Subscribe
Share
Share
Embed
A lot has changed since the last time Christian Selig was on the show. We talk Apollo post-morten, Third-Party APIs and of course Pixel Pals!
Guest
Related Episodes
- Now You Know What I'm Doing This Summer
- My Taylor Deep Dish Swift Heroes World Tour
- Pizza, Playpen, and Fastlane Funding with Josh Holtz
- Apples, Glasses, and HAL, Oh My!
- Elephant in the Room
- Third-Party APIs with Christian Selig
Related Links
Social Media
Credits
Music from https://filmmusic.io
"Blippy Trance" by Kevin MacLeod (https://incompetech.com)
License: CC BY (http://creativecommons.org/licenses/by/4.0/)
"Blippy Trance" by Kevin MacLeod (https://incompetech.com)
License: CC BY (http://creativecommons.org/licenses/by/4.0/)
- (00:00) - Origin Story
- (05:07) - What Happened
- (17:03) - Apollo Post-Mortem
- (19:41) - Third-Party APIs Revisited
- (23:40) - ChatGPT
- (27:24) - On Yearly Subscriptions
- (31:16) - Lost Features of Apollo
- (34:08) - Pixel Pals and SwiftUI
- Bertram Eber
- Edward Sanchez
- Satoshi Mitsumori
- Steven Lipton
Creators & Guests
What is Empower Apps?
An exploration of Apple business news and technology. We talk about how businesses can use new technology to empower their business and employees, from Leo Dion, founder of BrightDigit.
Leo Dion (host): Welcome to
another episode and EmpowerApps.
I'm your host Leo Dion today.
I'm joined once again by Christian.
See like Christian.
Thank you so much for coming on.
Christian Selig (guest):
Pleasure is all mine.
It's great to be back.
Leo Dion (host): This is
your second appearance.
So I'm glad to have you back.
We met in April back in Chicago and now
having you on quite a bit has changed
since then, which we'll get into.
But I'll let you go ahead and I'll
let you go ahead and introduce
yourself for people who don't know.
Christian Selig (guest): Yeah, for sure.
So I'm an iOS developer
from Eastern Canada.
I've been doing the whole
indie developer thing, for
almost about 10 years now ever
since I graduated university.
I started working on kind
of a Reddit app to scratch
my own itch called Apollo.
And I worked on that up
until a few months ago.
And that's my, claim to fame.
And on the side of that, I also
work on a little virtual pets app
called Pixel Pals, which has been
a lot of fun to work on as well.
Leo Dion (host): And we'll definitely
be talking about Pixel Pals in a
bit, but I think today, like the
Apollo story has been in the press.
We're like, what, three, two
months after everything blew over.
Christian Selig (guest):
Something like that,
Leo Dion (host): Yeah, I wanted
to first just talk about,
where'd you get the idea?
Let's look, cause I've had you on,
we've talked about this, but just
for folks who don't know, where
did you get the idea of Apollo?
And yeah, the origin story of it.
Christian Selig (guest): Yeah, yeah
I wish I had a super original story
for kind of how I come up with
apps, but most of the time it's just
scratching my own itch, like I'm
just there's something out there
that's missing that I'm looking for.
Maybe it's something I, like I
use a lot, like I used Reddit
a lot and you just want to.
When you're using something a
lot, you just want to have the
best experience possible doing it.
And as a developer, you have the
means to create that reality.
So for me, a lot of the stuff I
create is just I'm not completely 100%
pleased with everything that's out
there, even if they're like 99% great.
And it's just that little bit extra.
I want to maybe contribute a little bit
to Yeah, and that's where I get started,
and that was how Apollo started.
There's some phenomenal reddit
apps out at the time, like there's
Alien Blue is still in its heyday.
And yeah, don't get me wrong, these
apps are phenomenal, but it's just one
of those things where software is so
personal, and everyone has a different
approach to how they use things.
And for me, I wanted something just
a little different and something
that felt a little bit more like
if Apple themselves built it.
Like a really iOS first and foremost
Reddit experience, and that was what
I set out to build with Apollo, and
thankfully that resonated with quite a
few people throughout, like the initial
posts on Reddit where I was previewing
it, and it took on a life of its own and
became my full time job for a long time.
Leo Dion (host): This in the best
way, Apollo is a very addictive app.
It was good.
It was like, yeah, it was the best.
Did you win a design awards and
stuff besides being plugged as
a future app for the vision pro?
What other design awards did you win?
Christian Selig (guest): I don't know
if I actually got any award awards.
I got a few at least from Apple,
I'll say a few nice publications
gave me some accolades over the
years, which is really nice.
And, of course it was always really
cool to be in some of the Apple
keynotes being mentioned there.
But yeah, no, no actual Apple design
award yet, so we'll stay tuned for that.
Leo Dion (host): So the
other question that I wanted
to cover was like, besides.
The, what happened this
year what was your biggest
challenge developing the app?
Christian Selig (guest): It's, honestly
for me, it's just like designing stuff.
Being someone who does the design
work and the programming work there's
such different disciplines in that.
When you're programming something,
there's typically like a finite point.
Or you can put a bow on it and
ship it out the door when there's,
no bugs everything's scrolling
well, it's performant everything's
acting as it should but when you're
deciding things it's much more of a
a vague kind of nebulous concept and
reconciling those two things, I think
still gives me trouble to this day
where there's a lot of times where
I'll be trying to design something
and I just can't quite nail it down.
And in my head, there's like this
going to be this point where, it
just clicks like almost like a math
equation or something in programming
where I'm like, okay, that's it.
That has to be it.
But the reality is that never comes.
There's always compromises with design.
Like one of the big ones
with Apollo that I really
struggled with was having the.
Some people call it sloppy swiping
where you can swipe from like anywhere
on the screen to go back and forth.
And that's really comfortable because
you got these big ass phones and
you're holding them in one hand.
You're maybe carrying groceries
or a toddler in the other.
Being able to grab anywhere
on the screen to go back a
page, it's really comfortable.
But at the same time in doing so you
lose the ability to do any like cool
gestures on like a specific post
because that becomes the back gesture.
So it's little stuff like that
where I remember sitting with
that for a while and being like.
Gosh, how do you reconcile that?
And the answer, stuff like
that I've learned is you can't.
