WEBVTT

00:00:00.000 --> 00:00:04.759
So I walk up in a sleeper train in the morning and I'm looking at my Apple watch and the

00:00:04.759 --> 00:00:09.240
reason notification from GitHub. It doesn't just start with the holy shit, holy shit, it finally

00:00:09.240 --> 00:00:11.720
worked. And it was like, what a way to start a day.

00:00:14.839 --> 00:00:20.800
Hello and welcome to episode 61 of the Metacast Behind  scenes podcast. I'm Léa Bessilov.

00:00:20.800 --> 00:00:27.679
And I'm Arnaugteca. We are co-founders of Metacast, a podcast app best known for transcripts.

00:00:27.679 --> 00:00:32.079
So we are all about transcripts. We have transcripts for every episode ever published. If you

00:00:32.079 --> 00:00:35.920
don't, you can easily generate one. So the app generates transcripts for you on demand.

00:00:35.920 --> 00:00:40.159
So you can get access to any transcript that you can bookmark, you can search, you can

00:00:40.159 --> 00:00:45.878
share quotes from it, you can save quotes from there. And as we develop the app further, transcripts

00:00:45.878 --> 00:00:50.560
will be more and more advanced. So that's the thing that we want to be known for. And actually

00:00:50.560 --> 00:00:56.119
we will talk about that specific thing in the episode later today. We are in production

00:00:56.119 --> 00:01:02.280
on Android. So you can download Metacast on Google Play Store. And we are also in beta on Apple

00:01:02.280 --> 00:01:08.039
through TestFlight, which you can get through the Metacast.app website. And we will be very soon

00:01:08.039 --> 00:01:12.920
launching on App Store. And we are also going to be talking about that in today's episode.

00:01:12.920 --> 00:01:17.159
So this podcast is where we talk about how we build apps, how we build the business, building

00:01:17.159 --> 00:01:23.239
public stuff. So Arnaugte, what are we going to be talking about today? So I think first we have

00:01:23.239 --> 00:01:27.879
three or four topics. We'll see how long we can do it. We'll try to keep it to about 30,

00:01:27.879 --> 00:01:34.280
35 minutes today. I think loneliness of being in a very small company. That's one of the topics.

00:01:37.319 --> 00:01:42.759
Then we had a pretty cool request from a couple of users and we decided we don't have time to

00:01:42.759 --> 00:01:49.159
build the feature, but we hacked our way around it. So we'll talk about that. Then last few weeks,

00:01:49.159 --> 00:01:55.079
added to the loneliness I've been banging my head against some iOS and Flutter was. So we'll talk

00:01:55.079 --> 00:02:00.519
a little bit about that. We'll probably try not to go too deep technical in there. And finally,

00:02:00.519 --> 00:02:06.280
the thing that Ilya mentioned is what do we want Metacast to be primarily known for? So we're

00:02:06.280 --> 00:02:10.758
going to talk about that. And then of course, like we do, we'll close with what we're listening to

00:02:10.758 --> 00:02:15.719
and reading and all that. Let's go. Let's start with the confession of the lonely founder.

00:02:15.719 --> 00:02:23.240
Yeah. So last couple of weeks, Ilya and Jenny have been away. We have had times like this. So with

00:02:23.240 --> 00:02:31.319
Jenny, how long has it been October 1st, 2023, right? Yes. Yeah. It's about a year. Okay. So it's been

00:02:31.319 --> 00:02:36.840
just about a year, less than a year. Before that, you and I were talking on like WhatsApp and

00:02:36.840 --> 00:02:43.240
Signal and all that, I think mostly. But after Jenny started, we primarily transitioned into Slack.

00:02:43.240 --> 00:02:54.599
So our Slack has been, even though it's just three people, it's been pretty vibrant because we not just share what's going on in our work, but we also talk a lot about our life and what we're

00:02:54.599 --> 00:02:59.479
reading and listening. Interesting things that are happening in our life as well as not interesting

00:02:59.479 --> 00:03:06.120
things, I guess. But last two weeks, you both were away and you were away in a way that you both

00:03:06.120 --> 00:03:12.919
didn't want to use Slack. We were unplugged. Unplugged. Yes. Both of you. And we have had times where

00:03:12.919 --> 00:03:18.039
two of us or three of us were away for like a week. But I think we were still maybe checking

00:03:18.039 --> 00:03:24.118
Slack or responding to it sometimes things like that. This time was like truly unplugged for both of

00:03:24.118 --> 00:03:31.000
you. And I think in the mornings, my ritual is like, I wake up around 7 a.m. and I open Slack and

00:03:31.000 --> 00:03:36.838
see what's new because you both are time zone wise like ahead of me. So there's already like things

00:03:36.838 --> 00:03:41.799
happening or you may have responded to some things that I asked yesterday. So I go through it and

00:03:41.800 --> 00:03:47.479
unconsciously, I guess I was getting a bit of a dopamine hit kind of effect from it. Like people

00:03:47.479 --> 00:03:52.520
do from Instagram and Reddit and all that, right? Like you open it up first thing in the morning

00:03:52.520 --> 00:03:58.599
and doom scroll through it. And these last two weeks, I felt like, oh man, this is so sad.

00:03:58.599 --> 00:04:04.120
Like there's nothing happening, right? And I don't know if you noticed or Jenny noticed,

00:04:04.120 --> 00:04:09.800
but I kept posting things on Slack and we kept ignoring them. Yeah, yeah, which is fine, right?

