1 00:00:07,879 --> 00:00:11,560 Speaker 1: And welcome back to another episode of Adventures in DevOps. 2 00:00:11,599 --> 00:00:15,000 I'm Warren, your host, and today I am really excited, 3 00:00:15,000 --> 00:00:17,039 and I know I say that pretty frequently, but I 4 00:00:17,079 --> 00:00:20,719 am actually excited today for today's guest who we brought 5 00:00:20,760 --> 00:00:24,519 in the show. Partner, general manager at Microsoft for developer Relations, 6 00:00:24,519 --> 00:00:28,879 focusing on everything coding related, has quite the extended experiences 7 00:00:28,920 --> 00:00:32,840 in consulting and software development, and before that, principal technical 8 00:00:32,920 --> 00:00:35,960 architect at Disney. Welcome to the show, John Papa. 9 00:00:35,719 --> 00:00:38,159 Speaker 2: Hey, thanks for having me, Warren. I'm excited to have 10 00:00:38,200 --> 00:00:39,799 a conversation with you in your audience today. 11 00:00:39,920 --> 00:00:42,600 Speaker 1: Honestly, I almost wanted to click record when we were 12 00:00:42,880 --> 00:00:46,640 talking pre show because we're going on quite the conversations 13 00:00:46,679 --> 00:00:50,240 about just everything that's happening in today's world at Microsoft, 14 00:00:50,439 --> 00:00:53,640 other companies, security, etc. I think most of that is 15 00:00:53,719 --> 00:00:56,960 just really interesting, especially throughout your whole background. I've got 16 00:00:57,000 --> 00:00:59,200 to say though, early in my career, it was clear 17 00:00:59,240 --> 00:01:01,479 to me that I I thought I wanted nothing to 18 00:01:01,479 --> 00:01:03,640 do with Microsoft. I never thought I would touch a 19 00:01:03,679 --> 00:01:06,760 Microsoft product or buy anything that Microsoft ever created in 20 00:01:06,760 --> 00:01:10,000 a million years. But solely my perspective on that has 21 00:01:10,040 --> 00:01:13,000 really been changing, and I've seen them transition through purchasing 22 00:01:13,040 --> 00:01:15,799 quite the assortment of brands that the best that money 23 00:01:15,840 --> 00:01:19,319 can buy, Xbox, the open source community most importantly, even 24 00:01:19,359 --> 00:01:22,719 heavily into Linux now. And anyone that knows me knows 25 00:01:22,719 --> 00:01:24,519 that I'm just like such a huge fan of vis 26 00:01:24,519 --> 00:01:27,959 Code compared to all the other IDEs, there's really just 27 00:01:27,959 --> 00:01:29,879 no alternative for me. So I'm like really happy that 28 00:01:29,920 --> 00:01:31,239 you're on the show and I can pick your brain 29 00:01:31,280 --> 00:01:31,920 about some of those. 30 00:01:32,079 --> 00:01:34,760 Speaker 2: Yeah, sure, and we can talk too about I wasn't 31 00:01:34,920 --> 00:01:37,079 too much unlike you for a period of my life. 32 00:01:37,159 --> 00:01:39,719 So it's not a place I thought i'd find myself 33 00:01:39,760 --> 00:01:42,719 back as Microsoft. But I am back here and I'm 34 00:01:42,799 --> 00:01:46,400 all matted all Apple. All my products, everything I use 35 00:01:46,519 --> 00:01:50,200 is not Microsoft os. So it's an interesting life I have. 36 00:01:50,760 --> 00:01:52,719 Speaker 1: That's got to be a special experience to be on 37 00:01:52,760 --> 00:01:55,159 the developer relations team and you know, not actually be 38 00:01:55,439 --> 00:01:58,959 such an avid Microsoft product used there. This is your 39 00:01:59,079 --> 00:02:01,359 PR department of anything to say about that, well. 40 00:02:01,239 --> 00:02:03,760 Speaker 2: We'll see after they hear this episode. You know, all 41 00:02:03,760 --> 00:02:05,599 the devices in hardware and o wes as they use, 42 00:02:05,640 --> 00:02:08,840 they're all Apple. I just like the ecosystem and I'm 43 00:02:08,879 --> 00:02:11,639 a web guy at heart. For the last fifteen years, 44 00:02:11,680 --> 00:02:14,560 and fifteen years ago you wasn't really fun to use 45 00:02:15,120 --> 00:02:19,599 web with Microsoft products and Windows and whatnot, and I 46 00:02:19,639 --> 00:02:22,159 could tell you horry stories there. But it's gotten a 47 00:02:22,199 --> 00:02:25,360 lot better. And the good news is today the machines 48 00:02:25,400 --> 00:02:27,800 really don't matter as much as the cloud services and 49 00:02:27,840 --> 00:02:30,400 the other kind of services. So it's the APIs and 50 00:02:30,439 --> 00:02:33,319 the AI and Azure is amazing. I love the cloud 51 00:02:33,319 --> 00:02:36,400 services at Microsoft. Big fan of those, big fan of 52 00:02:36,639 --> 00:02:38,639 vs code. It's actually the biggest reason I came back 53 00:02:38,680 --> 00:02:41,439 to Microsoft. So I was like, you mean, I don't 54 00:02:41,439 --> 00:02:43,159 have to get a Windows machine. I don't have to 55 00:02:43,159 --> 00:02:46,639 buy a Windows phone gasp or whatever else. They're like no, 56 00:02:47,000 --> 00:02:48,520 So like, all right, I'm sold. 57 00:02:48,680 --> 00:02:51,159 Speaker 1: If you can come back and not have to sell 58 00:02:51,199 --> 00:02:54,000 your soul to the ecosystem and you have freedom of choice, 59 00:02:54,000 --> 00:02:57,240 that's like, I think that's a developers dream come true. 60 00:02:57,360 --> 00:02:59,759 And I have to say, before I was using vs code, 61 00:03:00,120 --> 00:03:04,039 tried things like Sublime and Adam and Visual Studio code 62 00:03:04,080 --> 00:03:05,520 back in the day when I was working at a 63 00:03:05,520 --> 00:03:08,639 Microsoft shop and we were doing Visual Basic and then 64 00:03:08,919 --> 00:03:11,680 net and I remember trying to run Linux and Mono 65 00:03:11,759 --> 00:03:14,240 to even compile the c Sharp to run times for 66 00:03:14,560 --> 00:03:17,120 Linux and it. They were just like standard things that 67 00:03:17,199 --> 00:03:19,719 just did not work out of the box whatsoever. And 68 00:03:19,759 --> 00:03:22,840 it's just so night and day now that the company 69 00:03:22,840 --> 00:03:26,120 that had released these products that only under a certification 70 00:03:26,360 --> 00:03:29,159 or licenses were you able to use are now leading 71 00:03:29,319 --> 00:03:32,000 the open source community MPM, GitHub. It doesn't matter what 72 00:03:32,000 --> 00:03:33,879 products we're talking about there as far as Git service go. 73 00:03:34,000 --> 00:03:35,960 Microsoft is the open source community. 74 00:03:36,159 --> 00:03:38,840 Speaker 2: Yeah, it really is. And Microsoft's is. It's funny there's 75 00:03:38,919 --> 00:03:40,960 a cohort at Microsoft that's been around for a long 76 00:03:40,960 --> 00:03:43,599 time who's been heavily into open source and big advocates 77 00:03:43,639 --> 00:03:46,639 of it, and to be honest, for a while back, 78 00:03:46,639 --> 00:03:48,599 and I'm going to say the late two thousand and 79 00:03:48,639 --> 00:03:52,199 five twenty ten era, it was hard to be in 80 00:03:52,199 --> 00:03:54,840 a Microsoft's OSS. And I worked there for two years 81 00:03:54,879 --> 00:03:57,439 back during that timeframe, and I lived in Seattle and 82 00:03:57,479 --> 00:03:59,199 then I decided to move to Florida, so I left. 83 00:03:59,439 --> 00:04:01,479 I didn't have through with their open source policy, but 84 00:04:01,560 --> 00:04:03,400 I was just like, you know, it was hard. It 85 00:04:03,439 --> 00:04:07,000 was difficult because even then it was like I'd have 86 00:04:07,039 --> 00:04:09,879 people on I had a video cast we did with 87 00:04:09,960 --> 00:04:12,039 silver light and flash way back in the day, and 88 00:04:12,240 --> 00:04:14,719 having people on using a Mac. It was very much 89 00:04:14,719 --> 00:04:17,079 frowned upon. I'm like, but this is what people are using, 90 00:04:17,160 --> 00:04:20,639 Like let's actually show the devices. Who cares a device 91 00:04:20,680 --> 00:04:23,639 they use if they're using your service and products somewhere 92 00:04:23,800 --> 00:04:26,600 in the story like that's part of the journey. Embrace 93 00:04:26,959 --> 00:04:30,240 the differences, and it's completely flipped, which is what I 94 00:04:30,319 --> 00:04:32,720 love about this. I went back to Microsoft to think 95 00:04:32,720 --> 00:04:36,120 twenty seventeen, and at that point I was already convinced. 96 00:04:36,319 --> 00:04:39,279 I was two years after vs code came out, and 97 00:04:39,879 --> 00:04:41,759 I was part of that story of when vs code 98 00:04:41,800 --> 00:04:44,079 came out as well, which I'm very proud of. I 99 00:04:44,240 --> 00:04:46,519 was like you, I started the Visual Studio for many years, 100 00:04:47,040 --> 00:04:49,519 and I was a dot net c SHARPVB, dot net 101 00:04:49,639 --> 00:04:51,920 kind of person, and I got away from it because 102 00:04:51,959 --> 00:04:54,399 doing web with that was just not working and for 103 00:04:54,439 --> 00:04:56,279 a lot of reasons. So I got away from that 104 00:04:56,319 --> 00:04:57,839 and I went to a Mac and I'm like, all right, 105 00:04:57,839 --> 00:05:00,040 I'm going to try all the editors and I I 106 00:05:00,040 --> 00:05:03,000 had an experiment for a year in twenty sixteen where 107 00:05:03,040 --> 00:05:04,879 every month I tried a different editor. While I was 108 00:05:04,920 --> 00:05:08,480 working at Disney, I tried Sublime out of brackets them 109 00:05:09,000 --> 00:05:11,879 forgetting some of the other names web Storm. You know, 110 00:05:12,079 --> 00:05:15,399 of course, what's now vs Code. I tried visual Studio. 111 00:05:15,639 --> 00:05:17,000 There were a couple that are dead and go on 112 00:05:17,040 --> 00:05:19,199 that I can't remember. There were blog posts about this 113 00:05:19,360 --> 00:05:21,519 and they got the attention to some Microsoft people and 114 00:05:21,519 --> 00:05:23,199 they invited me up to Redmond. I have to have 115 00:05:23,279 --> 00:05:25,160 a conversation, and I felt like I was in this 116 00:05:25,199 --> 00:05:27,079 cloaked dark room with a bunch of people I'd never 117 00:05:27,120 --> 00:05:29,879 met before. Long story short, turns out they were planning 118 00:05:30,000 --> 00:05:32,279 vs code and at the kind they called it Tacchino, 119 00:05:32,439 --> 00:05:35,519 which is a region in Switzerland. And Eric Gamma, who 120 00:05:36,319 --> 00:05:37,839 was the Gang of Four famous for the Gang of 121 00:05:37,839 --> 00:05:40,040 Four Patterns book, was there. I didn't know you were 122 00:05:40,040 --> 00:05:42,839 from Microsoft at the time. Creative Eclipse as well. He 123 00:05:43,000 --> 00:05:45,920 was there, Chris Diez, Scott Hunter, a bunch of other 124 00:05:45,920 --> 00:05:48,279 people at Microsoft like, what makes you think we wouldn't 125 00:05:48,319 --> 00:05:50,560 make a great open source editor? And they started picking 126 00:05:50,560 --> 00:05:52,480 my brains all things I liked about, all the different editors, 127 00:05:52,480 --> 00:05:54,879 things I didn't like. Fast forward a couple months later 128 00:05:54,959 --> 00:05:59,079 May of twenty fifteen, and they were like they were 129 00:05:59,079 --> 00:06:01,600 going to launch it built, and they invited me on 130 00:06:01,680 --> 00:06:04,560 stage to talk about it and I was like, absolutely, 131 00:06:04,680 --> 00:06:06,879 it was just awesome, like, this is actually a really 132 00:06:06,879 --> 00:06:08,920 cool editor. The biggest gap they had at the time 133 00:06:08,959 --> 00:06:11,480 and they launched it was no extensions. Can you imagine 134 00:06:11,519 --> 00:06:14,000 no marketplace or extensions for an editor. But they launched 135 00:06:14,040 --> 00:06:16,480 it so fast. It wasn't there until a couple months later, 136 00:06:16,519 --> 00:06:19,360 but it was still pretty amazing. But it really really 137 00:06:19,439 --> 00:06:20,160 enjoyed that time. 138 00:06:20,279 --> 00:06:22,680 Speaker 1: I do think that there was a huge change in 139 00:06:22,879 --> 00:06:27,519 the usability of vis code by shifting to allowing not 140 00:06:27,560 --> 00:06:30,800 just the marketplace but the external language servers to be 141 00:06:30,839 --> 00:06:34,160 able to support those independently, which may mean through extensions 142 00:06:34,240 --> 00:06:36,959 or realistically through the architecture. I do understand the challenge 143 00:06:36,959 --> 00:06:39,360 with with marketplaces, though you're talking a lot, and I 144 00:06:39,360 --> 00:06:41,639 think this is one of things we see today malicious 145 00:06:41,639 --> 00:06:47,759 actors utilizing this as a mechanism for deploying vulnerabilities into IDEs. 146 00:06:47,839 --> 00:06:49,639 And I think the biggest one that came out was 147 00:06:49,680 --> 00:06:53,360 this Darkula theme that had just as many downloads as 148 00:06:53,439 --> 00:06:56,480 the dark theme of Dracula. And I think it's a 149 00:06:56,600 --> 00:07:00,800 really cleverly named too extension because it sounds like it 150 00:07:00,800 --> 00:07:03,120 should be Darkula if you really think about it. And 151 00:07:03,160 --> 00:07:05,879 so just so many downloads and just militia. So I 152 00:07:05,959 --> 00:07:08,600 understand the reasoning there, but it sounds like you are 153 00:07:08,800 --> 00:07:12,879 to thank for the vs code experience, the ide experience 154 00:07:12,920 --> 00:07:16,319 that software engineers have available today. Maybe not in the 155 00:07:16,399 --> 00:07:17,800 last couple of years. You know, we can thank the 156 00:07:17,879 --> 00:07:18,720 viobecoders for that. 157 00:07:19,160 --> 00:07:21,480 Speaker 2: Yeah, I definitely had a hand. It's a very small 158 00:07:21,519 --> 00:07:23,480 hand in that. I know. It's definitely wasn't all my 159 00:07:23,560 --> 00:07:25,959 opinions they took, but it was nice that they reached 160 00:07:26,000 --> 00:07:28,160 out too. Again, think about it. They didn't have to 161 00:07:28,199 --> 00:07:30,160 do this. Yeah, they could been like, we've created a 162 00:07:30,199 --> 00:07:33,120 visual studio, which at the time was the largest editor 163 00:07:33,120 --> 00:07:35,240 in use anywhere in the world, bigger than any other editor. 