Critter Stack Update with Jeremy Miller 的完整文字稿

of 84 00:04:00.159 --> 00:04:01.379 college in London. 85 00:04:01.400 --> 00:04:03.729 Oh, wow. Saw the coronation. 86 00:04:04.849 --> 00:04:08.889 His first four or five weeks, two monarchs, three prime ministers. 87 00:04:08.949 --> 00:04:12.650 Yeah. They're still on the prime minister role here. What 88 00:04:12.689 --> 00:04:15.129 is it, 10 in four years? It's amazing. 89 00:04:16.490 --> 00:04:20.459 He just got there when Liz Truss was trying to 90 00:04:20.500 --> 00:04:22.100 outrace the lettuce and failed. 91 00:04:22.120 --> 00:04:27.930 Yeah, and failed. Cannot last longer than lettuce. 92 00:04:28.889 --> 00:04:32.720 Let's see. The Titan submersible disaster, the disappearance of Ocean 93 00:04:33.000 --> 00:04:35.879 Gates Titan during a dive to the wreck of the 94 00:04:35.920 --> 00:04:39.740 Titanic triggered a massive multinational search and became an enormous 95 00:04:39.779 --> 00:04:43.699 worldwide news story. You probably know more about that than 96 00:04:43.740 --> 00:04:47.050 I do, Richard. Don't use carbon fiber for submersibles. 97 00:04:47.399 --> 00:04:48.670 That's the end of that. Yeah. 98 00:04:48.759 --> 00:04:50.050 Catastrophic implosion. 99 00:04:50.230 --> 00:04:54.610 It imploded. Yeah, not exploded. Five people died. I mean- 100 00:04:55.360 --> 00:04:57.600 They'd ever felt a thing. It would have been instantaneous. 101 00:04:57.980 --> 00:05:00.160 So incredibly fast, but still. 102 00:05:00.259 --> 00:05:01.680 Chinese spy balloon incident? 103 00:05:01.759 --> 00:05:03.269 Yeah, that's a weird one, isn't it? It was a 104 00:05:03.310 --> 00:05:07.120 weird one. F-22 shoots down a spy balloon? Yeah. 105 00:05:08.410 --> 00:05:12.029 Okay. The Maui wildfires. Bad, bad, bad. 106 00:05:12.230 --> 00:05:13.170 Yeah. Terrible. 107 00:05:13.189 --> 00:05:18.129 There was an earthquake in Morocco, a catastrophic flood in Libya, 108 00:05:19.069 --> 00:05:22.610 or floods, more than one. Kevin McCarthy's removal as House Speaker. 109 00:05:23.610 --> 00:05:25.759 All sorts of And it's gone well since then. 110 00:05:25.779 --> 00:05:25.860 Yeah. 111 00:05:27.930 --> 00:05:30.430 Okay, we got to laugh to keep from crying. So, Richard, 112 00:05:30.990 --> 00:05:32.769 let's talk about space. 113 00:05:33.300 --> 00:05:35.860 I would also suggest that this was the end of COVID, right? 114 00:05:35.959 --> 00:05:39.019 In January, when China finally reopened. Right. But also announced 115 00:05:39.040 --> 00:05:43.139 that its population has fallen. You know, they start to 116 00:05:43.160 --> 00:05:45.610 understand that the way they were doing their assessment of 117 00:05:45.639 --> 00:05:49.769 population wasn't effective. And there was like 100 million fewer 118 00:05:49.810 --> 00:05:50.529 people than they thought. 119 00:05:50.750 --> 00:05:50.970 Wow. 120 00:05:51.029 --> 00:05:53.750 And then in May, when the World Health Organization officially 121 00:05:53.850 --> 00:05:55.129 ended the global emergency. 122 00:05:55.430 --> 00:05:55.589 Yeah. 123 00:05:55.930 --> 00:05:59.410 End of COVID. Yeah. As a pandemic anyway. All right, 124 00:05:59.490 --> 00:06:02.649 into space. Yes. 2023 is really the year that SpaceX 125 00:06:02.689 --> 00:06:09.399 begins to dominate with 98 of 223 launches in the world. 126 00:06:09.579 --> 00:06:15.300 So that's 44% comes from one company. So that's 91 127 00:06:15.300 --> 00:06:18.459 Falcon 9 launches, five Falcon Heavy launches, and the first 128 00:06:18.620 --> 00:06:22.149 two test flights of starship uh the previous year in 129 00:06:22.149 --> 00:06:25.670 22 it was 61 launches they had 99 booster landings 130 00:06:25.730 --> 00:06:28.129 which seems weird they only had 98 launches but 99 131 00:06:28.129 --> 00:06:31.189 booster landings remember that every falcon heavy has two boosters 132 00:06:32.129 --> 00:06:36.939 for a total of 1200 metric tons lifted to orbit 133 00:06:37.389 --> 00:06:41.470 um wow again a crazy record Now, 63 of those 134 00:06:41.470 --> 00:06:46.189 91 Falcon 9 launches were Starlink because Starlink is really 135 00:06:46.250 --> 00:06:49.519 coming into play here, especially in the context of the 136 00:06:49.680 --> 00:06:54.439 Ukraine war. This is the year that the Elon biography 137 00:06:54.480 --> 00:06:57.360 comes out where we find out that at the beginning 138 00:06:57.379 --> 00:06:59.319 of the war, he actually– well, not the beginning. It 139 00:06:59.339 --> 00:07:01.339 was like in November of 22, he actually turned off 140 00:07:01.379 --> 00:07:04.949 Starlink at the request of Putin. That was more than 141 00:07:04.990 --> 00:07:06.509 a little bit of a kerfuffle overall. 142 00:07:07.009 --> 00:07:07.569 Kerfuffle. 143 00:07:07.589 --> 00:07:08.350 Yeah. 144 00:07:08.550 --> 00:07:12.319 Yeah. Interesting times. I would also have to mention China, 145 00:07:12.360 --> 00:07:16.930 who had 67 launches that year compared to SpaceX's 98. 146 00:07:16.930 --> 00:07:20.189 But that is them operating their own space station. Three 147 00:07:20.230 --> 00:07:25.810 supply missions and a few crew changes in 23. In January, 148 00:07:25.870 --> 00:07:28.250 SpaceX does a Transporter 6 mission. I would normally not 149 00:07:28.290 --> 00:07:31.269 mention these. Transporter 6 is literally their flying CubeSats. I 150 00:07:31.310 --> 00:07:33.790 think they flew like 50 CubeSats on it. But one 151 00:07:33.810 --> 00:07:38.389 of them was Caltech's Space Solar Power Demonstrator. This is 152 00:07:38.410 --> 00:07:42.670 a little 50-kilogram CubeSat that actually has deployable solar collectors 153 00:07:42.730 --> 00:07:47.089 on it that had tested 32 different styles of solar collectors, 154 00:07:47.170 --> 00:07:51.529 different types of full of voltaics and so forth, and 155 00:07:51.589 --> 00:07:56.300 included a microwave transmitter and successfully beamed power to Earth. 156 00:07:56.560 --> 00:07:58.160 It wasn't a meaningful amount of

power, but it was 157 00:07:58.240 --> 00:08:01.920 an end-to-end test, essentially, that you can collect solar power 158 00:08:01.959 --> 00:08:06.490 in orbit and beam it onto the surface. It was microwaved, right? microwaves. Yeah. 159 00:08:06.529 --> 00:08:08.449 Microwave microwave is like nine gigahertz. 160 00:08:08.610 --> 00:08:11.810 Was that actually successful? That's been a lot of sci-fi books. 161 00:08:12.250 --> 00:08:12.529 Yeah. 162 00:08:12.589 --> 00:08:12.769 Yeah. 163 00:08:12.810 --> 00:08:15.509 Well, it's, that's the old, you know, back from the seventies, uh, 164 00:08:15.720 --> 00:08:18.240 you know, Gerald O'Neill and so forth. So this was 165 00:08:18.339 --> 00:08:20.560 a genuine end to end test worked a hundred percent. 166 00:08:20.920 --> 00:08:22.980 Like, uh, yeah, we were talking about that in the 167 00:08:23.019 --> 00:08:24.860 nineties that sure that this was a good. 168 00:08:24.939 --> 00:08:26.870 It goes back to the seventies. It's gone a long, 169 00:08:26.930 --> 00:08:29.759 long time. But my, um, We did the data centers 170 00:08:29.819 --> 00:08:32.100 in space talk, and I sort of end that one with, hey, 171 00:08:32.679 --> 00:08:35.350 if we can really lift 100 metric tons to orbit 172 00:08:36.169 --> 00:08:40.009 with a starship at a price of less than $ 100 million, 173 00:08:41.070 --> 00:08:45.250 maybe space-based power is actually feasible today. Because now we 174 00:08:45.909 --> 00:08:50.929 talk about the most advanced theoretical designs, the SPS Alpha 175 00:08:50.960 --> 00:08:55.500 designs from the British Aeronautical Society. They're talking about 3,000 176 00:08:55.500 --> 00:08:58.860 metric tons for a gigawatt. And that's outrageous if you're 177 00:08:58.899 --> 00:09:01.830 lifting them in 20-ton chunks. But if you're lifting them 178 00:09:01.970 --> 00:09:05.250 in 100-ton chunks... It starts to look practical. And speaking 179 00:09:05.269 --> 00:09:08.990 of that 100-ton lift, in April, Starship test flight number one, 180 00:09:09.049 --> 00:09:11.580 the most powerful rocket ever launched. This was the V-1. Today, 181 00:09:11.620 --> 00:09:15.080 we're dealing with V-3s. It didn't go particularly well, but 182 00:09:15.100 --> 00:09:18.580 it was a full end-to-end stack. They launched. The separation 183 00:09:18.620 --> 00:09:21.440 went a little bit wonky. Everything was terminated after about 184 00:09:21.480 --> 00:09:25.740 four minutes worth of flight. But this gigantic 9-meter wide, 185 00:09:25.759 --> 00:09:30.919 120-meter tall stainless steel rocket took off. More thrust than 186 00:09:31.000 --> 00:09:35.820 Apollo by Saturn V by a lot. And yeah, they're 187 00:09:35.860 --> 00:09:38.509 on their path. It's advancing to this day. 188 00:09:38.809 --> 00:09:41.480 Was that... That wasn't the one that sort of blew 189 00:09:41.559 --> 00:09:44.240 up on the launch pad and toasted everything around. 190 00:09:43.960 --> 00:09:44.399 It, right? 191 00:09:44.559 --> 00:09:44.820 No. 192 00:09:45.419 --> 00:09:49.299 Well, actually, what it was was it lost three engines 193 00:09:49.559 --> 00:09:53.429 at launch. And so, it lifted off very, very slowly, 194 00:09:53.490 --> 00:09:56.639 literally as it burned fuel lost weight. So it dug 195 00:09:56.659 --> 00:10:03.049 a 20-foot deep hole. Blowing concrete everywhere. 196 00:10:03.070 --> 00:10:03.169 Yeah. 197 00:10:03.250 --> 00:10:07.590 It destroyed, it really seriously damaged the pad. Destroyed the pad, yeah. Arguably, 198 00:10:07.669 --> 00:10:10.230 when they had three engines fail, they should not have 199 00:10:10.269 --> 00:10:11.990 attempted to launch it. It should have cut off immediately, 200 00:10:12.090 --> 00:10:16.409 but they didn't. And so, they also, this is the 201 00:10:16.929 --> 00:10:20.529 milk stool design. So, they just had legs to, they 202 00:10:20.580 --> 00:10:23.870 had no deflectors. They had no water suppression system. None 203 00:10:23.909 --> 00:10:25.230 of the normal things you would do with a rocket 204 00:10:25.269 --> 00:10:27.889 like this. They did not do that again. In the 205 00:10:27.929 --> 00:10:29.639 process of repairing the pad, they did put in a 206 00:10:29.649 --> 00:10:33.379 water suppression system. And subsequently, today, they have a proper 207 00:10:33.779 --> 00:10:37.700 full flame trench with water suppression redirecting the exhaust. 208 00:10:37.720 --> 00:10:38.320 What were they thinking? 209 00:10:38.340 --> 00:10:40.679 This massive rocket. He wanted to make it as simple 210 00:10:40.720 --> 00:10:43.379 as possible for flying from Mars. That was what he 211 00:10:43.419 --> 00:10:48.259 was thinking, which is dumb, dumb, dumb, dumb. Anyway, the 212 00:10:48.299 --> 00:10:54.669 good news is they learned. In August– The Indians, the 213 00:10:54.690 --> 00:10:59.529 Indian space program, Chang'e-N-3 successfully does a soft landing of 214 00:10:59.549 --> 00:11:02.580 their lander near the moon's south pole. This will be 215 00:11:02.600 --> 00:11:04.620 the only fourth country to ever land on the moon. 216 00:11:04.820 --> 00:11:06.899 The first to be in the Americans will actually be. 217 00:11:07.360 --> 00:11:11.059 The Soviet Union, then the Americans, and then China, but 218 00:11:11.120 --> 00:11:13.059 now India was up there as well, and also the 219 00:11:13.100 --> 00:11:16.159 first to land anywhere near the South Pole. That same month, 220 00:11:16.220 --> 00:11:18.879 Russia attempts to return to the moon with their Luna 25. 221 00:11:18.879 --> 00:11:25.399 It crashes. In September, OSIRIS-REx, this is a US mission 222 00:11:25.480 --> 00:11:28.799 to collect samples of the asteroid Bennu, returns those samples 223 00:11:28.860 --> 00:11:33.750 successfully to Earth. And in October, Falcon Heavy launches the 224 00:11:33.820 --> 00:11:38.220 Psyche mission. This is a mission to the largest object 225 00:11:38.440 --> 00:11:42.519 in the asteroid belt.

