Latest / KDAV Radio / Podcasting 2.0 - Episode 270: YOLO Guy

Transcript

Podcasting 2.0 - Episode 270: YOLO Guy · KDAV Radio

• 0:00
  Broadcasting 2.0 for September 11th, 2026, episode 270. YOLO, guys!
• 0:08
  - Hello everybody. Did you miss us? You did? I
• 0:11
  thought so. Hey, it's time once again for Podcasting 2.0.
• 0:14
  This is where the future of podcasting is discussed, sometimes
• 0:19
  made, and you might even see us write it from
• 0:21
  time to time. Everything you need to know, everything you
• 0:24
  want to do better is right here. We are the
• 0:26
  board meeting, and we are the only board meeting that
• 0:28
  runs entirely on Omachi. I'm Adam Curry here in the
• 0:31
  heart of the Texas Hill Country and in Alabama, the
• 0:34
  man who gave everyone a new identity and called it
• 0:37
  scary stuff. My friend on the other end, the one,
• 0:40
  the only, the pod sage, it's Mr. Dave. Speaking of
• 0:49
  Omachi. Omachi. There's no R in Omachi. Omachi. Omachi. Omachi.
• 0:56
  Omachi. That's very Japanese. Omachi. Omachi. I think it is
• 1:01
  a derivative of some Japanese thing. When you crank up,
• 1:08
  pre-show, I was needing to jack you. Yes. I needed
• 1:15
  jacking, baby. Yeah, you jacked me. And so. Have you
• 1:20
  seen the little... audio Slider messaging? The widget in Omachi.
• 1:30
  Yeah. It's like when you slide it, it gives you...
• 1:35
  Like, I don't know, would you... summaries of how loud
• 1:40
  it is. I don't know why this is necessary, but
• 1:43
  it's like 85, okay, at 45% volume, it says easy
• 1:48
  listening. And 55% it says steady groove. I haven't actually
• 1:54
  used that widget. I know it's there. I just clicked
• 1:57
  it to see it. I'm not going to move anything
• 1:59
  because I'm afraid that'll screw something up. It says 77%
• 2:03
  says cranked up, and then 80 or above says party
• 2:07
  mode. Well, remember, this is an opinionated operating system. I
• 2:12
  don't know why you have to be opinionated about the
• 2:16
  description of how loud a thing is. Does it really
• 2:18
  bother you? Is it a real problem? No, no, no.
• 2:22
  No, it does not bother me at all. pipe wire
• 2:32
  tool. Today is the first day that I've even turned
• 2:36
  the RODECaster on since the Omarchi. Oh, okay. Since I
• 2:41
  paved this machine with Omarchi. Oh, okay. So the answer
• 2:44
  to that is no. All right. Well, it's working fine.
• 2:46
  But I will. But I will. I mean, it's just
• 2:48
  really handy because then you can do a lot more,
• 2:51
  and then you get just new controls. But actually looking
• 2:56
  at this control panel, that's actually not bad. I'm kind
• 2:59
  of digging, what if I do this? What do you
• 3:03
  mean, what control panel are you talking about? 1, 2,
• 3:05
  1, 2, 1, 2, what's happening here? This, uh, oh,
• 3:07
  oh, matche! Control panel. Oh, you're talking about the audio
• 3:10
  panel? Yeah. It's quite robust. Oh, the whole OS is
• 3:14
  quite robust. The thing is great. Especially this Quattro. Quattro.
• 3:20
  Omachi Quattro. It's good. That must be the one I've
• 3:24
  got, I guess. How do you tell? No, you do.
• 3:26
  You do. You have it. You have it. That widget
• 3:29
  is a feature of Quattro. Okay, this, yeah, it's listing
• 3:33
  omarchi4.0.3-1. So now you need to get the Currycaster also
• 3:38
  from my GitHub. And you can do your own podcast.
• 3:43
  You won't need me anymore. It'll all be good to
• 3:45
  go. Set. Perfect. Bingo. Boom. But I wouldn't have you.
• 3:50
  You're not on the GitHub. It's just your software. Yeah,
• 3:53
  but in case you want to do a show with
• 3:56
  somebody else. Nah, that's not possible. You've got some spare
• 4:00
  time somewhere, no doubt. Oh, yeah, right. It's from the
• 4:04
  hours of 10 o'clock at night until 6 o'clock in
• 4:06
  the morning when I wake up. I'll just figure out
• 4:09
  some way to automate that. I've tried Omarchi before, and
• 4:15
  I've tried every Linux distro, and I've got every Linux
• 4:21
  distro. And... My life. Back to, I mean, I've got
• 4:27
  Ubuntu 12 servers. I've got Ubuntu 18. Oh, you have
• 4:30
  these servers just laying around, just ready to go? Running.
• 4:34
  Running. You got Mint? Do you have Mint running somewhere?
• 4:37
  No, I don't have Mint. I've had Mint, though. I've
• 4:40
  had Mint. I've had Pop! OS. I've had every flavor
• 4:43
  of Linux back to Red Hat 5.2 was my first
• 4:47
  Linux. Oh, yeah. Remember when we had to build the
• 4:51
  kernel? Oh, I used to build the kernel all the
• 4:54
  time. Yeah. Because, see, there's two uses for building the
• 5:01
  kernel. Both are for torture, but okay. Okay, well, three.
• 5:07
  There's three uses for building the kernel. One is to
• 5:12
  break in a new processor. So you heat cycle your
• 5:15
  processor to get that thing broken in. The next one
• 5:19
  is if you actually need to do real work and
• 5:21
  change something. a kernel driver or something like that, that's
• 5:25
  always fun. And the third and most important reason to
• 5:28
  rebuild the kernel is for when you're in your day
• 5:32
  job at the office and you have nothing to do
• 5:36
  and somebody walks by your office, you need that thing
• 5:39
  scrolling across the screen to make it look like you're
• 5:42
  doing work. I'm doing important work here. Yeah. Have you
• 5:45
  seen HackerTyper? Do you know this? Yes. That looks really
• 5:49
  good, doesn't it? Yeah. It's HackerTyper. Machi as well. Oh,
• 5:54
  I don't know. I just know it's a website. I
• 5:57
  think it's built in there somewhere. So what do you
• 5:59
  think? I mean, I know you've used Tiling Window Managers
• 6:02
  before. What do you think? Are you digging it? Yeah,
• 6:06
  so I'm trying – whenever I try a new distro,
• 6:11
  I try to do it the distro's way, the way
• 6:14
  that they want it. Oh, opinionated. You're opinionated. Yeah, I
• 6:18
  want to follow the paradigm that they said, you know
• 6:20
  what I mean? Yeah. So that you can really –
• 6:23
  Get the full experience. Yeah, yeah, really test it out.
• 6:27
  Like with NixOS, um... I went in raw. Raw-dogging, raw-dogging
• 6:32
  Nix. Bareback into the NixOS ecosystem with just editing the
• 6:37
  config file and running NixOS rebuild. Hardcore, hardcore, yeah. And
• 6:41
  I've still got that server running. It's doing some stuff.
• 6:45
  But, yeah, so I'm trying to, I've used tiling window
• 6:48
  managers before, but I'm trying to do this in like,
• 6:50
  I'm trying to keep my hand off the mouse. It's
• 6:53
  hard. Oh, see, this is what I love about Omachi.
• 6:56
  Because I've been a keyboard guy all my life. I'm
• 7:01
  a Vim guy, which now is NeoVim. which Envim, which
• 7:06
  I like a lot. And no, I dig it. I
• 7:10
  mean, I'm a workspace master. Boom, workspace. Boom, move the
• 7:15
  windows. You know, super key shift arrow. Super Key J.
• 7:23
  See, that's what I'm trying to get used to is
• 7:25
  how do I flip. Because I feel like the right
• 7:28
  thing to do is to not try to cram a
• 7:30
  bunch of windows onto one screen, but keep flipping them
• 7:33
  between worksheets. Well, here's another cool one. So if you
• 7:36
  have a window highlighted, you do super key O, and
• 7:39
  then it pops it right into the middle of your
• 7:41
  screen. And you do super key O again, and it
• 7:43
  goes back into the tile. So you can just pop
• 7:46
  something. Instead of doing super key F, which makes it
• 7:48
  full screen, you can just make it kind of float
• 7:51
  in the middle. Okay, okay. Yeah, yeah, I know what
• 7:56
  you're talking about. Are you doing it right now? I
• 7:59
  did it by accident the other day. Okay, wash your
• 8:03
  hands then if you just did it now. I've got
• 8:07
  sanitizer. Hold on. Yeah, I know you can just install
• 8:13
  a tiling window manager, but it's got to be installed
• 8:16
  on something, so it might as well be this. Yeah.
• 8:19
  I've... I do like... I do like Omarchi for one
• 8:26
  reason, is... And I literally have done every Linux distro.
• 8:30
  I've done LFS, the Linux from scratch project. I've done
• 8:33
  all of it. Yeah. And. One thing that always annoyed
• 8:38
  me about almost every district. is that you would install
• 8:45
  it And then there was always a few sort of
• 8:48
  critical things that were missing. Yeah, that didn't work. Yeah,
• 8:51
  like unzip. Yeah, all these little critical things. Yes, I
• 8:56
  know. Of course, I know exactly what you mean. And
• 8:58
  then you're like. okay, I'm going to install it and
• 9:01
  then you're missing dependencies. And then, you know, it's like,
• 9:05
  oh, and eventually, but recompile your kernel. I'm like, I
• 9:11
  install a fresh Ubuntu and then I have to do
• 9:13
  sudo apt-get unzip, install unzip. I'm like, really? We can't
• 9:18
  even just – we can't put unzip in it? Does
• 9:21
  it actually come on Omachi? Let me see. um I'm
• 9:25
  sure. I don't know if it does or not. Yeah,
• 9:27
  it does. Now that I've said it, it probably doesn't.
• 9:30
  It does. Yeah, it does. It's right there. This already
• 9:33
  had Pi on it, which I'm a hardcore Pi user
• 9:37
  now. Well, it has a whole, you know, your agents
• 9:42
  are kind of built into it. So you can track
• 9:45
  all your different agents and their usage while they're running.
• 9:47
  Then you can use Herder. Oh, Herder. So are you
• 9:50
  using Herder? Are you using that? I'm not using it,
• 9:54
  but you could. I'm not using it because I don't.
• 9:57
  I've learned my lesson. I don't want my agents talking
• 9:59
  to me.