You Best case scenario, you make
an educated guess, and if people
push back and want the other thing,
you just offer it as a setting.
And that tends to be the thing that
works the best in my experience.
But yeah, for me, honestly,
it's been, design is always the
trickiest thing to nail down.
And takes me the longest
time, and I don't think that's
going away anytime soon.
Leo Dion (host): I definitely want
to ask about that later when we talk
a bit about your work with SwiftUI.
But yeah, because I remember in
the last episode, you talked quite
a bit about design notebooks and
using those and those being a way to
To just get your ideas started and
but we'll talk about that in a bit.
Cause I definitely, I
have a few questions about
Christian Selig (guest): For sure.
Leo Dion (host): All right,
so let's just get into it.
For those who don't know, I
have been living under a rock.
What happened with Apollo and reddit?
Christian Selig (guest):
Historically, I think the Reddit
API was introduced in about 2008.
And so you've, you've had this API,
which is just a way for services like
apps to communicate with Reddit and
get like a two way communication street
where you can say Hey Reddit, could
you give me the post in the subreddit,
or the, or the comments on this
post, or could you upvote this post?
And it was this, this interface
for talking with Reddit.
And that's what API stands for,
Application Programming Interface.
And it was really helpful because for
a long time Again, I think most of
Reddit's history, they didn't have
a first party official Reddit app.
So they relied on third party apps.
If you wanted to browse Reddit on your
iPhone beyond just the website, they
relied on these third party developers
building these application experiences
for iOS or Android or Windows Phone,
what have you, using this API to build
these experiences that people loved.
Eventually, Reddit came and
released their own first party
app after buying up one of the...
Original third party apps, Alien Blue
and it's been smooth sailing since then.
And but the thing is, since it's been
a mutually beneficial arrangement
where Third party developers are
building these Reddit experiences for
users maybe if you're not jiving with
the first party app, you have this
alternate option where you can still
keep browsing Reddit, or, historically,
if there wasn't even an official app
these third party apps filled this
really important space there's this
mutual the API's free, but it gives
both parties some really nice benefits.
And...
As Reddit started tightening their
bootstraps a little bit, it became clear
that they were like, maybe looking to
maybe monetize the API a little bit,
and they reached out to developers.
I want to say, it is a little far
removed now, I want to say it was
late April that they reached out.
They made a post and said we're
looking to change the API a little bit.
And then they started calling
developers like myself with third
party apps later and said, Hey, we're
looking to monetize the Reddit API.
So insofar as like you guys generate
like a fair bit of server traffic and
we're not necessarily seeing like the ad
revenue from when users use your apps.
So we'd like, a slice of that
pie and you to pay your fair way.
And it was honestly like great.
That's that sounds totally reasonable
on paper and it's one of those things
where It really had the potential to be
awesome for both parties because there's
been extra features added on reddit
like reddit chat this instant messaging
thing and stuff like that where reddit
hasn't added that to the api So there
was initially a lot of excitement about
oh, like maybe if we enter in this more
formal paid arrangement with reddit
But we'll also get these benefits.
So great.
Leo Dion (host): the relationship
could be more stable.
Christian Selig (guest): Exactly.
Yeah, exactly.
We both know where both parties stand.
But it's like anything, like the
proofs in the pudding what is
the actual price going to be?
That kind of dictates everything.
And unfortunately for whatever
reason, they announced this big
pricing change without having
any pricing which was confusing.
But they ultimately said we'll get it
in a few weeks and a few weeks went by
and then they ended up, I think after
two months they finally got the pricing
out and it was like, eye wateringly
high to the extent that during the
discussions over those two months, they
were saying stuff like, we don't want to
pull a Twitter because Twitter famously
killed third party apps a few months
earlier and then made the API prices.
Astronomically high to the extent
that it just suffocated everybody
but Twitter themselves and Reddit was
like, no, no, we don't want to do that.
We're not going to pull a Twitter,
we're not going to pull a Nelon.
And then the pricing they came
out with was like remarkably
similar to Twitter's pricing.
So it was one of those things
where people were saying, okay,
like this is not jiving with
any of your previous statements.
Like this kind of
doesn't make any sense.
This isn't what you were saying.
And they also on top of that
saying, okay, and you have 30 days
from now before you start getting
charged for this or start incurring.
fees for using the API.
So you were hit with this one, two
punch of this, like astronomically
high price and like to put that
to a number, like Apollo's pricing
would be 20 million a year.
And when the previous year it costs 0.
So it was this astronomically
high price that they were.
Like saying developers have to pay or
we'll turn off your API access and you
have 30 days To make it happen And it
was just like it quickly it quickly
like I think every developer I talked
to Since we all were talking Of course
over the course of all this we're trying
to make it very clear to read it like
this is not tenable at All like we
don't have Millions tens of millions
of dollars at our disposal just to
throw at this especially with 30 days
notice and this is just gonna sink
us and Yeah, and that's where things
deteriorated from there, where it got
to the point that I don't think Reddit
wanted to listen anymore, and they were
adamant, and this is what it's going to
be and it fundamentally, I think, just
killed the entire third party ecosystem
as happened with Twitter as well.
Yeah, and it's weird because funnily
enough I also talked to them in the
January that stuff started in April, but
the January I was also talking to them,
and they were, like, very adamant that
hey, we have no plans to change the API
this year everything's good if anything,
it'll be in coming years, and it'll
be with positive changes, so it was
this very quick turn of events so that
they just started like absolutely like
hammering through these changes as fast
as humanly possible with little regard
for like timelines or, or talking to
developers or, or really anything that
would indicate forethought and reason.
Leo Dion (host): Do you, and two I'll
just mention they basically, try to
character assassinate you and make up
Christian Selig (guest): Yes,
Leo Dion (host): stuff and, yeah it
seems to me like go ahead, I'll let you,
Christian Selig (guest): No, no,
I was just gonna say, yeah, it got
weirdly personal at points where it
was like like we were having these
back and forths on phone calls a lot.
So we were like talking and
getting to know each other.
And I think there was this initial
I just don't think fundamentally
they had done their due diligence
in planning this change.
So I think there was when developers
started talking about these numbers,
like 20 million a year, like 30
days notice like posting these like
factual statements on reddit to
contextualize the situation There was
like immediately a lot of blowback
toward the administration on reddit
and I think they were caught off guard.