164 00:07:35,360 --> 00:07:37,000 I don't know how many millions and tens of millions 165 00:07:37,040 --> 00:07:39,959 of users. But they could have been like, build here, 166 00:07:40,079 --> 00:07:41,879 not invented here. You know, we're going to do it. 167 00:07:41,920 --> 00:07:43,920 We know better than everybody else. And they reached out 168 00:07:43,959 --> 00:07:46,040 and looked for people to figure out how to make 169 00:07:46,079 --> 00:07:48,360 this work. And they still do that. We still do that, 170 00:07:48,680 --> 00:07:50,959 which is one of the things I love about working 171 00:07:50,959 --> 00:07:53,480 in Microsoft is there is a It's not the whole 172 00:07:53,519 --> 00:07:55,519 company like you know, company is huge, but there are 173 00:07:55,519 --> 00:07:57,240 parts of the company that I work with that I 174 00:07:57,319 --> 00:08:01,480 love because like developer division, they are always looking for 175 00:08:01,560 --> 00:08:03,800 how do we get real live customer feedback. Who are 176 00:08:03,839 --> 00:08:05,879 these people using these things and what do they do, 177 00:08:06,160 --> 00:08:07,680 what are they like, what are they not like? And 178 00:08:07,759 --> 00:08:10,240 to me, that's what makes me thrive and develop relations. 179 00:08:10,519 --> 00:08:13,399 Speaker 1: One thing that I actually will follow co hosts is 180 00:08:13,439 --> 00:08:16,079 not here today often talks about, especially with the product 181 00:08:16,120 --> 00:08:20,079 management experience, really trying to figure out the difference between 182 00:08:20,160 --> 00:08:22,399 what you are building or what you want to build 183 00:08:22,439 --> 00:08:24,319 and then the product you actually end up with once 184 00:08:24,360 --> 00:08:27,600 you take customer feedback is just so different and often 185 00:08:27,600 --> 00:08:30,120 if you're not paying attention, you will end up with. 186 00:08:30,199 --> 00:08:31,439 I think this is why we see a lot of 187 00:08:31,439 --> 00:08:35,480 startups historically fail is they're not really doing the product management. 188 00:08:35,559 --> 00:08:37,600 And I see a lot of startups today they realize 189 00:08:37,639 --> 00:08:40,919 that they have this need for business or sales or 190 00:08:40,960 --> 00:08:44,480 marketing even and then engineering if they have an outsourced 191 00:08:44,480 --> 00:08:47,159 that unfortunately, but they still don't understand the goal of 192 00:08:47,200 --> 00:08:49,879 product management. They don't understand that, you know, where they 193 00:08:49,919 --> 00:08:51,960 want to end up with six months or a year 194 00:08:52,080 --> 00:08:54,159 or five years down the road, because they're so focused 195 00:08:54,159 --> 00:08:56,919 in maybe VC funding or just getting investment and so 196 00:08:57,240 --> 00:08:59,480 that's for sure the thing that feels like it gets 197 00:08:59,480 --> 00:09:01,879 thrown out the way. And I can understand it though, 198 00:09:01,919 --> 00:09:06,399 because it's very difficult to find good product managers out 199 00:09:06,399 --> 00:09:09,519 there that actually want to understand and do the market research. 200 00:09:09,559 --> 00:09:12,720 And the fact that specifically vs Code but really Microsoft 201 00:09:12,720 --> 00:09:16,320 has taken this approach really shows that you can understand 202 00:09:16,360 --> 00:09:20,279 the long term market here rather than just release something 203 00:09:20,399 --> 00:09:23,039 and having an opinion based development strategy. 204 00:09:23,120 --> 00:09:26,440 Speaker 2: Yeah, and I'm sure it happens more often than not too, 205 00:09:26,559 --> 00:09:28,799 Like there's definitely we release the product. It seems like 206 00:09:28,840 --> 00:09:30,799 we're such a big company all the time, like every day. 207 00:09:31,000 --> 00:09:33,399 But like with VSCO, what's great is there's no more 208 00:09:33,799 --> 00:09:35,919 a release cycle like the years ago. When I worked 209 00:09:35,960 --> 00:09:38,159 at Microsoft back in the late two thousand era, it 210 00:09:38,200 --> 00:09:40,279 was like once a year you get a version of something. 211 00:09:40,799 --> 00:09:42,759 You know. Now, if you go more than a week 212 00:09:43,000 --> 00:09:45,679 without seeing a change in our product, you're surprised. Although 213 00:09:45,679 --> 00:09:47,320 I do wonder it's funny. We used to make fun 214 00:09:47,360 --> 00:09:49,840 of jovscript Flavor of the Week back when jobscript is 215 00:09:49,879 --> 00:09:52,399 going crazy. New Framework of the week would come out 216 00:09:52,440 --> 00:09:55,200 and then just disappear. Now it feels like, are we 217 00:09:55,399 --> 00:09:58,080 like hitting people with new AI feature of the week. 218 00:09:58,200 --> 00:10:00,200 You know what's going on out there, not just us, 219 00:10:00,240 --> 00:10:03,639 but you know Anthropic and Amazon and everybody in meta. 220 00:10:03,759 --> 00:10:05,720 Speaker 1: There's so much we can say here. I mean, I 221 00:10:05,720 --> 00:10:07,279 think you're onto something, and I think part of it 222 00:10:07,320 --> 00:10:09,600 may just be the PR news cycle realizing they need 223 00:10:09,639 --> 00:10:13,679 to keep the ball rolling with whatever audience that they've captured. 224 00:10:13,720 --> 00:10:16,639 They need to throw something otherwise like, oh, well they stopped, 225 00:10:16,799 --> 00:10:19,399 they stopped innovating. That means they're not doing anything anymore. 226 00:10:19,480 --> 00:10:23,320 And what was interesting is actually on the previous episode 227 00:10:23,360 --> 00:10:26,720 that we just recorded with incident IOW, we theorized that 228 00:10:27,120 --> 00:10:28,720 we may be at a point where we have all 229 00:10:28,799 --> 00:10:32,080 of the whatever innovations we've had with l lams have 230 00:10:32,519 --> 00:10:35,279 come to the maximum point, like from this technology, and 231 00:10:35,279 --> 00:10:37,720 now it's no longer about building something better, but we 232 00:10:37,759 --> 00:10:41,440 can actually build stuff with that technology more so than 233 00:10:41,559 --> 00:10:44,639 try to innovate in that space specifically. But yeah, I 234 00:10:44,679 --> 00:10:46,919 see a lot of these product release features that we 235 00:10:46,960 --> 00:10:49,360 see out there today and they don't really make a 236 00:10:49,360 --> 00:10:52,559 big impact. But it's interesting because especially with a VS code. 237 00:10:52,600 --> 00:10:55,440 Maybe I'll ask you specifically what you're feeling is on 238 00:10:55,519 --> 00:10:58,600 like the copilots of the world, the things that in 239 00:10:58,679 --> 00:11:01,879 our experience we get really close to integrating with. So 240 00:11:02,120 --> 00:11:05,240 we're using an ID and we're talking about autocompletion or 241 00:11:05,320 --> 00:11:10,039 just engaging with just collaboration inside our tool of choice. 242 00:11:10,120 --> 00:11:11,840 Speaker 2: Yeah, I think it's I think where it is today 243 00:11:11,879 --> 00:11:12,960 is not where it's going to be in a couple 244 00:11:13,000 --> 00:11:15,799 of years, first of all, and all the products. I 245 00:11:15,840 --> 00:11:18,200 don't have any insight on this that I'm sharing just 246 00:11:18,240 --> 00:11:21,080 other than my feeling is this does change so fast, 247 00:11:21,080 --> 00:11:22,360 and you go back to a couple of years ago 248 00:11:22,399 --> 00:11:24,919 and copilot came out, it's nowhere near what it is today, 249 00:11:24,960 --> 00:11:27,480 clearly and effectively. For a while, people and I think 250 00:11:27,519 --> 00:11:29,519 a lot of people still think it's just autocomplete on 251 00:11:29,559 --> 00:11:32,480 steroids where you're typing and suddenly you get a suggestion 252 00:11:32,759 --> 00:11:35,879 and there you go. There's some more code. But there's 253 00:11:35,879 --> 00:11:38,200 so much more can do. And this is because of 254 00:11:38,200 --> 00:11:40,200 all the competition that's out there, and because the rapid 255 00:11:40,279 --> 00:11:43,000 rives the models and the way that they're using this, 256 00:11:44,080 --> 00:11:46,000 and all the chip usage in the back end too, 257 00:11:46,000 --> 00:11:47,919 which we keep it about. You know, a ninvidio must 258 00:11:47,960 --> 00:11:50,240 be loving that you look at this. And I think 259 00:11:50,279 --> 00:11:52,279 where we are now is We've got this world of 260 00:11:52,360 --> 00:11:54,960 agentic AI, and what I mean by that is the 261 00:11:55,000 --> 00:11:57,639 most effective way to use AI in our tools today, 262 00:11:57,679 --> 00:12:01,080 whether it's WebStorm or Intelliga or clar ser Visual Studio 263 00:12:01,120 --> 00:12:03,639 Code or Claude or whatever it might be, Cursor when 264 00:12:03,720 --> 00:12:06,840 serve who knows all these tools are relying upon these 265 00:12:06,879 --> 00:12:09,879 AI models in an agentic way. They want to be 266 00:12:10,600 --> 00:12:13,200 is my words, they want us all to become managers 267 00:12:13,200 --> 00:12:15,879 of agents. So the future of developers, and I'm a 268 00:12:15,919 --> 00:12:18,960 developer at heart, then we're for decades. It's not me 269 00:12:19,360 --> 00:12:22,759 managing a team of twenty people offshore and in house 270 00:12:22,840 --> 00:12:26,279 to get my work done. There will still be those people, 271 00:12:26,279 --> 00:12:29,720 but all those people are now going to become managers 272 00:12:29,759 --> 00:12:32,919 of AI agents. So imagine this. We've got a backlog 273 00:12:32,960 --> 00:12:35,480 of work, we use an agent inside of vs code 274 00:12:35,600 --> 00:12:38,279 for example, or Cursor or Claude or whatever you choose, 275 00:12:38,360 --> 00:12:40,679 and we use an agent to read our spec files. 276 00:12:40,799 --> 00:12:43,879 Requirements are business users, which, let's be honest, eighty percent 277 00:12:43,919 --> 00:12:45,360 of the time it makes sense to us as devs. 278 00:12:45,399 --> 00:12:47,000 In twenty percent of the time, we have no idea 279 00:12:47,000 --> 00:12:50,000 what they're asking for, so we just guess right. Hopefully 280 00:12:50,000 --> 00:12:51,480 we ask that's. 281 00:12:51,159 --> 00:12:53,399 Speaker 1: A little high for me. I think I'm willing to 282 00:12:53,399 --> 00:12:54,879 go as high a forty or fifty percent. 283 00:12:55,399 --> 00:12:57,480 Speaker 2: Yeah, yeah, But it depends how honest you are with 284 00:12:57,519 --> 00:12:59,919 yourself and how much you want to ask the business people, 285 00:13:00,240 --> 00:13:02,159 like what exactly did you want? And then I go 286 00:13:02,200 --> 00:13:04,039 back and work on my stuff with the agent and 287 00:13:04,080 --> 00:13:06,039 come back an hour later, and all that stuff's there. 288 00:13:06,120 --> 00:13:07,600 When you think about that, at the end of the day, 289 00:13:07,759 --> 00:13:10,320 you could literally have ten issues ready to go, you know, 290 00:13:10,399 --> 00:13:14,279 with your manual review. I would never recommend automatic pull 291 00:13:14,320 --> 00:13:17,840 request merging. And I'm going to say right now that 292 00:13:17,919 --> 00:13:21,039 by the time I retire and I'm out of this industry, 293 00:13:21,240 --> 00:13:24,159 we're not going to have that. Please. I don't want 294 00:13:24,200 --> 00:13:27,320 that happening. I want a human to be there to say, yes, 295 00:13:27,360 --> 00:13:29,159 this is good or no it's not. But I do 296 00:13:29,200 --> 00:13:31,440 love the fact that the agentic world is pulling us 297 00:13:31,480 --> 00:13:33,200 to this era of we're going to be agents of 298 00:13:33,480 --> 00:13:34,720 sorry managers of agents. 