Psyche is a metallic asteroid, possibly 226 00:11:42.580 --> 00:11:45.700 the core of a planet that was ripped apart by 227 00:11:45.720 --> 00:11:48.190 the tidal forces caused by Jupiter. And they want to 228 00:11:48.289 --> 00:11:50.529 seriously map that. That mission is still underway. It's still flying. 229 00:11:51.009 --> 00:11:54.850 Second test flight of Starship in November. And that's all 230 00:11:54.889 --> 00:11:56.710 I got for that. So it's not a big, I mean, 231 00:11:56.730 --> 00:11:59.009 there's a lot of missions, but they're mostly Starlink missions, right? 232 00:11:59.519 --> 00:12:02.840 Shall we talk about computing? Because it's 2023. Yeah. And 233 00:12:02.860 --> 00:12:06.399 this is when AI takes over the conversation. Yeah. 234 00:12:06.440 --> 00:12:06.679 Right. 235 00:12:07.120 --> 00:12:10.419 We had ChatGPT launched in November of 22. It is 236 00:12:10.480 --> 00:12:13.210 now the beginning of 23. This is the first time 237 00:12:14.169 --> 00:12:16.970 Sachin Adela ever writes an all-hands letter the way Bill 238 00:12:17.009 --> 00:12:20.769 Gates did, where he says, look, we've got this agreement 239 00:12:20.789 --> 00:12:25.450 with OpenAI. We've invested an additional $ 10 billion in them, 240 00:12:25.509 --> 00:12:29.240 bringing their total investment to $ 13 billion. And every team 241 00:12:29.379 --> 00:12:32.899 should take a look at these APIs and start immigrating 242 00:12:32.919 --> 00:12:36.259 them into their applications and make a co-pilot. I've heard 243 00:12:36.299 --> 00:12:39.580 rumor there were literally hundreds. But, you know, it's a 244 00:12:39.720 --> 00:12:41.740 first for Satya to direct the whole company in a 245 00:12:41.779 --> 00:12:44.580 new direction. And really the thing that I think he 246 00:12:44.620 --> 00:12:47.159 thinks he can hang his hat on as CEO, that 247 00:12:47.210 --> 00:12:49.309 this is what he brings to the company. First out 248 00:12:49.330 --> 00:12:52.509 of the gate in February will be Bing, formerly known 249 00:12:52.549 --> 00:12:54.269 as Project Sydney, where they were trying to put a 250 00:12:54.309 --> 00:12:56.889 natural interface onto the search engine. I hear both guys 251 00:12:56.909 --> 00:13:00.289 who tried it really liked it. I said that a lot. 252 00:13:00.330 --> 00:13:02.269 I feel, I feel bad. I mean, yeah. You know, 253 00:13:02.289 --> 00:13:04.779 it turns out that just putting an LLM onto your 254 00:13:04.799 --> 00:13:06.779 search engine doesn't make it a popular search engine, but 255 00:13:06.840 --> 00:13:07.879 you know, it was first out the gate. 256 00:13:08.960 --> 00:13:09.080 Right. 257 00:13:09.220 --> 00:13:12.429 In March we get GPT four. So we go from what? 258 00:13:12.429 --> 00:13:16.789 275 billion parameters up to a trillion parameters. Uh, It's 259 00:13:16.830 --> 00:13:19.490 amazing it happened that fast, you know, that less than 260 00:13:19.529 --> 00:13:22.980 a year after GPT-3, they had GPT-4. This is when 261 00:13:23.000 --> 00:13:25.639 the 10 billion bought them, was to build an incredibly 262 00:13:25.700 --> 00:13:29.000 massive model. And it's impressive, the step forward on that. This, 263 00:13:29.039 --> 00:13:32.039 of course, freaks out Google, who had been playing with 264 00:13:32.120 --> 00:13:34.360 BARD and had Google Brain and all those sorts of things. 265 00:13:34.379 --> 00:13:37.379 They're actually, internally, they called it a code red event. 266 00:13:38.200 --> 00:13:41.740 This becomes the big push for Gemini, right? In May, 267 00:13:42.519 --> 00:13:46.750 Jeffrey Hinton said, One of the godfathers of the generative 268 00:13:46.840 --> 00:13:49.820 AI movement leaves Google and goes on a press tour 269 00:13:49.840 --> 00:13:51.399 that he's still on to this day talking about the 270 00:13:51.440 --> 00:13:52.120 dangers of AI. 271 00:13:52.159 --> 00:13:53.320 He basically says, stop. 272 00:13:53.419 --> 00:13:57.159 Yeah. Well, now that all of my stock is vested 273 00:13:57.559 --> 00:13:59.980 and I get to keep my millions, now let's talk 274 00:14:00.000 --> 00:14:01.299 about how problematic it is. 275 00:14:01.480 --> 00:14:01.840 Yeah. 276 00:14:02.240 --> 00:14:06.480 Yeah. In April, Meta releases an open source large language 277 00:14:06.519 --> 00:14:12.120 model made by Meta AI, aka Lama. One of the 278 00:14:12.139 --> 00:14:14.879 first open source models out of the gate. And by July, 279 00:14:14.899 --> 00:14:17.779 there'll be Llama 2. And it really hits pretty hard 280 00:14:17.850 --> 00:14:19.669 because if you're going to start thinking local models, there 281 00:14:19.710 --> 00:14:22.649 it is. This puts a lot of pressure on everybody else. 282 00:14:22.669 --> 00:14:25.169 So by September, you have Amazon out there with Bedrock 283 00:14:25.230 --> 00:14:28.750 and Titan. Again, I think both guys have used that 284 00:14:28.809 --> 00:14:32.919 one too. But, you know, all of that, you recognize 285 00:14:32.960 --> 00:14:35.200 that the 23, everybody has to play, right? So we've 286 00:14:35.240 --> 00:14:39.960 got all the hyperscalers on board now and we're in motion. 287 00:14:40.340 --> 00:14:45.539 to try and grow something with all this technology. October, 288 00:14:45.610 --> 00:14:49.250 Microsoft completes the acquisition of Activision Blizzard for $ 68. 7 billion. 289 00:14:49.269 --> 00:14:51.809 It takes over a year, lots of battles with the 290 00:14:51.929 --> 00:14:54.470 FTC and so forth, and pretty much nothing has come 291 00:14:54.529 --> 00:14:57.009 of that. For the most part, somebody cashed in big, 292 00:14:57.029 --> 00:15:00.399 but goodness knows what Microsoft's strategy for gaming actually is. 293 00:15:01.559 --> 00:15:04.039 All right, now we get to the fun month in 294 00:15:04.039 --> 00:15:07.279 2023 when it comes to tech, and it's in November. Obviously, 295 00:15:07.299 --> 00:15:11.600 we get. NET Conf. with. NET 8, a long-term support

296 00:15:11.639 --> 00:15:13.480 version of. NET, a very good one too. We're a 297 00:15:13.519 --> 00:15:16.919 big push with native AOT, C Sharp 12, a lot 298 00:15:16.960 --> 00:15:19.980 of cloud-native tooling. And the first announcement of what at 299 00:15:20.019 --> 00:15:24.740 that time will be known as. NET Aspire. And our 300 00:15:24.820 --> 00:15:29.740 friend David Fowler, now distinguished engineer David Fowler, driving this, 301 00:15:30.360 --> 00:15:32.059 what we now refer to, we didn't know how to 302 00:15:32.100 --> 00:15:34.960 talk about at the time, about the scaffolding for building 303 00:15:34.980 --> 00:15:35.710 cloud-native with. 304 00:15:35.659 --> 00:15:35.840 NET. 305 00:15:35.940 --> 00:15:39.250 Of course, it'll advance a lot more. And a week 306 00:15:39.610 --> 00:15:42.870 after that event, or not even a week, the weekend 307 00:15:42.950 --> 00:15:47.970 of that event, on Friday, November 17th, the OpenAI board 308 00:15:48.509 --> 00:15:50.769 puts out a press release where they say, Sam Altman will. 309 00:15:50.649 --> 00:15:53.090 Be departing as CEO. And so began the drama. 310 00:15:53.509 --> 00:15:57.769 And Mira Muriati, the CTO, is appointed as interim CEO. 311 00:15:59.529 --> 00:16:02.549 The chairman of the board for OpenAI, a guy named 312 00:16:02.590 --> 00:16:05.509 Greg Brockman, who's also been part of the originators on this, 313 00:16:05.529 --> 00:16:08.230 he's an OpenAI president. He is removed as chairman and 314 00:16:08.309 --> 00:16:15.120 resigns in solidarity with Altman. On Saturday, by all accounts, 315 00:16:15.200 --> 00:16:18.279 all hell breaks loose. A very angry Satya Nadella who 316 00:16:18.320 --> 00:16:20.860 did not see this coming. Nobody thought to notify the 317 00:16:20.899 --> 00:16:24.909 guy who kicked $ 13 billion into your company that, hey, 318 00:16:24.950 --> 00:16:30.289 we're going to fire the CEO. This leads to on 319 00:16:30.429 --> 00:16:34.190 Sunday where the board announces a different interim CEO, Twitch 320 00:16:34.230 --> 00:16:39.200 CEO Emmett Shearer. After only two days of Marati, I 321 00:16:39.240 --> 00:16:41.220 think she sort of was backing away from all of this. 322 00:16:41.840 --> 00:16:44.159 This is the day that Satya announces that Sam Altman 323 00:16:44.179 --> 00:16:46.460 and Greg Brockman are going to join Microsoft and form 324 00:16:46.580 --> 00:16:47.799 a new AI group. 325 00:16:48.139 --> 00:16:48.600 I remember. 326 00:16:48.679 --> 00:16:51.149 Which leads to, on Monday, where 700 of 770 employees 327 00:16:51.169 --> 00:17:00.470 of OpenAI, including– One Ilya Sutskiver, who had publicly expressed 328 00:17:00.509 --> 00:17:02.740 a regret for his role in firing Sam Altman, also 329 00:17:02.769 --> 00:17:05.519 signs this letter where they say they demand the resignation 330 00:17:05.559 --> 00:17:09.059 of the board and will consider following Altman to Microsoft. 331 00:17:09.079 --> 00:17:11.839 So essentially the entire OpenAI says, we're going to go 332 00:17:11.859 --> 00:17:16.809 to Microsoft. This results in, on Tuesday, Sam Altman returning 333 00:17:16.869 --> 00:17:20.329 as CEO and the entire board being replaced by Wednesday. 334 00:17:20.390 --> 00:17:21.849 It's like a soap opera, isn't it? 335 00:17:21.970 --> 00:17:22.970 It's insane. 