Just because For some reason they
expected us to go over and I think
at that point especially some of the
upper management the ceo started to
take things like personally rather than
like taking a step back and viewing
it objectively and it got to points
where yeah there were there was like
discussions where I was saying like Hey,
if Apollo's costing you this amount,
this nebulous 20 million a year just
acquire it like you did with Alien
Blue or whatever for 10 million a year.
And they somehow interpreted
that as like a hush payment
Mafia style or something.
And I very quickly squashed that and
said, no that's not the intent at all.
And they apologized
for misunderstanding.
And you think that's that.
Leo Dion (host):
Canadian hush money, yes
Christian Selig (guest): yeah, exactly.
We're very well known for shaking
down billionaire corporations and so
I they apologized for misunderstanding
and I thought that was that.
And then you.
here through like on the moderator calls
like you a moderator friend messages you
and saying yeah the CEO's going on calls
saying you were like blackmailing them
for ten million dollars and stuff And
you're just like what on earth and like
and like thankfully like Canada's a one
party consent state for recording phone
calls So I was like just and not out
of any like nefarious intent, but just
Leo Dion (host): No, just record
keeping, you wanna, you forget stuff.
No, I get
Christian Selig (guest): Yeah,
we had probably 10 calls over the
course of those like several months.
So just being able to go back
and be like, okay, so this was
what was said, not this, like
you just lose track of things.
So I was saying like, and I read
that and I was like, Oh my gosh
did I have a brain aneurysm and
I somehow did threaten them?
Let me go back and listen to this.
And it was like, no, it was
exactly how I remember they
apologized for misunderstanding
the entire conversation.
So it was one of the things where you
post that and say Hey, look, they're
trying to say some very serious
accusations that are just, objectively
false based on a recording I have.
And I think that was the point
where they said Oh, okay.
Like now you want to go to war,
which is it was also very confusing
to me because it was a thing.
If the roles were reversed and the CEO
and I said like the CEO was blackmailing
me and said I'll allow Apollo to stay
alive for a million dollars and I posted
that and he had a recording saying like
the complete opposite, like I would very
much expect him to vindicate himself
and show the reality of the situation.
Like you have to defend yourself
when you're pushing that corner,
especially with such like a.
a massive character assassination.
So I was confused what they thought
like the, what the recourse was.
Like I was, it was very peculiar
and it felt like a very, like the
whole thing felt very unprofessional,
quite frankly, for a billion dollar
corporation to be engaging in.
Leo Dion (host): Assuming they are
a billion dollar corporation, right?
Christian Selig (guest): Yeah, maybe
not in liquid money, but like in terms
of valuations, I think they're still
at like several billion dollars and,
or at least they were at one point.
Leo Dion (host): yeah okay, so I'm
just gonna ask what I, what kind of
the feeling that I've gotten from the
whole story was that A, they never had
a plan, and B, the head didn't know
the foot didn't know what the head was
wanting, and the head, being the CEO
doesn't know oh, this is what we need to
do for the, Public a public IPO, right?
Oh, we got to do this.
We got to do that.
And they just keep changing their mind.
And who cares about Christian Siegel
eggs, Apollo app, even though like
people love it and moderators love
it, like we'll do whatever it takes.
And if he says something that sounds
like hush money there we can get them.
And then, I feel like they were just
like trying to figure out a way to
Christian Selig (guest): It was.
Yeah, it was very slapped together.
Leo Dion (host): but they
didn't like, they didn't care.
And it was like, yeah.
Christian Selig (guest): all felt
very much like somebody woke up one
day and saw like an expense like
an earnings report and said Oh my
God, we're coming way under, we need
to like, make up some money here.
Who do we squeeze?
And then from that conversation they,
within two hours, they had to play.
I'm obviously just speculating here,
but it felt very much like they were
just like, flying by the seat of their
pants and coming up with a plan very
impromptu because honestly, the amount
of times on phone calls where they
said something like, Oh, I'm not sure.
Let me get back to you.
Or to very fundamental questions
like things like, oh there's all
these APIs, Reddit apps don't have
access to, are those on the table?
And it would be like,
oh, that's a thing?
There aren't?
Oh, let me look into it.
And it was kind of like, if
you're talking about APIs,
like, how do you not know that?
Or they'd say do you know how big,
is compared to the first party app.
And I said I don't only, you would
know that, like I could theorize.
And they said, Oh, we don't know.
I was asking you.
And it was kind of like, like, why
wouldn't you look up that, prior to
the call and there was all these things
where it was just, it felt very much
Oh my god, like I could go on it.
They ended up removing ads.
Apollo doesn't have ads, but they
ended up blocking third party apps
from as far as a guideline goes.
You no longer can
display ads in your app.
Leo Dion (host): Remember that.
That was a big deal with Android, folks,
Christian Selig (guest): Exactly,
because so many of them did that.
And then the CEO kind of came
out and said Oh, I didn't
know that was such a big deal.
And it was like, how did you not
do the due diligence to realize
that's how so many of the Android
ecosystem, like that was how they
Leo Dion (host): they made money,
Christian Selig (guest): That's
how they kept the lights on.
Yeah, it was just, it was bizarre how
many things like that, that you were
just like, wouldn't that be, And Oh,
and the worst thing that almost bothered
me the most was that they, their iOS
app is like very poorly annotated
for like voiceover use from people
with like low vision or blindness.
And it took until them realizing that
the third party apps were doing a
really good job of annotating their
apps and making sure like Apollo's has
a very large blind community or had a
very large blind community using it.
And it took them until they like killed
all these apps to go, Oh my God, our
app doesn't work for blind people.
Blind people exist and like the whole
blind community was angry at them.
And it was just like these little
things where it was like, how did
you not talk to a group or kind of
formulate a plan here before executing
this because it just it felt very
11th hour trying to throw together a
homework assignment before the teacher
notices you didn't do the homework
the night before like It was bizarre.
Leo Dion (host): Do they how much
of a development staff do they have?
I would assume
Christian Selig (guest):
think it's I think it's pretty
substantial, but it's just like
Leo Dion (host): Or they just
don't communicate, because
Christian Selig (guest): yeah, or it's
Leo Dion (host): staff?