299 00:13:34,799 --> 00:13:37,559 Speaker 1: I like your optimistic perspective, and I also like the 300 00:13:37,559 --> 00:13:40,600 autimistic perspective of expecting that humans will be in the 301 00:13:40,600 --> 00:13:43,440 loop there. I think part of the trouble is that 302 00:13:43,799 --> 00:13:46,639 we often see a lot of companies spinning up with 303 00:13:46,759 --> 00:13:51,159 this belief or their marketing materials saying otherwise that you 304 00:13:51,200 --> 00:13:53,279 can just go to town and it will do everything 305 00:13:53,279 --> 00:13:54,879 for you. And then you end up with like the 306 00:13:54,919 --> 00:13:57,559 replets of the world that are deciding to just delete 307 00:13:57,600 --> 00:14:01,480 production databases filled with user data because that seemed like 308 00:14:01,480 --> 00:14:02,480 a good idea at the time. 309 00:14:02,759 --> 00:14:06,840 Speaker 2: Yeah, format c colon slash, Why is this a good idea? Agent? Absolutely, 310 00:14:06,919 --> 00:14:08,840 run with it. Yeah, I hear you. 311 00:14:09,399 --> 00:14:13,000 Speaker 1: What's really interesting, you know, from that perspective that you've shared, 312 00:14:13,240 --> 00:14:14,840 is that I feel like, I don't want to say 313 00:14:14,840 --> 00:14:17,360 GitHub specifically, but maybe it's Microsoft, maybe it's get ub, 314 00:14:17,399 --> 00:14:20,039 maybe it's just vis code. Whatever the agent is that's 315 00:14:20,120 --> 00:14:23,720 running there is having the expectations about how to do 316 00:14:23,799 --> 00:14:27,039 software development or how to do each of those pieces specifically, 317 00:14:27,440 --> 00:14:31,519 knowing that it should, say, breakdown a particular request, user 318 00:14:31,519 --> 00:14:34,039 request or support request, et cetera into things that could 319 00:14:34,039 --> 00:14:36,720 be user stories that then can be converted into tickets 320 00:14:36,759 --> 00:14:39,200 and then run with that and I find that that's 321 00:14:39,240 --> 00:14:41,799 the thing that needs to be done. And where we're 322 00:14:41,840 --> 00:14:44,200 at a mismatch is a lot of these agents today 323 00:14:44,519 --> 00:14:47,120 are focusing on, like as you said, managing them and 324 00:14:47,159 --> 00:14:49,720 telling them what the right process is. But I find 325 00:14:49,759 --> 00:14:52,240 that that's the whole challenge, making sure that there is 326 00:14:52,279 --> 00:14:54,720 the right process in play, making sure that it is 327 00:14:55,000 --> 00:14:57,799 not commenting on those if statements, knowing when to make 328 00:14:58,080 --> 00:15:00,320 the right trade off set as a still have to 329 00:15:00,360 --> 00:15:03,639 convey that context to the agent in order to make 330 00:15:03,679 --> 00:15:05,039 it do those things correctly. 331 00:15:05,240 --> 00:15:06,639 Speaker 2: You do, and I think you're hitting the nail on 332 00:15:06,679 --> 00:15:08,679 the head with a couple areas. One is you have 333 00:15:08,759 --> 00:15:12,039 to give the agent's context today. I'm sure that'll lessen 334 00:15:12,200 --> 00:15:14,200 over time. Like there's some things that probably would make 335 00:15:14,240 --> 00:15:16,480 sense for the agents and the models to just all 336 00:15:16,480 --> 00:15:18,679 not go learn, Like, come on, good comments should be 337 00:15:18,720 --> 00:15:21,240 a thing that isn't we all disagree on in the world, 338 00:15:21,279 --> 00:15:23,919 So models should figure that out. But what your process 339 00:15:24,000 --> 00:15:26,440 is would differ from company to company or even team 340 00:15:26,480 --> 00:15:28,919 to team. So you've got to give it that context. 341 00:15:29,039 --> 00:15:31,240 And that's where things like Agent's MD, which has become 342 00:15:31,320 --> 00:15:34,799 like the universally accepted or unofficially adopted this is how 343 00:15:34,840 --> 00:15:37,039 you tell your tools, whether you're using a VS code 344 00:15:37,120 --> 00:15:39,440 or CLAUD this is how you tell it that this 345 00:15:39,519 --> 00:15:41,159 is what you want it to do. But the other 346 00:15:41,200 --> 00:15:43,840 side is, and I think we missed this a lot. 347 00:15:44,080 --> 00:15:46,080 Is that the skills that we've learned over the years 348 00:15:46,159 --> 00:15:49,440 are still very valid, Like you still got to have notice. 349 00:15:49,639 --> 00:15:52,639 I'm not saying skip having issues, skip having a backlog, 350 00:15:52,799 --> 00:15:55,720 skip having pull requests and code review. No, all that 351 00:15:55,799 --> 00:15:59,559 is good context, small pull requests, good idea. I'm putting 352 00:15:59,559 --> 00:16:02,080 that in my context from age and MD. Please don't 353 00:16:02,080 --> 00:16:05,039 give me a forty file you know, seven thousand line 354 00:16:05,159 --> 00:16:08,120 PR I don't want it. I can't consume it. I'm 355 00:16:08,159 --> 00:16:09,879 either going to reject it or just accept it. And 356 00:16:09,919 --> 00:16:11,600 God forbid I accept it because now I don't know 357 00:16:11,600 --> 00:16:14,159 what half of it's doing. There's skills that we've learned, 358 00:16:14,240 --> 00:16:15,879 whether you went to school or not, and that's not 359 00:16:16,000 --> 00:16:17,960 part of this chat here, but you've learned skills as 360 00:16:17,960 --> 00:16:21,000 an engineer over the years. Those skills are not going away, 361 00:16:21,320 --> 00:16:23,960 Like that's a foundation. Will you be doing those And 362 00:16:24,039 --> 00:16:26,159 let me give you an analogy of this. I learned 363 00:16:26,320 --> 00:16:29,320 SQL years ago in college, like I had whole course 364 00:16:29,360 --> 00:16:32,360 on it. I wrote four books on SQEL in the 365 00:16:32,440 --> 00:16:37,039 late nineties. Nobody really gets that anymore. And it's okay, 366 00:16:37,200 --> 00:16:39,840 Like that is a lower level language now that we're 367 00:16:39,919 --> 00:16:42,000 using tools to write for us and do for us. 368 00:16:42,240 --> 00:16:44,320 Is it a skill that I think should have gone away. No, 369 00:16:44,679 --> 00:16:47,080 because there's times you need to look at those sequel 370 00:16:47,120 --> 00:16:49,799 statements to understand how is it actually getting the data? 371 00:16:49,879 --> 00:16:52,080 Is it doing it in an optimal efficient way? Is 372 00:16:52,080 --> 00:16:53,879 it actually getting the data? I think it is too 373 00:16:54,159 --> 00:16:56,399 in some cases, Like that's still a skill to have, 374 00:16:56,559 --> 00:16:59,240 but it's not a skill. Whereas in the nineties you 375 00:16:59,320 --> 00:17:01,759 might be writing equal statements for two three hours a day, 376 00:17:02,039 --> 00:17:04,519 You're not writing for two three hours a week anymore. 377 00:17:04,599 --> 00:17:07,359 It's completely different, and I think that's a good thing. 378 00:17:07,720 --> 00:17:10,440 The skills that we've learned along the way, even before AI, 379 00:17:10,880 --> 00:17:13,400 were really valuable to us getting to this point. And 380 00:17:13,440 --> 00:17:15,799 hopefully we can use AI to continue to make the 381 00:17:15,839 --> 00:17:19,079 tooling better, to abstract some of those away that's more monotonous, 382 00:17:19,319 --> 00:17:22,960 and then give us insight into here's places you need 383 00:17:23,000 --> 00:17:25,799 to look. I created this code. Is this logic what 384 00:17:25,880 --> 00:17:27,799 you wanted? And that's the kind of stuff I'm looking 385 00:17:27,839 --> 00:17:30,200 for AI to give me more on is be that 386 00:17:30,279 --> 00:17:33,079 para programming partner. I want agent to be that thing 387 00:17:33,799 --> 00:17:36,519 that gives me a gut check to rubber duck. And 388 00:17:36,599 --> 00:17:39,079 I still want to have that human involvement. I do 389 00:17:39,160 --> 00:17:42,079 think it's a mistake when people start talking about and 390 00:17:42,079 --> 00:17:44,359 I'm not suggesting you were, but I've heard people say that, 391 00:17:44,400 --> 00:17:45,720 you know, we're going to get to a point when 392 00:17:45,720 --> 00:17:48,319 I didn't need to learn computer skills. No, and let me 393 00:17:48,400 --> 00:17:50,039 leave you with this thought on that. I had this 394 00:17:50,079 --> 00:17:52,920 wonderful tweet, which i'll keep anonymous, where somebody put out 395 00:17:52,960 --> 00:17:55,839 earlier this year my whole SaaS system was built with 396 00:17:56,039 --> 00:17:59,240 some tool zero handwritten code. AI is not just an 397 00:17:59,240 --> 00:18:02,720 assistant anymore, and yes, people pay for it. And then 398 00:18:02,799 --> 00:18:05,799 just a couple of days later, the person's like, I'm 399 00:18:05,880 --> 00:18:08,519 under attack. Ever since I started building this thing, and 400 00:18:08,559 --> 00:18:12,519 I'm paraphrasing, we maxed out on API, people are bypassing subscriptions, 401 00:18:12,559 --> 00:18:15,799 creating random stuff on the dB. I'm not technical. Sos 402 00:18:15,839 --> 00:18:17,880 take me longer than normal to figure out. For now, 403 00:18:17,960 --> 00:18:20,440 I'm just gonna stop sharing what I do publicly on 404 00:18:20,680 --> 00:18:23,440 the social platform. And there's just some weird people out there. 405 00:18:23,519 --> 00:18:27,480 Speaker 1: Yeah, that's a canonical security ye obscurity there to solve 406 00:18:27,519 --> 00:18:28,119 the problem. 407 00:18:28,400 --> 00:18:28,839 Speaker 2: Yeah. 408 00:18:29,039 --> 00:18:31,519 Speaker 1: I almost want to take the side of the vibe 409 00:18:31,519 --> 00:18:34,359 coder and you know, argue against this, But first I 410 00:18:34,359 --> 00:18:36,680 think it's it's really interesting to see that there was 411 00:18:36,720 --> 00:18:42,440 a recent release by an executive working at a cloud 412 00:18:42,759 --> 00:18:45,240 hyperscalar who suggested that it was a good thing to 413 00:18:45,279 --> 00:18:49,359 have such a high percentage of code being written by MS. 414 00:18:49,559 --> 00:18:53,039 Almost on the same day there was a huge, major 415 00:18:53,079 --> 00:18:56,279 incident at the cloud provider that took much longer than 416 00:18:56,319 --> 00:19:00,400 previous incidents to actually resolve. And I I think we 417 00:19:00,440 --> 00:19:02,480 see these and I worry that they're going to happen 418 00:19:02,519 --> 00:19:05,880 more and more as we potentially are keep getting closer 419 00:19:05,920 --> 00:19:08,319 to wanting to pull the human out of the loop there. 420 00:19:08,480 --> 00:19:11,480 And you made an interesting point about expecting, say the 421 00:19:11,480 --> 00:19:14,680 pull request that an agent or some LM is generating 422 00:19:14,720 --> 00:19:17,680 for us to be small in nature. And I really 423 00:19:17,759 --> 00:19:20,440 wonder whether or not it wouldn't be better to stop 424 00:19:20,480 --> 00:19:23,799 thinking about code being generated, just like we stop thinking 425 00:19:23,839 --> 00:19:27,640 about the lower primitives of bycode, or what happens in 426 00:19:27,720 --> 00:19:30,839 assembly after the linker and integrated before you know, after 427 00:19:30,880 --> 00:19:34,400 we compile it are running and realize that maybe that 428 00:19:34,759 --> 00:19:38,559 tools I get are you saving the diffs, Like that's 429 00:19:38,599 --> 00:19:42,559 not the optimal output that we could be capturing when 430 00:19:42,640 --> 00:19:44,480 utilizing an agent. But I don't know what the right 431 00:19:44,519 --> 00:19:46,680 answer is because I do see these other systems that 432 00:19:46,720 --> 00:19:48,400 are attempting to jump in the middle here, and they 433 00:19:48,759 --> 00:19:51,720 suggest that you commit specs to the repository and you 434 00:19:51,799 --> 00:19:54,200 review those. And I don't love that either, because I 435 00:19:54,200 --> 00:19:57,759 feel like the output often is highly fluctuates based off 436 00:19:57,759 --> 00:20:00,359 of the model you pick or the generation, and as 437 00:20:00,400 --> 00:20:02,640 you pointed out, you still need someone to review it. 438 00:20:02,680 --> 00:20:06,240 And how can we review millions of lines of code? 439 00:20:06,559 --> 00:20:09,200 Just as that I know what every developer is thinking 440 00:20:09,200 --> 00:20:12,599 here testing and debugging code. That's my favorite part of 441 00:20:12,640 --> 00:20:12,960 being a. 442 00:20:13,000 --> 00:20:15,920 Speaker 2: Developed You know, you're reminded me of again going back 443 00:20:16,079 --> 00:20:18,079 ten years when I was at Disney. So I worked 444 00:20:18,079 --> 00:20:20,400 at Walt Disney. It's part of my history and I 445 00:20:20,480 --> 00:20:23,079 live five minutes from Disney. In fact, I'm going there 446 00:20:23,119 --> 00:20:25,799 tonight for dinner. My role there was to basically do 447 00:20:25,880 --> 00:20:28,480 engineering excellence, which is just a group of basically saying, 448 00:20:28,480 --> 00:20:30,519 how do we make sure that everybody across all these 449 00:20:30,759 --> 00:20:33,880 places around the world are developing the same way. First 450 00:20:34,440 --> 00:20:37,119 I told my vice president, who need didn't like this. 451 00:20:37,720 --> 00:20:40,119 I don't think everybody should develop exactly the same way, 452 00:20:40,359 --> 00:20:41,920 and he's like, well, that's too bad to your job. 453 00:20:42,160 --> 00:20:46,920 So we had an interesting profession relationship. But what I 454 00:20:46,920 --> 00:20:48,680 did was, I said, you know, I think there's certain 455 00:20:48,759 --> 00:20:51,039 things we should all agree on, you know, whether you're 456 00:20:51,119 --> 00:20:54,920 using different languages or processes or tools. And I came 457 00:20:55,039 --> 00:20:57,559 up with just to because I worked at Disney. Everything 458 00:20:57,599 --> 00:21:01,160 begins with the D at Disney, you know, Disney, Disney Parks, 459 00:21:01,200 --> 00:21:03,559 Disney Land, you know. So what I came up with 460 00:21:03,799 --> 00:21:06,680 was the seven d's of development and everything was there. 