Episode 239: Fuzzing

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Podcasting 2.0 for October 24th, 2025, episode
239.
We're fuzzing.
Hey, everybody, welcome back to the boardroom.
This is where it all happens.
The magic takes place right here.
It is time for podcasting 2.0, everything
going on in podcasting.
And there's a lot going on, as always.

(00:21):
We are, in fact, the only boardroom that
does not have an adoption process.
I'm Adam Curry here in the heart of
the Texas Hill Country, and in Alabama, the
man who will not backfill your stats or
mine.
Say hello to my friend on the other
end, the one, the only, the illustrious Mr.
Dave Jones.
I read Justin Jackson's adoption process.

(00:44):
I do too.
Lots of boxes with connectors.
Justin, I love you, man, but that's a
lot of stuff.
It felt a lot like a corporate process.
Yeah.
I mean, is that how it felt to
you?
Maybe it'll work.
Yeah, it did.
It felt like a PowerPoint at an offsite.

(01:11):
We have an away this weekend, and Justin's
bringing the PowerPoint.
Well, there was a lot of talk about
it on Power.
I don't know if you've listened to Power
podcast weekly review.
It was interesting.
Sam is very down.
I think he's very down on everything.
Oh, man, come on.
Sam, I'm sorry, brother.

(01:31):
Somehow he feels like adoption of everything is
key.
Well, not everybody's adopting.
These things take so much time that's just
waiting.
And I think no matter where I look
at the super secret WhatsApp group or the
Slack group or wherever I look, I've been
added to all these.

(01:53):
It's all still, well, is Apple doing it?
That is the same thing we've been hearing
for over 20 years.
And if Apple doesn't adopt it, then it
might as well be a tree in the
forest that fell.
And I just fundamentally disagree.
And also there's a little bit and it's
cultural.

(02:14):
I think the Brits and even though James
is in Australia and I don't know if
he considers himself to be a Brit, well,
it's a penal colony.
I'll get there.
It's like the complaining about the government.

(02:36):
And we come from a different culture.
It's like we, the government is always by
the people of the people for the people.
And there's like, well, no one's funding the
PSP to evangelize.
It doesn't work that way.
It'll never work that way.
And in fact, I think there's a gross
misunderstanding about what we're doing.

(03:01):
We set up the podcast index to allow
for non-censorship.
It's right there on the tin, as they
say.
We do not believe that Apple should be
the ruler of the universe as to what
is in podcasting, what isn't in podcasting.
That's how it's, that's literally the reason why

(03:22):
it started because they started de-platforming people
during an era of de-platforming.
And at the same time, we created a
namespace with the pod love chapters because it
was there and the value tag, because it
was a thought that at the time made

(03:44):
a lot of sense.
I think it still makes a lot of
sense.
And people came in and we set up
podcastindex.social as a way to, you know,
discuss things.
But I need to push back from time
to time, like Adam Curry and Dave Jones
aren't here to do anything.
We're not the government.
We're not your daddy.

(04:04):
You know, we're not here to do anything.
No, we're not.
Well, we do this podcast.
I would say the main reason is to
ensure that people are reminded that the podcast
index is a value for value project and
it needs funding.
And we accept that funding through booster grams
and all those little bits and all the,

(04:26):
all the 1% splits that helps.
It really helps over time.
Certainly we've never had to dip into the
node for any cash for the system.
And of course we have a big red
donate button at the bottom of the page
and that's for PayPal.
And I mean, lots of people have supported
this, continue to support it with monthlies, with

(04:48):
ongoing small bits and pieces.
Some of the hosting companies come in big
once a month.
And that literally, I mean, we're, I look
at our cash balance and, you know, we're
about the same.
We kind of roll on.
We did very well in the beginning and
now we're just about self-sustaining.
You know, it doesn't really go up.
It doesn't really go down.

(05:09):
It just stays there.
And we have enough for a couple of
years should all funding stop.
But that was really the idea.
And somehow when I hear like, well, you
know, what, what does Dave Jones determine?
I'm like, hold on a second, you know,
and well, you know, the, nothing has happened
in phase eight since November, 2024.

(05:30):
Hold on a sec.
There's been a lot of things that have
gotten into the namespace.
It's been, I would say by general consensus.
And, uh, I, I liked the idea of
the podcast standards project.
It doesn't really seem to be functioning.
I sat in the last meeting, one of

(05:50):
the last meetings, I liked it.
And, you know, people were talking about things
that were important to them, which of course
had nothing to do with the namespace.
It was all about HLS video, at least
the session for the time that I was
there.
And, you know, what do you want?
I mean, do you, it's like, it's a
fail, you know, it's deemed a failure because

(06:12):
these tags have gone nowhere.
Well, I disagree.
I think these tags are very valuable and
people can use them for other projects.
If anything, the mindset of podcasting development, both
on the app and the hosting side is
very stuck in a one track mind of
this is what we do, transmitter receiver.

(06:34):
And I, I fundamentally don't think that's the
future of anything really.
I mean, you know, it's, there's, there's ways
to expand and look at things.
We did this with Godcast.
We've created something completely different based on RSS
and the index.
And we've done something completely different.
And just as a quick aside, the pub,

(06:55):
because we were essentially accused of being horrible
people.
I'm exaggerating because we didn't use the publisher
feed in Godcaster.
I just need to read you from the
publisher tag in the namespace.
This element allows a podcast feed to link
to its publisher feed parent.

(07:16):
This is useful when a parent publishing entity
wants to attest ownership over all the podcast
feeds it owns.
These feeds are not owned by the people
putting the feeds together.
Stations don't own those podcasts.
Those are not the publishers.
So it was kind of like, well, you're
not eating your own dog food.
I'm like, simmer down, Sparky.

(07:37):
It's not, it's not, that's not the use.
That's not the correct use case for that.
No, we, we, we, and we do, we
actually do have a plan to use the
publisher feed because we recently, which is not
even released yet, have a publisher feature and
we're a, and we're the actual publisher, not

(08:00):
the station, but the actual publisher.
So somebody like Focus on the Family or
whoever can get in and, and we can
help, we can set a particular set of
podcasts to be owned by them.
And then we will produce a publisher feed
exactly on their behalf.
But that's still not, they have it.

(08:23):
That's just a convenience for them.
If they want to use that, they, they
should be publishing that themselves.
Why would we publish using the example of
Focus on the Family?
They have like, I don't know, 10 or
so podcasts out there in the world.
Why would Godcaster publish a publisher feed for

(08:44):
them?
Exactly.
That's their, that, that, who are we?
Right.
They should be doing that on the, with
their own system.
Exactly.
Exactly.
So it, it was just, it was, and
I understand, you know, but actually I don't
understand what people just want to listen to

(09:08):
podcasts.
And if you tell them there's a cool
feature in a certain app you can use
with your podcast feed, it turns out people
probably use that app for that very reason.
I do this all the time.
I say, use a modern podcast app, podcastapps
.com for two, well, three reasons.
One, you get the bat signal when we

(09:30):
go live and you can listen to our
show as we record it live in your
podcast app.
Hello, mind-blowing legacy apps.
Two, when we publish the show within 90
seconds, thanks to the pod ping technology, you
will be alerted.
So stop waiting around on your legacy apps.
And three, you get chapters in chapter art,

(09:50):
which are fun and entertaining.
And then of course there's the transcript and
other things, but that's what I highlight and
people use it for that reason.
You know, and do I expect the whole
world to listen to No Agenda on modern
podcast apps?
No, it's actually, it's quite a lot when
you look at the OP3 stats, but a

(10:10):
lot of people listen on the website.
To me, for baffling reasons, I would say
there's almost 15% of people listen to
No Agenda, three and a half hour show
at noagendashow.net.
Now there's 2.0 features that show up

(10:30):
there like chapters and transcript and I'm not
sure what else.
And that flows through nicely, but you know,
it's like, what do you want?
What are you expecting?
I mean, what is success?
To me, the fact that the index exists
and anybody can create any crazy idea they

(10:52):
have for podcasting is the success.
Open source projects are only successful and let's
define successful for
a

(11:17):
change.
And with that definition, there's a better word,
sticking around long enough to see meaningful impact.
So if your open source projects have success

(11:39):
in that way, only in so far as
they stay around and staying around is 90
% of the success.
That's an excellent point.
Because look at the carnage, look at the

(12:03):
litter of open source projects that are just
scattered everywhere that just don't exist.
They're dead.
They're dead projects.
They won't ever return.
And most of that is because of burnout.
People don't, most, I'd say like almost all

(12:28):
open source projects do not get funding.
They are projects that are built out of
love and passion for a particular thing.
And passion wanes over time.
If you're not getting paid for it, then
it's purely a passion project that fades.
It does.

(12:49):
And so that's why sticking around is so
important.
Because if you're not there, then the project
has no chance of making impact.
So my focus has always been first and
foremost on just sticking around.
I'm with you.
I'm with you on that.

