it's made for kids, comments are turned off, 322 00:24:27,799 --> 00:24:33,279 so people can't comment unless you set that value. And it has to 323 00:24:33,319 --> 00:24:37,359 be set when you're creating the video, when you're uploading the video. The 324 00:24:37,400 --> 00:24:41,720 reason and the reason I know that and I found it out is because I 325 00:24:41,759 --> 00:24:45,680 had this conversation with chat gpt about how to set that after the fact, 326 00:24:47,880 --> 00:24:51,440 and you know, it gave me the solution and I said, here's my 327 00:24:51,640 --> 00:24:56,680 code boom, you know, and it said, okay, well, if 328 00:24:56,680 --> 00:24:59,200 this is the case, then here are the things that you can check. 329 00:25:00,000 --> 00:25:03,319 I could see my code but it's still told me that I should do things 330 00:25:03,359 --> 00:25:07,240 that I was already doing in my code, right, and it's like you 331 00:25:07,279 --> 00:25:11,599 can see this, You see that I'm doing that. Why are you giving 332 00:25:11,640 --> 00:25:15,759 me? And the reason is is because it didn't have an answer. The 333 00:25:15,799 --> 00:25:18,799 answer was and it didn't know the answer that it's a read only value. 334 00:25:18,920 --> 00:25:23,960 You can only set it when you upload a video in the API. You 335 00:25:23,960 --> 00:25:27,480 can only set that flag it's available, and you can try to set it. 336 00:25:27,799 --> 00:25:32,119 But it didn't know that. Yeah. So chativity is smart but doesn't 337 00:25:32,160 --> 00:25:36,400 know everything right, doesn't Yeah. And there's a really really great keynote at 338 00:25:36,400 --> 00:25:41,759 one of the n DSA's recently on the different types of ways that charativity and 339 00:25:41,799 --> 00:25:45,480 all the ais can get things wrong. So one of them is it just 340 00:25:45,519 --> 00:25:51,319 doesn't have the facts, so it makes up something. Or the other thing 341 00:25:51,400 --> 00:25:57,200 is it doesn't have enough information and while it's providing you with information, that 342 00:25:57,400 --> 00:26:03,480 is what they call grounded. So it might be true for something else, 343 00:26:03,599 --> 00:26:06,640 but it's not actually true for the context you'll talk about there. It might 344 00:26:06,680 --> 00:26:10,279 have already been true in the past but not anymore, yeah, Or or 345 00:26:10,319 --> 00:26:12,319 it's true for something else. So for example, when I asked chatter TIPT 346 00:26:12,480 --> 00:26:15,440 to write a bio about me. It gets some things right, and then 347 00:26:15,440 --> 00:26:21,079 it says that I have a PhD in like chemistry, which is incorrect. 348 00:26:21,079 --> 00:26:22,440 I didn't know. You're so clever, I know, right, which is 349 00:26:22,519 --> 00:26:25,920 I mean, I don't have HAD. I didn't do chemistry, I don't 350 00:26:25,960 --> 00:26:30,559 have HD. But that information is grounded because there's another Michelle Manoring out there 351 00:26:30,559 --> 00:26:33,960 who lives in New York who potentially has PhD. Right, So the information 352 00:26:34,359 --> 00:26:38,519 is grounded, so it has a base in fact, but it is not 353 00:26:38,880 --> 00:26:44,519 actually correct for me, and chat to PT can't distinguish between what we're trying 354 00:26:44,519 --> 00:26:48,079 to ask about. The other example she gave was, you know, write 355 00:26:48,119 --> 00:26:53,720 me a or give me some information about this airline. I'm going to use 356 00:26:53,799 --> 00:26:59,920 Duke Airlines and exactly you know, tell me about Duke Airlines and chat units 357 00:27:00,119 --> 00:27:03,200 that would be you're going on how it was an amazing airline, it was 358 00:27:03,240 --> 00:27:08,720 always on time, top customer satisfaction, It flew like lots of roots around 359 00:27:08,759 --> 00:27:11,960 the world and had like a fantastic and it doesn't exist, and you're like, 360 00:27:12,279 --> 00:27:15,559 wow, this sounds really good. Doesn't exist, right, So that's 361 00:27:15,599 --> 00:27:21,000 the whole Like AI hallucination thing where it just the thing is when you come 362 00:27:21,039 --> 00:27:23,359 when it comes to the when it doesn't know the answer, it will just 363 00:27:23,400 --> 00:27:26,680 say, well, I really have no you should It should work right. 364 00:27:27,440 --> 00:27:30,599 It does. Sometimes it's like I can't tell you any further, and then 365 00:27:30,640 --> 00:27:34,240 other times I shouldn't. It doesn't say that something that just makes stuff up. 366 00:27:34,400 --> 00:27:37,839 Yeah, it just makes stuff up. Or it says, well, 367 00:27:37,880 --> 00:27:41,079 clearly there's something wrong and you should check all these things that you're already doing. 368 00:27:41,440 --> 00:27:44,720 But you know, a human would figure out a way to go look 369 00:27:44,759 --> 00:27:47,359 for the answer. I was going to say to you. What I think 370 00:27:47,440 --> 00:27:51,319 is interesting is that this is again a personal opinion. But what I think 371 00:27:51,400 --> 00:27:55,640 is interesting is I doesn't mimit humans. Because you're saying a human will go 372 00:27:55,680 --> 00:27:57,839 and look up the information, you'll find it out, not necessarily. I 373 00:27:57,839 --> 00:28:03,279 mean most people, most I know, if they don't know the answer, 374 00:28:03,319 --> 00:28:07,000 they'll make it up on the spot because they are incredible or worded. People 375 00:28:07,039 --> 00:28:11,759 believe them, so they're off, you know, providing incorrect information as well. 376 00:28:12,440 --> 00:28:15,319 I mean I'm definitely one of those people. I'm like, look, 377 00:28:15,359 --> 00:28:18,920 if I don't know the answer, I will go and find out for you 378 00:28:18,960 --> 00:28:22,160 because I know usually know where to get it from. It's the best of 379 00:28:22,359 --> 00:28:25,680 being human and the worst of being human at the same time. And that's 380 00:28:25,720 --> 00:28:27,960 what I think you're getting a chat chibert right, it's they train and everything 381 00:28:27,960 --> 00:28:30,799 available on the internet, you know, not everyone the internet is good. 382 00:28:32,440 --> 00:28:36,839 You do