00:04:09.800 --> 00:04:14.360
I didn't expect you to read it, but it's like, okay, when they come back, we'll talk about these

00:04:14.360 --> 00:04:20.120
things. I don't want to forget about it. So it kind of reminded me of that. And I'm glad that we're

00:04:20.120 --> 00:04:34.439
a small company, but still there's so much collaboration. Whereas I now feel like I don't know how indie solo developers, solo entrepreneurs go through it because they literally have nobody to talk

00:04:34.439 --> 00:04:39.639
to. Right. And you go to your family and they're like, when are you going to make money again? So it's like

00:04:42.120 --> 00:04:47.639
pressure from all sides, right? All sides. Yeah. And nowhere to release that pressure into.

00:04:47.639 --> 00:04:51.959
Right. And it's like, you almost don't want to talk about this stuff, the family sometimes.

00:04:51.959 --> 00:04:57.000
Stuff. Maybe that's why people go to co-working spaces and the meetups and all that, just to talk

00:04:57.000 --> 00:05:07.000
to other people, right? As you were talking about this, it made me think that let's say six months ago, we were also at a different stage where yeah, we were knocking out some features here and there,

00:05:07.000 --> 00:05:12.920
we were working mostly on smaller things where we were releasing every two weeks. We had a lot of

00:05:12.920 --> 00:05:23.879
progress and there was also a lot of collaboration going on. I think to the last couple of months, we entered in this kind of vlog stage where we needed to launch on the Google Play Store and now we

00:05:23.879 --> 00:05:35.399
are working on the iOS launch. It's just a bunch of lonely and interesting work that has to be done that's not around like, oh, how about this feature or like, how do you make these UX

00:05:35.399 --> 00:05:40.680
different, right? Or like, look how I redesigned this. It's more like, I'm going to my head on the wall.

00:05:40.680 --> 00:05:46.120
Why is this not working? Right. And it's like so complex that nobody else even wants to look into this.

00:05:46.120 --> 00:05:50.120
I'm like, we can't, I'm gonna figure it out. But it also creates this environment where,

00:05:50.120 --> 00:05:54.680
I mean, you're feeling lonely because you're just working on this and we can't help you really. So

00:05:54.680 --> 00:06:08.199
there is not much communication happening naturally, right? And when you're banging your head on the wall, I don't think there is a natural desire to like think about big picture and you're just so focused and also frustrated, which, I guess, compounds the problem, right?

00:06:10.439 --> 00:06:17.240
That's our CDO. He kind of, I think once in a while, reminds us that he's part of the company

00:06:18.279 --> 00:06:25.319
because he does nothing else. Although to be fair, I think that period of going through it was

00:06:25.319 --> 00:06:37.079
painful because we could not get it working. I literally have like over 100 commits trying to get this one thing to work, could not work from the tooling that we have Xcode and all that.

00:06:37.079 --> 00:06:43.000
There is no obvious hints or clues or errors at why it's not working, right? So that's why it was

00:06:43.000 --> 00:07:01.399
very difficult. But once it worked and that was on a beautiful Friday afternoon, I think it finally worked, then it felt like all of this was worth it because we got unblocked for our iOS we had like three features left to deliver for iOS before we launched on iOS and all three were

00:07:01.399 --> 00:07:07.959
actually coded, but we could not deliver it in a way to Apple devices where they would work.

00:07:07.959 --> 00:07:13.480
So that was unblocked all in one go and that was like amazing. Actually, I think it's a good segue

00:07:13.480 --> 00:07:20.360
into our next topic about iOS and flutter problems that we had. So those three features were links.

00:07:20.360 --> 00:07:31.000
I think it's called app links or something. The idea is so when you have a Metacast.app link that links to a podcast or an episode, if you have Metacast installed near your phone,

00:07:31.000 --> 00:07:34.920
instead of opening a browser for you, it opens the app with this episode. I think people should

00:07:34.920 --> 00:07:40.680
be familiar with this mechanics from apps like Airbnb and Twitter, I mean, app, right? You click on

00:07:40.680 --> 00:08:00.360
link it opens the app for you. But another like super cool thing we do is if somebody sends you a link to a Metacast, any podcast episode with a deep link inside the episode to a specific part and you tap that link, it'll open it up right there and you can read it or listen to specifically

00:08:00.360 --> 00:08:13.399
that segment. It's basically like YouTube, right? If a link has a time signature or time stamp, it goes straight there, which is really cool because if you have a very long legs episode and you want to share a specific moment with somebody, in other apps you have to tell people, okay, go to

00:08:13.399 --> 00:08:18.199
one hour, fifty three minutes and listen from there. Whereas on Metacast, you can just send a link

00:08:18.199 --> 00:08:22.519
from that moment on and then people will open it and listen right from there. So that thing wasn't

00:08:22.519 --> 00:08:27.959
working. It worked, but then at some point it broke and you can talk about why it happened. Basically,

00:08:27.959 --> 00:08:31.959
what was happening is you would click on that link and it would open the browser for you. We had no

00:08:31.959 --> 00:08:37.879
way to trigger an app to open when you click link. Whereas it works on Android and we, from our

00:08:37.879 --> 00:08:42.759
perspective, didn't change anything. We even had a consultation with Apple on this. I guess we can

00:08:42.759 --> 00:08:48.679
talk about that too. The other feature that was not working was Apple signing. It gave us the clue

00:08:48.679 --> 00:08:54.039
as to why the other thing, the links were not working too. And the third feature was the Bluetooth

00:08:54.039 --> 00:09:00.438
device integration. Like if you wear AirPods or if you're in your car, if you do like the play

00:09:00.438 --> 00:09:06.279
pause button in your car or on your headphones, it would play in pause. But skips were not working.

00:09:06.279 --> 00:09:10.438
Like if you wanted to skip forward for 30 seconds or skip back for 15, actually I forgot what

00:09:10.440 --> 00:09:16.279
our defaults are. 10 and 30. So we skipped backwards for 10 and skip forward for 30.