336 00:17:23.130 --> 00:17:23.829 It really is. 337 00:17:23.990 --> 00:17:26.930 Yeah. And then look, I think everybody knows Sam Altman's 338 00:17:26.990 --> 00:17:29.630 not a good man. Right. But boy, you sure screwed 339 00:17:29.730 --> 00:17:30.299 this up. 340 00:17:30.390 --> 00:17:30.680 Yeah. 341 00:17:30.980 --> 00:17:33.859 Like, there was a way to do this that might 342 00:17:33.880 --> 00:17:37.940 have worked. This was not that way. And fundamentally, you know, 343 00:17:38.099 --> 00:17:41.309 it cemented Sam Altman's control of the company in the process. Like, 344 00:17:41.349 --> 00:17:44.170 he ultimately benefited from this and turned OpenAI from its 345 00:17:45.220 --> 00:17:47.859 Original idea, which I don't know how good that idea 346 00:17:47.920 --> 00:17:50.319 ever was of doing AI in the open, which they 347 00:17:50.519 --> 00:17:55.119 very quickly diverted from into just another for-profit company that's 348 00:17:55.319 --> 00:17:58.359 racing to try to IPO. Little goodness knows if that's 349 00:17:58.420 --> 00:18:01.099 ever going to happen. Yeah. Last but not least, I 350 00:18:01.119 --> 00:18:04.599 think I've been mentioning quantum computing in the past few 351 00:18:04.640 --> 00:18:06.619 of these just because we had an interesting wave of 352 00:18:06.660 --> 00:18:10.829 quantum computing in the late 2010s, beginning of 2020s. I 353 00:18:10.849 --> 00:18:12.710 sort of see 2023 as sort of the end of 354 00:18:12.769 --> 00:18:14.680 the road of the race to the larger qubit at 355 00:18:14.700 --> 00:18:20.500 the time with IBM's Condor processor, a thousand qubits. Now 356 00:18:20.539 --> 00:18:22.339 we sort of have this debate about what a physical 357 00:18:22.440 --> 00:18:25.089 versus a logical qubit is actually because there's so much 358 00:18:25.269 --> 00:18:28.589 error correction that needs to happen that Under the current 359 00:18:28.630 --> 00:18:31.150 model at that time, they said, we're probably not going 360 00:18:31.170 --> 00:18:34.450 to useable work until we get into possibly millions of qubits. 361 00:18:35.009 --> 00:18:37.750 And that'll put a lot of pressure on this error 362 00:18:37.809 --> 00:18:40.839 correction problem for quantum computing, which we'll see more of 363 00:18:41.079 --> 00:18:41.920 in subsequent years. 364 00:18:42.220 --> 00:18:42.420 Yeah. 365 00:18:42.440 --> 00:18:43.059 And that's what I got. 366 00:18:43.460 --> 00:18:48.200 All right. As again, Jeremy Miller's here. Jeremy, you can 367 00:18:48.240 --> 00:18:51.329 jump in anytime you want, as you have been doing. 368 00:18:51.940 -->

00:18:54.220 But right now we're going to do better know framework. 369 00:18:54.460 --> 00:19:04.990 Awesome. Roll the music. All right, man. What is Simon 370 00:19:05.009 --> 00:19:05.769 Kropp up to? 371 00:19:06.029 --> 00:19:08.730 Yeah, I know. We should just call it Simon Kropp. 372 00:19:08.769 --> 00:19:14.539 Better know Simon Kropp. This is EF query complexity. 373 00:19:14.700 --> 00:19:15.180 What is this? 374 00:19:15.400 --> 00:19:20.549 So it detects overly complex entity framework core queries. and 375 00:19:20.609 --> 00:19:24.490 either logs them or throws. Each check has two levels. 376 00:19:25.049 --> 00:19:28.049 The level a query is logged at, which has defaults, 377 00:19:28.289 --> 00:19:33.750 and the level a query throws at, which is opt-in. So, yeah. So, 378 00:19:34.069 --> 00:19:34.309 you know. 379 00:19:34.329 --> 00:19:36.609 So this is just a, if you've written a link 380 00:19:36.650 --> 00:19:40.750 expression so complicated for EF that it's going to make 381 00:19:40.809 --> 00:19:42.690 a bad query, this will give you a heads up? 382 00:19:42.789 --> 00:19:45.630 Yeah, before calling it and hanging your database. 383 00:19:46.089 --> 00:19:46.309 Yeah. 384 00:19:47.690 --> 00:19:49.069 You might want to run it through this. 385 00:19:49.390 --> 00:19:51.599 Yeah, no, that's true. a great idea god of course 386 00:19:51.779 --> 00:19:54.180 you're so clever i know you know you know he 387 00:19:54.220 --> 00:19:56.660 got here the hard way yeah that he wrote he 388 00:19:56.680 --> 00:19:59.170 wrote a link expression that pinned the database to the 389 00:19:59.210 --> 00:20:00.950 wall for i don't know maybe a day and he's 390 00:20:00.990 --> 00:20:01.890 like hmm. 391 00:20:01.940 --> 00:20:04.460 Let's give him a little more credit. He probably inherited 392 00:20:04.500 --> 00:20:05.559 someone else's code base. 393 00:20:05.940 --> 00:20:07.000 You're probably right. 394 00:20:07.039 --> 00:20:07.359 Fair. 395 00:20:07.599 --> 00:20:09.720 You're probably exactly right. He probably didn't write the query. 396 00:20:09.799 --> 00:20:11.440 He probably didn't stop this from happening again. 397 00:20:12.039 --> 00:20:12.259 Yeah. 398 00:20:12.539 --> 00:20:15.119 Well, anyway, that's it. That's what I got. Who is 399 00:20:15.180 --> 00:20:16.299 talking to us today, Richard? 400 00:20:16.319 --> 00:20:19.539 Let's grab a comment off of show 1860, the one 401 00:20:19.559 --> 00:20:21.660 we did with one Jeremy Miller. Never heard of him. 402 00:20:21.700 --> 00:20:23.680 Talking a little bit about minimal architecture. 403 00:20:23.740 --> 00:20:24.619 Right before the Civil War. 404 00:20:24.640 --> 00:20:25.119 There we go. 405 00:20:25.180 --> 00:20:25.980 Yeah. 406 00:20:27.420 --> 00:20:28.720 Now you're in the groove, Jeremy. 407 00:20:28.769 --> 00:20:29.289 You get it. 408 00:20:30.450 --> 00:20:32.819 Imagine if we started this in the 1800s. It was like, 409 00:20:32.839 --> 00:20:34.960 I think it was like, you should go back and listen, 410 00:20:34.980 --> 00:20:37.720 but I think it was like in episode 1914. I'm like, hey, 411 00:20:37.740 --> 00:20:40.299 start a quote World War I. Right. The good news 412 00:20:40.400 --> 00:20:43.920 is it's almost over. A couple more shows. This comment 413 00:20:43.940 --> 00:20:46.339 comes from Trevor. He said, I love this discussion. Enjoyed 414 00:20:46.359 --> 00:20:50.880 the comments on microservices versus monoliths. I got pushed heavily 415 00:20:50.900 --> 00:20:53.400 into microservices approach with a product we had built. and 416 00:20:53.420 --> 00:20:56.059 we're re-architecting it to microservices, and it was the worst 417 00:20:56.200 --> 00:21:00.119 mistake ever. Things just became more complicated. It was harder 418 00:21:00.140 --> 00:21:03.160 to maintain. It added a lot of latency and security issues, 419 00:21:03.200 --> 00:21:06.759 and the complexity was just not worth it. So I 420 00:21:06.779 --> 00:21:13.660 came up with a new acronym, appropriately side service, or ASS. 421 00:21:13.680 --> 00:21:16.619 I 100% believe in services, separation of concerns, and clean architectures, 422 00:21:16.660 --> 00:21:18.690 but the approach must be appropriate to the complexity of 423 00:21:18.710 --> 00:21:23.230 the solution and the size of the team. It makes 424 00:21:23.349 --> 00:21:25.289 no sense to have 100 separate services with a team 425 00:21:25.329 --> 00:21:27.470 of 10 people. And then it also makes no sense 426 00:21:27.490 --> 00:21:29.700 to have one massive single deployment with a code base 427 00:21:29.799 --> 00:21:32.480 for a team of 250 people. The services need to 428 00:21:32.539 --> 00:21:35.240 work with the cognitive load and be appropriate to the 429 00:21:35.279 --> 00:21:38.799 organization and team structures. Love any discussion. But on this 430 00:21:38.940 --> 00:21:41.869 one point, stop making the CTO out to be the 431 00:21:41.930 --> 00:21:51.309 bad guy. Love from Trevor, CTO. Maybe we were teasing 432 00:21:51.450 --> 00:21:55.930 on the non-technical CTO. Clearly, Trevor's not that guy. He's 433 00:21:55.970 --> 00:21:58.269 the one who's actually trying to get things right and 434 00:21:58.609 --> 00:22:00.210 be part of the team and so forth. So, sorry 435 00:22:00.250 --> 00:22:02.349 about teasing you. Too mean to you there, Trevor. But 436 00:22:02.789 --> 00:22:06.430 we're with you. Appropriate size service is correct. So, thank 437 00:22:06.450 --> 00:22:08.319 you so much for your comments. And CopyMuse2CodeBuy is on 438 00:22:08.339 --> 00:22:09.500 its way to you. And if you'd like a copy 439 00:22:09.519 --> 00:22:12.960 of Muse2CodeBuy, write a comment on the website at. netrocks. com 440 00:22:12.220 --> 00:22:13.900 or on the Facebooks. We publish every show there. And 441 00:22:13.920 --> 00:22:15.519 if you comment there and I read it in the show, 442 00:22:15.539 --> 00:22:16.579 we'll send you a copy of

Muse2CodeBuy. 443 00:22:16.940 --> 00:22:18.640 And if you want to just get Music to Code By, 444 00:22:18.700 --> 00:22:21.359 which will help you stay in a state of flow 445 00:22:21.680 --> 00:22:24.819 to be calm and all that stuff while you're writing code, 446 00:22:25.500 --> 00:22:31.349 go to musictocodeby.net and just get it. MP3s, waves, and 447 00:22:31.509 --> 00:22:36.730 FLAC formats. All right. And that voice that you heard before, 448 00:22:36.789 --> 00:22:40.460 that third voice, was that of Jeremy Miller. He's been on. 449 00:22:40.390 --> 00:22:43.390 NET Rocks many times. He started his career as a 450 00:22:43.490 --> 00:22:47.799 quote-unquote real engineer. but wandered into software because that looked 451 00:22:47.839 --> 00:22:51.240 like more fun. Since then, Jeremy has worked in and 452 00:22:51.319 --> 00:22:56.829 led software development teams in the computer manufacturing industry, finance, insurance, healthcare, 453 00:22:56.869 --> 00:23:00.910 and banking industries. Lately, Jeremy has been focused on leading 454 00:23:01.009 --> 00:23:06.650 software architecture teams and helping mentor other software architects. Having 455 00:23:06.670 --> 00:23:10.650 had roles both as an in-house software architect and as 456 00:23:10.690 --> 00:23:13.309 a software consultant, Jeremy has a great deal of insight 457 00:23:13.329 --> 00:23:17.349 into the challenges that confront companies developing and maintaining enterprise 458 00:23:17.369 --> 00:23:20.549 systems over time. He's well known for his open source 459 00:23:20.569 --> 00:23:24.750 software tools, starting with StructureMap and continuing today to Martin 460 00:23:24.849 --> 00:23:28.490 and Wolverine. Jeremy is also a frequent author and technical 461 00:23:28.509 --> 00:23:32.750 speaker at software conferences and can be found at jeremydmiller.com. 462 00:23:32.930 --> 00:23:33.349 Welcome back. 463 00:23:33.450 --> 00:23:33.930 Hey, all. 464 00:23:34.089 --> 00:23:35.630 Thanks, guys. Thanks for having me on again. 