(13:10):
I mean, look at, look at how many
hosting companies have added podcast index to their
lineup of places that they publish.
That to me, I look at that and
go, yeah, win right there.
That's the decentralization we were always looking for.
Right.
And, and the, the sticking around means, sticking

(13:40):
around looks like life.
You know, life, people's actual lives are messy.
And you go through phases in your life
where you have, where things are, where things
are busy, things take your attention away, that
kind of thing.

(14:00):
And so you have to expect that there's
ebbs and flows in the activity of an
open source project.
And that's, that's fine.
That's to be expected.
People become, there's a flurry of activity around
the images tag and then things settle down
and go quiet for a while.
Right.
And actually, if you look at what James
did with the location tag, I think that's

(14:21):
how you drive something.
He did a great job on that.
For sure.
And then there's, so there's, there's these bursts
of activity.
And then as there's bursts of activity, as
recognition begins to build around the necessity or
the attractiveness of a certain set of features.

(14:42):
And then a lot of people pour their
intensity into that for a period of time
and get it across the finish line.
And then there's a pullback, there's a rest
and that's fine because you can't, the quickest
way to burn out an open source project
is to expect consistent output on an unrealistic

(15:06):
basis.
Amen.
And you know, in a funny way, the,
the analog of government kind of fits here
because the podcast index is the true federal
government.
It's only job is to protect your right,
your right to podcast.
That's it.
And then over time, just like the United

(15:28):
States, but this of course is not a
United States project.
We have all these States and all these
States have different things they want and don't
want.
And you can see those as hosting companies
as app companies and, and, and groups like
value for value, the, the music group, there's

(15:50):
all these different groups who have pieces they
want and pieces they want to use for
different reasons.
If we really think that all podcast apps
are going to have this, all this set
of features and all hosting companies are going
to have some standardized set of features.
I say, dream on.
I don't think that's ever going to happen.
And maybe it shouldn't.

(16:13):
Well, Buzzsprout, Buzzsprout is a huge, they've, there's,
they're a huge supporter of, of, of us
financially.
They keep us in the black when we
had a couple of big drop-offs and
they teach every month.
And they're also more than just financially supportive.

(16:36):
I hung out with Tom Rossi a couple
of weeks ago here in Birmingham.
He was up here and we had, you
know, we had a beer and are you
an advisor?
Are you an advisor?
You need to disclose if you're an advisor.
I have nothing to disclose anywhere because I'm
not any of that.
But you know, we hung out and talked
for a few hours.
We did not even mention podcasting.

(16:57):
I think maybe we mentioned it once during
the whole evening of spending, we talked, we
talked about family and school and kids and
and Christianity, sports ball.
No, we did not talk about sports ball.
We just talked about all kinds of stuff.

(17:17):
So they're more than just a financial supporter.
They're also a, you know, just a, a,
a motivator and a, and a, just a,
a people that help just like, just like
RSS.com and, and, and, and Todd and
them, they, they, they pour more than just
money into these things.
They also pour themselves into it.

(17:38):
But the example of Buzzsprout, I brought that
up for a reason because they have not
adopted a ton of podcasting 2.0 tags.
They've got a few, you know, they've got,
they've got a handful, but they are going
by what they're going by a timetable that
makes sense to their business.

(18:01):
And when, when they go and make their
next plan for their next feature build or
whatever that is, everything's available to them.
The whole, you know, the whole arsenal of
next generation podcasting features is available.
You know, now they, I think they have

(18:22):
transcripts, they have a pod role, they support
the GUID tag, they support chapters, they support
a few of these things, but they do
not support all 20 something tags that are,
that are developed.
But every time they do a new cycle,
they have the chance to, and if they
determine it makes sense for them at that
moment, they will pursue it.
And that, and I was thinking about Apple

(18:43):
podcast the other day, took them a long
time to adopt their first 2.0 tag,
which was a transcript tag.
It now looks like they're going to probably
adopt the chapters tag as well.
Well, they, they also support the TXT record,
but they, so they support two, it looks,
it looks like they're going to support the

(19:04):
chapters tag.
And I think the way that my impression,
and I, you know, Ted may laugh when
he hears me discuss, talk about this, but
my impression is that you have to have
transcript first because I feel like what they're
going to do with chapters is they are

(19:25):
going to do it similar to the way
they did transcript.
They're going to auto-generate chapters for every
podcast based on the transcript.
But then if you bring your own chapter
URL, they're going to support, they're going to
let you override it.
That's the same way they did the transcript
tag.
They're going to generate a transcript, but if
you bring your own, they'll use that instead.

(19:46):
As long as they have the override automatic.
I don't know if they do that for
transcripts yet, because you had to go into
Apple connect and tell them explicitly.
I don't know if that's still the case
or not.
I think, yeah, I don't know, but I
feel like that's what they're going to do

(20:07):
here.
So if that's the case, and there's a
reason I'm kind of laying it all out
like this, if that's the case and they
do, and they auto-generate chapters, you know,
using whatever tools, then you can sort of
reverse engineer and say, okay, they needed transcripts
first because they'll base the chapters on the

(20:28):
transcript.
They need the transcript in order to generate
the chapters.
So there's your development cycles progressing there.
You have to build this before you can
build this before you can build this.
And that's just the way all of this
works.
So that's why when you've come up with
a new feature, it can take 10 years

(20:51):
to actually get into production.
You know, it, it doesn't, I think, so
like, like we can do things, I can
do things in the index fast because it's
just me.
And James can do things in his stuff

(21:13):
fast because it's just him.
We're, we're single or small teams and we
can move quick and make all kinds of
changes to our systems.
But the bigger they are, these bigger institutions,
they just, it just takes time because they
have to, they have to develop based on
a strategy.
And the strategy is just, they take multiple

(21:34):
layers in order to come to, you know,
something like a stable usable product that can
go out to, you know, millions of people
potentially.
And I still remain that it doesn't all
have to be uniform.
Fountain is what Fountain is because Fountain serves
a very specific group currently.

(21:57):
And they're great at serving that group.
And that group uses Fountain.
It's like, that's perfect.
There's no reason why every single app should
have to support Nostr and Bitcoin value for
value.
And there's no reason for it.
In my mind, it's like, that's, that's their
USP.
And it's fantastic.
You know, Podverse started initially as a, as

(22:21):
a clipping tool.
That was the focus, clipping.
And I think accessibility was big for them
as well, which is highly appreciated with our,
with our gunky eyed blind mother, brothers and
sisters.
Yeah, that's right.
They really appreciate it.
You know, so, and Overcast, which has, let
me count them, zero podcasting 2.0 features

(22:44):
is adored by its user group.
I think mainly for its sound processing, sound
processing, yeah.
Sound processing and just the, and the UI
or UX or whatever it is.
Yeah.
And it's developer people like him.
And sometimes you, sometimes you use a product
just because you want to support the person
that you have some sort of parasocial relationship

(23:05):
with.
And that's fine too.
Exactly.
And you know, I, the stats are broken
and you know, they're the, we produce stats.
I produced two stats files.
I say produce, they haven't been actually producing
for weeks now.

(23:28):
So there's, what happened is there, let me,
let me just talk about that.
Cause that that's, this is going to tie
into this whole thing.
Stats were stopped working, I think sometime maybe
in September.
So these two, these two JSON files that,
that I produce have a lot of baseline

(23:54):
stats about things in the index.
So some are just the high level, how
many shows have published an episode in the
last three days, seven days, 10 days, blah,
blah, others are other stats that it, that
it puts or how many feeds have soundbites,
how many soundbites are there, how many chapters,

(24:16):
how many feeds have chapters, how many episodes
have chapters on and on and on all
the, trying to break down a lot of
useful information about what's in the index without
having to go and query the database.
So the stats broke, James told me, he
said, Hey, you know, heads up, these, the
stats have not updated in a long, the
stats files haven't updated since sometime in September.

(24:39):
So I was like, okay, well, um, I
went and looked at it.
I ran the stats file generation script by
hand and it got to the transcripts tag
and just stuck.
Barfed.
It's, it just hung for half an hour.

(24:59):
Bimro says breaking stats.
You know, you need the little, uh, you
need the little, the red alarm lights, flashing
lights.
Yes.
Breaking stats.
Yes, exactly.
That's every other, that's every third post on
Twitter breaking with the real life.
Exactly.
Um, so I ran and I'm like, okay,

(25:23):
well this is just, it ran for like
30 minutes and I'm finally, I just control
C.
I'm like, okay, this thing is frozen.
Well, it turns out that the explosion of
transcripts in the recent months broke stats.
Ha!
Killed by our own success.
That's right.
There you go.

(25:44):
Proof.
So, and, and it broke, here's how, here,
here is the way that it broke.
The stats, the transcripts, the, the NF items
transcript table has currently about seven and a
half million rows in it, which is a

(26:06):
lot.
That's as many rows as the entire podcasts
table.
Wow.
And the reason is because some episodes produce
multiple transcripts, sometimes as many as five.
Is that multiple languages or what is that?
Possibly and some, and they're multiple formats.

