Latest / WGKM / Podcasting 2.0 - Episode 272: "Is this Y'all?"

Transcript

Podcasting 2.0 - Episode 272: "Is this Y'all?" · WGKM

• 0:00
  podcasting 2.0 for september 25th 2026 episode 272 is this
• 0:06
  y'all hello everybody welcome once again to podcasting 2.0 this
• 0:13
  is your weekly board meeting The boardroom is filled and
• 0:16
  we're ready to talk about what is going on with
• 0:18
  podcasting, podcastindex.org, and all the things we're talking about on
• 0:22
  podcastindex.social. We are, in fact, the only board meeting that
• 0:25
  doesn't care about the midterm elections. I'm Adam Curry here
• 0:29
  in the heart of the Texas Hill Country and in
• 0:31
  Alabama. The man who went to war with the feed
• 0:33
  aggregators before breakfast and posted a decadence. database graph as
• 0:37
  a riddle. Say hello to my friend on the other
• 0:39
  end, the one, the only, the pod sage himself, Mr.
• 0:43
  Dave Jones! Man. Man. I tell you. Tell me something,
• 0:53
  Dave. Tell me something, Dave. If I can not die
• 0:56
  here. No, no, please. I can only handle one friend
• 1:00
  dying a year. I'll wait until it's three months. New
• 1:04
  rule, new rule, new rule. I got a few months
• 1:07
  to do it. Yesterday was nutty. Yeah, you don't say.
• 1:14
  There was a lot of me staring at the screen
• 1:16
  and typing things yesterday. Well, the first thing that happened
• 1:20
  is I'm like, hey, man. My podcast is no longer
• 1:25
  showing up in search. No, wait. It was Wednesday is
• 1:27
  the first thing that happened. I'm like, I hate doing
• 1:31
  this to you because I go into the dashboard and
• 1:33
  I try everything. I'm like... And I'm like, scan for
• 1:36
  we get to do this. Like, okay, let me hit
• 1:39
  the pod ping again. No, it's not happening. And then
• 1:42
  at a certain point, you know, after like an hour,
• 1:44
  I'm like, I got to tell Dave because it's probably
• 1:47
  not just me. I was like, Dave, it's not updating.
• 1:50
  And then while you were working on that, someone sent
• 1:54
  me an email and said, hey, you know, no agenda
• 1:57
  doesn't show up at the top of the search, which
• 2:01
  you would expect. And then it wasn't showing up at
• 2:04
  all in the search for a while there. All these
• 2:07
  adjacent shows were showing up. I'm like, okay, well, oh,
• 2:11
  man, I hate to do this to Dave, but there
• 2:14
  it is. What went wrong, my friend? What was happening?
• 2:18
  Where do you want to start? I mean... What do
• 2:20
  you want to know first? I'm dying to know why
• 2:23
  no agenda dropped out of the top in search, and
• 2:26
  then I'd love to know what happened with the aggregator.
• 2:29
  I mean, there's nothing but pain is what I love
• 2:31
  hearing about, so bring it on. Yes. So the search
• 2:35
  is easy. We'll do that one first. Of course, when
• 2:39
  you told when when you sent me that email saying
• 2:41
  that no agenda was not showing up in search anymore.
• 2:45
  Internally, my first thought was, well, it's not it's nothing
• 2:50
  I did. Curry probably screwed it up again. Yeah. Oh,
• 2:55
  this is clearly somebody else's problem. But. But that's not
• 2:59
  what comes out of your mouth because you learn over
• 3:03
  time that every time you think that, it ends up
• 3:06
  being your problem. Right. So I'm like, well… What's the
• 3:12
  what's the most likely explanation for this? And clearly it's
• 3:17
  the podcast identity layer schema change thing that I've been
• 3:22
  doing for the last three weeks. That has to be
• 3:25
  part of it. It had to be. Yeah. And the
• 3:29
  first thing I did was I checked no agenda. Can
• 3:32
  I ask you a question? Can I ask you a
• 3:34
  question? Yep. My thinking was this has to be something
• 3:39
  with a feed consolidation or merge or something you were
• 3:42
  doing. But I stopped myself from saying, could it be
• 3:46
  or it might be related to this as a developer
• 3:50
  troubleshooting these things. Are you happy that I didn't do
• 3:54
  that or would you have preferred me to do that?
• 3:57
  It would not have been it would not have made
• 3:59
  any difference because I well, I say for most people.
• 4:05
  Most of the time when you get reports like that
• 4:07
  as a developer, you sort of mentally chop that off
• 4:09
  the end anyway. It's like it may be dot, dot,
• 4:12
  dot, and you're not even reading. You're already on a
• 4:15
  different browser tab. I figured. All right, good. But when
• 4:20
  you say stuff, though, I tend to – Read through
• 4:26
  it and consider it because – Oh, I get through
• 4:28
  the gate. All right. The gate. I'm load-bearing. That's why.
• 4:33
  I'm load-bearing. Yeah. You're the load-bearing gate of the index.
• 4:39
  But you know how this stuff works on the back
• 4:43
  end so well that you're probably in the ballpark. Right.
• 4:47
  And that's so I pull up no agenda, which is
• 4:51
  podcast ID forty one five oh four. I pull that
• 4:55
  up in the index dashboard. And the first thing I
• 4:59
  see is our new little badge here that says podcast
• 5:02
  forty one five oh four currently serves feed five nine
• 5:06
  two six oh six four. Oh. And so... And it
• 5:13
  also gives another little piece of text that says, not
• 5:16
  41504, so this row is an abandoned or superseded feed.
• 5:21
  Oh, no. You've been superseded. So this was the first
• 5:26
  clue of, yeah, well, this is obviously related because –
• 5:32
  So podcast ID 41504 used to serve, used to be
• 5:36
  used to be identical to literally identical to feed 41504.
• 5:43
  But now that the 41504 has moved up in the
• 5:48
  conceptual hierarchy to a podcast. Yeah. And the feed that
• 5:54
  is serving can be any feed. Now it's not pointing
• 5:59
  to 41504 anymore. It's now pointing to feed 5926064, which
• 6:04
  and the reason is because of what I texted you
• 6:08
  trying to get clarity was that. Feed.nashownotes.com slash RSS XML
• 6:16
  is the feed we have in the index. Right. But
• 6:20
  it's actually being served from feeds.noagendaassets.com slash noagenda.xml. Yes. So
• 6:26
  there legitimately is two feeds of the same thing, and
• 6:33
  it picked – The later one, because this seems to
• 6:37
  be the more widely distributed one, and it's what's in
• 6:40
  Apple's directory. Got it. So it's in our attempt to
• 6:44
  try to stay consistent with Apple as much as possible.
• 6:47
  When they have a feed and we have a feed,
• 6:49
  excuse me, when they have a podcast and we have
• 6:52
  that podcast. Because we act as a fallback, we try
• 6:56
  to just stay with what they have unless what they
• 7:00
  have is absurd. So that that means that the search,
• 7:05
  the search still works on feeds. Not podcasts because I
• 7:12
  had not gotten to that point yet. So basically it
• 7:16
  was just – it was seeing 41.504 as a dead
• 7:20
  feed. So it was not surfacing in the search. So
• 7:24
  I just had to hastily – Finish that work that
• 7:28
  I'd already designed and get the Sphinx search indexer SQL
• 7:34
  remade to respect podcast identity table and it fixed it.
• 7:43
  So, I mean, that was a fairly straightforward issue. I
• 7:45
  sort of knew that. quickly what was going on. What
• 7:48
  determines the order of search ranking? Well, I don't know.
• 7:54
  Because we don't have first-party data or we have very
• 7:58
  little of it, Uh... We get a couple. Just whoever
• 8:04
  your friend is, basically. Pretty much Sam and Mitch. That's
• 8:09
  pretty much it. They're our only friends. But other than
• 8:13
  that, we have very little first-party data. So we have
• 8:17
  to just do what we always do. Heuristics based on
• 8:23
  API traffic. Yeah, yeah, yeah. And so it's a, you
• 8:27
  know, it's just a number that gets cooked up through
• 8:31
  what we think is going on. But it could be
• 8:34
  completely wrong, but it's it generally tends to get it
• 8:38
  mostly right. Yeah. It does. I mean, it's decent. Although
• 8:42
  Podcasting 2.0 has also been dropping down the charts, I
• 8:48
  noticed. Hold on. So if I go to – I
• 8:51
  like going to pod.link. Okay. So podcasting 2.0. And it's
• 8:59
  an Apple podcast. It's number two with number one being
• 9:04
  podcasting explained 2.0 the podcast. Okay. And on podcast index.
• 9:11
  We are not in there. We are like, I think
• 9:16
  we're like 10 down or something like that. 1, 2,
• 9:19
  3, 4, 5, 6, 7, 8, 9, 10. Now we're
• 9:22
  not in the top 10. I'm seeing it on my
• 9:24
  dashboard search. Well, I'm just doing, so in Podlink, I
• 9:28
  couldn't even link to it. pod.link because at least not
• 9:33
  in maybe nathan bandus oh that would make nothing but
• 9:37
  sense yes because we called him because we called him
• 9:40
  undefined guy for so long he was like screw these
• 9:43
  guys yeah that's interesting i don't know why that happens
• 9:48
  when i saw podcasts index.org. Let me see. Podcasting 2.0.
• 9:57
  Well, shouldn't we be at the top?
• 10:02
  Oh, man. We're like 11 down. Yeah. That's what? 1,
• 10:05
  2, 3, 5, 6, 7, 8, 9, 10. Yeah, we're
• 10:10
  number 11. We're number 11. It goes to 11. That's
• 10:13
  why we're not showing up in pod.link because they only
• 10:16
  go to 10. We are below – in our own
• 10:18
  search rankings, we are below – Gay Talk 2.0. Yes.
• 10:23
  And Esto... Alpha Male 2.0. Esto es Podcast 2.0. The
• 10:29
  Rural Church Podcast 2.0. And the Daddy Mines Podcast. Daddy
• 10:32
  Mines. But yeah, I was just looking at Daddy Mines.
• 10:35
  Gay Talk 2.0. Let's see what happens. And Earthlings 2.0
• 10:38
  Podcast. Let's see what happens if someone goes there. What
• 10:41
  if it's... It's just a gay version of you and
• 10:43
  me talking to each other. You smell that? What? Smells
• 10:47
  like gray sweatpants and bearded daddies. Yeah, baby. That's a
• 10:51
  podcast right there. Sweatpants and gray bearded daddies. Show title.
• 11:02
  There we go. Sweatpants. Smells like daddy talk. Yeah. Yeah,
• 11:10
  so that was the easy one to figure out. Search
• 11:14
  ranking is always a constant weird thing. I demand you
• 11:20
  put us at the top. I'm sorry. I demand you
• 11:23
  cheat. Put us at the top. Come on, man. Come
• 11:26
  on. The aggregator issue of why your podcast wasn't getting
• 11:33
  picked up by Podping, that was a complete rabbit hole.
• 11:36
  That was where the fight was happening. Yeah, talk to
• 11:39
  me. So… When I got in there to look and
• 11:44
  see why we get to do this was not being
• 11:50
  told. It looked like it. I look in the database.
• 11:53
  There's one hundred and thirty five thousand backlog. Of shows.
• 11:57
  Whoops. And they're all in this. They're all in the
• 12:01
  seven thousand. Excuse me. The seven million ID range. I'm
• 12:04
  like, oh, ag seven again. Ag seven. That bastard. It's
• 12:10
  this thing. It's. Gives me like aggregator anxiety like it
• 12:15
  is always every time there's an aggregator problem. It's always
• 12:20
  seven. I don't know why, but... So I get in,
• 12:26
  and of course it's crashing. So it's been crashing for
• 12:29
  who knows how long. I don't know, maybe a day,
• 12:32
  24 hours at least. Weeks and weeks. No, no, no.
