00:00:04 ◼ ► Yeah, yeah, it was a lot of things at once because I mentioned in my post, but it was my girlfriend's 30th birthday on the Monday.
00:00:10 ◼ ► So I was kind of like, how am I gonna get down there without being the worst person in the world?
00:00:16 ◼ ► So I was kind of like, oh, I'll bring you down with me. We can like have some fun. And then everything kind of went to crap pretty quickly.
00:00:21 ◼ ► Yeah, we could go down there and celebrate. I mean, let's face it, anything with a zero is a big round number. It's always important.
00:00:30 ◼ ► It really, when I read that in your post, it really was like the extra kick in the nuts, right?
00:00:36 ◼ ► Yeah, it was. She was so good about it, though. We ended up going mini golfing one night, so I think it was salvaged somewhat.
00:00:41 ◼ ► I, for anybody out there listening who's been living under a rock, Christian Selig is, you are, is there a company behind Apollo or is just Christian Selig?
00:00:57 ◼ ► I think technically in some capacity, but yeah, it's basically just me. I have a friend named Andre who works on the server back end part time because that's not my forte and he's really good at it. But it's effectively just me.
00:01:09 ◼ ► Right. But for the last eight years, you have been the one person show behind Apollo, which is literally an award winning, right? You've won awards with Apollo.
00:01:22 ◼ ► Yeah, yeah, yeah. No, I think there's been some backstory ones and some editor's choice stuff. So I've been pretty lucky.
00:01:38 ◼ ► I'm crying inside. Reddit has gone from a free API since forever to instituting a paid API that your estimates would peg your bill if usage of Apollo continued as it has been would cost you $20 million a year.
00:02:03 ◼ ► And and to top it off as if that weren't bad enough, you got about 30 days notice or 30 days on.
00:02:11 ◼ ► Yeah, yeah. Yep. 30 days from when they kind of announced the pricing to when we started incurring charges, which is kind of funny because I did the math last night and it took them 43 days from the initial announcement for them to figure out what the pricing would even be. And then 30 days from that to actually implement it. So
00:02:29 ◼ ► And so you came to the conclusion that you're going to have to because of this shut Apollo down at the end of June, which again is not even that is not free. This is going to it really is deja vu all over again with Reddit as a third party.
00:02:49 ◼ ► Exquisitely well designed, beloved paid app for Reddit is exactly in the almost exactly in the same position that apps like tweet bot and Twitter if we're in for Twitter. Just a few months ago, I mean, just
00:03:07 ◼ ► no. And it's funny because like starting out iOS development, like Paul and Craig and them were just like people I looked up to so much. And it's kind of funny, like I've talked to them quite a bit over the last two weeks. Like they've been honestly so kind and like giving tips and like sharing code for how they handled it and contacts.
00:03:21 ◼ ► I wish I got to talk with them so much under better circumstances, but it's yeah, they've been awesome.
00:03:26 ◼ ► Yeah, and it's again, it's unfortunate. It really is. It's tragic. We'll get into the weeds on some of these details. But you know, part of looking at this looking for the silver lining in a cloud with the Twitter situation months ago. And I'm really good friends with Craig Hockenberry. And I know Paul Haddad and Mark at tap bots. So I'm not surprised. And I know from as much as I know those guys, I wasn't the least bit surprised that those guys helped each other.
00:03:56 ◼ ► Go through the process of unwinding this and sort of coordinating and working behind the scenes with Apple to coordinate how to because part of the mess of this whole thing and it is again, it is like deja vu all over again, where Twitter rank yank the rug out from under those guys read it is giving you 30 days notice, but it's not that much different.
00:04:24 ◼ ► 30 days is closer to zero than it is to a reasonable amount of time. And the problem is that it's you Apollo and tweet bot and Twitter effect were monetized through subscriptions, which included annual subscription. I mean, this is it really when you think about this from a business perspective, it's just a gut punch because you've got customers who might have renewed their annual subscription a month ago. Right?
00:04:50 ◼ ► Exactly. I mean, there's no Do you know, are your subscribers more or less because you've been around and Paul has been around for eight years, I would guess that your subscribers are evenly distributed throughout the year. There's no real reason for Yeah, yeah, it's pretty much exactly evenly distributed through the year. And it's funny. I'd also add that I had a call in January with Reddit, funnily enough, where they said they had no plans to change the API in 2023. So when when you say stuff like that, you can kind of make projections that okay, a yearly subscriptions probably a safe bet. But then three months later,
00:05:22 ◼ ► Yeah. And I would think and part of what makes the Reddit situation extra frustrating 100 times for you, but even just for outside observers like me and the people listening is that at the outset, when they first started saying, Hey, we're going to have to change our API rules, we're going to have to start charging or something, blah, blah, blah. They they emphasize at the outset, but we're not going to do what Twitter did. And it's like, okay, that sounds right. You understand Reddit to business. I think
00:05:50 ◼ ► there were suspicions until recently that they weren't profitable. And now, Steve Huffman confirmed that it's kind of that they're not profitable. I mean, it's obviously in your interest for Reddit. If all if things had gone well, and this this weren't happening, you would still want Reddit to thrive because you've built a client that is completely specific to Reddit. In an ideal situation,
00:06:20 ◼ ► it's a symbiotic relationship where Reddit is thriving financially. Reddit is thriving as an overall community. And the the API's are priced in a way that allows third party apps to thrive. Right? You would think
00:06:37 ◼ ► Yeah, you would like I don't think any developer I talked to at the beginning of this on the Reddit side was like, this seems unreasonable or like it was there was a sense of excitement that this will kind of formalize the relationship a little bit better. Right? Like I remember Paul from tweetbot was even saying like, dang, I wish Twitter kind of did this thing. Like, this sounds a lot better. So it sounded like a good deal at the beginning. But yeah, the devil's in the details, I guess.
00:06:56 ◼ ► And again, like you said earlier this year, not that long ago, you know, we're only in June, we're not even halfway through the year. For them to say we don't anticipate changing the API's this year. It certainly didn't give you any reason to think that you should rethink the fact that you have annual subscribers, right?
00:07:15 ◼ ► Oh, no, they were they were very like, we love having you like, we're not thinking this year, it might be the tune of like, several years. And even then it would be improvements. Like it was very like, keep doing what you're doing. Yeah.
00:07:26 ◼ ► What do you think happened? What changed from Reddit's perspective between the beginning of this year and May?
00:07:35 ◼ ► Um, if I had to guess it would be a combination of seeing what happened at Twitter and seeing you can kind of get away with this killing of the third party apps thing and a combination of wanting to IPO and try to extract all this money from services that use your use the Reddit API, like the AI platforms of the world, which is again, totally fair, but like you could pretty simply in a way Apple almost does with reader apps, like carve out an exception for certain kinds of apps that maybe contribute back to the platform in some capacity.
00:08:04 ◼ ► Yeah, and I'm not like a financial expert, but it's no secret that Reddit is looking to do an IPO. And it's also not a secret. It's been publicized tech crunch was the place I linked to but that the one of the lead investors in Reddit downgraded their investment and downgraded their estimate of the company's valuation to beneath the level they were looking to IPO at. And so in some, I and I think it's fair.
00:08:34 ◼ ► Again, I'm not a financial expert. I've never done an IPO, but I think the very, very broad perspective that a big part of a company having an IPO. Obviously, you'd want it to be very logical and have actual experts looking at the actual the books as it were have accountants and finance CFO types looking at the revenue and the costs and the projections for the future.
00:09:03 ◼ ► And make a truly logical calibrated defendable mathematical estimate as to what the company is worth. But the truth is the stock market isn't entirely logical. Everybody knows that in every way. And an IPO is largely about also about momentum and public consensus.
00:09:26 ◼ ► And Reddit's momentum has been on the downswing recently. Just the news coming out that your lead investor has cut billions off your valuation is not the direction you want to be headed when you're looking at an IPO.
00:09:42 ◼ ► And I guess this is and I think it's this AI thing where they're like looking at Reddit's business and who's using Reddit and what it's been used for. And the fact that Reddit has been around for so long. When did Reddit start? Like 2006, 2007?
00:10:00 ◼ ► That sounds about the right time. Somewhere around there. A couple years, four or five years after I started writing Daring Fireball. And it took off very quickly. I guess by today's standards of Reddit, those early years were probably small. But I seem to recall that it gained traction alongside dig.
00:10:18 ◼ ► Yeah, exactly. And therefore it had 15, 20, close to sometime between 15 and 20 years of massive amounts of, I guess, largely English language. I don't know. Is there are there?
00:10:38 ◼ ► I was going to I, I that was a bad assumption on my part. But I realized that if there is like a very strong German or Spanish language contingent, I guess I just never ran into it. But I don't see it. But lots and lots of text written by real people.
00:10:55 ◼ ► Open and indexable to search engines. This is something that people have already said Google is already suffering with search results because it's just a very it's super common. I mean, I'm sure you're aware of this, that it's like super common suggestion that when you're looking for how to blank or recommendations on blank, just include the word Reddit in your search string.
00:11:21 ◼ ► To steer Google or whichever search engine you use towards Reddit results, and you're way more likely to get good results, right? I mean, it's it's the truth.
00:11:31 ◼ ► Oh, absolutely. I imagine that's how a lot of people find their way to read it in the first place.
00:11:35 ◼ ► Right. I told this story a couple weeks ago where I rented a car, a stupid Kia Sorento to drive my son home from college a couple weeks ago. And I could not figure out how to open the hatch on the thing. And I felt so stupid. I'm like, how is this possible? And after trying for like, literally 10 minutes, I googled it and the top result was a Reddit thread. And it was exactly it was somebody in my exact situation. Except this woman was like, she even said in a Reddit post, look, I know this is stupid, but my husband and I just rented a 2020 Kia Sorento.
00:12:04 ◼ ► We can't figure out how to open the trunk. And somebody was like, oh, it's under the E in Sorento. And it's really a small button. And it's like, oh, thank you. God only knows how what the second place result was. And I remember looking at the search results. And it wasn't that great. It was like this one thread on Reddit gave me the answer instantly. And second place was, I don't know, read the manual. I don't know.
00:12:31 ◼ ► And but so all these in addition to search engines being a great source for search engines for answers, because it's really human stuff, and the ranking system promotes good answers. It also turned out to have been a very good data set for these AI language models. And they got to train for free off an enormous corpus of very good text for training real language.
00:12:59 ◼ ► And now they're looking, I guess, like Steve Huffman is looking at OpenAI having a multi hundred billion dollar valuation. And they're like, well, give me some of that. How about we set the API price super high so that the next time somebody wants to train on our data set, we can make a fortune.
00:13:17 ◼ ► And then our IPO will look AI flavored. The hype buzzword of 2023. Reddit's part of it because we're the training set for this. And we're going to make a mint on AI training, blah, blah, blah. And then it's the old dot, dot, dot gnomes and underpants profit.
00:13:41 ◼ ► And yeah, and I see that to a certain extent. But it's one of those things where I wonder if that is how they landed on this. It just feels a little short sighted in so far as you have these companies like OpenAI that just vacuum up all of Reddit's data and then package that up and sell it as their own product, essentially.
00:13:56 ◼ ► And I don't blame Reddit at all for wanting to monetize that. And I also don't blame them for wanting to monetize the greater API. But for services that inherently give a certain amount back to Reddit through like moderator tools and alternative experiences and whatnot, there's an inherent value for Reddit there that I think would almost warrant carving out a separate pricing tier that might be a little bit more affordable for the non billion dollar companies.
00:14:21 ◼ ► Because at a certain, from a certain perspective, if you're just looking at the bill from AWS or whoever you're Reddit's paying hosting bills to, an API call is an API call and incurs some each call incurs some fraction of a penny's cost to keep the server running and for the bandwidth, blah, blah, blah.
00:14:42 ◼ ► But strategically, it really doesn't make any sense at all to price all developer even if you're going to go to a paid API, which is reasonable, right? Everybody, I don't think anybody is disagreeing that it's unreasonable to go from totally free API usage to some sort of paid model.
00:15:01 ◼ ► But it doesn't make any sense to consider clients, just usable clients for normal users like Apollo or specialized ones for moderators, which I get the impression there's a bunch of tools like that to something like open AI or somebody who wants to train on this massive data set.