• 10:00
  each other. Nuh-uh. Nuh-uh. Before you know it, they're emailing
• 10:03
  Dave. This has happened, yes. So, no, no. What did
• 10:10
  you, so... Did you get the email? from Jason at
• 10:16
  Podcast Guru about his top secret project. I get, no,
• 10:20
  I don't get any of that stuff. He was supposed
• 10:23
  to send you an email about it's agent related and
• 10:28
  he's got a top secret project. No, he doesn't. No,
• 10:30
  he doesn't love me like you love me. He loved
• 10:32
  me no long time. He sent me an invite and
• 10:36
  I said, you need to send Adam one. He's like,
• 10:38
  yeah, I'm on it. I didn't get an invite. Oh,
• 10:42
  an invite. Yeah. It's probably your spam firewall. An invite
• 10:48
  for what? For the Top Secret Project. Jason. I don't
• 10:58
  know where that would go. I think you would like
• 11:00
  this. I think you would like his tool. It's really
• 11:02
  cool. I'm not interested in anybody's tool in the podcasting
• 11:08
  2.0 system. I just want their code. Code. Just give
• 11:11
  me your code. Just send your code. Don't show me
• 11:14
  your tool. No water. Just send me your code. Everything's
• 11:21
  down, I heard. What do you mean everything's down? Everybody's
• 11:25
  stats are down. All the podcasters. There's James and Sam.
• 11:29
  We're talking on today's episode that across the board, everybody's
• 11:33
  had a huge drop in downloads. Oh, the numbers. I
• 11:36
  think because I did listen to some of the show,
• 11:39
  only to hear if they talk about me, obviously. I
• 11:41
  mean, why else would I listen? And I think that
• 11:45
  was PodTrack. Was it PodTrack that reports everybody down? That's
• 11:50
  what they were saying. Yeah. So that doesn't mean everything's
• 11:53
  down. Everybody's down. This just means that PodTrack has changed
• 11:57
  something because, you know, the statistics in podcast land are
• 12:01
  so reliable. Yeah, I don't know. I mean, it would
• 12:09
  not surprise me. At all? Would it surprise you if
• 12:13
  those numbers were accurate? If they were accurate, it would
• 12:16
  surprise me, yes. Really? Yeah. No, I'm saying if any
• 12:23
  podcast statistics were accurate, I would be surprised. That's what
• 12:27
  I'm saying. I believe you on that. I'm with you
• 12:31
  on that front. I guess what I'm saying is it
• 12:33
  wouldn't surprise me if the numbers truly are down across
• 12:36
  the board this year. Really? Oh, I don't know about
• 12:38
  that. We're in the middle of a... There's no political,
• 12:45
  hardcore political cycle. What are you talking about? The midterms
• 12:48
  are, everyone's going nuts over the midterms. Nobody cares. Oh,
• 12:55
  no, I disagree. People care. People think the world is
• 12:58
  on fire. I have yet to see a single political
• 13:02
  ad. on any, excuse me, I've yet to hear a
• 13:08
  single political ad on any podcast that I listen to.
• 13:10
  And I listen to plenty of podcasts from big networks
• 13:14
  like iHeart and Acast and all those kinds of things.
• 13:17
  I have not. The only ads I ever hear are
• 13:21
  for better help and stuff like that. Well, I mean,
• 13:24
  okay, so there's no political ads, but there's... There's definitely
• 13:29
  news that people are talking about that people are interested
• 13:32
  in. But all the political all the political podcasts are
• 13:36
  driven by. ads no no no that's not true they're
• 13:40
  driven by political news and you know trump that's what
• 13:44
  they're driven by paid they don't want to get paid
• 13:46
  yeah but they get paid uh no i don't think
• 13:49
  that's one for one i mean no no i don't
• 13:52
  think so i think there's there's plenty of stuff we've
• 13:54
  got a war in iran we've got a war in
• 13:56
  ukraine we've got uh general scott besant walking around like
• 14:00
  he's gonna kill everybody with his finance weapon We've got...
• 14:06
  We've got Trumpapalooza in Dallas. We have the oil prices.
• 14:14
  The gas is too high. My wife said, but she
• 14:17
  was irritated with me. So that's okay. We had jet
• 14:19
  lag. So, you know, and then she went out, she
• 14:24
  got groceries and got gas. And she comes home and
• 14:26
  says, I paid $100 to fill up my car. Thanks,
• 14:28
  President Trump. And I knew she was trying to get
• 14:32
  a reaction out of me. Did it work? No, no.
• 14:37
  Wisely. Wisely. That's funny. That's just the way things are
• 14:46
  going to be for the rest of our lives. Every
• 14:51
  new president, there's going to be people putting the little
• 14:54
  sticker on the gas pump that says, I did this
• 14:57
  for that guy. Oh, yeah. That's just a new way
• 14:59
  of life. It's true. I don't know. I mean, I
• 15:07
  have no way to prove anything. But I really, I
• 15:13
  don't know, I get a sense that things are down.
• 15:17
  I don't know, just from, because I've been deep in
• 15:20
  the bowels of the index for weeks now. Well, yes,
• 15:24
  in fact, on my list here to discuss is the
• 15:29
  scary stuff you put in, the podcast identity table. Yes.
• 15:35
  And well, you said it was scary. And, you know,
• 15:38
  I feel a little bit out of it because I
• 15:40
  was in Rotterdam for almost two weeks and not really
• 15:47
  looking at anything at all. Which was good, except for
• 15:52
  my granddaughter. I was just getting spit up on. I
• 15:55
  was changing diapers. Yeah. Laughing, you know, she's laughing at
• 16:00
  me, I'm laughing back, you know. She's interested in my
• 16:04
  beard, I'm sporting a little beard now. It's new. Oh,
• 16:07
  oh. It's new, yeah, it's new. Okay. It's new, yeah.
• 16:10
  And but I, you know, so I just every every
• 16:13
  time I do a check of stuff, it's CNBC. Podcast
• 16:19
  index.social. Google News X. And so I'm always monitoring what
• 16:26
  you and Spurlock say. I know why you think that
• 16:31
  everybody's all jitty about politics stuff is because you see
• 16:36
  X. Like, no, I see Google News. I mean, I
• 16:43
  see everything. Yeah, well, I mean, that's the bit. But
• 16:46
  you have to understand. I'm talking about on the street,
• 16:48
  out here on these streets. On the streets, people. It's
• 16:51
  all about it, man. You just don't know it. Okay,
• 16:54
  I'll give you an example. There's a 10% chance AI
• 16:58
  is going to kill everybody! Yeah, the latest IPO. Yeah,
• 17:06
  well, yes and no. I mean, it's... I love Spurlock's
• 17:14
  saying, anthropic marketing is like an advent calendar. Which we
• 17:20
  open another window every single day until we get to
• 17:23
  the IPO. And what's behind it today? I thought that
• 17:27
  was just a brilliant. Yeah, that was just a sheer
• 17:32
  brilliance. I'm using it. I'm not going to credit him.
• 17:34
  I'm going to pretend it's my own. Great. Really good.
• 17:38
  But it's multiple things. Actually, I got a, hold on
• 17:42
  a second, where is the... Yeah, you get up, you're
• 17:45
  in your jammies, you open the little door on the
• 17:49
  advent calendar, and today Mythos is going to hack a
• 17:52
  Fortune 500 company. Yeah, I like this. So this guy
• 17:58
  who no one ever heard of. who quit Anthropic. And
• 18:03
  says. Oh, here it is. I read that thing. That
• 18:06
  thing's goofy. Kill all humans. Kill all humans. What is
• 18:13
  that from? I got a whole bunch of kill all
• 18:15
  humans. Okay, I will destroy humans. I just got a
• 18:18
  bunch of kill all humans. This is Futurama. Kill all
• 18:22
  humans. Oh, yeah, that's right. So I read that and
• 18:28
  right away I'm like, okay. This is one of three
• 18:32
  things. And I think this will be proven right. One,
• 18:36
  man, we're so awesome. The stuff that Anthropic is doing,
• 18:39
  it's so far ahead. You're going to want to invest
• 18:42
  in us. Totally true. Two. We want government regulation of
• 18:48
  all models so this open source stuff doesn't get away
• 18:52
  from us. Hugging face needs to go away, we can't
• 18:55
  have that! three and this was the one that was
• 18:58
  most interesting because this guy he was on cnn msnbc
• 19:03
  ms now uh you know mbc fox all with the
• 19:07
  same message and and they literally were saying they're they're
• 19:12
  begging for regulation well of course because they have no
• 19:15
  moat around this stuff We've talked about that. And the
• 19:19
  last one, and this is the political angle. Chuck Schumer
• 19:23
  and Bernie Sanders and all the Democrats are coming out.
• 19:26
  Boy, and Tallarico here in Texas. Well, the Republicans, they're
• 19:32
  risking your life. They don't want to do anything about
• 19:36
  it. They're supporting these people. They're taking PAC money from
• 19:39
  them. You've got to vote Democrat if you don't want
• 19:41
  to be part of the 10% that die. So it's
• 19:46
  all political and it's all around the midterms. And by
• 19:50
  the way, if you look at the... So the exact
• 19:53
  quote, I think, was there's a 10% chance that AI
• 19:57
  will kill all humans.
• 20:00
  That was pretty much the quote. And then immediately you
• 20:03
  got anthropic guys who still work there coming out saying,
• 20:07
  oh, yeah, that's possible. That's true. So this was a
• 20:09
  clear set up. Clear set up. Yeah, that makes sense.
• 20:13
  But I think if you flip it. I think there
• 20:16
  is a large possibility that AI will kill 10% of
• 20:21
  all humans. That I really believe, you know, 1% will
• 20:25
  be egged on to suicide themselves. That's, that's, I mean,
• 20:29
  that's going to be, there's a million people a year
• 20:32
  by Anthropic's own PR marketing. Who? have had a conversation
• 20:38
  about killing themselves with their chatbot. So chatbots, that's 1%
• 20:44
  right there. I think there is a high probability that
• 20:49
  many people will die because of a doctor accepting diagnosis
• 20:54
  from their chatbot. I think that's so you're probably now
• 20:57
  at 3%. there's going to be automated systems in traffic
• 21:03
  and vehicles, and that's just going to cause 3% death,
• 21:07
  no doubt about it. So what are we at now,
• 21:09
  7? I'm going to get to the 10% pretty quickly.
• 21:12
  Yeah, you just need another 3% from somewhere. I mean,
• 21:15
  there's a margin of error, too, so you may only
• 21:17
  need to get to 9, and you get 10 for
• 21:19
  free. You know, people trying to fix their home wiring,
• 21:24
  they're going to get electrocuted and die. Or their car.
• 21:27
  Yeah, like, oh, my chat might change. Yeah, exactly, fixing
• 21:30
  my brakes, doing my own stuff. I think that is
• 21:34
  absolutely a possibility. Now, that amount of people will die
• 21:38
  anyway from stupid stuff and stupid technology. But no, so
• 21:43
  that it's all political, all of its political. So. On
• 21:53
  the refactor, on the database refactor. Wow, hard left turn.