Yeah,
Christian Selig (guest): I don't
know or it's just too many cooks in
the kitchen or kind of the message
gets lost or what is it but like
they certainly have a substantial
impressive amount of engineers and
like I've met some of them at WWDC
like the engineers and the people who
work on like the apps and whatnot are
like super nice very smart people.
I think it's just one of those
things where like upper management
just is I think they're just like
siloed away and not necessarily
like talking to the right people
Leo Dion (host): How do
we get the public IPO?
What do we do about the public IPO?
I don't care about anything
but the public IPO.
Christian Selig (guest): Yeah,
that's the impression I got.
Leo Dion (host): Yeah
would you do it all again?
Christian Selig (guest): Oh my gosh.
Yeah.
In a heartbeat.
Like it was, oh no, it was Like
I would do it again tomorrow.
Like it was so much
fun building that app.
Like even today, like the Reddit
community is so phenomenal.
I don't hold the higher up management
and maybe the highest regards, but
like the community itself at its core
is just, it's a phenomenal community.
Very, has a lot of there's no more.
diverse area of the internet in terms
of interests where you can go and find a
community to talk about, like basketball
shoes versus the best ice skates or
there's just something for everything.
And then the community is so phenomenal
there and getting to work alongside them
for so many years, building something
that I had so much fun building.
Oh my God.
Yeah.
And a heartbeat.
Leo Dion (host): Yeah, and I
mean you lived off of it for
like at least almost a decade so
Christian Selig (guest):
yeah, it was, yeah,
Pretty much my everything.
Leo Dion (host): I think How many
years did you get out of it eight?
Christian Selig (guest): It was,
I think seven were monetized.
Like the first two years, it
was in like a beta where it was
just free for everyone to do.
I figured out all the bugs
and got more feedback on what
was working and what wasn't.
But after about two years,
yeah, the initial 1.
0 came out.
Yeah, it was, all in all, it was about
nine years of like public work on it.
Leo Dion (host): That's a decent run for
Christian Selig (guest): Yeah, no,
I was, like I was saying I think
I said to someone else it's almost
hard to be disappointed about this
because who gets to build a quite
successful app for nine years alongside
hundreds of thousands of really
awesome supporters who gave incredible
feedback and are really passionate
about the app, and that's something to
be, like, feel very fortunate about,
not oh, God, now it's gone it's sad,
of course, but it's holy shit not
many people get that opportunity,
and I do, that's not lost on me.
Leo Dion (host): oh,
you know could be worse.
You could be having a bill
for 20 million dollars in
Christian Selig (guest): Exactly.
Exactly.
So
Leo Dion (host): What so
Christian Selig (guest): side.
Yeah.
Leo Dion (host): One of the questions I
got on Macedon was people were wondering
about Lemmy what any hope for Apollo for
Lemmy and what would be your deciding
factor for it or how do you come up
with that decision that you have,
Christian Selig (guest): that's fair.
It's a tricky decision for sure.
It's one of those things where I
think at this stage I'm good and now
that I've been like removed from it
for a month, I'm excited to maybe
try out a few different things.
I like that Apollo went out on top.
And Frankensteining it together for
something else isn't something I'm
super excited to do at the moment.
And I think it's in the same way
that Alien Blue kind of maybe
passed the torch to other apps.
I think maybe it's like Generation
3 now where I'm curious to see
what other developers do with them.
LEMME apps and whatnot and how the
community grows and all these kind
of open, federated communities do.
Because I think it's like a really
exciting time on the internet
to be an internet user and have
all these services popping up
that are offering such a thing.
Leo Dion (host): I want to ask
this cause the last episode we
talked about third party APIs.
What's like going into
any new app at this point?
What's your opinion of using anybody's
third party API, and how would you be
able to be like, judging whether, yeah,
this is stable, or no, these people are
gonna screw me, or oh, it's federated
I don't have to worry about anything,
like, how do you make those decisions?
Christian Selig (guest):
That's a great question.
I think at this stage, I'm like a
little burnt, so I think I'm almost
overly cautious, so I might not
be the best person to ask that,
because I feel a little cagey, but
Leo Dion (host): no, no,
Christian Selig (guest): one of those
things where it's like, if, if it's
clear that both parties benefit.
I think like very much I
thought that was the impression
that, that Reddit understood.
And I think it was unfortunately
lost on them that this was something
that really benefited like you,
especially power users to have.
And so there was this kind of
relationship where it's a give and take,
and both parties are benefiting and
the APIs exists for beneficial reasons.
And I think building for services
like that I think are a relatively
safe bet, at least for a foreseeable
future until maybe something
changes as it did with Apollo.
But even if you can have nine
years of good times building for
a really awesome community and
building something fun go for it.
But for me, if I was to build something
today, I think it would have to be,
and it was using an API in such a core
capacity as the core of the app, I
think it would have to be something
that was federated in some capacity.
So that there wouldn't be, one
central power that could just
ruin everything for everybody.
But if it, if, but if it was like a
side thing, like if I wanted to have
I don't know, YouTube thumbnails
like Apollo used the YouTube API
to show thumbnails for YouTube
videos when people post them.
That feels pretty safe to use
because if that ceases to exist
you just lose a thumbnail.
So stuff like that, I'm
less worried about but stuff
that's at the core of the app.
Yeah.
I think I'd be a little cagey
about it at this stage, just
because I am so recently burnt.
Leo Dion (host): I think like
something federated that's
going to be open no matter what.
Since the Elon apocalypse, we've
seen a variety of like social
media apps out there, none of them.
And that's probably wasn't
the case 10 years ago.
None of them have third party APIs.
And so like that, that definitely to
me, like anything that's of social
media app is pretty much even if
they did offer a third party API, I
Christian Selig (guest): Oh,
Leo Dion (host): Rely on it at this
point because they want the ownership of
the whole experience ads and everything
and tracking and all that stuff.
So I
Christian Selig (guest): yeah, 100%.
Leo Dion (host): that,
Christian Selig (guest): The internet
was much more innocent and young
and exciting in a way that like it
was people were excited to build and
reddit Especially was a very like and
to this day is like it's a very like
techie nerdy user base and I think
that played a lot to their Having
an api was that they were tinkerers
themselves and they liked offering
Leo Dion (host): yeah.