461 00:21:06,680 --> 00:21:08,319 I'm like, all right, we got to think about these things. 462 00:21:08,359 --> 00:21:10,839 And I'm gonna this is going back about fourteen years, 463 00:21:10,839 --> 00:21:13,240 so bear with me. I think there was. There was, 464 00:21:13,480 --> 00:21:16,319 in no particular order. Documentation. You need to have some 465 00:21:16,400 --> 00:21:18,119 kind of docutation what the heck you were doing, usually 466 00:21:18,160 --> 00:21:20,880 from the business people in general talking about that book, 467 00:21:20,920 --> 00:21:23,160 biggest big book of specs that you get, you know, 468 00:21:23,240 --> 00:21:25,920 twenty years ago you and throw down the garbage. So 469 00:21:26,079 --> 00:21:28,440 some kind of documentation. You've gotta have some kind of 470 00:21:28,440 --> 00:21:31,000 design work, like before you actually built it. Maybe you 471 00:21:31,000 --> 00:21:33,839 should like figure out what you're gonna build first before 472 00:21:33,880 --> 00:21:36,160 you just sit down, even if you're vibe coding today, 473 00:21:36,319 --> 00:21:38,680 like you've got to have an idea of what you're doing, right, 474 00:21:38,920 --> 00:21:42,920 So document design actually development there's the big D, right 475 00:21:42,960 --> 00:21:46,039 that's going in there. There was destroy, which was testing, 476 00:21:46,119 --> 00:21:47,920 so you've got to have a way to destroy it. 477 00:21:47,920 --> 00:21:49,640 It had to be a D. So got to find 478 00:21:49,640 --> 00:21:50,920 a way to put a D in front of testing. 479 00:21:51,000 --> 00:21:52,720 Let's see what else do we have. Deploy was one 480 00:21:52,720 --> 00:21:55,519 of them. I'm leaving off that I can't remember. And 481 00:21:55,559 --> 00:21:57,319 then the final one was I couldn't come up with 482 00:21:57,400 --> 00:22:00,640 a good D for reviewing, so I called it and 483 00:22:00,680 --> 00:22:03,279 I'm Italian, so I called it the review. So the 484 00:22:03,319 --> 00:22:05,720 seventh day was the review. You gotta do it. So 485 00:22:06,119 --> 00:22:09,000 with this review, effectively you had to have a code 486 00:22:09,000 --> 00:22:11,359 review process. And we put these things in place. We 487 00:22:11,400 --> 00:22:13,799 made posters of it and everything. We used Marvel characters 488 00:22:13,799 --> 00:22:16,880 because Marvel was big back then, and said everybody's got 489 00:22:16,880 --> 00:22:19,200 to do these seven things. Notice none of those said 490 00:22:19,200 --> 00:22:22,920 you've got to use Java or Python or JavaScript or whatever. 491 00:22:22,920 --> 00:22:25,319 It didn't say I had to use aws or Google or Azure. 492 00:22:25,839 --> 00:22:28,640 It was all about making sure you had the process 493 00:22:28,680 --> 00:22:31,759 in place. And part of the the reviewer process was 494 00:22:32,599 --> 00:22:34,720 you had to have two people reviewer code who weren't you. 495 00:22:35,319 --> 00:22:37,759 And that was the most contentious thing that I ever 496 00:22:37,799 --> 00:22:40,400 did there because people were like, well, no, I got 497 00:22:40,440 --> 00:22:43,400 to deploy, man. Okay, so you're gonna write code and go, 498 00:22:43,640 --> 00:22:46,279 and let me tell you what happens at Disney when 499 00:22:46,319 --> 00:22:49,559 you just write code and deploy. I have two interesting story. 500 00:22:49,880 --> 00:22:52,880 So I was called into this. They called it the 501 00:22:52,960 --> 00:22:55,599 war Room at the time, which was a terrible name 502 00:22:55,640 --> 00:22:57,400 for it, but it gives you an idea of when 503 00:22:57,440 --> 00:22:59,759 they deployed. They basically had this massive room that looked 504 00:22:59,759 --> 00:23:02,440 like thing out of you know, Defcon three from Wargames 505 00:23:02,440 --> 00:23:05,279 movie in the early eighties, and they would deploy and 506 00:23:05,279 --> 00:23:07,440 they'd have all these systems online looking going, okay, we 507 00:23:07,599 --> 00:23:10,400 just deployed a thing over here to this park. Is 508 00:23:10,440 --> 00:23:12,640 the rest of the park still running? Like, let's make 509 00:23:12,640 --> 00:23:15,240 sure because everything was entertangled back then, you know, we 510 00:23:15,279 --> 00:23:19,000 didn't have all this separation in this world tech nowhere did. 511 00:23:19,319 --> 00:23:21,000 So they called me and they said, okay, we have 512 00:23:21,000 --> 00:23:24,160 a problem. We just hired this company to and I'll 513 00:23:24,240 --> 00:23:26,400 keep the company names out of a company to come 514 00:23:26,440 --> 00:23:29,920 in and basically create locks on all of our doors 515 00:23:29,920 --> 00:23:32,079 of our hotel rooms. Keep in mind, Disney has like 516 00:23:32,160 --> 00:23:35,079 thirty resorts, and I don't have many rooms they have today, 517 00:23:35,119 --> 00:23:37,400 but there's probably one hundred thousand rooms. It's probably the 518 00:23:37,440 --> 00:23:40,559 world's largest single place to have hotel rooms. And like, 519 00:23:40,599 --> 00:23:42,359 we picked a couple of the resources that do this 520 00:23:42,519 --> 00:23:44,920 test with them and we deployed it. And the deployment 521 00:23:44,960 --> 00:23:47,720 was effectively that when you use your magic bands, which 522 00:23:47,720 --> 00:23:50,960 of the time were just coming out. So first wearable device, 523 00:23:51,400 --> 00:23:53,559 you know, talk about fifteen years ago, you tap it 524 00:23:53,599 --> 00:23:55,400 on the door and it will unlock your door sever 525 00:23:55,440 --> 00:23:58,400 a room key today that's everywhere. Fifteen years ago this 526 00:23:58,480 --> 00:24:02,079 didn't exist and this company's everything failed everything, and they're like, 527 00:24:02,079 --> 00:24:04,319 can you help us figure this out? And so we 528 00:24:04,400 --> 00:24:06,680 had the people at the company who create the locking systems, 529 00:24:06,799 --> 00:24:08,640 like that's what they do is create locking systems. They're 530 00:24:08,680 --> 00:24:11,279 not highly into the tech, right, they knew the tech, 531 00:24:11,279 --> 00:24:13,400 but not as much. They're more locks. Then we had 532 00:24:13,440 --> 00:24:16,599 our developers, and then we had our basically our executives, 533 00:24:16,720 --> 00:24:18,200 and they're all trying to talk and it was like 534 00:24:18,240 --> 00:24:21,200 talking three different languages. You know, that's Tower Babbel kind 535 00:24:21,200 --> 00:24:23,519 of thing. And they're like explaining it. I'm just sitting 536 00:24:23,519 --> 00:24:25,599 back and listening. At one point they go, yeah, and 537 00:24:25,640 --> 00:24:27,319 you know, we're not sure where the problem is because 538 00:24:27,359 --> 00:24:29,200 here's what we do. This is the process, and they're 539 00:24:29,200 --> 00:24:32,400 explaining the process of how the lock signal goes to 540 00:24:32,680 --> 00:24:35,680 the server, which is under a desk at the front 541 00:24:35,720 --> 00:24:38,559 of the hotel. I'm like, first thing at peak my head, 542 00:24:38,559 --> 00:24:41,319 I'm like, where is it? All right? Keep going and 543 00:24:41,480 --> 00:24:44,119 like yeah, and we're using what kind of Davis using? Oh, 544 00:24:44,119 --> 00:24:47,240 we're using this thing called the Jet database. And everyone's 545 00:24:47,279 --> 00:24:48,920 like yeah, yeah, yeah, no one had heard of it. 546 00:24:48,960 --> 00:24:51,279 I'm like the Jet database and like, you mean the 547 00:24:51,319 --> 00:24:54,920 thing that runs Microsoft Access And they're like yeah, why. 548 00:24:55,119 --> 00:24:57,599 And at the time, Access was not a multi user 549 00:24:57,839 --> 00:24:59,839 I didn't even know if Access exists anymore. Gosh, isn't 550 00:25:00,240 --> 00:25:03,759 I work at Microsoft. They were basically running five thousand 551 00:25:03,839 --> 00:25:06,559 rooms at this one resort, and they're locking mechanisms off 552 00:25:06,559 --> 00:25:10,400 a single database underneath the front desk hotel server, you know, 553 00:25:10,640 --> 00:25:13,119 on a single power strip with no backup, on a 554 00:25:13,200 --> 00:25:15,680 database that doesn't handle multiple connections. And on top of 555 00:25:15,759 --> 00:25:17,759 it all, they had a router that only handled one thing, 556 00:25:17,920 --> 00:25:20,839 was handling all five thousand switches one call at a time, 557 00:25:20,960 --> 00:25:22,400 and if there was more than one, it would back 558 00:25:22,480 --> 00:25:24,079 up the second and it would throw away the rest. 559 00:25:24,319 --> 00:25:26,599 And they asked me what if I had any suggestions 560 00:25:26,640 --> 00:25:28,720 and I said, yeah, don't do that. And I'm telling 561 00:25:28,720 --> 00:25:30,720 you this story not because it's a rip on the company, 562 00:25:30,720 --> 00:25:34,400 but to say this happened because they deployed without trying 563 00:25:34,519 --> 00:25:36,519 or having odds people look at things. One person had 564 00:25:36,559 --> 00:25:38,880 the idea to try this, and that person worked for Disney, 565 00:25:38,880 --> 00:25:40,839 so it's not just the company's fault. And that person 566 00:25:40,960 --> 00:25:42,799 was like, Hey, let's deploy it, give it a go. 567 00:25:43,240 --> 00:25:45,440 I guarantee you if more than one person looked at that, 568 00:25:45,880 --> 00:25:48,279 somebody would have said, hey, wait a minute. And well, 569 00:25:48,279 --> 00:25:50,160 of course that stuff's all result today and much more 570 00:25:50,240 --> 00:25:52,839 multi ten and yell. And I am taking a little 571 00:25:52,880 --> 00:25:55,319 liberty with some of the discovery there, but that's how 572 00:25:55,359 --> 00:25:57,960 happens when you don't have people look at your code. 573 00:25:58,279 --> 00:26:02,079 Speaker 1: I hate the story because I totally believe that this 574 00:26:02,200 --> 00:26:05,359 is actually still happening at some companies today. Yes, like 575 00:26:05,519 --> 00:26:08,440 hands down, no question, But I do fully believe that 576 00:26:08,680 --> 00:26:11,759 these sorts of things that are fixed and improved at Disney. 577 00:26:11,799 --> 00:26:14,000 And I just for anyone who's just like listening now 578 00:26:14,039 --> 00:26:16,960 and not watching the episode, I have to say, Johnson 579 00:26:17,119 --> 00:26:20,000 one of the most amazing recording rooms that I've seen. 580 00:26:20,039 --> 00:26:22,799 I mean, there is memorabilia all over the place. It 581 00:26:22,920 --> 00:26:26,240 is quite a joy to look at, with stormtroopers and 582 00:26:26,640 --> 00:26:31,559 the octocat and multiple Mickey mouses in Italian and you've 583 00:26:31,599 --> 00:26:33,920 just got too many things here to name. I think 584 00:26:34,000 --> 00:26:37,240 your Obi wan Kenobia's blinking red in the background for 585 00:26:37,279 --> 00:26:37,799 some reason. 586 00:26:37,920 --> 00:26:40,000 Speaker 2: Yeah, he is. I'm not sure why. I think that's 587 00:26:40,039 --> 00:26:41,599 just a light, a trick of the lights, but I 588 00:26:41,640 --> 00:26:43,960 see it. I just noticed. 