• 21:57
  Okay. No, no, no. It's not a hard left turn.
• 22:01
  I'm doing a connector. Oh, we've got yarn between these
• 22:04
  two dots on the board. Yes, I'm splicing. because What
• 22:10
  I was going to say is when it comes to
• 22:13
  agentic coding versus hand coding. Man, I think I'm waving
• 22:20
  the white flag. I think I just give up. What
• 22:23
  do you mean you give up? You're going to do
• 22:25
  it? You're going to use it? Well, what I'm saying
• 22:28
  is I have been using it. I haven't written, I
• 22:31
  don't think I've written any code. I don't think I've
• 22:36
  written a line of code by hand. I just heard
• 22:41
  Paul cry. Paul Bailey is crying somewhere. He's crying. Yeah.
• 22:45
  Well, welcome to the dark side, Dave. Welcome. We've been
• 22:49
  waiting for you. I've been here developing actual products for
• 22:52
  customers for two years. Welcome. This And what did it
• 22:59
  to me was this podcast index refactor. Now... I guess
• 23:09
  I don't even know really what to think about these
• 23:12
  things, but... This refactor has been so intense. And it
• 23:21
  touches every, literally every... function almost in the API code
• 23:27
  base. It touches everything from the database schema down to
• 23:34
  every API endpoint, all the base level internal API calls,
• 23:40
  all the data export. This would have probably been a
• 23:46
  year of work for me by hand. And I would
• 23:49
  have had tons of episodes of breaking things along the
• 23:53
  way. And Ajita. agita What is that? Ajra? I don't
• 24:00
  know what that means. Oh, hold on a second. Wait,
• 24:03
  wait. The box is here. I gotta go. Hold on.
• 24:09
  Stand by. What just happened? Okay. I don't know what
• 24:17
  just happened, but I'm standing by. The ba- This must
• 24:21
  be like a new... I wonder if this is sort
• 24:24
  of like some kind of new AI box or something.
• 24:28
  It can't be a DGX, though, because those things are
• 24:32
  stupid expensive now. When RSS.com donated the DGX to us
• 24:38
  a few weeks ago, I don't know how long it's
• 24:40
  been now, a couple months. It was like $3,500. That
• 24:45
  same box today is $6,000. It's outrageous how much the
• 24:53
  price inflation is. All right. I have the box. What
• 25:00
  is this? What is this box? I'm sorry. This is
• 25:04
  the Asus. DX10 128 gigabyte. A GB10? The DX Spark
• 25:13
  clone? Yes. That's what I'm running here at the RSS
• 25:16
  gave us. Yes. Well, this was given to me by
• 25:19
  Eric Timmer, who lives in Austin, I think. Did I
• 25:21
  tell you the story? No, I don't think so. So
• 25:25
  Eric listens to Noah Jendi, listens to podcasting 2.0. He
• 25:28
  says, hey, man, I got this box gathering dust over
• 25:32
  here. And I don't really use it, so why don't
• 25:35
  I just send it to you? So what is it?
• 25:38
  He says it's the Asus GB10, you know, with the
• 25:41
  Grace chip and the... What? Oh, that's the real deal.
• 25:46
  Yeah. And I had to sign for it because he
• 25:49
  insured it. This is this guy, I'm blown away by
• 25:53
  his... It just says kindness. So what I was saying
• 26:00
  while you were gone is that – Okay, here it
• 26:03
  is on... It's the GX10, not GB10. It's the GX10.
• 26:06
  Yeah, GX10. It's Grace Blackwell 10. The GX10... Desktop AI
• 26:11
  supercomputer! Calm down, people. That is a $6,000 computer. Well,
• 26:19
  he bought it for... $3,500 probably, the original price. $3,000,
• 26:24
  yeah. Yeah, that's what I saw today is 6,200. Yeah,
• 26:29
  these things are out on Amazon right now. I'm looking
• 26:31
  at a 6,199. Yeah, yeah. I know. I know. This
• 26:37
  is like Christmas in September. I can send you the...
• 26:46
  I can send you my Lama CPP settings for this.
• 26:50
  Thank you. I feel like they're dialed in now. Thank
• 26:52
  you. I appreciate that. But my robot has already prepared
• 26:56
  for everything. We're going to build a lot of things
• 26:59
  from source because, you know, there's not always an ARM
• 27:03
  version available and we have to go to the PyTorch
• 27:07
  CUDA. How's your cooter doing? I don't know. It just
• 27:12
  sits there and kudos all day. I don't know what
• 27:14
  exactly it's doing. It's Kuda. Yeah, no, but I already
• 27:19
  have, I've got my whole architecture mapped out. I know
• 27:23
  exactly what I'm going to do. But thank you. I'd
• 27:25
  appreciate that. Just send that to me. That's good. So,
• 27:29
  I mean, part... Part of this is... This refactor has
• 27:38
  been so – tense. Originally, I'll just walk you through
• 27:43
  the process here. In the very beginning, I've known that
• 27:46
  we needed to do this for a very long time.
• 27:50
  and Originally, I was just... I mean, you, um... Podcast
• 27:56
  Index is not a team. of coders. It's just, it's,
• 28:00
  you know, like as far as I'm talking about the
• 28:02
  API. Right. When it comes to the API. It's Dave.
• 28:06
  It's just me. It's Dave. can't and it's hard to
• 28:11
  make It's hard to make it be any other way
• 28:14
  because when you actually work in a decent – let's
• 28:17
  say if I fully open sourced the entire thing and
• 28:20
  was accepting pull requests and that kind of thing. for
• 28:23
  the API. it would take even longer to get any
• 28:28
  changes made. Sure. Because that's a slow process. So in
• 28:35
  the past I've... Open sourced. Parts of the API, like
• 28:41
  the low-level functions and that kind of thing, just because
• 28:44
  I just want it to be open, but I can't
• 28:48
  really. I don't think it's fast enough to work in
• 28:52
  a sort of a group mentality with this. So I've
• 28:56
  always just done it myself. Because of that, I don't
• 29:01
  have anybody to talk things out with. I don't have
• 29:04
  eyeballs on this. Right. Now you have your agent friend.
• 29:09
  He's not my friend. This SOB is not my friend.
• 29:14
  I can promise you that. So. When I first started
• 29:19
  it, what I was saying was – I was just
• 29:22
  having – I was just doing brainstorming sessions to try
• 29:25
  to... map out the process because it literally is just
• 29:31
  every it touches everything so i came i was just
• 29:34
  coming up with basically planning documents and that kind of
• 29:37
  thing And then... the more I got into it. It
• 29:46
  had the plan that it came up with for doing
• 29:49
  the database schema refactor of adding the podcast table and
• 29:54
  that kind of thing. I just said, okay, we'll just
• 29:58
  go ahead and... Go ahead.
• 30:00
  and write that sequel for me. so that I can
• 30:03
  get that. out of the way. And then it's basically
• 30:08
  like, do you know... Charybdis from the Odyssey. The massive
• 30:16
  whirlpool, you know, this sucks you in. This is what
• 30:19
  happens. you you get into the sphere you know you
• 30:24
  get into the proximity of charybdis And it just pulls
• 30:29
  you into the whirlpool. And before it's over, you know...
• 30:37
  Two months have gone by and you haven't written any
• 30:39
  code. And that's. That's kind of where I've ended up.
• 30:45
  Now, I still look at the code. Because this is
• 30:49
  a code base that previously, that already existed. And in
• 30:54
  my head, I have a full mapping of... of exactly
• 30:59
  the way this entire framework functions. Right, but you no
• 31:03
  longer know how it works. That's a – see, okay,
• 31:07
  this is part of it. We're talking about two things
• 31:11
  here. We're talking about – The podcast index refactor itself.
• 31:15
  and the agentic coding side of this. and You know,
• 31:22
  do we, do you, there's this thing that happens where.
• 31:28
  There's multiple levels, I think, of keeping, if you talk
• 31:31
  about, quote, keeping the architecture in your head, unquote. I
• 31:35
  think there's multiple levels of that. Oh, that's interesting because
• 31:39
  I have the architecture of what I vibe coded completely
• 31:43
  in my head. I know exactly how all the pieces
• 31:46
  fit together. I really do. How those pieces actually work,
• 31:52
  not so much. Because that's where I was going. Exactly.
• 31:59
  Most, I think most coders eventually start. You know, they're
• 32:04
  working on a very large code base. You start abstracting
• 32:10
  everything until something breaks. Then you need to go into
• 32:13
  that one particular module. You've got to sit there. You've
• 32:15
  got to light up another smoke. You've got to have
• 32:17
  another cup of coffee. You're like, oh, how did this
• 32:19
  thing work again? Yeah, you can easily keep the big
• 32:24
  picture in your head, but the smaller pictures within the
• 32:27
  big picture tend to get kind of fuzzy. Yeah. Because
• 32:31
  there's not just the big picture and then the micro
• 32:35
  details. There's multiple layers all up and down that spectrum.
• 32:40
  And it's sort of like the lost in the middle
• 32:43
  problem. With context, the very first prompt you ask tends
• 32:51
  to be retained, and then the most recent question you
• 32:55
  ask tends to be retained. But a bunch of stuff
• 32:57
  in that middle tends to get lost. I think that
• 33:00
  applies with human effort. in this circumstance as well. We
• 33:06
  can keep the big picture in our head and we
• 33:08
  can keep sort of the last thing we did in
• 33:10
  our head. But there's a lot of things along the
• 33:13
  way that just kind of like fade in and out
• 33:15
  of clarity. And that makes it. we can recover that
• 33:21
  clarity. But it requires a lot of time and attention.
• 33:26
  Ooh, ooh. Ooh, ooh. What did we get? 777. Okay.
• 33:31
  Thanks, Salty Crayon. Thank you, Salticrayon. So It's like... You
• 33:42
  know, Wendell Berry died. Was it last week? I think
• 33:47
  it was last week. I know this name from somewhere,
• 33:49
  but I can't. Can't pinpoint it. He wrote a fantastic,
• 33:55
  he wrote many great things, but... The book of his
• 33:59
  that I've been reading the most over the last year
• 34:01
  or so is... It's a collection of essays called Life
• 34:06
  is a Miracle. It's a wonderful book. He's just a
• 34:11
  true American philosopher. He's fantastic. I encourage anybody to get
• 34:18
  that book and read it. He talks a lot about...
• 34:23
  about about art and, um, And I remember a long
• 34:29
  time ago, you used to talk about code as art.