• 12:36
  Luckily not weeks. The oldest one non-polled I think was
• 12:40
  from the 21st. It was like 24 hours or right
• 12:44
  around the right time. And I'm like, okay, what am
• 12:48
  I doing? I've always – again, I've given up on
• 12:52
  index coding by hand. I'm now just full agent guy
• 12:56
  when it comes to the podcast index. Heck yeah. I'm
• 12:59
  so deep into this thing now. Mr. Agent Guy. Yes.
• 13:04
  So – I'm like, all right, here we go. I'm
• 13:07
  just going to start throwing the coding agent at it.
• 13:10
  And first thing that pops up is – because I
• 13:15
  knew what the problem was. It was running out of
• 13:20
  – it was overflowing – Node.js was overflowing the heap,
• 13:23
  and so it was just hitting a feed file. Loading
• 13:27
  the feed file into the parser was just overflowing the
• 13:30
  amount of memory that was given. And it crashed it.
• 13:37
  But I'm like, I've put – I know I've put
• 13:38
  in a 35 meg limit. I know that. And so
• 13:41
  I go start going through, well, syntax problem or just
• 13:45
  a logic problem in there. It just wasn't hitting. It
• 13:50
  was not picking up the limit. So I'm like, okay,
• 13:53
  this is an easy fix. The coding agent can do
• 13:55
  this. So it goes, it refactors a couple of lines
• 13:58
  of code, 35 limit kicks in, back in business. But
• 14:03
  then, so I'm like, okay, this thing parsed now. Well,
• 14:07
  when it gets to, these run in detached screen sessions.
• 14:10
  Mm-hmm. So I get in the screen session. And I'm
• 14:15
  looking at it. I'm watching it. I'm watching it roll.
• 14:19
  And, yeah, Eric, we are going to move to that,
• 14:25
  but just so many other things that I haven't got
• 14:27
  there yet. But the – so I'm watching it roll,
• 14:30
  and then – I see it when it goes to
• 14:34
  do its next cycle. It just hangs for life. Five
• 14:41
  minutes. Then I see it blow through and start doing
• 14:45
  aggregation again. And then it goes to its next cycle.
• 14:47
  Hang. For like five minutes. I'm like, okay, what is
• 14:51
  going to start looking at the other aggregators? They're all
• 14:53
  doing the same thing. So there's all the, there's this
• 14:57
  like anywhere from three, To eight minute hang that's happening
• 15:01
  at the beginning of each feed polling. This is hang
• 15:04
  time in this case is not good. This no is
• 15:06
  the opposite of what you want. Right. And. So I'm
• 15:12
  like, okay, well, here we go. Coding agent. What's happening
• 15:16
  here? Is that literally the prompt you gave it? No,
• 15:20
  I was a little more eloquent than that. So then
• 15:25
  it's a it's it starts looking at the code and
• 15:28
  it says, well, here's your problem. You are... You're trying
• 15:36
  to use a SQL query. to pull feeds to poll,
• 15:46
  but some of the columns you're using are not indexed.
• 15:51
  I can't be right. I mean, I'm I've been doing
• 15:55
  this a long time. I know how to properly construct
• 15:59
  an indexed SQL statement, or at least I thought I
• 16:03
  did. And but here's the issue. Sometimes SQL can get.
• 16:15
  At a certain size, a SQL statement becomes almost incomprehensible.
• 16:20
  You really need – An agent. At least a syntax
• 16:25
  parser to help you make your way through this. And
• 16:31
  I don't typically do that. I'm typically writing these things
• 16:34
  just – text by hand. And I don't like it's
• 16:38
  almost like a regex. But, you know, you need something
• 16:42
  to visually like give you colors and stuff because this
• 16:45
  stuff can just get away from you. If you have
• 16:47
  like a. Yeah. Your colors is good. Yeah. If you
• 16:50
  have like a 40 or 50 line long SQL statement
• 16:54
  with multiple subselects, this kind of thing, it really gets
• 16:58
  difficult to understand what's happening. And so what. here's to
• 17:05
  make it concise or try to The way that each
• 17:12
  aggregator pulls in the feeds that it's going to scan
• 17:17
  is, is there's nine aggregators, zero through eight. And each
• 17:24
  one of them takes a slice of the index. But
• 17:27
  that's not the only criteria that it needs. It also
• 17:31
  bases it on priority. So there's a priority number each
• 17:35
  feed gets. and update frequency. So there's multiple criteria. And
• 17:42
  the way I have that set up is, first of
• 17:44
  all, each aggregator looks for its own slice of the
• 17:47
  index. So aggregator zero gets podcast IDs one through a
• 17:53
  million. Aggregator one gets one million and one through two
• 17:57
  million and so on. Each one starts with. give me
• 18:04
  all the feeds where dead equals zero. and the ID
• 18:10
  is within my slice range, And it's pull now flag
• 18:15
  is set to something greater than zero. So that means
• 18:19
  I'm supposed to pull it. I'm supposed to pull this
• 18:22
  feed. But then there's a second set of criteria, and
• 18:26
  those are all indexed columns. Then there's a second set
• 18:31
  of columns that are also all indexed. And so, for
• 18:35
  instance, on aggregator zero, it's saying, I want that. Or
• 18:40
  so either that criteria or if nothing is in that
• 18:45
  criteria set, I want to I want to also scan.
• 18:51
  for feeds where the update frequency is nine. And the
• 18:56
  priority is greater than negative one, which means like don't
• 18:59
  ever scan. Negative one is don't ever scan. And the
• 19:02
  pull now is greater than negative one. is is equal
• 19:06
  to zero so what i'm trying to say there is
• 19:08
  i'm trying to say this aggregator needs to always look
• 19:12
  at at flagged for for pulling feeds within this slice.
• 19:18
  And also another set of work. And so and what
• 19:21
  I'm trying to do there is maximize all the aggregators
• 19:25
  so that even if there's nothing to pull at any
• 19:28
  given time or if the number of feeds in their
• 19:30
  slice is very small, They still get a bunch of
• 19:33
  extra feeds that also need work to do. Ladies and
• 19:37
  gentlemen, this is a value for value project. I hope
• 19:39
  you're listening very closely and understand what work goes into
• 19:42
  this. Please consider supporting the index podcast index dot org
• 19:47
  down at the bottom of red donate button. This is
• 19:50
  a lot of work that's going on here. And we
• 19:53
  need to keep it running. Back to you, Dave. Thanks,
• 19:57
  Jim.
• 20:00
  Um, um, So, All of the columns in all of
• 20:06
  that, they're all indexed. The thing that I did not...
• 20:14
  Think of the thing that I missed is that ordering
• 20:17
  those results, you want to give some of those a
• 20:20
  priority because what if you come back with, let's say,
• 20:24
  what if you come back with 1,100 results, but your
• 20:28
  aggregator is limited to 250 feeds? You want to have
• 20:33
  some sort of priority in there where you say, OK,
• 20:36
  I got back a bunch of these, but these are
• 20:38
  the ones I want to do first. And the order
• 20:41
  by is the pull now column because that can be
• 20:44
  great. It can be zero means don't pull this feed
• 20:47
  currently. Everything gets a one by default, but if I
• 20:53
  want to jump something to the top of the stack
• 20:55
  and make sure it gets pulled right this second, I
• 20:59
  set it to two. So that's what I did with
• 21:02
  we get to do this. I set it to two.
• 21:07
  Well, when you were pod pinging it, it was setting
• 21:10
  it back to one. So that's why we kept like,
• 21:12
  that's why I told you not to pod ping it.
• 21:15
  I didn't. I didn't. No, no, no. But you did
• 21:17
  the first time. I saw in the second time. I'm
• 21:19
  sorry. Yeah. That's fine, but then there's another order buy,
• 21:27
  which is last check. And that's an ascending column that's
• 21:31
  just a Unix timestamp. So that's saying if if the
• 21:35
  pull now is is two, I want those first and
• 21:39
  everything else. I want to check it in the oldest,
• 21:43
  oldest to newest. So if we haven't checked it in
• 21:46
  a long time, well, that last check column. Did not
• 21:50
  have an index. Okay. Yeah, because the last check had
• 21:54
  been like two hours before I started publishing. And so
• 21:58
  in my mind as I was writing, I remember as
• 22:02
  I'm writing this sequel, in my mind I'm thinking, okay,
• 22:06
  by the time we get to the order by… We've
• 22:09
  already we're already down to our limited set. So we're
• 22:12
  just doing an ordering. We're not doing this is not
• 22:15
  a criteria for this for the select body itself. Well,
• 22:21
  that's not accurate because MySQL's plan engine where it determines
• 22:28
  how to execute the plan does not always do what
• 22:31
  you think it's going to do. And if you're ordering
• 22:36
  by a multi... congruent set of criteria It expects that
• 22:45
  thing to be indexed or else it's going to order
• 22:49
  it before the limit check and not after. So it
• 22:53
  was basically doing a lack of it. It was yeah,
• 22:56
  it was pulling all every time any aggregator pulled its
• 22:59
  slice, it was pulling all. All 8 million feeds and
• 23:02
  then grabbing the 100 off the top is what it
• 23:07
  was doing. Oh, wow. And that was the hang time,
• 23:12
  the 8 million ingest. Yeah, and it was hammering the
• 23:16
  database all day long. Okay, so is that the change
• 23:20
  that – was that the graph that you showed? That
• 23:22
  was the change that you made? Yes, and it went
• 23:26
  from – on the metrics, it cut database server load
• 23:31
  by 65%. Wow. Just by refactoring. I thought my Linode
• 23:36
  alerts were low today. I'm like, I don't have as
• 23:38
  many. You weren't getting Mindy alerts? No, I didn't. I
• 23:41
  got no Mindys. I got Mindy. Who names a server
• 23:46
  Mindy? I can tell you why it's named that, because
• 23:49
  originally we had two database servers, and one's name was
• 23:52
  Mork, and the other one was Mindy. But we killed
• 23:54
  Mork because it was too expensive. Mork was from Ork.
• 23:59
  Get rid of him. Yes. Yeah, Mindy is all that
• 24:01
  was left. Yeah. So there was two solutions to this
• 24:05
  problem. One is we… The last check is critical. So
• 24:11
  in order – we need an index on that column,
• 24:13
  so I had to do an alter on the table
• 24:15
  to add the index. Well, then in trying to do
• 24:20
  the alter on that table, it would not get a
• 24:25
  lock. So when you alter a table that's the size
• 24:28
  of the – that's got 8 million rows in it.
• 24:31
  It needs to sort of find a place to jump
• 24:34
  in, so to speak. Yeah. Because it's going to have
• 24:36
  to do a temporary table lock. And then it can
• 24:40
  do the online alteration. It doesn't have to stop the
• 24:43
  database. But then at the end of the process, it'll
• 24:46
  have to do another brief table lock. Well, if it
• 24:48
  can't ever get that very first table lock because there's
• 24:51
  something else going on, it'll just keep rejecting it. And
• 24:55
  I fought that for, gosh, an hour and a half
• 24:58
  and finally figured out. It was the Sphinx person server.
• 25:06
  That thing had been running a constant hit on the
• 25:11
  database for like 12 hours. So I had to kill
• 25:14
  that thing, refactor that thing's sequel to go in batches.