00:09:16.279 --> 00:09:20.279
Right. And we also make it configurable at some point. It just wasn't working. So like if you

00:09:20.279 --> 00:09:25.399
in the car, if you click next to previous, it just would do nothing. It just wouldn't react.

00:09:25.399 --> 00:09:28.919
Which was very annoying for people who are driving and they want to skip an ad or something

00:09:28.919 --> 00:09:35.000
because it's like five minutes of ads. And you have to open your car dashboard and like tap on

00:09:35.000 --> 00:09:40.519
the screen. It was a very inconvenient experience. So that was not working. It's been going on for a

00:09:40.519 --> 00:09:46.279
long time. I want to say months that we were struggling with this. That's when I think at some point

00:09:46.279 --> 00:09:51.240
we decided this is like taking too long. Let's just launch on Android. And I think we talked about

00:09:51.240 --> 00:09:55.559
this in the last episode. I think one episode to go. Episode 59, I think that's when we talked

00:09:55.559 --> 00:10:00.039
about that. Yeah, features we decided to punt on, right? But now that we have to launch an app store,

00:10:00.039 --> 00:10:05.958
we had no options to get figured out. It was existential, right? So yeah, Arna, tell us the

00:10:05.958 --> 00:10:20.438
glory story of this. Yeah. So these features Apple sign in and like universal links. They are native iOS features and they require you to like set up what Apple calls it or capabilities

00:10:20.438 --> 00:10:31.958
for your app. They're like extra capabilities. Like my app is able to send push notifications or it supports these universal links or it supports Apple sign in and you have to configure it on

00:10:31.958 --> 00:10:39.559
Apple's side like a website and also on your app. And there's a bit of madness in the way that

00:10:39.559 --> 00:10:46.919
Apple has created to be fair. So Apple created this whole ecosystem app ecosystem first. They created

00:10:46.919 --> 00:10:53.000
a lot of things that are only Apple specific. You require Xcode to build all this. Xcode is a tool

00:10:53.000 --> 00:11:06.120
that Apple provides for like building iOS apps. So you have to use Xcode and Xcode uses like all sorts of certificates and profiles and basically like security or cryptography management features.

00:11:06.120 --> 00:11:12.679
All of this have to work in conjunction to make such features that are native to iOS work on your app.

00:11:12.679 --> 00:11:19.720
So what was happening is before we move to our like automated build mechanism CI CD, we use

00:11:19.720 --> 00:11:25.639
tram line and GitHub actions to build the app every time we push a commit. Before that we were

00:11:25.639 --> 00:11:31.720
building it manually. So Xcode when you're building it manually on your MacBook, there's a flag that

00:11:31.720 --> 00:11:36.679
automatically manage profiles and certificates. And so we're doing that and it was beautiful like

00:11:36.679 --> 00:11:48.840
everything works. You don't have to think about anything. When you want to automate the build and you still need to use Xcode, then you have to basically manually configure all these certificates

00:11:48.840 --> 00:11:54.759
and provisioning profiles. Basically like if you're on a Mac keychain, you have to automate a lot

00:11:54.759 --> 00:12:00.759
of keychain actions and all that on our build servers. In the human language, it's a huge pain.

00:12:01.559 --> 00:12:07.879
Yes. Yes. And to be fair to Apple, they came up with it first. So that's why I think Google

00:12:07.879 --> 00:12:29.080
Android did a much more simplified, more standard way of building apps than they did. So what happened is when we moved our build into our build servers on GitHub, Xcode silently was basically saying, oh, I can't find this profile or actually not saying that's why it was silent, right?

00:12:29.080 --> 00:12:34.679
Was nowhere it was probably unable to find a profile or the certificates in the right place.

00:12:34.679 --> 00:12:39.159
And it said, oh, these features, I'll just drop them. So that's what it did. And there was no

00:12:39.159 --> 00:12:45.958
indication in the build logs itself that it was doing that. I mean, directly. So the end effect is

00:12:45.958 --> 00:12:52.919
that there is no errors. The build works. It ships the app to test flight or app store. And it just

00:12:52.919 --> 00:12:59.318
doesn't enable these features on the device, which is why it was very hard to debug. If you see errors

00:12:59.318 --> 00:13:04.198
or if you get a clue as to what is going wrong, you can figure it out. But if you don't see any

00:13:04.200 --> 00:13:09.559
effect, it's harder to figure out what's going on. It's kind of like you're back in JavaScript 20 years ago.

00:13:09.559 --> 00:13:15.960
Yeah. So I went through like a hundred steps of trial and error of configuring things in

00:13:15.960 --> 00:13:21.080
slightly different ways with Xcode and all that. All of this is scripted and automated thankfully,

00:13:21.080 --> 00:13:26.759
it was not manual. But like we would do one little change and then we have to push a new release

00:13:26.759 --> 00:13:31.720
to actually test it out. And then after a while once the app is deployed, you find out that it's

00:13:31.720 --> 00:13:38.279
still not working. So that was about a hundred or so at times of that. Eventually, I just deleted all

00:13:38.279 --> 00:13:44.919
our automation and started from scratch reading the documentation again. And even then it didn't work.

00:13:44.919 --> 00:13:51.240
But I think by going deeper and deeper into all this Xcode details and documentation,

00:13:51.240 --> 00:14:02.200
eventually, I think I figured out what was wrong. And that was that Friday afternoon when the first time Universal Links and Apple signed in and everything worked on our debug build.

00:14:02.200 --> 00:14:07.240
And I think it was a couple of days before that. I sensed so much frustration from you.

00:14:07.240 --> 00:14:12.600
I mean, you were not shy saying that you were frustrated. It had been like 10 days every day. I just

00:14:12.600 --> 00:14:17.399
keep trying and error because we have no other option. We have to make it work.