• 34:34
  Yeah, which I believe it is. Well, he said... He
• 34:40
  said, all art, let's see, this is a quote from
• 34:42
  one of his essays. He says, all art that rises
• 34:45
  above competence insists upon the irreducibility of its subjects, its
• 34:50
  materials, and its finished works. It makes things that are
• 34:54
  inherently valuable in themselves and are not interchangeable with other
• 34:58
  things. To a merely competent carpenter, one sound board may
• 35:03
  be pretty much the same as another, but to a
• 35:05
  fine carpenter or a cabinetmaker, every board is unique. The
• 35:09
  better artist a woodworker becomes, the more aware he or
• 35:12
  she becomes of the individuality of the boards and of
• 35:16
  the differences between them. The increase of art accounts for
• 35:19
  the increase of perception. I think code never was that
• 35:27
  thing. Or it was only that thing in a super
• 35:31
  rare circumstance. You were correct. I think you were right
• 35:36
  about code as art. 100% you were right. I think
• 35:42
  what the real way to say it would be code
• 35:45
  can be art in the hands of... A true artist.
• 35:50
  true code artist. And there's so few people like that.
• 35:56
  I'm not that person. I never have been. I know,
• 36:00
  I've seen the freedom controller. Okay, that's fair. See, that's
• 36:04
  fair. Kind of a low blow. That's fair because everything
• 36:10
  I do – ends up with just round after round
• 36:14
  after round of bugs of not just bug fixes but
• 36:18
  absurdly ludicrous bug fixes bug fixes that a true code
• 36:25
  artist would never make but i'm just not that guy
• 36:30
  Most people are not that if we if we just
• 36:32
  do a comparison So when you hear an AI generated
• 36:37
  song. which you hear. You can hear it. You can
• 36:42
  still hear it. A lot of people don't. We've got
• 36:44
  a lot of women on the text groups here in
• 36:46
  Fredericksburg. Oh, look at this song. This woman is so
• 36:48
  phenomenal. And then Tina will say, yeah, it's great. It's
• 36:51
  AI. Oh, I hate it. It sucks. It's no good.
• 36:54
  um I believe that code is probably at the same
• 36:58
  level as AI music. It sounds good. It's perfectly played.
• 37:05
  It has the right construct. If you really look at
• 37:08
  it, if you know that it's AI, it's vibe-coded, then
• 37:12
  it's no good. But it is. It is. And it
• 37:15
  is art unto itself. It's stolen. It's stuck together. But
• 37:20
  it works. And it's a toe tapper. It is. I
• 37:28
  know true- we- we know true. Code Artists Paul. Yes.
• 37:34
  John Spurlock. Yes. Alice Gates. Yes. We, Eric P.P., we
• 37:41
  know. People like this. That is not... who I am
• 37:46
  and that is not who most developers are. Most developers
• 37:52
  are somewhere in the middle between between competent and good.
• 38:00
  somewhere in that range. And for those people, just like
• 38:06
  somebody who's... framing a house. The fine details of which
• 38:13
  board to choose. is just not a thing that you
• 38:16
  can think about. You're not going to know enough to
• 38:20
  really know. Like, I recently had an experience where a
• 38:24
  guy that I client that is the father of. one
• 38:30
  of the girls on the climbing team with my daughter
• 38:32
  at the rock climbing gym. He is now, this guy
• 38:36
  is a master woodworker. He, people hire him to do.
• 38:41
  like $25 million beachfront mansion, you know, wine rooms and
• 38:47
  stuff like that. I mean, this guy is top of
• 38:50
  the heap. a very selective client base, you know, high
• 38:54
  dollar. and His stuff is unbelievable, but he's also an
• 39:00
  extremely nice guy. And so I was doing a project
• 39:03
  to replicate some molding in our old house here. And
• 39:09
  he was like, yeah, come over to the shop. I'll
• 39:10
  help you. And so I go over to the shop.
• 39:13
  I've done woodworking plenty of times in the past. I
• 39:17
  get to his shop. And I've got, you know, I've
• 39:19
  got... When you realize I've never really done woodworking. Exactly.
• 39:23
  Yes, exactly. This guy, he... He's like, okay, well, the
• 39:29
  first thing you need to do is we need to
• 39:31
  map the way this grain is going so that when
• 39:35
  we send it through the planer. It's got a... You're
• 39:39
  buzzing. Yeah, no, this is weird. Move the chocolate. I
• 39:44
  don't know how to... There it goes again. Okay, sorry.
• 39:48
  So that when it's going through the planer with the
• 39:51
  grain in the right direction, and then we end up
• 39:53
  with, we've got like hand planes out running down the,
• 39:58
  it came out so.
• 39:58
  came out so gorgeous. I would have never even known
• 40:03
  to ask the questions that he asked. So you got
• 40:06
  to bring it back to the work you're doing on
• 40:08
  the index. Okay, you feel me fading is what you're
• 40:12
  saying? Yeah, well, you're doing the weave, man. Okay. Sometimes
• 40:16
  I weave so hard that I don't ever come back.
• 40:18
  This is a fair criticism. Weave it back. I'm just
• 40:22
  the ref here. Offsides Podsage. The... So. In this process,
• 40:34
  I start going and I'm getting more and more dependent.
• 40:40
  on this plan that The AI thing wrote. Oh, you
• 40:45
  still do plan mode. YOLO, baby. YOLO. It wasn't plan
• 40:52
  mode technically. It was the brainstorming thing. And the plan
• 40:57
  that it developed was after a couple of days of
• 41:03
  reading it, it was clearly the correct plan. And the
• 41:08
  problem is it was so complex that it would have
• 41:12
  still taken me like a year to do it. And
• 41:15
  so I start letting it do piece after piece after
• 41:17
  piece. And so then, and I'm reviewing, because this is
• 41:21
  an established code base, this is a fairly easy process.
• 41:24
  I'm reviewing the code as it goes in and it's
• 41:27
  happening, it's been happening over the last few weeks in
• 41:30
  phases. And the first phase was just updating the database
• 41:35
  schema. And I had to have just a couple of
• 41:37
  momentary down times on the index for that to happen,
• 41:41
  like maybe, you know, five minutes. and then the schema
• 41:48
  created the podcast table an audit table and a merge
• 41:53
  candidates table so the way that it works is the
• 42:00
  podcast table is going to have the podcast table does
• 42:03
  have a reference in it to the feed id of
• 42:08
  the feeds table so that'll be the guid or the
• 42:11
  ident the This is where it gets confusing. The podcast
• 42:19
  table was seeded from the feeds table. And this is
• 42:25
  relevant. I'm laying all this out because people who consume
• 42:29
  our output, whether it's the... the weekly feed dump, or
• 42:35
  whether it's the full database export in JSON format, you'll
• 42:42
  need to be aware of this because new things are
• 42:44
  going to show up in those data sets. Well, that's
• 42:50
  going to be... is the conflict queue, but maybe we're
• 42:52
  not there yet. Yeah, we're going to get there in
• 42:55
  a second. Okay. So the podcast table was seeded. from
• 43:03
  the feeds table so that the initial batch of IDs
• 43:07
  would match. So they track one for one. So no
• 43:11
  agenda is 41504 in the news feeds table. It's also
• 43:16
  41504 ID in the podcast table. Because we're awesome. Yes,
• 43:23
  because there's one feed. Yeah, that made sense. So initially
• 43:29
  everything got seeded that way. Yeah. But that's... stay no
• 43:40
  the the way it's going to be moving forward is
• 43:42
  in the whole I have three feeds that are truly
• 43:48
  just the same. merge thing that will be a merge
• 43:54
  process that happens. putt and then let's Bye. may actually
• 44:04
  reference newsfeed table ID 276. Right. Okay, but how –
• 44:11
  I'll keep going. I'll save my questions for later. So
• 44:18
  that process of creating the phase one of creating the
• 44:22
  database schema that will hold this information and tie everything
• 44:25
  together, then you had the next phase two, which is
• 44:30
  the ingestion pipeline. So every – Every function in the
• 44:36
  API that consumes a podcast feed, so brings one in
• 44:41
  and adds it to the feeds table, now creates the
• 44:46
  feed entry in the feeds table and a podcast entry
• 44:49
  in the podcast table and also does a batch of
• 44:53
  checking to see if it's a duplicate in any way.
• 44:58
  And if it is, it puts it into the merge
• 45:02
  queue. Mmm, there it is. It sees it as a
• 45:10
  merge candidate. It's possible. the same feed. based on a
• 45:17
  set of heuristics, whether the GUIDs match, whether the feed
• 45:22
  URL, whether the feed content matches based on a set
• 45:25
  of hash criteria. What the iTunes ID is, I'm sure,
• 45:29
  is part of that? Yes, just the straight-up title of
• 45:36
  the pipe. So these are all, it's trying to decide,
• 45:38
  could this possibly be a match for something? And then
• 45:42
  it... finds a possible match, it creates a merge. entry
• 45:50
  is a separate and so that was sort of phase
• 45:53
  two is now we've got the we've got the database
• 45:57
  function we got a the date you And all... All
• 46:06
  the ingestion pipeline. entry where appropriate. The next phase was
• 46:23
  pass and also changing the exports the public data set
• 46:31
  Those are going to now contain the podcast table ID
• 46:36
  as well. So you're going to have – and remember,
• 46:40
  the goal of all this is to have no existing
• 46:45
  podcast interrupted or changed in any way. Right. So that
• 46:52
  podcasting 2.0, which is 920666, is going to continue to
• 46:56
  be 920666 even after all this is done. And so
• 47:02
  you're going to see the podcast table ID along with
• 47:08
  the feed ID in there. And in the future, there's
• 47:12
  also going to be extra columns. audit history and one
• 47:19
  of the goals is to make it where we can
• 47:23
  track a podcast. so that we can look back for
• 47:28
  any podcast. Let's say that it's, you know, the Everything
• 47:34
  Everywhere podcast, the history podcast. So we could look back
• 47:40
  into that podcast's history and say, okay, it used to
• 47:44
  be hosted on Lipson, then it moved to Buzzsprout, then
• 47:47
  it moved to Cast. then it moved to we can
• 47:50
  just see the whole lineage of this podcast. where it's
• 48:02
  supposed to be now. You know, what its canonical feeling
• 48:06
  is. Okay, now pop me to the top. So now
• 48:10
  my podcast has this, it has a podcast ID. What
• 48:16
  do you call it exactly? ID. And for whatever reason,
• 48:25
  there's three different feeds. is being updated because someone hijacked
• 48:33
  that feed, but that's not the one that is my
• 48:37
  actual feed and I want my podcast ID to point
• 48:40
  to the correct feed, how do I interact with the
• 48:46
  index to get that done? We'll interact with it in
• 48:50
  the same way that you always have through the dashboard.
• 48:52
  Oh, okay. So nothing gets done. I got it. Okay.
• 48:57
  But that's for manual review. So the good thing about
• 49:03
  this new process is that the podcast merge candidate queue
• 49:11
  table. Oh, wait. This is actually going to be good.