splash. So yeah, I think you're going to get those times 383 00:28:37,200 --> 00:28:38,960 just give to just like to see, you know, at one point, 384 00:28:40,000 --> 00:28:41,759 you know, when it doesn't know the answer, say well, aliens must 385 00:28:41,799 --> 00:28:45,880 have something to do with this, right, you know, just like start 386 00:28:45,000 --> 00:28:49,519 spouting conspiracy theories at least it that way that this is how I want you 387 00:28:49,559 --> 00:28:53,599 to respond when you can't answer, I suppose. So, yeah, you 388 00:28:53,640 --> 00:28:59,799 know, just making software do even dumber things. At the same time, 389 00:28:59,839 --> 00:29:03,680 I've they're talking to some pms and they're I mean, it's not small productivity 390 00:29:03,720 --> 00:29:08,559 provements. It's thirty percent a more code in less time, higher quality, 391 00:29:08,880 --> 00:29:12,920 less remediation. Like these are big numbers. Yeah, yeah, I've got 392 00:29:15,200 --> 00:29:18,400 some stats here. Actually they're just pulled up from a potent exactly what company, 393 00:29:18,480 --> 00:29:26,279 But they said that their developers reported ninety percent increase in writing better code, 394 00:29:26,799 --> 00:29:32,319 which I just think is insane. It's kind of subjective statement. It 395 00:29:32,440 --> 00:29:38,000 is a little bit, but fifty percent more builds, which is not measurable, 396 00:29:38,319 --> 00:29:44,799 which is measurable, eighty four percent increase in successful builds. I think 397 00:29:44,799 --> 00:29:48,079 it's really cool. That's pretty big number. It's a pretty big number again 398 00:29:48,119 --> 00:29:51,599 and a little bit more like subjective on like ninety percent of them feeling like 399 00:29:51,720 --> 00:29:55,720 more satisfied and fulfilled in their jobs. Well, and that's the interesting part 400 00:29:55,720 --> 00:30:00,119 about this, right, is that here's this tool arguably doing some of you 401 00:30:00,200 --> 00:30:03,200 work for you, but you love it. You love it, and what 402 00:30:03,279 --> 00:30:06,400 I think is really good, And this is something that came out from some 403 00:30:06,400 --> 00:30:10,799 of the original research that we did when get ub copilot was first introduced, 404 00:30:10,799 --> 00:30:15,000 and this is before all the chat stuff too, is that developers reported that 405 00:30:15,039 --> 00:30:18,200 they, yes, get ub copilot was writing a lot of code for them 406 00:30:18,440 --> 00:30:22,279 and you know, doing that, but they actually had to think less when 407 00:30:22,319 --> 00:30:23,599 writing code, which I was like, it's not a bad thing, but 408 00:30:23,680 --> 00:30:27,720 they follow up that statement with I have to think less, but when I 409 00:30:27,920 --> 00:30:33,079 do have to think, it's about the fun stuff that sets off that spark 410 00:30:33,160 --> 00:30:36,480 that makes coding fun again. And I think that's what I think is really 411 00:30:36,480 --> 00:30:40,839 great about a lot of these ais. It takes away you know those blockers. 412 00:30:40,920 --> 00:30:44,279 You know, we often get writer's block or coder's block, So it 413 00:30:44,400 --> 00:30:47,559 removes those blockers, so we're not seeing there going or where do we start 414 00:30:47,599 --> 00:30:51,839 again, So it removes that side of it. It writes a lot of 415 00:30:51,880 --> 00:30:55,480 that repetitive, mundane code for you so that you're actually able to sit there 416 00:30:55,480 --> 00:30:59,920 as a developer and do the thing that you would train best to do, 417 00:31:00,119 --> 00:31:03,240 which is solve the problem, like what is the best way to do something? 418 00:31:03,319 --> 00:31:10,759 If I give a problem or a task or project to fifty developers, 419 00:31:11,240 --> 00:31:15,160 I'll probably get fifty. Some of them might be similar, but there's going 420 00:31:15,200 --> 00:31:18,440 to be different ways of writing things. We all know as developers there's multiple 421 00:31:18,440 --> 00:31:21,359 ways of doing something, and that's what you know. Some get up copilot 422 00:31:21,400 --> 00:31:25,920 as well. It will often give you multiple suggestions or if you rephrase a 423 00:31:26,000 --> 00:31:29,240 question, you'll get a different response. So I think the best thing about 424 00:31:29,279 --> 00:31:33,720 being developer is actually looking at what we're doing and going what is the best 425 00:31:33,720 --> 00:31:38,599 way to actually solve this and what is the best way to take this on? 426 00:31:38,759 --> 00:31:42,480 I think what's really exciting. This is another personal opinion. We got 427 00:31:42,519 --> 00:31:47,160 to take a break, but we'll be right back after these very important messages, 428 00:31:51,400 --> 00:31:53,480 and we're back. It's starting at rock some Carl Franklin, that's Richard 429 00:31:53,480 --> 00:32:00,680 Campbell and that is Michelle Mishmanner's duke, and we're talking about get hub, 430 00:32:00,759 --> 00:32:04,599 copilot and AI and all those cool things. And you were just saying that 431 00:32:05,319 --> 00:32:09,200 you think it's really great for removing those blocks, those things that you were 432 00:32:09,240 --> 00:32:12,559 working on that you may have said, you know what, I'm going to 433 00:32:12,640 --> 00:32:15,680 sleep on it or whatever. Now you can get there quicker. And I 434 00:32:15,759 --> 00:32:21,400 was just going to add that it's also good for you to take inventory of 435 00:32:21,920 --> 00:32:27,319 write down the top ten things that you haven't learned because you were afraid to 436 00:32:27,319 --> 00:32:30,839 too busy, or you thought it was over your head, you could never 437 00:32:30,880 --> 00:32:34,960 figure that out. Whatever. Take that list, go through it one by 438 00:32:35,000 --> 00:32:38,359 one with co pilot or you know, your favorite AI, and you'd be 439 00:32:38,400 --> 00:32:42,799 surprised at how fast you learn these things. Yeah, and so yeah, 440 00:32:42,839 --> 00:32:46,759 what I was going to blow everyone's mind was my personalia. So what I 441 00:32:46,880 --> 00:32:50,680 think a lot of these iis and things are going to do. Like if 