00:15:22 ◼ ► It makes sense to have totally different classes. You brought up the reader exception. The app store guidelines are full of weird little exceptions for different classes of apps.
00:15:32 ◼ ► Yeah, no, yeah. That way I don't get it. And they even said server costs aren't really the biggest thing here. It's the opportunity cost of using one app versus the other. So that's where I get almost confused where it is so contingent on you're making 100 versus 200 API calls.
00:15:47 ◼ ► If it's more so the user, why are we focusing so much on the server fees? So it's very confusing.
00:15:53 ◼ ► Yeah, it's definitely confusing. I guess I know you just talked to the Verge on the Verge cast this week. And just a couple days ago, they asked you, how sure are you that come June 30, this is really going to be the end? And your answer was 90%. Are you still at 90%? Or is that crept up?
00:16:12 ◼ ► I think, yeah, it's crept up a little. I just have pretty much no expectations at this point. Not to say I don't hope they won't come around on this, but I just don't 100% see it happening. And I feel like getting my hopes up in that way would not be smart. And I'm kind of operating on the assumption that if they do, that's a nice little surprise. But there's certain kind of shutdown procedures I need to go through with Apple that I need to get done. But I can't kind of wait around for them to hopefully finally listen to the community some.
00:16:40 ◼ ► Yeah, it's like you got to start moving. All right, we'll come back to this. But let me take a break here and thank our first sponsor of the show. And it is our good friends at Memberful. Memberful is where you can go as a creator to monetize your creative efforts with memberships, and grow your revenue. Memberful is passionate about your success as a creator. And they do everything to be on your side, to give you access to everything you need to have
00:17:10 ◼ ► a successful membership system. Number one, they always give you full control and ownership of your audience and your brand. So people sign up for your membership when you're running a system on Memberful. They don't see Memberful first, they see your brand. And Memberful is just the back end. And people think they're interacting with you because they are. And if you ever want to leave Memberful and go somewhere else, you get to take your whole membership list with you.
00:17:36 ◼ ► It is so they make it easy to leave because they are so confident that you're not going to want to leave. But it gives you confidence. Your brand stays foremost in you maintain ownership of your membership. And they why would you want to use them because they seamlessly integrate out of the box with all of the tools you already use, including services like MailChimp, WordPress, Stripe, Discord, and more. They're your accounts. These are your fans who are signing up for your
00:18:06 ◼ ► membership and they just want to easily plug into the stuff you already use. They have everything you need to run a membership program. Powerful, easy to use checkout, easy to use member portal so your members can come in and change their email address or do change their from monthly to annual or anything like that.
00:18:24 ◼ ► All of that is all built in. They give you advice on how best to monetize because they only make money when you make money and they know what works. They have so much experience with so many creative professionals out there. You can get started for free and try Memberful with no credit card required upfront. Free trial. See how everything works without giving them your credit card. And then when you're ready, just go to memberful.com/thetalkshow.
00:18:53 ◼ ► And they will know you came from the show and it couldn't be easier. Memberful.com/thetalkshow.
00:19:01 ◼ ► So let's talk about other ideas. What if? So why not? Here's why not just charge your customers more to make up for the fees that Reddit would charge you if you kept Apollo going?
00:19:15 ◼ ► That's a great question. I think there's a lot of reasons, but I think if I was just to give a few, the main one for me is that, not even the main, but one of them is that at this point it kind of seems evident that they're not really happy having third party apps around. And this was kind of like an indirect way to kind of try to push them out. Seeing that through some of the actions that they've taken and decisions they've made, I don't think they're particularly happy. And that kind of creates an environment where it doesn't feel very productive to build it in if you're not wanted there. It doesn't feel like there'll be a good working relationship going forward and a healthy future.
00:19:45 ◼ ► So that gives me a lot of concern. And in terms of why not just say, okay, everybody's paying five bucks a month now if you want to stick around. That would be a very easy solution on the surface at least. And it becomes one of those things. Apollo's average user uses around 345 requests a day. To get mathy, that's an average. So for users closer to 200, a subscription in existing users closer to 500. So you've got an existing subscription user using about 500 requests a day on average, which makes sense that they'd use more. They're paying for the app, they probably don't have the money.
00:20:15 ◼ ► They probably like the service. You're looking at closer to like $3.70 to keep them around. So you take Apple's cutter, the five bucks, you're 10, 20 cents in the red. So your average subscription user already can't stay around. But then even beyond that, I've got a subset of users around, I think it's like almost 6%, which is almost like 100,000 users where they use between one and 2000 requests a day. So you're looking at seven to $14 in fees a month if they'd incur. So you've got kind of this system where you're like, is it going to be like, I'm turning the lights off once you hit a certain amount of usage of this app?
00:20:45 ◼ ► Am I going to have to build Apollo in a way where, okay, I can't make this feature too good because they'll use the app more and it'll cost me a bunch of money. I've kind of got to shittyify everything. And that's not really a good way to build an app. And then even beyond all that, if I could make all of that make sense, just the fact I've got a lot of existing subscribers that, as we mentioned earlier, have already paid based on operating costs, I kind of assumed earlier.
00:21:07 ◼ ► And starting July 1st, the 50,000 or so yearly subscribers I have now will start incurring at minimum a dollar, but more realistically like $2 each. So you'll have a bill of $100,000 a month coming until their subscription kind of all lapses. And that's almost at a point where it would be cheaper rather than giving hundreds of thousands of dollars to Reddit just to kind of close up and say you can have your money back.
00:21:30 ◼ ► How important has it been that Apollo also heretofore up until now can be used for free?
00:21:40 ◼ ► I think it's just been one of those... It hasn't been super important. It's just kind of been a nice thing to have. The thing that drew me to app development in the first place was that you could kind of just be on the bus and see somebody using something you created and that whole concept was kind of infectious.
00:21:52 ◼ ► So if somebody wants to use the app and they're not paying me and they're just enjoying something I like, that always sounded super cool to me. And maybe if they use it after a few months, they might decide to support it in some capacity. And that was really cool.
00:22:03 ◼ ► I love that aspect. And having people like moderators who kind of like are the reason the communities are alive, being able to use the app for free was really cool. So it was a really nice thing to have, but it wasn't something I was 100% tied to.
00:22:13 ◼ ► Because I do understand at a business level why Reddit can't potentially give those out for free indefinitely. It's just the price they landed on to facilitate that was so high and so fast that it made things quite difficult.
00:22:26 ◼ ► Reddit aside, and Reddit's obvious control over their own platform aside, it just seems like the modern way to have a commercial indie app is a free to download app.
00:22:46 ◼ ► I mean, and obviously there are exceptions to every rule, but basically make the app free to download and have in app subscriptions probably monthly and annual where annual is something less than 12 times monthly to encourage people to save money on an annual basis.
00:23:06 ◼ ► But then to sort of regularize your income by having annual subscribers instead of living month to month. And the free download is sort of what people it's what used to be the shareware model of you download an app from somebody's website, try it for 14 days and then decide to pay.
00:23:29 ◼ ► That doesn't it's never been supported by the app store. There's never been a free trial thing. You can have a free app and then you have in apps purchases or subscriptions.
00:23:38 ◼ ► It's it. That's how most of these apps work. I don't I guess tweet bot didn't have a totally free to use tier. I mean, but yeah, I think that was a limitation of that whole token thing they went right.
00:23:52 ◼ ► There so there can be exceptions. I can imagine a scenario where they came up with some sort of Reddit had come up with some sort of pricing.
00:24:00 ◼ ► That probably would have meant you couldn't have a free tier or it would have required some other hypothetical like if Reddit had said you can keep using the API, but if it's a client app like Apollo where it's this is what you as a user used to access Reddit instead of the official Reddit client that you'd have to show Reddit's own ads to use the API.
00:24:26 ◼ ► I mean, this is obviously where Twitter sort of lost interest in third party clients a decade ago, right? One of the weird one of the differences between the Twitter situation and a Reddit situation is at least.
00:24:40 ◼ ► With Twitter developers like the icon factory and tap bots kind of knew they were on shaky ground for years as you just emphasized earlier in the show even just earlier in January of this year. Reddit was like, yeah, we're good till next year at least and then we'll come up with something fair.
00:24:59 ◼ ► But would you have been amenable to that if they had said look one of the ways we're going to want to monetize this is we want to put our ads in front of everybody who's using Reddit.
00:25:09 ◼ ► Absolutely. I think there would have been so many different ways that they could have done this that would have been ultimately better for everyone. That would have been an option and you could have it so maybe the ads wouldn't come through if you had Reddit Premium or something.
00:25:21 ◼ ► Or just you need Reddit Premium to use third party apps. There would have been so many aspects that they could have done this as that I think ultimately would have made a lot more sense than what they landed on and could have potentially been easier.
00:25:31 ◼ ► But even with the ad thing, the unfortunate thing for a lot of developers was not only are they integrating their own ads, they're effectively not allowing you to have your own free tier because they're also banning third party ads frameworks like Google's for instance where a lot of these platforms historically also use those to power their free tier.
00:25:49 ◼ ► Apollo never did because I wasn't a big ad guy but Narwhal and a lot of the Android apps were kind of double hit because on the same day as that initial announcement they were like you have a week to get rid of your ads and they were like hold up what? How do I monetize? That was kind of the whole thing. So it's very tricky here.
00:26:07 ◼ ► Right and I can imagine that on the Android side of things, I can see why that's probably more common. Everybody knows that for whatever demographic reasons, iOS users are more likely to pay than Android users.
00:26:24 ◼ ► And so paid apps are more of a thing, paid subscriptions to get advanced features are more of a thing in the iOS world than the Android world and so the fact that Android equivalence to Apollo, third party clients you put on your Android phone access Reddit and offer X, Y and Z that are better than the official Reddit client or just different, right?
00:26:46 ◼ ► Even if it's just different and therefore better for some people monetize through their own ads. For them to just say what? You have a couple weeks notice or one week notice?
00:26:59 ◼ ► Yeah well what happened was initially they said it was very confusing because as a backstory, when the first announcement came out in April I messaged a bunch of the other developers and said let's make a group chat and kind of talk here. And then a week later they all got an email that was like we notice you have ads in your app, you have one week to get rid of them.
00:27:16 ◼ ► And they were like, okay so but your API guidelines say they don't come into effect until like June or July or whatever like what's the deal there? And then they kind of ended up backpedaling thankfully and I think they gave them up until July 1st or maybe a little bit before then to remove them.
00:27:30 ◼ ► But there was definitely some like running around with their hair on fire like this is the entire monetization model for the app and it's being yoinked with no opportunity to replace it with really anything.
00:27:39 ◼ ► Right because they're not even offering their own. And again it would be bad enough to give such short notice and say because we were going to insist that you show our ads and we're going to have some sort of 70/30 revenue split or whatever the revenue split would be on the ads. Something like that. Nope.
00:27:59 ◼ ► And I mean say what you want about Twitter but to my knowledge at least Twitterrific had their own if you didn't pay they had some sort of ad banner platform so Twitter at least was okay with advertising in some capacity through their API so it's yeah it's a weird situation.
00:28:14 ◼ ► I guess the way and it's the we're back to the things that I find similar between the two companies and I find it almost baffling that the leadership at Twitter and Reddit doesn't seem to see how damaging these moves are to their company's credibility.
00:28:33 ◼ ► Because when you're the bigger company and when we say third party client I mean when you're the first party you have a certain responsibility and I'm not just saying like hey be a good person I realize business is business and sometimes tough decisions have to be made.
00:28:49 ◼ ► But it you're you're as a company your reputation is worth something a reputation and it's weird because you can't really put a dollar sign on reputation.
00:29:04 ◼ ► Right so if if you conduct a poll and people say I trust Amazon and do you trust Google do you trust Apple and those companies tend to do very well in those polls Amazon in particular people trust.
00:29:19 ◼ ► What what is that trust worth where you can't it's impossible to put a dollar you can't sell it it's not something you can sell, but you can make moves that damage it and that therefore damage the company and damaged future actions that people have.
00:29:36 ◼ ► And when you're a first party and you offer an API you're you're implicitly saying to developers trust us you can build on these API's and you can invest time you can build a strategy around it and this is something you can count on.