00:14:17.399 --> 00:14:22.759
Exactly. Yeah. And I remember I'm like, let's talk and you obeyed it. Our meeting. But then finally,

00:14:22.759 --> 00:14:26.919
you picked up the phone and I think you were walking with our studio. And I think we talked for a good

00:14:26.919 --> 00:14:31.958
hour or so. For my perspective, I felt like it helped relieve some of that frustration.

00:14:31.958 --> 00:14:35.479
Even though we didn't really solve anything, but it's more like talking about the founder

00:14:35.479 --> 00:14:41.399
Londoners. Because I felt like I had to talk to you. You asked deeper and deeper questions about

00:14:41.399 --> 00:14:48.120
what is wrong and like what's happening and all that. And it has also a rubber duck effect.

00:14:48.120 --> 00:14:53.958
It makes me think twice about what is actually happening and think about all of this again. So yeah,

00:14:53.960 --> 00:15:00.440
that definitely helped. I think finally getting it worked made it feel like, okay, this was worth it.

00:15:01.399 --> 00:15:06.200
I mean, to be fair, we had no other option other than to make it work. But when it finally worked,

00:15:06.200 --> 00:15:18.200
it felt amazing. And now I feel great that all these freezers are working. Yeah. And actually, it was so interesting to find out about you making it to work because it was just before I became

00:15:18.200 --> 00:15:23.320
unavailable for my travels. Right? So I woke up in the sleeper train in the morning and I'm

00:15:23.320 --> 00:15:29.240
looking at my Apple watch and the reason notification from GitHub. What did you say? Holy shit. I made

00:15:29.240 --> 00:15:34.200
it work. Something like that. It doesn't just start with the holy shit. Holy shit. It finally worked.

00:15:34.200 --> 00:15:39.639
Finally worked. Yeah. And it was like, what a way to start a day. Yeah. You should post that

00:15:39.639 --> 00:15:44.840
screenshot in the release notes somewhere of this episode. Yeah. Maybe you should make it the cover

00:15:44.840 --> 00:15:51.080
image for this podcast episode. Yeah. Yeah. I wanted to ask you how about the other one, the audio issue.

00:15:51.080 --> 00:16:05.080
The background audio. Yeah. So we have had background audio working for a while with Android. But like you said, Bluetooth and car controls for skip and rewind were not working on iOS only on iOS

00:16:05.080 --> 00:16:12.679
only. Yes. It was working on Android. And that was also pretty weird because this one comes down

00:16:12.679 --> 00:16:19.639
to be doing enough trial and error. So the thing is when you're building in development mode on a

00:16:19.639 --> 00:16:27.240
Macbook, you're using an iOS simulator. It's not a real iOS device. It's supposed to simulate an

00:16:27.240 --> 00:16:33.240
iPhone pretty accurately. And for the most part, it does pretty much 99% it does. But for these

00:16:33.240 --> 00:16:40.120
kind of things where it requires system native feature integration, it behaves differently. So we

00:16:40.120 --> 00:16:54.438
could not debug or we could not figure out what is wrong with this thing on a simulator. Basically, your code sitting on your Macbook running on an actual Apple device is more complicated than like

00:16:54.438 --> 00:17:01.240
you could probably imagine. So that's why we have to like again, go deeper, read a lot of code,

00:17:01.240 --> 00:17:06.440
like underlying libraries and all that. And I think eventually yesterday we got this feature

00:17:06.440 --> 00:17:12.759
also working. It was a similar, I think kind of frustrating path. But once it works, it feels great

00:17:12.759 --> 00:17:18.598
because we've been struggling with this on and off for months like you said, right? And eventually,

00:17:18.598 --> 00:17:23.480
we decided we punt on it because we need to focus our energies. We made the Android launch and then

00:17:23.480 --> 00:17:31.878
we came back to this. Right. Well, huge congrats. Yeah. So our iOS launch is coming very, very soon.

00:17:31.878 --> 00:17:37.720
Now that we are all done with everything, hopefully Apple approves it quickly too. Yeah. Very

00:17:37.720 --> 00:17:42.680
accordingly. So September 4th. So this episode will come out in a couple of weeks by that time. We

00:17:42.680 --> 00:17:48.440
will hopefully have submitted to Apple at least. Yeah. You are doing the product hunt launch first and

00:17:48.440 --> 00:17:53.240
then we'll do iOS and then do it again. Yeah, we are doing the product hunt launch actually tomorrow,

00:17:53.240 --> 00:18:03.720
which means for you listeners 13 days ago. Yeah, next episode we can talk about how the product hunt thing went because at first I was actually very excited about product hunt and like let's do it.

00:18:03.720 --> 00:18:13.559
I was reading all of these posts people talking about how they got the product of the day. And then the same time you open the Reddit and they say like product hunt is like I think they use the word scam.

00:18:13.559 --> 00:18:17.480
It's like it's designed to just drive more people to the platform. It's like not helpful. You

00:18:17.480 --> 00:18:22.359
know getting the real users from that unless you're building four startup founders. Yeah, I think for

00:18:22.359 --> 00:18:30.440
SaaS focused at in B or other founders, it works. Yeah, my point is that it's mixed. So we will go

00:18:30.440 --> 00:18:36.440
through that ourselves and see how it goes. I'm curious. All right. So the next topic we have is a

00:18:36.440 --> 00:18:41.640
request that we heard from a couple of users who were migrating from other apps. In the podcast

00:18:41.640 --> 00:18:47.160
world, there is a thing called opml. I think it has much to do with podcasting actually the actual

00:18:47.160 --> 00:18:54.039
abbreviation. Anyway, so opml is an XML file format. So you can export all of your podcasts in

00:18:54.039 --> 00:19:01.079
opml file and then other apps can import it. It's outline processor markup language. It's very

00:19:01.079 --> 00:19:11.638
generic basically. Anybody could use it. Yeah. But essentially like when you listen to podcasts, if you have been listening to podcasts for a while, you probably have 30, 40, 100 plus maybe

00:19:11.638 --> 00:19:17.878
podcasts that you have subscribed to over the years. And when you switch from one app to another,

00:19:17.878 --> 00:19:23.720
you don't want to go through manually and hunt those hundred and subscribe again. And we did not

00:19:23.720 --> 00:19:30.119
build this feature right away. So what is the feature? Like you can export from some apps, not all

00:19:30.119 --> 00:19:35.720
apps. I think support exporting it, right? Not all apps. Yeah. But some of the independent apps do.