• 25:20
  Before I could finally come back and do the alter
• 25:22
  to add the index, Then after that happened, that only
• 25:26
  took like 10 minutes. Then after that, go in and
• 25:29
  refactor the sequel to because that sequel was still going
• 25:33
  to be slow when the when the limits drop below
• 25:37
  100 because it's still going to fall back to a
• 25:41
  big scan. So then we refactored that out from being
• 25:46
  like select where and or to being a union, which
• 25:51
  is basically two separate selects. Then do a join and
• 25:55
  take like a Venn diagram of of the contents that
• 25:58
  you're pulling out of the middle. So you both of
• 26:01
  those things run at full speed. So. Like basically the
• 26:04
  aggregators are pulling out the database at like 300 milliseconds
• 26:08
  now. They're like blazing fast. And so they got like
• 26:12
  a million times faster and dropped the load on the
• 26:16
  database by like over half. Wow. Well, that's a big
• 26:20
  win. Congratulations. That's a big one. Yeah, I mean, the
• 26:25
  end result is fantastic. It was just – but, I
• 26:28
  mean, there was more. I just – I kept going
• 26:31
  because I'm like, okay, well, we're in here. We might
• 26:35
  as well just keep going down the rabbit hole and
• 26:39
  ended up spending more time on the aggregators. Staggering their
• 26:46
  start times. So now none of them launch their polling
• 26:50
  at the same minute mark. So they randomize across the
• 26:57
  minute. There were some feeds that were stuck in limbo
• 27:03
  that were just getting – reprocessed over and over and
• 27:07
  over because there was a couple of, um, There's a
• 27:12
  couple of HTTP status codes I was not handling. And
• 27:16
  so they were just they had been getting pulled like
• 27:19
  once every 10 minutes for, I don't know, years. We're
• 27:23
  just hammering those things. And there's a couple of. Stop
• 27:29
  the hammering. Yeah. Yes. Somebody the the A.I. agent literally
• 27:34
  told me it was like. It said something like, this
• 27:38
  is how you get banned. I was like, well, sorry,
• 27:41
  man. I mean, like, geez, come give me a break.
• 27:44
  So and then we were keeping an extra copy of
• 27:47
  every feed and memory even after the XML parsing was
• 27:53
  done in order to potentially flag it later. So we.
• 27:58
  It refactors some of that. It's just a bunch of,
• 28:02
  you know, a bunch. Oh, and it found, it did
• 28:06
  find another bug with one of the. It being the
• 28:09
  robot. The agent, yeah. Yeah. Another runaway API query. Yeah.
• 28:15
  that was a result of this recent refactor. So that
• 28:18
  was a bug it created, so it had to fix
• 28:21
  it. So anyway, it's probably... It was an all-day adventure,
• 28:28
  let's say that. Well... It's highly appreciated, brother. The stuff
• 28:32
  you do. The stuff you do. Let's bring in our
• 28:35
  guest. Let's bring in our guest because he's been waiting
• 28:37
  very patiently. And if we let him wait too long,
• 28:41
  he'll be drunk because he likes to drink wine around
• 28:43
  this time. Because our guest today is coming in from
• 28:46
  Marlowe. He is the CEO of True Fans, co-host of
• 28:49
  Pod News Weekly Review. And as of last weekend, a
• 28:52
  sexagenarian who would like it noted that growing old is
• 28:56
  mandatory, but growing up is optional. Ladies and gentlemen, say
• 29:01
  hello to Sam Sethi. Ah, good evening, gentlemen. Good evening.
• 29:06
  Can I just point out all I understood from what
• 29:08
  Dave said was nanu nanu. It meant nothing to me.
• 29:13
  I have no idea what you said, but it sounds
• 29:15
  amazing, Dave. Thank you. Well. If amazing means fixing problems
• 29:21
  that you yourself created, I agree with you. It was
• 29:23
  amazing. I got to ask you one before we talk
• 29:26
  about podcast stuff, though. Your wine business, are you seeing
• 29:31
  an increase? Because I have a friend here in town.
• 29:36
  who also is a wine importer, and he said the
• 29:40
  amount of non-alcoholic wine is way up. Yeah, it is.
• 29:47
  Are you seeing that? Yeah, more people are going for
• 29:50
  lighter wines or non-alcoholic wines. It's a shame. This is
• 29:53
  a travesty. This will not stand. It's the young, not
• 29:57
  the old. It's actually demographically young.
• 30:00
  But I mean, how do you even make non-alcoholic wine?
• 30:03
  Do you, I mean. It's called Ribena, I think. He
• 30:06
  said they, he said they like centrifuge it or something.
• 30:10
  It's just, it's not, I mean, there's trends like orange
• 30:13
  wine now as well. It's like, it's just becoming silly.
• 30:18
  i mean with with the warming that we've had this
• 30:20
  summer a lot of the european wines it's just grape
• 30:24
  juice yeah but they've they've actually ripened too quickly so
• 30:29
  the problem is you're now getting white wines are great
• 30:33
  but red wines uh it's not good for it so
• 30:35
  it's not going to be a good summer for wine
• 30:37
  anyway we we digress So it's juice boxes, basically. Juice
• 30:43
  boxes. Juice boxes and shoko love. What more could you
• 30:47
  ask for? It's a beautiful thing. So Sam did what
• 30:50
  everybody should do. He sent us an email and said,
• 30:54
  Hey, man, I want to be on the show. And
• 30:57
  that's how you do it. And because you got stuff
• 31:01
  to talk about, because if there's any... For any member
• 31:05
  of the podcasting 2.0 band of misfits, it has got
• 31:09
  to be true fans that is literally implement. I mean,
• 31:13
  they run out of tags. It's like we don't have
• 31:15
  another thing to put in there. Could you please make
• 31:17
  another tag? We need something else. We need more tags.
• 31:20
  We reached the end of the namespace. We need more.
• 31:23
  I literally sent an email to Mo saying feature complete.
• 31:26
  there's nothing left that's where you need to start making
• 31:33
  bugs so that you can fix them so you can
• 31:35
  have something to do now we've got enough of those
• 31:39
  still so you now what I always appreciate about Sam
• 31:42
  is he will update the the podcast index.social with the
• 31:48
  latest things that he's working on. Although it always kind
• 31:51
  of sucks the way it shows up because it comes
• 31:53
  in reverse order. You know, I have to go, oh,
• 31:55
  let me go find out, you know, okay. And I
• 31:57
  have to go and go to your click on it
• 31:59
  to get the first one and they go down the
• 32:02
  list. But, man, you've you've done a lot of new
• 32:04
  things, particularly on the value for value side with new
• 32:08
  payment stuff. Where do you want to start? I think
• 32:13
  one thing I'd love to start because it's related to
• 32:15
  Godcaster is publisher feeds versus network feeds, if you don't
• 32:19
  mind. Yeah. So Oscar came up with the idea of
• 32:23
  publisher feeds and that was about two, three years ago.
• 32:27
  It's been ratified as a standardised tag within the podcast
• 32:30
  namespace. Brilliant. Brilliant. No one's implemented it and I don't
• 32:34
  know how to get other hosts to do it. But
• 32:37
  what is a publisher feed? It's fundamentally podcasts from one
• 32:42
  single producer that you aggregate. So it's a great form
• 32:46
  of discovery. I land on... Smart list as an example
• 32:51
  and then we say the publisher is wondering you click
• 32:54
  on the one tree link and boom You get all
• 32:56
  the other podcasts from wondering you go. Oh my god,
• 32:58
  okay I didn't know about this this isn't okay become
• 33:01
  a fan become a fan Click on the playlist and
• 33:03
  run it as a whole playlist. So that was really
• 33:06
  cool, but no one's doing it so We then had
• 33:09
  this issue that a lot of independent podcasters were saying,
• 33:14
  well, that's great for the big boys, but I want
• 33:16
  to form a wine podcast society. And I want to
• 33:21
  be able to start with my own podcast, but I
• 33:23
  want to invite people. I don't know, five or six
• 33:26
  other little podcasters. And we want to aggregate all of
• 33:29
  our listen times and our plays and then go back
• 33:31
  to the advertiser or sponsors and say, look, instead of
• 33:34
  having 100 plays on one podcast and 10 plays on
• 33:37
  another, we've got a thousand plays as a network. Can
• 33:41
  we now get sponsorship or can we now get sponsorships?
• 33:45
  some advertising or whatever it may be, right? And so
• 33:49
  the difference between a publisher and a network feed is
• 33:51
  a publisher is a single publisher with multiple podcasts within
• 33:56
  it that you can use for discovery and a network
• 33:58
  is you are not the publisher of any of those.
• 34:01
  You aggregate and invite people to join you But the
• 34:04
  analytics in your dashboard shows you all of the aggregated
• 34:09
  totals so that you can then act like a big
• 34:11
  player and take it to somebody else for monetization. This
• 34:15
  is a very good explanation. And as it relates to
• 34:20
  Godcaster, I would say that because we have an integration
• 34:25
  network. partnership with RSS.com, and we're seeing a definite uptick
• 34:31
  of stations who are creating podcasts, you know, like –
• 34:40
  uh their own station podcast but they may have one
• 34:43
  from a local pastor or something from the city council
• 34:45
  you know some other ones and this would be a
• 34:48
  great place to integrate that i'm the funny thing is
• 34:51
  sam is that people they really don't seem to care
• 34:54
  that much about stats it's so crazy i i the
• 34:59
  for the um uh For the linkage side, you know,
• 35:04
  this podcast belongs to this network, that kind of stuff
• 35:06
  they understand. Our customers, they're not, I mean, they don't
• 35:09
  have ads or sponsorships, most of them, I would say.
• 35:13
  But in general, it seems just like the hosting companies,
• 35:17
  this is something they need to jump on board with,
• 35:19
  and we should probably start asking them to do that,
• 35:22
  because I presume True Fans Hosting does this. Yes, we
• 35:25
  do already. Of course. Why am I even presuming? And
• 35:30
  it's like, yeah, I mean, I put a link in
• 35:32
  the email I sent you, but the idea… is that
• 35:36
  we copied the publisher feed format in the structure. So
• 35:40
  in the RSS feed for the podcasts we host, we
• 35:43
  also put podcast colon network, and then that links to
• 35:48
  a file which lists all of the podcasts within that
• 35:51
  network. So we copied the same structure, and I'm going
• 35:53
  to put it up on the GitHub page, So that
• 35:55
  hopefully, you know, everyone can review it and then maybe
• 35:58
  we can get that approved as well. But the idea
• 36:01
  is that within the RSS feed, we now have pod
• 36:05
  roles. We now have publisher feeds. We now have network
• 36:08
  feeds. We have merch feeds. So we're trying to aggregate
• 36:11
  information. Lots and lots of things. I call it a
• 36:14
  super feed, but that's just our internal name for it.
• 36:19
  The thing you're – it sounds like you're describing –
• 36:25
  is less of a network and almost like a collective
• 36:28
  or a co-op. Correct. Yeah. And that was something that
• 36:31
  your hosting customers asked for? Yeah. So a lot of
• 36:35
  them were saying, look, I keep going and asking somebody
• 36:38
  to sponsor my podcast or I ask somebody to put
• 36:41
  DAI type advertising into it. and they go how many
• 36:46
  plays have you got how many listeners have you got
• 36:48
  what how many fans and the number is always small
• 36:51
  and so in the long tail it's really hard for
• 36:53
  those people to go and find people to you know
• 36:56
  pay them so if you can then say well my
• 36:58
  podcast is about wine and i've got 20 other wine
• 37:02
  podcasts and collectively we can then act as one So
• 37:06
  the whole is greater than the sum of the parts,
• 37:08
  then it works. And we also, in the analytics, drill
• 37:11
  down. So in terms of the money splits, it's easy
• 37:15
  to work out which one's got 10% of the plays,
• 37:17
  which one's got 5%, etc. And so you can split
• 37:21
  the monies back very easily as the admin of that
• 37:24
  network. Okay. What I thought was most interesting, and you've
• 37:29
  been working on this for a long time and I've
• 37:31
  been hearing you talk about it. By the way, two
• 37:33
  weeks of no pod news weekly review is tough, I
• 37:36
  have to say. We're doing it tomorrow morning. I can
• 37:40
  tell you why we had. So James was in New
• 37:42
  York for one week and then... We recorded two hours
• 37:45
  on Thursday as we normally do. And unfortunately, James sounded
• 37:50
  like he was in a water tunnel. Oh, no. What
• 37:54
  happened? What happened? There is a super secret new recording
• 37:58
  tool from our friends at Buzzsprout. And you decided to
• 38:01
  use the super secret new recording tool? Well, we've done
• 38:06
  it once before, but actually James says it was somewhere
• 38:09
  between the Rodecaster's SD card failing and obviously what was
• 38:13
  being tested. I hate it when that happens. I mean...