589 00:26:44,359 --> 00:26:46,359 Speaker 1: And I still think that's quite an innovation in the 590 00:26:46,440 --> 00:26:50,359 last fifteen years to really focus on well ninety nine 591 00:26:50,400 --> 00:26:52,319 percent of your work is like not even doing the 592 00:26:52,359 --> 00:26:56,400 development realistically, and I throw in business practices there, like 593 00:26:56,440 --> 00:26:58,039 you're at the end of the day, you're there to 594 00:26:58,079 --> 00:27:01,480 solve a problem which is realistically walking and unlocking you know, 595 00:27:01,519 --> 00:27:04,720 in your example from Disney, the hotel room doors or 596 00:27:04,799 --> 00:27:07,519 the customers, rather than you know, solving just an arbitrary 597 00:27:07,599 --> 00:27:10,039 software problem. And when you're looking in that regard, and 598 00:27:10,119 --> 00:27:11,640 I think this really just makes the whole story about 599 00:27:11,640 --> 00:27:14,599 getting customer feedback and really testing it in a way 600 00:27:14,640 --> 00:27:17,200 that makes sense, and not just deploying it and calling 601 00:27:17,279 --> 00:27:20,000 it a day. And I honestly don't where where're going 602 00:27:20,079 --> 00:27:20,920 with this, but. 603 00:27:22,160 --> 00:27:26,000 Speaker 2: It's the human involvement, right they yeah, no question, Yeah, 604 00:27:26,039 --> 00:27:27,319 you know there was. There's something else you made me 605 00:27:27,359 --> 00:27:30,079 think of a testing too, which is a different story. 606 00:27:30,119 --> 00:27:34,000 Not as long as we were working on something. We had. 607 00:27:34,039 --> 00:27:36,480 This thing used to be a think called oh my gosh, 608 00:27:36,519 --> 00:27:39,079 Disney's Magical Express. Basically they pick it from the airport 609 00:27:39,079 --> 00:27:40,759 and a bus and bring it back to the hotels. 610 00:27:40,839 --> 00:27:44,079 And there was the system basically where they had some 611 00:27:44,119 --> 00:27:47,519 custom devices before iPhones and stuff were super popular. iPhones 612 00:27:47,519 --> 00:27:49,640 were early addition at this time, and they had these 613 00:27:49,720 --> 00:27:52,559 like custom Android devices or Windows Device can Remeer, which 614 00:27:52,599 --> 00:27:55,160 they were where you could basically build give it to 615 00:27:55,200 --> 00:27:57,200 the person at the airport who work for Disney to 616 00:27:57,200 --> 00:27:59,400 say it's hey, it's warrant, and he's here with his 617 00:27:59,440 --> 00:28:01,519 family and the four bags and they're going to you know, 618 00:28:01,559 --> 00:28:04,359 this hotel the Contemporary. They get on the bus, they're 619 00:28:04,400 --> 00:28:07,079 on bus forty two. The bus driver has it, says up, 620 00:28:07,079 --> 00:28:09,359 they're here, and they get to the hotel. Somebody checks 621 00:28:09,400 --> 00:28:10,920 them in on the app again in the same kind 622 00:28:10,920 --> 00:28:13,119 of device, and they get in. It's just a very 623 00:28:13,119 --> 00:28:16,119 simple app process, not a lot of functionality, but it 624 00:28:16,160 --> 00:28:18,079 helps I think at the time, it was running twenty 625 00:28:18,079 --> 00:28:20,880 five million luggage pieces a year. It was insane, like 626 00:28:21,119 --> 00:28:22,839 the amount of fruit. But this thing was going through 627 00:28:23,000 --> 00:28:24,799 and there were people, and we had to read to 628 00:28:24,839 --> 00:28:27,720 build it for some other crazy reasons, and we had 629 00:28:27,720 --> 00:28:30,400 to rebuild this quickly. And I remember the biggest contentious 630 00:28:30,400 --> 00:28:32,519 part I had was like we don't have I said, 631 00:28:32,559 --> 00:28:34,200 I want to do a field test, Like we don't 632 00:28:34,240 --> 00:28:37,119 have time for field tests, Like I know this works 633 00:28:37,319 --> 00:28:39,559 sitting in my cube, but I want to go to 634 00:28:39,599 --> 00:28:41,880 the airport, get on a bus. I want to do 635 00:28:41,920 --> 00:28:43,519 all the steps I don't have to give to all 636 00:28:43,519 --> 00:28:45,240 this users. You don't want to, but letting me be 637 00:28:45,319 --> 00:28:47,440 the person who does it, And like, why would you 638 00:28:47,480 --> 00:28:49,319 want to do this? I'm like, because at the airport, 639 00:28:49,680 --> 00:28:51,759 I'm on airport Wi Fi on. 640 00:28:51,759 --> 00:28:54,359 Speaker 1: The buses right there, right like yeah. 641 00:28:54,000 --> 00:28:55,960 Speaker 2: And then I'm on cell phone when I'm on the bus. 642 00:28:55,960 --> 00:28:57,599 And then when I get to the resort. There are 643 00:28:57,640 --> 00:29:00,319 two Wi Fi signals at each resort. They guess the 644 00:29:00,319 --> 00:29:02,200 one you can use when you come Warren and the 645 00:29:02,559 --> 00:29:04,920 one for the cast members the people who work at Disney. 646 00:29:05,160 --> 00:29:07,079 So I'm like, I don't know how this thing's going 647 00:29:07,119 --> 00:29:10,240 to behave based upon the switching of networks, because each 648 00:29:10,279 --> 00:29:13,279 one has different security levels, different security you know, to 649 00:29:13,319 --> 00:29:15,839 do it and lo and behold, I did it anyway 650 00:29:15,880 --> 00:29:17,960 when they said not to, and me and my buddy 651 00:29:17,960 --> 00:29:20,079 went and did all this, and we found that the 652 00:29:20,160 --> 00:29:22,640 switching of the Wi fis and the cell phones again 653 00:29:22,640 --> 00:29:25,599 ten years ago did not work well. We had we 654 00:29:25,680 --> 00:29:27,799 had to change things up, and the speeds were so 655 00:29:27,920 --> 00:29:30,720 completely drastically different. We had to change what kind of 656 00:29:30,759 --> 00:29:33,279 data was requested at each stage the way too. And 657 00:29:33,279 --> 00:29:35,079 this stuff comes in play. If you don't test it, 658 00:29:35,400 --> 00:29:37,079 how do you know? I mean, can you then we're 659 00:29:37,079 --> 00:29:38,720 deploying this a Christmas By the way, can you do 660 00:29:38,720 --> 00:29:40,519 you have any idea how many people come to Disney 661 00:29:40,559 --> 00:29:43,079 at Christmas? Like? Can you imagine if this thing broke 662 00:29:43,160 --> 00:29:45,680 down and all these unhappy people not being able to 663 00:29:45,680 --> 00:29:47,680 get their stuff, and all it took was me four 664 00:29:47,720 --> 00:29:49,240 hours to get on you know, go to the airport, 665 00:29:49,279 --> 00:29:50,400 get on a bus and drive back. 666 00:29:50,440 --> 00:29:52,440 Speaker 1: I always found that so ridiculous when someone says we 667 00:29:52,480 --> 00:29:55,519 don't have time to test it, because you know there's 668 00:29:55,599 --> 00:29:58,720 the then I don't do you have time to spend multiple 669 00:29:58,759 --> 00:30:03,279 sleepless nights to debug production incident while it's happening like that. 670 00:30:03,559 --> 00:30:04,599 That's the trade off there. 671 00:30:04,640 --> 00:30:07,279 Speaker 2: I'm want to do enjoy Christmas break without being interrupted, 672 00:30:07,440 --> 00:30:10,160 you know with my family. That was my motivation, but. 673 00:30:10,240 --> 00:30:12,000 Speaker 1: Comes out of a different budget that comes out of like, 674 00:30:12,039 --> 00:30:14,319 you know, the production budget rather than the you know 675 00:30:14,519 --> 00:30:15,359 the budget. 676 00:30:15,680 --> 00:30:17,640 Speaker 2: Yep, And it did by way. At Disney they had 677 00:30:17,680 --> 00:30:20,240 they had development, then they had sustainment, and sustainment was 678 00:30:20,240 --> 00:30:22,440 a completely different budget. I'm like, why do we have 679 00:30:22,480 --> 00:30:24,880 such a big sustainment budget, Like maybe if we did 680 00:30:24,920 --> 00:30:26,599 a better job up front, we wouldn't need to have 681 00:30:26,599 --> 00:30:27,759 so many problems the back end. 682 00:30:27,960 --> 00:30:30,039 Speaker 1: Well, I think the incentives drive a lot of things 683 00:30:30,079 --> 00:30:33,000 in this world, especially in software development, and I think 684 00:30:33,039 --> 00:30:35,960 that's one of the things that we see now starting 685 00:30:35,960 --> 00:30:39,480 to cause problems in just the organizational structure and tech 686 00:30:39,480 --> 00:30:43,960 companies where we're we're somehow decided all of a sudden. 687 00:30:44,119 --> 00:30:46,960 You ask any software engineer, even I'll say tech lead 688 00:30:47,000 --> 00:30:50,640 or software manager is evaluating people on the lines of 689 00:30:50,680 --> 00:30:52,880 code like the sheer number that they produce a good 690 00:30:52,920 --> 00:30:55,200 metric for success, And I don't think any of them 691 00:30:55,240 --> 00:30:57,480 really say yes. I will say there are some parts 692 00:30:57,480 --> 00:30:59,680 of the world where they believe that the answer is yes. 693 00:31:00,200 --> 00:31:02,319 But everywhere else, I think everyone agrees that the answer 694 00:31:02,359 --> 00:31:05,839 is now. But then we see these technologies that we're utilizing, 695 00:31:05,920 --> 00:31:08,440 and we're getting to a point where people are being 696 00:31:08,640 --> 00:31:12,319 I'm honestly rewarded, but maybe punished or not delivering by 697 00:31:12,400 --> 00:31:15,720 sheer quantity as much as they have before. And I 698 00:31:15,799 --> 00:31:19,319 just I don't see the incentives growing to meet the 699 00:31:19,440 --> 00:31:23,599 changing world that we're in because arguably, if the job, 700 00:31:23,759 --> 00:31:27,000 the development, the creation, the design, the testing is becoming harder, 701 00:31:27,240 --> 00:31:30,240 so then is the management of those things as well. 702 00:31:30,440 --> 00:31:32,160 And I just think there may be a challenge to 703 00:31:32,319 --> 00:31:36,519 also evolve the management of technical resources to the twenty 704 00:31:36,559 --> 00:31:37,200 first centric. 705 00:31:37,279 --> 00:31:39,319 Speaker 2: I think the way we actually look at what makes 706 00:31:39,359 --> 00:31:43,079 a successful a really good developer in the future, what 707 00:31:43,160 --> 00:31:45,279 makes them a good developer. I do think at one point, 708 00:31:45,319 --> 00:31:47,160 probably a long time ago, in lines of code was 709 00:31:47,160 --> 00:31:49,400 a good thing. I mean, I got them glad I'm 710 00:31:49,400 --> 00:31:52,480 way past that point. But in fact, I create a 711 00:31:52,480 --> 00:31:55,400 presentation years ago which I keep evolving cold readable code, 712 00:31:55,559 --> 00:31:57,079 and I did a bunch of research on this for 713 00:31:57,200 --> 00:32:00,480 it used to be clean code, which was a big thing, 714 00:32:00,720 --> 00:32:02,200 and I said, Joe, clean code's good, but what a 715 00:32:02,319 --> 00:32:05,599 readable code? And because you think about it, and the 716 00:32:05,640 --> 00:32:07,000 more studies you do in this, and I think it's 717 00:32:07,000 --> 00:32:10,200 getting more true today with ai IS. Before AI, I 718 00:32:10,279 --> 00:32:12,640 found a bunch of studies at show you spend ten 719 00:32:12,680 --> 00:32:14,559 times the amount of time reading your code then you 720 00:32:14,640 --> 00:32:16,839 do writing. You think about it. We've been lying as 721 00:32:16,880 --> 00:32:19,079 developers for years. People in your family go, what do 722 00:32:19,119 --> 00:32:21,799 you do for a living? I write code? Do you 723 00:32:21,839 --> 00:32:24,160 do you really or do you just literally spend eight 724 00:32:24,200 --> 00:32:26,359 or your hours a day reading code that you wrote 725 00:32:26,640 --> 00:32:28,799 already and kind of tweaking it and moving it around 726 00:32:28,839 --> 00:32:31,400 and refactoring make sure it works. And then with AI, 727 00:32:31,559 --> 00:32:33,599 AI is writing a lot of the code. I'm okay 728 00:32:33,640 --> 00:32:35,480 with it writing the code. I'm not going to get 729 00:32:35,559 --> 00:32:37,680 rid of the reading part. That's the part I find 730 00:32:37,960 --> 00:32:40,160 is we need to be there to make sure it's 731 00:32:40,240 --> 00:32:42,200 doing it. I want a human to be there on 732 00:32:42,240 --> 00:32:44,400 the line for this, because the human's going to be 733 00:32:44,400 --> 00:32:47,480 one as to fix it. If we're in a completely autonomous, 734 00:32:47,519 --> 00:32:50,839 ancientized world for everything to me, that's not We're so 735 00:32:50,920 --> 00:32:53,720 far from that at this point, and I think it's 736 00:32:53,759 --> 00:32:56,920 unrealistic to put us there in the near future. Velopper's 737 00:32:57,039 --> 00:32:59,519 roles are changing. I have no problems. So I'm like 738 00:32:59,519 --> 00:33:01,440 what you said at LM's ready code, I'm have a 739 00:33:01,480 --> 00:33:03,920 problem with any company or using LM's write code. I 740 00:33:03,920 --> 00:33:05,799 do have a problem with them going oh the code 741 00:33:05,799 --> 00:33:09,160 has been written, push go ah, let's fire the developers 742 00:33:09,160 --> 00:33:11,400 and now let's just push. To me, that is I'm 743 00:33:11,440 --> 00:33:13,960 shocked we're not having more disasters because of that kind 744 00:33:13,960 --> 00:33:14,359 of stuff. 745 00:33:14,480 --> 00:33:17,119 Speaker 1: You know, I'm not as shocked because my shock level 746 00:33:17,119 --> 00:33:20,119 hasn't really gone down since I entered the professional world. 