00:29:55 ◼ ► And if we're going to change we're going to do it like the golden rule do unto others as you would have others do to you.
00:30:01 ◼ ► You will work with you on changes like who in the world would trust Reddit going forward at this point it's the same same thing with Twitter right I mean who in the world is going to get back to building on Twitter as a platform.
00:30:16 ◼ ► Well and it's funny you say that because I'm one of the calls that was exactly what they said we saw with Twitter and once you do one of those moves like you can't get developer trust back after you make one of those moves.
00:30:24 ◼ ► And it's fascinating to me because like I don't think you can attribute a specific number but I think almost in abstract terms at least if you compare it to a company like Apple say they decided to shut down their app store tomorrow like that would piss a lot of people off but Apple has like they sell incredible machines they have incredible patents and hardware and software that they sell that gives their company a lot of inherent value.
00:30:43 ◼ ► Where Reddit doesn't have any of that like their inherent value is solely based on the community they have and the content they create that's the reason you're going to the website not because there's a Reddit book Pro or something.
00:30:54 ◼ ► And I think and even to a greater extent than services like Twitter or Facebook where even more so where they their content moderators are paid employees like I was looking up and I think like Facebook pays pays like $100 million a year contract for their moderators and that's something where it gets completely for free.
00:31:10 ◼ ► And so I think when you have these like nine figure amounts of money that you're getting out of goodwill from the community and the community creates the fact that your company has any value whatsoever I think it's really important to balance that reputation even almost more so than Apple would have to or a hardware company like that is where all your money is.
00:31:27 ◼ ► Historically I found Reddit and Steve too being the founder like had an inherent knowledge of that how Reddit operates as a company in that community aspect and it's just been really weird over the last few weeks to kind of see it completely go in a different direction.
00:31:39 ◼ ► Right because it was founded as a company it wasn't like it turned from some sort of nonprofit entity which is sort of the sleight of hand that OpenAI pulled right I mean it's right there in the name OpenAI is supposed to be OpenAI and now it's not open at all.
00:32:12 ◼ ► And it's funny Reddit does operate a lot like that where it is completely volunteer driven.
00:32:30 ◼ ► Very volunteer driven I mean in terms of where does the manpower for the moderation and content creation come from the closest I can think of is Wikipedia both in terms of scale and importance to the internet overall.
00:32:50 ◼ ► I mean I think Wikipedia is more important than Reddit but Reddit's in that same ballpark at least but it is it's volunteer.
00:33:04 ◼ ► Yeah, it's just almost an unfathomable amount of volunteer effort certainly the moderators of each subreddit but then within the subreddit the content itself comes from users who just choose to spend their time typing and writing the either their thoughts or their advice or whatever the nature of the subreddit is and just generating content for free.
00:33:35 ◼ ► There's no such thing though in Twitter as the moderator role right for better or for worse and probably for worse.
00:33:42 ◼ ► That's just a fundamental difference between the concept of Twitter and the concept of Reddit.
00:33:47 ◼ ► But the moderate I mean how much time do you think the moderators on a big subreddit spend?
00:33:53 ◼ ► I mean to me it's at least a part time job and for some it might be closer to full time in terms of.
00:33:58 ◼ ► Oh I would imagine some of these subreddits are like just deal with like literally millions of people and like and not only the time investment but like the toxicity you'd have to deal with from users and the content they'd be posting and trying to get through and like dealing with like angry mobs with people like I just can't even imagine like the mental strain in addition to the time like the mental health deduction per minute would be a very high even if the minutes weren't very high so it's yeah it's definitely not an easy job.
00:34:22 ◼ ► Right one of the reasons that we and again Facebook is all of these successful the more successful they are the more different they are from each other but one of the things we know from I know the Verge did a lot of research on it a few years back that with the paid contractors that Facebook has to do content moderation across their products.
00:34:42 ◼ ► It's incredibly stressful a lot of people burn out very quickly they effectively no joke get the like post-traumatic stress and they have paid counselors it's a big part of the infrastructure that meta slash Facebook has for their content moderators even if they're not officially meta employees a big part of it is counseling and helping people deal with the anger the hate for some of the stuff and I'm sure Reddit right now is like.
00:35:11 ◼ ► And I'm sure Reddit runs into it too with just some of the imagery that comes through you know for sure and you see these things that are just whether it's violence or stuff to do with children or whatever it might be that you some stuff it just isn't good to see any of it let alone to see it on a regular ongoing basis because you're moderating the subreddit and for Reddit that's just something they didn't have to pay for I mean.
00:35:38 ◼ ► So it does bring me to one question I've had which is how in the world is Reddit not profitable at this point that it kind of boggles my mind I mean I'm not saying that it's easy to turn a profit.
00:35:49 ◼ ► But it doesn't it seems to me that for as long as they've been around there's no excuse for them not to be profitable.
00:35:56 ◼ ► Yeah even on a small scale like it's I honestly don't get it there's like I think they've shown or we've shown through third party apps that there's a large amount of people who are willing to like contribute financially to the service they care a lot about if what they get in return is something they see as advantageous and it's one of those things where it's like yeah like you as a company might not be profitable but about a lot of the third party apps have figured out a way to be in some capacity and like maybe a revenue share model would have like where we could have helped you.
00:36:24 ◼ ► To become more profitable could have even been on the table and it's like you can just run through so many situations in your head where the outcome would have been better than the one they landed on.
00:36:34 ◼ ► It's like a recurring theme for me companies that aspire to punching above their weight in terms of how big they imagine they should be and like my business during fireball is very very successful business for a one person company I do well it's very nice.
00:36:56 ◼ ► It is not a billion dollar idea it's orders of magnitude away from that it is a very nice living for one person and I'm fine with that because it's the work that I want to do and I get to write what I want to write and I get to do this show which I enjoy doing.
00:37:11 ◼ ► And there's just no way to turn my business into a 50 million dollar a year business without ruining what it is I mean it would be nice if it was a 50 million dollar a year business I would that would be very fun but it's not but I'm okay with that.
00:37:28 ◼ ► And I kind of feel like a lot of times companies like it sounds to me like what Steve Huffman has in mind with this IPO is that Reddit should be valued at like a 20 30 billion dollar company.
00:37:42 ◼ ► And it's probably like a billion a 1 billion dollar company, which is nice you know a billion dollar company should be could be very nice there's a lot of companies out there that would love to be a billion dollar company, but if you have it in your head that you should be a 20 billion dollar a company or a 50 billion dollar company, then being a 1 billion dollar company sounds like a failure.
00:38:06 ◼ ► But just because you want it to be true doesn't mean it is true right and I just don't think the nature of Reddit lends itself to being as big as they think it should be.
00:38:17 ◼ ► And I think that's why they're not doing things like the again, it's like people who want to keep using Apollo going forward like how about you just charge us $5 a month to be Reddit premium subscribers will just give you $5 a month each and then let us use third party clients.
00:38:33 ◼ ► Yeah, and especially when you see that like Reddit has historically so much farther below other apps or other services in terms of average revenue per user just I don't know if that's because of like the more anonymous nature of Reddit makes it harder to advertise to but they don't have a very high average revenue per user.
00:38:49 ◼ ► So being able to say like give us $5 a month to use these third party apps like that would that would be an incredible amount of margin for them to have so to the extent that it just seems like a slam dunk.
00:38:59 ◼ ► Right and I just can't help but think that they look at it and don't think well look our poo I don't know is that how people pronounce ARPU average revenue per user.
00:39:09 ◼ ► I'll pretend it well I've made it up but I'll just pretend average revenue per new but ours is lower than theirs and to me the reasonable way to look at that is well we can try to turn that dial up a little and and increase our average revenue per user.
00:39:27 ◼ ► But the that's going to be like a slow boil right it's like if successful if maybe we can do let's try these three things and maybe we can gently increase that but it seems to me like they're like why can't we just jump ahead to a much higher ARPU instantly.
00:39:45 ◼ ► Again it comes down to like the gnome selling underpants how I don't know we'll do something but we've got to do something because it's got to be much higher because if it stays where it is we're not a 20 billion dollar IPO and it's like well tough noogies maybe you're not a 20 billion dollar IPO you're just a nice little smaller business.
00:40:09 ◼ ► Yeah, no, I wish I wish I wish I could make sense of it too it just doesn't add up to me and it's like yeah, I mean Instagram sold for a billion dollars and I think everyone was really proud of that but I think like the metric and the goalposts for what what is deemed as successful changes every year to just going so high that if it's at the cost of the company so be it.
00:40:26 ◼ ► I guess the other reason that their average revenue per user is so much lower it's like you said part of it is the nature of a lot of Reddit content doesn't lend itself to advertising as well I mean I'm not just talking about not safe for work content.
00:40:41 ◼ ► There's probably large swaths of Reddit that just aren't as interesting to advertisers Instagram is sort of one of the reasons it's such a there were a lot of people when when Facebook bought it for 1 billion thought you know Zuckerberg's crazy I can't believe they're buying a stupid picture sharing site for a billion dollars and of course in hindsight is one of those all time great acquisitions in business history not just like tech history but business in general it's like Apple buying
00:41:10 ◼ ► next in 1997 just a bargain, but the nature of Instagram in hindsight because it's so visual and that they develop this culture of influencers when everything you see in Instagram is supposed to be attractive and pretty and aesthetically pleasing it just seems very natural to put ads next to it and of course you gain these insights into what the person is interested in and you can advertise against that and I guess some people
00:41:39 ◼ ► some parts of Reddit aren't like that maybe they just haven't maybe they could be more successful I don't know that it's it's so interesting to me though because like I hear what you're saying but I'd almost challenge that like you have versus Instagram and sites like that like if you're like a cycling company like you can go to like our slash bicycles or whatever and like you can put an ad specifically for that community that you can go I am very confident anyone in this community will find this product interesting whereas Facebook they'd have to do some like
00:42:06 ◼ ► AI and detection of like has this guy looked at a bicycle in the last 12 months but Reddit has like these super hyper focused communities you can almost dial into
00:42:14 ◼ ► right and that's my business right that's how daring fireball sponsorships work it's people who come in there like oh I people sometimes ask like have I ever done like a reader survey or something like that and I haven't maybe I will someday but I always say your guess as to who reads daring fireball is probably very accurate I don't think it's very hard to sort of guess the general interests of people who read my website and when you look at the companies that sponsor my website they they're they're seldom surprising
00:42:43 ◼ ► right it's people who appreciate nice things and people who are computer nerds and there's great and like you said bicycle subreddit seemed like it what do you think people who read bicycling urban people who who talk about getting bike lanes in cities and how to increase the number of people who can commute on bicycles what type of stuff do you think they're interested in
00:43:06 ◼ ► guess what they're interested bikes. Yeah, it's pretty easy sell right how bikes helmets biking clothes lights and there's that it is literally an industry and maybe they would like to advertise there just doesn't seem that complicated. It just also just doesn't seem like a license to print money right it's right it it's it's the nature of being in the business it's a good business it should be fulfilling work it's such an important resource
00:43:35 ◼ ► on the internet read it it really there's nothing else quite like it. The fact that it's not the most profitable thing in the world shouldn't be the end of the world for them but it seemingly drives all decisions. Yeah, it's like if you're if you're not profitable maybe maybe focus on that in the short term rather than like catapulting to the moon.
00:43:55 ◼ ► Anyway, let me take another break here while I'm exasperated with this positive attitude. Tell you about our good friends at collide collide k o l i d e. They've got big news if you're an octa user, they can get your entire fleet of devices to 100% compliance how well if a device in your fleet isn't compliant, then the user whose device it is can't log in to your company's cloud apps until they have finished fixing the problem.
00:44:23 ◼ ► It's that simple collide patches one of the major holes in zero trust architecture, which is device compliance. Without collide it department struggle to solve basic problems like keeping everybody's operating systems up to date with security updates or keeping their browsers up to date with the latest versions to because it's if not as important as OS security updates or browser updates to patch security holes.
00:44:48 ◼ ► Browsers not up to date, you can't log in, they tell you collide will tell your user. This is why because your browser is out of date. Here's how to go update the browser. They update the browser. Now they can log in and they haven't gone to the IT department for help.