• 38:17
  That's the beauty of streaming live is there's always some
• 38:20
  crazy guy out there who's recording it. I've had that
• 38:23
  happen many times. Like, holy crap, everything failed. And I
• 38:26
  got a copy here for you. Don't worry about it.
• 38:29
  Yeah. It'll be out on Sunday. Oh, okay, good. So
• 38:34
  you're talking about it more. You're posting about it more.
• 38:38
  The idea of hosting no longer just being podcast hosting,
• 38:43
  but more like a creator portal. Yeah. Now, look, you've
• 38:48
  talked about this in the past. You've called it the
• 38:50
  Rachel Maddow problem. Well, Rachel Maddow is a problem, but
• 38:55
  yes. Okay. Continue. I think what I'm seeing is that
• 39:01
  podcasters doing audio is great. And I still insist that
• 39:06
  a podcast can only be a podcast if it has
• 39:09
  an RSS feed. Otherwise, it's not a podcast. Now, because
• 39:12
  the alternative enclosure, that can be video as well. It
• 39:16
  doesn't have to be just audio. but it has to
• 39:18
  have an RSS feed. Anyway, that's just me. But what
• 39:21
  we've seen is that people have started to want to
• 39:23
  aggregate all of their content that's been disparately spread across
• 39:28
  the web. So in the past, you might have had
• 39:30
  a WordPress site, and then you might have had a
• 39:32
  sticky meal site or a Wix site. And all your
• 39:36
  contents everywhere, your merch, your blogs, Now what we're seeing,
• 39:40
  not just with Truefans, I think, you know, Substack, Supercast,
• 39:44
  Memberful, Supercast, all of these sites are now becoming creator
• 39:48
  portals. And I think that is the arrow of direction
• 39:52
  I hope and see that creators will go down. So
• 39:56
  what would that include? For us,
• 40:00
  That's so audio, video, a blog, live events, ticketing, merch,
• 40:06
  network or publisher feeds. And all of that gets aggregated
• 40:11
  into one RSS feed as well from Truefans. So when
• 40:14
  we publish to the podcast index, you get that full
• 40:18
  RSS feed. Of course... Any app can then read it,
• 40:21
  but most apps won't read half of the RSS reader.
• 40:24
  And I think you've also talked about adding newsletter functionality.
• 40:28
  Yeah. Yeah, so we've just built our own newsletter. So
• 40:32
  we're just going to push one out very quickly. And
• 40:34
  then, yes, that's the last bit we want to add
• 40:37
  for creators so you can have an own newsletter. your
• 40:40
  own email list and then send that out to people
• 40:42
  who have subscribed to your newsletter that's that's gotta be
• 40:46
  a hard thing to get into yes i mean like
• 40:49
  with all email is so restricted and broken broken broken
• 40:54
  broken we use send grid um you know is it
• 40:58
  good is it bad we'll find out um i think
• 41:01
  You know, we're coming from a very hard place. I
• 41:05
  mean, you've got Substack whose core was the newsletter, you
• 41:08
  know, and they're now building audio, video, live premium content.
• 41:13
  They're building all of this stuff around it. And, you
• 41:16
  know, So we're coming the other way. We're coming from
• 41:19
  an audio first and then trying to build these other
• 41:21
  features onto it. I will say one thing that I
• 41:23
  tell people over and over again, particularly with value for
• 41:27
  value. So if you're supported by listeners, the newsletter is
• 41:32
  essential and people have no idea how essential it is.
• 41:38
  So to be able to send out an email the
• 41:41
  day before you have an episode come the day, the
• 41:43
  day before and the timing also matters. And that's, you
• 41:46
  know, that's a testing issue. And what I've what we've
• 41:50
  been doing for 19 years. is hey we've got a
• 41:54
  show tomorrow here's a funny picture remember you have to
• 41:57
  support us donate here's what's come on up on the
• 42:00
  show and remember you have to support us donate and
• 42:03
  uh here's another fun picture remember you have to support
• 42:06
  us donate and it is typically responsible for i'd say
• 42:10
  close to 40 of donations come in after the newsletter
• 42:14
  goes out Because it functions as just a reminder? Yeah.
• 42:22
  Is that what you see? That's what it is. We're
• 42:24
  all attention deficient, so we need things that land just
• 42:28
  in time. So we get to do this. We've been
• 42:32
  using Substack. Now, unfortunately, Substack won't just let you... keep
• 42:37
  your own RSS feed and import it. If anyone's listening
• 42:40
  at Substack, you're idiots. Because that would be like PodPage.
• 42:45
  That would just be perfect if you allowed for that.
• 42:48
  But no, we have to use our feed and all
• 42:50
  this stuff. But what they have built in is they
• 42:54
  have... the financial model built in. So with one click,
• 42:58
  you're paying, which is great. On the other hand, it's
• 43:02
  not great because they're always popping up little things like,
• 43:06
  hey, you should spend money on this, which I hate.
• 43:10
  You know, it's like, no, no, it's value for value.
• 43:12
  Don't put the free option at the bottom. Let people
• 43:14
  know. You know, so there's some problems with that. But
• 43:17
  that integration, if I was starting from scratch and I
• 43:21
  didn't, for whatever reason, I would trust them as a
• 43:25
  host, which I just, I'm not seeing that right now.
• 43:29
  I want features. I want 2.0 features. They don't offer
• 43:31
  that. I can see why people go with that. So
• 43:34
  I think that integration and once you sub stack has
• 43:37
  has mass adoption with people who've already put their their
• 43:41
  payment details in. So the value part is like one
• 43:45
  click, boom, done. Okay, you're good to go. That's the
• 43:48
  amazing part. And look, they're building on a really strong
• 43:53
  core function, which is the newsletter and the blog. So
• 43:57
  it's – you know, Substack is being – it's being
• 44:00
  inshittified. I mean, you can just see – It is.
• 44:02
  It is a bit, yeah. It's gotten demonstrably worse over
• 44:06
  time. More stay on our platform, please don't leave. It
• 44:10
  just – It's a worse experience in every way. The
• 44:14
  app is horrible. It is. It's so bad. It's not
• 44:18
  as bad as Medium. You've got articles, notes, comments, posts.
• 44:25
  I don't know what's what anymore. It's too much. Well,
• 44:28
  the CTO has just left... and also to join OpenAI.
• 44:33
  Really? From the factory to the fire. Nice. Well, to
• 44:37
  start a creator network on OpenAI. I mean, that's what
• 44:41
  his announcement was. Yeah, so... And, you know, they added
• 44:45
  a podcast button and stroke tab on Substack because people
• 44:50
  couldn't find podcasts. Right. But you go try and look
• 44:52
  for that and find that tab. I still can't find
• 44:55
  it. So I think the UI is really a mess
• 44:57
  right now. It is. And it makes me – it
• 45:02
  brings up the – Something that me and Adam talked
• 45:05
  about last week, which is just everybody's trying to do
• 45:09
  everything now. It's not like you have hosting companies. Well,
• 45:15
  I think it started with things like Riverside and Descript
• 45:19
  putting in hosting. And now it's spreading to where, well,
• 45:23
  everybody feels like they have to do everything because everybody
• 45:27
  else is doing everything. And so now you're going to
• 45:30
  – everybody is going to have to – I think
• 45:33
  there's this pressure going on for all platforms to give
• 45:37
  you every possible thing to do. Well, I talk about
• 45:41
  – Sorry, thank you. No, no. I was just going
• 45:44
  to finish up by saying true fans, at least you
• 45:47
  understand podcasting. Like Substack, I mean, they may understand it
• 45:51
  a little bit, but they started out as a completely
• 45:55
  – it's a bolt. Just like Spotify started out with
• 45:58
  music and they bolted on podcasts later, I feel like
• 46:01
  the podcast first world – It makes more sense to
• 46:05
  bolt on other podcast features to start out as a
• 46:08
  podcast hosting company and then release like a like a
• 46:11
  recording tool or something like Buzzsprout instead of starting out
• 46:15
  as a. You know, it's like a meeting tool like
• 46:18
  Zoom and then adding podcasting onto the side of it.
• 46:20
  I have to say I just because Sam said it
• 46:22
  challenge accepted. I went to the I got the Substack
• 46:26
  app. I hit podcast and it shows a whole bunch
• 46:29
  of podcasts. And this is the interesting thing. So I
• 46:32
  put in we get to do this. And it shows
• 46:36
  it as the top hit in the podcast search. Then
• 46:40
  I tap and it shows that. Now, these episodes are
• 46:43
  just articles. I have no player embedded or anything, just
• 46:47
  purely a link to our own player. I think we
• 46:51
  link to the bridge page. Godcaster player. I tap on
• 46:56
  it. It pops up a page. It still has the
• 46:59
  click here for this episode, but it has a listen
• 47:02
  button. And I hit the listen button and boom, a
• 47:04
  player opens and it's playing our podcast. I don't know.
• 47:09
  So they recognize it somehow. And they I give them
• 47:13
  no feed. I give them nothing. Somehow they recognize it.
• 47:17
  That's that's rather outstanding. I have to say. Huh. So
• 47:22
  are you saying that you don't have a presence on
• 47:24
  Substack? No, we do. Somehow they have your content. No,
• 47:27
  we we our newsletter is on Substack. And. We post,
• 47:34
  so we'll create a new article when the episode comes
• 47:36
  out, which is basically show notes, and we post a
• 47:39
  link. But I didn't put the listen button on there.
• 47:41
  They did that. So they're smarter than I thought, actually.
• 47:45
  So they're not pulling your feed, is what you're saying?
• 47:48
  No, they're not. As far as I know, they're not.
• 47:50
  So it sounds like they're almost re- Building you another
• 47:54
  feed on their side and in effect. Because you have
• 47:58
  a play button next to all of your episodes, but
• 48:01
  it's only in their silo. Yeah, but if I say
• 48:05
  – if I'm looking for other episodes and it doesn't
• 48:08
  – I don't understand it. This is the problem. This
• 48:13
  is the problem is it's confusing. It is confusing. And
• 48:17
  you're the one who's – It's your account. Yeah. Yeah.
• 48:21
  Well, it's better than what I thought, but it's also
• 48:24
  much worse than I thought. Oh, well. I was just
• 48:27
  going to say, Deb, I think what we're seeing is
• 48:29
  a change from what I call horizontal to a vertical
• 48:33
  world. So horizontal was Buzzsprout was the host and then
• 48:37
  Overcast was the app or Pocketcast was the app. And
• 48:40
  then you've got WordPress for your blog and you've got
• 48:43
  something else. And I think we're now spinning that to
• 48:46
  a vertical where... Everyone does hosting through and you know
• 48:50
  Fountain and ourselves became hosts not because we wanted to
• 48:54
  I spoke to Oscar about this. He didn't particularly want
• 48:57
  to, but we were frustrated. asking all the existing hosts,
• 49:02
  add video, add this. Why is that feature not there?
• 49:06
  Who's got a pod role? Why don't you do this
• 49:08
  tag or that tag? And it was like, okay, well,
• 49:10
  if you're not going to do it, guys, we're just
• 49:12
  going to come and go, we'll do the hosting part.