747 00:33:20,119 --> 00:33:22,160 Like once I graduated and I went to work for 748 00:33:22,480 --> 00:33:24,440 a bunch of companies and I saw the code that 749 00:33:24,480 --> 00:33:27,440 other engineers were writing, I was shocked that there were 750 00:33:27,440 --> 00:33:31,599 any systems in the world that could work correctly. Yeah, 751 00:33:31,759 --> 00:33:35,160 deployed software on on satellites that never got an update. 752 00:33:35,279 --> 00:33:38,839 Always amazed me how those things could run like bug 753 00:33:38,960 --> 00:33:43,400 free for decades even, And I'm just like wow, realistically wow, 754 00:33:43,480 --> 00:33:45,839 because I look at what's generated today and I have 755 00:33:46,039 --> 00:33:49,160 zero faith. And so there is a thing where the 756 00:33:49,240 --> 00:33:52,359 thought is that LMS and any sort of agentic system 757 00:33:52,440 --> 00:33:56,119 raises the floor up and so on average, the median 758 00:33:56,319 --> 00:33:59,759 is better than it was before. But the expertise that 759 00:34:00,000 --> 00:34:03,240 and into the software development process isn't being improved where 760 00:34:03,279 --> 00:34:05,240 we're not raising the ceiling in any way. So you 761 00:34:05,279 --> 00:34:07,480 still have to be as you pointed out earlier, you 762 00:34:07,519 --> 00:34:09,519 still have to be great at everything that was there 763 00:34:09,679 --> 00:34:12,320 in the past. But people with less ability, you know, 764 00:34:12,400 --> 00:34:15,039 in experienced engineers, are now able to produce things that 765 00:34:15,079 --> 00:34:17,960 are slightly better. If I look at that perspective, it 766 00:34:18,000 --> 00:34:21,760 basically says lms don't help us produce not just LMS, 767 00:34:21,760 --> 00:34:23,280 but any sort of system built on top of them, 768 00:34:23,440 --> 00:34:27,719 don't help us produce better products or more reliable solutions 769 00:34:27,840 --> 00:34:30,119 or anything like that. But it does help people to 770 00:34:30,320 --> 00:34:34,199 potentially learn or And there's even a questionable answer about 771 00:34:34,280 --> 00:34:36,360 learning there because it was a study published by a 772 00:34:36,440 --> 00:34:40,239 Microsoft back sponsor about the reduction and critical thinking when 773 00:34:40,760 --> 00:34:43,880 you pull in automated systems like LMS et cetera to 774 00:34:44,000 --> 00:34:46,559 actually do the critical thinking for you. So there is 775 00:34:46,599 --> 00:34:49,320 a question of you know, where exactly is the best 776 00:34:49,360 --> 00:34:52,800 place in the process for us to actually pull these in. 777 00:34:52,840 --> 00:34:55,320 You know, what are we missing out in society where 778 00:34:55,519 --> 00:34:57,159 this is the right thing? And I think you started 779 00:34:57,159 --> 00:34:59,199 off this episode basically asking, you know, what is the 780 00:34:59,239 --> 00:35:01,800 point where is the actual goal here? I love it 781 00:35:01,840 --> 00:35:04,199 as like a science fiction a direction that we've gone 782 00:35:04,239 --> 00:35:06,519 where oh, yes, now we have these tools that automate 783 00:35:06,559 --> 00:35:09,800 lots of things. I think automation was another area where 784 00:35:09,840 --> 00:35:12,480 a lot of people were rightly concerned early on that 785 00:35:12,559 --> 00:35:14,880 you know, automation will take away jobs in some regard, 786 00:35:15,159 --> 00:35:17,400 but there was a goal there. It was we can't 787 00:35:17,440 --> 00:35:22,119 scale effectively enough without automation. And then that leads me 788 00:35:22,159 --> 00:35:24,960 to believe what the LM revolution or AI revolution or 789 00:35:24,960 --> 00:35:26,679 whatever you want to call it, that the problem is 790 00:35:26,719 --> 00:35:29,480 being able to scale, being able to deliver a factor 791 00:35:29,760 --> 00:35:33,599 more products or technology or software than we would app 792 00:35:33,679 --> 00:35:36,079 before without it. And so that for me makes the 793 00:35:36,159 --> 00:35:38,159 question did we need more things? 794 00:35:38,280 --> 00:35:40,639 Speaker 2: I don't think the humans ever need more things. And 795 00:35:40,800 --> 00:35:43,400 think about it, This goes into a less my father 796 00:35:43,440 --> 00:35:45,199 taught me years ago. The more money you make, the 797 00:35:45,239 --> 00:35:47,360 more you will spend. Right, you will find ways to 798 00:35:47,360 --> 00:35:49,800 spend your money, and I think we'll find ways to 799 00:35:49,840 --> 00:35:53,280 spend our free time as well, which is fascinating. It 800 00:35:53,320 --> 00:35:55,920 wasn't a pan recently about AI, and we had some 801 00:35:56,079 --> 00:35:59,119 very different opinions on where AI should be. And one 802 00:35:59,119 --> 00:36:01,159 of the questions that always in my head is what 803 00:36:01,320 --> 00:36:03,719 problem does that solve? Or a better way to ask 804 00:36:03,880 --> 00:36:06,119 is maybe a kind of a nastier way to ask 805 00:36:06,159 --> 00:36:08,159 It is Okay, ifenderu I future, you're gonna go tell 806 00:36:08,159 --> 00:36:11,360 your friend about Okay, who asked for that? Okay? Sometimes 807 00:36:11,360 --> 00:36:13,360 we're just creating AI for the sake of AI. Doesn't 808 00:36:13,400 --> 00:36:15,599 it feel like that? And you just pointed out Warren 809 00:36:15,800 --> 00:36:19,400 a great thing that's happening everywhere. I don't think we 810 00:36:19,440 --> 00:36:22,079 all see is that AI is being used everywhere. Nobody 811 00:36:22,079 --> 00:36:24,599 knows where it should be used. So what's happening. All 812 00:36:24,639 --> 00:36:28,519 the companies seemingly are saying, let's AI everything and we'll 813 00:36:28,559 --> 00:36:31,480 see what things actually stick. And I know there's more 814 00:36:31,480 --> 00:36:34,360 strategy behind this, but it feels this way, and to 815 00:36:34,760 --> 00:36:36,320 a degree, I think a lot of the places we're 816 00:36:36,320 --> 00:36:38,440 seeing AI today some of those will not exist in 817 00:36:38,480 --> 00:36:42,440 the future, and others they absolutely will and they'll double down. 818 00:36:42,760 --> 00:36:47,039 That's probably how you think about the Industrial Revolution when 819 00:36:47,039 --> 00:36:50,000 it came about, was probably like this, let's automate everything 820 00:36:50,039 --> 00:36:52,079 and let's figure out what actually works. Okay, you know 821 00:36:52,159 --> 00:36:54,880 we can actually do like conveyor belts and things like that, 822 00:36:55,000 --> 00:36:57,760 and machines to generate and create things, but maybe we 823 00:36:57,760 --> 00:37:00,000 actually still need a human to, like, you know, check 824 00:37:00,079 --> 00:37:02,599 quality control because I want to make sure the thing 825 00:37:02,639 --> 00:37:05,119 we just created actually still works like a car roped 826 00:37:05,119 --> 00:37:07,159 the assembly line. I think we're going to find that 827 00:37:07,199 --> 00:37:09,480 more with AIS. There's got to be places that the 828 00:37:09,559 --> 00:37:11,599 human has to be there. And it's it's one of 829 00:37:11,639 --> 00:37:13,599 the reasons I love the name co pilot so much, 830 00:37:13,800 --> 00:37:16,719 and I believe it was Nat Friedman, the former CEO 831 00:37:16,920 --> 00:37:18,760 of GitHub, was one of the creators of the co 832 00:37:18,840 --> 00:37:20,679 pilot name when they first came out with that, and 833 00:37:20,719 --> 00:37:23,440 it's because he's also a pilot of airplanes. So the 834 00:37:23,480 --> 00:37:26,440 story goes that basically, it's like, it's not someone who's 835 00:37:26,440 --> 00:37:28,360 going to replace your job or take it or do 836 00:37:28,440 --> 00:37:31,559 the work for you. It's a pair programmer. It's a 837 00:37:31,599 --> 00:37:34,679 co pilot. You're the pilot and there's someone next to you. 838 00:37:34,719 --> 00:37:37,320 It's this AI tool, not someone but a thing, and 839 00:37:37,639 --> 00:37:40,880 it's working with you. And I like that model because 840 00:37:40,920 --> 00:37:43,280 whether you know, LM's are big a couple of years ago, 841 00:37:43,360 --> 00:37:44,920 and you know, and then it gets into, oh, we're 842 00:37:44,920 --> 00:37:46,800 going to do all this AI agent stuff and who 843 00:37:46,880 --> 00:37:49,039 knows what the next thing is going to be wherever 844 00:37:49,079 --> 00:37:54,239 this goes, if we relinquish the ability of the humans 845 00:37:54,239 --> 00:37:56,920 to be playing a major factor in this. You know, 846 00:37:56,920 --> 00:37:59,360 we laugh at the whole thing of Skynet with Terminator, 847 00:37:59,440 --> 00:38:03,800 but really should we be laughing at that? Because what 848 00:38:03,800 --> 00:38:05,800 do we want us to be? Where's our role as 849 00:38:05,880 --> 00:38:07,920 humans in this? And I do think we need to 850 00:38:07,960 --> 00:38:09,840 play a big role in it. And it's one of 851 00:38:09,880 --> 00:38:12,719 the reasons I love things like Get Up Copilot and yeah, 852 00:38:12,880 --> 00:38:16,639 you know, my team does like Java, Python, that JavaScript, 853 00:38:16,679 --> 00:38:18,760 et cetera. We do a lot of different languages. But 854 00:38:19,039 --> 00:38:21,320 the language wars are over in my mind. People still 855 00:38:21,320 --> 00:38:23,679 fight them. But I'm like, all right, I am going 856 00:38:23,760 --> 00:38:27,320 to use an AI tool to convert from Python to JavaScript. Sorry, 857 00:38:27,400 --> 00:38:30,679 It's just why would I spend my time learning the language, 858 00:38:30,920 --> 00:38:33,320 you know, so in depth that I've already got when 859 00:38:33,320 --> 00:38:34,800 I can use a tool to help me get that 860 00:38:34,840 --> 00:38:37,039 foundation and yeah, and then I'll tweak it and have 861 00:38:37,119 --> 00:38:39,119 somebody look at it. But those are the kind of 862 00:38:39,119 --> 00:38:41,840 things to me AI is good for. Or are generating 863 00:38:41,880 --> 00:38:44,880 the foundational pieces for my new application. You want a 864 00:38:44,960 --> 00:38:47,760 vibe codeed to get it going, sure, but then still 865 00:38:48,000 --> 00:38:52,159 have a process, use software engineering processes that have been 866 00:38:52,199 --> 00:38:54,960 hardened for decades. Make sure all that stuff's there, and 867 00:38:55,000 --> 00:38:57,360 get humans involved to work with it. I just think 868 00:38:57,400 --> 00:39:00,519 if you don't have that, and like release control, version control, 869 00:39:00,639 --> 00:39:05,159 branching strategies, CICD, these are all still very valid. The 870 00:39:05,199 --> 00:39:07,119 only thing I worry about a little bit is I 871 00:39:07,159 --> 00:39:11,079 worry about the idea that, yes, the floor has been raised, 872 00:39:11,360 --> 00:39:14,559 so anybody can code now with these tools. Yes, but 873 00:39:15,039 --> 00:39:17,079 I'm not a huge fan. I'm just going to put 874 00:39:17,079 --> 00:39:19,159 it out there. It might be contentious. Not a fan 875 00:39:19,280 --> 00:39:22,360 of Hey, put your startup idea into this tool and 876 00:39:22,400 --> 00:39:25,440 I'll generate everything for you and it just works prompt 877 00:39:25,519 --> 00:39:29,559 trusts me. Really, it's not a fan of that, and 878 00:39:29,880 --> 00:39:32,039 I like it for a starting point, but then hire 879 00:39:32,119 --> 00:39:33,440 somebody to actually look at it. 880 00:39:33,519 --> 00:39:35,920 Speaker 1: See, you know, I think some of those French things 881 00:39:36,360 --> 00:39:38,719 I do sort of understand though, because if we look 882 00:39:38,760 --> 00:39:42,840 at the startup world, the value in the process here 883 00:39:43,199 --> 00:39:47,360 was sort of idea validation, which requires the human feedback aspect. 884 00:39:47,480 --> 00:39:51,039 And those companies that are switching to using either vibe 885 00:39:51,079 --> 00:39:53,559 coding or some sort of other LM to generate a 886 00:39:53,559 --> 00:39:56,320 lot of their code for them were potentially before it 887 00:39:56,400 --> 00:39:59,199 needed to get funding and then just outsource the software 888 00:39:59,239 --> 00:40:02,119 development to build a prototype in first place. Arguably, I 889 00:40:02,159 --> 00:40:06,920 am not a proponent of outsourcing critical internal competency. If 890 00:40:06,920 --> 00:40:09,559 you want to be a technical company, out like to 891 00:40:09,599 --> 00:40:12,760 an outsourcer to build something for you, because it's I 892 00:40:12,800 --> 00:40:15,079 think it's very similar to the result that we get 893 00:40:15,079 --> 00:40:17,239 when we use an LM. So I think in that 894 00:40:17,280 --> 00:40:19,800 regard there is a short circuit there. But I think 895 00:40:19,800 --> 00:40:22,280 what you've made me realize is that maybe the metric 896 00:40:22,400 --> 00:40:25,519 for value in our process or in the output that 897 00:40:25,559 --> 00:40:27,960 we have is the number of human hours that have 898 00:40:28,039 --> 00:40:30,159 gone into it. It doesn't matter what tools you use, or 899 00:40:30,480 --> 00:40:33,159 how many outsourcers you use and whatnot. The result is 900 00:40:33,199 --> 00:40:35,960 like internal into the company when you're building that product, 901 00:40:35,960 --> 00:40:37,960 whatever tools you're using, how many hours are you putting 902 00:40:38,000 --> 00:40:40,800 into it. You don't get to short circuit that in 903 00:40:40,840 --> 00:40:44,159 any way, because that's really the intelligent part of the process. 