00:19:35.720 --> 00:19:41.240
And also with Google podcast shutting down, they had the feature to export to opml. So people will

00:19:41.240 --> 00:20:00.920
ask us to import their opml files. We decided to do the what's called Wizard of Oz approach, where instead of building the feature to import the file, we created a runbook for doing it manually, which consists of parsing the file, getting the feeds from it, then doing some magic to find them

00:20:00.920 --> 00:20:13.160
in our catalog. If they're not in our catalog to import them and then using a backdoor sort of script kind of thing to update the production database with those records to subscribe the user

00:20:13.160 --> 00:20:19.720
to those podcasts that you want to be subscribed to. Yeah. Pretty much what you would want the app to do,

00:20:19.720 --> 00:20:25.880
but like you said, it's a lot of complex steps. And there's a lot of edge cases in the process. So

00:20:25.880 --> 00:20:31.480
we decided it's not worth building this whole thing right now. Let's semi-automate it, right? That's

00:20:31.480 --> 00:20:35.319
what we did. To be fair, we were not really thinking about the edge cases. We were like, we just

00:20:35.319 --> 00:20:45.480
don't have time for that. And then when we did the first import, I was premiering myself and I spent probably a good day, maybe two days going through that, figuring out the process, figuring out

00:20:45.480 --> 00:20:56.599
all of those edge cases. And I had this holy shit moment where like if you started building this, it would have taken us probably three or four times longer than our best estimates, just because of

00:20:56.599 --> 00:21:00.839
all of those edge cases. We're not even talking about the edge cases here, but we're like, okay,

00:21:00.839 --> 00:21:05.640
it'll probably be a while before we actually have this feature. But we did this for one user and

00:21:05.640 --> 00:21:11.880
another user sent us an email with their file. So the first user had I think 120 also podcasts,

00:21:11.880 --> 00:21:17.960
which we were able to import about 117. I think the second user had over 300. And they were like,

00:21:18.599 --> 00:21:25.400
wow, okay. So that was way beyond what we imagined a person would have in their library.

00:21:25.400 --> 00:21:29.960
The interesting thing though is after we did that, it was great. And then we get a

00:21:29.960 --> 00:21:35.640
regid thread from the first user. He's like, I can't follow any more podcasts. He was just getting an

00:21:35.640 --> 00:21:44.519
error. And the error wasn't super helpful. It was something like cannot follow a podcast. So we started looking and figured out that we actually had a limit of maximum 100 podcasts that you could

00:21:44.519 --> 00:21:49.319
follow. We did not hit this limit when we were importing the podcasts for the user because we

00:21:49.319 --> 00:21:53.720
get it through the back door. It did not have to go through those checks that the app goes through.

00:21:54.519 --> 00:21:58.680
So that's why it couldn't follow anything because it just wasn't working. It was hitting the

00:21:58.680 --> 00:22:04.359
limit. And we had to urgently change that limit and push a new version to the Google Play Store

00:22:04.359 --> 00:22:09.720
to fix that. We still do have a limit just because we have to protect ourselves too. Because there

00:22:09.720 --> 00:22:15.160
is cost associated with the podcast that people follow. And also there could be some technical issues,

00:22:15.160 --> 00:22:20.359
I guess. No, I think we went deeper and we figured out there are no scaling or technical

00:22:20.359 --> 00:22:25.240
concerns here. It's just a cost related. Yeah. And the reason why we initially set the 100

00:22:25.240 --> 00:22:30.200
limits, we were like, who's going to follow more than 100 podcasts? You and I, we listened to a

00:22:30.200 --> 00:22:37.960
lot of podcasts. I just looked at it right now. You're at 78. I'm at 58. But we do have a lot of

00:22:37.960 --> 00:22:46.119
users in the 40s and some users in the 60s and 70s. So I think initially we set this because of

00:22:46.119 --> 00:22:50.599
the feature that we decided not to build actually. It was something to do with home. We couldn't

00:22:50.599 --> 00:22:59.959
figure out how to make it computationally inexpensive to show a big number of podcasts. So we set it to 100 and we were like, if somebody has more, they will complain and then we'll figure out what to do.

00:23:00.599 --> 00:23:05.638
Yeah. And then somebody complained and we had to undo that decision. Increase the limit here.

00:23:05.638 --> 00:23:09.079
But yeah, it was an interesting exercise. The reason actually this approach is called

00:23:09.079 --> 00:23:13.878
Wizard of Oz. It kind of looks like the app is doing something on the background, but it's actually

00:23:13.878 --> 00:23:19.000
human behind the curtain that does the actual work. Well, we didn't automate it to this extent.

00:23:19.000 --> 00:23:28.599
People were sending it to us by email, but you were thinking about creating a form where people can submit the file and then it would actually look like something is happening automatically, but it was

00:23:28.599 --> 00:23:39.559
yeah, a person kind of. So the last topic of today's positioning, right? I think I mentioned in the last episode that I felt like I've not been reading enough nonfiction professional literature

00:23:39.559 --> 00:23:45.079
in the last couple of years. So I picked up a few marketing books. One of them was 22 laws of

00:23:45.079 --> 00:23:49.878
branding by Al Ries. I think the author is and there's the second author I don't remember.