00:45:03 ◼ ► The it's teach a person, give a person a fish they eat for a day teach a person to fish they eat for a lifetime collides in the business of teaching your users in your company, how to fish where if by fishing I mean not getting fished with a pH, which is something they help with to.
00:45:22 ◼ ► They are the only device trust solution that enforces compliance as part of authentication, and it is built seamlessly to work with Okta their method collides method means fewer support tickets, less frustration and most importantly 100% of fleet compliance.
00:45:39 ◼ ► Visit collide.com/thetalkshow to learn more or book a demo. That's K O L I D E dot com slash the talk show. Let's zoom back out to the origins of Apollo. So when did you first get the itch to write a Reddit client?
00:45:58 ◼ ► I can actually remember pretty clearly I was walking home from school or university one day and I think alien blue is like the hotness at the time, which read it then eventually turned into the official app and it was a great app.
00:46:10 ◼ ► I just for me I wanted something that felt a little bit more like if Apple themselves had built a Reddit app like something that really like took the iOS human interface guidelines to heart and kicked in all the cool iOS new things.
00:46:21 ◼ ► And I kind of got to work on playing around with that would look like and I got something like up and running a little bit and some designs and I posted that to the Apple and the iPhone subreddits and that thankfully resonated with a lot of people.
00:46:32 ◼ ► It was something they were interested in. So I that summer I think that was after I graduated university that I ended up making the post and I was kind of like, okay, this has legs.
00:46:44 ◼ ► I think I want to kind of see where I can take this rather than going and getting like a proper job. I'll see if I can see this through and yeah, it's kind of been history since I think a year and a half later.
00:46:52 ◼ ► I released the actual first build took a little longer than I thought but it ended up blowing up even more and people were really happy to see it.
00:47:00 ◼ ► Did you have it in mind to make it a like a paid product from the get go like hey, I should I could do this if I do this because there are the Reddit client is free. So when there's already a default first party client to be free it.
00:47:15 ◼ ► It's it it seems more natural to think my third party client could be a paid product that I could maybe build a business around.
00:47:23 ◼ ► A hundred percent it was like funnily enough when I started I took so long to finish building it that when I started there was no official app alien blue was independently run and then over the course of that like one and a half years or whatever it got bought out.
00:47:36 ◼ ► Turned into the official app sunset and then the official app came out in its own right. So there was a whole saga of things that happened in between there.
00:47:44 ◼ ► But my goal from the outset was kind of alien blue. I think had like a one-time unlock for like a pro version where you got some extra features and that was kind of just what I am elated because people seem to like that use the app for as long as you want and then there's some extra stuff that if you like that you can support the developer and hopefully you can keep building this for a while.
00:47:59 ◼ ► And again reminiscent of Twitter where Twitter's original first party client was Tweety from Lauren Brikter the right sort of well Twitter if it was first but Tweety sort of broke out as as the sort of the most iOS E Twitter client famously Lauren invented pulled a refresh which is hard to believe hasn't been part of the iPhone from day one but was actually originally in Tweety.
00:48:27 ◼ ► You'd get to the top of the timeline and instead of a reload button just pull down and a spinner would spin and if there were more tweets that it would reload. It was so early days that sometimes there weren't more tweets.
00:48:40 ◼ ► I remember that where it's like you'd pulled a refresh and it'd be like no you're caught you're still caught up to close close this go do something else.
00:48:48 ◼ ► It's been a long time since Twitter ever ran out of new tweets, but then Twitter bought Tweety hired Lauren and very very quickly changed the app into something else.
00:49:04 ◼ ► Yeah, it's funny. I actually like I back in high school. I was like, I love Tweety and I was like, this is really cool. I want to I'd love to build an app of my own. I ended up emailing Lauren and he gave me like a few books to like start to get started out with and wish me luck and it was kind of like a first starting step for everything for me for him.
00:49:20 ◼ ► So he always seemed like the nicest guy he kind of did. I hope he's doing well. No, he is. I just talking to him recently because he actually built well not single-handedly but was sort of primarily responsible for building the website for the make something wonderful.
00:49:36 ◼ ► Steve Jobs book. Oh, wow. It was Lauren Brikter who did the which is in my opinion if you're not going to if you didn't if you're not lucky enough to get one of the print versions the website is a much better experience than the e-book because there's it's just much more dynamic.
00:49:51 ◼ ► But yeah, Lauren still around he's doing well. That's yeah that website was incredible. It's very well and when you know that Lauren Brikter had a hand in it, you're like, oh, yeah, of course look at these nice things.
00:50:04 ◼ ► Remember how nice to I this it makes me so sad because it's I always think first of Tweedy is an iPhone app, but it was sort of at the end of his keeping it as his own little product, but he came out with an iPad app and it was like the most innovative iPad app.
00:50:22 ◼ ► So I think still to this day that I've ever seen because it wasn't just take the iPhone Tweedy and make it fill the iPad screen. It was like a totally different metaphor for sort of layering different sort of tabs on the left and the structure on the right.
00:50:40 ◼ ► It's so hard. It's impossible to explain out of audibly here on the podcast but was so totally different and nothing at all like a big iPhone app and nothing like a Mac app just put on a touch screen on iPad.
00:50:55 ◼ ► It was just touch first very innovative and Twitter just scrapped it just it's funny and it kills me too because in continuing my Lauren Brikter homage like Apollo's current iPad app isn't quite where I'd want it to be in it.
00:51:09 ◼ ► And in some ways it is a really a big blown up iPhone app and what I've been working like on for like a lot of the last year was almost building something that reminded me of Tweedy and that you had like the multi-pane swiping around and then it's made me realize how difficult like that is to do now that like for him to have done that back then is just a work of art.
00:51:25 ◼ ► But yeah, I was I think like if I had September I would have been able to ship something like somewhat reminiscent of that and it's yeah. Yeah, there's not a lot of iPad apps that work like that nowadays and it's a shame.
00:51:41 ◼ ► It was a friend of mine. We were kind of just walking around campus talking and I couldn't names are the hardest thing and I was like Reddit is space themed.
00:51:49 ◼ ► You're kind of like traversing space and exploring it through this app and we kind of got on the topic of spaceships and I can't remember if it was me or him but one of us said like, okay, how about like Apollo?
00:51:59 ◼ ► That's a cool it's kind of powerful. It's a nice name like sometimes you just hear a name and you're like, that's it and it instantly felt right.
00:52:05 ◼ ► Yeah, it's hard to come up with a good name and it's for anything and it's also hard to explain why a good name is a good name.
00:52:12 ◼ ► It just it just is trust me as the guy who came up with daring fireball. I know it's hard to explain but when you when you hear it one of the neat things that to me Apollo sort of trailblazed.
00:52:24 ◼ ► It's at least one of the apps that I first became aware of this as a successful monetization strategy is having access to alternative app icons as a paid tier.
00:52:43 ◼ ► It crossed a hundred earlier this year. So I think yeah, like if I go to have had all the unreleased ones. I think it probably a close to 120 I think it's it because another thing.
00:52:53 ◼ ► Names are hard icons are hard and anybody who's ever developed an indie app or had anything to do with an indie app knows how passionate users are about app icons.
00:53:08 ◼ ► Oh a hundred percent and and sometimes trends change they change in the world around us graphic design wise obviously iOS 7 which were 10 years into now, but it's was a complete reset of the visual aesthetic of iOS.
00:53:26 ◼ ► I mean could not be more of a wipe the slate clean moment in design those things happen apps that are around long enough will live through transitions like this.
00:53:37 ◼ ► There's no app that it's my mom taught me this years ago. You were talking about your girlfriend turning 30 like when you cross these things people the older you get the more people feel a little like bummed like 40 now I'm 50 but it's like my mom taught me years ago.
00:53:53 ◼ ► I forget if it was when she turned 40 maybe when she turned 40, but she just said well, it kind of sucks getting old but either you turn 40 or you die before you're 40 and I'd rather turn 40 than die.
00:54:08 ◼ ► Right and so apps are like that right? It's like you're either around for a long time and you have to go through these transitions where you change the look of your app to stick with modern times or the app isn't around anymore.
00:54:22 ◼ ► People get so angry when app icons change sometimes because you can't please everybody all of the time and this idea that you could supply a dozen two dozen or like in Apollo's case a hundred different icons.
00:54:38 ◼ ► It's brilliant because you kind of can make everybody happy you can have an app. I'd like to meet the person who subscribes to Apollo and still can't find an app icon that they like.
00:54:52 ◼ ► Honestly, yeah, and it's so and it's like one of those things that in hindsight like makes so much sense because you look at like the video game industry for instance and people love having those like cosmetic skins they can like pay to customize their character for and it's great because it doesn't necessarily make the game worse for anyone else.
00:55:06 ◼ ► It doesn't feel like an inherent unfair advantage. It's just a nice little thing to customize it and people took to it really quick and I don't think to this day.
00:55:12 ◼ ► I've had like really any complaints where people are like I have to pay for this. It's just it's a fun thing.
00:55:16 ◼ ► Yeah, it's the app equivalent of like the Fortnite style of gameplay right and people are happy to pay for those costumes and custom dance moves in Fortnite.
00:55:28 ◼ ► But it doesn't give you any advantage in the game and because it doesn't give you any advantage in the game everybody feels good about it right so it's like oh I didn't win I lost but you know you didn't lose because the guy you lost who paid more money.
00:55:42 ◼ ► Right exactly. That person they just beat you fair and square and just look better doing it. Yeah they just look better doing it or maybe you looked cooler on your way to losing because you had a costume that you really liked.
00:55:58 ◼ ► But it's so funny because we like to think human beings we all like to think we're so rational especially when it comes to money and I was talking about that with trying to set the valuation of a company for an IPO.
00:56:09 ◼ ► It's not entirely rational right there's there's an emotional context to it and our purchasing decisions as human beings are like that too and people would like to think that they pay for features right and it's like I'm a rational human being and if I'm going to pay five dollars a month for a pro Reddit client.
00:56:28 ◼ ► It's going to be because of technical features like a superior threading presentation or the best markdown editor in any Reddit client possible.
00:56:39 ◼ ► But the truth is people will pay happily it just satisfies a part of your brain to pay for cool icons.
00:56:46 ◼ ► Oh 100 percent yeah there's some synapse that fires off when you see the right colors come together.
00:56:51 ◼ ► Right and for you personally it's also must be very fun because you've gotten to meet like and develop a professional relationship with like what all probably all of the top icon designers out there who.
00:57:05 ◼ ► Yeah I guess that that takes I take some solace in the fact that if it all goes belly up now I don't think there was really anybody that I didn't get to talk to that I didn't get to work with because yeah that's the best part and like if you're like I'll get sometimes questions be like oh like if you're looking for an icon designer like who should I go with.
00:57:20 ◼ ► And I can just kind of say scroll through the icons and Apollo pick out when you want and like look at the who the designer was and reach out to them like it's almost a gallery.
00:57:26 ◼ ► Right and that's one of the things I also like about the way you do it is that the credit is right there. I love this is why I like the way I link to things on Daring Fireball I don't just say the New York Times reports this I like to say so and so reporting for the New York Times so that the reporter gets credit and both to give them credit.
00:57:44 ◼ ► And so that readers of my site can kind of get to know some of these names and like either know oh that's a good that's somebody who I really like at the verge or this is a reporter at the Washington Post who always makes me roll my eyes so I'm already braced for it but you know credit is important.
00:58:00 ◼ ► I love that you give the credit. Yeah but at the end of a movie. Yeah it's nice to see. Well it's more like the old movies right like as it used to be like you could get through the movies and in the credits would take one minute.
00:58:12 ◼ ► And everybody was like involved with the movie now that the credits are like eight minutes long and it's like, again I'm not putting down the caterers of the movie or the people who drove the assistant director to the set, but it's like do they really need to be in the credits.
00:58:30 ◼ ► I don't know. The line has to be drawn somewhere and to me, Hollywood sort of let that line be erased and made the credit something, you know, it's always a source of contention in my family. My wife has never been a fan of my like when we go to the movies I like to watch the credits.