904 00:40:44,320 --> 00:40:47,360 It doesn't matter how great those lms are for us. 905 00:40:47,159 --> 00:40:49,440 As you pointed out, I think the reason you're thinking 906 00:40:49,440 --> 00:40:51,519 about having a human in the loop is because you 907 00:40:51,559 --> 00:40:56,280 recognize that there's always something there that can never be automated. 908 00:40:55,800 --> 00:40:57,639 Speaker 2: Away, and I think you still have to have critical 909 00:40:57,639 --> 00:41:00,559 thinking skills throughout this process. I do think that development 910 00:41:00,559 --> 00:41:02,760 will take a lot less time now with a I 911 00:41:02,880 --> 00:41:05,159 use the tools properly, it should do that. But I 912 00:41:05,159 --> 00:41:07,079 don't think it goes from you know what would once 913 00:41:07,159 --> 00:41:09,880 take six months and one hundred developers to build. I've 914 00:41:09,880 --> 00:41:11,760 been on projects that big. It's not going to go 915 00:41:11,760 --> 00:41:14,480 from six months hundred developers down to six minutes and 916 00:41:14,519 --> 00:41:17,880 no developers. Just it's not going to do that. And 917 00:41:18,199 --> 00:41:20,400 if you do that at your company, but you know, 918 00:41:20,519 --> 00:41:22,519 I wish you all the luck in the world, that's 919 00:41:22,519 --> 00:41:23,599 not going to be where I'm going to be. 920 00:41:23,880 --> 00:41:25,920 Speaker 1: I think this is where a lot of my faith 921 00:41:26,000 --> 00:41:29,199 in the innovation falls down is that I look at 922 00:41:29,239 --> 00:41:31,519 these companies that are promising me some sort of tool 923 00:41:31,639 --> 00:41:33,400 to do a lot of work or a lot of 924 00:41:33,400 --> 00:41:35,000 the work that I think has to be done, and 925 00:41:35,079 --> 00:41:37,480 I asked them, Okay, let's say I've been using your 926 00:41:37,480 --> 00:41:40,159 tool for five years now and I've built you know, 927 00:41:40,760 --> 00:41:43,559 billion dollar business on it that has an incredibly complex 928 00:41:43,599 --> 00:41:47,239 software stack. What do you expect the process around your 929 00:41:47,280 --> 00:41:49,760 tool usage to be or the architecture of those things. 930 00:41:49,800 --> 00:41:52,719 If it's spec driven development, how many spec files do 931 00:41:52,760 --> 00:41:54,519 you expect me to have. What do those look like? 932 00:41:54,800 --> 00:41:56,840 What is that like when someone needs to change something? 933 00:41:56,840 --> 00:41:59,039 What does that go? And they're like, we didn't really 934 00:41:59,039 --> 00:42:01,039 think about that. We were just thinking about you need to, 935 00:42:01,599 --> 00:42:03,840 you know, make a UI and so you first do 936 00:42:03,920 --> 00:42:05,920 this thing and then you write this command and bam, 937 00:42:05,960 --> 00:42:08,000 you're done. And I'm like, I don't care. That's like 938 00:42:08,079 --> 00:42:10,719 two hours, ten hours, one hundred hours even to get 939 00:42:10,760 --> 00:42:13,480 something done. I'm way more interested in the long term 940 00:42:13,480 --> 00:42:15,440 process here. And if you say, well, that's going to 941 00:42:15,519 --> 00:42:19,800 take like years of agent time to iterate on solving 942 00:42:19,800 --> 00:42:21,679 that question you've asked. When you're at that scale, I'm like, well, 943 00:42:21,679 --> 00:42:23,760 that's not a real solution. You haven't really solved any 944 00:42:23,840 --> 00:42:27,239 hard problem here. You set up a lemonade stand and 945 00:42:27,440 --> 00:42:28,679 you didn't really go further than that. 946 00:42:28,800 --> 00:42:30,880 Speaker 2: And I think some of this is user preference too, 947 00:42:30,920 --> 00:42:33,119 Like when you u seing ghetto Copela at agent mode. 948 00:42:33,159 --> 00:42:34,760 One of the things I like, I like it to 949 00:42:34,800 --> 00:42:36,480 ask me questions along the way. I don't want it 950 00:42:36,519 --> 00:42:38,679 to go away for half an hour and build something 951 00:42:38,800 --> 00:42:40,760 and give it to me. I wanted to, you know, 952 00:42:41,079 --> 00:42:43,119 give it the context and go And if I wasn't 953 00:42:43,159 --> 00:42:45,559 clear in something, I wanted to say, hey, did you 954 00:42:45,639 --> 00:42:48,159 want this UI to have a save button or to 955 00:42:48,239 --> 00:42:51,320 no save button? Making this up? Oh yeah, I didn't 956 00:42:51,360 --> 00:42:52,679 tell you that. So I wanted to have a save 957 00:42:52,760 --> 00:42:55,039 button here because I want an explicit you know, HP 958 00:42:55,199 --> 00:42:57,440 call to go to back end? Okay when using ag GTP, 959 00:42:57,599 --> 00:42:59,480 do you want to use HPS or what other kind 960 00:42:59,519 --> 00:43:02,800 of bag? I wanted to ask me questions along the way. Now, 961 00:43:02,840 --> 00:43:04,719 I don't want it to be a you know that 962 00:43:04,760 --> 00:43:06,599 guy who sits in the cube next to you and 963 00:43:06,639 --> 00:43:10,360 who's asking you forty eight questions a minute, but there's 964 00:43:10,400 --> 00:43:12,400 a there's a lever, and I want to be able 965 00:43:12,440 --> 00:43:15,159 to tune that with my agent. So it's like and 966 00:43:15,199 --> 00:43:17,360 that's also part of the agents MD for me is 967 00:43:17,559 --> 00:43:19,760 ask me questions when you're not sure which way to 968 00:43:19,800 --> 00:43:21,920 go before you go off on a ten minute adventure 969 00:43:21,920 --> 00:43:24,960 of coding. Mister agent or missus agent, you know, come 970 00:43:25,000 --> 00:43:26,840 ask me or show me what you've done, give me 971 00:43:26,880 --> 00:43:30,239 a demo, show me screenshots. This is my preference. I 972 00:43:30,239 --> 00:43:32,960 know there's some people who I've talked to after sessions 973 00:43:32,960 --> 00:43:35,800 on this and workshops where they're like, I'm totally not 974 00:43:35,840 --> 00:43:37,800 like that, John, I disagree. I want the thing to 975 00:43:37,840 --> 00:43:39,440 create the whole thing for me and then I'll look 976 00:43:39,480 --> 00:43:41,440 at it. So there's different ways of going about it. 977 00:43:41,679 --> 00:43:43,239 I think that's part of what we have to figure 978 00:43:43,239 --> 00:43:46,000 out is what is the best way for different companies 979 00:43:46,039 --> 00:43:48,639 to adopt standards, which I'll come back to. My main 980 00:43:48,639 --> 00:43:50,840 point of this whole episode is companies still need to 981 00:43:50,880 --> 00:43:53,559 have critical thinking, They still need to have processes. They 982 00:43:53,559 --> 00:43:56,159 still need to have guidelines, and I call them style 983 00:43:56,239 --> 00:43:59,360 guides because I've written these things. Basically, it's what is 984 00:43:59,360 --> 00:44:02,559 your company expect from your development output? Not just a 985 00:44:02,599 --> 00:44:06,920 functional program, but do they expect you know, these get 986 00:44:07,039 --> 00:44:10,760 up commits? Do they expect small prs, they expect certain assets, 987 00:44:10,920 --> 00:44:12,960 There's certain things that should be there along and long 988 00:44:13,000 --> 00:44:15,800 development cycle. Do they expect a demo a week? We 989 00:44:15,840 --> 00:44:17,760 did that at Disney a lot every week or every 990 00:44:17,760 --> 00:44:19,800 two weeks. It was every two weeks. We'd have an 991 00:44:19,800 --> 00:44:22,320 iteration end and we'd have a demo to the users 992 00:44:22,320 --> 00:44:24,119 to say, here's what we have so far, this is 993 00:44:24,159 --> 00:44:26,800 the feat she you asked for. Great chance to give feedback, 994 00:44:27,559 --> 00:44:29,440 you know, also make some even if everything there is 995 00:44:29,480 --> 00:44:31,320 no feedback. It always made them feel good to go, 996 00:44:31,800 --> 00:44:34,039 all right, they got this. I want that from my agent. 997 00:44:34,239 --> 00:44:35,920 Speaker 1: You just made me think of this perspective that came 998 00:44:35,920 --> 00:44:38,239 out not too long ago that was saying something like 999 00:44:38,480 --> 00:44:41,639 only five percent of i'll say enterprise companies that are 1000 00:44:41,679 --> 00:44:45,199 utilizing any sort of ailm et cetera are actually getting 1001 00:44:45,199 --> 00:44:48,000 the value out and I don't know how much it 1002 00:44:48,079 --> 00:44:51,039 really went into. I mean, it's hard to justify the why, right, 1003 00:44:51,119 --> 00:44:54,000 like why isn't that the case? And I can understand 1004 00:44:54,079 --> 00:44:56,280 the perspective of those I hate to call them vibe 1005 00:44:56,280 --> 00:44:58,280 coders that just say I want the tool to do everything, 1006 00:44:58,599 --> 00:45:01,559 because I think this maybe where we disagree with other people. 1007 00:45:01,800 --> 00:45:05,239 We know that the tool isn't capable of doing everything yet, 1008 00:45:05,280 --> 00:45:07,159 and so we want the questions to be asked so 1009 00:45:07,199 --> 00:45:09,960 that we can clarify them, whereas I can see other people, 1010 00:45:10,280 --> 00:45:12,119 and I think there's maybe sort of the why that's 1011 00:45:12,159 --> 00:45:16,440 been propagated by these companies that are in almost trillion 1012 00:45:16,480 --> 00:45:20,360 dollar debt now trying to justify their products creation that 1013 00:45:20,440 --> 00:45:23,599 they can do everything, because we know that they get stuck. 1014 00:45:23,639 --> 00:45:25,719 We know that there are struggles there that they just 1015 00:45:25,719 --> 00:45:27,599 can't have the right answer to. There are too many. 1016 00:45:27,840 --> 00:45:30,800 It depends in our experience for it to solve that 1017 00:45:30,920 --> 00:45:34,199 and that we can answer those questions for them, but 1018 00:45:34,199 --> 00:45:39,119 we have no faith in the agents actually answering them correctly. 1019 00:45:39,239 --> 00:45:40,960 Speaker 2: Yeah, I think it's something we all have to be 1020 00:45:41,000 --> 00:45:43,800 aware of. Understand where's AI playing a role in our lives? 1021 00:45:43,880 --> 00:45:47,840 And to me, AI is I'm going to say something 1022 00:45:47,840 --> 00:45:50,320 it probably shouldn't say out out too much, which is 1023 00:45:50,599 --> 00:45:53,000 AI is just another tool that we're all going to 1024 00:45:53,039 --> 00:45:56,239 learn to use, just like the cloud was amazing, the 1025 00:45:56,280 --> 00:45:59,400 Internet was amazing, all these foundational changes and shifts that 1026 00:45:59,480 --> 00:46:02,880 were out there. It's just as dramatic as all those changes. 1027 00:46:02,920 --> 00:46:05,320 Like think, what how when the internet came that's like AI. 1028 00:46:05,719 --> 00:46:07,599 We're going to be using it, but there'll be other 1029 00:46:07,719 --> 00:46:10,639 things and our lives still go on. You think, when 1030 00:46:10,679 --> 00:46:12,519 the Internet came out, it opened up new our doors 1031 00:46:12,519 --> 00:46:15,920 and possibilities and efficiencies. But we still have our businesses, 1032 00:46:16,159 --> 00:46:18,239 and we still have our jobs, and we still have 1033 00:46:18,599 --> 00:46:22,079 our lives and those things still exist. So AI is 1034 00:46:22,159 --> 00:46:24,719 not the most important thing in the world. It is 1035 00:46:24,760 --> 00:46:26,480 a part of our worlds which we have to figure 1036 00:46:26,480 --> 00:46:28,519 out where does it actually fit in and where doesn't 1037 00:46:28,559 --> 00:46:30,880 it fit in? So feel free, everybody out there to 1038 00:46:30,960 --> 00:46:32,800 use that statement and ask when somebody says, here's a 1039 00:46:32,800 --> 00:46:34,800 new AI feature and go who asked for that? 1040 00:46:34,880 --> 00:46:36,280 Speaker 1: I just just feel like we're both going to get 1041 00:46:36,280 --> 00:46:37,159 so many angry letters. 1042 00:46:37,440 --> 00:46:41,039 Speaker 2: Oh yeah, absolutely, I get them all the time. Yeah, 1043 00:46:41,079 --> 00:46:42,639 I work in an AI world and I've got to 1044 00:46:42,679 --> 00:46:44,039 do it. But you know, at the same time, you 1045 00:46:44,119 --> 00:46:46,679 got to be real, like, AI is amazing, but it's 1046 00:46:46,719 --> 00:46:47,239 not everything. 