(26:27):
So you'll have VTT, SRT, TXT, HTML.
You'll have, you'll have sometimes five different formats
in the same episode.
So, you know, if you have a million
episodes with podcasts and they all produce at
least three formats, well then boom, 3 million
rows.
So there's seven, there's over seven, there's like

(26:49):
7.3 million rows in the transcripts table.
And that happened quick.
It went, I mean, it went way up
really fast.
Cool.
And so the way that my SQL database
has what I consider to, it has many

(27:13):
flaws, but one of the most annoying is
it's seeming inability to count accurately and quickly
in a table.
It's like an LLM.
Exactly.
Yeah.
How many R's are in the word strawberry?
So, um, yes, Eric, the transcript table just

(27:38):
tracks the number, you know, like what's in
the transcript tag as it comes in.
Yeah.
Uh, the, the URLs to the transcript file.
So going, you know, uh, that's what happened
is trying.
So the way that the, that that table
is structured and, and yeah, it, it has

(27:58):
an index on it, Eric, but let me,
I'll tell you what the issue is here.
The issue is that this is, there's a
trade-off between, between like elegance of design
and real world needs.

(28:19):
The, the, the elegance part of your, of
the development cycle is you don't want to
duplicate data in multiple tables.
So you have a newsfeeds table in this,

(28:40):
you know, I'll meet, let me speak in
regular language and not freedom controller language.
You have a podcast table, then you have
an episodes table, then you have a transcripts
table.
Okay.
I think we all can understand that layout.
The podcast, the, the, the podcast table links
to the episode table through a column called

(29:02):
feed ID.
And so you've got a link back to
this, uh, this episode belongs to this podcast,
the transcript table links to the episodes table
on an episode ID.
So now you have this transcript goes with
this episode.

(29:22):
Well, you can walk back, right?
You can walk back from the train.
If you want to know what feed that
transcript came from, you can walk back up
the chain and say, this transcript came from
this episode, which came from this podcast.
Well, that's great.
If all you're doing is that sort of

(29:43):
work, if you want to actually count how
many feeds, how many podcasts are using the
transcript tag.
Oh, you can't walk at all backwards.
That would take forever.
But exactly.
Wow.
I'm a sequel developer.
And so you, what you could do this,

(30:07):
what I was doing before is I was
just grabbing every row out of the transcripts
table.
And then walking it back and finding out
what podcast it belonged to.
Right.
With a join.
Yes.
With a join.
And then going through and sorting it all
out.
Well, as soon as it gets over a
particular size, then MySQL begins to have to

(30:31):
create a temporary table.
Because what you're counting is the number of
podcasts that use transcripts, not the number of
transcripts.
Right.
Okay.
Got it.
And so you can easily solve the how
many episodes use transcripts by doing what's called
a select distinct.
So you can say select, select count distinct

(30:54):
ID in the transcripts table.
And that will give you on the distinct
on the episode ID row.
And that will give you how many episodes
quickly, it will quickly give you how many
episodes use transcripts.
You, there's no, as soon as you do
a join, you immediately have to create a

(31:16):
temporary table on disk and you're toast.
Because the count day you're, I mean, it
could take you 45 minutes to complete that
query.
And that is just, it's just too, it's
too costly.
And so there's, there's, you have this elegant
design where you're not duplicating data across tables,
but that elegance of design harms you because

(31:40):
now it, you can't actually in the real
world get the data out that you want.
Can I ask an Adam Curry question?
Sure.
Can I fix this with money?
It's not worth fixing with money.
Okay.
You could technically, you could technically scale way

(32:03):
up and pay ridiculous amounts of money to
get an insane piece of horsepower VM that
would make this go faster.
But it does, it would, it would eventually
drop as well with scale.
You just can't, it's not, that's not the
right solution.
The right solution is to, as a developer,

(32:24):
get over yourself and say, it's not, elegance
is not the, is not the most important
thing here.
What's the most important is to get something
that's usable.
So you get something that works well and
fast for all of the use cases that

(32:44):
you need it to be.
And what that means is I'm going to
have to put another, another column on the
transcripts table that's going to hold, that will
hold the podcast ID, the feed ID.
Now that it, that's clearly a duplication of
data.
Now I've got the pod, now I've got
the podcast ID in two places, which, you

(33:07):
know, it doesn't make any, it shouldn't have
to be that way, but it is.
And I've been trying to find, you don't
do changes like this on a production database
without planning.
And so I've been, I've been trying every
SQL trick I can think of to make,

(33:32):
to avoid having to do this.
And I can't, it's just not going to
work.
So now in the last couple of days,
I've decided I'm going to have to make
this change.
I'm going to have to make this schema
change on the database.
And which is going to be a, it's
going to be a three-step process.
It's going to be number one, add a

(33:53):
column to the transcripts table.
Now remember, the key to all this is
the transcript table has seven and a half
million rows.
So it's not small.
Now it's, it's not gargantuan.
It's not like the episodes table, which has
170 million rows.
But it is, but it's big enough to

(34:13):
where it is a problem.
If you lock, if you lock the table,
because if you lock the table, everything stops.
No API requests, no aggregators, everything halts for
the duration of that table lock.
Because there's relation, there's foreign key constraints across

(34:37):
these tables.
And when you lock, when you lock a
table that has a foreign key constraint, it
locks it all the way up the chain.
So there's going to be, you have to
do it, do it in three steps.
Step one is going to be adding a
column using the instant algorithm with my SQL.
I've done this before with other tables.
It's, I know how to do it, but
it does take planning.

(34:59):
Um, you add it with instant algorithm and
with no foreign key constraints initially, and then
you add a nullable column.
And so everything comes in there, null, stop
after that test, make sure nothing is broken.
Once you do that, uh, you go in
and you, I'll write a script that will

(35:20):
backfill all the rows in that transcripts table
with the appropriate feed IDs.
Then step three, add the foreign key constraint
to connect the, uh, to relate the feed
ID column on the transcripts table to the

(35:41):
podcasts table.
Um, so you, you, you do it in
that order because you'll have, because if you
try to, if you try to, I have
not had very much success trying to add
it all at once, even though it should
technically work.
I don't, I've not, I have seen my
SQL eight fall back to an in-place

(36:06):
locking table change, even when you give it
the instant algorithm.
So if you try to do it for
a key constraint that it doesn't, that is
not simplistic.
Um, so anyway, that's, that's what stats are
broken because of that.
Wow.
So that's, that's great as a part of
the conversation.

(36:26):
As I said, it's broken because of the
success of transcripts.
This is kind of good news.
You know, I don't know if you remember,
we talked about it at the time, but
there was a, uh, Lex Friedman did an
interview with John Carmack and you know, John
Carmack, right?

(36:47):
The developer, they created Doom and Quake.
Oh yes.
Okay.
Sure.
Yeah.
And he, and he said in that, uh,
uh, yes, Eric, there is a lot of
exceptions to the instant algorithm.
Yeah.
It's, it's very hit or miss, but, uh,
Carmack at the time, um, in that interview,
he said that he was a big fan

(37:09):
of hard limits on things.
So when you're writing, when he writes software,
he'll put in some limit.
Like if he thinks that this thing could
probably handle, um, a million, whatever's a million
transactions, uh, you know, I don't know, we're

(37:29):
making up stuff here.
Then he would put in a hard limit
of like 750,000 in the code, just
a number in the code.
And that way, when he hits that number,
the key knows that the code's going to
break, but he'll, it'll be, it's like a

(37:50):
way point to say, okay, I thought that
this could do, I thought this particular metric
was achievable and we've gotten there and it's,
how did it go?
Right.
Is it broke or did we never even
get there because it broke?
And I think that that's kind of similar

(38:11):
to what's going on here.
We have some, we have some metrics that
we're hitting that actually are important.
They, they mean, they mean things.
Um, they mean that, you know, the, that
the index doesn't just exist on its own.

(38:32):
That means that there's a person behind it
who's having to actually make changes when this
stuff gets popular.
And we try to, we're probably a pretty
good canary in the coal mine because our,
I mean, our directory is bigger than Apple's.
It's bigger than, I guess it's bigger than
everybody's.
It's huge.
It's huge.
Um, and so we're, you know, when things

(38:56):
break and when they break, it means that
we have to stop.
That's a good time to stop and say,
okay, well, is it, is everything working?
Okay.
Do we need to change things?
And that's the mental process I've been going
on going through for the last couple of
weeks is trying to sort of figure this
out.

(39:16):
Um, and these are the, this is what
I meant earlier by, you can't expect out
of an open source project, consistent, like unreasonable
levels of output.
Because when you only, you know, when you
have a loose affiliation of people that are
working on things, um, it takes time.