00:58:47 ◼ ► Well I've had to give up on it. Even as somebody who used to try to stick through the credits, because I always thought they were part of the movie, you know when they're literally eight or nine minutes long, I mean forget about it. I mean I gotta go pee.
00:59:03 ◼ ► Oh, 100% but it's like, and a lot of the time on streaming services they don't even give you the option anymore it's like, on to what's next, right show you like the first three seconds of the credits before it jumps somewhere else.
00:59:17 ◼ ► I'm going to talk about my friend Brad Ellis, he does all the, he does the monthly covers for dithering, me and Ben Thompson's podcast does great work and Brad is a just a top notch person he's just a personal friend, he's super talented. He amazes me though because he's one of these designers who doesn't have one style at all, he could do anything.
00:59:38 ◼ ► He had these Reddit, or not Reddit, Apollo icons in the works that sort of were like in the style of the monopoly chance and community chess cards. Oh, they're so good. It's like the Reddit mascot in the style of the monopoly man I forget his name Mr. Pennyworth or something like that.
00:59:58 ◼ ► Right, right. The guy with the top hat monopoly like, and it just exquisite like cardboard texture design. Oh, just so and he did my favorite thing where you're like can you do an icon for me and he comes back with like three incredible ones, you're like, Oh, that's perfect.
01:00:13 ◼ ► And it's, yeah, it's it and that's my favorite part of working with people is like I always get the question like what do you tell them as like the design brief and like with people like Brad especially like you can kind of just go like go buck wild like as long as it vaguely resembles Apollo like, like your imagination flow and they come back to you a few weeks later like the coolest thing.
01:00:30 ◼ ► Yeah. Let me take another break here and thank our next sponsor. Well, while I'm thinking of and it's our good friends at Squarespace. You guys know Squarespace. It's the all in one platform where you can build a website. And it's got everything you need for your own website, starting with even just at the before you even start designing the website, you need a domain name.
01:00:51 ◼ ► Squarespace does domain name registration. They have a great domain name registration feature. Get your domain name, pick a template. They have hundreds of professionally designed templates to choose from. You can just use them as is swapping your website's name, your logo, that sort of thing, or start tweaking it. You can tweak it a little. You can tweak it a lot.
01:01:12 ◼ ► If CSS and design, you can design your own template completely from scratch. So you can just use a complete out of the box template. You can start with one out of the box, tweak it a little. You can do your own custom design from the ground up or you can meet somewhere in the middle. It's entirely up to you.
01:01:30 ◼ ► The level of technical expertise you want to use is entirely up to you, but it works all the way up to the level of not even if you don't even know the difference between HTML and CSS, you're fine because you can use Squarespace in a completely wizzy wig drag and drop manner.
01:01:45 ◼ ► What else do they have built in? They've got analytics where you can see which parts of your site people are coming to, where they're coming from in a really good presentation that is really easy to understand.
01:01:58 ◼ ► Online stores, if you want to sell stuff, you can do that at Squarespace. If you want to sell your time because you're doing consulting or something like that, or maybe you know somebody who's a trainer and you do training like fitness training over the web, you can do that.
01:02:15 ◼ ► Anything you can sell, you could sell through a Squarespace website. Email campaigns, more. Anything you can do, you want to do, you can do. Where do you go to find out more? Go to squarespace.com/talkshow.
01:02:31 ◼ ► You get a 30-day free trial with no limits, no watermarks on the website as you build it for a whole month, and then when you're ready to launch, just use the offer code "talkshow," same thing as the URL, and you get 10% off your first purchase. Go to squarespace.com/talkshow for more.
01:02:48 ◼ ► All right, you mentioned that you were vaguely, I guess we have to start talking about it in the past tense, unfortunately, but vaguely dissatisfied with the iPad version of Apollo. And I don't know why. I've always thought of it as an iPhone app first. I don't know.
01:03:05 ◼ ► I guess it's just where I tend to use it. But no Mac app. And I'm curious just to ask if you ever thought about that, or were you too busy already, but especially with the last few years with Catalyst and SwiftUI and things that make it easier to share a code base across iOS and Mac, whether you ever looked at it or what your thoughts are.
01:03:28 ◼ ► Well, I was biding my time because I finally found that AppKit historically hasn't been, I think the Mac apps built that way are just wrong, and I've been waiting for SwiftUI. No, I'm just kidding.
01:03:41 ◼ ► No, I have. When I was redesigning the iPad app to hopefully take advantage of the iPad a little bit better, I was in the back of my head trying to design it in a way that would also grow to the Mac. But it was kind of just a factor of one man shop figuring out that the iPhone has X percentage of users, the iPad was 5% of that, and the Mac was like, the people who currently run the iPad version blown up on the Apple Silicon stuff, they were 1%.
01:04:06 ◼ ► So it was kind of just slowly walking down the stairs to try to figure out how to hit everyone, and it's one of those things where I would have loved to have got there. I use my Mac probably more than I use my iPhone most days, and having something there would have been really cool. But yeah, just never quite got around to it.
01:04:20 ◼ ► And I can't help but think that part of it too, it's like a Mac app can be the most intricate, truly major app, like Xcode, right? It's truly one of the most complex apps for any platform anywhere in the world. It's a super complex app.
01:04:41 ◼ ► The full version of apps like Photoshop tend to be bigger, the iPad versions are often subsets. We see that even with Apple's own apps, like with the new to the iPad versions of Logic and Final Cut Pro are impressive for iPad apps, but especially Final Cut is a subset of the Mac app.
01:05:03 ◼ ► But conversely, the other thing that's weird about desktop platforms like the Mac and Windows is web apps work better there than they do on mobile, right? So like just using the reddit.com or for a lot of people old.reddit.com in a browser on the Mac is closer to what you expect and want anyway than using the website on the phone.
01:05:30 ◼ ► I think traditionally, yeah. I almost find reddit is a very specific case where if you're not using old.reddit, I find the new site, while technically impressive in a lot of ways, is quite slow.
01:05:41 ◼ ► And having done a lot of API requests through reddit, you can kind of know how long it takes. I don't know if many people know, but you can put just .json at the end of a reddit URL and it'll load all the JSON.
01:05:52 ◼ ► And so you can kind of, if a post is taking a while, you can see how long it takes just to get the raw JSON response from reddit and the rest is them kind of doing all the other stuff. So like out of a selfish reason almost, I've long wanted a way just to say, get me the JSON and represent that in a way as fast as possible just so I can quickly browse reddit because I'll be browsing the current site sometimes and it just takes forever to load a comments thread.
01:06:12 ◼ ► And I think even just building an app based off of that foundation alone I think would appeal to a lot of people.
01:06:18 ◼ ► I just, just a couple episodes ago on this show, I had Neil Gervarry from MimeStream on and very similar, you know, MimeStream is a Gmail client just for the Mac. So it's sort of the inverse of Apollo where it's Mac only for now.
01:06:33 ◼ ► But people were like, oh, that was a good interview, very interesting and he's a very interesting, very friendly person. But people who hadn't tried MimeStream are like, well, I still don't get it, why would I spend $50 a year when I can just type gmail.com in my browser and I've been doing this for a long time.
01:06:50 ◼ ► And it's like, you don't understand how much faster MimeStream feels, right? And it's like, to me, I mean, and I'm super picky about the Mac specific UI idioms and having all the Mac consistency and keyboard shortcuts and support for system services and stuff.
01:07:09 ◼ ► But even putting aside that MimeStream is a very good Mac app and adheres to the Mac style of design, although I think that's partly what makes me faster, it's just a very fast app, right?
01:07:21 ◼ ► And the other side of those comments I got from people who listen to the show is they're like, I tried it after I listened to your show, and I've been using Gmail forever, I can't believe how fast this is.
01:07:30 ◼ ► And they thought, like, I don't know what people think, people who aren't programmers, I got a couple of emails like this about MimeStream in particular, where they thought it would definitely be slower, because wouldn't just typing gmail.com always be faster?
01:08:02 ◼ ► There's no less complex and no less back and forth between the device you're on and Google servers than a client, and making fast software is, it's a skill.
01:08:17 ◼ ► Absolutely. And I feel like you said with old dot Reddit, to a certain extent you can see where the advent of JavaScript and loading 8 zillion things has kind of perverted the web, and you can see how Reddit in a lot of ways is much slower than it was,
01:08:29 ◼ ► and it's a version that I don't know if they've touched since 2009, and it becomes a weird tradeoff of features versus, do you want the thing to load in the first place?
01:08:38 ◼ ► Right, and I can't help but think, it's on my list of questions to ask you, why are there adherence to old dot Reddit dot com?
01:08:47 ◼ ► And I don't think it's because of the way it looks, even though people get used to the way something looks and they don't want it to change, I think it's just that it's faster.
01:08:54 ◼ ► Yeah, I think that's a lot of it. And there's so much cruft I find on the normal website, like with NFTs and all these different things they're trying to try.
01:09:04 ◼ ► It's a new thing every week, which I think is a direct result of them trying to find new ways to monetize.
01:09:08 ◼ ► At its core, the old dot Reddit is kind of like what gave the Reddit magic spark from the get-go.
01:09:14 ◼ ► It's just like, you're just posting links, you're just reading comments, it's just this inherently simple concept, and it's kind of drifted away from that more and more as they try to nail more features on top of this simple foundation.
01:09:37 ◼ ► But I don't think people who just use Gmail have as many complaints as Reddit users do.
01:09:47 ◼ ► It doesn't feel weird, until recently, until the last month, it didn't feel weird that Reddit kept old dot Reddit around, even though that's an unusual move for a company that totally redoes the way they do their website.
01:10:02 ◼ ► But for Reddit it felt natural, because yeah, that feels like something they would do to make the community happy.
01:10:08 ◼ ► But I saw a couple of comments in the Ask Me Anything that CEO Steve Huffman did, disastrous in my opinion.
01:10:18 ◼ ► Oh yeah, I'm sure you did. But I saw people mentioning old dot Reddit dot com, and their official line is that they don't have any plans for it to go anywhere, and there were dozens of people saying, "Yeah, right."
01:10:39 ◼ ► I think putting third party clients out the door seems like, at this point, old dot Reddit is, even though it's from them, it's more like a third party client from their perspective, from the reasons that they're pushing you out, that it doesn't have the metrics that they're measuring, it doesn't have a zillion JavaScript libraries.
01:11:01 ◼ ► I wonder if they're thinking to force apps like Apollo out the door, it seems like it would apply just as much to old dot Reddit dot com, and it seems like people who love old dot Reddit dot com, or at least just prefer it, totally, you can smell it coming, right?
01:11:18 ◼ ► And I think Reddit's almost unique in that their entire upstart was from when everyone migrated from Digg because I think they had a similar redesign issue. So I think Reddit was acutely aware that if we botched this redesign and have no way for people to go back, this could shut down everything.
01:11:32 ◼ ► In that same way, I thought that would almost be their strategy with third party apps where they'd say, we're going to charge you an amount that is so expensive that you'll lose 95% of your users who don't want to pay, but you'll still be able to stick around for those who do, and in doing so will shrink your monthly active users from 1.5 million to 50,000 or 100,000.
01:11:52 ◼ ► And then you'll basically have the old dot Reddit effect where, sure, it's still around, but the amount of people that you've choked it down to is so low that if they go one step further at that point, then kill it, it's going to be a lot worse than if they did it from the outset.
01:12:04 ◼ ► But it seemed like they skipped that step with the third party apps, I don't know, intentionally or otherwise, to go right to the death blow.
01:12:15 ◼ ► Overall, it was just a very… I think I said on a different interview that that was when I was flying home from DubDub, and I had to pay 10 bucks for airplane Wi-Fi for that, and I was really disappointed that I felt like I didn't get my money's worth.
01:12:29 ◼ ► I genuinely thought… I've had honestly a high opinion of him as a CEO historically just because he is one of the founders, and I think he's understood how Reddit ticks really well, and I expected him to take the really easy slam dunk win there and just go in and say, "Look, we did not handle this well.
01:12:46 ◼ ► We're going to go back to the drawing board, talk to developers, figure out how we can make this right. We hear you guys, and we apologize for how we handled this."
01:12:53 ◼ ► And I genuinely think the community would have been like, "Okay, he heard us. There's going to be steps going forward. This is reparable."
