Under the Radar

240: File Edit WWDC Window Help


00:00:00   Welcome to Under the Radar, a show about independent iOS app development.

00:00:04   I'm Marco Arment.

00:00:05   And I'm David Smith. Under the Radar is never longer than 30 minutes, so let's get started.

00:00:09   So it has finally happened. I was a little nervous that today or this day would never come,

00:00:15   but WWDC 2022 has been announced.

00:00:19   Yup.

00:00:20   Which is very exciting. I was starting to get, I think the previous years it had been announced

00:00:25   in kind of early-ish March or mid-March and then late March, but like now this was announced

00:00:31   beginning of April, and so I was getting a little nervous.

00:00:34   And so I'm very, A, I'm very glad that it was finally announced just because I don't know,

00:00:37   this is probably just the reality for being an iOS developer, but summer plans are always

00:00:42   complicated for us because I, you know, want to go on vacation with my family.

00:00:46   I want to do things. I want to plan things. But until I know when WWDC is, I can't really

00:00:51   make commitments for essentially June and July are just like complete write-offs for

00:00:56   my calendar until I can know when WWDC is going to slot in there because the reality

00:01:00   is it's a week that I'm dedicating to being an Apple developer first, you know, sort of

00:01:05   other commitments because people start to be, you know, it's not a week that I'm going

00:01:09   to intentionally go on vacation for because it's kind of an important week.

00:01:13   So that's, I was very excited and glad to finally be able to, you know, sort of take

00:01:18   a pen out right on our paper calendar that this is WWDC week. I can set it aside. I can,

00:01:23   and now I can start to make plans for the rest of the summer. So very exciting both

00:01:27   just that it exists and also, you know, the general form of what it will look like and so on.

00:01:32   Yeah, I found this announcement to be very good news because, you know, as listeners

00:01:36   know, I am a big fan of the remote conference and I think overall it is better and it

00:01:41   serves more people way better than the old format did. And so I'm a huge fan of what

00:01:47   they are doing. The watch party thing is interesting with that but otherwise, you know,

00:01:53   like having the conference be effectively fully remote again for almost everybody and

00:01:58   therefore having it being produced in that way, you know, made for that, optimized for

00:02:03   that, that is all very good news to me. And so I'm very happy with this. I'm glad they

00:02:08   decided to do this and, you know, as, again, as listeners probably know by now, I'm a

00:02:12   big fan of them keeping it in this format in the future even after in-person events

00:02:17   become a little bit easier and more responsible to hold. I still like this format better.

00:02:22   I really do. I think, you know, the old format really priced out so many people and

00:02:29   such a large percentage of the developer base. And even among those who were willing

00:02:34   and able to spend the money to go to the in-person conference, very little of it was

00:02:38   available. Like, you know, very few tickets were available relative to the demand. And

00:02:42   so this new format is just so much more inclusive and so much better fit to the modern

00:02:49   world that we live in. And not talking about COVID, just like the modern world of

00:02:53   massive demand for this conference that, you know, can only hold a few thousand people

00:02:57   in practice when it's held in person. So I'm very happy with this decision they've

00:03:02   made to keep it this way. And, you know, as for the in-person viewing party component,

00:03:08   which we don't really know a lot about yet, but I think it's interesting because, you

00:03:13   know, they said that, you know, they're going to have basically a small gathering of

00:03:18   some subset of students and other interested parties who will be able to gather

00:03:23   somewhere on the Apple campus and watch the event, quote, "live," but they're watching

00:03:29   the video being played live. It's not being produced live, it's being pre-recorded.

00:03:35   So basically a viewing party and presumably some kind of, you know, meet and greet with

00:03:41   some people at Apple or something, you know, some kind of component there. But what's

00:03:45   interesting is that the in-person component there is seemingly not including things

00:03:49   like labs, where normally the value of being in person is, you know, part, obviously,

00:03:56   socialization and meeting with the developers and everything, that's a big part, but

00:03:59   also a big part of being in person is having access to things like labs, where that's

00:04:04   harder to do remotely, although now they have been doing that remotely. And so I think

00:04:08   it's interesting, like, the in-person component now seems purely social or, you know,

00:04:14   purely for, like, the value of the event. And I'm curious, you know, what the appeal

00:04:19   of that is, obviously if you're local and it's not super hard to get there and you

00:04:23   don't have to, like, stay in an expensive hotel, obviously that's a very different

00:04:26   calculus. But, you know, for people like me and you, Dave, you know, we would have to,

00:04:31   if we wanted to go to the in-person thing and if we could even get tickets, which is

00:04:34   its own thing, but, you know, we would have to fly out to California, which, you know,

00:04:39   is significant for both of us, but especially for you because you're much further away,

00:04:43   and, you know, stay in the expensive hotels and, you know, figure out transportation

00:04:47   and everything. That's all, you know, it's a lot of logistics, a lot of travel. And so

00:04:52   for me, I'm not going to try. I've already decided, you know, that this year for this

00:04:57   in-person thing, like, I'm not going to go. If I get it, if I happen to get, like, a

00:05:02   press invitation, I might consider it, but even that I think would be a tough sell just

00:05:08   because of the logistics involved and, you know, for the, you know, risk-reward calculation

00:05:12   here, I think it's a little bit tricky. But I'm curious, what do you think? Like,

00:05:15   are you going to try to go to the in-person thing and try to get a ticket?