442 00:32:50,680 --> 00:32:53,920 we look back through history, we say where a lot of new technologies are 443 00:32:53,920 --> 00:32:59,759 coming and people being really upset or skeptical and one I like to look at 444 00:32:59,839 --> 00:33:05,720 is writing. So when writing was first like widely invented and adopted, people 445 00:33:05,880 --> 00:33:09,680 got really worried that we're all going to become dumb because we no longer needed 446 00:33:09,680 --> 00:33:13,960 to remember the things we need to remember. So if I needed to note 447 00:33:14,000 --> 00:33:15,880 something back then, I was there. You were there, you were there, 448 00:33:16,000 --> 00:33:20,400 Yeah, I was there. If I needed to know something I had 449 00:33:20,400 --> 00:33:22,359 to remember, I could just write it down. I mean even today, 450 00:33:22,480 --> 00:33:25,119 like I don't have to remember. I mean I still remember because I've just 451 00:33:25,160 --> 00:33:29,599 got stupid photographic memory. But I don't have to remember. You're absolutely right 452 00:33:29,640 --> 00:33:31,519 about that. Like I don't have to remember my husband's phone number, my 453 00:33:31,559 --> 00:33:35,640 family's phone number, my work phone number. I mean, I remember all 454 00:33:35,680 --> 00:33:37,119 of them because again, photographic memory. But we don't have to. We've 455 00:33:37,119 --> 00:33:40,000 got a phone that does that for us. We were talking to Richard, 456 00:33:40,039 --> 00:33:47,720 we were talking to Grant Barrett from Away with Words, and he mentioned that 457 00:33:49,559 --> 00:33:55,519 in Aristotle's time, when you know, when people were writing handwriting, you 458 00:33:55,519 --> 00:34:00,640 know they were these guys were against it because they thought, well, if 459 00:34:00,640 --> 00:34:02,799 we're if that's cheating, we write it down, we can refer to it 460 00:34:02,880 --> 00:34:07,839 later. We should keep everything in our mind. Exactly. But what actually 461 00:34:07,880 --> 00:34:13,239 happened instead of everyone becoming dumb, is there's this massive wave of innovation invention 462 00:34:13,400 --> 00:34:16,239 because all of a sudden, people's minds were freed up from having to remember 463 00:34:16,280 --> 00:34:21,159 stuff that they could actually think creatively. And I think that's what's going to 464 00:34:21,199 --> 00:34:23,360 happen. I think I'm bigger problems at a higher level exactly. So I 465 00:34:23,440 --> 00:34:27,360 actually think that's what's going to happen over the next five to ten years, 466 00:34:27,440 --> 00:34:31,079 is we're actually going to solve some of the world's biggest problems because all of 467 00:34:31,119 --> 00:34:35,079 a sudden, people's time is free to and you know, it's the fun 468 00:34:35,079 --> 00:34:37,239 stuff. We don't have to sit there, you know, doing a lot 469 00:34:37,280 --> 00:34:40,519 of that mundane competitive stuff. We can do what those original developers are same 470 00:34:40,519 --> 00:34:45,400 when we did the research, start to get up copilot, which is it 471 00:34:45,440 --> 00:34:47,599 sets off this spark which makes things fun again, right, so we can 472 00:34:47,679 --> 00:34:52,199 now spend more time actually, you know, thinking about how to solve these 473 00:34:52,239 --> 00:34:57,119 problems. I think we're going to get this big wave of just cool stuff 474 00:34:57,159 --> 00:34:59,480 happening. Again, this is all personal opinion, but I think we're going 475 00:34:59,519 --> 00:35:01,119 to see a lot We've already had a wave in Richard, I'd be interested 476 00:35:01,159 --> 00:35:04,920 in what you have to say about this. But we've already had a big 477 00:35:04,960 --> 00:35:12,599 wave of people using it for nefarious purposes before you know, the and it's 478 00:35:12,639 --> 00:35:15,760 a game of cat and mouse and ketch up with the you know, the 479 00:35:15,800 --> 00:35:20,440 authorities in the security business. But not in gethub copilot. No, no, 480 00:35:20,440 --> 00:35:22,880 no, no no, I'm just talking AI in general. Yeah, 481 00:35:22,079 --> 00:35:24,920 but you know, I was thinking more in terms of the productivity boosts we 482 00:35:24,960 --> 00:35:28,599 got when we got the dot net framework. Oh yeah, and you already 483 00:35:28,639 --> 00:35:32,199 had an encryption library. I already had sockets and like you didn't have to 484 00:35:32,199 --> 00:35:36,320 write all that stuff. And all the C plus plus programmers were like, 485 00:35:36,480 --> 00:35:39,639 Dad, I don't want that's so lazy. Yeah, real programmers allocate and 486 00:35:39,679 --> 00:35:44,400 deallocate their memory manually. Come on. Like, I just think it's a 487 00:35:44,440 --> 00:35:46,599 really exciting time. Like I know a lot of people to get worried about 488 00:35:46,599 --> 00:35:50,320 the whole you know, what if it takes my jobs and what if it 489 00:35:50,360 --> 00:35:53,880 does this? And you know what, I again, personal opinion, there 490 00:35:53,880 --> 00:35:58,800 are going to be some jobs that get get lost. That's just the reality 491 00:35:58,880 --> 00:36:01,079 of every industrial revolution should move hard over to us, you know, since 492 00:36:01,079 --> 00:36:05,079 the dawn of time. But I think with this song. What's going to 493 00:36:05,079 --> 00:36:07,960 be really interesting is the majority of people aren't going to lose their jobs to 494 00:36:07,000 --> 00:36:10,199 AI. They're going to lose their job to another human who's using AI. 495 00:36:10,960 --> 00:36:17,360 So the people who can understand how how these things work, so understand AI, 496 00:36:17,639 --> 00:36:22,719 understand the capabilities of AI, but more importantly understand the limitations of AI, 497 00:36:22,920 --> 00:36:25,519 and you therefore, yeah, therefore, where they can add value as 498 00:36:25,519 --> 00:36:30,480 a human, they're the people who are going to have jobs. It's exactly 499 00:36:30,519 --> 00:36:35,760 my position. Yeah again tall personal opinion. I may play the role of 500 00:36:35,760 --> 00:36:40,480 a lad eight for you know, for communic effect, but I'm absolutely with 501 00:36:40,679 --> 00:36:45,079 you on that