01:13:00 ◼ ► But just to go completely so far in the other direction of just doubling down and being accusatory and…
01:13:08 ◼ ► It's so strange to me that he seems so personally angry about stuff like me recording and leaking that phone call of the employee with him, where it's like…
01:13:23 ◼ ► I genuinely love to ask him, in the most polite way possible, genuinely, if the roles were reversed and I said something potentially like a career-ending accusation against you as the CEO, and then you realized, "Oh, I've got evidence that shows that to be completely fine."
01:13:43 ◼ ► If the answer is yes, why are you so bothered by it? Why wouldn't that be an opportunity to say, "Oh, crap, I really messed up here. Let's dial this back and fix it."
01:13:51 ◼ ► So I guess that would have been my favorite part, where I could not understand the strategy of it at all.
01:13:58 ◼ ► Yeah, so for people who don't know what you're alluding to, you've had some communication with Reddit during this process.
01:14:11 ◼ ► Right, but at one point, in talking about this, you mentioned something jokingly like, "Well, look, if it's about…" And it seems to be before they backtracked from saying, "It's not really about server costs, it's about the opportunity cost of what we could be making per user if they weren't using these clients."
01:14:30 ◼ ► But when it seemed to be talking about actual server costs, you said something to the effect of, "Well, look, if you think Apollo's usage is going to rack up $20 million a year in costs on your point, why don't we just meet it halfway and you buy it for $10 million and we'll call it a deal?" As sort of a joke?
01:14:53 ◼ ► Yeah, kind of. And it was almost like that was like when I made the post saying the $20 million figure, that was half of what people were saying too, was half the comments were like, "If they're saying you're an opportunity cost or a server cost or what have you of such a massive amount, wouldn't the obvious call be just to acquire you and integrate you into the existing app?"
01:15:09 ◼ ► And so it was like the very end of the phone call and I kind of just vocalized that and they kind of took it the wrong way at first and then kind of asked for clarification. I was like, "Okay, no, what I mean is Apollo uses a lot of API requests, wouldn't this just be easier for both parties if you basically acquired it?"
01:15:24 ◼ ► And they were like, "Oh, okay, so we see what you mean. Okay, apologies for misunderstanding that. Okay, yeah, that sounds good."
01:15:29 ◼ ► Their first impression in this bit of seconds, for like 30 or 40 seconds, they misinterpreted you as saying, "Why don't you just give me $10 million and I'll stop making a fuss?" Which is a shakedown and that's not what you meant at all.
01:16:21 ◼ ► I got the most random Macedon message from, I assume it would have been a Reddit employee saying, "Can you respond to the allegations internally that you're trying to shake them down for $10 million through blackmail?" And I was kind of like, "What?"
01:16:34 ◼ ► And then a week passes and then Steve did a call with a bunch of moderators and said a similar thing like he was trying to coerce us and blackmail us and threaten us. I thought we were very clear here, like you apologized. So it was very strange.
01:16:50 ◼ ► Yeah, and the term "gaslighting" gets thrown around too loosely these days. It's like we didn't, I know that it's a reference to I think a movie from, I don't know, like 80 years ago. So it's not a new term.
01:17:04 ◼ ► But I don't know, in the last five, six years, it just seems like it's a term that came into vogue and it filled a void in our language and then it suddenly got overused. But this is a case where it really applies, where it's like, you clarified this on the same call and you recorded it, which is legal for you.
01:17:31 ◼ ► Is that all of Canada or is it just like in Nova, it's like a Canadian law, one party consent. If you're on the phone, it is perfectly legal to record the phone call. It's very clear from the, you can read the transcript, you can listen to the call. It's very clear. This wasn't like an ongoing thing where Reddit spent days thinking you were trying to shake them down. It was clarified instantly.
01:17:50 ◼ ► But the only honest way to approach the fact that Steve Huffman was still sort of going with it would be that he read like the first 15 seconds of this transcript and never read the rest of it and just went with it.
01:18:07 ◼ ► And I find that very hard to believe. I honestly, it's very hard for me to walk away from his disparagement of your negotiations and alluding internal to Reddit that you were trying to shake down the company.
01:18:23 ◼ ► It just seems disingenuous. And I'm very generally, I try to be very, very hesitant to accuse anybody of being a liar or dishonest, but it really, there's no other conclusion I can come to. He's either dishonest in a very personal way, which angers me, or he spouted his mouth off in a disparaging way when he, all he would have had to do is read another 30 seconds of transcript.
01:18:58 ◼ ► No, and I even gave multiple opportunities to talk about it because after that first, I got that first kind of anonymous message where somebody was alluding to it. I reached out to him and said, "Hey, if there's any miscommunication here, I want to clear that up." That was not the case.
01:19:12 ◼ ► And then days go by and then he says it again. And it's kind of one of those cases where you've been given opportunities, I corrected it on the call, sent you another email, and then gave you the audio.
01:19:24 ◼ ► And then at this point, for you not to say, "There must have been some wires crossed here where I didn't understand everything. I apologize." To keep going on it was just kind of bewildering.
01:19:35 ◼ ► I did feel gaslit for a bit where it was kind of like, "Man, if I didn't have the foresight to record that, it would have been a really gross situation for me where you have a multi-billion dollar company CEO going, "Hey, this guy over here is blackmailing our company for 10 figure money."
01:19:51 ◼ ► That could have been potential. You Google my name, I'm going to a job interview, and they're like, "Oh, it's that guy? The guy who blackmails companies? Yeah, we're not going to hire him."
01:19:58 ◼ ► And I'm like, "Thank God I did record it because, yeah, a game of he said, she said with a billionaire company isn't going to typically go too well."
01:20:07 ◼ ► Right. Or even if it's not even a hiring type situation, if it's just, "Do we want to grant this person developer APIs to our product?"
01:20:15 ◼ ► Or, "No, no, this is somebody who you let them in the front door and they start shaking you down."
01:20:22 ◼ ► Everybody knows there's people like that out in the world, but it's just so antithetical to Reddit and their brand and the sort of symbiotic relationship they have with the community overall.
01:20:39 ◼ ► Because it is. There is no Reddit without a community. It couldn't be more community driven.
01:20:47 ◼ ► And it's just a weird, almost unique thing on the internet, but part of that is trust, right?
01:20:54 ◼ ► In a way that like Facebook, for example, just doesn't need that level of trust. It just is what it is.
01:21:01 ◼ ► And what else are you... They kind of have the position of, "Well, what else are you going to do?"
01:21:06 ◼ ► Right? Because there is no other platform with three billion users all on the same platform.
01:21:12 ◼ ► Nobody else has that. Whereas with Reddit, there is sort of the, "Well, what else are you going to do?"
01:21:18 ◼ ► And then we saw it this week, right? We saw pretty significant protests across all of Reddit.
01:21:23 ◼ ► But I'm curious now, as we record this show, it's Thursday, June 15th, it does seem like it's sort of...
01:21:31 ◼ ► I hate to say it, but it does seem like his sort of, "Hey, this is going to blow over."
01:21:40 ◼ ► And I think I said something to that effect in my post where I made that shutting down post before the separated blockouts were to start.
01:21:52 ◼ ► But it's one of those things where the management at Reddit seems so sticking their heels into this and unwilling to kind of meet the community.
01:22:04 ◼ ► It's not something that I see them caring about, honestly. I'd love to be wrong, but yeah, I just didn't want to get any expectations up just to have it come crashing down again.
01:22:13 ◼ ► So it does seem sad that they don't want to listen, but I don't think it's the most surprising given the actions so far, I suppose.
01:22:21 ◼ ► Yeah, the protests are playing out. I know some people were hoping they would stay in perpetuity until they relented.
01:22:31 ◼ ► And I didn't think that was realistic, but I thought that more subreddits went dark or private, whatever you want to call it, for a day or two days and maybe even a third day than I expected.
01:22:45 ◼ ► Especially non-technical ones. Like, it's not surprising at all that our Apple shut down and the ones that you would think would care the most about Apollo in particular or just third-party clients in general.
01:22:59 ◼ ► But there are funny, I think, shutdowns. I mean, some of the very top subreddits, ones with, I don't know, like 40 million active users. I mean, it was impressive.
01:23:10 ◼ ► I didn't think, though, that it was going to last. It's just not the nature of how a protest like this works.
01:23:17 ◼ ► But it certainly, if I were an executive or in leadership at Reddit, even the fact that it does seem to be passing in terms of actual subreddits staying private, I would be in a panic because it's not a place you want to be where you've left a huge swath of your community with a very bad taste in their mouth.
01:23:45 ◼ ► Yeah, I think it's had a significant effect on Steve's legacy as CEO. And it's one of those things where it's like, you almost have to wonder if it's been worth the cost because I feel like this will kind of just come up whenever anybody's talking about stuff now. It's kind of this blemish on Reddit.
01:24:03 ◼ ► And it was for what? Like, if you came back to the table and said, "Look, we apologize for how we handled this. We want to give you more time to make sure this transition can go off without your companies completely shutting down because we recognize 30 days was not nearly enough time."
01:24:18 ◼ ► I feel like that would have been such an easy win that would have kind of reversed everything. To have this massive stain just because you didn't want to give a solitary inch is just, even just from a strategic standpoint as a leader, I don't understand.
01:24:33 ◼ ► Yeah, I don't understand it either. And it's like I said earlier, you can't put a dollar figure on reputation, but it is valuable. It's incredibly valuable. And speaking of reputation, the other thing, aside from this clearly provably, honestly, you have a recording.
01:24:49 ◼ ► I mean, like as Jason Snell said, you brought receipts that you did not try to blackmail them into buying your silence or support or whatever. But there's also the broader implication and the one that I don't think it can't really be backtracked or proven false with like, "Ah, it's the Perry Mason courtroom moment. Here's the recording of a phone call that proves it otherwise."
01:25:14 ◼ ► But it's the implication that Apollo is, honestly, for lack of a better term, poorly engineered, that it's noisier than the average Reddit client would be, more API calls per user.
01:25:30 ◼ ► And you've made the point in your posts on this that one of the reasons is that your users are, the reason they paid for a paid client for Reddit is that they're highly active Reddit users, right? I mean, can you talk me through this?
01:26:00 ◼ ► There are a lot of posts that confuse me there where it's just like, even at the high level, I think I made the analogy on Quinn Nelson's YouTube channel too, but it was kind of like, historically, I guess as a reference point, Apollo's average calls per day per user is about 350. And Reditt's, up until two weeks ago, their rate limit, they're saying like, "This is how many times we'll allow you to interact with the Reddit API was 60 times per minute."
01:26:24 ◼ ► So if you do the math over the course of the day, that's 86,400 times a day. So if Apollo's got a limit of 86,400 and it's doing 350, that's 0.4%. I would argue that's quite efficient.
01:26:39 ◼ ► And I use the analogy that if you scale those numbers, like if I was to borrow your car and you said, "Don't drive more than 864 miles," and I return it with three miles, I would say I did a pretty good job. But Reddit's saying, "Oh yeah, but Bob over there only used one mile."
01:26:54 ◼ ► And they're saying, "That's great, but I didn't know that was kind of the measuring stick we were going for."
01:27:02 ◼ ► Yeah, exactly. And I just don't understand why that's a comparison. And beyond that, it's like, yeah, there's also the fact that you can literally look at different swaths of my user base and say, "Oh, the people who pay a subscription use even more than that 350."
01:27:18 ◼ ► There's an amount here where people who enjoy Reddit use it more, and that shouldn't be surprising.
01:27:23 ◼ ► And then there's almost a third aspect where it's like, okay, if you think I am being inefficient in some ways in comparison to other clients, I do some things like I check your product to see if you have any notifications every so often automatically in the background.
01:27:36 ◼ ► And I check if I think you might try to load the next page. I automatically kind of load that ahead of time just to make it a little faster for you as the user and make things a little nicer.
01:27:44 ◼ ► Those are calls I could get rid of. If we're moving the tent pole as to what efficient is, I can get rid of stuff like that and drop that countdown.
01:27:53 ◼ ► But when you give us 30 days to do all that and completely revolutionize the payment model of how we're doing the app and transition all the existing users over, test all these things, get it through App Review, the laundry list becomes so massive to do in 30 days that it's not possible.