(39:41):
It takes time to, to, to reason through
these things and make, and not just willy
nilly push to production, you know, you have
to figure this stuff out.
And once there's, it's not just, it's not
just this as well.
I mean, we, I'm hoping to have, um,
uh, Archie on the show.
I need to reach back out to him,
uh, in email, uh, Archie helps, uh, Podverse

(40:04):
a lot with their database stuff and infrastructure.
And I want to have him on the
show to talk about open sourcing the index
even more and maybe redesigning our database.
Some who's Archie, uh, he's on a podcast
index dot social as SUCD.
Oh, okay.
Yes.
Yeah.
Um, yeah, it seems like a nice guy
and, uh, I want to get him on

(40:26):
the show.
He's, uh, to, he's been at, he's been
asking for a while.
He's like, Hey, can I help you all?
You know, I'd be glad to get in
and do some infrastructure work and be glad,
you know, glad to help.
And so, uh, I just want to see
if he wants to get on the show
and we, it'd be a good opportunity to
just talk out how can we make the

(40:47):
index be more, uh, managed by the community.
Um, because rather than trying to kind of
figure all this stuff out myself in isolation,
it would be helpful to talk this out
and in the, you know, in the podcasting
2.0 group and get pull requests and

(41:07):
that kind of thing for more than just
the website.
Excellent.
Yeah.
So I think it's, all of this wraps
up into, you know, the broader discussion of
just, you know, we're an open source project
and James is part of it.
Sam's part of it.
We're, we're all in this together and Justin

(41:28):
and I, and the PSP and everything.
We're all, we're all a part of this
and there's just going to be some, um,
there's going to be ebbs and flows of
activity.
And I, I don't, I don't want to
try to push it to be more than
that because I think if you try to
push the, put the pedal to the floor
and expect constant output, I think it that

(41:49):
people will just get tired and leave and,
and, and move on to other things.
And I just don't want that to happen.
I'm with you.
And I want to give big props to
Chad F who continues tirelessly to work on

(42:11):
wallet solutions.
And he's, I was blown away.
He built this Vercel app for V4V music
with the Breeze SDK.
And I'm really impressed by all of this.
And I'm actually quite impressed with the Breeze

(42:33):
SDK because they have a Vercel demo.
And I created a wallet within three seconds.
And in fact, I changed our, um, our
splits because people have been saying, Hey, you
guys want LN URL, but you're using keysend.

(42:53):
So I changed our splits and it's currently
going into a Breeze just on the web,
just a Breeze Vercel app, demo app wallet.
And it's working fine.
It's just, it's, it's working.
Just popped it up.
Chad, Chad's going to be on the show
next week.
Good, good.
Now, of course, there's no metadata coming through.

(43:15):
So I still have 1% going to
my helipad and actually I'll have to send
whatever comes in so far, a whopping 1
,342 sats.
Whatever comes in, I'm going to send to
Barry's wallet because somehow I made a copy

(43:36):
paste mistake when I released the last episode.
And it was, uh, I forgot the T
at the end of .NET. So he missed
out on some Instagrams.
So I'll send this to him.
But it's amazing how well this works.
And even the, and of course I still
have to talk to Graham over at Voltage

(43:57):
since we can't seem to set up, uh,
um, anything on top of our node, any
kind, anything that'll do LN URL.
I can't set it up because it just
breaks the minute, uh, the minute I activate
it due to our humongous girth, I guess,
or whatever we may be running on special.

(44:18):
I, I, I always had the feeling he's
got like five servers running just to serve
the podcast index.
I don't know.
You know, I don't, I don't know how
he, it's always a special case.
He's always taken very, very, very good care
of us.
He has, yes.
Downtime zero.
As far as I can tell, we never
had down except when we had to reboot
the node for an update downtime, seven hours,

(44:39):
but okay.
Yeah.
All night.
Minor, minor thing, minor thing.
Um, but it is, it's beautiful to see
because you, you see that, you know, this
type of solution, which can easily be baked
into any app.
Yeah.
You don't have your onboarding.
You don't have your buy sats and all
this stuff, but they're already doing liquid, which

(45:02):
is, I don't even know.
I don't even understand liquid very well, but
it's not really even a layer two solution
at some mystical other thing.
Side chain.
I thought it was a side chain between
two that I thought it was a side
chain that, that exchanges used between themselves.
Yeah.
But it's not like a settlement on chain

(45:23):
is I'm not sure what it is, but
I'm sure that breeze will be doing any
other type of system that comes along.
I just have that general feeling.
Did you, did you see that arc launched?
What's arc?
That was the, that's the competitor to lightning.
That's the other layer two.
No, I didn't.
I didn't.
I didn't.
What, what, what are their, what are their
benefits?

(45:43):
What do they have in arc?
Let's see.
Arc layer two, the big, I think the,
the arc protocol, I think the biggest benefit
with it was, um, off chain receive.
Oh, I mean, I'm sorry.
Not off chain, uh, offline receive.
Offline receive.
Yeah.
Oh, very cool.
Um, I wonder how it does that.

(46:04):
Yeah.
Arc layer arc is layer two protocol for
making off chain Bitcoin transactions.
Uh, we talked about it at the time
and I remember it actually, it made a
lot of sense when we went over it.
Uh, the arc, see arc protocols built around
shared UTXOs while lightning also uses shared UTXOs.

(46:27):
Okay.
I got it.
They're only shared between two channel catapult and
arc UTXOs are shared among a large number
of users, potentially hundreds of thousands.
Hmm.
Oh, um, sounds a bit like a cashew
thing in a way, in an odd kind
of way.
Yeah.
I remember when we went over it on
the show, it makes like seeing how they

(46:47):
did it made a lot of sense.
I was actually pretty.
Oh, I remember this is where you could
eat.
You can basically email somebody.
I think so.
Yeah.
I think so.
Interesting.
See.
Yeah.
I don't, I'm sorry.
I don't remember much about it, but I
did see it launched.
I expected it to just be kind of
like a, a failed, you know, like an

(47:09):
idea that never You mean like, like podcasting
2.0 basically.
PSP.
You know, one more thing on that.
We, as an open source project that just
came out, you know, just kind of grew
up out of nowhere.
You can't argue that we have not been

(47:31):
successful.
Exactly.
I mean, wow.
I mean, we have a self-sustaining open
source project, self-sustaining by the users and
stakeholders.
I don't know if that's ever been done.
It, the, the growth, the growth first mindset,
just, I don't know.

(47:53):
It, it, it kills, it kills momentum.
Oh, this is a, this is the philosopher,
Dave.
You've been, you've been posting stuff about this.
The growth, the growth mindset.
I guess I did.
Yes, you did.
Yes, you did.
Yeah.
I'm reading Wendell Berry's book called Life is
a Miracle.
An essay on modern superstition.

(48:17):
And this is the book that you will
be getting as your Christmas present this year.
So don't buy it.
You ruined it.
I already have your Christmas present picked out,
but I'm not going to tell you what
it is.
Yes.
Because I received something.
I'm like, oh, I got to send this
to Dave.
Oh, nice.
Thank you, brother.
But that, this is, um, he goes off

(48:42):
on...
By the way, you're the only other person
in my life who I give a Christmas
gift to.
I mean, besides my, my kids and my
wife.
Oh, that's, I'm like, I'm in a rarefied
air.
Yes.
Yes.
It's very rare up here.
You just roasted my chestnuts.

(49:05):
Bring...
So the, what, one of the things that
he mentions in there is, you know, we
didn't see the earth for fully for the

(49:25):
first time when, uh, when, when astronauts went
to space and started taking the pictures of
the big blue orb hanging in the blackness,
we, we, we saw the earth, artists saw
the earth from the first, you know, centuries
before, um, through poetry and through, uh, philosophy.

(49:49):
He quotes Dante and we just, we, we
just forget that we have a, we have
sort of a, um, a physical, a very
physicalist mindset.
Physicalist?
Yeah.
You know, like, like, uh, if it's not,
if it's not a, um, if it's not

(50:10):
material, we don't, it's not, it's not real.
Okay.
Yeah.
And we forget that we've seen...
Science is just one part of the human
experience.
It's not every, it's not all of it.
And we, we make ourselves poorer, you know,

(50:32):
spiritually by taking one aspect of human life
and putting it over all else, whether that
be economics, you know, uh, philosophy, the, you
know, the sciences, the hard sciences.
We, we need as, as Westerners, we have

(50:55):
a, we, we only use one half of
our brain really most of the time.
Some people use much less, I found.
Some people only use one quadrant.
That's right.
That's right.
Um, you know, but we really have, um,
we really have a, a physicalist growth at

(51:15):
all costs mindset, uh, the, the capitalist urge
and, you know, capitalism is fine.
It's just the worst.
It's just the best of, it's the best
of all the worst systems.
Everything else, capitalism is fine.

(51:36):
It's just that everything else sucks worse.
It doesn't mean that it's actually perfect in
any way, shape, form, or fashion.
And the, the one cost that it comes
with is just this perpetual never ending growth
at all costs mindset.
And the, and it elevates the, you know,

(51:56):
the, it elevates the materialist, uh, outcome based
model above everything else.
And I just think that that has been,
and we apply it to everything.
We apply it to every aspect of our
life.
It's an urge within us in the West.
And we just can't like, it's an infection.
And I think that it comes out in

(52:17):
these kinds of ways too.
Yes.
The number must go up.
It can never go down.
Number go up.
That is just all we care about.
And you know what?
I mean, sometimes when the number goes up,
your database breaks.
Precisely.
You break shit.
I mean, like it's, it's not, it's sometimes
it's not, it's not fun to go for

(52:40):
the number to go up, but, but I
just think that we, sometimes you build something,
you some, sometimes you build something just because
it's beautiful or just because it's meaningful or
just because it's useful.
And if the rest of the world picks
it up, great.
If they don't, it's still great.