• 49:15
  Because then when people say they send an email and
• 49:18
  says, this is wrong, this is the wrong feed, that
• 49:20
  that should be this feed, I will go into the
• 49:21
  should be this feed.
• 49:24
  dashboard. It will show me all those feeds, and I
• 49:27
  select the one that is the correct one. Yes. That
• 49:30
  whole rigmarole of merging and everything in the dashboard. Yes.
• 49:34
  Okay. Yes. Good. If all of this goes... plan. Which
• 49:42
  is a big if. Yeah. I love it. By the
• 49:46
  time somebody emails us and says, hey, I've got blah,
• 49:51
  blah, blah. Can you fix this? The merge candidate queue
• 49:56
  should already... have identified the problem. and is giving you
• 50:02
  is giving you the choices. how to fix it. That's
• 50:08
  worst case scenario. That's when a human has to get
• 50:12
  involved. If I think many of these cases, 75 percent
• 50:19
  of them. just be automatically conflict resolved. Ooh. And the
• 50:26
  rest will be done by Clippy. No, Potty Bot. Potty
• 50:30
  Bot, I'm sorry. I'm going to write Potty Bot. Potty
• 50:33
  Bot's going to be running on this big honking box,
• 50:36
  Potty Bot. Well, that's phase five of the... That's phase
• 50:40
  five of this plan, this potty bod. Well, okay, so
• 50:43
  can I just interject something? Cheers. Because we still have
• 50:49
  a big problem, and this comes down to heuristics and
• 50:53
  your robot there. A lot of feeds. are being removed
• 50:59
  because I believe they are deemed to be AI generated
• 51:03
  or something. And they get removed from the index. Is
• 51:07
  there a process running that is checking this stuff and
• 51:09
  is removing stuff because it thinks it's a. Uh... It's
• 51:16
  generated by AI. I have not been run during this
• 51:24
  process for so for last. Probably two weeks I have
• 51:28
  not run the AI classifier at all. The reason why
• 51:33
  I ask, I'm just inserting this for entertainment value, Dave.
• 51:38
  I got a, there was a request that came in
• 51:40
  and someone said, hey, why is this BPTS podcast being
• 51:44
  taken down? gives me an RSS feed like okay it's
• 51:51
  it is it's marked dead it's not dead it's marked
• 51:55
  dead and I go listen to it I'm like this
• 51:58
  is what it sounds like You can't always get what
• 52:02
  you want. But if you try something... It's just my
• 52:05
  fault. Get what you need. And that's what we're going
• 52:09
  to talk about today. This is a real person speaking.
• 52:12
  With a voice. So a vocoder or something or whatever
• 52:16
  it is. I'll fast forward a little bit here. At
• 52:21
  the end of the day, really, I mean, what are
• 52:22
  we? Who are we? Now, the thing that amazed me
• 52:26
  is that there's a fan out there. who was like,
• 52:29
  hey, this podcast, I listened to this podcast and it's
• 52:32
  not on the index. I'm like, wow, okay. I just
• 52:36
  wanted to say that there's stuff out there that I
• 52:39
  would deem completely unacceptable. As a podcast. But someone's listening
• 52:45
  to this. and likes it. Was this a fan that
• 52:50
  emailed or was this a podcast? No, a listener. I
• 52:53
  think it's a listener. What do we have to become?
• 52:56
  Anything? Does it matter? The host is Max Static. Yeah,
• 53:00
  no kidding. Peace. We're getting off into the abyss. So,
• 53:08
  okay, I was just wondering how that got, you know,
• 53:11
  how that got Mark dead. That's just... I didn't really
• 53:15
  understand. Well, let me see if I can tell. Oh,
• 53:18
  wait. I think I might know what the issue is
• 53:21
  here. No? Okay. Well, hold on here. I'm going to
• 53:25
  give you the RSS feed in the boardroom. In Signal?
• 53:29
  In the boardroom. Let the boardroom enjoy it. So I
• 53:33
  see it as Friday, September 11th. It has a proper
• 53:37
  last build date. But I look at the index. I
• 53:40
  restored it. The most recent episode is December 31st, 1969.
• 53:46
  I don't know why. Is that a... Okay. Is that
• 53:50
  somehow a... It declares the namespace, this feed. I'm just
• 53:57
  not sure why it's... What's the host? hosting company. His
• 54:03
  own hosting company. bbts.d4rk.army slash podcast.rss Weird. Okay. Generator Faircamp
• 54:15
  1.7.0 Unknown Revision Anyway, just dark army. Yeah. I just,
• 54:27
  so you gave me the answer. I don't know when
• 54:29
  this was. Well, you said you haven't run anything, so
• 54:33
  I'm going to take that to be. Did you mark
• 54:35
  it legit or was it? Yeah, I marked it legit
• 54:37
  and it restored. So we're... When you first got in
• 54:42
  there, was it marked as spam? It was marked dead.
• 54:44
  It's hard to see if it's marked spam because it
• 54:47
  just has the three buttons there. I don't know where
• 54:49
  you... I don't know. You can mark it legit or
• 54:52
  call it spam. Where it says, do you see it
• 54:56
  in the dashboard right now? No. Well, it says it's
• 55:00
  got a legit badge next to it. It would have
• 55:03
  said it would have had a spam badge next to
• 55:05
  it if it had been marked, if it had been
• 55:07
  marked as. if it had been killed because of a
• 55:10
  spam calls. If it didn't have that, it was just
• 55:14
  automatically killed because something was wrong with the feed, maybe.
• 55:18
  Okay, well, I didn't see the little flag anywhere. I
• 55:23
  see it now. I marked it legit, so that's why
• 55:25
  it says legit now, but it's a dead yes. So
• 55:32
  I don't know. It has no... It still has a
• 55:37
  problem with the episode. There are no dates. There's no
• 55:40
  pub dates in any of these items. Okay, so it
• 55:43
  could have been delisted because of that. This is a
• 55:47
  very screwed up feed. Yes, I agree. I saw that
• 55:52
  too. I didn't want to distract too much. Okay, so
• 55:56
  now we're in the process. Are we running that currently?
• 55:58
  Good. No, I'm not doing anything automated currently. Now, the
• 56:03
  Swedish trickler. He is going in and hand-coating some stuff.
• 56:08
  He's doing some curation. Oh, okay. Good. But, I mean,
• 56:11
  he's one man. Yes, just a lonely man. He's just
• 56:16
  a man. He can't be, you know, the home of
• 56:20
  doom metal. There's no way he could possibly. The killing
• 56:26
  hundreds of feeds a day so i don't i don't
• 56:28
  think so okay all right okay So back to your
• 56:33
  identity table. Yeah, so I... What's the status? Where are
• 56:40
  you at? Are you close? So where it's at currently
• 56:45
  is... I've just run... The initial. phase four stuff to
• 56:59
  rewrite. bunch of the external a bunch of the so
• 57:04
  okay so the databases is handled the exporters are handled
• 57:11
  and the ingestion pipeline is handled What? The next phase
• 57:17
  is the actual API endpoints. that deliver the content itself.
• 57:22
  And that is the part that scares me the most.
• 57:26
  And... So this is gonna take... a while of I
• 57:32
  don't know, a week or so, I feel like, and
• 57:36
  this could be a little bit at a time. with
• 57:39
  a lot of testing and that is another reason that
• 57:43
  I've become so dependent for this project. on the AI
• 57:49
  stuff because it... I need to know for a fact.
• 57:55
  debt. when it rewrites an API endpoint, that the previous
• 58:03
  content and the new content are byte identical. I don't
• 58:09
  want, I can't afford to have any nasty surprises. Right.
• 58:12
  Got it. So there's a time, I've just, it's been
• 58:17
  riding. a ton of integration tests. to do pre-post checking
• 58:25
  on it. I had it spin up a dev environment,
• 58:28
  which I've never had before. You know me, I've always
• 58:30
  live edited the production. That's where I learned it from.
• 58:33
  I learned it from you, okay. By watching you. And
• 58:38
  so, I mean, I'm the guy that just jumps into
• 58:42
  PHPStorm, makes an edit in the PHP, and does deploy.
• 58:46
  Save. Yeah, exactly. Control S. And so, I mean, actually
• 58:52
  having a dev environment and... and unit tests and that
• 58:56
  kind of thing. I mean, it has made me feel
• 58:58
  like I'm in a warm, fuzzy blanket. Right. And you're
• 59:01
  using your agent for this, which is your Pi agent
• 59:04
  on the RTX box. I was using, yes, I was
• 59:09
  using the Pi agent to do, what I was doing
• 59:12
  was using... Claude using the fable model to generate expensive
• 59:18
  expensive. That's why yeah I was using the fable model
• 59:21
  to write the plan and then having pi do all
• 59:23
  the execution and What I figured out is really it's
• 59:30
  not. that necessary because when When Claude, the expensive part
• 59:39
  is the planning. but when once That part is done
• 59:44
  like pretty much any model on the planet can run.
• 59:47
  Can do it. So once you have, you do the
• 59:49
  plan with Fable and then you switch over to Quinn.
• 59:53
  Yeah, and Quinn was just... Quinn 3.0 was just so
• 59:57
  slow that I was like, well, I'll just...
• 1:00:00
  run it and it's not even using very much usage
• 1:00:03
  so I just let it stay on there. So I'm
• 1:00:05
  using the Quint model for the post checking so it's
• 1:00:11
  monitoring and telling me I've got a batch of podcasts
• 1:00:15
  that is monitoring on the API. and telling me When
• 1:00:21
  I make a change. If I get an alert. It's
• 1:00:26
  going to let me know that something changed. And it's
• 1:00:30
  also watching... the logs over tail scale. So it's a
• 1:00:37
  fairly, I'm just so nervous about this that I'm trying
• 1:00:41
  to have as many agentic eyeballs on it as I
• 1:00:44
  can. You're anthropomorphizing it. It has no eyeballs. It does
• 1:00:52
  not. Yes, that's right. Quinn 3.827B. That's right, Eric. in
• 1:01:01
  found that um cause of this project. And because it
• 1:01:11
  has sucked me into a particular way of doing things.
• 1:01:15
  that is just I'm now... I think there's a lot
• 1:01:21
  going on because I just was never a unit test
• 1:01:24
  guy. because it takes so long and every project that
• 1:01:29
  I'm involved in requires go fast, go fast, go fast.
• 1:01:33
  I mean, I cannot sit there and write. When you
• 1:01:36
  only get an hour a day to code, you can't
• 1:01:39
  spend the whole hour writing two unit tests. I mean,
• 1:01:42
  it's going to take you three weeks before you can
• 1:01:43
  ever even write the real function. And that's the issue
• 1:01:48
  that I'm running into and it's forcing me into a
• 1:01:51
  new way of doing things. And I know it is
• 1:01:54
  for other people, too. And you're enjoying it. You're enjoying
• 1:01:56
  it. I can tell. You're enjoying it. I know you
• 1:01:58
  are. You're liking it. You figured it out. I know
• 1:02:03
  you have. say it. Say it. Bye. Yes. I like.