I remember you mentioned
that in the last episode.
Christian Selig (guest): that yeah,
Leo Dion (host): yeah.
It's just much more of a techie crowd.
So they were like more
comfortable with that.
Yeah.
And if there's more of like the
developer social media thing that I
could see, maybe that having an API,
Christian Selig (guest): yeah,
but i'd be pretty careful
Leo Dion (host): yeah.
But the other thing is what you'll see
companies do well now, a, they offer
the social media app like this, like
threads is not have a web app iPhone.
So they like go whole hog and that's
what they focus their development on.
But then the other thing is like,
unless it's something that you can,
it's an API that you can somehow
no longer need and do it yourself.
I don't know what that means
exactly, but let's say you
have some API that does.
transcription and you
rely on a third party API.
Eventually you could just be like,
screw it, I could just use machine
learning locally and just do it
Christian Selig (guest):
Or a different provider.
Leo Dion (host): yeah, that's
always a possibility as well when
it comes to a third party API.
Christian Selig (guest): Oh, for sure.
I think you just have to it would be
sad to live in an age of the internet
where nobody relies on anyone else
or talks to anyone else just because
everyone's so afraid of each other.
So I think there's, it's
understandable to have some sort
of skepticism, but it's good
that's the beauty of the internet.
Everything talks to each other
and I think relying on that to a
certain extent is totally fair.
Just make your choices calculated.
Leo Dion (host): Yes.
Yes.
Put.
I wanted to ask you this.
What do you think of chat GPT
Christian Selig (guest):
Honestly, I like, it's such
a it feels seeing the Wright
Leo Dion (host): or a open
Christian Selig (guest): and then Yeah.
Yeah.
And then being like, what do you think
of airplanes like, it, it's such a,
it's so young that I almost don't.
Have a fully formed thought about
what it means or what it can do.
I will say some of the stuff it can do
and some of the stuff like mid journey
can do is just, is mind boggling
and fundamentally really helpful.
Like the other day I was like trying
to do something where I was like
the app store was returning like
country codes and I was like, Oh,
it'd be really nice to have that as
just like the country flags emoji.
And I was like, started going
down the list and just like
manually converting them.
And I looked and there was like 150
or something and I was like, surely
like chat, GPD can just do this.
And yeah, sure.
No, but just.
Three seconds later, it had the
entire like Swift dictionary
mapped and stuff like that.
I was like that's really cool.
And I don't feel like that's threatening
me as a developer in the same way,
like a nail gun didn't threaten like a
carpenter who used a hammer typically.
Like it's I'm like, I it's a really
weird thing for me to like, Speculate
on because I think it's really cool.
It can potentially be really scary.
I don't think I'm as quite as oh
my God, like I have six months
left of programming before my
job's replaced as like waving
the flag like some people are.
But I 100% do see the fear and
how powerful they're getting
and how that can be scary.
So it's definitely going to be an
interesting few years in terms of
where stuff like that and deep fakes
and mid journey and just all this
AI stuff is going because it really
feels three years ago, this was
like not a topic we were having.
And now suddenly it's all
anyone is talking about.
Leo Dion (host): I think it's definitely
a good replacement for a spreadsheet.
Like I would have done the country code
stuff in a spreadsheet, but like any
like advanced code, I usually run into
a brick wall or like podcast titles.
Honestly, I've used that
a few times here and they
Christian Selig (guest): Oh yeah.
Leo Dion (host): are very
formulaic after a while.
Christian Selig (guest): And it's,
yeah, it's just those formulaic
ones that I think that especially
that there's almost like a
lot of repetitiveness with it.
I think that's where I'm really
excited to have it as a tool.
Because one of the dumb things I
did in Apollo was I had separatists,
for whatever reason, when you
create them, you're stuck with
the capitalization you stick with.
Created the separate ask Reddit
and you capitalize the a and the R.
So it looked pretty good.
Bob's your uncle.
If you were dumb like me and just
wrote Apollo app and all lowercase
that's how it displays in the app
and you can never update that.
So one of the things that drove
me crazy with Apollo was you'd be
scrolling through the feed and you'd
see this, like everything lowercase.
And sometimes I was having trouble,
like maybe I'm just dumb, but
knowing like where each word was.
So it's I started the painstaking
process of I just had a P list
where I was like, okay, ask Reddit,
like all lowercase maps too.
And so whenever it saw a subreddit
in the app, it would do a find
and replace for the properly
human capitalized version.
And gosh, I probably did this six
years ago and it was like literally
thousands of subreddits that I just
did like a hundred a night for a month.
And it was honestly fun.
It was just watching TV and
just slowly going through it.
And I didn't mind it too much, but
I went like the other, maybe like
a year ago, I went on chat GPT and
just pasted like a hundred of the
raw ones in, and it just absolutely.
Every single word capitalized properly.
Like even ones that like, I had to go
to the subreddit, just I can't think
of an example off the top of my head,
but ones that like, maybe it's ask
T O and you're like, you don't know
if that's like T O is in Toronto.
So both should be capitalized or
it's ask to something like all
these things that I had to go to the
subreddit to get the context for, or
it wasn't immediately clear to me.
And I had to guess it just,
it knew the context of, and it
just capitalized it perfectly.
And that was it was like, Holy crap,
like this stuff would be so handy.
Leo Dion (host): Yeah, it's
like the busy work stuff.
Christian Selig (guest): Yeah, like that
would have saved me like, yeah, like 12
plus hours of work and it probably would
have done it in like under a minute.
And it, yeah, boggles the mind.
Leo Dion (host): Any
thoughts on subscriptions?
Because that was a big thing.
It was like how we were getting screwed.
Because you have annual subscriptions
that you still have to Like people
could ask for refunds, you want
to explain the issue with it?
Christian Selig (guest): Yeah.
Yeah.
So basically the gist of it is
if say January you subscribe
to Some magazine for a year.
And you say okay, I want their
subscriptions for, I want a
magazine a month for a year.
I mean you pay for that year upfront.
It's 10 bucks.
And then say March comes along and the
magazine says, Oh, we've gone bankrupt.
We don't exist anymore.
I guess in the real world,
in that context, you probably
just lose your money.