00:23:50.599 --> 00:24:03.398
The same people who wrote 22 laws of marketing. I read maybe four or five chapters and the thing that really stuck to me there was what is the word that you want your customers to be associated

00:24:03.398 --> 00:24:08.519
in their mind with your product? And they were making an example of what do you think about when

00:24:08.519 --> 00:24:13.159
you hear Mercedes? Do you think about prestige? What do you think about when you hear BMW? It's like

00:24:13.160 --> 00:24:17.720
driving, right? Or maybe like a different version of that, but it looks like luxury. But also the book

00:24:17.720 --> 00:24:22.680
was written like 20 years ago. So maybe there were fewer luxury cars back then. So Mercedes was

00:24:22.680 --> 00:24:26.519
probably like the top of the pack, right? And they have multiple other examples which I was reading.

00:24:26.519 --> 00:24:38.680
I was like, yes, it makes sense. Like photocopying is zero. PCs were IBM's back in a day, right? Volvo is safety and it really got me thinking what should be that word that we should own or should try

00:24:38.680 --> 00:24:43.720
to own, right? Who is the app for? What is it known for? What is the first association that comes with it?

00:24:43.720 --> 00:24:48.920
I think only today I was walking with my kids on the street and it just came accidentally out of nowhere

00:24:48.920 --> 00:24:54.039
transcripts because that's what we started with. But we were sort of shying away from the like transcript

00:24:54.039 --> 00:24:58.920
is just a feature. We also have powerful playlists and there is a lot more kind of in the roadmap that

00:24:58.920 --> 00:25:03.079
transcript is just implementation in detail for. There also like other apps have transcripts to now.

00:25:03.079 --> 00:25:07.880
Like Apple has them, Spotify has them. I think it will become a table stakes feature at some point.

00:25:07.880 --> 00:25:26.200
Maybe within a couple of years in all podcast apps. But what we've been always thinking about and we'll continue to be really doubling down on is how do we make knowledge extraction from podcasts really, really powerful and transcripts is the core fundamental user experience and the underlying

00:25:26.200 --> 00:25:31.799
infrastructure for that. Knowledge extraction and retention and recall, I guess. Yeah.

00:25:31.799 --> 00:25:36.359
Yeah. Yeah. So it's like learning, knowledge learning, information. So these are the words that

00:25:36.359 --> 00:25:39.638
we want to be associated with, but they have to weigh great. But the transcript is something that

00:25:39.638 --> 00:25:44.759
user can see, user can touch quite literally, right? Because we have these things like bookmarking

00:25:44.759 --> 00:25:50.199
and the sharing quotes and we'll be making more and more powerful enhancements to those. So

00:25:50.199 --> 00:25:54.599
basically we want to be the app that is the best in transcripts. Just like if you think about the

00:25:54.599 --> 00:26:04.359
indie apps, you could think about overcast that's been the best in the audio enhancement. Because they like skip pauses, they have the smart speed control that you know it's really good and

00:26:04.359 --> 00:26:09.000
dynamically good features, right? Which other apps don't have and we could be kind of similar in

00:26:09.000 --> 00:26:22.839
a way, right? For transcripts. So transcripts should probably be the words that we are associated with and we should be using that more and more and as we are continuing to evolve features that are centered around the transcript, I think it will be more and more apparent as to how we are

00:26:22.839 --> 00:26:29.000
differentiated from other apps in this regard. Yeah. The path ahead I think is very long,

00:26:29.000 --> 00:26:36.759
but the power that the transcript provides is amazing. I think we'll get there now that we've

00:26:36.759 --> 00:26:42.839
implemented, we are about to release on iOS. We have launched on Android and I think we're very

00:26:42.839 --> 00:26:49.880
close to actually being feature complete as far as a podcast app goes. Then I think that's when we

00:26:49.880 --> 00:27:01.000
start doing the things that nobody else is doing. Well, we already have some features that nobody else is doing, but that's when we actually focus maybe 80% of our time on these kind of things

00:27:01.000 --> 00:27:18.920
that nobody else has. Actually, I'll tell you this now in iOS because we are part of the internal users group and I get paywall in the iOS app that prevents me from seeing the transcript and the problem with the sandbox testing of paywall is that you have to like click the button,

00:27:18.920 --> 00:27:28.279
you enter your Apple iG passport which you know is paying the ads by itself. I think it only works with 15 minutes and 15 minutes later it resets back to the paywall and I started listening to

00:27:28.279 --> 00:27:32.440
fewer podcasts because of that. It makes me anxious that if I want to highlight something,

00:27:32.440 --> 00:27:42.359
the transcript is not there and I may not be in the position to enter the password. And also one of my favorite podcasters, he's primary channels YouTube, but they were also cross posting to podcasts.

00:27:42.359 --> 00:27:47.480
I think I mentioned that show declaration that I was listening to for the half a year or so.

00:27:47.480 --> 00:27:52.440
The last two or three episodes, he only posted on YouTube and I even wrote him like, why they

00:27:52.440 --> 00:28:05.400
post on podcasts, but he has a large volume, he never responded to me. I almost didn't want to listen to those on YouTube even though I have YouTube premium, I can listen to this in the background, but not be enabled to highlight because he has a lot of interesting stuff there that I always

00:28:05.400 --> 00:28:10.680
highlight. It makes me anxious because like, how am I going to make notes of what I heard?