• 49:14
  And the cost of hosting is minuscule. you know the
• 49:19
  the cost per gigabyte is literally 0.001 it is not
• 49:24
  expensive anymore so you know in the technology escalator it's
• 49:29
  now the bottom of the escalator in terms of feature
• 49:32
  and function and you're now seeing all the hosts are
• 49:34
  having to up their game to differentiate and Yeah. So
• 49:38
  we just went through the whole video upgrade. Everybody pretty
• 49:41
  much did that. Do you see a huge uptake as
• 49:44
  was expected? No. Yeah, I didn't think so. No, pretty
• 49:48
  much. That's as expected. I think it's good that we've
• 49:53
  done it. I think leaving the video to YouTube and
• 49:57
  Spotify was stupid. Um,
• 50:00
  I'm glad that we've all upped our game and done
• 50:02
  it. Will we see a massive uptake? I don't think
• 50:05
  so, but, you know, at least we've ticked that box.
• 50:08
  I mean, you know, Amazon this week have added the
• 50:10
  alternative enclosure. They've added video in the alternative enclosure, which
• 50:14
  is really well done. So congratulations to them. But Apple...
• 50:19
  So as a host, I'm still having to use the
• 50:21
  proprietary APIs from Apple, Spotify and YouTube to send video.
• 50:25
  It's not, you know, they don't support the alternative enclosure.
• 50:29
  They don't support in the primary enclosure HLS. So we've
• 50:32
  still got a long way to go in order to
• 50:34
  get the adoption of video properly in those clients. So
• 50:38
  is TrueFans hosting? Are y'all publishing in the alternate enclosure?
• 50:45
  Yes. Yeah. So we, before any of the Apple API
• 50:50
  stuff came out, we had made the decision to upload
• 50:54
  MP3 and MP4 and convert to HLS. So we did
• 50:57
  that. Why did we do that? it goes back to
• 51:00
  the other big issue that's going around what is a
• 51:03
  play how long should a play be and how can
• 51:06
  you get first party data and we were like okay
• 51:09
  we can't beg borrow or steal apple will not give
• 51:12
  it to us spotify or youtube and you look at
• 51:16
  bumper you know they have to ask their clients to
• 51:19
  give them a login and then they act and then
• 51:22
  they aggregate in some dashboard but it's micro data just
• 51:25
  for their clients now we worked out and i'm gonna
• 51:28
  you know i told you in the email i'm now
• 51:30
  gonna reveal how we're doing it because i think everyone
• 51:33
  should be doing it because we've got to break the
• 51:35
  monopoly of these three big players so There is something
• 51:39
  that we did, we converted all of the content, MP3
• 51:43
  and MP4 into HLS, and then we use this HTTP
• 51:47
  206 partial content. What that does is it actually asks
• 51:53
  for a range request. So we're streaming content down to
• 51:57
  the client. and the range request is being pushed. At
• 52:01
  the point when the person stops requesting content, i.e. they
• 52:05
  stop playing, we know exactly that A, they've played it,
• 52:09
  and B, how long they've played it, and then we
• 52:11
  can put that in our metrics, and that's how we
• 52:13
  do it. So we can do that across all apps,
• 52:15
  and it works really well. Mm-hmm. So the issue that
• 52:20
  leaves, though, is the thing that we're facing right now.
• 52:26
  I've gotten two questions in the last couple of weeks
• 52:30
  of people texting me and saying, hey, is this y'all?
• 52:37
  And – They gave us a screenshot of they gave
• 52:41
  me a screenshot of a user agent claiming to be
• 52:46
  podcast index. And it is not us. And that's the
• 52:50
  and I this we know this is I mean, obviously,
• 52:54
  this has been happening forever. But it seems like this
• 52:59
  is now a huge problem. User agent faking is now
• 53:04
  just the norm. You have no idea whether or not
• 53:08
  Apple Podcasts or Overcast or Podverse or Truefans, you'd have
• 53:14
  no idea if that's legitimately who they are. But you
• 53:17
  cannot tell anymore. No, and so my learning curve has
• 53:23
  been vertical since we became a host. So the first
• 53:25
  thing was, okay, how do we do this? How do
• 53:27
  we get our own CDN? How do we make the
• 53:30
  optimisation? So great learning. then we did the hls learning
• 53:33
  and then everyone jumped on the hls bandwagon brilliant and
• 53:37
  so that's the next thing and then we started to
• 53:39
  get all the streaming data coming back and it was
• 53:41
  like hours and hours of stuff and we were seeing
• 53:44
  bots all over the place and so we built a
• 53:47
  wath which is a firewall uh and we basically allowed
• 53:51
  the individual creators then see all the bots we worked
• 53:55
  with john spurlock's oporg list to begin with and so
• 53:59
  we could see the the ones that had truly said
• 54:03
  they were a user agent and they were bots and
• 54:06
  then as a creator you can then turn them on
• 54:08
  or off so the reason why we gave the option
• 54:11
  was because some people said no i want to be
• 54:13
  indexed by open ai no i want to be indexed
• 54:16
  by these bots and other people like well that's costing
• 54:19
  me money no turn it off so we gave that
• 54:22
  option but now we're getting uh unknown um data in
• 54:27
  terms of streaming it's quite a large percentage and We
• 54:31
  don't know how to fix it right now. Seems like
• 54:33
  it was 60% or something when I saw that graph
• 54:35
  you posted. Yeah, it's really high. And I think the
• 54:39
  answer is Apple. And so if these bots are apps,
• 54:45
  forget web for a minute. If these bots are apps
• 54:48
  and that bot that James highlighted is. which was removing
• 54:53
  the ads and providing some skipping capability. That's up in
• 54:57
  the app store. But the first version was spoofing itself
• 55:02
  as overcast. Now it's worked out that he can turn
• 55:05
  off that and he doesn't have to even give a
• 55:08
  user agent. Apple banks 30% from that app because it's
• 55:12
  a paid app. Apple could enforce every app to have
• 55:16
  a user agent before it is allowed into the app
• 55:19
  store. And that would fix it because then it would
• 55:22
  be a whitelist that we as hosts could then point
• 55:26
  ourselves to, oh, that's what it is. Okay, that would
• 55:29
  fix it, but Apple's never going to do this. I
• 55:32
  don't know if they're interested. So you're saying that this
• 55:36
  particular app is now claiming to be something like a
• 55:40
  web browser? Yeah, it can claim to be safari. It
• 55:44
  knows now that if it claims or reports itself, it
• 55:47
  will be blocked. So now it says, well, I'm going
• 55:50
  to illegally claim to be something else. And no one's
• 55:53
  policing me. So happy days. You know, that is interesting
• 55:57
  because that could be I'd seen I'd seen James mention
• 56:00
  that or heard him mentioned it. And I wasn't really
• 56:03
  following what he was saying, but now that you lay
• 56:06
  it out like that, I think I understand because if
• 56:09
  there was a user agent enforcement mechanism on apps, that
• 56:12
  really would make a big difference. That would help. Yeah,
• 56:15
  it definitely would help. And it's not like it's completely
• 56:18
  out of line with the other enforcement. I mean… Apple
• 56:22
  and Google enforce lots of things on app developers, and
• 56:26
  we know this because they – sometimes app developers have
• 56:29
  to email us and say, hey, I submitted my app
• 56:32
  for app review, and they say that I'm not allowed
• 56:35
  – they won't approve it unless I get written approval
• 56:39
  from Podcast Index that I can use your API. So
• 56:43
  they already enforce lots of different requirements like that. That
• 56:47
  is something that Apple and Google – I don't know
• 56:51
  that I agree with you, Sam, though, but I think
• 56:53
  that sounds like something they absolutely would do. I don't
• 56:56
  see how it hurts them at all. Well, how do
• 56:59
  we get them to do that? You know, we have
• 57:02
  friends in both camps, but nobody seems to want to...
• 57:08
  Nope. Hello? Sorry about that. I'm sorry about that. I'm
• 57:13
  sorry. I'm sorry. I was on the phone with Apple.
• 57:16
  Okay? I felt my brain turned inside out for a
• 57:19
  second. I'm sorry. I was on the phone with Apple.
• 57:21
  Yes. I don't know how you – that's the question
• 57:26
  of how do you get giant mega – Mega national
• 57:31
  corporations do anything. We can hear you coming back, Adam.
• 57:37
  You still hear me? Yeah, I'm hearing it. You're bleeding
• 57:40
  through, brother. I'm bleeping through. I'm sorry. All right. Thank
• 57:43
  you. I'll stop now. How do you get these humongous
• 57:47
  corporations to do anything? That is an unknown thing. Well,
• 57:52
  I think if you talk about it more and talk
• 57:55
  about it as a pain point, I mean, this week
• 57:58
  was crazy in podcast movement, you know. Tom Webster stood
• 58:05
  up on stage and said, you know, Netflix is basically
• 58:07
  one of the top three podcast platform apps. I'm like,
• 58:10
  where's that come from? Then Rob Walsh said, yeah, but
• 58:13
  Netflix has one in 67,000 podcasts, you know, so yours
• 58:18
  isn't going to be part of it. and then apple
• 58:21
  has a play that is zero you know literally one
• 58:24
  second um youtube has a play that is a hover
• 58:28
  um The amp record totally tried to get something to
• 58:32
  be 30 seconds. That lasted about 30 seconds. Yeah. That
• 58:38
  died as predicted. We knew that would die. Yeah. So
• 58:42
  nothing is, you know, so we've got this going off
• 58:44
  in the background about what is a play, what's the
• 58:46
  analytics. And then we've got user agents. There's a lot
• 58:49
  in the air right now I think that we have
• 58:51
  to fix and not quite sure how we go about
• 58:53
  it. Well, I think from experience, we all know that
• 58:57
  you that you can't fix stupid things at once. Well,
• 59:00
  there's that too. You can't you can't fix 10 things
• 59:04
  at once. You got to you got to like make
• 59:07
  make your list and hit the biggest ones. To me,
• 59:11
  the list of things that have to be fixed always
• 59:14
  come down into two camps. It's like, is it realistic
• 59:17
  this will ever be fixed? Well, okay, once you get
• 59:21
  through that filter, which one is the most likely to
• 59:24
  be fixed? Not which one is the thing that hurts
• 59:28
  the most, but which one is the – Which which
• 59:31
  issue that you're trying to solve is most likely to
• 59:35
  receive a favorable ear from the people from from this
• 59:39
  other group? And if you can if you hit that
• 59:44
  one. And then the next most likely, instead of trying
• 59:48
  to solve your hardest problem first or your worst problem
• 59:51
  first, if you solve the one that you're most likely
• 59:53
  to find agreement on first, then over time you build
• 59:57
  a consensus or a base in order to start working.
• 1:00:00
  next hardest thing. And I mean, I think at least
• 1:00:04
  Spotify has shown zero interest in playing in the open
• 1:00:09
  podcast world at all. They just seem to not care.
• 1:00:13
  Apple, on the other hand, I think Apple Podcasts, that
• 1:00:16
  team over there, I think they would love nothing more.
• 1:00:20
  than to see App Store review enforce user agents to
• 1:00:25
  make these other podcast apps identify themselves. If they do
• 1:00:27
  it, everybody will do it. So that would be great.
• 1:00:31
  Yeah, I think the biggest problem is it's mini-thiefdoms inside
• 1:00:34
  of Apple. So the podcast team won't have App Store...