that you know this is not something to be feared. You 502 00:36:45,119 --> 00:36:49,440 have to you have to jump on the horse and write it. So if 503 00:36:49,480 --> 00:36:52,559 anyone is listening of like, okay, that sounds good, what do I 504 00:36:52,639 --> 00:36:54,599 do, like, actually go and start using AI. If you're not using 505 00:36:54,679 --> 00:37:00,840 AI, you need to start using it so you really understand what it can 506 00:37:00,880 --> 00:37:04,519 actually do and then what it can't do if you And it's hit me that 507 00:37:05,199 --> 00:37:08,679 developers uniquely sued for these kinds of tools kind of sad. Is the reason 508 00:37:08,760 --> 00:37:13,039 Get Help Copilot took off first? Yes, because we get it. It's 509 00:37:13,119 --> 00:37:16,079 just another kind of coding to describe a task to the tool, But do 510 00:37:16,119 --> 00:37:20,320 you still have to evalue it. We're used to criticizing other people's code and 511 00:37:20,440 --> 00:37:23,639 our own, and the compiler always gets the same. And you're right, 512 00:37:23,760 --> 00:37:25,920 Like, you know, a lot of people just think, well, you 513 00:37:25,960 --> 00:37:29,239 know, is it going to be this age where get up copil it just 514 00:37:29,280 --> 00:37:30,840 writes code and ships it off. It's like, no, that's that's really 515 00:37:30,840 --> 00:37:34,360 stupid, Like no one would do that, Like, no one would you 516 00:37:34,679 --> 00:37:38,519 give a project to an intern and ship it off before you know, before 517 00:37:38,599 --> 00:37:42,320 checking it. Right, it's kind of the same thing to get up copile. 518 00:37:42,360 --> 00:37:44,400 In a way, we kind of go to treat it a little bit 519 00:37:44,440 --> 00:37:47,519 like an intern in that sense that you know, we as developers are responsible 520 00:37:47,559 --> 00:37:50,960 for the code we ship at the end of the day, regardless of how 521 00:37:51,039 --> 00:37:53,079 much of it was written by an AI, which means as developers, we 522 00:37:53,199 --> 00:37:58,559 have to be responsible for checking the code, checking the security of it, 523 00:37:58,639 --> 00:38:01,159 you know, doing all those kinds of normal things that we would do before. 524 00:38:01,480 --> 00:38:05,960 And I like your actorism, Michelle, because I'm sure people are going 525 00:38:06,039 --> 00:38:10,320 to do that. They're just going to get well, you know, back 526 00:38:10,360 --> 00:38:15,480 in the early days of chat GPT with a lawyer asked for citations from chat 527 00:38:15,519 --> 00:38:20,960 GPT. Yeah, go fictional ones and fictional case. The soft there wasn't 528 00:38:20,960 --> 00:38:23,719 in trouble, the lawyer was, the court doesn't care. But there was 529 00:38:23,760 --> 00:38:30,159 an article recently about someone who used chat GPT and they didn't even bother changing 530 00:38:30,320 --> 00:38:32,800 like you know, it was like hi, insert name here, like they 531 00:38:32,800 --> 00:38:37,360 didn't changing any of that stuff. And you're like why, like what's what 532 00:38:37,559 --> 00:38:43,840 is wrong? Like lazy? Are you? Yeah? Exactly? I have 533 00:38:43,920 --> 00:38:50,880 to bring up something here. My friend and aphoenex colleague, Brian uh sent 534 00:38:50,920 --> 00:38:59,239 me this thing called pseudo no Suno suno app dot suno dot ai and it's 535 00:38:59,280 --> 00:39:06,880 apparently this thing that can generate music and it sounds like anything that you could 536 00:39:06,880 --> 00:39:12,639 hear in a genre specific radio station, right, and we complete with singing 537 00:39:12,719 --> 00:39:15,199 and lyrics and everything. And he thought this was the coolest thing, and 538 00:39:15,239 --> 00:39:17,840 he said, so what I've I let it go for about a week. 539 00:39:17,880 --> 00:39:21,960 Hecaes, what, no comment, And I'm like, Okay. From a 540 00:39:21,960 --> 00:39:28,440 technology perspective, it's pretty impressive, but from a music perspective, it's soulless 541 00:39:28,480 --> 00:39:35,559 and soul draining and just like crushing. Yeah, because because what it does 542 00:39:35,719 --> 00:39:39,960 is you know, if somebody actually puts out a hit record, you know 543 00:39:40,000 --> 00:39:46,320 that's completely generated by AI. It takes away all the integrity of real artists 544 00:39:46,320 --> 00:39:50,920 that are doing real things. And it's you know, for the musicians out 545 00:39:50,960 --> 00:39:53,360 there, get busy, you got a tour, Yeah, you got to 546 00:39:53,400 --> 00:39:57,159 let people know that you're doing it for real. It's the same, right, 547 00:39:57,280 --> 00:39:59,239 right. You know, I mentioned in my bio I'm a journalist, 548 00:39:59,360 --> 00:40:01,639 right, so a natural a lot of people asking me, you know, 549 00:40:01,719 --> 00:40:06,559 are we using chattivt And it's like, well, we're not. We know 550 00:40:06,639 --> 00:40:10,719 a lot of publications who are using chat activity. We're basically told, like 551 00:40:12,280 --> 00:40:15,840 in our our sphere that like if you use chatchivit, you're gonna get fired 552 00:40:16,280 --> 00:40:22,079 because you're right, like, it's a lot of it is generic. It's 553 00:40:22,119 --> 00:40:24,320 soulless. It's not like I mean they said, like, you know, 554 00:40:24,400 --> 00:40:28,760 sure, go use it to like get some you know, get some like 555 00:40:28,880 --> 00:40:30,719 a base of something where to start, but actually write something yourself, because 556 00:40:30,760 --> 00:40:36,360 otherwise it is it is still plagiarism, right, rather, you're plagiarizing something 557 00:40:36,400 --> 00:40:38,360 that's that just hadn't been written. But that's why I think a lot of 558 00:40:39,000 --> 00:40:44,280 you know, we our news site, so I write for Upcoming. It's 559 00:40:44,280 --> 00:40:47,639 a gaming and like esports kind of publication. So we kind of moved away 560 00:40:47,679 --> 00:40:51,760 a lot from the like the generic news stuff, and so we write a 561 00:40:51,760 --> 00:40:57,719 lot of you know, guides and interview type pieces because it's it's much more. 562 00:40:58,199 --> 00:41:01,000 It's it's different content, but it all so is like that more