465 00:23:35.869 --> 00:23:38.150 Yeah, great to have you back. Great to have you back. 466 00:23:38.299 --> 00:23:39.450 There was a comment on another show that I was 467 00:23:39.490 --> 00:23:44.500 very tempted to read where it mentioned alt.net. Just a 468 00:23:44.539 --> 00:23:47.109 callback to The first show we did with you, which 469 00:23:47.130 --> 00:23:50.430 was in 2008, was the Alt.net show. 470 00:23:50.569 --> 00:23:50.769 Yeah. 471 00:23:50.950 --> 00:23:52.170 It was a different day and time. 472 00:23:52.289 --> 00:23:52.690 Certainly. 473 00:23:52.930 --> 00:23:55.950 So if we're ready to be, let's just start with 474 00:23:56.009 --> 00:24:00.069 a get off my lawn kind of moment. One thing 475 00:24:00.170 --> 00:24:05.009 I've thought pretty recently. So after Twitter became, you know, 476 00:24:05.029 --> 00:24:09.730 the Nazi infested Star Wars, the Star Wars cantina theme 477 00:24:09.779 --> 00:24:10.279 it is now. 478 00:24:10.299 --> 00:24:10.859 Yeah. 479 00:24:14.450 --> 00:24:17.309 You know, the technical conversations kind of went everywhere, but 480 00:24:17.349 --> 00:24:19.710 it seems to be mostly on LinkedIn, or at least 481 00:24:19.950 --> 00:24:24.180 as far as I can tell, weirdly, right? One thing 482 00:24:24.319 --> 00:24:26.640 I've noticed, I think it's a gripe, and maybe it's 483 00:24:26.660 --> 00:24:30.339 just the algorithm. What I see are the. NET content creators. 484 00:24:31.559 --> 00:24:34.759 The people are the. NET content creators that are the 485 00:24:34.799 --> 00:24:37.000 same age that I was back in the all. 486 00:24:37.160 --> 00:24:37.619 NET days. 487 00:24:37.640 --> 00:24:41.750 We were big bloggers on CodeBetter. It is very narrowly 488 00:24:41.869 --> 00:24:47.309 focused on a handful of topics. It's very narrowly mainstream. 489 00:24:47.349 --> 00:24:50.869 They're not expanding the scope of software development. They aren't 490 00:24:50.990 --> 00:24:54.630 trying to innovate. It's here's how to use EF Core 491 00:24:54.789 --> 00:24:59.230 to do modular monoliths. So interesting. We were better in 492 00:24:59.269 --> 00:25:02.670 the all.net days than the content creators are now. Just 493 00:25:02.690 --> 00:25:05.109 a little bit of some trivia first. You know, you 494 00:25:05.130 --> 00:25:07.910 call that Simon Kropp. I do want to call him out. 495 00:25:08.329 --> 00:25:13.160 If you've ever looked at any critter stack related documentation website, 496 00:25:13.579 --> 00:25:16.500 you'll see his works. We use markdown snippets. That's a 497 00:25:16.500 --> 00:25:19.539 really valuable way to create living technical documentation. 498 00:25:19.779 --> 00:25:19.960 Right. 499 00:25:20.799 --> 00:25:24.269 And then just on your theme, the history theme, you know, 500 00:25:24.289 --> 00:25:28.569 you were talking about 2023. We missed a couple really 501 00:25:28.609 --> 00:25:35.559 important things. I founded Jasper FX in June of 2023. 502 00:25:35.559 --> 00:25:37.880 Went solo for the first time at the tender age 503 00:25:37.920 --> 00:25:39.960 of 49. 504 00:25:40.509 --> 00:25:41.130 Never too late. 505 00:25:42.309 --> 00:25:44.750 Well, and that's the lesson I think I want. I 506 00:25:44.789 --> 00:25:46.829 want everybody to learn, but also do it while you're younger. 507 00:25:47.289 --> 00:25:49.750 And we'll bring 1.0 in 2023. Yeah. 508 00:25:49.930 --> 00:25:55.160 Yeah. Awesome. Well, congratulations. It's been a fun few years 509 00:25:55.200 --> 00:25:57.059 and we've done some interesting shows around it just because, 510 00:25:57.099 --> 00:26:00.740 you know, we've really built out a lot of tools 511 00:26:00.819 --> 00:26:03.160 to make it easier for us. And if Simon likes them, 512 00:26:03.630 --> 00:26:04.869 That's a heck of a compliment, man. 513 00:26:05.130 --> 00:26:07.950 I'll take it. I'll put words in his mouth. Simon's 514 00:26:07.990 --> 00:26:09.710 also a

contributor to the Critter stack. 515 00:26:09.829 --> 00:26:11.160 That's awesome. Yeah, I love it. 516 00:26:11.410 --> 00:26:12.789 So new Critters? 517 00:26:13.509 --> 00:26:17.230 Yeah. Again, I mean, AI is going to drop in 518 00:26:17.269 --> 00:26:20.970 here sooner or later. AI has allowed us to go 519 00:26:20.990 --> 00:26:23.130 a lot faster to deal with a lot of ideas 520 00:26:23.170 --> 00:26:25.170 that have been out there for a long time. Interesting. 521 00:26:25.190 --> 00:26:25.890 And that's real. 522 00:26:26.730 --> 00:26:30.359 So I'm still going to say Martin, based on Postgres, 523 00:26:31.039 --> 00:26:36.720 is still the most robust Critter. widely used event sourcing 524 00:26:36.980 --> 00:26:37.980 solution in all of. 525 00:26:38.109 --> 00:26:38.369 NET. 526 00:26:40.549 --> 00:26:43.490 Still going to say it's the most capable. But we've 527 00:26:43.529 --> 00:26:46.150 been told for over a decade that, hey, you know what? 528 00:26:46.170 --> 00:26:48.759 It would be much more successful if it was based 529 00:26:48.819 --> 00:26:52.700 on SQL Server instead. And I've always said, I don't 530 00:26:52.740 --> 00:26:55.720 have time. That's too hard. And SQL Server doesn't have 531 00:26:56.430 --> 00:26:58.710 SQL Server is way behind in JSON support. 532 00:26:58.970 --> 00:26:59.450 Sure. 533 00:26:59.609 --> 00:27:02.289 Now we have SQL Server 2025. It has a native 534 00:27:02.329 --> 00:27:06.339 JSON type. It has its own capability to do vector support. 535 00:27:06.380 --> 00:27:08.960 It has a lot of the same capabilities that Postgres has. 536 00:27:09.500 --> 00:27:11.160 So now we have a tool called Polecat. 537 00:27:11.240 --> 00:27:11.859 Polecat. 538 00:27:11.880 --> 00:27:18.200 That's another critter that has probably 80% to 90% of 539 00:27:18.259 --> 00:27:21.960 the functionality of Martin and at this point probably shares 540 00:27:21.980 --> 00:27:24.710 a lot of the same a lot of the same 541 00:27:24.750 --> 00:27:29.759 infrastructure and, um, compliance tests of behavior. So it gives 542 00:27:29.779 --> 00:27:32.859 you the same kind of document database capability right on 543 00:27:32.900 --> 00:27:36.299 top of SQL server. So you guys can stop wasting 544 00:27:36.359 --> 00:27:39.579 so much time with EF core mapping and, you know, 545 00:27:39.680 --> 00:27:43.650 to the crazy EF core link query is too complex. 546 00:27:44.329 --> 00:27:47.509 Using event sourcing with projections, you can project to exactly 547 00:27:47.529 --> 00:27:52.009 the view you want. You can sidestep silly queries, just saying, um, 548 00:27:53.660 --> 00:27:57.349 So we have Polecat on top of SQL Server, SQL 549 00:27:57.390 --> 00:28:01.130 Server 2025. And just a little more recently, we've added 550 00:28:01.170 --> 00:28:05.109 another library called Fisher that gives you the same functionality 551 00:28:05.250 --> 00:28:07.059 on top of SQLite. 552 00:28:07.259 --> 00:28:08.279 Is that like FisherCat? 553 00:28:08.440 --> 00:28:09.319 Yeah. 554 00:28:09.880 --> 00:28:10.099 Yeah. 555 00:28:10.160 --> 00:28:12.440 There's a lot of mustelids in this conversation, let me 556 00:28:12.460 --> 00:28:12.759 tell you. 557 00:28:12.940 --> 00:28:15.700 Exactly. And that's mostly the theme. 558 00:28:16.299 --> 00:28:17.900 Yes. 559 00:28:17.940 --> 00:28:20.349 When we first moved into our house, we were told 560 00:28:20.410 --> 00:28:22.609 that the woods out back were full of FisherCats. 561 00:28:23.069 --> 00:28:23.230 Yeah. 562 00:28:23.450 --> 00:28:26.269 Oh, that's awesome. I've never seen one in real life. 563 00:28:26.329 --> 00:28:27.910 The dangerous little buggers. Of course. 564 00:28:28.190 --> 00:28:30.190 That was the day my children vowed never to go 565 00:28:30.329 --> 00:28:31.309 for a walk in the woods. 566 00:28:32.309 --> 00:28:37.130 So, they're supposed to look just like Martins, but bigger, right? Mm-hmm. Yeah. Cool. 567 00:28:37.150 --> 00:28:38.930 And they're hyper-aggressive. They are. 568 00:28:38.950 --> 00:28:39.950 Vicious. 569 00:28:40.289 --> 00:28:42.690 We've got a mink hanging around here right now. And 570 00:28:43.029 --> 00:28:44.990 he scares me more than the otters that are five 571 00:28:45.039 --> 00:28:51.460 times his size. They'll give you. They'll go. They're looking 572 00:28:51.480 --> 00:28:51.940 for trouble. 573 00:28:52.380 --> 00:28:52.500 Huh. 574 00:28:52.880 --> 00:28:53.920 All right. So Fisher. 575 00:28:54.460 --> 00:28:55.900 Sorry, guys. There's actually more. 576 00:28:56.319 --> 00:28:59.589 Oh, yeah. Okay. So Fisher is SQLite. 577 00:28:59.890 --> 00:29:00.410 SQLite. 578 00:29:00.430 --> 00:29:02.609 So you started on Postgres. You've done SQL Server 2025 579 00:29:02.609 --> 00:29:06.750 because of the JSON. type. And now you are also 580 00:29:06.789 --> 00:29:09.970 got it over to SQL light as well, which is cool. 581 00:29:10.150 --> 00:29:12.339 So that is also helping us with some of our 582 00:29:12.380 --> 00:29:15.000 own products. We have a commercial tool called critter watch, 583 00:29:15.640 --> 00:29:20.160 you know, that's the inevitable management console. Um, but it's 584 00:29:20.200 --> 00:29:25.420 also an MCP hub that connects connected to your AI 585 00:29:25.500 --> 00:29:28.420 tools and it will give you complete visibility into a 586 00:29:29.160 --> 00:29:32.319 static view of your application and a dynamic view as well. 587 00:29:33.279 --> 00:29:36.150 It'll give you access to alerts. It'll even call out 588 00:29:36.210 --> 00:29:41.200 to LLMs to maybe be able to take proactively take 589 00:29:41.279 --> 00:29:46.559 amelioration act actions, replay things off, off of dead letter cues. 590 00:29:46.640 --> 00:29:48.950 Maybe it can be smart enough to know, Hey, I 591 00:29:48.990 --> 00:29:52.589 need to go reboot the server, something like that potential. 592 00:29:52.750 --> 00:29:52.990 Wow. 593 00:29:53.250 --> 00:29:55.880 And then,