In the iOS world, Apple is
a really nice middleman.
And I don't mean that sarcastically.
They're nice in that they're very
good at protecting consumers.
And they say, okay, this app
went away, but it can no longer
provide the remaining say, nine
months of service it promised you.
When you paid that 10 so
you have nine months left.
So we'll give you, nine over 12 is, 75%.
So we'll give you 75%, like whatever
was remaining of that 10 back because
they were unable to fulfill that.
So if you have, say, so
that's 7 and 50 cents.
If you have say 10, 000
subscribers, like that suddenly.
75 grand or say a hundred grand that you
were paid all in January all at once.
And then suddenly you're
getting a bill for that.
Hey, Apple, give us that 75 grand back.
Which is like devastating as like an
indie developer, because it's not like
you went in all, you went to the casino
and everything or spent it, but like
you, you have like ongoing costs that
maybe you put that toward and, or you
put it in savings or what have you.
And then suddenly it's all being
clawed back like this money that
you thought you had and that
can be like very devastating.
And Apollo had that issue where
it was like literally that where
there's a lot of people who maybe
a few months before everything
went down paid for a year up front.
And then Apollo was suddenly unable
to give that because Reddit was
effectively shutting down the API.
Apollo through Apple would have had
to refund any the prorated amount
remaining on your subscription.
And that was where Apple was awesome.
Like really nice to work with.
They were Hey, what we're going to
do is we're going to work with you.
And you can add like an option
in your app where a user can say
Hey I'm a really nice person.
I don't need that refund.
I feel I got my money's
worth with Apollo.
You know what?
Don't worry about it.
So if they had that seven
dollars and fifty cents left,
they'd just say, call it a wash.
Don't worry about it.
And that was a really nice option
because every single one of those
person who did that was basically
money I didn't have to pay back.
So yeah, so I guess
that's the explanation.
It was it was one of those things
where it was costly, but it was
one of those things where it's
also Not much you can do either way
Leo Dion (host): Did you,
Christian Selig (guest): hard to escape.
Leo Dion (host): how did people handle
did most people handle that pretty well?
Christian Selig (guest): Yeah.
Oh, yeah people are very generous
with them With yeah, like I was
very happy with how people acted
Leo Dion (host): It seems do you
think there's a better solution
that Apple could implement when it
comes to subscriptions for that?
Did you talk to them?
Because I, we all know like the
Christian Selig (guest): I did
Leo Dion (host): votes and stuff
had run into the same issues
Christian Selig (guest): Yeah, and
honestly they were phenomenal too
I was talking to like Craig and
Paul and they gave me like a lot of
advice and even shared code with how
they handled it on like the kind of
systematic side They were phenomenal
help Apple was like phenomenal help.
I had like calls with them lots
of emails back and forth how to
do things and they were, very kind
and apologetic that not like it
was their fault, but they're like
sympathetic, would be the word.
And and yeah, and they were phenomenal.
And in terms of anything they could
do differently, honestly, not really.
Like it's, that's just
how the system works.
If the developer wants to offer a
yearly subscription they have to be
able to follow through with that.
Yeah.
And if they can't, that's like
where Apple stays comes in as
the middleman of the platform.
And where kind of being the.
The middleman between developers
and the user, that's like a big
benefit for users that they can
be protected in that capacity.
And yeah, outside of so yeah,
no, I don't think there's really
anything Apple can do, nor would I.
Really want them to like it was just
an unfortunate situation for everybody.
I suppose.
Leo Dion (host): What were some
of the features you wished you
got in before the end of Apollo?
Christian Selig (guest): Got so many
I Guess three three I can think of off
the top of my head was the iPad app It
still had a decent amount to go but I
was so happy with how that was coming
It had this cool like I teased the
picture on Twitter But it was basically
this cool multi pane layout that took
a lot more advantage of the screen
real estate the iPad had to offer And
I was really excited to get that out
and it felt very much kind of like
if you remember Twitter for iPad way
back in 2011, that multi pane, like
very gestural UI I was really excited
to, to get people to play with that
and a lot of work went into that.
So it was, and I learned a lot, so
it's not all for loss, but it was I
was sad to see that not make it out.
I had a, kind of like a gallery feature
where you could view a subreddit more
picturely, for lack of a better term.
So if it was very media heavy.
Leo Dion (host): An Instagram,
Christian Selig (guest):
yeah, yeah, exactly.
So it would be like...
Yeah, so you could go in and just
blow the subreddit out and just view
all the thumbnails and whatnot and
dive into them and that was a really
nice way to view some like picture
heavy or video heavy subreddits.
And the other thing like that I put
a lot of work and I still might do
a blog post on sometime because it
was a lot of trial and error was
getting like a cloud syncing system
working where if you read one post.
On your iPhone, like it would sync
that you read that post to your
iPad or certain like preferences
like what theme you have enabled.
Because that was one of those things
that was like, it's, I had so many users
be like, why don't you just enable that?
It must be so easy.
And it's it is so difficult to get
Just so many, so many things that
can go wrong with cloud syncing.
Leo Dion (host): Especially
multiple devices.
Yeah.
Christian Selig (guest): And that was
one that I had a a solid amount of
work done and would have been really
cool to see make it out the door,
but it also would have been a really
stressful one, just to make sure I
didn't do something colossally wrong,
so maybe I'm glad I didn't have to
deal with the stress of that, but
Leo Dion (host): you're
back in Ruby, right?
If I'm
Christian Selig (guest): No, actually
it might have been at one point, but
it's mostly Go now and yeah I liked
Go because it was pretty similar
to Swift in a And it was very fast.
It was compiled, which was nice.
And by the end of it, I wasn't
writing any of the Go myself.
I had someone who used Apollo reached
out to me on Twitter and was like, his
name's Andre, and he was super nice.
And we ended up working together
probably like the last year and a
half, and he ended up like doing
all the server stuff, and it
got so much further and faster.
And that was the other great
thing about Go, is that it's...
I'm sure like Ruby, but there is,
there was, there's no shortage of
people who are really good at it.
I think Andre was a cut above
even a normal good person.
But it was,
Leo Dion (host): but
his expertise was go.
Christian Selig (guest): yeah, exactly.
He was, I, when I met him, he was
working on Amazon at Amazon and
someone like the backend stuff.