heartfelt 563 00:41:01,000 --> 00:41:07,440 and thought provoking stuff, and like writing a guide and exactly, and so 564 00:41:07,599 --> 00:41:12,079 writing a guide like I write a guide, and this is why I love 565 00:41:12,079 --> 00:41:14,800 writing my technology guides too, which you can read a lot of them on 566 00:41:14,840 --> 00:41:20,360 dev two or like on dev is, I write a guide as I'm doing 567 00:41:20,400 --> 00:41:25,480 it, so and like I'm got no knowledge of what I would have had 568 00:41:25,519 --> 00:41:29,400 to do in the first place. That's why I love writing beginner content as 569 00:41:29,400 --> 00:41:34,519 well. So and AI can't do that. They can't necessarily put themselves and 570 00:41:34,760 --> 00:41:37,599 you could tell it to go, hey, write this on the perspective of 571 00:41:37,639 --> 00:41:40,639 someone who's never done this, but it's really hard for an AI to do 572 00:41:40,719 --> 00:41:45,159 that because it doesn't quite understand that human comment. It's like when you watch 573 00:41:45,480 --> 00:41:51,079 a video or see some artwork. I mean, there's so much AI generated 574 00:41:51,159 --> 00:41:54,760 artwork out there, and I can probably bet that majority of people, regardless 575 00:41:54,800 --> 00:42:00,760 of whether they're working in technology or not, can pick up AI content like, 576 00:42:02,119 --> 00:42:06,559 right, I've been able to pick up dally generated content because it has 577 00:42:06,559 --> 00:42:08,480 a look about it. Yeah, exactly right. Do you notice this? 578 00:42:08,639 --> 00:42:12,239 Yeah? Yeah, there's a specific look. Like I'm looking at this stuff 579 00:42:12,320 --> 00:42:15,639 now, going that's a really cool picture of like a Pika two. But 580 00:42:15,920 --> 00:42:19,519 I know that's been AI generated. And then you look into it has been 581 00:42:19,679 --> 00:42:21,800 and everyone's like, how do you know? And you're like, you just 582 00:42:21,920 --> 00:42:25,280 know, like you just have this like does this look and a feeling about 583 00:42:25,320 --> 00:42:29,960 it where you're just like, it's been AI generated. However, okay, 584 00:42:30,039 --> 00:42:34,079 However, the thing is is that we keep complaining about this stuff not being 585 00:42:34,599 --> 00:42:37,840 as good as humans, and then it keeps raising the bar and raising the 586 00:42:37,880 --> 00:42:40,840 bar, and you know, twenty years from now, who knows, right, 587 00:42:42,239 --> 00:42:45,119 exactly right, Eventually it's going to catch up with our complaints. And 588 00:42:45,559 --> 00:42:50,599 I am one of those people. I fall into categy again more personal preference 589 00:42:50,760 --> 00:42:54,920 and personal opinion. I love the AI generated stuff because I think it's really 590 00:42:55,000 --> 00:43:01,920 good for people who don't necessarily have that creative I want to have something different 591 00:43:02,000 --> 00:43:05,760 in like I'm thinking here, too, Like I use a lot of it 592 00:43:05,800 --> 00:43:09,119 for my presentations, right because I want something that is a little bit unique, 593 00:43:09,159 --> 00:43:14,000 a little bit different, and I can't find exactly what I'm looking for. 594 00:43:14,119 --> 00:43:19,000 So being able to use some sort of AI generated thing to create pretty 595 00:43:19,039 --> 00:43:21,320 much what I want is really good. Now, a lot of people say 596 00:43:21,360 --> 00:43:22,840 to me, oh, but that's taking work away from artists. I'm like, 597 00:43:23,079 --> 00:43:25,519 well, on the flip side, let's look at how much work I 598 00:43:25,559 --> 00:43:30,480 have on my computer and within my office that is created by an actual artist 599 00:43:30,519 --> 00:43:35,280 that I commission work for. I do a lot of that kind of stuff 600 00:43:35,280 --> 00:43:37,599 because those are the kind of like I have stickers I have on my Twitch 601 00:43:37,679 --> 00:43:42,039 branding all that kind of stuff. But I'm not going to go and commission 602 00:43:42,039 --> 00:43:45,679 an artists for every single PowerPoint slide that I want done, Like that's really 603 00:43:45,719 --> 00:43:50,559 costly and it's actually not a good use of the artist time either. Like 604 00:43:50,599 --> 00:43:54,280 I've got artists she is actually doing a PowerPoint slide for me at the moment, 605 00:43:54,280 --> 00:43:58,400 but it's a reusable slide, like I'll be using it for every presentation. 606 00:43:58,639 --> 00:44:00,440 Yeah, you're not taking work away. It's worked that wouldn't be otherwise 607 00:44:00,440 --> 00:44:02,679 be done. It's the work that want to be done, because she's telling 608 00:44:02,679 --> 00:44:06,199 me, she's like, I love doing this stuff, but it takes a 609 00:44:06,199 --> 00:44:08,039 long time. Like I think she'd been working on it for about three months 610 00:44:08,039 --> 00:44:12,840 now, like on a slide. Like it's you know, like if we 611 00:44:13,440 --> 00:44:15,840 all wanted to go and commission it, like we just do way too much 612 00:44:15,920 --> 00:44:20,679 work. It's I just think it's a bit insane. So I think having 613 00:44:21,320 --> 00:44:25,039 that good balance of like being able to create things that are quick and easy 614 00:44:25,079 --> 00:44:30,199 that we wouldn't have gotten artists to do, and then still having artists for 615 00:44:30,639 --> 00:44:35,000 the really high quality, good work that we always want to do, folks 616 00:44:35,000 --> 00:44:37,960 who have been Usingdali to generate the rough of what they want and then taking 617 00:44:37,960 --> 00:44:43,119 that to the artist saying I want this but different, you know, in 618 00:44:43,159 --> 00:44:45,840 this style looking at way good for a sketch. All right, So, 619 00:44:45,960 --> 00:44:47,800 Michelle, I have a hypothetical situation for you. Let's say you have a 620 00:44:47,800 --> 00:44:52,880 little girl, and you know she's five or six years old, about the 621 00:44:52,880 --> 00:44:58,239 time that you might want her to start taking piano lessons and learning an instrument 622 00:44:58,320 --> 00:45:00,719 is a really good thing for a child because you know, it can be 623 00:45:00,760 --> 00:45:06,320 a