00:28:10.680 --> 00:28:16.680
So, I think that whole transcript thing is really, really powerful. We just need to find the

00:28:16.680 --> 00:28:21.160
way to talk about it, to market it, to the right people because that's nothing. I think we figured

00:28:21.160 --> 00:28:26.359
out our work. Now, the next thing is like, who is our tribe? Who is the tribe that you market to?

00:28:26.359 --> 00:28:29.880
I think it's least kind of knowledge junkies like us, but where do you find them?

00:28:30.519 --> 00:28:36.680
Yeah, I think reaching to those right people, marketing to them, that's the next, I think,

00:28:36.680 --> 00:28:48.519
growth area for us. I'm enjoying that journey because even if we were in a team building and app like this inside a big corporation, we wouldn't be getting all these other experiences of like,

00:28:48.519 --> 00:28:57.638
how to figure out which channel to market to and how do you market it and all that. So, I'm really enjoying all of these other things that I'm learning to.

00:28:57.638 --> 00:29:01.720
Even though you had like 15 users if you were a family, on a launch date.

00:29:03.240 --> 00:29:05.878
Or at least people will check it out immediately.

00:29:05.878 --> 00:29:10.680
Yeah, but then you probably get flooded with basic requests and you year away from your vision.

00:29:10.680 --> 00:29:20.920
You know what's interesting? One of our friends sent me a text probably a year ago with the announcement that Amazon Music had transcripts in on the same 2017, 2018, 2019.

00:29:20.920 --> 00:29:26.119
So like a while back. So they did a pilot with a few podcasts and I found those.

00:29:26.119 --> 00:29:30.039
And only those specific episodes of those specific podcasts had the transcript.

00:29:30.039 --> 00:29:34.599
So apparently they abandoned it for whatever reason, right? But we can really keep to the vision

00:29:34.599 --> 00:29:42.039
and really like keep going down that route. Right. Okay, so let's close this out. What are you

00:29:42.039 --> 00:29:46.359
listening to or reading? So I already mentioned the 22 laws of branding.

00:29:46.359 --> 00:29:49.720
In it's interesting. It's one of those books where I read five chapters and chapters are very short.

00:29:49.720 --> 00:29:54.759
They're like five six pages. And I feel like I've gotten out of this book what I need it. So I can stop.

00:29:54.759 --> 00:29:58.920
The other book that I'm actively reading is actually a Russian book called

00:29:58.920 --> 00:30:03.960
National Dazor, which is translated as Night Watch. I think it was translated into English.

00:30:03.960 --> 00:30:15.398
It's sort of fantasy, sci-fi-ish kind of story about vampires and magic and the interesting thing I noticed about this one is I think it was released in the late 90s and there was a movie.

00:30:15.960 --> 00:30:19.799
And the movie really made the whole series of books. It's a three book series.

00:30:19.799 --> 00:30:24.599
It just exploded after the movie in early 2000s. So I read it when I was 20 or so.

00:30:24.599 --> 00:30:28.839
And I barely remember it. So now I just picked it up again and I started reading it.

00:30:28.839 --> 00:30:32.679
What's interesting when you read the book half the life ago because I'm 14 now,

00:30:32.679 --> 00:30:38.759
I'm seeing just how much more I'm able to relate to the book. The whole thing happens in the like

00:30:38.759 --> 00:30:43.960
mid 90s Moscow. I wasn't from Moscow but I was part of that. I was in nature at that time.

00:30:43.960 --> 00:30:54.920
There could be like a phrase from five words. I can build a whole experience by just reading like somebody entering the whole way of the house and he sees X. And by just that X which is like one or two

00:30:54.920 --> 00:30:59.480
words, I'm able to reconstruct the whole picture which I'm not able to do with some of the American

00:30:59.480 --> 00:31:09.720
books because I didn't grow up in that environment which is pretty cool. Yeah. I'm halfway through the first book and I'm really enjoying it and I'm trying to get my son hooked on that but has some

00:31:09.720 --> 00:31:15.880
romantic stuff in there as well. And as far as teenagers go he will probably be like,

00:31:17.240 --> 00:31:21.319
and stop dreaming. The moment he sees and you can romantic notion in there.

00:31:21.319 --> 00:31:28.440
Are these the movies available outside of Russia or so? They might, I mean they're over 20 years old.

00:31:28.440 --> 00:31:33.079
The movies are not as impressive as the book. It's one of those movies where you watch the movie,

00:31:33.079 --> 00:31:37.878
the movie is good but then you read the book. You're like, okay, that movie is just nowhere near the

00:31:37.878 --> 00:31:47.960
book. The book has just so much more depth. It must have been translated. I think it's one of the sort of seminal Russian contemporary books and I've never been listening to podcasts much recently.

00:31:47.960 --> 00:31:52.439
If you got a lunch on App Store and remove that freaking paywall, it's for me to get back into that.

00:31:53.559 --> 00:31:59.799
Okay. 4b. So I have some family visiting and some friends. Actually the same friends that were

00:31:59.799 --> 00:32:05.559
here when you were here last year just around this time. So they're back here. Even when we are in

00:32:05.559 --> 00:32:10.680
the car, I'm not listening to podcasts because I don't want to bore them with my podcasts. Other

00:32:10.680 --> 00:32:16.279
than that there's of course US Open going on this year. Farmula one is really exciting because there's

00:32:16.279 --> 00:32:21.559
like at least three or four different people who can win the whole thing. Whereas the last few years

00:32:21.559 --> 00:32:27.159
it's been just one person who is one everything. But outside of that the Dave Girl book by the way

00:32:27.160 --> 00:32:32.759
quick callback. So there is music in that book in the Dave Girl book but it's their own music.