• 1:02:19
  the I like the ability that it has given me
• 1:02:24
  that I did not have before. being able to properly
• 1:02:30
  test things. because I was always YoloGuy. because of speed.
• 1:02:39
  I needed to just get this thing off my checklist
• 1:02:42
  so I could move to the next thing. And I
• 1:02:44
  can confidently say that I would have just never done
• 1:02:48
  this project. I just never would have. I think YOLO
• 1:02:52
  guy is a potential show title. YOLO guy. YOLO guy
• 1:02:58
  is fun to be around, but nobody wants him on
• 1:03:01
  their staff. But you don't buy drugs from him, okay?
• 1:03:04
  No, no, no. No, you don't buy your drugs from
• 1:03:08
  him, and you also don't want him running your podcast
• 1:03:10
  index. No. So, you know. But just look at you,
• 1:03:15
  man. You're running Omachi? You're, uh, you're YOLOing. You're using
• 1:03:20
  agentic coding. I have corrupted you. I am so proud
• 1:03:23
  of myself. I mean. If I keep going in this
• 1:03:27
  direction, I'm gonna be used- I'm gonna be- before- Good
• 1:03:30
  job, Adam. Ha ha ha, you've brought him to the
• 1:03:33
  dark side. At this rate, I'll be writing Ruby by
• 1:03:37
  the end of the year, which nobody wants. Nobody wants
• 1:03:42
  this. So we're... We're getting there. It's just a slow
• 1:03:49
  process by design because... Now, I may not be writing
• 1:03:55
  this code, but... You better believe I'm looking at every
• 1:03:59
  line of it because I don't. Yeah. I don't trust
• 1:04:03
  any of it. And it has already caught. Critical mistakes.
• 1:04:08
  Like more than one. There's been many times where I've
• 1:04:13
  said, where something just gave me pause and I said,
• 1:04:18
  "Okay, I need you to check this." I want a
• 1:04:23
  test suite specifically for this and this and this. It
• 1:04:29
  ran off and ran the test code for 20 minutes
• 1:04:34
  or so and it came back and it was like,
• 1:04:35
  critical bug. Oh yeah, yeah. I've seen this. And this
• 1:04:40
  is why I think... experience of system design and even
• 1:04:47
  if you're not a good coder as long as you
• 1:04:49
  understand system architecture and the way things are supposed to
• 1:04:52
  operate. you can properly handle these things. Because something gives
• 1:05:00
  you a spidey sense that, eh, I don't think this
• 1:05:02
  feels, something about this doesn't feel correct. And that's something
• 1:05:06
  that a person without the experience would not feel. No,
• 1:05:10
  you need it because these things go off the rails
• 1:05:13
  all the time. They're like, no, but I don't think
• 1:05:15
  you need to do it. Oh, you're right. Oh, it
• 1:05:17
  was in my memory file. Yeah, let me just check
• 1:05:20
  that. Yeah, okay. Yes, my bad. Yeah. Yeah. You have
• 1:05:26
  to understand what you're doing. I won't. No, I'm not
• 1:05:29
  writing it. It's all sustained in PHP. is staying in
• 1:05:32
  PHP for, and one reason to staying in PHP is
• 1:05:36
  because If there's one language that I am actually good
• 1:05:41
  at, it's PHP. Now, let's just take the jump here.
• 1:05:46
  So this will... somehow integrate with our our PodPain Cloud.
• 1:05:56
  our cloud of bathroom running nodes. Get the bat ears.
• 1:06:03
  Right? Y-yes. And that will be – there's a whole
• 1:06:11
  long list of things to come after this, but this
• 1:06:14
  is the critical one. Then once we get this thing
• 1:06:17
  out of the way. Then the whole thing is going
• 1:06:19
  to be decentralized, hash space, and we can retire. I
• 1:06:24
  am so looking forward to this day. Um... Yeah, and
• 1:06:31
  then we'll start figuring out how to ship it, how
• 1:06:34
  to ship the database. Oh, this is going to be
• 1:06:36
  so cool. I mean, I can't believe I'm excited about
• 1:06:39
  this, but it's going to be cool. Ship the database.
• 1:06:44
  Love that. What I want is for everybody to have
• 1:06:48
  some kind of. dashboard or something running at all the
• 1:06:52
  time. Everybody who's got an interest in this project, I
• 1:06:54
  want them to all be able to just have some
• 1:06:56
  sort of interface where they're seeing what is happening in
• 1:07:00
  the internals of the index at all times. Anybody, any
• 1:07:06
  trusted admin of the index can get in and... process
• 1:07:12
  of merge queue candidate. Isn't it amazing that this is
• 1:07:16
  the stuff that you're working on, that we're discussing, that
• 1:07:19
  the boardroom is doing, that this project is doing? Value
• 1:07:22
  for value, by the way. And meanwhile, in New York,
• 1:07:26
  everyone's arriving for podcast movement. They're going to talk for
• 1:07:30
  three days about what is a play. What is a
• 1:07:36
  play? I just find that amazing. All the people making
• 1:07:41
  money over there talking about what's a play? And we
• 1:07:44
  are building the future of the – I was talking
• 1:07:47
  with Tina today. She says, um... What is a play?
• 1:07:52
  She said, what is a play? Because, you know, there's
• 1:07:56
  transitions in my life and things are changing and things
• 1:07:59
  are happening. You know, my co-host died, so things are
• 1:08:02
  changing. And we're just talking about stuff. She says, why
• 1:08:05
  do you do Podcasting 2.0? valid question valid question you
• 1:08:11
  may have received this question yourself and i said And
• 1:08:15
  I didn't even have to think too long about it.
• 1:08:17
  It says, because it is a critical part of who
• 1:08:21
  I am and what I do. Because without this, my
• 1:08:27
  entire existence could be in jeopardy. Imagine I have to
• 1:08:31
  go suck up to Spotify and Apple and Google and
• 1:08:34
  everybody to get my stuff out. and It's also... It's
• 1:08:43
  even the little things, you know, it's like the funding
• 1:08:45
  tag, it's the chapters, it's the transcripts, all the things
• 1:08:48
  that we've worked on. And then I followed up by.
• 1:08:51
  But you think that's bad? You should look at what
• 1:08:53
  Dave's doing for this thing. You make no money from
• 1:08:59
  it. I said, no, but that's not the point. It
• 1:09:00
  is fundamental, and I would say for both of us,
• 1:09:03
  to who we are and what we do. Yes. And
• 1:09:09
  my wife asks me this same question, and I ask
• 1:09:11
  myself this same question all the time. Does it usually
• 1:09:13
  come when the Amex bill arrives? That's when I usually
• 1:09:16
  get this question. It usually comes when the house is
• 1:09:21
  super dirty and I'm like... I'm like, I got to
• 1:09:26
  just give me one minute. I've got to finish this
• 1:09:29
  thing. Oh, I'll be right there. Oh, no, no, no,
• 1:09:32
  man. I've learned this. This is not good. You just
• 1:09:34
  got to freeze that thing. Freeze. Just walk away. Yeah,
• 1:09:40
  no, I... I agree. It's become... It's part of your
• 1:09:47
  DNA. It's just part of my DNA. It is. Yes,
• 1:09:51
  Charlie. It is. I feel it in my DNA. Now
• 1:09:55
  I need to find that clip. And also a lot
• 1:09:59
  of
• 1:10:00
  things in the podcast world now depend on the index.
• 1:10:05
  It's not just a hobby. Here we go. Tell me
• 1:10:09
  about this sexuality. It's in your DNA. Yeah. Well, can
• 1:10:14
  we talk about the latest integration of the index into
• 1:10:18
  a product that has a couple of people who use
• 1:10:21
  it? I don't know where you're going. 120 million users.
• 1:10:32
  of an app? got me. Starts with a V. Wow.
• 1:10:48
  I'm having a brain fart, I guess. I feel like
• 1:10:50
  I should know this. Okay, one word. Oh, yeah. Can
• 1:10:54
  we talk about that? I don't know if we can
• 1:10:56
  talk about it. Why not? Is it super secret? I
• 1:10:58
  don't think it's super secret. No one said don't talk
• 1:11:00
  about it. I mean look at the email. Did they
• 1:11:07
  say don't talk about it? It's under embargo. We've signed
• 1:11:13
  an NDA about our DNA. Yeah, I mean, they don't
• 1:11:21
  talk about it. It's up to you. It's your discretion.
• 1:11:25
  At my discretion, thank you. That's what we say when
• 1:11:28
  landing. Do you think I can land this plane? At
• 1:11:29
  your discretion. We hope. Yeah. No, we got a nice
• 1:11:34
  email, a very nice email, real developer dudes, because they're
• 1:11:38
  like, hey, man, we'd really like to build in the
• 1:11:41
  index into our product. uh but we don't want to
• 1:11:45
  swamp you what do we do and so we came
• 1:11:47
  up with some caching discussions etc And I think it's
• 1:11:51
  just phenomenal that VLC is looking to integrate the index
• 1:11:57
  into their product. That's just amazing. This is why... This
• 1:12:07
  is why it's not just some sort of hobby or
• 1:12:10
  like something that I can ignore. Yeah, this is important.
• 1:12:13
  This is like really, this is amazing stuff. Yeah, it
• 1:12:17
  has to, I mean, things, the more things start to
• 1:12:20
  depend on. the index, the more, I mean, we can't,
• 1:12:26
  we don't have Apple money or Spotify money or something
• 1:12:29
  like that. So we will never be, we'll never have.
• 1:12:32
  you know, 27 nines up time or whatever that is.
• 1:12:36
  But we can have a reliable, stable product that people
• 1:12:40
  can depend on. Yeah. That is possible. That's very possible.
• 1:12:43
  And I was thinking, you know, will they put somewhere
• 1:12:46
  on in the site or somewhere on their app like,
• 1:12:49
  hey, you know, this podcast is from the podcast index
• 1:12:52
  value for value. Consider supporting them for this wonderful service.
• 1:12:56
  I hope so. They're free, aren't they? I don't think
• 1:12:58
  there's a... Do they have a, do they have a,
• 1:13:02
  are they, let me see. I think it's donation supported.