comfortable comfort in times of struggle. It expands your horizons, like it's 624 00:45:06,440 --> 00:45:09,760 just good for people to learn an instrument, whether you plan on being a 625 00:45:09,800 --> 00:45:15,320 performer or not. And then the kid is you know, hip with the 626 00:45:15,360 --> 00:45:17,599 tech and the YouTube and AIS that comes to you and says, but mom, 627 00:45:19,440 --> 00:45:22,760 I don't need to learn piano. I have a program that does that. 628 00:45:22,119 --> 00:45:24,800 Okay. So I got two responses for you. The first one is 629 00:45:24,800 --> 00:45:30,880 I have four nieces, so this is a very much a hot topic of 630 00:45:30,920 --> 00:45:35,760 conversation amongst my siblings. And second thing, I actually had someone come to 631 00:45:35,800 --> 00:45:39,199 me when I was at State of Open Con, which was less than a 632 00:45:39,199 --> 00:45:45,079 month ago. Young girl, she's in year ten or eleven, and she 633 00:45:45,320 --> 00:45:50,360 literally said to me, why do why should I study computer science and software 634 00:45:50,400 --> 00:45:52,320 engineering? I don't want to be a developer because AIS are going to do 635 00:45:52,360 --> 00:45:58,320 everything for us, Like literally said that. This isn't even a hypothetical anymore. 636 00:45:58,400 --> 00:46:02,079 This is young kids in high school saying I don't want to study yourself 637 00:46:02,159 --> 00:46:06,000 by engineering because an AI is going to do it all for me. And 638 00:46:06,079 --> 00:46:09,559 my response, yeah, exactly. And so my response to her was very 639 00:46:09,599 --> 00:46:14,000 similar what we're already telling you on this podcast is that like AI can't do 640 00:46:14,119 --> 00:46:19,000 everything like and she was like oh, but it's it is just growing exponentially. 641 00:46:19,000 --> 00:46:22,239 I'm like, yeah, but if you have done exponential blog, it 642 00:46:22,320 --> 00:46:25,079 does go like this and plateaus out right. We're still in the you know, 643 00:46:25,159 --> 00:46:29,320 the upward phase. But I think there's going to gain personal opinion, 644 00:46:29,679 --> 00:46:32,679 I think we're going to hit a point where we can't go any further because 645 00:46:32,719 --> 00:46:37,000 of it. Whether yeah, we're basically there, but whether it's because of 646 00:46:37,320 --> 00:46:42,760 limitation. There is another order of magnitude of data to be collected off the 647 00:46:42,760 --> 00:46:46,599 internet exactly. So it's like whether we're going to be limitated by you know, 648 00:46:46,920 --> 00:46:52,519 literal physical data points, the speed of electricity, whatever it's going to 649 00:46:52,559 --> 00:46:54,719 be. There's going to be a limit point, right. What I think 650 00:46:54,960 --> 00:46:58,360 is that there will, Yeah, there will be that limit. But the 651 00:46:58,440 --> 00:47:05,960 limit that I think is more important is the public's rejection of AI and because 652 00:47:06,159 --> 00:47:10,000 because it's going to produce so many Charlatans and so many so much phony stuff 653 00:47:10,719 --> 00:47:15,039 that people are going to want to they're gonna learn, they're going to yearn 654 00:47:15,840 --> 00:47:22,679 for human to human communication, contact sharing of arts and culture. Again, 655 00:47:22,880 --> 00:47:27,519 it's funny they said the same thing about the telephone. It was going to 656 00:47:27,599 --> 00:47:30,159 limit human communication. I think I think we're not there, Richard. I'm 657 00:47:30,159 --> 00:47:35,559 saying, like, in the future, I think we're going to have a 658 00:47:35,639 --> 00:47:39,599 situation where there'll be a rejection of artificial anything. I think to certain agree, 659 00:47:39,639 --> 00:47:43,519 But my response to this girl was very somewhat we're saying here like a 660 00:47:43,639 --> 00:47:45,920 counter everything you as the developer is still in charge, like we still need 661 00:47:45,960 --> 00:47:49,800 to direct all those kind of things. So after a while she was like, 662 00:47:49,840 --> 00:47:52,039 I actually want to starty computer sides now, So she got really excited 663 00:47:52,400 --> 00:47:58,480 about the prospect of being able to then use AI as opposed to being wary 664 00:47:58,519 --> 00:48:02,079 of it. And to your point about like, you know, the public 665 00:48:02,159 --> 00:48:07,920 rejecture, I don't necessarily think it's going to be like a public rejection over 666 00:48:07,960 --> 00:48:09,519 all of AI, because there's so many people in tech that are just like, 667 00:48:09,519 --> 00:48:13,000 oh my gosh, it's stuff so cool. What I think we're going 668 00:48:13,079 --> 00:48:16,440 to get again more personal opinion, is we're going to have the Ikea model, 669 00:48:16,920 --> 00:48:21,119 which is there's gonna be a lot of stuff out there that's built by 670 00:48:21,159 --> 00:48:22,880 AI and it's all got and you can tell it's being built by AI. 671 00:48:23,239 --> 00:48:28,880 But then there's going to be certain pieces where you're going to go that was 672 00:48:28,920 --> 00:48:32,199 built by a human and therefore it is bespoke, it is custom, it 673 00:48:32,280 --> 00:48:37,239 is more expensive. So same with the Ikea furniture thing, is why I 674 00:48:37,320 --> 00:48:39,519 use IQ. So many people have Ikea furniture in the house, right, 675 00:48:39,599 --> 00:48:44,280 it does the job. But if there are no more furniture makers because they 676 00:48:44,320 --> 00:48:46,039 all think, well why should I learn how to make furniture if I can 677 00:48:46,079 --> 00:48:50,039 just buy it here? And then I think, because there will be a 678 00:48:50,119 --> 00:48:53,239 lack of those people, that they will be the ones that are left being 679 00:48:53,239 --> 00:48:57,960 able to do that will be sought out. So it's like a renaissance, 680 00:48:58,079 --> 00:49:01,119 right exactly happened because of the depress and all of that. I think there 681 00:49:01,119 --> 00:49:06,519 may be another renaissance, but in this in this situation, not a rejection 682 00:49:06,599 --> 00:49:10,599 of technology, but just a recognition of human value. Yeah, basically, 683 00:49:10,679 --> 00:49:14,440 