01:28:08 ◼ ► So it becomes such a confusing message from Reddit as to what you want us to do here realistically as developers. Like I said, April 18th was when they were thinking about doing API pricing.
01:28:19 ◼ ► And then June 1st, 43 days later, it took them that long to just even decide what that number would be. But then they only give us 30 days to actually implement that number.
01:28:28 ◼ ► It's just like I don't know how that passed a sniff test, let alone made it up the door.
01:28:33 ◼ ► Yeah. And speaking of sniff tests, I would say the accusation that Apollo is "inefficient" doesn't pass the sniff test. Right?
01:28:41 ◼ ► And it's like, you've written this, you've said it, you just said it here on this show, that if you had more time and they said, "We would like to recalibrate our usage guidelines from an average number per day to like a peak amount per five minutes or something like that."
01:29:02 ◼ ► That there's obvious things you could do, you'd be more than willing to do. Whatever they would ask you to do, you would try to do.
01:29:09 ◼ ► And in your best efforts to date, you've stayed well within their only published guidelines for how much API usage you should be using.
01:29:19 ◼ ► But the sniff test to me would be a huge part of the reason that people love Apollo is how fast it is. It is a very, very fast client. It is fast to load, it is super fast to scroll, which is super...
01:29:35 ◼ ► Anybody who knows, I'm like a very, very, very... My understanding of how to do complex table views in UIKit or SwiftUI or whatever is super...
01:29:45 ◼ ► I'm not an expert in that, but I know that it's hard, right? And it's like the complex scrolling views are just notoriously a source of slowness in iOS apps.
01:29:57 ◼ ► It's just a hard thing to do. And the nature of Reddit is incredibly intricate because it's got infinite hierarchical threading, right?
01:30:08 ◼ ► And to me, it looks simple, and the more you think about it, and the more... It's like you said, you set out to make the Apollo client that Apple would design if they made...
01:30:18 ◼ ► Or the Reddit client that Apple would make if Apple did a Reddit client. And I think Apollo absolutely fits that bill.
01:30:25 ◼ ► But when you really look at Apollo's threading view, you're like, there's really nothing in iOS like this with seven, eight levels of threading,
01:30:35 ◼ ► and how you use color to indicate the threading level. And doing threading hierarchically is you push in, you indent.
01:30:46 ◼ ► But the nature of a phone is that you've got this super skinny screen, right? It's a design problem. It is a technical problem, right?
01:30:54 ◼ ► And Apollo... The whole reason people love it. It's like it does this really complex thing really, really fast.
01:31:06 ◼ ► No, and yeah, there was weird calls, or one of the calls he said, "Apollo doesn't make any effort to be efficient."
01:31:14 ◼ ► And it's like, I can literally go back through my emails, and I have conversations where I've talked to Reddit and said,
01:31:19 ◼ ► "The way this API works isn't that efficient. It would be great if you could do it this way, or if the existing system has to stay in place.
01:31:27 ◼ ► Is there any way I could talk to someone at Reddit and make sure I'm optimizing this to the fullest extent?"
01:31:31 ◼ ► Because Apollo has a server component as well that has a lot of functionality to notify users when a new message comes in,
01:31:38 ◼ ► or a subreddit they're watching, for instance. And I have a really talented part-time server engineer who works on that.
01:31:43 ◼ ► And we've had so many conversations with Reddit where we've tried to make sure we're doing as best as absolute possible.
01:31:48 ◼ ► And for them to say they haven't made any attempts, it's another thing where it's demonstrably false.
01:32:02 ◼ ► So it does a few things. It's mostly around notifications. So if you have a Reddit account, Reddit's API is quite primitive in so far as if you want to know if you get a private message or somebody mentions you on Reddit, similar to Twitter.
01:32:15 ◼ ► The way the Reddit API works is I have to go to your inbox and say, "Is there anything new?" And it says yes or no.
01:32:21 ◼ ► And then I go, "Okay, so maybe wait 10 more seconds ago. Is there anything new?" "No." "Okay, anything new?"
01:32:27 ◼ ► And you just do that repeatedly. And if you only get a comment once every three days, 99.9% of the requests are going to be just saying, "No, stop asking me."
01:32:34 ◼ ► And the way a lot of more modern APIs get around that is they kind of flip the tables where it's an event-based, WebSocket API where you just connect and you just say, "I'm just going to sit here.
01:32:46 ◼ ► And that's something we've proposed to Reddit historically or said, "Is there a way that we could maybe get the JSON size of the response down so that the request would be smaller?"
01:32:55 ◼ ► And it's always been like, "Okay, maybe." And it's like those inefficiencies that you're kind of stuck with. But you kind of have to do your best within that system to make them work.
01:33:04 ◼ ► And there's other ones like if you were saying, if there's a Philadelphia subreddit and you want to know the next time tickets for an event that's local are being sold, you could add a subreddit watcher for Philadelphia and get notified when a certain keyword match occurs.
01:33:17 ◼ ► And that's one of those things that we try to be really efficient with too, where if there's 10,000 people watching the Philadelphia subreddit, we don't have 10,000 watchers.
01:33:24 ◼ ► We have one watcher that then just requests it every so often and then talks to those 10,000 people.
01:33:29 ◼ ► So there's like, it's funny, I open sourced the server just to show we do a lot of complex stuff to make sure that this is efficiently done.
01:33:37 ◼ ► And yeah, so it's a lot to get because there are published, you have to stay under this many requests.
01:33:44 ◼ ► That has been the limit they've imposed on clients so far. So we've done a lot of work to make sure we're being as efficient as possible within those limits. So it's weird.
01:33:53 ◼ ► Yeah, so it's the complete opposite of what they were accusing you of doing. When you're developing an iOS client for Reddit, to have a server component, it's proof that you were looking at ways to be more efficient, not inefficient.
01:34:10 ◼ ► Right, exactly. And I reached a limit as to what I could do efficiently on the server as an iOS developer. So I even went out and found somebody, his name's Andre, who's incredibly, he works full time as a server engineer.
01:34:21 ◼ ► And he came and revamped all my code, made it 100 times better. So there's been money and lots of time and code that's gone into making this as efficient as possible.
01:34:29 ◼ ► And hopefully by opening sourcing the server component, it was kind of showing my cards there and saying, you can go in and see the work that's been done.
01:34:36 ◼ ► Well, I have to say you've been not just on this show, but it shows even in your writing on the Apollo app subreddit that you're remarkably upbeat about what can only be described as a total shit sandwich.
01:34:55 ◼ ► I mean, you've got eight years, you've built an entire livelihood around Apollo, and you're just going to, as things stand as you and I talk right now, you're going to have to pull the plug in two weeks.
01:35:08 ◼ ► And your attitude is remarkably upbeat in my opinion, but I'm just curious, with the Twitter clients, well, Twitterrific hasn't come back, but Tapbot was able to sort of take the bones of Tweetbot and turn that into the foundation for Ivory on Mastodon.
01:35:29 ◼ ► And people, there are, I know there's Lemmy, which is sort of a, I think, activity pub, which is sort of similar, or same foundations as Mastodon, a sort of open source Reddit type thing.
01:36:00 ◼ ► It's tricky because it's one of those things where it's like, it would be possible, but it would be so much work.
01:36:05 ◼ ► And to be honest, the last few months have been pretty mentally draining, so it's not something that I have a lot of fuel in the tank for necessarily.
01:36:12 ◼ ► But even if it was, you look at Mastodon, and I can't remember what it was, but years ago, I think it was when Twitter did stuff with the tokens, there was kind of an initial burst to go to App.net and Mastodon.
01:36:23 ◼ ► So many of us had accounts from five years ago when Mastodon came back up, so there was kind of this already understood Twitter thing that was brewing in the background that had a foundation.
01:36:32 ◼ ► And with Reddit, I don't think there's anything quite to that extent yet. There's a few that are kind of spreading up, which are great to see, and I honestly wish them the best, but there's not anything as clear as Mastodon, I don't think.
01:36:43 ◼ ► And even if there was something as big and powerful as Mastodon, I've even seen, with Mastodon, it's unfortunate, there's still so many people I follow that just haven't made it over there for whatever reason.
01:36:55 ◼ ► The vast majority have, but if I'm thinking even the most technical iOS developers I know haven't been able to maybe necessarily grok Mastodon for whatever reason, it kind of creates this fear in my mind that I'm like, is this going to be the most long-term thing?
01:37:11 ◼ ► As much as I love it, is that a hurdle where it's the de-federated, decentralized thing? As amazing as it is to me, is that going to be the same thing for everybody? And kind of investing all that time and effort into something that I'm not 1000% sure will have long-term legs, even if I wish it did, is like, I don't know if I have the energy to commit to that right now.
01:37:31 ◼ ► Yeah, and the other thing that's different, in my opinion, is as unhappy as the broad Reddit community, again, like I said a couple minutes ago, it's not just the nerd subreddits that protested, it really went Reddit-wide for a couple of days this week to voice their displeasure.
01:37:51 ◼ ► As unhappy as people are, fundamentally, Reddit is still Reddit, and whatever drew people to Reddit is still there.
01:38:05 ◼ ► And the difference with Twitter is that under Elon Musk's ownership, he's continuously steering Twitter. From the moment he took over till we record right now, he is steering Twitter in a direction that is antithetical to what Twitter used to be, and offensive to people for whatever reasons, many reasons, multiple reasons.
01:38:31 ◼ ► There's a large number of people who Musk's decisions pushed away, and it has nothing to do with third-party clients, right? And for some of us, that's just one more straw breaking the camel's back. It's one thing among many, but it's not just that, it's just the overall tone of everything on Twitter.
01:38:54 ◼ ► Like, "Okay, I'll use your first-party client, fine. I'll just give up on that, I'll just concede the point, I'm just going to use Twitter.com and the Twitter official app on my phone, fine."
01:39:06 ◼ ► People are still like, "This site sucks. I hate it. I don't like the feed." Whereas you don't have that with Reddit, right? It would almost be better if Elon Musk bought Reddit and Twitterfied Reddit to push people to do what large communities have done by moving from Twitter to Mastodon. And I don't think that's going to happen.
01:39:28 ◼ ► If only Elon Musk bought Reddit is not a sentence I thought I would have been agreeing with earlier this year.
01:39:34 ◼ ► Well, it's like I've said, I've had to eat my words where I was optimistic before he took over. Because I thought, the one thing I still think it was true is that Twitter did need a shakeup. They needed to be shaken up in multiple ways. And obviously his leadership didn't go the way I hoped it would.
01:39:54 ◼ ► But in some ways, I do feel like I was right that it's a better outcome for me because the people who I care about most, reading their tweet-like posts and interacting with them for people who @GruberMe about a Daring Fireball post,
01:40:16 ◼ ► all of that that has moved to Mastodon is my overall Twitter-like experience is better today than it was before Musk bought Twitter. Because I've moved to Mastodon, right?
01:40:31 ◼ ► I don't see that happening with Reddit, unfortunately. It's almost like, I guess where I'm going is it would be better if Huffman's decisions were even worse. So that it would just drive more people away.
01:40:42 ◼ ► Yeah, yeah. No, I honestly would agree with that. It's a shame he didn't even go further.
01:40:47 ◼ ► Right. And in a perverse way, it's good that Musk is so bad for Twitter because it's made an entire segment. And again, it's like you said, there are the cognitive load aspects to understanding Mastodon's federation are just complex enough that they keep people who, even people who could understand it, or maybe even do,
01:41:11 ◼ ► it just gives them enough resistance to like, I don't know, it's just too much of a hassle. Whereas Twitter, it's just, I've got a username, I type in a box, I hit return, and now that's my tweet. And that's it.
01:41:23 ◼ ► Yeah, the amount of people who I still think are incredibly smart, I met at like DubDub who were like, yeah, I just couldn't understand the whole Mastodon thing. It was a little fascinating to be like, wow, really?
01:41:34 ◼ ► You are very intelligent to me. And not that that changed, but like, it was an interesting data point.
01:41:41 ◼ ► Yeah. All right, let me take one more break here and thank our last sponsor of the show. It's our good friends at Rocket Money. Rocket Money is a personal finance app that finds and cancels your unwanted subscriptions, monitors your spending, and helps you lower your bills all in one place.