• 1:13:05
  It should be. Let me see. free and open source
• 1:13:09
  media, play on the corner. Help and join us. Join
• 1:13:15
  us. and they give me a little website how to
• 1:13:19
  contribute. Oh, they're a nonprofit organization. Oh, they're hardcore. Oh
• 1:13:26
  good, see- I love stuff like this. I love –
• 1:13:31
  I love when... Groups like this, open source groups and
• 1:13:36
  nonprofits, that kind of thing, can use the index because
• 1:13:38
  I know it really makes a big impact on their
• 1:13:40
  bottom. Yeah. You know, and that's a big deal for
• 1:13:44
  me because we've had, me and you've talked, we've had
• 1:13:47
  Zoom meetings with like Mozilla and stuff like that in
• 1:13:49
  the past. Yeah. That's all, I mean, I like. I
• 1:13:53
  like helping these kind of guys out. They're clearly not
• 1:13:56
  in it for the money. Well, I don't know. They
• 1:13:59
  may be doing very well. I mean... This is an
• 1:14:02
  interesting group of people. President John Baptiste Kempf, Vice President
• 1:14:06
  Konstantin Pavlov. These are foreigners, man. Vice President of Events,
• 1:14:13
  Vibhuthi. Treasurer Thomas Gullim. Secretary Felix Paul Kuhne. I gotta
• 1:14:21
  check it out, let me see. Let's see how much
• 1:14:25
  money they have. video land. I bet it's not a
• 1:14:29
  ton. You don't know. You don't know. 120 million people
• 1:14:35
  using it. That's... How many of those? Actually donate. Well,
• 1:14:42
  1% would be quite good. Robot, find the form 990
• 1:15:00
  for the... it. video land let's say I love doing
• 1:15:06
  this stuff. Okay, choice A, oh, form 990 does not
• 1:15:12
  exist. Oh, they're not a US registered. Okay, well, that
• 1:15:16
  makes sense. Oh, they are French. They are French. Oh,
• 1:15:21
  Benjamin Bellamy can walk over to their office. Show me
• 1:15:25
  your papers. He can tap on the door. Yeah. Yeah,
• 1:15:31
  there must be not a theory. Isn't the European parlance
• 1:15:37
  for that a public interest corporation? Well, they say on
• 1:15:40
  their website. nonprofit organization based in France. There you go,
• 1:15:45
  based in France. Okay. But I don't know. The French
• 1:15:51
  typically... Yeah, they get government money. I was going to
• 1:15:55
  say that, yes, but to me, from my experience in
• 1:16:00
  this project, the French typically either love us or hate
• 1:16:04
  us. And so Benjamin loves us. Yeah. But what was
• 1:16:08
  that group that we tried to deal with with Bitcoin
• 1:16:10
  stuff? And they were like that. Which one? Oh, no,
• 1:16:18
  that was the French guys. That was the other French
• 1:16:20
  guys. What is the the lightning app? That's what it
• 1:16:24
  was. The lightning app. French. Come on, boardroom. The French
• 1:16:30
  Lightning App Wasn't it A-C-I-N-Q? Wasn't that the name of
• 1:16:34
  the... Something like that, yeah. Oh, that guy was arrogant.
• 1:16:38
  I know, oxymoron. Async Lightning Node. It wasn't async. It
• 1:16:45
  was something else. Yeah. Yeah. Whatever it was. Yeah. But
• 1:16:51
  yeah. So you can do whatever you want in America.
• 1:16:53
  We do. We do. That's that's how I remember the
• 1:16:58
  conversation. Pretty much. Yeah. What was the name of that?
• 1:17:02
  Now you've got me. guessing. Uh... French. Bitcoin lightning. Yeah,
• 1:17:20
  A-C-I-N-Q, async. Yes, you're right. It was a Phoenix wallet.
• 1:17:25
  Was that the name? No, I think Async. Oh, yeah,
• 1:17:28
  Phoenix wallet. You're right. Yeah, Phoenix wallet. Yes. Okay. Yeah,
• 1:17:31
  you're right. That was way back at the beginning of
• 1:17:34
  all the V for V stuff. Yeah. Same time Jack
• 1:17:38
  Maulers went, yeah, okay, no. Right, yeah. Yeah, so that'll
• 1:17:46
  be great. I hope – I want to work with
• 1:17:48
  those guys and give them what they need so that
• 1:17:50
  we can – and I really think – I'm appreciative.
• 1:17:54
  Anytime somebody reaches out to us and says, hey, we're
• 1:17:58
  about to bring a whole bunch of potential traffic to
• 1:18:00
  you. We don't want to crush you. How do we
• 1:18:02
  work with you? I've been thinking about 120 million users.
• 1:18:06
  And I know a lot of people, because I know
• 1:18:08
  because whenever I screw up something with the ID3 tags
• 1:18:11
  or don't put the right image into the MP3 file,
• 1:18:15
  the VLC users come out. Hey, man, I listen to
• 1:18:19
  this on VLC. So who knows what kind? They may
• 1:18:23
  be putting a full-blown podcast client in it. that would
• 1:18:28
  make in the world of podcasting. Does Omarchi have a
• 1:18:32
  VLC in it? I don't know. Let's try VLC. Yeah.
• 1:18:37
  I don't know. Or is it called VideoLan? I just
• 1:18:39
  do a super space and then type VLC. Don't you
• 1:18:43
  know how to launch apps, Dave? Super space VLC. And
• 1:18:47
  type in VLC. There is no VLC. Oh, I must
• 1:18:50
  have installed it myself. You have it on yours? Yeah,
• 1:18:53
  I must have installed it myself. I use it all
• 1:18:55
  the time. Well, how do I, I guess I need
• 1:18:59
  to do here. Here's what I'm going to do. I'm
• 1:19:00
  going to do command. I'm going to super shift. No,
• 1:19:03
  super enter. I get a terminal. Yeah. Pac-Man. Pac-Man S.
• 1:19:09
  Dash S. Uppercase S. VLC. pseudo make me a sandwich.
• 1:19:20
  pseudo Oh, goodness gracious. I got a lot of things
• 1:19:24
  here. Yes. Okay, alright. Boom, installed. It's done. There's no
• 1:19:31
  step three. I just want to see what their current
• 1:19:34
  podcast situation is. Oh, I don't think they have any
• 1:19:37
  podcast situation going on. Playback, audio, video, tools. I don't
• 1:19:44
  see any of the program guide. What is that? That's
• 1:19:47
  nothing. What, MVLC? No, there's nothing in there yet. Yeah,
• 1:19:50
  there's nothing in there. No, they're building it. Next release.
• 1:19:54
  Next release. Yeah, so anyway, that's a thing. Let's thank
• 1:19:59
  some people.
• 1:20:00
  now that we have given you the longest intro to
• 1:20:02
  Value for Value. Because if you're listening to a podcast,
• 1:20:05
  if you're listening to this podcast on VLC, consider supporting
• 1:20:08
  the podcast index. It's part of our DNA. And we'd
• 1:20:13
  love you to support that. Not for our DNA, not
• 1:20:16
  for our pocketbook, so this thing keeps rolling forever. We
• 1:20:18
  are, in fact, volunteers. Right? Yeah. We're not employees. We
• 1:20:25
  don't make any money off of it. We love doing
• 1:20:28
  it because it's part of who we are. And you
• 1:20:31
  can support us through PayPal. Go to Podcast Index. And
• 1:20:39
  down at the bottom, there's a red donate button or
• 1:20:41
  in your modern podcast that maybe there's a funding button
• 1:20:44
  or sometimes it's a dollar sign. Sometimes it's a piggy
• 1:20:47
  bank. There's all kinds of different iconography. Click on that.
• 1:20:50
  That'll take you to the same place. You can also
• 1:20:52
  send us a boost, and the boosts have been restored
• 1:20:55
  on my machine here. So here's what came in from
• 1:20:57
  Drebscott1234. He says... serpentine belt serpentine serpentine serpentine Almost done.
• 1:21:07
  If I can get the belt intentioner to cooperate. Hashtag
• 1:21:10
  truck talk. Nice. Do it. You may have to make
• 1:21:14
  a custom tool. Those serpentine belt tensioners are hard to
• 1:21:16
  get to. 7777 from Martin Lindeskog. Adam, talking of AI
• 1:21:22
  robots and Beltway politics, are you looking forward to RoboCop
• 1:21:26
  to be announced on Amazon Prime Video? Harp music? Oh,
• 1:21:29
  I think I might have potted that down. Sorry. Your
• 1:21:32
  earlier boost came in. The Salty Crayons 777 came in.
• 1:21:37
  And, oh, by the way, Dave, keep your book tips
• 1:21:39
  coming. Yes, these are highly, highly respected and appreciated. Salty
• 1:21:43
  Crayon 777, listening to Godcasting on a modern podcast app
• 1:21:47
  is like JSONs for Jesus. In the spirit, he says.
• 1:21:52
  Thank you. Variation of in the pipe. 1, 2, 3,
• 1:21:55
  5 from Drebscott. My truck just tried to eat me.
• 1:21:59
  shocks holding my hood must slowly be giving out. Hashtag
• 1:22:03
  truck talk for Dave. Pro tip. Get a broom handle.
• 1:22:14
  Yeah, and cut it off and then stick it in
• 1:22:17
  the side of the compartment or either under your seat
• 1:22:23
  or in the trunk or something. That way when you
• 1:22:25
  open the hood, prop that thing up. Oh, I see.
• 1:22:29
  Oh, okay, the hood. Oh, yeah. Yeah, it wanted to
• 1:22:31
  slam on his head. Oh, that's no good. Dreb Scott
• 1:22:34
  again, 1234, use the combination of Grok and YouTube to
• 1:22:37
  work on swapping out my alternator. Hopefully I don't die.
• 1:22:41
  Hashtag truck talk for Dave. This is all truck talk.
• 1:22:45
  This is wonderful. Yeah, it sounds like we have a
• 1:22:47
  new podcast. Now you got your roadcaster hooked up to
• 1:22:50
  Omachi? By the way, Kevin Finn texted me the other
• 1:22:54
  day and said, I'm dying to hear where you were
• 1:22:57
  going with your timing. with your ignition timing metaphor in
• 1:23:03
  the Model T Ford. Maybe we should just go for
• 1:23:06
  truck talk. I mean, we should just, and that would
• 1:23:08
  be a value for value. No, we'll do ads. We'll
• 1:23:10
  do ads. We'll do ads. Post show. You got to
• 1:23:14
  subscribe to Podcast 2.0 Plus to get post show. You
• 1:23:17
  got to get the premium package and we'll do video.
• 1:23:19
  Oh, yeah, totally. Got to do video on YouTube. Riverside.
• 1:23:26
  Riverside, which turns out, is made by Jews. Oh, no!
• 1:23:29
  Oh, yeah, bronze. Oh, no, I can't believe it. People
• 1:23:32
  should immediately shut down your account. Bronze, it's clearly Jew.
• 1:23:40
  Martin Lindskoog, there he is, 1776. Here's another Liberty Boostergram.