Sorry to keep overloading you. We have a 594 00:29:55.980 --> 00:30:01.319 new tool that's not quite 1.0 called Bobcat. Bobcat? It's 595 00:30:01.339 --> 00:30:06.000 been hanging around long enough that it predates the Critterstack naming. 596 00:30:06.019 --> 00:30:09.079 I always thought they were cool, and I've got a 597 00:30:09.079 --> 00:30:12.380 rural background, so Bobcat means a steer skid to me. 598 00:30:12.769 --> 00:30:15.190 That's just kind of a thing you use to do 599 00:30:15.289 --> 00:30:19.490 crazy hard jobs. So Bobcat is going to be our 600 00:30:19.549 --> 00:30:25.420 tool for spec-driven development and integration testing. has a Gherkin function, 601 00:30:25.559 --> 00:30:27.380 but I mean, I don't know if anybody's going to 602 00:30:27.420 --> 00:30:28.960 care about that. Yeah. 603 00:30:29.160 --> 00:30:30.960 And it's a feline. So, you know, you're cheating. 604 00:30:32.440 --> 00:30:35.460 Theoretically our neighbor says that there's a Bobcat that comes 605 00:30:35.529 --> 00:30:38.140 out in our neighbor, our, our yard at night, but 606 00:30:38.180 --> 00:30:38.900 I've never seen. Yeah. 607 00:30:39.269 --> 00:30:40.450 They're yeah. They're ghosts. 608 00:30:40.730 --> 00:30:42.230 We're getting Bobcats in Connecticut. 609 00:30:42.329 --> 00:30:42.549 Wow. 610 00:30:42.650 --> 00:30:43.170 Yeah. 611 00:30:43.190 --> 00:30:45.190 They came back to where they came back to Missouri 612 00:30:45.230 --> 00:30:48.690 where I grew up because the, the wild Turkey population 613 00:30:48.750 --> 00:30:51.190 finally recovered from the great depression. Yeah. Yeah. 614 00:30:52.880 --> 00:30:56.190 I think that's happening in general. Generally, we hunt less. 615 00:30:56.390 --> 00:30:59.230 Our otter population is way up. Having to mink around 616 00:30:59.250 --> 00:31:03.170 is crazy. Definitely, the birds, there's more of everything. Bit 617 00:31:03.190 --> 00:31:06.160 by bit, we're going to have more encounters. 618 00:31:06.339 --> 00:31:10.500 My late mother is probably solely responsible for the wild 619 00:31:10.619 --> 00:31:14.380 turkey population of Groton, Connecticut skyrocketing. 620 00:31:14.480 --> 00:31:15.269 She fed them all the time. 621 00:31:15.569 --> 00:31:19.470 Because she fed them corn, you know, cracked corn every 622 00:31:19.519 --> 00:31:20.799 day along with the deer. 623 00:31:21.099 --> 00:31:21.480 Oh, man. 624 00:31:21.680 --> 00:31:21.839 Nice. 625 00:31:22.140 --> 00:31:24.480 And I always wanted to taste one that had been 626 00:31:24.539 --> 00:31:28.779 living on cracked corn, you know, for generations. But they're 627 00:31:28.799 --> 00:31:31.640 still wild turkeys. But if the only food they eat 628 00:31:31.720 --> 00:31:35.009 is her cracked corn, they must be delicious, you know? Maybe. 629 00:31:35.180 --> 00:31:35.440 Maybe. 630 00:31:35.990 --> 00:31:38.910 Anyway, she never let me kill one. So I didn't. 631 00:31:39.289 --> 00:31:39.829 And now they're gone. 632 00:31:41.039 --> 00:31:43.970 She stopped. She's gone. So they're gone, right? There's nobody 633 00:31:44.069 --> 00:31:44.630 feeding them anymore. 634 00:31:44.670 --> 00:31:45.630 She's gone. They're gone. 635 00:31:45.869 --> 00:31:46.109 Anyway. 636 00:31:46.309 --> 00:31:46.630 All right. 637 00:31:46.650 --> 00:31:46.980 All right. 638 00:31:47.000 --> 00:31:48.359 Well, this seems like a good time for a break. 639 00:31:48.440 --> 00:31:50.980 So we'll be right back after these very important messages. 640 00:31:51.039 --> 00:31:56.880 Don't go away. And we're back. It's on the rocks. 641 00:31:56.920 --> 00:31:59.500 I'm Richard Campbell. Let's call Franklin. Yes, I am. Talking 642 00:31:59.519 --> 00:32:03.339 to our friend Jeremy Miller about the expanded critter stack. 643 00:32:03.380 --> 00:32:06.440 So I could see the AI tools allowed to accelerate. Although, 644 00:32:06.619 --> 00:32:09.549 I mean, Polkat and Fisher are adjacent to Martin in 645 00:32:09.569 --> 00:32:12.329 that sense. So they are some ways the tool will 646 00:32:12.349 --> 00:32:13.849 be really good at, you know, how do I do 647 00:32:13.890 --> 00:32:14.710 this in this other stack? 648 00:32:14.829 --> 00:32:16.890 They are. But to get there and to get there 649 00:32:16.990 --> 00:32:20.309 and to make it sustainable is, We've had to do 650 00:32:20.390 --> 00:32:23.329 a lot of work over time to lift a lot 651 00:32:23.369 --> 00:32:29.230 more common functionality out, lift reusable compliance tests to get 652 00:32:29.279 --> 00:32:30.119 a lot of behavior. 653 00:32:31.319 --> 00:32:32.710 You don't want to do that over and over again. 654 00:32:32.779 --> 00:32:35.099 Yeah. So, you know, we can also dive into a 655 00:32:35.119 --> 00:32:37.099 little bit of a theme of you aren't going to 656 00:32:37.140 --> 00:32:40.119 be able to just vibe code something like a polecat 657 00:32:40.220 --> 00:32:41.640 over a weekend. 658 00:32:41.900 --> 00:32:43.839 Well, that's a classic, isn't it? Right. Like, why do 659 00:32:43.920 --> 00:32:46.599 I buy a framework and I just make my own? 660 00:32:46.839 --> 00:32:48.359 Yeah, that's delusional. 661 00:32:48.970 --> 00:32:49.890 Um, sorry. 662 00:32:50.329 --> 00:32:50.950 I agree. 663 00:32:51.710 --> 00:32:55.059 Sorry. Coming from, I support one that's sort of couple 664 00:32:55.400 --> 00:32:57.480 that are widely used by a lot of different people, 665 00:32:58.079 --> 00:33:00.819 even with the AI stuff. And we're all using AI 666 00:33:00.880 --> 00:33:03.519 tools to do the fixes. I use it to do 667 00:33:03.559 --> 00:33:07.750 a lot of exploratory testing. There's just so much input 668 00:33:07.789 --> 00:33:10.890 that comes from people using it in real life. It's, 669 00:33:11.369 --> 00:33:15.799 it's all kind of, and it's the, It's the upset conditions. 670 00:33:16.099 --> 00:33:21.990 Kubernetes shuts down a node suddenly. Your DBA suddenly kill 671 00:33:22.029 --> 00:33:25.210 a

database late at night and it will actually skew 672 00:33:25.710 --> 00:33:29.069 Postgres sequence values if you're not careful. We learned that 673 00:33:29.109 --> 00:33:33.180 one the hard way. Being able to have a system 674 00:33:33.220 --> 00:33:38.940 that's very highly in flight and then try to take 675 00:33:38.980 --> 00:33:42.079 it down gracefully without losing work as you go. That's 676 00:33:42.119 --> 00:33:46.460 actually really hard. Yeah. And the only way you get 677 00:33:46.559 --> 00:33:48.720 to a tool like a Wolverine or a Martin and 678 00:33:48.759 --> 00:33:51.079 make it good is you gotta have a lot of 679 00:33:51.119 --> 00:33:53.579 people beat up on it in real life and then, 680 00:33:54.259 --> 00:33:57.009 then iterate and improve. You're not just going to get 681 00:33:57.049 --> 00:33:59.049 that with a single time vibe coding pass. 682 00:34:00.029 --> 00:34:02.930 Have you paid any attention to the use what works group? 683 00:34:03.309 --> 00:34:04.589 That's yeah. 684 00:34:05.549 --> 00:34:09.369 A little bit. I'm going to admit that I've been 685 00:34:09.409 --> 00:34:12.670 a tiny bit cynical about it because it looks like 686 00:34:12.809 --> 00:34:14.929 the conference drinking club. 687 00:34:17.199 --> 00:34:17.960 You're not wrong. 688 00:34:19.820 --> 00:34:23.139 I appreciate what they're trying to do, but I'm not 689 00:34:23.179 --> 00:34:23.739 involved with it. 690 00:34:24.639 --> 00:34:29.159 The message is essentially the same. This idea that your 691 00:34:29.179 --> 00:34:32.579 job wasn't to build a framework. Your job was to 692 00:34:32.960 --> 00:34:36.500 provide value to your customers. And these frameworks facilitate that. 693 00:34:36.639 --> 00:34:37.320 Use what works. 694 00:34:37.579 --> 00:34:40.000 Yes. So one of the things I think I would 695 00:34:40.079 --> 00:34:42.369 argue or what I do argue to people of why 696 00:34:42.429 --> 00:34:45.309 you should still use something off the shelf. Like you 697 00:34:45.349 --> 00:34:49.190 look on these LinkedIn conversations, they'll say, oh, just let 698 00:34:49.230 --> 00:34:54.550 AI build you an outbox. something that's widely used is curated. 699 00:34:55.150 --> 00:34:58.030 Somebody is paying attention to it, reviewing the pull requests 700 00:34:58.070 --> 00:35:02.699 coming in, driving AI, watching the quality, the problems that 701 00:35:02.739 --> 00:35:08.019 people are having, and evolving it over time. You're just 702 00:35:08.039 --> 00:35:10.300 not going to build something over a weekend, no matter 703 00:35:10.340 --> 00:35:10.659 how good. 704 00:35:10.920 --> 00:35:12.900 I wonder if this is even a temporary concept now 705 00:35:12.920 --> 00:35:15.409 that we're starting to actually pay for tokens. It's like, 706 00:35:15.570 --> 00:35:16.949 it ain't worth the tokens. 707 00:35:17.750 --> 00:35:20.090 It's probably valuable in a way that it makes it 708 00:35:20.150 --> 00:35:21.369 easy to quantify efforts. 