(53:03):
It's fine.
It's okay.
We, we did, we did that.
One of the funnest things we've ever done
in podcasting 2.0 is going, is going
to the Bitcoin National Conference a couple of
years ago and doing that live thing and,
and all that stuff.

(53:23):
I mean, in the grand scheme of things,
maybe there was a couple of thousand people
that participated like online and that kind of
thing.
Those are tiny numbers.
I mean, that's, that's just like.
But how fun was it?
But it was so fun.
And everybody got such a kick out of
it.
And it showed what was possible.

(53:43):
And that, that the, the, the showing of
what was possible will stick in somebody's mind
and some, at some point in the future,
fruit will be born from that.
In some other way, maybe in some other
way, not even connected to podcasting.
I don't know.
I'm with you.
I'm with you on that.

(54:04):
Absolutely.
I plan on being here doing this show
five years from now and, and whether we
got a single new tag at that point
or not, I don't care.
Well, we've, we've held onto that for, are
we in our sixth year already?
Is this our fifth or sixth year?
We started in 20, we're in the sixth

(54:26):
year now.
Let me see.
Let me see.
Our first 2020, uh, August.
August, June.
August 28th, 2020.
So we're, you know, yeah, we're in our
sixth year.
Wow.
Wow.
Think about all the stuff that's happened in
those six years.
And you know, all these things, it's like,

(54:50):
you know, so on Sunday, Dvorak and I
celebrate 18 years of no agenda.
Wow.
I mean, that's almost as long as my
first marriage.
Okay.
And he's been through all three of them.
And the third one is still going strong.
I'm happy to report.
But you know, the first three years and

(55:11):
we was all value for value.
Um, it was not a lot and, you
know, it took consistency and, well, that's it
really a consistency, just consistently being there every
single, initially Sunday, then, or yeah, Sunday, then
Thursday and Sunday.
And this was amidst the rise of Facebook,
Twitter.

(55:33):
YouTube, uh, all of this was happening right
in those first years and none of it
slowed us down.
It was a 5% per year gain.
It took a while and it took, it
just took just being there, just being there,
doing it every single time over and over
again.
And I remember the first time I was
in Texas that really, it wasn't until I

(55:55):
was in Texas that, uh, it was like,
okay, I can barely live off of this.
And I had a very expensive second wife.
That was, that sucked, but that's a problem.
Yeah.
That's I sold my plane.
I remember, I completely gobbled it up.
Um, but it all works.

(56:17):
It all can work, but yeah, it's not
like today's everything's free.
Number go up.
It's all going to be great.
I'm an influencer.
No, it's just not the way it goes.
And so, and luckily I'm seeing this whole
video conversation, except for Rob Greenlee.
I'm seeing the whole video conversation finally died
down.
You know, everyone's okay.

(56:38):
Boy, that was really scary.
Mainly due to Google PR and marketing.
They're good at it.
Well, they have, let's put it this way.
They have a lot of money and they
have complete, they're putting all they're all in
on AI all in.
By the way, I saw the numbers.
The revenue numbers got broken out somewhere for
YouTube.
I saw $8 billion was the YouTube revenue.

(57:00):
On an annual basis.
No, really?
That tells me that I think we're right.
I think they're losing money on it.
I don't see the more I think about
it, the more I don't see how they
can be making money.
Yeah.
Now, you know, of course they, they have
all these other, other pieces that are highly
profitable, but even that I, I predict before

(57:21):
we're done with this podcast, we'll see someone
else usurp Google for search because they are
literally killing their own business model.
In so many ways, one, where are you
going to stick your ads in, uh, in
your AI search results to, if you, if
you stop sending people to websites, there'll be

(57:41):
no one to buy ads from you.
It's just that simple.
It doesn't make any sense.
It really doesn't.
It really doesn't.
And I'm, I'm sure they're sitting around going,
what are we going to do it now?
Sure.
They're $40 billion a year.
I'm sure, but still, I think they're sitting
around going, what are we going to do?
How are we going to, how are we
going to manage this?
How are we going to make this happen?
Yeah, we're, they, from all, from all I

(58:04):
can tell, they get off with just basically
a slap on the wrist with all the
antitrust stuff.
Um, but that's not, but that doesn't matter
because, you know, if you, if you're, if
your whole product drives up, well then antitrust
is the least of your concerns.
By the way, Eric PP is just going
back to no agenda.
There were episodes.

(58:24):
We threatened to quit the show for low,
low donations.
Yes.
How about every episode?
Because it's value for value for value goes
both ways.
If I don't get enough, the value I
need, which is honestly not all that much
in return for the, and Hey, if people
stop supporting the index, guess what?
Dave and Adam will probably dig into our

(58:47):
own pockets to keep it running, but there
won't be much other activity.
No, I mean, like I plan on running
this thing in for the rest of my
life, but, you know, I mean, uh, if
people don't use it or support it financially,
I mean, we have a lot of, now
we have a lot of use.

(59:08):
I mean, um, usage has steadily usage of
the index by apps has steadily gone up
over the years and it never, that number
does not go down.
Number go up.
Number go up.
Well, it is.
It is absolutely beautiful resource.
It's so beautiful.
There've been so many people have tried to

(59:29):
do commercial indexes and while they've certainly had
success, you know, with 40 employees, 140 employees,
like, no, you don't need that.
This just needs to be around.
And I'm sure that the biggest fear I
have is you, that's the problem.
When you're like, as long as I'm around,
when you're not around, it's like cut off

(59:50):
his thumb, put it on the iPhone app,
get the biometrics, take the imprinted into some
clay, please quickly, you know, and then we're
going to have to figure out how you
did it.
The fly out to Alabama and be camping
out, you know, your body's still going to
be there, you know, laying in state like,
Hey Dave, how you doing?

(01:00:11):
We're going to be thinking.
I've got a dead man's switch.
No, you don't.
No, you don't.
You don't.
Uh, this, the, uh, yeah, the, yeah, the
number just keeps going up and I'm, I
tried desperately to keep from having to upsize
servers and all that stuff.
I mean, I've, I just pull out all

(01:00:31):
every trick I've got to try to keep,
because I do not want our expenses to
go up any more than they already are.
Uh, I just, even though traffic has gone
up in the last year and a half,
probably 30%.
Um, and most of those people don't, they
don't ever pay us a dime.

(01:00:52):
No, no, you know, but the bus, bus
route pays us every month.
Uh, fountain pays us every month.
Mitch at podverse pays us every month.
They pitch in stuff, you know, periodically, but
that's about it.
I mean, we, we have monthly, we have
a lot of monthly supporters, you know, the

(01:01:12):
$25, the $5, all of that matters.
Believe me, $5 is a Linode server.
Uh, a little one gig serve.
That's a party time party time.
But all the, uh, Brandon at pod page,
but most of the people, like we have,
we have a person that people would probably

(01:01:32):
know that their entire product is built on
the index, paid a paid product.
Um, and they don't ever give, they don't
ever donate anything.
And w which is fine.
That's, that's fine.
But I'm saying like this, the only time

(01:01:54):
I ever hear from this, from this guy
is when he emails me to say, Hey,
this, I'm trying to get this podcast and
it's broken and I've got a paid customer.
So I'm like doing tech support for this
person for their broken stuff.
When they moved to us because they were
having trouble with Apple's API, they moved over

(01:02:15):
to ours and they, it's a paid service
and they, they don't donate a thing.
So we're, we're supporting tons of, of apps
that, that make money that don't pay us.
So like, and that costs money for us
and time, you know, and so like, that's,

(01:02:35):
it's not a, some, some days it's not
fun to do this.
But you know, you just keep on, you
just keep on doing it.
Keep on keeping on baby.
Keep on keeping on.
We give us some of that AI cash,
baby.
Yeah.
Marco gave us 500 a month for years
and I will always be appreciative of that.
Certainly in the beginning.

(01:02:56):
Yeah.
Yeah.
Yeah.
And he knows, he knows that even at
any moment I've corresponded with him before.
He knows that at any moment if Apple's,
if Apple's API goes down, he can switch
over to us.
Drop in replacement.

(01:03:16):
Yeah.
Drop in replacement and then when they come
back up, he can flip back over.
That's always the case.
Nobody has to pay us in order to
use our service.
But it does, but it is expensive and
it has real world consequences.
So we do actually have to have some
money.
Speaking of such, if you feel so moved

(01:03:37):
by the value that the podcast index provides
to you, your project, your listening pleasure, the
app you use, consider supporting us by sending
us a booster gram, which you can do
from many of the modern podcast apps.
You can even do it on podcastindex.org
by connecting your Albi hub.
I just did that today.

(01:03:58):
Yeah, I sent a sad puppy boost to
This Week in Bitcoin because I've been using
the Godcaster to listen.
So I'm not streaming sats anymore at the
moment.
Down the road, that'll change.
But because that app has gotten so good,
it has all the podcasts I want and
he was low on donations.