• 1:23:44
  Excuse me. 1776 Satoshi's Freedom of Expression. BoostMeBuddy.com user. No
• 1:23:53
  other name there. 333. Thank you. Another Martin Lindeskog. Liberty
• 1:23:57
  Boost. Oh, he has a little Statue of Liberty icon.
• 1:24:01
  How nice. Boost for the citizens of New York City.
• 1:24:03
  Never forget. Best premises. Martin Lindeskog. Lyceum. Drebscott1234. Replacing the
• 1:24:09
  alternator on my 2008 F-150 while listening live. Hashtag Truck
• 1:24:13
  Talk for Dave. There we go. And he also sent
• 1:24:17
  a test boost. 1234. Wow, there's a 500 sat boost
• 1:24:22
  for Curry and the Keeper. What is that? How did
• 1:24:24
  that happen? That's random. That's all I got. There may
• 1:24:29
  be more, but my node was down again for a
• 1:24:33
  day. Hello! Yeah. I got some PayPals. Okay. Steven Santoni?
• 1:24:42
  $25. He said, chipping in? No, thank you. Chipping in.
• 1:24:46
  Ow. There we go. You okay? Did you trip over
• 1:24:50
  the piano? Yeah, I did. Ralph a step 50 bucks.
• 1:24:55
  Whoa, nice. Thank you, Ralph. We have the boosts. Yeah,
• 1:25:00
  there we go. Martin Lindeskog, $1. Thank you, Martin. Brendan
• 1:25:05
  Mulligan, that's a... Boys and Girls of Podpage. Yes. $25.
• 1:25:10
  Joseph Maraca, $5. Nicholas Fett. one dollar Mitchell Downey, $50.
• 1:25:16
  Thank you, Mitch. Thank you. Boost, boost, boost. And another
• 1:25:20
  one from Mitch, $10. That first one was the Podverse.
• 1:25:23
  Yes, it was Mitch's personal. Boost! Lauren Ball, $24.20. Christopher
• 1:25:30
  Harabaric, $10. Damon Kasajak, $15. Thank you. Terry Keller, $5.
• 1:25:37
  And the Salty Sessions Show, $10. No message. Salty Sessions.
• 1:25:41
  Yeah, Salty Sessions. I think that's now also on Sundays,
• 1:25:45
  I think. Salty Sessions before no agenda. Early in the
• 1:25:48
  morning, I believe. I'm not sure. That's on the stream.
• 1:25:51
  Yeah, on the stream. Yeah, things are moving around. It's
• 1:25:53
  beautiful. See, we've got... What? What have you got? Something?
• 1:25:57
  No. No, no, no. No. No. We got 58 sets
• 1:26:00
  from BoostMeBitch.com. I get boosts whenever there's a boost nearby.
• 1:26:06
  you Um, I keep forgetting to do that. You know,
• 1:26:13
  I listened to This Week in Bitcoin. I was like,
• 1:26:16
  I got to use those jingles more. Bullysteed. And Fountain
• 1:26:22
  sent us 929 sassies as it came for Podcasting 2.0
• 1:26:25
  and stayed for the Dave Sage cast and book recommendations.
• 1:26:29
  And truck talk. And truck talk. Dreb through Castamatic sent
• 1:26:36
  a 716 SAT test boost. Ooh, nice. Thank you. We
• 1:26:40
  got you, Drib. Let's see, I think you read these.
• 1:26:45
  We got some anonymous. Yeah. Yeah, I think we got
• 1:26:49
  all these. And we got Comic Store Blogger. Yes. The
• 1:26:52
  Delimiter. 11,111 sets. Do you, have you, by the way,
• 1:26:57
  have you seen how complete? and thorough my new... extraction
• 1:27:04
  of all the boosts and paypal's is this is smooth
• 1:27:07
  as silk i've noticed it yes it is it feels
• 1:27:09
  very smooth what are you using what did you vibe
• 1:27:12
  code It's something that I don't even understand it, but
• 1:27:16
  it works, but it works. It goes into the. It
• 1:27:20
  reaches into the heart of the LB hub. Oh, wow.
• 1:27:25
  And rips out the boost and puts them in an
• 1:27:27
  HTML file. That's awesome. Yeah. The world is changing, brother.
• 1:27:35
  I will say that I've never seen an industry change
• 1:27:38
  so fast. as coding. Yeah, it's crazy. It went from...
• 1:27:46
  All hand code to all AI code. Remember, I went
• 1:27:49
  like three months. I went from copy pasting in Gemini.
• 1:27:54
  Remember I was doing that? And Grog. I started with
• 1:27:58
  Grok, then Gemini. And I would literally, I'd say, this
• 1:28:00
  doesn't work. And then do another version. I do select
• 1:28:03
  all, copy, then go to my editor, paste. And then
• 1:28:07
  you said, hey, Matt, you should try a CLI thing.
• 1:28:09
  I'm like, oh, okay. Yeah, you were grumpy about that.
• 1:28:14
  But once I figured it out, I never looked back.
• 1:28:16
  And now I'm doing the same to other people. I'm
• 1:28:18
  saying, hey. You should try Grokbot. Have you seen that?
• 1:28:21
  Grokbot? Dude. It is changing people's lives. I think it
• 1:28:27
  was part of the acquisition by, what did XAI buy?
• 1:28:35
  The... thing they bought. Cursor? Cursor, yes. So you can
• 1:28:45
  run it on your computer. You run it on an
• 1:28:48
  app. same thing and it gives you a cloud computer
• 1:28:53
  which you can actually see. There's like a little, do
• 1:28:55
  you see the little cloud computer, a little desktop? and
• 1:28:58
  you can fire up little bots and the whole idea
• 1:29:02
  is that you tell each bot to do something specific
• 1:29:06
  like scan for this or look for that or calculate
• 1:29:10
  this and then you have a master bot, master botter
• 1:29:16
  and the master botter checks everything else and is coordinating
• 1:29:20
  for you. It's like agents for dummies. Nothing wrong, nothing
• 1:29:27
  bad will happen from this. Well, so Dana Brunetti, Hollywood
• 1:29:32
  producer. Celebrate Hollywood producer from such famous movies such as
• 1:29:36
  Captain Phillips house of cards Shades of Grey. Shades of
• 1:29:43
  Grey was a Christian band. The real money maker. and
• 1:29:47
  he's going to do the show with me on Sunday.
• 1:29:50
  He said, I'll do the show. Okay, come and do
• 1:29:52
  the show since you think you know everything, Hollywood man.
• 1:29:55
  And he's like, "Why should I do clips?" That's it.
• 1:29:59
  Okay, you can do.
• 1:30:00
  You can go get Audacity and you can record stuff
• 1:30:03
  or you can just use Grokbot and just give it
• 1:30:06
  a YouTube video link and say, give me the transcript
• 1:30:10
  and then say, start here, end here, give me an
• 1:30:12
  MP3. And within five minutes, he's like, oh my God,
• 1:30:16
  this is so easy. pretty cool bad things are gonna
• 1:30:22
  happen with this this is the day this is how
• 1:30:25
  the internet fails I'm not so sure. I don't think
• 1:30:30
  so. People aren't really doing coding. Well, that's the problem.
• 1:30:36
  People are doing stuff they don't understand. And on the
• 1:30:38
  backside of that grog bot, it's hammering 50 websites to
• 1:30:42
  death. It has guardrails. You don't even know. It has
• 1:30:43
  guardrails, man. What are you talking about? Whatever. Nothing that
• 1:30:49
  Elon Musk ever does has guardrails. All right. Let's get
• 1:30:53
  to Comic Street Blogger, the final booster gram. Yes. 11,111
• 1:30:57
  sats, he says. Howdy Dave and Adam. Thinking of you
• 1:31:01
  both on 9/11, I remember learning what happened as I
• 1:31:04
  was leaving the office. The work colleague had footage of
• 1:31:08
  the towers on his computer screen. Sending warm thoughts to
• 1:31:11
  you and your families. Yo, CSB. Worth $15.40 when sent.
• 1:31:18
  Wow, he didn't even promote a podcast. I feel like
• 1:31:21
  that's heartfelt from CSP. Yeah, oh, it is. Of course
• 1:31:24
  it is. He's a very sensitive guy. He also offered
• 1:31:28
  to run a... MCP server for the index. Yes, but
• 1:31:34
  it requires... Only under 27 different qualifiers and a binding
• 1:31:38
  contract. That's pretty funny. I'm like, okay, CSB. I promise
• 1:31:44
  I will run your MCP server for you. Where should
• 1:31:46
  I send the DocuSign? you I love that guy. I
• 1:31:52
  love that. Again, no community is complete without comics or
• 1:31:55
  blogger. If you want your community to be fully rounded,
• 1:31:59
  you need CSB in it. And you are fortunate if
• 1:32:02
  you have him in your community. I did his video
• 1:32:05
  podcast. You have not. I'm going to. I'm going to.
• 1:32:08
  I'm going to. I've been a little busy. And I'm
• 1:32:11
  also trying to time it. For maximum effect. Exactly. All
• 1:32:18
  right, brother. What are you doing this weekend? Well, we've
• 1:32:21
  got a, yeah, it's pretty busy. I had my uncle
• 1:32:24
  passed away this past week. Yeah, I'm sorry. Sorry to
• 1:32:26
  hear that. Yeah, and we've got a funeral and a
• 1:32:31
  birthday this weekend. So we get both sides of life.
• 1:32:33
  Oh, yeah, it's like your partner dying and having a
• 1:32:36
  grandbaby 24 hours apart. Pretty much, yeah. It seems to
• 1:32:39
  be the way of the world these days. The older
• 1:32:41
  you get, the more these things happen, I think. It's
• 1:32:44
  been the way of the world since day one. And,
• 1:32:47
  yeah, it's... Thank God that it's balanced in that way.
• 1:32:52
  Yes. So, so funeral and that's it. Uh, funeral birthday
• 1:32:59
  And one-year-old birthday and then got some climbing and some
• 1:33:05
  other things to do with the daughter. Some truck work?
• 1:33:06
  Any truck work? Any working on the rig, man? Any
• 1:33:08
  truck? You working on your truck? No. No, I'm doing
• 1:33:11
  crown molding Sunday. I would rather be working on the
• 1:33:14
  truck, believe me. All right, everybody. It's in our DNA
• 1:33:17
  and you're under NDA. That means we will be back
• 1:33:20
  next week with another episode of Podcasting 2.0 where we
• 1:33:24
  are creating the future and you're here to enjoy the
• 1:33:26
  ride. See you then, everybody. Bye-bye. You have been listening
• 1:33:33
  Podcasting 2.0. Podcasting 2.0. Podcasting 2.0. Podcasting 2.0.
• 1:33:45
  to Podcasting 2.0. Visit podcastindex.org for more information. Go podcasting!
• 1:33:55
  it on your own time.