• 1:00:40
  input i mean you look at apple it it's got
• 1:00:44
  siloed all over it right so the the app music
• 1:00:47
  the app audio books the app for podcasting all separate
• 1:00:50
  look at amazon exactly the same audible is different to
• 1:00:54
  amazon music is different to wandry is different to art
• 1:00:57
  19. um this is corporate um, fiefdoms being exposed at
• 1:01:01
  the front end, whereas Spotify is an aggregated app. At
• 1:01:05
  least it appears to the consumer as one out with
• 1:01:08
  audio books, podcasts, and music. Well, we had, we had
• 1:01:11
  this solution kind of bubbling for a little bit, which
• 1:01:14
  is the idea of registering your, um, your ID, your,
• 1:01:22
  what's the word I'm looking for? Your, user agent? Yeah.
• 1:01:29
  Are you talking about the HTTP signing? Oh, whatever it
• 1:01:32
  is. I mean, yeah, HTTP signing. I think that's what
• 1:01:35
  it was. In general, look, Google forced HTTPS on us
• 1:01:39
  for everything. Yes. So why don't we get them to
• 1:01:44
  enforce this for everything? This would actually, it doesn't matter
• 1:01:48
  too much to me, but it does matter to people
• 1:01:50
  who are paying for bandwidth, although we just heard I
• 1:01:53
  should start a CDN because it's pennies. It's fractions of
• 1:01:56
  pennies now, so that's a new business I'm working on.
• 1:01:59
  But at the same time, it is costing people money,
• 1:02:02
  and it would be nice if we had the ability
• 1:02:05
  to say, no, not you, you know. But yeah, no,
• 1:02:09
  not you or only you or whatever it is. And
• 1:02:12
  that should go for browsers, too. Well, I don't really
• 1:02:18
  know about – I don't want to go back down
• 1:02:22
  the signing argument. You already tried that. Yeah, and we
• 1:02:27
  do it. We do it. That's how you can tell
• 1:02:30
  if it's us. If you want to know if it's
• 1:02:33
  the podcast index aggregator – We sign our request with
• 1:02:37
  a public key. You can verify that it is us.
• 1:02:39
  So we do – we put our money where our
• 1:02:42
  mouth is. We're walking the walk. And I don't want
• 1:02:45
  to like re-litigate that whole thing. Oh, my God. You're
• 1:02:49
  buzzing like crazy. I know. Well, not anymore. Move the
• 1:02:53
  chocolate. I swallowed my chocolate. But – Like I don't
• 1:02:57
  want to relitigate that whole thing, but at the same
• 1:02:58
  time, this – I think what we're seeing is an
• 1:03:04
  escalation in – The issue of apps being impersonated –
• 1:03:08
  it's an impersonation and escalation. That's what we're seeing. And
• 1:03:14
  like – so if you're an app – like I
• 1:03:17
  know that there's – you know, Spurlock and Mark and
• 1:03:20
  all these. I know that they have – There are
• 1:03:24
  legitimate reasons not to want to do this, but at
• 1:03:26
  the same time, if you see millions of requests claiming
• 1:03:29
  that they're you but they're not, that's also – you
• 1:03:32
  have to admit that's a problem. Because what if they're
• 1:03:37
  being abusive and it's not your app doing it? Yeah.
• 1:03:40
  Well, the biggest problem is if you keep spoofing that
• 1:03:43
  you're overcast and suddenly people block overcast. Yes. And the
• 1:03:47
  legitimate overcast thing gets blocked. Then, you know, Marco's going
• 1:03:51
  to be very upset. And I would be too if
• 1:03:53
  it was true fans getting blocked. So... Before we get
• 1:03:57
  to that problem, let's try and fix it because we
• 1:03:59
  know that's the problem that will bite us in the
• 1:04:01
  future. And I think that was the biggest misunderstanding of
• 1:04:05
  that whole argument is that it was looked at from
• 1:04:08
  the standpoint of if you have signing, then you can
• 1:04:12
  get blocked. But the whole point of it was to
• 1:04:15
  not get blocked. That was the whole point of it
• 1:04:18
  was if you sign your stuff, we'll know how we'll
• 1:04:20
  know you're that you are who you are and we'll
• 1:04:23
  know not to block you. And so, like, I think
• 1:04:26
  the incentive to. The incentive to not block a legitimate
• 1:04:34
  app is stronger than the incentive to block a legitimate
• 1:04:39
  app. So at what point would a hosting company or
• 1:04:46
  provider... Want to block legitimate download traffic from a real
• 1:04:52
  app? I can't think of one. So Tom Rossi at
• 1:04:58
  the PSP has proposed this. I think it's on the
• 1:05:01
  GitHub as well. The only other negative I heard was
• 1:05:06
  the podcast apps that are serverless wouldn't be able to
• 1:05:10
  participate. Now, I don't understand that enough. But that's one
• 1:05:13
  of the arguments. Say that again. Say that again. So
• 1:05:16
  the apps that are serverless would not be able to
• 1:05:20
  participate in a signing because they don't have a back
• 1:05:24
  end to sign against. I understand the argument, but I
• 1:05:27
  think that's easy to work around with like an escrow.
• 1:05:30
  Yeah. Yeah, I agree. I don't think that's a problem
• 1:05:33
  at all. I mean, I would – the podcast, I
• 1:05:36
  would be totally willing to host an escrow on behalf
• 1:05:38
  of serverless apps. I mean, like, completely open. We could
• 1:05:43
  – it's – it costs us nothing. Why don't we
• 1:05:46
  just – how about this? Let's just do it. Let's
• 1:05:50
  just build it. Build an escrow for all apps, for
• 1:05:53
  all I care. And just make it a thing. And
• 1:05:57
  then maybe someone will pick it up and say, oh,
• 1:05:59
  look what these guys are doing. And that's good. I
• 1:06:01
  mean, that's how things usually work in this world. You
• 1:06:03
  build it. You push it. It's like any other type
• 1:06:06
  of content. Oh, here's a new tag. Oh, here's a
• 1:06:09
  key to sign with. Oh, I kind of like that.
• 1:06:11
  We can get a nonce or some sort of key
• 1:06:17
  signing material, give it to an app developer. They can
• 1:06:21
  sign it with their key, put their public key in
• 1:06:25
  our escrow, and then we have verified that they are
• 1:06:28
  who they say they are, and then we'll basically co-sign.
• 1:06:32
  I'm fine with that. You know we'll join you. You
• 1:06:35
  know that. As always. We've got nothing else to do.
• 1:06:40
  We've been waiting for you. You already finished the internet.
• 1:06:45
  We got to the end. We're like, now what? How's
• 1:06:49
  business doing in general, Sam? Pretty good. Daily signups of
• 1:06:55
  users about five, 600 people. Wow. Yeah, and hosts about
• 1:07:00
  20 to 30 a day minimum. With some days we
• 1:07:05
  spike at 50. So it's good. It's good. It's going
• 1:07:08
  well. We... I mean, I'm going to reveal something that
• 1:07:11
  you might not be aware of. So Mo, my developer
• 1:07:16
  who is stuck in Afghanistan. Afghanistan. Yes, I know. He
• 1:07:20
  was in Iran. He got out just before the Internet
• 1:07:23
  was turned off. He was in Afghanistan living in his
• 1:07:27
  uncle's. a farmyard just outside a car, but how that
• 1:07:31
  guy delivers anything to me, I have no idea. Sam,
• 1:07:33
  one day you're going to have to tell the story
• 1:07:36
  of how you know this guy. I mean, like, it
• 1:07:39
  sounds like he's always in the deepest, darkest. I'm thinking
• 1:07:43
  since he's from Afghanistan, it was probably Sam's old import-export
• 1:07:47
  business. Yeah. No, no, no, no, no. No, I won't
• 1:07:52
  tell you who, but I was at a podcast movement
• 1:07:55
  three years ago and they wanted to invest into true
• 1:07:59
  fans and we shook hands, quarter of a million. And
• 1:08:01
  then I got a phone call at eight o'clock the
• 1:08:04
  next morning because I'd spoken and said, oh, Mo's in
• 1:08:07
  Iran. You know, that's where my CTO is. And he
• 1:08:10
  went, I do work with the government. We can't do
• 1:08:12
  this. And that was it. It was just removed. I've
• 1:08:17
  never revealed this. I'm quite happy to do it on
• 1:08:20
  this show. Yes. We've raised two million out of 10
• 1:08:26
  million valuation. That's what we did. We did that back
• 1:08:28
  at the beginning. Congratulations. No, not excellent because while Mo
• 1:08:32
  is in Afghanistan, they won't give me the money. It's
• 1:08:35
  stuck in a bank account. So he's currently at an
• 1:08:38
  airport. All two million? All. Nothing will be. So he's
• 1:08:41
  currently at an airport on his way to Dubai. We
• 1:08:45
  have to now send a photo next week with a
• 1:08:47
  geolocated photo. Can't be AI. It's like, no. Proof is
• 1:08:52
  out. Of him standing in front of the Dubai airport,
• 1:08:55
  right? Something, anything. It's KYD, know your developer. You've got
• 1:08:59
  to do a whole – wait a minute, wait a
• 1:09:01
  minute. Hold on. Stop for a second. Right. So this
• 1:09:04
  is 2 million pounds sterling, I'm presuming. Yes. Yep. And
• 1:09:09
  from what I understand, you did this under an interesting
• 1:09:13
  – UK specific financing law for small business. EIS, yeah.
• 1:09:20
  Yeah, so you got a bunch of... High net worth.
• 1:09:24
  Yes, that's the term I'm looking for. High net worth
• 1:09:27
  who actually get a tax benefit for making this investment.
• 1:09:30
  Their tax benefit probably doesn't count until they make the
• 1:09:33
  actual investment, but I'm trying to make the connection to
• 1:09:37
  why this is being held because, Is it being held
• 1:09:40
  by them? Yes. The lead investors basically through due diligence.
• 1:09:45
  I'm not going to lie because it'll come out in
• 1:09:47
  discovery anyway. So when you tell them where he is,
• 1:09:51
  they go, oh, no, no, no, no, no, no, no.
• 1:09:53
  No, that's too dangerous. We won't do it. But if
• 1:09:55
  you can get him out and put him somewhere else
• 1:09:57
  where we think it's a safe country, we'll
• 1:10:00
  We'll release the money. And we're like, okay, come on.
• 1:10:02
  So we've been trying and trying. Oh, man. Hey, but
• 1:10:05
  listen, if you need help, I do have contacts in
• 1:10:08
  that area. Right. I do. I do. Okay. Certainly getting
• 1:10:14
  people out of a country. If he doesn't mind the
• 1:10:17
  military, C-130 military transport. I don't care how you get
• 1:10:22
  him out. It's not very comfortable, but we probably can
• 1:10:26
  bust them out, man, if he has some kind of
• 1:10:28
  paperwork. My wife will agree with you because our savings
• 1:10:31
  are dropping like a rocket right now with me funding.
• 1:10:34
  Yeah, I hear you. Yeah, it'll be lovely if we
• 1:10:36
  can because it's so, it's awful. I can see the
• 1:10:39
  money because I've got access to the account. It's sitting
• 1:10:42
  there, but I can't access it. Because if I do,
• 1:10:46
  then obviously the investors will come at me. I mean,
• 1:10:49
  the EIS is a great scheme in the UK. It
• 1:10:52
  stands for Enterprise Investment Scheme. And what it means is
• 1:10:55
  if you've got a tax bill, let's say a hundred
• 1:10:57
  thousand pound tax bill, you can put that money legitimately
• 1:11:00
  into a certified startup. Yeah, that's great. And then as
• 1:11:04
  long as that startup survives three years, you get 75%
• 1:11:09
  of that back as, you know, relief. Tax relief. Yeah.
• 1:11:14
  And if I fail, they get 75%. And if I
• 1:11:17
  make a killing, then it's capital gains free. So they
• 1:11:21
  don't pay tax on the upside. I love how the
• 1:11:24
  British I love how the British call. They just straight
• 1:11:27
  up call this a scheme. It took me a while
• 1:11:30
  to figure that one out. Yeah. You have the BBC
• 1:11:34
  licensing scheme as well. Everything's a scheme over there. Schemers.