(01:04:19):
I'm like, oh no, this is no good.
And I just went over there and you
do have to hit it multiple times for
the because of the splits that the Albi
plugin asks you to.
I guess if you pre-approve it, then
it's OK, but I don't use it that
often.
But it works really well.
And it went straight from through Albi's, you
know, my Adam at AlbiHub.com, I guess,

(01:04:43):
or Albi.com to my node here at
home.
Worked fine.
It's amazing.
Any of the stuff works at all.
It was quite beautiful.
So you can do it that way.
And of course, on podcastindex.org, if you
scroll down to the bottom, there's a big
red donate button.
We gleefully accept your Fiat fund coupons through

(01:05:03):
PayPal.
There we go.
See, Laszlo on Linux comes in with one,
two, three, four sats.
He says, can the stats page show other
things too?
Request amounts or bandwidth or other.
What?
I guess he's looking more for stats from

(01:05:24):
the index on the stats page.
Like how many requests we're getting or how
much bandwidth we're using.
Oh, oh, like, OK, I see.
So like more than just the content of
the index, but actually like the performance metrics
of the index.

(01:05:44):
Yeah, yeah, yeah.
A dashboard.
I see a dashboard coming.
Dashboard, yeah.
Uh, the health, the health dashboard.
That's an interesting idea.
If you're in there anyway, if you're doing
stuff anyway, you know, you got some spare
time.
Yeah, I guess there's plenty of that.
Yeah, that's an interesting idea.

(01:06:05):
Yeah, I'll think about that.
RoaDux from Eric PP.
Value!
Exclamation mark.
Just got two sats from IPFS Podcasting.
I still run my IPFSPodcasting.net node.
I'm hosting a lot of stuff, which, by
the way, what a beautiful thing to see
working perfectly when all these, uh, when AWS

(01:06:27):
outage happened.
And I had no idea how many people,
how many podcast company hosting companies were dependent
on AWS.
That was rather interesting.
DynamoDB was the culprit, evidently.
And I've used it.
Well, it's DNS basically, right?
Yeah, but I think, yeah, it took every,
like DynamoDB, I think, if I'm not mistaken,
was the very first AWS service.

(01:06:50):
I think.
Oh, really?
Other than just VMs. I think DynamoDB was
like in their original service launch because they
used it.
They used that internally.
That's like part of Amazon's core infrastructure.
They depend on DynamoDB.
Yeah, they depend on it.
Yes.
So Eric PP sent that from CurioCaster.

(01:07:11):
Class on Linux through Fountain.
Salty Crayon sends three, three, three from Podcast
Guru.
Uh, says backup Boostergram.
Howdy, boardroom.
It's day 24 here at the DOW.
And I'm just here keeping the lights on
while everybody is off on vacation in other
states.
We living in peak clown world, five by
five in the pipe.

(01:07:31):
I would say DOW's Department of War is
what I'm thinking.
Thank you.
Thank you for your service, Salty Crayon.
The 1701 from Lyceum.
That's Martin Lindeskog.
I'm sticking around, but I could be lost
in space now and then.
Starship Boost with 1701, Satoshi's PS.
I like the real money, silver and gold
talk at the end of the previous episode.

(01:07:53):
Wait.
Yeah.
Did we talk about that?
Yeah, we did.
And then I hit the delimiter.
So you're up, Davey.
Uh, we, let's see.
Oh, I just, of course, I lost my
page.
You got Buzzsprout.
Speak of the devil.
Thousand dollars.
Bam.
There it is.
Baller, shot caller.
20 inch blades, only in baller.

(01:08:15):
Thank you, boys and girls of Buzzsprout so
much.
That, that makes a huge difference.
It really, really does.
Yep.
It basically pays the bills.
Yep.
Yep.
Thank you guys.
And J, uh, see donors.
A Jacob Salinas sent us $10.
Thank you, Jacob.
Thank you, Jacob.
Appreciate that.

(01:08:35):
Uh, let's see.
We got, uh, oh, oh, Oistein Barra.
$5.
I guess he's coming up after the show.
Yes.
Mutton and Mead, Mutton Mead Music is, uh,
always takes over the screen right after the
board meeting.
Michael Goggin, $5.
Thank you, Michael.
Jorge Hernandez.
That's $5.
Thank you, Jorge.
James Sullivan, $10.

(01:08:56):
Thank you, James.
I hope the, I hope the person or
company or whoever is, uh, is making you
do tech support for free.
I hope they hear this and they, and
they understand that these people are actually paying
for the support they need.
Although it's not paying you, but they're supporting
the service they use.
You must tell me offline who this is

(01:09:16):
so I can pray for them.
That was what I always appreciated so much
about, uh, about Marco's donations is he was
funding other, he's funding his competitors.
Yeah.
Christopher Reimer, $10.
Thank you, Christopher.
Appreciate that.
And, uh, Cohen Glotzbach, $5.
Uh, Randall Black.

(01:09:38):
Hey, Randall.
Thanks, brother.
Uh, $5.
And, uh, that's it.
We got some Boostergrams.
Let me flip over here to the Boostergrams.
Let's see.
Someone was asking me the other day, how
does Dave get the Boostergrams?
And I think the answer is you have
a script that runs, um, probably through, what's
the command line tool we always like?
Uh, you pull it out of the node

(01:10:01):
directly, right?
Yeah, I have a, I have some PHP
scripts that run, that pull, uh, boost, that
pull all of our, uh, for tax, for,
to keep track of taxes for, for us,
I have a script that downloads everything, every

(01:10:22):
transaction out of our lightning node, uh, once
a minute.
Nice.
And so then it, and it sticks it
into a local database on that machine.
And so then at any moment I can
go and give it a range of time
and say, give me, extract all the Boostergrams
out of these donations for this rain time
range.
And so what does the, uh, what does

(01:10:43):
the communication with the node?
Uh, it's just, it's the REST API, context
of the REST API.
Huh.
That's a cool little widget.
Yeah, it's cool.
It's just a, uh, it's just a PHP
script.
Yeah, I like it.
Yeah.
It's a set of them.
Yeah.
And then I got another set of them
that exports everything at the end of the
year so we can do our taxes and
all that.
Hmm.

(01:11:03):
Um, it's homegrown stuff, you know, uh, let's
see, Bruce, the ugly cracking duck, 2222, do
podcast guru.
Thank you.
It says, uh, what?
You mean that video is the zombies platform
and audio still shines life into our brains?
Wait, you guys probably didn't say that.
Haha.
I just dreamed it.
73s.
73s.
Cole McCormick, Satchel Richards, one, one, one, one

(01:11:25):
through fountain.
He says, while you were discussing short form
video, my mom sent me five different reels
on Instagram and Facebook in a 30 second
span.
Yeah, we, I'm sorry for that.
We set up this, uh, this group.
We had about 14 people over the other
night at our house and, uh, it's kind
of hokey, but I'll tell you what it
is.
So we, um, there's a movie by Kirk

(01:11:47):
Cameron called fireproof and this.
Yeah, I remember seeing that.
Yeah, I don't think I saw it.
15, 20 years ago or something.
And, you know, he's going through a marital
problems and then his dad gives him this
book and it's called the love dare.
And for 40 days you do this thing
and it's, it's based on scripture and it's,
it's kind of fun.

(01:12:07):
And so, uh, someone said, Hey, let's do
this.
And we'll, you know, so, and some of
us have been married for one year.
Some of us have been married for 30
years.
So we all got together and, uh, and
we, and, and, you know, so we did,
you know, we talked about it, did this
and we set up the startup.
It's basically hang out, you know, everyone brought
food, but then they put together a text

(01:12:29):
group and I'm like, okay.
And then the purpose of the text group
was supposed to be for the organizers to
post at in the evening, post the love
dare for the next day.
You know, so you don't have to guess
it turned into nothing but memes, dude, dude,
like pictures, Instagram, little mini reels.
I'm like, this is, this is the problem.

(01:12:51):
People I'm out.
It's like, please.
And of course everyone hates me because I
turned the whole group into green bubble with
my Android phone.
Like who has the Android?
Yeah.
You're gross.
Chad F.
Sasser Richards, one, one, one, one.

(01:13:11):
Thank you, Chad.
Great hearing Barry from Noster.
He's got a Noster handle there on the
show.
He's building some cool stuff and I always
hear great things about Podhome.
And again, Chad will be on the show
with us next week.
Chad's voice sounds very different than I envisioned
it.
Oh yeah?

(01:13:31):
Yeah.
He posted a video of his, uh, of
his, his, uh, test app.
I'm like, ah, I never thought that that's
how he would sound.
You know how you, you put a, I
don't know if you had, you put a
voice to a, to someone who you only
see online.
And somehow I thought he would have this
really high pitched whiny voice.
I'm sorry, Chad.
I can't help it.
That's just what I thought.