00:32:32.759 --> 00:32:43.559
For the other songs like ACDC or Metallica he talks about he just says like okay we can't afford to pay royalties for every listen of this audiobook back to them so that's why they didn't

00:32:43.559 --> 00:32:49.640
include it. So I'm listening to that. I'm slowly progressing to that Dave Girl book. So far so good.

00:32:49.640 --> 00:33:00.519
I love it. I think the podcast I will mention today. So if you like to hear about app development or the struggles of getting big and doing a rewrite and all that then I think this was great

00:33:00.519 --> 00:33:13.319
because Marco who's the creator of the very popular overcast app. I think at some point it was maybe even now it's the biggest podcast app on iOS outside of Apple and Spotify probably.

00:33:13.319 --> 00:33:18.519
It's one of the big ones. It's a single developer written. It's been over 10 years. It had become

00:33:18.519 --> 00:33:31.079
hard for him to manage that app. He decided to do a whole big rewrite with a new app and that's been a journey because he decided to change some things, cut some features and people got really

00:33:31.079 --> 00:33:39.240
mad about that. Marco is in two podcasts. One is called accidental tech podcast and episode 596

00:33:39.240 --> 00:33:49.799
of it called a new foundation for progress. That's the one just before the launch they did talking about the launch and what he's looking forward to and all that and then there was a lot of

00:33:49.799 --> 00:34:03.720
backlash after the launch. So in his other podcast called under the radar episode 300 the aftermath that's when they talk about what were the problems and what is he doing about it and how does he

00:34:03.720 --> 00:34:21.880
think about all that. These are both from the creator himself and other people but then I've also found this really interesting another podcast called comfort zone and they did an episode called I like to read my podcast recently in the last two weeks and this is a pretty independent look at

00:34:21.880 --> 00:34:28.119
the rewrite and what worked, what didn't work and how they think about it as a user so I like that

00:34:28.119 --> 00:34:35.079
episode quite a lot too. Yeah, I started listening to the ATP podcast too. That's like an hour and a

00:34:35.079 --> 00:34:46.199
half long and in the middle is a whole segment about very deep technical stuff. So yeah, I've never developed in Swift but it was interesting to listen about the problems he was facing and the

00:34:46.199 --> 00:34:51.239
trade-offs he was making between the user experience and the technical simplicity. It was

00:34:51.239 --> 00:34:55.798
interesting trade-offs. It also gives us a peak inside the mind of our competitor which is

00:34:55.798 --> 00:35:03.639
pretty cool. Competitive research. Okay, so let's close this out. So yeah, the app is on Android right

00:35:03.639 --> 00:35:11.559
now and beta on iOS launching very soon on the app store. You can find the links at metacast.app

00:35:11.559 --> 00:35:17.878
that's our website. There also you can find this podcast and our newsletter that Ilya does

00:35:17.878 --> 00:35:23.079
give it a listen. If you have any feedback or thoughts or features or requests for us and it

00:35:23.079 --> 00:35:39.960
are way at hello at metacast.app. Hello at metacast.com. That's for the podcast. Yeah, for the app it's my feedback at metacast.app and we start publishing blog posts on our blog at metacast.app and we

00:35:39.960 --> 00:35:46.599
also cross-post them on our newsletter at metacast.com. I think what I need for previous previous one.

00:35:46.599 --> 00:35:52.599
Basically I took the discussion that we had and then wrote in a say about the topics that we

00:35:52.599 --> 00:35:56.599
discussed specifically about pricing. I think pricing was the main topic that we covered in

00:35:56.599 --> 00:36:01.719
episode 59. It covers more than the podcast and also it goes into more depth. It has links.

00:36:01.719 --> 00:36:08.278
I generally think that our newsletter is getting better in that regard and it's worth to be

00:36:08.278 --> 00:36:20.679
subscribed to at metacast.com. If you're a person who likes to read rather than listen then use the newsletter more but then I'm wondering why are you here 35 minutes into this episode

00:36:20.679 --> 00:36:25.159
listening to it if you don't like listening. Exactly. I think it's for par with social connection

00:36:25.159 --> 00:36:29.159
which actually didn't know if we ever covered this term in the podcast. I remember it was Justin

00:36:29.159 --> 00:36:33.878
Jackson on one of our interviews. He mentioned this term. Then I had to google it because I didn't

00:36:33.878 --> 00:36:38.039
know what it meant. Par social connection is a connection that is developed between people

00:36:38.039 --> 00:36:44.358
is like a one-sided connection. If somebody is listening to us talk they develop bond with us

00:36:44.358 --> 00:36:51.318
but we may or may not know them. We are feeling like they know us really well whereas we may not

00:36:51.318 --> 00:36:56.440
be aware of their existence. I definitely have that feeling towards some of the podcasters that I

00:36:56.440 --> 00:37:06.119
regularly listen to. That's what's called par social connection. We've met a few of such people who listen to our podcast and all that and those have been wonderful to me in person.

00:37:06.119 --> 00:37:21.880
You remember when we met one of those people who also seemed like it feels so weird because I know so much about you and like you're just the only for the first time but they only know the voice so they only know us as like a voice plus some images probably conceived in their brain.

00:37:21.880 --> 00:37:26.280
They make a watch the YouTube videos. That's true. We do post on YouTube. I should have

00:37:26.280 --> 00:37:39.000
done my hair. My wife was just meeting someone she calls an Instagram. We never discussed it before and she told me it just felt so weird that she knows so much about that person and to that person she is just like a random person who wrote to her on Instagram and they decided to meet.

00:37:39.000 --> 00:37:48.440
Yeah. Okay so till the next time stranger on the audio waves will or you'll hear us next time

00:37:48.440 --> 00:37:59.000
I guess we will hear you unless you send us an email at helloatmetacastpodcast.com. Yes. Bye. Bye.