• 1:11:38
  I'm telling you, the British schemers at least is honest.
• 1:11:41
  Yeah. Well, Sam, send me a note later. I'm happy
• 1:11:44
  to see if I can. can put you in touch
• 1:11:46
  with somebody who can help him. As long as he
• 1:11:47
  has some kind of paperwork, everything's possible. He literally got
• 1:11:51
  his passport about a week ago. Don't ask. Well, I
• 1:11:55
  don't ask. And he got a visa as well. Again,
• 1:11:59
  I don't ask. A visa for Dubai or for the
• 1:12:02
  U.S.? Yeah, Dubai. Well, this guy is critical to the
• 1:12:06
  podcasting 2.0 ecosystem. We need to help. We got to
• 1:12:10
  stick him in a hubcap and spin him out of
• 1:12:13
  there. Hold on to that wheel, mate. You're going out.
• 1:12:20
  Things are going to get a little dicey. Wow. Oh,
• 1:12:23
  I'm sorry to hear that, but we'll see what we
• 1:12:25
  can do for you, man. We'll definitely see what we
• 1:12:27
  can do for you. Well, I just find it amazing
• 1:12:28
  when I go to Mo every day, because we talk
• 1:12:30
  every day, and I go, "Hey, we've got this crazy
• 1:12:32
  idea for this feature or function, whatever." And he goes,
• 1:12:35
  "Yeah, leave it with me." And a day or two
• 1:12:36
  later, it's done. And I go, "How are you doing
• 1:12:39
  this?" the the you know i moan about my broadband
• 1:12:42
  connection in the uk and he's probably got you know
• 1:12:45
  one hand pointed out the window with an air to
• 1:12:48
  try and get some internet access, you know. Yeah, and
• 1:12:51
  this is pre, and he was doing this pre-AI. I
• 1:12:54
  mean, he was coding features in 24 hours way before
• 1:12:58
  AI even existed. Exactly. Yeah. Now, thanks to Claude, we...
• 1:13:04
  We are rebuilding all of TrueFans at the moment. Oh,
• 1:13:08
  really? Oh, okay. That's interesting. I have found that, I
• 1:13:12
  don't know if you're in the same boat, but when
• 1:13:14
  you build things very fast, you have a lot to
• 1:13:17
  go back and clean up after the fact. Oh, yeah.
• 1:13:20
  I mean, refactoring is essential. But one of the... By
• 1:13:25
  hook by crook, we got lucky. We didn't plan it.
• 1:13:28
  We built it as a PWA to begin with because
• 1:13:31
  we didn't have the iOS and Android native skills. So
• 1:13:35
  we said, right, we'll build it as a PWA. And
• 1:13:38
  then we put the wrapper around it and submitted it
• 1:13:41
  to the app stores and they approved it. What's really
• 1:13:45
  interesting is we don't then have to put for every
• 1:13:48
  new feature we push to the PWA. Yeah, you just
• 1:13:51
  push it to the PWA. Yeah, we don't have to
• 1:13:53
  get approval every time we have a new feature, which
• 1:13:56
  is why we've been able to accelerate so fast with
• 1:13:58
  all the things we do. But are you going to
• 1:14:00
  go fully native eventually? Exactly. now with now we're feature
• 1:14:04
  complete or at least i think we're feature complete well
• 1:14:06
  just stand by new new tags coming up but yeah
• 1:14:09
  i mean that's what we're doing right now because we
• 1:14:13
  can't get are you Yeah, exactly. Android Auto and CarPlay.
• 1:14:18
  Exactly. Now, are you considering putting everything in there, like
• 1:14:21
  with the kitchen sink, with the newsletter capability, basically creating
• 1:14:24
  a sub stack? It's going to be a pure native,
• 1:14:27
  yeah. We have so many micro issues that our user
• 1:14:32
  experience. experiences that you know limit us so good example
• 1:14:37
  when it goes to the iphone screen when you're playing
• 1:14:40
  a podcast yeah don't lock it button doesn't yeah the
• 1:14:42
  play button doesn't yeah you have to open it in
• 1:14:45
  order to play right it just doesn't work you don't
• 1:14:47
  have to tell us we've been we've been through this
• 1:14:50
  pain many many times And so I just bit the
• 1:14:53
  bullet and said to Mo, look, forget it. Let's just
• 1:14:55
  stop building new stuff. Let's get to a point where
• 1:14:58
  we're happy. And then let's just build a native and
• 1:15:01
  then submit them. So that's what we're doing. Yeah. And
• 1:15:04
  then you get to get on the bandwagon of waiting
• 1:15:06
  two weeks for approval. Yeah. Yeah. But that's just the
• 1:15:09
  way life is. I mean, it's just the way it
• 1:15:11
  is. You can't change that. You've got to play that
• 1:15:13
  game. Yeah, and I mean, Paul, because one thing Paul's
• 1:15:18
  doing right now with Godcaster is having to re-architect the
• 1:15:21
  whole thing for Android. What started it was Android Automotive.
• 1:15:27
  Mm-hmm. You know, which is the new version that's replacing
• 1:15:30
  Android Auto. And evidently, it's a completely different animal. And
• 1:15:34
  so that just really got none of the existing stuff
• 1:15:38
  framework that he was using would work. So he had
• 1:15:41
  to just refactor the whole app from scratch. And like
• 1:15:44
  that's going to be the new. But you're going to
• 1:15:46
  have to because Google. is way more aggressive at deprecating
• 1:15:50
  stuff than Apple is. Oh, we lost access to the
• 1:15:55
  Google store recently because we wanted it. Again, the UK
• 1:16:00
  is weird. We've got these codes that determine what your
• 1:16:04
  company is, right? So they're called SIC codes, S-I-C. It's
• 1:16:08
  a scheme. Yeah. Yes. And Audio UK has now got
• 1:16:12
  the UK government to agree on two new denominations. One
• 1:16:17
  is podcast producer and one is podcast sort of app.
• 1:16:21
  That's not the exact ones, but that's roughly what it
• 1:16:24
  is. There's two codes. And so we wanted to be
• 1:16:27
  listed as a podcast app. I don't know what the
• 1:16:31
  benefit or downside would be, but I wanted to do
• 1:16:33
  it. So we took the opportunity to take Trufans FM
• 1:16:39
  Limited as our company name from what it was before,
• 1:16:41
  which was Trufans Technologies. Anyway, because we've just changed the
• 1:16:47
  limited company name, Google just took us out of the
• 1:16:49
  app store. Oh, no. So we had to reapply. Apple,
• 1:16:53
  no change. Didn't care. Don't worry. They'll screw you with
• 1:16:56
  something else. They always do. That sounds like the American
• 1:17:00
  NAICS stuff. It's like a company... like industry code. Yes,
• 1:17:06
  exactly. Designate, yeah. I think that, isn't that, doesn't that
• 1:17:10
  all come down to like insurance or something? I'm not
• 1:17:12
  sure what the value of it is, but it was
• 1:17:15
  one of those, it came out, Audio UK had done
• 1:17:17
  the work. I thought, oh, let me try and do
• 1:17:19
  that for my own company then. you know, trying to
• 1:17:21
  eat the latest, greatest thing. And then, of course, I
• 1:17:24
  screwed myself over, but hey. So how do you remedy
• 1:17:27
  that? What do you do? We create a new account
• 1:17:31
  on Google in the Google Play Store. We resubmit and
• 1:17:35
  then we put it in. Start over from zero. Yeah,
• 1:17:38
  exactly. Wow. Well, this is depressing, Sam. Tell me something.
• 1:17:43
  No, it's not. Got to ramp up that wine business,
• 1:17:48
  man. Podcasting is a bust. No, I think what I'm
• 1:17:56
  seeing is – Premium content is now becoming much more
• 1:18:01
  prevalent. I think people are beginning to accept painful quality
• 1:18:06
  content. Stripe, I don't know if you saw, I posted
• 1:18:11
  yesterday, they've got pay by bank, which I think is
• 1:18:13
  really interesting. We've just enabled it. Again, I think there's
• 1:18:18
  stuff happening. We still haven't got that micropayment, you know,
• 1:18:22
  the way that we all want it to be. I
• 1:18:24
  think there's still hope there. I'm not sure how yet.
• 1:18:29
  And look, I think, you know, we are moving the
• 1:18:31
  needle forward. You know, HLS is there now. We have
• 1:18:34
  got audio, video. We are seeing the alternative enclosure used.
• 1:18:39
  If you look at all the tags... and the lag
• 1:18:42
  between thought and implementation is about three years but it's
• 1:18:46
  all happening So I'm really positive. Okay. I'm glad you
• 1:18:50
  are. If we can avoid – I guess the thing
• 1:18:57
  I'm mostly concerned about is just avoiding an economic crash
• 1:19:02
  at this point. I feel like the AI bubble is
• 1:19:05
  so close to popping. Well, it's only 3.8% of GDP.
• 1:19:12
  3.8% does a lot. I had some clips I was
• 1:19:16
  going to play today before we knew you were going
• 1:19:19
  to be on, Sam. And, like, I mean, I'm pretty
• 1:19:22
  concerned about that. And I think that's one of the
• 1:19:26
  reasons why Bitcoin is going up is – Not because
• 1:19:30
  of money printing, but because of – like what I
• 1:19:35
  was hearing was that the AI debt markets have grown
• 1:19:39
  so large that they're now competing with treasury auctions. That's
• 1:19:46
  right. That is insane. That cannot be a good thing.
• 1:19:52
  And you're talking about they're competing with treasury markets, and
• 1:19:54
  they're already at like 10% on a note. Really? Oh.
• 1:20:00
  This is not going to end well. This is the
• 1:20:02
  beginning of a... Because when you're... When you see that,
• 1:20:06
  that means the only way that the treasury market survives
• 1:20:08
  is by increasing to match those interest rates. You've got
• 1:20:12
  to increase interest rates, yeah. Yeah, you've got to increase
• 1:20:15
  the yield. Well, then we have Buffalo Besant. His idea
• 1:20:18
  is to get everybody to use our debt as a
• 1:20:20
  stable coin, which I think is a good gambit. I
• 1:20:24
  really do. I still think it's a good gambit. Yeah,
• 1:20:27
  and what I'm saying is Get Mo out of there
• 1:20:30
  as soon as possible. Get your money. Yeah, yeah. Yeah,
• 1:20:35
  put that stuff in cash and hold on. We're going
• 1:20:37
  to have to do an extract. You know how the
• 1:20:39
  CIA extractions work, Sam? It's like Mo has to go
• 1:20:42
  into a field and he has like this big rubber
• 1:20:45
  band. And the plane comes down with the hook. The
• 1:20:47
  hook. I'm not kidding you. Yeah, the hook. And it
• 1:20:50
  – and it whisks them off into the air behind
• 1:20:52
  the plane and then they reel them in slowly. So
• 1:20:55
  don't drop your laptop. Yeah. Exactly. He needs a backpack.