01:41:57 ◼ ► Over 80% of people, according to Rocket Money's own polling, have subscriptions that they have forgotten about. So in other words, if they came to you and you brought all your credit card bills and you just went through it item by item, are you going to find at least one subscription that you've forgotten about that you're like, "I don't know why I'm still paying for that."
01:42:17 ◼ ► 80% of people have at least one subscription that they've forgotten about. So the odds, four out of five chance is you're one of them. Rocket Money can help you identify them. It'll quickly and easily find subscriptions for you.
01:42:29 ◼ ► And for any of the ones you don't want to pay anymore, you can just hit cancel right in Rocket Money and Rocket Money will cancel it for you right there.
01:42:36 ◼ ► They also help you manage all of your finances in one place. Set a budget, track as your month goes on where you are in your budget, whether you're going out to eat in restaurants budget, are you over budget, are you spending more than you thought you were on takeout and stuff like that.
01:42:51 ◼ ► You can track anything you want right there in Rocket Money in one place and they automatically categorize your expenses. So as you're spending money throughout the month, you don't have to sit there and enter stuff in one by one and pick a category and pick an amount. You just let Rocket Money have access to your accounts. The things come in, they categorize them for you automatically.
01:43:11 ◼ ► And so you can easily just observe it like a dashboard and see where you're going and adjust your budget in real time. Over 3 million people have used Rocket Money and the average user has saved up to $720 a year by using Rocket Money.
01:43:27 ◼ ► So stop throwing your money away. Cancel your unwanted subscriptions and manage your expenses the easy way by going to rocketmoney.com/thetalkshow. That's rocketmoney.com/thetalkshow. They put it in one more time. Here, I'll read the URL one more time. rocketmoney.com/thetalkshow.
01:43:47 ◼ ► Alright, bonus period. It's sad because it had to be. Again, I opened by saying WWDC must have been bittersweet because you must be used to paying attention to WWDC each year and then thinking, "Oh, that's cool. I could use that for blank." Right?
01:44:05 ◼ ► Anything that stood out to you? Did you catch yourself during the keynote saying, "Oh my God, that's great for repo-"
01:44:12 ◼ ► A few times. Well, it was funny when the keynote came out, I was still at the stage where they hadn't alleged blackmail and all this stuff and kind of stopped responding to my emails.
01:44:21 ◼ ► So I was honestly looking up there pretty starry-eyed like, "This will be a fun summer." Especially around the Vision Pro stuff. I got lucky that a few months ago, Apple kind of called me very coyly and was like, "Hey, I guess in order to put your app icon on a slide, apparently they need kind of your approval."
01:44:37 ◼ ► And they were saying, "There are certain apps that are going to be compatible with the future platform we're working on and Apollo is through our testing. Could we put it up there?" And I was like, "Oh, okay. That's so cool."
01:44:46 ◼ ► So when I saw it up there, I was like, "Okay, the Vision Pro looks really cool. That'd be something like, I don't know what that would look like for Apollo, but I sure as hell love to try." So I was really excited for that. So we'll see.
01:44:56 ◼ ► And yeah, there was even some more of the interactive widget stuff looked really cool across the iOS and whatnot. So there was definitely a bunch of stuff that would have been potentially cool to play around with.
01:45:10 ◼ ► Mr. Federighi mentioned the word Apollo. What was the context? I have it in my notes somewhere.
01:45:15 ◼ ► He was talking about how you can run your iPhone widgets on your Mac, I believe. And he said Apollo for Reddit is.
01:45:27 ◼ ► Yeah, it seems like that's one of the sleeper features of this year's cycle, this sort of- because the people who are already brave enough to have the betas for everything on all the- I've got like a spare phone with iOS 17, but I have not. I'm too old to put developer beta one on anything I actually use.
01:45:51 ◼ ► I remember years past developer beta ones. It seems like Apple's, they've gotten better in recent years, but it's always- who knows? One year that might be really terrible. But I just remember like five o'clock, four o'clock, people meeting on Monday of WWDC week and going to get a beer or something.
01:46:08 ◼ ► And one out of three people have like an iPhone that doesn't work anymore because they put the beta on and it's like, "Oh, shit, my phone's locked. I don't- I can't-"
01:46:20 ◼ ► And it's like, DFU, you gotta like- they gotta go back to their hotel and connect it to iTunes so they can do a hard reset. It's like, that hasn't happened. No, but this widget thing seems really cool.
01:46:31 ◼ ► It does, yeah. There's so many cool things like that. And I will say that's almost the one thing that I'm somewhat looking forward to not having to deal with is like, they've gotten so much better Apple with letting anybody install betas on their phone, which is great for testing, but you also get a lot of noisy users who are annoyed that there's bugs in beta one that relate to your-
01:46:50 ◼ ► -app and the amount of emails I get every year where it's like, "The text kind of looks funny on this one screen. I hope you fix that soon." And it's like, "Come on. Beta two, it's always gone. It's not me." So I'm somewhat looking forward to not having to deal with that this summer.
01:47:02 ◼ ► Yeah, and I could imagine something- and again, this show is about Apollo and Reddit, and I'll do another show later soon talking more about WWDC and my experience with Vision Pro, but I can sort of see that.
01:47:20 ◼ ► And I'm sure you're thinking the same thing where at a basic level when you're using Vision Pro, you do, you have the biggest canvas in front of you imaginable, right?
01:47:32 ◼ ► And so in a way that like when you go from your phone to like a big studio display or a 27-inch iMac or something, even just one display, but a big display on your Mac, and you feel like, "Ah, this is great. I can spread out.
01:47:48 ◼ ► I've got notes next to Safari, and I can still see messages next to them." It's very clear that in the Vision OS, you can spread out in a way that makes even using a big Mac display feel cramped, right?
01:48:03 ◼ ► And for something like Reddit, I could imagine, even if it's just two apps, but you could have Reddit open on the left and scroll and then have Safari open on the right and click lots and lots of stuff people post to Reddit or Lynx, and you click it or tap it over here, pinch it, whatever our verb is going to be.
01:48:23 ◼ ► And it's open over there, and you see them both. You just turn your head to go back from Reddit to Safari to whatever else you have to the side of them.
01:48:33 ◼ ► Yeah, there's so many. Or I was even thinking almost like in that scene in The Dark Knight where Morgan Freeman's character has the 800 screens of Gotham.
01:48:41 ◼ ► Where you can have like 800 different Reddit posts up at once just watching all the videos and pictures and discussions at once.
01:48:47 ◼ ► Yeah, one of those questions that only came to me on the airplane flying home, it's like, "Ah, I should have asked while I was there." It's like, what does a screenshot even look like?
01:48:58 ◼ ► Are we going to be able to take screenshots from the headset? And what would that look like? Because it wouldn't be a rectangle. It's not a rectangle in front of you. It's a whole field of vision.
01:49:13 ◼ ► Yeah, or push the camera button with the digital crown button. I don't know. It seems like every time they come out with a new platform, there's no screenshots immediately.
01:49:24 ◼ ► And then they come out with this. Or maybe the watch had a screenshot feature right away. I forget. But I don't know. Just mash all the buttons and take a screenshot.
01:49:34 ◼ ► But I can't wait to see. I do think that part of it, for some users, they're going to have so many windows open at a time. And you're going to see people complaining like, "Hey, how come VisionOS only lets me put eight apps open at a time?"
01:49:49 ◼ ► And it's like, "Yeah, but you're looking at eight apps all at once. You're just panning around." Yeah, of course people want to have the Morgan Freeman 50 of them.
01:49:57 ◼ ► And as the ears go on and RAM goes higher, maybe you'll be able to set it up. But people who want to see lots of stuff at the same field of view, on the same canvas, as it were, it's an unprecedented platform.
01:50:10 ◼ ► And for something inherently Reddit is sort of a multitasky, and Twitter's the same way, where people post links, so you're bouncing between this thread and where you want to go read the links. And having them in the same field of view seems super cool.
01:50:40 ◼ ► Thank you. Yeah, no, it was the silliest little thing I had an idea for in Apollo, where the dynamic island thing, when it came out, that was like 12 pixels above it, that were kind of just for content to pass through.
01:50:51 ◼ ► That I was like, "It would be kind of cool if you could put something up there." So I played around and worked with a few artists and had a few little cute ones.
01:50:57 ◼ ► And it got really popular on Twitter and people really liked it. And then it got to the point where I got a lot of emails where they're like, "I don't get what Reddit is. Can you just make it a separate thing?"
01:51:07 ◼ ► So I did that, and thank God I did, because it ended up taking on a life of its own on TikTok and whatnot outside of Apollo, to the extent that I looked at the analytics and it's like 90% of the people don't even have Apollo installed.
01:51:19 ◼ ► Yeah, yeah. So it's like, thankfully I've kind of got the secondary income super, so I don't have to go down to Philly and try to find a couch to sleep on this month. But it's a really different app to work on, but it's a lot of fun.
01:51:30 ◼ ► Are any of the changes that were announced this year for interactive widgets going to be helpful for PixelPouch?
01:51:38 ◼ ► Oh, yeah. Yeah, yeah. I think I can do some cool stuff with that. Before there was, yeah, you could do some kind of hacks with like tapping on it and launching the app.
01:51:46 ◼ ► And I saw some apps do some cool stuff with that, but the fact that Apple's got some first-party stuff where you could feed them or play with them or in some capacity through the widget system looks like, yeah.
01:51:55 ◼ ► I don't feel too bad that Apollo's gone in that sense, that I'll have a lot of fun time to work with that this summer.
01:52:00 ◼ ► Yeah, it taps into, and David Smith's super popular Widget Smith scratches the same itch, and I think, and David has very humbly taken his tremendous success with that with too much humility.
01:52:19 ◼ ► Right, but just vastly underestimated. He was like, "This seems like a fun idea that a lot of people would like," and just vastly underestimated the latent demand from iPhone users to just customize the way their lock screens and their home screens look, right?
01:52:38 ◼ ► And it used to be, back in the classic Mac days, we were always getting icon sets from the icon factory. In the early days of Mac OS X, they had the candy bar app that they made in conjunction with Panic where you could just do whole sets of icons and replace the way folder icons looked in the Finder and change the way your Windows looked and stuff.
01:53:00 ◼ ► People liked doing stuff like that, and iPhone was so locked down at a system level for so long that when this came up, it's like so many people just, they just wanted to customize it.
01:53:12 ◼ ► People like to play with it. It's like being a real developer is like you're making your own injection molds for toys.
01:53:22 ◼ ► But these are like Legos where anybody who's not like, "Oh, I'm not a professional developer, but I just want to snap some pieces together to make my own car, toy car," and that's what WidgetSmith and PixelPals really, it just satisfies people's urge to like, "I just want to put something fun on my lock screen that's like my choice."
01:53:41 ◼ ► Yeah, no, and it's one of those things where in hindsight, of course, it's popular, but yeah, it's just for so many years, there was just no way to do that, and now people are just, they use this thing so much every day that to have a little bit of personality sprinkled on one of the screens they see 10,000 times a day is just, something kicks off in your brain that makes you a little happier.
01:53:59 ◼ ► Well, thank you very much for your time. I'm very, very sorry for the overall situation, but I'm glad to have it as an excuse to get you on the show for the first time.
01:54:12 ◼ ► Yeah, well, I don't know about that. That's too kind, but whatever I can do to help in your future endeavors, you just let me know because I've been a fan of Apollo for as long as it's been around, and it might be the end of Apollo, but it's certainly not the end of software from you.
01:54:31 ◼ ► That's the goal, thank you. But yeah, you covered Apollo very early, and I think that was a big uplift at the beginning, so I thank you for that as well.
01:54:37 ◼ ► Well, it was like a lightning bolt, like, wow, this is a good app, right? So it's like, it just always, I'm sure there's some that I miss out there, but I just love identifying and rewarding with whatever attention I can give to exquisitely well-made apps, and that is certainly the work that you do, Christian.
01:55:04 ◼ ► Let me also thank our sponsors for the day. We had Collide, where you can manage your cross-platform fleet of devices, get them into compliance.