So he was like a super talented
guy and still is, and we
still work together on stuff.
But it was one of those things
where it was just so nice to have
I'm more of an iOS developer.
I'm not like the backend stuff was
me feeling around in the dark and
having somebody who would come in and.
Handle that who actually
knew what they were doing.
It was just like, ah, so nice.
So much
Leo Dion (host): Yeah, especially as
a designer like you don't want to have
to deal with Database maintenance and
Christian Selig (guest): Yeah.
Leo Dion (host): All right fun stuff.
What are you working on now?
Christian Selig (guest): So the main
thing I'm working on now is I've been
doing a lot of work on pixel pals for
iOS 17, because there's a lot of good.
goodies for iOS 17 around like a lot
of the new widget stuff that I'm really
excited to play around with because it
feels like something that's almost born
for PixelPals and having some fun there.
Outside
Leo Dion (host): were saying something
about like interactive widgets being a
Christian Selig (guest): Yeah, yeah, so
it's one of those things where that's
something I would have loved to have
had like the PixelPals 1.0 launch is
being able to interact with them in
some capacity when they're like in the
widget state or in the And the fact that
you can do that now on iOS 17 and it's
really powerful is I've, I don't have
all of it implemented, but there's just
so many like fun little ideas for little
stuff that I'm really looking forward
to building and playing around with.
So yeah, that's I'm.
Like, in a, it's terrible to say,
but in a way it's almost nice
that everything happened with
Apollo because I get to put so
much effort into this other thing.
Which is, a good way to, to for me
to look at it on the positive side,
Leo Dion (host): green app
is always easier and more
fun when it first starts off
Christian Selig (guest): Exactly yeah.
Leo Dion (host): for 10 years I'm
Christian Selig (guest): Yeah, the
code base has a lot less technical
debt in some areas, too, which is nice.
It's, like, all Swift
UI it's all very modern.
Leo Dion (host): I wanted to, so
going back to what you had talked
about earlier about design and
notebooks and things like that.
So you've obviously, I think when
you first started off, I would
assume you did Objective C and UIKit
Christian Selig (guest): Yeah.
Leo Dion (host): the first versions
and now you're doing, especially
with widgets, you have to do
Swift UI Swift UI and design.
How do you feel as comfortable?
Are you like.
You do feel more restrained
or you're just like, screw it.
I'm just going to put this
in a UIViewRepresentable
and do it that way or
Christian Selig (guest): of those.
That's a good question.
I think I do find myself more
limited in SwiftUI just because
there are limitations versus UIKit.
But I don't think every year it
gets less and less and it's to
the point now where I don't think
it's anything egregious and I
don't feel like overly limited.
There's just a few areas where once
in a while where I will just be like
for me the way I have PixelFile set
up is, it's all SwiftUI, but the the
life cycle is still the old school
AppDelegate and everything, so I
still have, it's all view controllers.
Leo Dion (host): Oh, yeah.
Yeah.
Christian Selig (guest): I do need
to do something in UIKit, it's easy
enough to just be like, screw it,
I'll just do this screen in UIKit,
it's not the end of the world.
So I have that nice flexibility,
so I don't feel too limited, but
SwiftUI has been able to do 99% of
the things I want to throw at it.
With Apollo, it was trickier because
so much of it is it's this massive
scroll view basically at its core.
And that's, scroll views are one
of the areas we're even to this
day, SwiftUI is a little rough at
in terms of moving you around and
knowing where you are and whatnot.
And even just being like as
performant as UIKit is in terms
of scrolling frames per second.
So that was an area where I would
have loved to have taken Apollo.
Maybe given a few more years,
I would've been able to.
But that was all UI kit and
would've stayed that way
for probably quite a while.
But all like the little side screens,
like if I was doing like a new
paywall screen or if I was doing
like a screen where you could edit
your notifications, like that would
all be Swift ui just 'cause it was
so much faster and so much easier.
But I, but as far as the design
perspective goes I still I've
graduated from like a notebook
that I do mostly like iPad.
Like I'll draw on like the
notes app or something.
Versus like a pen and paper but then
yeah, oh because I just found I didn't
like, like there's the tactility
of a pencil is beautiful, but just
having a big stack of notebooks
in the closet was getting annoying
not being able to reference them
if I was like away for the weekend,
Leo Dion (host): Yeah
Christian Selig (guest): not being
able to analyze be like, Oh, I
know I wrote this string somewhere.
Like, how do I find that?
Just all the short sides Of manual input
Leo Dion (host): So I think last
summer like I started doing design
notebooks and what I did was I wrote
it in a notebook and then I Just took
a picture and put it in like I use
bare for notes And then I would just
attach the picture to a note like yeah.
Yeah, but I
Christian Selig (guest): I would
go through 15 pages though, and I'm
just like I, knowing myself, I would
Leo Dion (host): click click
Christian Selig (guest): to
go back and take 15 pictures.
I just, it would not have happened.
And it's one of, and it's it's nice
being able to like, undo something
that you didn't like how it came
out or grab it and resize it around.
It's, once you get past the
initial lack of control,
maybe versus a pen and paper.
There's so many nice benefits
That I'm pretty happy with it.
But beyond
Leo Dion (host): is what, I was
just gonna say, this is what middle
school Leo learned early on is why
he started doing word processing
instead of writing his essays.
It's just save, undo, all that
crap you can't do that on paper.
Yeah, I get it.
Christian Selig (guest): That, no,
that's a good way of looking at it.
Yeah.
You don't see too many people
handwriting essays anymore.
Which I guess is the same thing.
But yeah, but I, but beyond that,
I still I'll go to Figma after that
to I still haven't quite got to the
point where like I'll go right to
SwiftUI to like design design it
because I still find them a little
faster and able to iterate a little
faster through like Figma or something
by just like having the preview app
open on my iphone and just dragging
something There's no compilation.
I can just I don't get attached to
anything but I know there's some people
who are probably even faster than me
just doing it in pure swift ui So it's
definitely more of a personal thing
than anything, but it's yeah I have
that three pronged approach I suppose
Leo Dion (host): yeah.
Yeah, I think, did Apple do a
talk on that this year where
they're like designing Swift UI?