• 1:20:58
  That's got all the code on it. No, don't do
• 1:21:00
  that, please. Get him a backpack. Hey, Sam, we appreciate
• 1:21:03
  so much what you do, man. We joke around. We
• 1:21:07
  have fun. But you've been leading so much in all
• 1:21:11
  of the 2.0 development. And we really want you to
• 1:21:15
  succeed. So whatever we can do. Thank you. Obviously, we
• 1:21:18
  want to help. you can hang around if you want
• 1:21:20
  I'm just going to thank a few people because I
• 1:21:22
  did hear some some boosts coming I'll hang in there
• 1:21:24
  hang in there very good I have to tell you
• 1:21:27
  my family by the way is not allowed between 6.30
• 1:21:30
  and 8 o'clock to talk to me on a Friday
• 1:21:32
  night I understand we've had arguments it's like my wife's
• 1:21:38
  booked theater tickets I'm not going Oh no, the dinner,
• 1:21:42
  oh no. I don't go anywhere. It's like, nope. at
• 1:21:45
  least you went to your own birthday party in the
• 1:21:47
  back garden well that's a Saturday no it had to
• 1:21:50
  be a Saturday okay okay let me see this is
• 1:21:54
  so we are value for value everything you support this
• 1:21:58
  podcast with doesn't go to us it goes straight into
• 1:22:01
  the account for podcastindex.org and obviously there's a lot going
• 1:22:06
  on a lot of machines. Mindy needs to be fed,
• 1:22:09
  so please, you know, support us. Podcastindex.org at the bottom
• 1:22:13
  there is a big red donate button. You can support
• 1:22:16
  us with PayPal. We also got some boosts coming in.
• 1:22:18
  Actually, we got a lightning invoice payment. I'm not quite
• 1:22:22
  sure where that came from. 100 sats just came in.
• 1:22:26
  So I don't know who it's from, but it says,
• 1:22:28
  good show today. I see it, yeah. Martin Lindeskirk, 1701.
• 1:22:33
  There it is. He says... And he comes in for
• 1:22:36
  true fans, as he always does. Fascinating conversation in cyberspace.
• 1:22:40
  Adam, Dave, Sam, go podcasting. More power to Moe, CTO,
• 1:22:43
  KYD. Fascinating. And he also sent another 1111 with a
• 1:22:48
  cheers to Sam's Wine Business. Indeed. Dreb Scott, 1234, coming
• 1:22:53
  in from Castomatic. I'm here. I'm here. Stuck in our
• 1:22:56
  on-prem data center tracking down fiber or SFP issue for
• 1:23:00
  some new servers. Yikes. Yeah. And I hit the delimiter.
• 1:23:04
  So, Dave, you are up with your spiffy news. new
• 1:23:07
  poll? Oh, Randall Black, $5. These are PayPal, so we
• 1:23:12
  got $5 from him. We got John Hind, $12.79. I
• 1:23:17
  think John had an... He had a message. Why did
• 1:23:21
  it not come through? It was Eric PP who sent
• 1:23:23
  the lightning invoice from Boost Me Bitch. So, okay. Thank
• 1:23:28
  you, Eric PP. Oh, thank you. Man, I cannot find
• 1:23:31
  his note. I'll have to find it after the fact.
• 1:23:34
  James Sullivan, $10. Thank you. Thank you, James. Cohen Glotzbach,
• 1:23:38
  $5. Christopher Reamer, $10. Uh, Jorge Hernandez, $5. Michael Goggin,
• 1:23:44
  $5. And the boys and girls at Buzzsprout, $1,000. Yo!
• 1:23:49
  Shot collar, 20-inch blades on the Impala. They are a
• 1:23:54
  great bunch of guys. They support a lot in podcasting.
• 1:23:58
  Thank you so much. We really appreciate it. Let's see.
• 1:24:02
  I think I got John's note here. Yeah, you. Oh,
• 1:24:07
  yeah. He says you accompany me while whilst marshalling the
• 1:24:11
  park run on Saturday mornings, nine o'clock UK time. Yes,
• 1:24:16
  I don't know what the park run is, but it's
• 1:24:18
  a run in the park morning. Yeah, it's a run
• 1:24:20
  in the park. Let's see, Dreb Scott, 716 sets. He
• 1:24:25
  says, bat signal received. Sorry for all the truck talk
• 1:24:29
  last week. I was successful in getting the alternator replaced
• 1:24:33
  last Friday. Very good. Anonymous 578. We got a few
• 1:24:39
  anonymous of various amounts. 920, 189. Quite a few of
• 1:24:43
  those. We got a fountain user. Another anonymous. 6691. Thank
• 1:24:48
  you user 9401. Blah blah blah. Uh... Let's see, read
• 1:24:53
  BTC at Fountain1176. As building a community around a podcast
• 1:24:58
  is the best way to add value for everyone. Noster
• 1:25:01
  as an identity and messaging layer is a perfect fit
• 1:25:03
  for podcast communities. Find your community at onlyboosts.social. I get
• 1:25:09
  boost whenever there's a boost nearby. Yeah. Only Boosts. This
• 1:25:14
  sounds like an OnlyFans competition. OnlyBoost.social. Interesting. I have to
• 1:25:20
  tell you, most people in my village think I run
• 1:25:22
  OnlyFans. Oh, you run everything else. Why would you not?
• 1:25:25
  There's that guy. He's loaded. He runs OnlyFans. That guy.
• 1:25:28
  Yeah. As you cruise by in the Ferrari. He's like,
• 1:25:32
  what does he – what does Sam even do? Does
• 1:25:34
  anybody know? Love down Babylon at Fountain. One, two, three,
• 1:25:40
  four. Is there something self-hostable as a value split destination
• 1:25:44
  in my podcast feed that could automatically split again to
• 1:25:47
  multiple wallets? Like a wallet switch where a band subsplits
• 1:25:52
  what others split to them but for a changing list
• 1:25:54
  of charities. Thanks. I think that's a combination of BoostMeBitch.com
• 1:26:01
  and the TARD box. That's the thing. That is a
• 1:26:07
  sentence that I never expected to hear in my life.
• 1:26:11
  It is a combination of Boost Me Bitch and Tart
• 1:26:14
  Box. You know, Chad F., he's vibe coding a lot
• 1:26:18
  of cool stuff, but the commercial value of his stuff
• 1:26:22
  – You know, the marketing is a little askew. I
• 1:26:25
  think it's hilarious. I love it. But I can just
• 1:26:28
  see, you know, hey, Pastor Jimmy, I just want everyone
• 1:26:31
  to go to boostmebitch.com. That's not going to work. That's
• 1:26:34
  going to be a bit of an issue. We need
• 1:26:36
  some different marketing there, Chad F. That's where you say
• 1:26:39
  boost me, buddy. Boost me, buddy. Boost me, buddy. Yeah.
• 1:26:43
  uh anonymous fountain user 2007 sats thank you anonymous user
• 1:26:48
  uh we got comic strip blogger 10 000 sats there
• 1:26:51
  he is and uh yes 10 000 i thought it
• 1:26:53
  was oh that's interesting your numbers must be off because
• 1:26:57
  i i see that's the part that went to us
• 1:27:00
  right but the total number is 18. Mm-hmm. We got
• 1:27:04
  10, though, to the podcast index. Right, but don't you
• 1:27:08
  want to give them the – Well, I don't see
• 1:27:11
  it. It's not in my export. Okay, all right. You
• 1:27:17
  tell me what the – what is the total? 18,000
• 1:27:20
  zats. Okay, we get, okay, 18,000 sets through Comicster Blogger.
• 1:27:23
  And I say this because otherwise you know that he'll
• 1:27:26
  never give up on us. I gave you 18. By
• 1:27:30
  the way, what I really loved this past week and
• 1:27:33
  a half, Comicster, something was up with Comicster Blogger, and
• 1:27:37
  he posted something. And it was beautiful to see our
• 1:27:40
  community say, hey, man, are you okay? Is everything? I
• 1:27:43
  texted him. I said, you know, everything okay? He's like,
• 1:27:47
  no. Like, well, let me know what I can do
• 1:27:49
  to help. But I just love seeing everybody who –
• 1:27:53
  everyone's had an issue with comics or blogger at some
• 1:27:55
  point. But it was really nice to see. Everyone's like,
• 1:27:59
  hey, bro, you okay? What's going on? That was just
• 1:28:01
  beautiful to see. And comics or blogger, if there's anything
• 1:28:04
  we can do, let us know. Howdy, Damon, Adam. Today,
• 1:28:08
  I recommend not a podcast, but YouTube channel. YouTube.com slash
• 1:28:14
  at puppet, E-D-G-A-R. Puppet Edgar. Puppet Edgar. So, Puppet Edgar,
• 1:28:20
  who is a blue puppet akin to Triumph Insult Comic
• 1:28:24
  Dog, but is operated by no agenda fan. Yo, CSB,
• 1:28:28
  still stonewalled by Adam Curry. This gram was $15.10 when
• 1:28:33
  sent. I wonder if it's my stonewalling that is making
• 1:28:37
  him feel so down. I hope not. You're going to
• 1:28:40
  have to solve this problem. Yeah. It's now firmly in
• 1:28:45
  your camp. Yeah. Okay. I'm going to solve this. I've
• 1:28:48
  got to solve this. My life is a mess. It's
• 1:28:51
  all right. Yep. Yeah. Okay. Was that it? That's it.
• 1:28:56
  That's it. That's it. There's nothing else. There's nothing. It's
• 1:29:00
  the end of the road. We've reached the end. Sam,
• 1:29:04
  is there anything else we can do for you? Can
• 1:29:08
  we help besides the things we already discussed? Anything else
• 1:29:11
  you want to close out with? No, no. I mean,
• 1:29:14
  look, you know, it's cool. I think... We've, what, what,
• 1:29:18
  is it five years, six years now we've been on
• 1:29:19
  this journey? Yeah, six. Yeah, I mean, you know, and
• 1:29:22
  we are moving the needle forward now. It's cool. I
• 1:29:26
  love it. That's why. The needle just does not, it
• 1:29:28
  just does not move very fast. It's a needle. No,
• 1:29:31
  but if you look back. Over the time, it's really
• 1:29:35
  funny, the arrow of time moves forward slowly, but when
• 1:29:38
  you look back, it's moved really quickly. You know... You
• 1:29:42
  see things, you go, "Wow, we came up with that
• 1:29:45
  crazy idea," or let's say Ukraine came up with a
• 1:29:48
  crazy idea, and it's implemented, and then suddenly it's adopted
• 1:29:52
  by the big players and suddenly it's everywhere. Transcripts, chapters,
• 1:29:55
  person tag. It's great. And I think the thing that's
• 1:29:58
  often missed is that it is
• 1:30:00
  The board meeting that keeps it all together. You know,
• 1:30:04
  every open source project should be doing some kind of
• 1:30:07
  podcast, some kind of board meeting. The idea of doing
• 1:30:10
  it live with a boardroom and people chiming in and
• 1:30:13
  we have fun. You know, we're goofing off this. You
• 1:30:16
  know, no one's a no one's the police officer of
• 1:30:18
  anything. Yeah. It is such a beautiful example of what
• 1:30:22
  podcasting can do. And Tina asked me the other day,
• 1:30:25
  she says, so why do you do this podcasting 2.0
• 1:30:28
  podcast? So, well, one – Otherwise, I'd never catch up
• 1:30:32
  with Dave. I just want to talk to my buddy
• 1:30:34
  Dave. I'd be lonely on a Friday night as well.
• 1:30:37
  Yeah, we don't want Sam to be lonely. We want
• 1:30:39
  a commentary blogger to have something to do. But it's
• 1:30:43
  also, it's in our DNA. It's in our DNA. It's
• 1:30:46
  in our DNA. It's who we are. It's a part
• 1:30:48
  of who we are. And I just love that this
• 1:30:50
  community has grown and that people continue to contribute. And
• 1:30:54
  it's a beautiful thing. It's a very beautiful thing. Sam,
• 1:30:57
  thank you so much, brother. Dave, thank you. Everybody have
• 1:31:01
  a great weekend. And we will return next Friday for
• 1:31:04
  another Board Beating of Podcasting 2.0. I hope you will
• 1:31:08
  join us. Somebody bring the donuts. We'll see you then.
• 1:31:11
  We'll see you then. Thank you. Thank you. You have
• 1:31:28
  been listening to Podcasting 2.0. Visit podcastindex.org for more information.
• 1:31:36
  Go podcasting! It's an impersonation and escalation.