709 00:35:21.750 --> 00:35:22.429 Yeah. Yeah. 710 00:35:22.760 --> 00:35:25.480 Just like, look, spend the tokens on the value that 711 00:35:25.780 --> 00:35:29.300 you really bring to the customer. Buying a framework at 712 00:35:29.340 --> 00:35:31.179 a fixed price is a heck of a lot more 713 00:35:31.260 --> 00:35:33.360 sensible where you at least know what it's going to 714 00:35:33.400 --> 00:35:35.510 cost and that somebody else is paying attention to it. 715 00:35:35.539 --> 00:35:37.059 You don't have to own that code base and you 716 00:35:37.079 --> 00:35:39.519 don't have to spend tokens on it either. All right. 717 00:35:40.079 --> 00:35:40.420 Awesome. 718 00:35:40.480 --> 00:35:44.250 So, I mean, a few new products, but CQRS is 719 00:35:44.269 --> 00:35:46.909 sort of CQRS. How do you evolve event sourcing at 720 00:35:46.929 --> 00:35:49.429 this point? What's changing in this world? 721 00:35:49.840 --> 00:35:52.780 Oh, quite a bit. I know you've had some recent hosts. 722 00:35:54.139 --> 00:35:56.739 So the last couple of years, last couple of years, 723 00:35:56.760 --> 00:36:00.050 there's been a trend. I don't entirely buy into it, 724 00:36:00.119 --> 00:36:04.889 but there is the idea of dynamic consistency boundaries where 725 00:36:04.929 --> 00:36:07.570 you stop. I don't know. Forgive me if you guys 726 00:36:07.590 --> 00:36:09.289 have already had somebody talk about it on the show. 727 00:36:10.389 --> 00:36:15.269 But breaking away from the model of strictly tracking events 728 00:36:15.349 --> 00:36:18.139 by event streams instead of doing it. really doing things 729 00:36:18.179 --> 00:36:18.980 more by tagging. 730 00:36:19.199 --> 00:36:19.420 Yeah. 731 00:36:19.800 --> 00:36:24.199 Um, where an event could, you know, the, the classic 732 00:36:24.260 --> 00:36:26.940 example is you have a classroom, you have events that 733 00:36:26.960 --> 00:36:30.059 apply to both the student and the classroom and maybe 734 00:36:30.099 --> 00:36:33.980 the instructor, uh, with the theory that this makes it easier. 735 00:36:34.460 --> 00:36:37.239 This makes it easier for you to be softer in 736 00:36:37.460 --> 00:36:42.960 designing your, um, stream boundaries. Um, If you'll let me 737 00:36:43.000 --> 00:36:45.539 be snarky for a second, I think this is partially 738 00:36:45.639 --> 00:36:49.269 driven by the weaknesses of some of the commercial event 739 00:36:49.289 --> 00:36:53.019 sourcing tools that can't do strong consistency or between event streams. 740 00:36:54.530 --> 00:36:58.530 But even without that, people think that makes modeling easier. 741 00:36:59.269 --> 00:37:03.880 That's trend number one. The biggest trend that's taking up 742 00:37:03.889 --> 00:37:07.460 a lot of my time right now is, again, it's 743 00:37:07.500 --> 00:37:11.880 more AI driven, but people getting very seriously about event 744 00:37:11.920 --> 00:37:16.369 modeling and honestly, for us old guys, trying to bring 745

00:37:16.590 --> 00:37:20.489 the old model-driven development idea that never worked, trying to 746 00:37:20.530 --> 00:37:24.610 bring that into the modern world with event modeling, where 747 00:37:24.849 --> 00:37:31.090 you build models, maybe you complement or supplement that with 748 00:37:31.239 --> 00:37:37.219 kind of BDD-style specifications, and then AI-driven tools and scaffolding 749 00:37:37.739 --> 00:37:42.449 builds you an application or constantly evolves your application based 750 00:37:42.469 --> 00:37:46.610 on almost a two-way mapping between the event model of 751 00:37:46.750 --> 00:37:49.050 this is what the application is. These are the slices. 752 00:37:49.110 --> 00:37:52.250 These are the events. These are commands. Generate me code 753 00:37:52.349 --> 00:37:56.010 from this and also generate code based off of BDD 754 00:37:56.050 --> 00:37:59.559 style specs. So those are the two big trends I 755 00:37:59.579 --> 00:38:01.539 see or things that are causing me to take a 756 00:38:01.559 --> 00:38:04.460 lot of time. And then all the old challenges of 757 00:38:04.579 --> 00:38:09.949 concurrency and scaling, those are all still real. And we're 758 00:38:10.150 --> 00:38:11.949 still working on new solutions for those all. 759 00:38:11.969 --> 00:38:13.449 I mean, I don't have a big problem with event 760 00:38:13.489 --> 00:38:16.650 modeling because you do need to think through the event flow. 761 00:38:17.269 --> 00:38:19.929 I think it almost feels like they're going too far 762 00:38:19.989 --> 00:38:22.530 down the path. We had Adam on a while back 763 00:38:22.699 --> 00:38:25.920 talking about the event sourcing. And it was always that 764 00:38:25.960 --> 00:38:28.559 question of like, how far do you go with this model? 765 00:38:29.679 --> 00:38:32.099 It's not the whole application. It's just one part of 766 00:38:32.820 --> 00:38:34.889 the flow of data and how you want to manage it. 767 00:38:35.170 --> 00:38:39.829 So everybody is the hero of their own story, right? 768 00:38:40.210 --> 00:38:42.349 If you talk to a tester, they think the most 769 00:38:42.409 --> 00:38:45.650 important thing is getting testing. Well, if you talk to 770 00:38:45.670 --> 00:38:49.230 a business analyst, it's getting the requirements coders, you know, 771 00:38:49.269 --> 00:38:56.030 obviously it's the code. Um, they've been modeling people, obviously their, 772 00:38:56.050 --> 00:39:01.210 their analysis work is key. Um, I think my take 773 00:39:01.429 --> 00:39:04.170 is it's definitely valuable. I hear from our own users 774 00:39:04.210 --> 00:39:07.320 that they think it's valuable. I think even though that's 775 00:39:07.360 --> 00:39:10.340 exactly what I'm working on today is the scaffolding from 776 00:39:10.360 --> 00:39:12.539 event models, I think that it does a lot of 777 00:39:12.559 --> 00:39:14.679 the easy work for you, and there's still a lot 778 00:39:14.699 --> 00:39:17.179 of technical challenges after the event modeling. 779 00:39:17.260 --> 00:39:17.480 Sure. 780 00:39:18.099 --> 00:39:20.519 And I think the event modeling is maybe a little 781 00:39:20.539 --> 00:39:22.690 bit oversold by some of its proponents. 782 00:39:23.750 --> 00:39:27.210 Yeah, and one could argue that for architecture every time too, right? 783 00:39:27.289 --> 00:39:31.320 It's like every one of these all of who was it? 784 00:39:31.340 --> 00:39:33.579 Was it Eisenhower who said, you know, the plan won't 785 00:39:33.619 --> 00:39:36.329 survive content to the enemy, but the planning was worth it. 786 00:39:36.760 --> 00:39:38.019 So it's not a bad thing. 787 00:39:38.079 --> 00:39:42.099 Yeah. You've seen, and you've seen that for, for younger folks, 788 00:39:42.699 --> 00:39:44.619 we used to quote that over and over again in 789 00:39:44.679 --> 00:39:47.269 the extreme programming, early agile days. 790 00:39:47.349 --> 00:39:47.570 Right. 791 00:39:47.690 --> 00:39:47.889 Right. 792 00:39:48.230 --> 00:39:50.710 Yeah. It's good to make a plan. It's also good 793 00:39:50.750 --> 00:39:55.150 to let that plan glow. It's contact with reality. Like 794 00:39:55.170 --> 00:39:57.250 you work through an event model and then real data 795 00:39:57.269 --> 00:39:59.469 starts to flow and you run into some issues and 796 00:39:59.510 --> 00:40:00.449 you have to change. 797 00:40:00.570 --> 00:40:00.820 Yes. 798 00:40:01.489 --> 00:40:05.019 Well, and of course I'm, I'm a history nerd. So, um, 799 00:40:05.159 --> 00:40:08.880 just happened to be listening to a podcast on the 800 00:40:08.900 --> 00:40:11.320 day for as horrible as that was, but how, um, 801 00:40:11.760 --> 00:40:16.840 how U.S. and British destroyers had to deviate from the 802 00:40:16.940 --> 00:40:20.239 plan quite a bit when all the communication broke down 803 00:40:20.300 --> 00:40:21.179 on the day of the landing. 804 00:40:21.219 --> 00:40:24.010 Right. And find a way to still be effective. 805 00:40:24.349 --> 00:40:27.889 There was decoys and stuff too, right? Wasn't there like 806 00:40:27.929 --> 00:40:31.829 some fake spy stuff that was going on that let 807 00:40:31.909 --> 00:40:32.809 the Germans believe? 808 00:40:32.929 --> 00:40:36.309 There's one famous story of U.S. Rangers have to do 809 00:40:36.329 --> 00:40:39.550 a crazy heroic assault up a cliff And when they 810 00:40:39.590 --> 00:40:43.789 got there, the guns were telephone poles. 811 00:40:44.090 --> 00:40:45.010 They were decoys. 812 00:40:46.269 --> 00:40:48.250 And there was, like I was about to say, there 813 00:40:48.269 --> 00:40:50.550 was some spy stuff that was going on that made 814 00:40:50.590 --> 00:40:52.530 the Germans think that it was happening at a completely 815 00:40:52.570 --> 00:40:53.150 different place. 816 00:40:53.409 --> 00:40:57.559 Oh, yeah. We set up a whole fake army around 817 00:40:57.659 --> 00:40:59.840 that to make them think we were trying to land 818 00:40:59.880 -->