(01:13:52):
And he doesn't.
I'm like, oh, that's kind of cool.
I can't wait to talk to him.
Okay.
I always do the, uh, the reverse.
I hear somebody's voice and I picture what
they look like.
And then I see them and I'm like,
whoa, that's not what I thought.
Yes.
Um, see, uh, see loss on Linux 2222
through fountain.
He says, Dave talking about a second vibe

(01:14:13):
coding pass to check for errors and issues.
There's at least one service called code rabbit
.ai. You connect it to your GitHub or
other.
And then on a pull request, it will
do things like read the code, make a
change, uh, see, make a changes graph, spin
up linters and static analysis tools and check
your code for all sorts of issues.

(01:14:33):
They have a CLI and plugins for editors
too.
It's good.
Even if you don't vibe code, well, that's
good.
I mean, yeah.
Fuzzing would be a good thing to do
for AI fuzzing fuzzing.
Yeah.
That's where you take a piece of where
you take a piece of code and you
just throw all kinds of, let's say that
your code expects a particular protocol.

(01:14:55):
You throw every different weird bug and permutation
and wrong thing at this code to see
how it breaks.
That's called marriage.
My friend, baby.
Stop fuzzing me.
Yeah.
Yeah.
That's a, a marital fuzzer is what it
is.

(01:15:16):
Oh, beautiful.
Seals and laces back again.
One, two, three, four.
Uh, through founding, he says, Dave said no
one under 30 or whatever number he said
doesn't use eggs and used Instagram.
I'm 25.
I'm probably addicted to eggs, but not using
Instagram.
And I see many people similar age and
younger use eggs, maybe just nerds, developers, gamers
and other weirdos.

(01:15:36):
We'll see part of CLAs part of one
of the, one of the privileges you gain
when you get to be, you know, old
in your old of any caliber is that
you get to just say things like no
one under 30 dot, dot, dot.

(01:15:58):
Yeah.
You don't have to have justification approved.
You just get to say it.
It's a right.
It's, it's, it's a human right.
Exactly.
Yeah.
And you just have to, and you as
an under, as an under 30, you just
have to take it.
Take it, take it like a man.
Yeah.
Take it and like it.
CLAs again, four, three, two, one.
He says, uh, four, three, two, one sats

(01:16:18):
through found.
He says, get to work is actually a
good motivation.
It's not like I haven't been coding.
I just didn't spend a lot of time
on bounce pad anymore after realizing most podcast
app developers can kind of take back in
the costs.
And then that's it, which is too small
for me.
I think I don't need to build Facebook
level things, but a thing with barely any
users isn't fun either.

(01:16:39):
I was supposed to build it out a
bit more and then get on a call
with Todd because he wanted a specific thing
built to integrate with, but then his passing
was basically stopped me fully for now.
What qualifies someone to get on the show?
Well, there's no qualification.
Let me just product helps.
Yeah.
Having something to talk about.
Yeah.

(01:16:59):
It qualifies when you have something to talk
about.
So let us know what you want to
talk about.
We are, we are open and, um, and
equitable, equitable.
If you, I mean, if you're, if you're
developing stuff and doing stuff with podcasting and
two dot oh, come on the show.
Yeah, for sure.
Doors open, hit us up.
DMs are open.
Hit me up.
Comic strip blogger, delimiter, the perpetual delimiter 14

(01:17:23):
0 1 5 through fountain.
He says, howdy fellow bit corners, Adam and
Dave, this promo written by grok based on
his website.
Join Martin Lindeskog as he navigates the gig
economy, sharing candid stories, triumphs and hacks from
podcasting trends to LinkedIn strategies.
Get inspired to thrive as a freelancer.

(01:17:46):
Tune in for insights and community vibes.
Listen now at gig.elitoo.com or search
in podcasting 2.0 apps for swing that
gig exclamation point.
Yo, CSB AI arch wizard.
Of course, he's an arch wizard.
He's interesting to see what the vibe coders

(01:18:06):
do.
I mean, I think Chad F is basically
a vibe coder.
I believe he is.
I want to ask you this as, as,
as, I mean, I've, I consider you to
be a certified vibe coder at this point.
I, I, I, I concur.
My question is, do you know, no, I
know sort of the answer to this already.

(01:18:28):
I know that you are part of what
you're going to say is that you do
this because you, of the pure enjoyment of
it.
But the other side of this question is,
would you, do you think that vibe coding

(01:18:48):
this product that you've created, do you think
that it has gone any faster than it
would have if you would have had a
developer to communicate with and build the product
per your instructions?
I've thought about this a lot.

(01:19:10):
It would not be the product that I
wanted you and I now we've worked together
for a long time.
So I think that what I've spent now,
the product has a number of different angles
to it.
There's an, there's a bunch of different pieces
that you could say is all part of
the same product, but it's kind of separate

(01:19:31):
products.
Um, because we've done this, we, we built,
we built Freedom Controller.
And, and of course, what was most beautiful
about Freedom Controller is that your creativity went
in there.
Um, and we had lots of conversations like,

(01:19:53):
how about this?
Let's try this.
Let's try that.
But ultimately your creativity was what created the
end product with my input, what I have
built, the end result, which is streaming products,
live takeover products, a proxy for metadata and,

(01:20:15):
uh, uh, something else.
Um, we could have probably built this in
three weeks time together.
But because of the vibe coding aspect, the
way it came together, I didn't even envision
it to be doing what it does.

(01:20:36):
I mean, it's, it's mind blowing what came
out of this and I'll give an example.
So as I'm, and this, this was the
first time I was really blown away by
the vibe coding.
And of course it comes from something in
the corpus and Lord knows how it got
in there or how it, how, how it
wound up in my project.

(01:20:57):
Um, so it's a streaming radio station system.
Very, very simple.
That's intended for really for churches and ministries.
If, if you're managing a YouTube channel and
an Instagram thing, and you're doing live streams,
you should technically, if you have a little
bit of understanding of scheduling, then you should
be able to use this product.

(01:21:20):
And, um, and so I gave it to,
I don't think you've had time to really
mess around with it, but I gave it
to Gordon.
Gordon's a long time radio guy.
And, um, and so I got some valuable
feedback from him.
And so there's a, there's a page that,
you know, so there's basically production, a scheduling
page where you schedule everything, whether you want,

(01:21:42):
you know, your lineup hour to hour, what
elements you want in there, your format, as
we would say in radio.
If you want to see, you know, most
recent episode from an RSS feed to play
at a certain time.
And so that's kind of the, the, the,
that's what people are doing most of the
time, but then there's also a studio page.
And I built the studio page, um, because

(01:22:03):
not everybody is able to, um, get someone
to record a local report.
And so it's connected to 11 labs and
you just, you know, you, you copy paste
or you type it in whatever you want.
And, uh, and it spits it out and
you can preview it there.
And Gordon said, well, okay, I put it
in a folder, but what else is in
the folder?

(01:22:24):
Can I listen to the other things in
that folder?
Well, that's actually on a different page.
So I'm like, oh, I need to put
a preview section on this studio page so
that you can preview other stuff and determine
where you want to put your newly created
AI voice stuff into.
And without me asking for it, it automatically,

(01:22:48):
um, when you create a new report, let's
say, and you want it to be in
the local folder, the local bin, as it
were, the preview section automatically selects that bin
and that file that you just created negating
the whole necessity for having a separate player
pop up to preview what you just did.
It was something that I loved it.

(01:23:10):
I'm like, wow.
And actually against all my rules, it went,
good job.
I was so happy with it.
And that, and that was something that I
don't know if that would have come up
if you and I were working on it
together.
And there's, there's a lot of other small
little things.

(01:23:31):
Um, but time-wise, dude, I've been working
on this for almost eight months and, you
know, I started off with liquid soap.
I killed that.
And I wound up building with the Swiss
army knife of all things, audio and video
FFmpeg.
I want, and that was actually after podcast
movement, talking to Rocky and her husband, like,
oh man, you should, you know, you should

(01:23:52):
take a look at FFmpeg.
And it's, I mean, it's rock, rock solid.
It's, it's unbelievable how, how well that works
compared to, I mean, and I don't want
to get into a, you know, a religious
war about liquid soap versus FFmpeg, uh, but
the way it functions, how it works.
Um, I don't think it would have, it

(01:24:14):
would have been a great product.
Uh, but it wouldn't have been what I've
wound up with now, because I was just
jamming with the machine, you know, I'm just
like, oh, let me try that.
Oh, you give me this.
So it was a different kind of jam
than you and I would have had, but
we would have been done with this in
three weeks, guaranteed hands down, instead of six
to eight months.

(01:24:35):
So would you, I think what I hear,
I think what I'm hearing you say is
that the ability for you to do this
yourself, um, you're trading time for sort of
inspirational control over the end result.