And so like you might have lots of buy Kea furniture in your house, 684 00:49:14,440 --> 00:49:16,599 but there might be one or two pieces that aren't ike and you spend a 685 00:49:16,639 --> 00:49:20,320 lot of money on Now, as people, we usually can't you know, 686 00:49:20,599 --> 00:49:23,880 generally and especially the cost of living, we can't afford to have every piece 687 00:49:23,960 --> 00:49:28,960 in our home built by you know, a beautiful furniture maker. In the 688 00:49:29,000 --> 00:49:31,480 same way we can't have every piece of artwork you know, in our house 689 00:49:32,079 --> 00:49:36,840 because it's too expensive, but there will be those, you know, really 690 00:49:36,840 --> 00:49:40,760 bespoke pieces. And I think you've hinted on something really important is that those 691 00:49:40,880 --> 00:49:46,159 people who really understand that in their craft will get better at it so that 692 00:49:46,320 --> 00:49:51,920 it has something different that AI doesn't. And I use that example because my 693 00:49:51,960 --> 00:49:55,599 brother is a photographer and a videographer and he loves the AI stuff. He 694 00:49:55,760 --> 00:49:59,280 was like, you know, it makes my job a lot, you know, 695 00:49:59,360 --> 00:50:00,960 a lot of the things automated a lot easier, He's like, but 696 00:50:01,000 --> 00:50:06,559 it's also made me a better photographer and a better videographer because now I'm learning 697 00:50:06,599 --> 00:50:12,519 techniques and ways of shooting that's completely different that AI cannot do. And so 698 00:50:12,599 --> 00:50:17,239 I think, you know, the smart people and the people who you know, 699 00:50:17,400 --> 00:50:22,360 I have that more optimistic viewpoint are going to look at their craft and 700 00:50:22,400 --> 00:50:27,400 say, how do I add a bespoke human element that people are going to 701 00:50:27,440 --> 00:50:30,920 cover it, and that people are going to are going to be willing to 702 00:50:30,000 --> 00:50:34,199 pay the extra money for it. And you can only see the horizon if 703 00:50:34,239 --> 00:50:36,920 you get on the horse and ride. That's exactly right, And I think 704 00:50:36,920 --> 00:50:40,800 it doesn't really matter. I think what kind of profession you're in. I 705 00:50:40,840 --> 00:50:45,280 think that's going to be across the board, whether you're a developer writing, 706 00:50:45,360 --> 00:50:49,519 you know, building a website, whether you're an artist, you know, 707 00:50:50,400 --> 00:50:53,599 creating artwork, whether again you're you know, even physical things like furniture and 708 00:50:53,599 --> 00:50:57,000 stuff like that, where I mean, that's already we're already seeing that, 709 00:50:57,119 --> 00:51:00,559 because you know, there's robots that can manufacture for a future like at ridiculous 710 00:51:00,559 --> 00:51:06,239 speeds. But I still want a nice bespoke custom. You want a story 711 00:51:06,360 --> 00:51:10,119 rather, Yeah, AI has no story. Oh I generated this on AI. 712 00:51:10,239 --> 00:51:14,360 Okay, but what's the story behind it? Yeah? The story. 713 00:51:14,719 --> 00:51:19,760 The story is the prompt you wrote. Yeah, a lot of things like 714 00:51:19,960 --> 00:51:23,559 especial occasions and things like that. People want that human touch. Weddings, 715 00:51:23,679 --> 00:51:29,199 anniversaries, birthdays, those kind of things where people are going to want that 716 00:51:29,280 --> 00:51:34,880 extra that extra quality and that extra human touch and knowing that that it's been 717 00:51:35,199 --> 00:51:39,159 human touched. Yeah, yeah, very very good. I agree. Oh 718 00:51:39,199 --> 00:51:45,159 my god, what else should we talk about, Richard? I don't know. 719 00:51:45,280 --> 00:51:46,800 I think there was some copilot in here. Somewhere, there was some 720 00:51:46,840 --> 00:51:52,119 copilot. I'm so we got off on that tangent. It's good because I 721 00:51:52,159 --> 00:51:55,320 think it does prompt a broader discussion around everything. You know, I go 722 00:51:55,400 --> 00:51:58,960 into the office, we need to keep that discussion going. Yeah, And 723 00:51:59,000 --> 00:52:01,159 I go into the the startup office now and people come up to me and 724 00:52:01,199 --> 00:52:04,719 go, hey, you're into this AI stuff. What do you think about 725 00:52:04,719 --> 00:52:07,599 this? This? And this nothing to do with you know, get up 726 00:52:07,639 --> 00:52:13,159 a copilt. But because it's just so ingrained in every conversation now, I 727 00:52:13,239 --> 00:52:16,119 just think it's you know, every it's touching everyone. That that is why 728 00:52:16,199 --> 00:52:22,320 we're getting this real just uproar in society about like everything happening, because it's 729 00:52:22,639 --> 00:52:25,760 it's all, it's everywhere. You cannot turn away, and you can't. 730 00:52:25,840 --> 00:52:29,119 You can't get away from it, essentially unless you want to go live on 731 00:52:29,119 --> 00:52:31,840 a deserted island by yourself. Yeah. I know it's a lot of gardeners 732 00:52:31,880 --> 00:52:36,000 who are too immersed in it right now. I think we are immersing it 733 00:52:36,039 --> 00:52:37,719 in the industry. I do think it's just another set of tools. The 734 00:52:37,719 --> 00:52:43,880 biggest problem that it has is the name. We've spent sixty years making crazy 735 00:52:44,039 --> 00:52:47,079 movies where the AI tries to kill everybody. So gee, I wonder why 736 00:52:47,119 --> 00:52:51,280 people are jumpy about this name. Well, it was like, I want 737 00:52:51,280 --> 00:52:53,199 to did my popcom talk last year. I was like, didn't we learn 738 00:52:53,239 --> 00:53:00,480 anything from like Terminator one or two or three or four or five? You 739 00:53:00,480 --> 00:53:05,440 know, you're right, like, but the technology in those movies has nothing 740 00:53:05,440 --> 00:53:07,440 to do with the technology of making here except for the name. It's well 741 00:53:07,480 --> 00:53:13,079 said Wretchard, Yeah exactly. And I think what's interesting is I love looking 742 00:53:13,119 --> 00:53:19,519 at some science fiction movies is because it shows what people's minds were actually thinking 743 00:53:19,599 --> 00:53:23,760 would be potentially capable in you know, twenty or thirty years time. And 744 00:53:23,800 --> 00:53:27,880 I think like some of the Cape Billy people always come to me and say, 745 00:53:27,880 --> 00:53:30,119 oh, how far are we away from getting like Iron Man's you know, 746 00:53:30,360 --> 00:53:34,800 Javis am I afforded there? Like, you know, that kind of 747 00:53:34,880 --> 00:53:39,079 capability is already available. While it might not be commercially available in the sense 748 00:53:39,079 --> 00:53:44,000 that you and I can't just go out and buy our own Javas, a 749 00:53:44,039 --> 00:53:47,360 lot of those functionalities are already there. And you know Open Eye has already 750 00:53:47,400 --> 00:53:52,920 put Chapter VT. I think it was five they put into a humanoid robot. 