00:41:00.400 somewhere else. 819 00:41:00.480 --> 00:41:00.659 Yeah. 820 00:41:01.019 --> 00:41:02.039 Yeah, as one does. 821 00:41:02.059 --> 00:41:04.480 I wonder if those techniques could still be used. I 822 00:41:04.519 --> 00:41:04.880 doubt it. 823 00:41:04.900 --> 00:41:05.920 I don't know. 824 00:41:06.340 --> 00:41:08.949 There's funny things going on in the Ukraine these days. 825 00:41:09.630 --> 00:41:12.110 With inflatable tanks and all kinds of nutty stuff. 826 00:41:12.179 --> 00:41:12.780 The decoys. 827 00:41:12.900 --> 00:41:13.139 Yeah. 828 00:41:13.179 --> 00:41:14.079 The decoys for drones. 829 00:41:14.539 --> 00:41:15.599 Decoys for drones. 830 00:41:15.619 --> 00:41:18.219 During World War II, there was also a fake Paris 831 00:41:18.719 --> 00:41:21.699 set up somewhere with an Eiffel Tower and everything to 832 00:41:21.739 --> 00:41:22.860 confuse the pilots. 833 00:41:23.420 --> 00:41:24.440 Oh, I didn't know about that. 834 00:41:24.460 --> 00:41:29.199 Oh, yeah. That was a classic thing. Anyway, we digress. 835 00:41:29.559 --> 00:41:32.210 Yeah, we do. But that's not a bad thing, too. 836 00:41:32.230 --> 00:41:37.269 It's interesting to explore all the ways that you actually 837 00:41:37.289 --> 00:41:39.510 have to build the software to You're going to bump 838 00:41:39.550 --> 00:41:41.750 into issues. I guess it's always a debate about how 839 00:41:41.789 --> 00:41:43.630 many of the features you end up building in a 840 00:41:43.670 --> 00:41:46.909 tool like Martin end up being the reality check of. 841 00:41:47.010 --> 00:41:49.199 And then when we actually did this, you're going to 842 00:41:49.239 --> 00:41:49.900 have this problem. 843 00:41:50.199 --> 00:41:54.800 So, I mean, as always, we learned, again, one of 844 00:41:54.820 --> 00:41:57.380 the lessons that came out of the Agile software development 845 00:41:57.400 --> 00:41:59.309 days before it became Scrum and got ruined. 846 00:41:59.329 --> 00:41:59.849 Yeah. 847 00:42:00.570 --> 00:42:04.170 That the design, the features you built should be pull, 848 00:42:04.269 --> 00:42:08.050 not push. You pull things in based on there is 849 00:42:08.090 --> 00:42:12.090 a demonstrated need to do this. So the features that 850 00:42:12.110 --> 00:42:14.480 have done the best for us, there's a feature in 851 00:42:14.519 --> 00:42:20.780 Wolverine we call global partitioning. What it is, is being 852 00:42:20.820 --> 00:42:23.699 able to take a look at some kind of business identification, 853 00:42:24.039 --> 00:42:27.769 you know, say an event stream, a classroom, and I'm 854 00:42:27.789 --> 00:42:30.869 going to make sure that all messages for this classroom 855 00:42:30.909 --> 00:42:33.670 are handled in sequential order, but I'm going to let 856 00:42:34.010 --> 00:42:40.480 other classrooms be handled in parallel. It's a really robust 857 00:42:40.559 --> 00:42:46.849 way to sidestep concurrency problems. That came out of... observing 858 00:42:47.090 --> 00:42:49.530 user needs you know what was happening in real systems 859 00:42:49.630 --> 00:42:52.610 right but other features that maybe we thought about for 860 00:42:52.710 --> 00:42:54.369 years that i thought was going to be a really 861 00:42:54.429 --> 00:42:58.449 big deal got it out there promoted it yeah not 862 00:42:58.469 --> 00:42:59.510 so much yeah. 863 00:42:59.360 --> 00:43:01.480 Yeah you never know it's hard to anticipate you just 864 00:43:01.500 --> 00:43:03.639 gotta go to collect the data and realize okay now 865 00:43:03.679 --> 00:43:04.659 we do this a different way. 866 00:43:04.619 --> 00:43:06.940 Now which you know that only works if you can't 867 00:43:07.000 --> 00:43:10.320 actually collect data and talk to people that use your 868 00:43:10.360 --> 00:43:12.980 tools And I'll tell you, that got a whole lot 869 00:43:13.039 --> 00:43:14.960 easier once I had a company and I was working 870 00:43:14.980 --> 00:43:19.219 for clients instead of getting secondhand stuff off of Discord. 871 00:43:19.480 --> 00:43:23.570 Right. Yeah. How much telemetry do you put into these 872 00:43:23.610 --> 00:43:26.889 tools as well? It's always great to collect data in utilization. 873 00:43:26.969 --> 00:43:27.889 I just don't know how. 874 00:43:27.829 --> 00:43:29.590 Comfortable people will be with that. So to be clear, 875 00:43:29.610 --> 00:43:31.449 we don't have anything that communicates back to us. 876 00:43:31.510 --> 00:43:31.710 Right. 877 00:43:31.969 --> 00:43:32.389 That's good. 878 00:43:32.659 --> 00:43:35.539 Don't have anything like that. The Critter Watch tool I 879 00:43:35.579 --> 00:43:39.210 talked about before. does have the ability, you could use 880 00:43:39.269 --> 00:43:41.730 that to kind of look into a system, but it's 881 00:43:41.750 --> 00:43:45.010 not sending anything back to us. The best way I 882 00:43:45.050 --> 00:43:48.210 can get that, it's not fine-grained on features, but just 883 00:43:48.269 --> 00:43:52.489 looking at, you know, the Nougat extensions, like I know, 884 00:43:52.889 --> 00:43:55.630 you know, Wolverine has so many downloads, and then I 885 00:43:55.650 --> 00:43:58.610 can see that most, probably a little over half are 886 00:43:58.670 --> 00:44:00.809 using it with Martin as well, and I can see 887 00:44:00.849 --> 00:44:02.829 how many people are using it with EF Core and, 888 00:44:03.289 --> 00:44:07.719 you know, down to the RabbitMQ and google gcp pub 889 00:44:07.739 --> 00:44:11.150 sub i get that a little bit but no right 890 00:44:11.489 --> 00:44:12.889 there's nothing snooping on you. 891 00:44:13.090 --> 00:44:15.690 Yeah which at the same time it's like that means 892 00:44:15.730 --> 00:44:19.309 invariably you build the products based on either customer feedback 893 00:44:19.349 --> 00:44:21.829 which is filtered or your own projects which are only 894 00:44:21.849 --> 00:44:24.969 going to be a subset like i i get it's uh. 895

00:44:24.969 --> 00:44:27.659 That's very true but at the same token i'm also 896 00:44:27.699 --> 00:44:31.909 going to say that jasper fx clients get um They 897 00:44:31.949 --> 00:44:34.440 are the highest priority on what gets built. So what 898 00:44:34.460 --> 00:44:36.960 they need is what gets built. And we would be 899 00:44:37.019 --> 00:44:37.829 happy to work with you. 900 00:44:37.840 --> 00:44:38.719 Yeah. Happy to help. 901 00:44:39.039 --> 00:44:40.300 No problem. All right. 902 00:44:40.920 --> 00:44:42.599 And if you need a feature built, I know a guy. 903 00:44:42.619 --> 00:44:48.719 I appreciate that. It's cool. All right. What else? What 904 00:44:48.739 --> 00:44:50.039 have we not dug into here? 905 00:44:50.340 --> 00:44:55.300 Well, I can throw something in the fire here, which 906 00:44:55.420 --> 00:45:01.280 is the tendency for unseasoned developers to want to find 907 00:45:01.300 --> 00:45:04.079 a way to use every feature of a tool, even 908 00:45:04.119 --> 00:45:07.619 if they don't need it. Um, you know, I find 909 00:45:07.659 --> 00:45:10.420 a new tool and then, you know, it's great if 910 00:45:10.519 --> 00:45:13.019 it helps you identify problems, but then you would go 911 00:45:13.119 --> 00:45:16.679 further and say, Oh, well we should use this on 912 00:45:17.039 --> 00:45:21.119 our code base or whatever when you may not need to. 913 00:45:21.590 --> 00:45:24.269 And that's just comes with experience. But I mean, I've 914 00:45:24.289 --> 00:45:25.570 seen that happen over and over again. 915 00:45:25.769 --> 00:45:28.059 I would give you the old, uh, I think it's 916 00:45:28.079 --> 00:45:30.110 a quote from Martin Fowler, but I don't remember where 917 00:45:30.139 --> 00:45:32.789 it came, where he did it. The only way to 918 00:45:32.849 --> 00:45:37.530 know how far a technique or a tool goes is 919 00:45:37.610 --> 00:45:40.210 to take it too far and then rein it back. 920 00:45:40.690 --> 00:45:41.230 Yeah. 921 00:45:41.449 --> 00:45:41.869 Yeah. Yeah. 922 00:45:42.860 --> 00:45:46.380 You kind of have to, right? Over on the system inside, 923 00:45:46.400 --> 00:45:50.880 we talk about minimizing constraints on a new tool so 924 00:45:50.900 --> 00:45:53.199 that people will actually engage with it. And then you have, 925 00:45:53.300 --> 00:45:54.960 it's like, you got to let the puppy run. Just 926 00:45:55.000 --> 00:45:58.559 don't let the puppy throw up. Like, Let them go. Yeah. 927 00:45:58.699 --> 00:46:00.719 You know, we saw this with SharePoint. We see it 928 00:46:00.739 --> 00:46:03.820 with Teams, like all these things where we're going to 929 00:46:03.900 --> 00:46:07.059 need governance. But if we put it in before anybody starts, 930 00:46:07.860 --> 00:46:10.320 it gets very little adoption. If we kind of let 931 00:46:10.340 --> 00:46:14.550 them run roughshod for a while and explore and cause problems, 932 00:46:14.969 --> 00:46:18.590 then people want governance. And you've

already got some adoption 933 00:46:18.610 --> 00:46:21.030 and value from it too. Like I think you kind 934 00:46:21.050 --> 00:46:23.469 of expect people to go too far. Yeah. I would 935 00:46:23.489 --> 00:46:24.480 agree with that. Yeah. 936 00:46:24.820 --> 00:46:27.619 Funny, funny reality. Like there's never a perfection on any 937 00:46:27.659 --> 00:46:31.420 of this. Well, that was a short lived topic. Brought 938 00:46:31.460 --> 00:46:32.920 the conversation to a screeching halt. 939 00:46:33.519 --> 00:46:36.500 What about the AI skills side of Critterstack? What's this about? 940 00:46:36.820 --> 00:46:39.449 Well, one, I've got to stay in business. So there's 941 00:46:39.469 --> 00:46:43.730 kind of, there's gotta be something commercial related to AI. Yeah. 942 00:46:44.630 --> 00:46:48.920 So the AI skills is, Think of some of the 943 00:46:48.960 --> 00:46:52.110 sci-fi books you've read over the years where, oh, I'm 944 00:46:52.150 --> 00:46:54.929 going into combat. I'm going to put my combat skills 945 00:46:55.070 --> 00:46:58.550 program into active memory, right? 946 00:46:59.110 --> 00:46:59.809 Very Matrix. 