(01:24:57):
Um, well, you're seeing things as you, since
you're, you're doing it and seeing things and
getting the feedback that inspires you to do
perhaps different things than you thought you intended
to.
Yes.
It's because it's so wonky and so crazy
and so like stuff that, I mean, once

(01:25:18):
I got over the, why did you change
something that I liked five hours ago?
And I figured out, okay, you can usually
stop it from doing that.
The things that would come out of it
sometimes were incredibly surprising.
I'm like, oh, oh, okay.
I can use that in this way.
So it was a, it's a very creative

(01:25:39):
process because I don't know what I'm doing.
If I just sat down and I, you
know, I, I went through the, the Python
course and I, and I learned everything and
all these bits and bobs, but you know,
it'll just pull, it'll pull in, um, a
library that I'd never, I'd never even heard
of or considered like, oh, that's interesting.
Okay.
Well, now I see that we can do

(01:26:00):
this in this way or we can add
this.
So it is a very creative aspect to
it, but it was like, you know, like,
uh, you know, like doing, um, pet sounds
album, you know, it's Brian Wilson, 15 years
to complete or whatever.
Um, so, so it is, I find it

(01:26:20):
very creative because I don't really know what
I'm doing.
And so what came out of it were
surprising, surprising things that perhaps someone who was,
um, you know, I, I kind of know
how you work and you set out a
plan, you structured.
And I mean, there's a lot of things
I can never achieve that you can do

(01:26:41):
never.
I mean, just the beautiful, like how you've
done the dashboard in Godcaster is such a
work of art.
So everything makes sense.
My dashboard is like, oh, okay.
You know, it feels like someone just threw
some spaghetti against the wall and it looks
kind of, that looks like art.
Yeah.
I'll keep it because there's not much I

(01:27:02):
can change because I just don't know what
I'm doing.
But aren't there old, successful, very, very successful
softwares that have like, I think of things
like e-tools and the UI is just
a complete disaster, but everybody, but it's like,
once you learn it, you can be highly
productive.

(01:27:23):
Well, and, and that's kind of where I
started because I looked at every single streaming
platform.
I looked at, you know, liquid soap has
a lot of UIs.
I want to say CentoCast.
I mean, there's all kinds of UIs that
have been on top of it and it's
just not how I wanted it to work.
I really just wanted it to be something.

(01:27:44):
In fact, the, where I started was, I
want you to be able to do this
from a Google calendar and that turned out
to not be so handy in the end,
but the idea was kind of good.
I remember that phase of you.
Yeah.
And, uh, and you know, I built the
bridge, the GCAL and all that stuff, but
it's just, yeah, it just didn't feel right.

(01:28:04):
That's a mess.
And then, you know, and I felt very
dependent upon, upon Google, of course.
Um, and, but what, what came out of
it is, is pretty close to, to what
I want.
I just want something very simple that doesn't
deal with, you know, format clocks and all
this stuff.
I want it to feel like a calendar,
like, okay, I understand every day at eight,

(01:28:26):
I want this devotional to run.
I want, you know, at the top of
the hour, I want Salem radio news.
And then I want a live break in
for the Sunday sermon, all this stuff.
And I, and I, it just came out
kind of the way I want it, but
to just through constant re-recording, if you
will.
It's like, this track's no good.
Let's do it again.

(01:28:46):
All right.
But now, now I've got this mix is
good over here, but now it sounds crap
with this track in it.
So it feels very creative, but I personally
would have preferred to have done it with
you.
I would, because we've done it before and,
and I, and I love how we work

(01:29:06):
together, but you know, you're, you're working on
other things.
So it was out of necessity that I
did this, not because I was so gung
ho on becoming a developer.
But yeah, I mean, I've been meaning to
ask you about that.
Cause it's interesting to like from, from a,
there are times when trading time for other

(01:29:27):
things makes sense.
Yeah.
And I was curious as to if that
was the act, sort of an active thing
in this case, cause you're, I agree with
you.
I think you, you, you have to live
with a product in a way, like when
you, if you're building it, you're just going
to see things that, you know, in a

(01:29:49):
way there it's, it's two, there's two sides
to it.
Like if you're building it, you see things
that the user will not see.
Right.
And if you're using it, you're going to
see things that the developer's not going to
see because they're too close to it.
Yeah.
And so like you're, you're, you don't normally
see it from the developer side.
And so you're having seen it, like doing

(01:30:10):
it from that side of things is, I
think, I think one thing you, one thing
about the way you approach things as well
is that you tend to look at something
visually and kind of, you visualize something and
you immediately get like an explosion of ideas.
Yeah.
This is true, right?

(01:30:31):
Yes.
Yes.
You, but you, cause you do this all
the time.
You're like, Hey, I just saw so-and
-so, you know, we could do this and
this and this and this and this.
I mean, you just, like, there's just like
an orgasm of explode.
Ah, there it is.
An orgasm of ideas.
Okay.
Yes.
Um, that, that happened.
And then, but, but when you see it

(01:30:53):
as a developer, you can actually test the
idea like immediately.
Yes.
You don't have to like hope and explain
it to somebody.
You can just immediately like boom, boom, boom,
boom, boom.
And then you feel like, then you find
out, ah, that's not going to work.
Ah, that's not going to work.
Oh, wait, that, that works.
You know?
Yeah.
That's interesting.
I'm, I'm, I'm, I'm in, I have enjoyed

(01:31:15):
watching you do all this.
Well, uh, and I will say for those
who are interested, um, I, I actually asked
God, why are you putting me through this?
That's, that's my prayer.
Maybe three or four times a week.
Yes.
And, uh, and interesting.
And this was a several months ago after

(01:31:36):
I kind of completed the first phase and,
and the word I got back was, I
wanted you to understand what AI is and
what it can and what it can't do.
And that was, I was like, wow.
Okay.
That's very helpful because, and, and also I'm
just like, I'm, I'm amazed at this period
of time that we can do this for
$20 a month.
I'm just blown away by it.

(01:31:57):
And I'm, I'm afraid for the day will
come, you know, when it's like, well, the
only business model we have is to charge
you what it really costs.
You know, I, I, I just like what,
what an amazing $350 a month, you know,
it's, it's an amazing time to be alive,
to be able to do this.
Um, well, by the time that day comes,
we'll have enough or we'll have an enough

(01:32:20):
horsepower to just do it at home because
I have actually, I have tried that and
it does actually work.
I mean, I've, I've done, I've done a
couple of, uh, you know, a couple of
small pieces of the coding, um, on my,
uh, on my own, uh, start nine that
has, uh, you know, was it the web,
web, what does that call?
WebView, WebGBT, you know, what is it called?

(01:32:42):
Yeah, right.
Yeah, I know what you're talking about.
You can just load them up and you
can, yeah, yeah.
You can load the models in there and
it spits back code and like, oh, that's
pretty cool, but it took forever.
Yeah, it's slow.
You have to have some, you have to
have a decent GPU.
Yeah.
So, you know, that, that, that didn't work,
but yeah, yeah.
That, that day will come.
We'll see.
Wow.
Okay.

(01:33:02):
Brother Dave, uh, what are you doing?
You hanging out this weekend, taking some time
off?
We have a very busy weekend.
We're going to a rock climbing competition for
my daughter in Atlanta.
We'll probably be going all day tomorrow.
All right.
And then, uh, we're going with them.
My daughter has, um, she's 15.

(01:33:23):
She started, she's, she teaches part-time at,
uh, a local, uh, outfit called the, uh,
Firehouse Community Arts Center.
Oh, wow.
And they teach, uh, she teaches kids how
to play instruments.
And so she, they, yeah, they have a
big block party about three times a year.
And so all the kids form into bands

(01:33:44):
and play.
Oh, that's cool.
And they play their favorite tunes.
And so she's playing is she, her band
is playing, and then she's playing on about,
she's playing bass on one song and drums
on another song.
So that's a, that's an all day affair
too.
So I love that the kids are still
playing music at all, ma'am, these days.
They only play music from the old, they
only play old music.
They, they, they, nobody plays anything new.

(01:34:07):
They all love the oldies.
Like eighties, oldies, seventies, oldies, sixties, oldies.
What kind of oldies?
You, you go to this, you go to
the block party.
On the last one, I heard, uh, War
Pigs, uh, the, um, the song that, uh,
duet between Dolly Parton and Kenny Rogers.

(01:34:28):
Islands in the sun.
Islands in the sun.
Yeah.
That, uh, that there was an old, uh,
old radio head song.
Wow.
Uh, there was a David Bowie, uh, star
man.
It's really old stuff.
Holy crap.
That's amazing.
Probably the newest song I heard was from
like 2000.

(01:34:48):
That gives me hope, brother.
That gives me a lot of hope.
I'm loving that.
Even the kids, the kids may listen to
new music, but they don't want to play
it.
All right.
You have a good weekend.
Bored room.
Thank you very much for being here.
Remember to support us.
Podcast index.org.
We'll be back next Friday with another board
meeting of podcasting.
2.0. You

(01:35:23):
have been listening to podcasting 2.0. Visit
podcastindex.org for more information.
Go podcasting.
He just roasted my chestnuts.

All Episodes

Episode Transcript

Popular Podcasts

Stuff You Should Know

Dateline NBC

On Purpose with Jay Shetty

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}Episode 239: Fuzzing

Episode Transcript

Popular Podcasts

.css-r6mb8g{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:1;overflow:hidden;}Stuff You Should Know

Dateline NBC

On Purpose with Jay Shetty

All Episodes

Episode 239: Fuzzing

Stuff You Should Know