I
Christian Selig (guest): Oh probably
Leo Dion (host): Yeah.
Christian Selig (guest): Yeah,
it honestly is great for that
Leo Dion (host): yeah, yeah, yeah.
So let's, anything else from
dub dub this year that you
really sparked your interest?
Christian Selig (guest): Oh, there's
probably a ton if I, if I was more
diligent and look through the videos.
But like for me, so much of the
stuff is I am almost like always
perpetually one dub dub behind because
so much of the stuff is like, Oh,
it's great if you're iOS 17 only.
But I'm, I'm
Leo Dion (host): You're
an indie app developer.
You don't even work for a big
company and you worry about that.
Christian Selig (guest): I normally
stick like one version behind.
Because like it would be nice, like
statistically in my apps, I find
like by like October, November, if.
The new iOS came out in September,
like you're at least at 50% of the
user base on the new version of
iOS by then, but I don't know many
businesses, small or large, who are
just happy to lose half of their
customers on that like on that day.
So I normally wait until like
it reaches about 90% which.
It truly, honestly only
takes until maybe April.
Yeah.
For my apps, I find like
Apollo was very quick.
Pixel Pals is a little more general
audience, so it's a little slower.
But and it depends if they
dropped any like major iPhones.
Like it wouldn't surprise me if iOS
17 is a little slower because they
dropped like the iPhone 10 which
is like a pretty, beloved phone.
So I could see people
holding onto that a little.
So we'll see.
But yeah, like stuff like, like the new
Swift data, like I'd love to play around
with that, but I just don't see myself
going iOS 17 only for the near future.
Oh, is it
Leo Dion (host): it.
Yeah.
Yeah.
Every beta it's like a roll of the dice.
Let's see which, what they're
going to break this time.
So
Christian Selig (guest): good,
okay, okay, so maybe it's
good I haven't done that.
Leo Dion (host): yeah, sorry.
Christian Selig (guest): No,
no, no, that's good, that's
very valuable information.
But, and honestly, that's one
of the nice things about waiting
like a year, is you get these
things ironed out a little bit.
Which is really handy.
But yeah, outside of that I'm almost
looking more at the iOS 16 stuff
in SwiftUI, like the view that fits
modifier, which I find like super
handy being able to use that full
time like some of the new app intent
stuff that started on iOS 16, like
all that stuff I'm really excited
for because now I get to use it like
fully but yeah, I'm, and of course
like for DubDub, like the Vision OS
stuff looks really exciting, but it's
one of those things where it's there's
no actual date for it yet and the
date, which is like early, early 2024.
That's for the U.
S.
too.
So I'm in Canada.
So who knows when we'll get it.
But I'm excited for that, but it's it's
a little bit of a wait and see situation
where, we're not like wait and see until
it's out for six months and then build
something, but it's kind of like, I, I
want to see more of their plans for what
they're hoping to do and maybe if I can
get a developer kit to play around with.
But I'm super excited for that.
I did, yeah, yeah, because I have
some ideas for it that I'd really love
to play around with but it's yeah,
I'm just curious to see where that,
where they go with it as a platform
and with all this development stuff.
Yeah.
Leo Dion (host): Have you been doing
anything with swift package manager?
Christian Selig (guest): I've been
like, I use that exclusively versus
CocoaPods or anything in pixel files.
And I've been looking to do more I
forgot who I saw on Twitter doing
it, but just like the be more
modular with your project in terms of
breaking it down into like packages
or frameworks or what have you.
If anything, just to increase like the
the portability of the little modules
of the project is nice, but also just
SwiftUI preview times sounded really
cool I was looking more into that.
That's
Leo Dion (host): Maybe we'll
talk about that offline
because I have a few thoughts.
Christian Selig (guest): Oh good.
Okay, perfect.
Leo Dion (host): What else
did you want to talk about?
And any big challenge, any what's
coming out with pixel pals next or
Christian Selig (guest): hopefully
for iOS 17 Day 1, I'll have some juicy
interactive widgets in some capacity.
I'm not 100% sure completely what form
that'll take but I think I'll have some
really fun stuff for that, for Day 1.
Outside of that I have some...
Some other kind of stuff swimming
around in my head for ideas.
But I'm very one track minded in
that I know if I start planning
anything else or digging into
anything else, like the pixel pal
stuff and the short term will suffer.
So I'm kind of like, okay,
get that out the door.
Step
Leo Dion (host): and this is why I did
not apply for a development kit for
Christian Selig (guest): yeah,
that's probably a fair point as well.
Leo Dion (host): Christian.
Thank you so much for coming on.
Where can people find you online?
Christian Selig (guest): I'm at
Christian Seelig pretty much everywhere.
I'm on Macedon threads, Twitter.
Yeah, that's, and that's my website too.
So it's, it should be pretty easy to
Leo Dion (host): forgetting,
you're forgetting another
social media platform?
Or do we not mention that anymore?
Christian Selig (guest): Reddit.
Leo Dion (host): Ha ha ha, yeah,
Christian Selig (guest): oh I
honestly don't really, you like
I I'll check my city subreddit
once in a while, but no, I
Leo Dion (host): Are you serious?
Christian Selig (guest): yeah, yeah.
Like in terms of that
being a lot or a little,
Leo Dion (host): No, no, but
that I'm just, I don't even check
Reddit anymore since Apollo's been
Christian Selig (guest): Oh, it's
it's, I don't really check it like
compared to what I used to check it.
No, not at all.
But there's once in a while There's
some stuff going on in my city that
I just to poke my head in on but
But yeah, i'm definitely that reddit
would not be the best place to find
me right now to be honest Not in
any shade just out of practicality.
We'll
Leo Dion (host): Yeah.
Thank you, Christian.
I really appreciate it.
It's
Christian Selig (guest): Oh,
yeah, it's always a pleasure
Leo Dion (host): People can find
me on Twitter at Leo G Dion.
My company is bright digit.
If you're watching this
on YouTube and subscribe.
I'd really appreciate it.
And if you're listening to this on a
podcast player review would be helpful.
If there's anything you want to talk,
come and talk about, have a new product
or have a new API you want to dive into.
Got a book, a talk, whatever.
Let me know.
I'd love to have you on.
Thank you everybody.
And I look forward to
talking to you again.
Bye.