751 00:53:52,320 --> 00:53:55,800 I don't know if you saw the new in Video Press the video Big 752 00:53:55,840 --> 00:54:00,159 Conference in who are there? But you know, robots like they're just it's 753 00:54:00,159 --> 00:54:04,039 like we created human eyed robots and you know, stuff them with all the 754 00:54:04,800 --> 00:54:07,679 stuff from the Internet and you're like, okay, I can't see what's going 755 00:54:07,719 --> 00:54:13,320 wrong at all. Yeah, all right, So ride the horse, get 756 00:54:13,320 --> 00:54:16,679 GitHub co pilot, kitt gethub co pilot chat, use it. It's good 757 00:54:17,320 --> 00:54:22,480 and you know you and find your way in the you know, in the 758 00:54:22,639 --> 00:54:25,440 in the wave of the future. Find your spot on that wave and ride 759 00:54:25,519 --> 00:54:34,159 it. That's that's it, and stop arguing with scenes from movies because those 760 00:54:34,159 --> 00:54:38,840 aren't real. Well, and you know, anthromorphizing software is never good. 761 00:54:39,039 --> 00:54:42,519 No, it's not. You're you're right, Richard, But it is easy, 762 00:54:42,719 --> 00:54:45,480 right, just like it's easy for us to oh, I won't get 763 00:54:45,519 --> 00:54:51,559 into the religious, but it's easy for us to answerpomorphize everything that is a 764 00:54:51,599 --> 00:54:55,239 force that we don't understand. Humans prefer to most of the time. My 765 00:54:55,360 --> 00:54:59,920 concerned is we're supposed to be the professionals and the regular mortals are being confused. 766 00:55:00,159 --> 00:55:06,599 It's a freakular more they ask you those questions the muggles do. The 767 00:55:06,679 --> 00:55:09,000 muggles come up to you and ask questions, Richard. That's a good bog 768 00:55:09,519 --> 00:55:16,800 like qootball. Yeah, in a wizard. Well, we're the wizards. 769 00:55:17,239 --> 00:55:22,599 We're having a lot of fun and it's useful for us to be responsible absolutely 770 00:55:23,079 --> 00:55:28,159 healthy and health these these ordinary people who are freaked out because they've watched too 771 00:55:28,159 --> 00:55:30,800 many movies. Yeah, and we're playing with the same thing, which we 772 00:55:30,880 --> 00:55:35,360 know we're not. We're not, but they don't know. Yeah. And 773 00:55:35,440 --> 00:55:37,599 on that note, Uh, Michelle, where are you going to be next? 774 00:55:37,679 --> 00:55:42,039 Or what's in your inbox? Oh? So many things coming up, 775 00:55:42,079 --> 00:55:45,760 but it's really exciting. We have kind of like our GitHub broad show happening 776 00:55:45,760 --> 00:55:51,360 at the moment. So I GitHub Galaxy is really exciting, a bit more 777 00:55:51,480 --> 00:55:54,599 geared towards enterprise. So if you go on to if you just google GitHub 778 00:55:54,639 --> 00:56:01,039 galaxy, but it's galaxy dot GitHub dot com. And there's some really exciting 779 00:56:01,119 --> 00:56:07,199 stuff that we're doing all around the world. We've got galaxies happening all over 780 00:56:07,280 --> 00:56:10,000 Apak, all of the Americas, and all over Europe. So I'll be 781 00:56:10,119 --> 00:56:15,119 doing the Melbourne one, which is happening in May. I'll be in the 782 00:56:15,159 --> 00:56:20,760 Hong Kong one which is in April, and I'll be Interestingapore and Thailand ones 783 00:56:20,880 --> 00:56:24,119 that are also happening towards the end of May. Then there's a couple of 784 00:56:24,239 --> 00:56:30,480 NDC's coming up soon which is really exciting. We'll be there Serverallest Days. 785 00:56:30,559 --> 00:56:37,239 I'm doing Serverallest Days in Sydney and Auckland for those people over down on note 786 00:56:37,280 --> 00:56:42,440 with me, so you come see me there, and I think they're kind 787 00:56:42,440 --> 00:56:45,599 of the main ones that are coming up. And then I'll be at NDC 788 00:56:45,760 --> 00:56:49,079 also again, which wuld be really all right. We'll be there, yeah, 789 00:56:49,199 --> 00:56:53,440 I'll be looking. Yeah. Well, Michelle, thank you. This 790 00:56:53,480 --> 00:56:57,400 has been a great conversation. And I know we didn't stick to get help 791 00:56:57,519 --> 00:57:00,039 co pilot, but I think our listeners like that about this show. We 792 00:57:00,239 --> 00:57:06,199 tangentize, don't we Yes, that's not a word. Okay, sometimes we 793 00:57:06,280 --> 00:57:10,199 go places tangent urize. This is all right, Michelle, thanks a lot, 794 00:57:10,280 --> 00:57:14,440 and for you, dear listener. We'll see you next time on dot 795 00:57:14,519 --> 00:57:38,360 net rocks. Dot net Rocks is brought to you by Franklin's Net and produced 796 00:57:38,360 --> 00:57:45,000 by Pop Studios, a full service audio, video and post production facility located 797 00:57:45,000 --> 00:57:49,639 physically in New London, Connecticut, and of course in the cloud online at 798 00:57:49,679 --> 00:57:53,480 pwop dot com. Visit our website at d O t N E t R 799 00:57:53,519 --> 00:57:59,320 O c k S dot com for RSS feeds, downloads, mobile apps, 800 00:57:59,480 --> 00:58:02,440 comments, and access to the full archives going back to show number one, 801 00:58:04,000 --> 00:58:07,800 recorded in September two thousand and two. And make sure you check out our 802 00:58:07,840 --> 00:58:12,079 sponsors. They keep us in business. Now go write some code, See 803 00:58:12,119 --> 00:58:24,639 you next time. You got Jack middle Vans