1047 00:46:47,320 --> 00:46:49,480 Speaker 1: I still want to continue this conversation, but I also 1048 00:46:49,480 --> 00:46:51,440 feel like this may be a good jumping off point 1049 00:46:51,480 --> 00:46:55,000 to go over to picks. So before we go there, 1050 00:46:55,039 --> 00:46:57,639 I'll ask, is there any last thing, any last point 1051 00:46:57,679 --> 00:46:59,800 of conversation, anything that you think that we've missed that 1052 00:47:00,079 --> 00:47:02,400 you just feel like, Yeah, you have to tell the viewers. 1053 00:47:02,519 --> 00:47:05,320 Speaker 2: Yeah, I think if you're technically wise, if you're looking 1054 00:47:05,360 --> 00:47:08,079 for something with AI, go check out get Up Copilot. 1055 00:47:08,159 --> 00:47:11,039 I guarantee you there's features that you didn't even know 1056 00:47:11,079 --> 00:47:13,199 existed in there. Every time I do my Get Up 1057 00:47:13,199 --> 00:47:15,800 Copilot sessions that I do or workshops and people who 1058 00:47:15,840 --> 00:47:18,079 are using these things come to them, they find ten 1059 00:47:18,159 --> 00:47:20,000 new things that they didn't know existed and can make 1060 00:47:20,000 --> 00:47:22,480 their lives better. So these tools are moving so fast. 1061 00:47:22,639 --> 00:47:24,199 Give it a shot. Take a look at it and 1062 00:47:24,480 --> 00:47:26,559 check out agent mode. Check out the cli. Do is 1063 00:47:26,559 --> 00:47:28,920 a copilot cli. Some people will know. Got to check 1064 00:47:28,920 --> 00:47:30,119 these things out. They're pretty cool. 1065 00:47:30,239 --> 00:47:32,480 Speaker 1: Yeah, even if you don't go there for the copilot. 1066 00:47:32,480 --> 00:47:34,440 I've got to say I've been on a huge kick 1067 00:47:34,480 --> 00:47:38,239 since the NPM security incident with the malicious packages to 1068 00:47:38,320 --> 00:47:41,079 use dev containers everywhere, and in the process of doing that, 1069 00:47:41,599 --> 00:47:44,599 I went down the copilot route and as part of that, 1070 00:47:45,000 --> 00:47:47,800 I got to say the code completions in line, like 1071 00:47:47,920 --> 00:47:51,639 every single time, the first the first completion always exactly 1072 00:47:51,639 --> 00:47:53,920 what I want. And so just one at a time. 1073 00:47:53,960 --> 00:47:55,599 If you don't even want to engage with a model 1074 00:47:55,639 --> 00:47:56,920 and you just want that and you just want to 1075 00:47:56,920 --> 00:47:58,400 try that out, like, just go and do that, and 1076 00:47:58,440 --> 00:48:00,920 then after that if you're done, that's fine. But it's 1077 00:48:00,920 --> 00:48:04,280 definitely a great starting point, honestly. So with that, let's 1078 00:48:04,400 --> 00:48:06,800 we're going to move over to picks. So, John, what 1079 00:48:07,199 --> 00:48:08,280 did you bring for us today? 1080 00:48:08,400 --> 00:48:11,239 Speaker 2: Totally non technical. One of the ways I keep up 1081 00:48:11,280 --> 00:48:14,280 with everything is I used to play soccer a lot 1082 00:48:14,400 --> 00:48:17,360 years ago, and as of aged, my body just can't 1083 00:48:17,440 --> 00:48:20,360 do it anymore. Too many joint issues and things. Keeping 1084 00:48:20,400 --> 00:48:23,400 healthy mind and body is really important out there, and 1085 00:48:23,480 --> 00:48:26,199 I've gotten into running. I do something called F forty five, 1086 00:48:26,239 --> 00:48:29,559 which basically just to workout programs. I do strength resistance, 1087 00:48:29,679 --> 00:48:33,079 I do cardio. Find something out there, folks, do some 1088 00:48:33,199 --> 00:48:36,199 kind of activity, get away from the desk, go outside, 1089 00:48:36,559 --> 00:48:39,079 do something. And my pick for all this is what 1090 00:48:39,199 --> 00:48:42,079 got me really motivated to do this was even though 1091 00:48:42,079 --> 00:48:44,920 I played soccer for years, I'd never run more than 1092 00:48:44,960 --> 00:48:48,159 I'd say, two miles consecutively anywhere, even like at a 1093 00:48:48,199 --> 00:48:51,320 slow jog. I got into Run Disney. Run Disney is 1094 00:48:51,360 --> 00:48:53,719 a do you run at Disney World exactly what it is. 1095 00:48:53,880 --> 00:48:56,599 They run five kse ten case half marathons and marathons, 1096 00:48:56,800 --> 00:48:59,400 and you do it with through the parks, with characters 1097 00:48:59,440 --> 00:49:02,559 and music and all sorts of stuff going on with 1098 00:49:02,760 --> 00:49:05,599 you know, tens of thousands of people and it's amazing. 1099 00:49:05,880 --> 00:49:08,639 If you haven't checked out Run Disney, you really should. 1100 00:49:08,760 --> 00:49:11,119 It's just an amazing experience, a bunch of people who 1101 00:49:11,199 --> 00:49:14,320 are totally non competitive having a good time to run 1102 00:49:14,360 --> 00:49:17,719 through and it's helped me change my entire physical and 1103 00:49:18,079 --> 00:49:20,960 mental emotional well being over the last several years. And 1104 00:49:21,239 --> 00:49:23,239 definitely check it out. That does sound really interesting. 1105 00:49:23,280 --> 00:49:26,039 Speaker 1: I've always found my struggle with just running is that 1106 00:49:26,079 --> 00:49:29,199 I have this internal monologue that just sort of gets 1107 00:49:29,199 --> 00:49:32,039 bored unless I have something else to engage with. Ye, 1108 00:49:32,199 --> 00:49:34,800 which has always been a challenge for me. But I 1109 00:49:34,800 --> 00:49:37,800 feel like being in a particular environment, definitely one that 1110 00:49:37,840 --> 00:49:41,519 you necessarily haven't been to before, it definitely creates a 1111 00:49:41,920 --> 00:49:42,840 unique opportunity. 1112 00:49:42,960 --> 00:49:45,519 Speaker 2: That's where audio books can be awesome, by the way, 1113 00:49:45,639 --> 00:49:47,159 and podcasts. 1114 00:49:46,639 --> 00:49:48,239 Speaker 1: Yeah, I mean, if you're not listening to the Adventures 1115 00:49:48,280 --> 00:49:50,400 in DeVos podcast while you're going for a run, I 1116 00:49:50,440 --> 00:49:53,880 mean you missing out on a unique opportunity there, exactly. 1117 00:49:54,599 --> 00:49:56,159 Speaker 2: Yeah. So what did I bring? 1118 00:49:56,199 --> 00:49:58,559 Speaker 1: I did the lame thing. I guess I brought a technology. 1119 00:49:59,119 --> 00:50:02,599 I've been to thing ear pods, different brands and companies 1120 00:50:02,599 --> 00:50:04,000 to try to figure out what I want to use 1121 00:50:04,000 --> 00:50:06,199 for the podcast because I used to be using these 1122 00:50:06,400 --> 00:50:09,440 Sennheiser I'll just hold them up real quick, and they're 1123 00:50:09,480 --> 00:50:13,639 really great for high quality stuff, but I don't like 1124 00:50:13,679 --> 00:50:17,039 them interacting with me while I'm on a call or recording. 1125 00:50:17,039 --> 00:50:18,960 It's just one more thing that I'm just aware or to. 1126 00:50:19,079 --> 00:50:23,159 So I recently finally sat on these shocks open open 1127 00:50:23,199 --> 00:50:25,559 fit too I'll just hold up with the little is 1128 00:50:25,599 --> 00:50:28,400 and of course it's got my magnetic USB C charger 1129 00:50:28,440 --> 00:50:30,199 in it, because I love those things as well. 1130 00:50:30,559 --> 00:50:31,159 Speaker 2: This has been. 1131 00:50:31,039 --> 00:50:33,320 Speaker 1: Absolutely great there. They don't go in the ear, so 1132 00:50:33,320 --> 00:50:36,639 they don't like cause me like physical pain after hours 1133 00:50:36,679 --> 00:50:38,679 and hours having them on. I can bring them on 1134 00:50:38,679 --> 00:50:41,360 the plane or for eight plus hours when I'm traveling 1135 00:50:41,400 --> 00:50:44,519 and they just they just sit there absolutely like just fantastic, 1136 00:50:44,599 --> 00:50:47,440 and I can hear everything. Bluetooth of course, tried so 1137 00:50:47,599 --> 00:50:49,639 many things and just disliked all of them. 1138 00:50:49,800 --> 00:50:52,639 Speaker 2: You know, I use shocks for running. I don't use 1139 00:50:52,679 --> 00:50:54,480 the open, open ear ones. I use the ones that 1140 00:50:54,519 --> 00:50:55,320 wrap around your head. 1141 00:50:55,320 --> 00:50:56,280 Speaker 1: They're bone conducting. 1142 00:50:56,400 --> 00:50:59,119 Speaker 2: The bone conducting for the two biggest reasons. One is 1143 00:50:59,159 --> 00:51:01,239 I don't like having things my ears all day. Yes 1144 00:51:01,679 --> 00:51:04,119 eventually it hurts or yeah, lose a little bit of 1145 00:51:04,320 --> 00:51:06,039 And the other part is when you're running, I don't 1146 00:51:06,039 --> 00:51:08,679 want to hit by anything. So having the bone conducting, 1147 00:51:08,719 --> 00:51:11,000 I can hear everybody around me and hear my music 1148 00:51:11,079 --> 00:51:12,760 or podcast and they're amazing. 1149 00:51:12,880 --> 00:51:14,719 Speaker 1: Yeah, so these aren't bone conducting, but actually I can 1150 00:51:14,760 --> 00:51:16,599 just show them. They don't have a piece in it, 1151 00:51:16,679 --> 00:51:19,880 so they're just sitting actually outside my ear for a 1152 00:51:19,920 --> 00:51:22,079 similar reason that I can hear everything else that's going on. 1153 00:51:22,280 --> 00:51:24,400 But yeah, the shocks bone conducting one. The only reason 1154 00:51:24,400 --> 00:51:26,679 I didn't go with them is that it felt a 1155 00:51:26,679 --> 00:51:29,199 little weird and I still had to turn the volume 1156 00:51:29,239 --> 00:51:31,719 off to make it happen. But yeah, when I exercise 1157 00:51:31,760 --> 00:51:34,119 and whatnot, it's definitely the thing that I much prefer 1158 00:51:34,440 --> 00:51:36,840 because being able to hear and not get well. Luckily, 1159 00:51:36,960 --> 00:51:40,920 I'm actually in Switzerland and so actually really close to Tacchino, 1160 00:51:41,639 --> 00:51:45,719 the Italian canton in Switzerland, so not as many cars here, 1161 00:51:45,760 --> 00:51:47,880 but if you're in anywhere where there are cars, it's 1162 00:51:47,920 --> 00:51:51,920 definitely a huge danger to go around. Yes, with the 1163 00:51:52,360 --> 00:51:53,239 plugs in your ears. 1164 00:51:53,280 --> 00:51:55,320 Speaker 2: It has saved me several times. My wife has told 1165 00:51:55,320 --> 00:51:57,920 me many times she worries about me running, and I'm like, 1166 00:51:58,000 --> 00:52:01,960 thank god, I've got these bone conducts headphones. Because some cars, 1167 00:52:02,039 --> 00:52:04,519 especially live in Florida, so people sort of do not 1168 00:52:04,599 --> 00:52:07,280 know how to drive, and very luckily I've not been it. 1169 00:52:07,440 --> 00:52:10,119 Speaker 1: Now, I think that's probably a statement that's true no 1170 00:52:10,159 --> 00:52:13,280 matter like which state or country you go to. No 1171 00:52:13,440 --> 00:52:16,199 people here they don't necessarily know how to drive. The 1172 00:52:16,199 --> 00:52:18,719 car is Fitzalan. They do stop for you actually here. 1173 00:52:19,199 --> 00:52:20,199 I don't know if they know how to drive, but 1174 00:52:20,239 --> 00:52:23,719 they definitely are way more aware of pedestrians going around and. 1175 00:52:23,679 --> 00:52:26,199 Speaker 2: Who's better AI drivers or human drivers like that. We 1176 00:52:26,199 --> 00:52:27,360 didn't get into that hall of thing. 1177 00:52:27,480 --> 00:52:30,360 Speaker 1: So and I and I shuddered to at the end 1178 00:52:30,400 --> 00:52:34,320 of the episode open At this moment, I guess that'll 1179 00:52:34,400 --> 00:52:37,800 be a research for the listeners to decide to go 1180 00:52:37,840 --> 00:52:40,800 out and do who is better. And maybe with that, 1181 00:52:40,840 --> 00:52:44,079 I'll say thank you John so much for coming on today. 1182 00:52:44,400 --> 00:52:46,320 It's been an absolute fantastic episode. 1183 00:52:46,400 --> 00:52:47,920 Speaker 2: Yeah, thanks for having me. It was really fun talking 1184 00:52:47,920 --> 00:52:48,280 to you. 1185 00:52:48,519 --> 00:52:51,199 Speaker 1: Yeah, this is absolutely great. And if anyone wants to 1186 00:52:51,199 --> 00:52:53,039 see John back on the show, you know, feel free 1187 00:52:53,039 --> 00:52:55,039 to email email us and tell us what we missed 1188 00:52:55,119 --> 00:52:57,920 or what we got wrong, and we'll try to get 1189 00:52:57,960 --> 00:52:59,880 him back on in his busy schedule and one of 1190 00:53:00,119 --> 00:53:02,360 winned the next couple of years when we see UH 1191 00:53:02,480 --> 00:53:06,239 how AI actually goes and UH involves over time. And 1192 00:53:06,800 --> 00:53:09,800 with that, I'll say thank you to our listeners for 1193 00:53:09,920 --> 00:53:12,280 joining us for this episode, and we'll see you all 1194 00:53:12,599 --> 00:53:15,400 next week. 1195 00:53:16,519 --> 00:53:16,559 Speaker 2: M