947 00:46:59.929 --> 00:47:02.079 Yes. Or if you've read any kind of Peter F. 948 00:47:02.090 --> 00:47:05.550 Hamilton books. So the AI skills, they're going to teach 949 00:47:05.989 --> 00:47:09.860 your CLOD, your ASTRA, GPT. They're going to teach it, hey, 950 00:47:09.880 --> 00:47:13.000 I mean, the really markdown files, but a little more 951 00:47:13.039 --> 00:47:15.960 concentrated in documentation. They're going to try to give the 952 00:47:16.019 --> 00:47:19.480 AI tools... Guardrails on here is how to use the 953 00:47:19.539 --> 00:47:24.070 tools the best way. Certainly a lot of stuff around 954 00:47:24.199 --> 00:47:29.130 Martin itself, but taking Wolverine is a better example. Wolverine's 955 00:47:29.170 --> 00:47:32.449 value is that it allows you to create much, much 956 00:47:32.530 --> 00:47:37.900 simpler application code than... any of its competitor tools. There's 957 00:47:37.920 --> 00:47:41.760 just more things it does to simplify code. But if 958 00:47:41.860 --> 00:47:44.889 you let an AI tool go, it will try to 959 00:47:44.989 --> 00:47:48.989 use Wolverine just like its mass transit or mediator or 960 00:47:49.550 --> 00:47:51.400 on the HP side, it'll try to use it just 961 00:47:51.440 --> 00:47:55.139 like its MVC core minimal API. and you will lose 962 00:47:55.159 --> 00:47:58.969 a lot of the value of what Wolverine possibly can 963 00:47:59.010 --> 00:48:02.070 do as far as shrinking your code. The AI skills will, 964 00:48:02.289 --> 00:48:05.090 will teach it of, no, this is the idiomatic way 965 00:48:05.110 --> 00:48:08.429 to use it. Nice. You know, for Wolverine, it's no, 966 00:48:08.719 --> 00:48:11.900 you want to reduce the code to a pure function. 967 00:48:12.039 --> 00:48:14.940 That's easy to test. I want to reduce layering. You 968 00:48:14.960 --> 00:48:18.119 want to, you know, Think of what somebody would do 969 00:48:18.239 --> 00:48:20.179 using clean architecture and don't do that. 970 00:48:20.960 --> 00:48:21.159 Right. 971 00:48:21.400 --> 00:48:23.429 Every possible way to simplify a code, there's a lot 972 00:48:23.469 --> 00:48:27.670 of skills to be able to utilize testing. Not be 973 00:48:27.710 --> 00:48:32.429 able to just to utilize the testing helpers in Wolverine 974 00:48:32.469 --> 00:48:35.380 and Martin, of which there's a lot, you know. But 975 00:48:35.570 --> 00:48:38.559 when you get into asynchronous testing, it also teaches your 976 00:48:38.619 --> 00:48:42.360 AI tool, hey, here's how to use all the built-in 977 00:48:42.539 --> 00:48:46.199 CLI tools, command line tools, diagnostic tools that we've built 978 00:48:46.340 --> 00:48:50.099 into Wolverine and Martin. Things like, okay, I want to 979 00:48:50.139 --> 00:48:54.159 go test. This message seemed to disappear. I can run 980 00:48:54.179 --> 00:48:56.280 a command line and it can tell me this is 981 00:48:56.340 --> 00:48:58.349 where the message will be sent to this route. If 982 00:48:58.550 --> 00:49:03.829 you end point or topic or whatnot, it'll explain the 983 00:49:03.889 --> 00:49:06.650 command line tools will explain to an AI agent. Here's 984 00:49:06.690 --> 00:49:12.719 what it's doing. Every possible bit of information about how 985 00:49:12.760 --> 00:49:16.280 the application is configured. If you have a Critterwatch tool, 986 00:49:16.300 --> 00:49:19.650 the AI skills will teach you how to use MCN points, 987 00:49:20.820 --> 00:49:24.059 in Critterwatch to be able to reach out into your 988 00:49:24.099 --> 00:49:28.239 open telemetry to say, hey, I'm working on this saga. 989 00:49:28.679 --> 00:49:31.820 Tell me what's going on. That'll teach your AI agent 990 00:49:31.880 --> 00:49:35.940 how to reach into Jaeger or App Insights or whatever 991 00:49:35.960 --> 00:49:39.579 it's using through Critterwatch. And it will grab all the 992 00:49:39.619 --> 00:49:43.320 information about here is all the work that's already happening 993 00:49:43.380 --> 00:49:46.679 for the saga. Here is the information that's yet to come. 994 00:49:46.699 --> 00:49:49.260 Maybe it can tell you, hey, this message failed. This 995 00:49:49.280 --> 00:49:53.929 message happened at this time. But this AI skills, it's 996 00:49:53.989 --> 00:49:57.510 trying to unlock every bit of information about your system, 997 00:49:57.550 --> 00:50:00.630 both at rest and at movement. Cool. 998 00:50:00.989 --> 00:50:01.250 Nice. 999 00:50:01.630 --> 00:50:03.250 Yeah, it makes a lot of sense. And it just 1000 00:50:03.269 --> 00:50:06.489 saves time, right? You're just going to get more done 1001 00:50:06.510 --> 00:50:07.230 in less time that way. 1002 00:50:07.489 --> 00:50:12.530 Potentially saves time. And this is us going through the 1003 00:50:12.570 --> 00:50:16.590 full cycle. I'm sure everybody in my

position went through 1004 00:50:16.659 --> 00:50:20.900 a cycle of, oh my God, we're dead meat. The 1005 00:50:20.980 --> 00:50:25.619 AI is going to crush our business model to, okay, 1006 00:50:26.159 --> 00:50:28.679 it's helping us deliver a lot faster, but we can 1007 00:50:28.760 --> 00:50:32.570 use AI related functionality to build a lot more value 1008 00:50:32.639 --> 00:50:35.269 in for our customers and maybe come out the other 1009 00:50:35.289 --> 00:50:40.949 side and maybe have something that's That's hopefully something that 1010 00:50:41.010 --> 00:50:43.170 people want to actually buy and pay for. 1011 00:50:43.190 --> 00:50:45.510 Yeah. No, no question. 1012 00:50:45.710 --> 00:50:48.949 Do you have anything else planned or are you done 1013 00:50:48.989 --> 00:50:50.050 making critters for a while? 1014 00:50:50.570 --> 00:50:54.250 So the biggest thing for us in a, so there'll 1015 00:50:54.269 --> 00:50:57.820 be a new critter watch 1.1. And by the time 1016 00:50:57.880 --> 00:51:01.820 this podcast drops, we'll have a big new release. It's 1017 00:51:02.300 --> 00:51:04.920 going to give you an option to be able to 1018 00:51:04.940 --> 00:51:07.860 embed critter watch in your own application. like at dev 1019 00:51:07.920 --> 00:51:11.219 time even. So you get a lot more insight at 1020 00:51:11.340 --> 00:51:15.440 what's happening at development time. And more features, it's finally 1021 00:51:15.500 --> 00:51:20.210 adding cron-based scheduling and then some other features. And then 1022 00:51:20.889 --> 00:51:25.670 our big grand epic of the critter stacks version of 1023 00:51:25.690 --> 00:51:29.210 event modeling and spec driven development, being able to sit 1024 00:51:29.230 --> 00:51:31.590 in a room and, you know, be able to project 1025 00:51:31.710 --> 00:51:34.380 up a meeting, meeting of, Hey, let's look at the, 1026 00:51:34.679 --> 00:51:37.420 the visualization of the event model as we talk about 1027 00:51:38.119 --> 00:51:41.840 what it's supposed to do. And is everybody agree with this? 1028 00:51:42.099 --> 00:51:44.320 And then push play and let an AI set of 1029 00:51:44.380 --> 00:51:47.980 AI tools, scaffold it all out for you, including the tests. 1030 00:51:48.119 --> 00:51:49.219 Nice. Awesome. 1031 00:51:49.239 --> 00:51:52.219 So we're going live in two days. Did you, mean 1032 00:51:52.260 --> 00:51:54.380 to say that by the time this podcast drops? 1033 00:51:54.679 --> 00:52:00.079 So, Carl, I went to Rice University in Houston. There's 1034 00:52:00.139 --> 00:52:03.719 a famous story. So, believe it or not, Rice was 1035 00:52:03.760 --> 00:52:05.159 really good in football once upon a time. 1036 00:52:05.420 --> 00:52:06.559 No. 1037 00:52:07.079 --> 00:52:11.019 But like in the 50s. And sometime in the early 60s, 1038 00:52:11.079 --> 00:52:13.159 they decided, hey, we're going to use a whole bunch 1039 00:52:13.199 --> 00:52:16.280 of leftover concrete from a Houston freeway, and we're going 1040 00:52:16.300 --> 00:52:19.139 to go build the giant Rice Stadium that's still there today. 1041 00:52:19.159 --> 00:52:19.570 Hmm. 1042 00:52:19.980 --> 00:52:22.090 And we're going to get it done before the next season. 1043 00:52:22.869 --> 00:52:26.809 So at one point, the story is that one of 1044 00:52:27.289 --> 00:52:29.409 the board of governors is asking one of the Brown 1045 00:52:29.449 --> 00:52:33.050 brothers from Brown and Root, famous company, hey, is the 1046 00:52:33.070 --> 00:52:35.380 stadium going to be done for the first game? And 1047 00:52:35.389 --> 00:52:37.940 he looks at him deadpan and asks, well, is it 1048 00:52:37.960 --> 00:52:39.239 going to be a day game or a night game? 1049 00:52:43.000 --> 00:52:44.920 It might be a couple of days after this goes live. 1050 00:52:44.940 --> 00:52:46.099 You guys used to be slower. 1051 00:52:46.119 --> 00:52:46.840 I know. 1052 00:52:46.880 --> 00:52:49.969 Well, we're going to be there really soon. Thursday. 1053 00:52:49.989 --> 00:52:53.110 Thursday we're going to do a big demo. I looked 1054 00:52:53.190 --> 00:52:55.590 up the Kelly Blue Book value of an old car 1055 00:52:55.650 --> 00:52:58.190 and it said, is that with a tank full or empty? 1056 00:52:59.469 --> 00:53:00.329 Kind of the same joke. 1057 00:53:00.590 --> 00:53:00.889 Nice. 1058 00:53:02.949 --> 00:53:05.090 All right, good. Well, we'll be looking forward to that. 1059 00:53:05.579 --> 00:53:08.219 Jeremy Miller, thank you very much for catching us up 1060 00:53:08.300 --> 00:53:09.900 on everything that's going on in Critterland. 1061 00:53:10.039 --> 00:53:11.679 All right, guys. Thank you so much. 1062 00:53:11.699 --> 00:53:12.059 You bet. 1063 00:53:12.179 --> 00:53:14.340 And we'll talk to you next time on. 1064 00:53:14.320 --> 00:53:14.599 NET Rocks. 1065 00:53:14.619 --> 00:53:15.190 NET Rocks. 1066 00:53:36.119 --> 00:53:38.820 NET Rocks is brought to you by Franklin's Net and 1067 00:53:38.880 --> 00:53:43.610 produced by Plop Studios, a full-service audio, video, and post-production 1068 00:53:43.639 --> 00:53:47.650 facility located physically in New London, Connecticut, and, of course, 1069 00:53:47.710 --> 00:53:51.329 in the cloud. Online at pwop.com. 1070 00:53:52.050 --> 00:53:59.929 Visit our website at dotnetrocks.com for RSS feeds, downloads, mobile apps, comments, 1071 00:54:00.369 --> 00:54:02.849 and access to the full archives going back to show 1072 00:54:02.889 --> 00:54:05.630 number one recorded in September 2002. 1073 00:54:05.630 --> 00:54:08.929 And make sure you check out our sponsors. They keep 1074 00:54:08.989 --> 00:54:12.150 us in business. Now go write some code. See you 1075 00:54:12.210 --> 00:54:12.610 next time.