00:05:19   Yes, this is the short answer to that. But I think the broader answer is, A, I think

00:05:24   I'm 100% with you that I think it is, I'm excited, and I think it's the right path

00:05:30   for Apple to have an online-first W2C, that it is a week that is optimized around the

00:05:38   online experience, around the experience that can scale both in terms of the number of people

00:05:43   and in terms of to the broadest audience of users and honestly to the highest quality

00:05:47   that the last few years, W2C online has been really good. And that's coming from an experience

00:05:53   of, you know, I've been to every W2C since 2009. Like, I know what W2C can be. Some of

00:05:58   those years have been great. Some years have been a bit mixed. But the last two years,

00:06:02   I think from a communicating what's new for inclusivity for lots of reasons, I think

00:06:08   it's just been great. But I am also really kind of glad and excited to see that they're

00:06:13   starting to dip their toe into having it be a hybrid event, having it to be an event for

00:06:19   which there is some in-person component. And in a weird way, what makes me think of is

00:06:24   the same for the same reason that they have, you know, press coming on site for a press

00:06:31   event in previous years, that there is something different about being in person than just

00:06:37   watching a video online. That this is the same reason that it's different to, you know,

00:06:42   it's different to go and watch a concert, you know, of your favorite band versus just

00:06:47   sitting at home and listening to their music. There is something different just about that

00:06:52   experience and about the excitement, about the energy, about your ability to pick up

00:06:56   on nuance and sort of gauging the people you interact with that if there's a meet and

00:07:03   greet, you know, or what Apple is talking about, or the things that other developers

00:07:06   around you are interested in, that there's something unique about that. And what I don't

00:07:11   know if this is exactly where Apple is going with this, but what intrigued me about when

00:07:15   they said they're having a special day for developers and students is in the same way

00:07:20   that some press goes on site in past years for Apple events, that I think there is something

00:07:26   potentially that Apple is doing here of trying to get developers there to talk to about what

00:07:34   they're excited about doing, what they're coming for this coming year. And essentially

00:07:38   it's developer press, that isn't press in the sense of like people who are writing for

00:07:43   newspapers, but I think about, you know, the communicators, people who are in our community

00:07:49   who talk about what they're doing, who talk about the thing I think about, you know, like

00:07:54   Paul Hudson or some of the other kind of people who are educators in this area, like getting

00:08:00   them on site and being able to talk to them and about getting them jazzed and excited

00:08:04   about what's coming makes a lot of sense. And as someone who does a fair bit of, you

00:08:10   know, like what I'm doing right now in some ways is developer communication, I would be

00:08:15   excited to go in person, to be able to kind of, well, it's the same content, it's a

00:08:20   different context. And I think that context would be interesting and useful. And so I

00:08:25   would certainly try. It's like, I have no idea if I would be able to get a ticket. I

00:08:28   have no idea what the criteria for that kind of going to be, if it's just going to be

00:08:31   a lottery, if it's going to be some kind of selection criteria. But the reality is,

00:08:35   if I was invited, I would absolutely go and I would be super excited to go. And I think

00:08:40   too, there's a second part of that, that in addition to I think it being a useful

00:08:43   thing to really get the full flavor of what this year's announcements are all about,

00:08:49   I think there's also an element that's just like, I'm all in on Apple developing.

00:08:56   Like that's my career, you know, I've been doing this for a long time and it's something

00:09:00   that I've dabbled with other platforms, I've dabbled with other things, but I'm an iOS

00:09:05   developer with a little bit of watchOS in there whenever I can squeeze it in. And that's

00:09:10   what I do. And I'm all in on that as a career path. And so I want to sort of like, anything

00:09:16   Apple is offering on the developer side, I want to take advantage. I want to do anything

00:09:20   I can to educate myself, to motivate myself, honestly, which is a small part of it too,

00:09:25   that like, there was something super motivating about in person WWDC that as much as it was

00:09:30   lovely on the online years to get the videos, be able to scrub them back and forth, do all

00:09:34   the things, it is not the same energy as walking out of a keynote in the Presidio. And obviously

00:09:41   it is a different thing, you know, if we're watching a video and it's not quite the same,

00:09:45   it's not like there's Steve Jobs on the stage. Wow. It's a different thing. But I imagine

00:09:50   there's a different level of motivation and excitement. And I would love that if it was

00:09:54   possible. And I think it's like, because I'm all in on this platform, it's like, if they're

00:09:57   offering, I will say yes, and I will try. And if it doesn't work out, no problem. I'm perfectly

00:10:01   happy to do another year of online WWDC. But like, if there's an opportunity, I am going

00:10:07   to try to do it, you know, do whatever I can to take advantage of it.

00:10:11   That's, you know, I didn't think about that angle of it about, you know, the motivation

00:10:14   and you know, that's, that is a pretty good point. It's not a good enough point for me

00:10:20   to want to try to do it this year. But you are right, though. Like, I always have been

00:10:25   very jazzed. Like every time I would go to the conference and I would, you know, I'd

00:10:30   see the sessions, I'd see the, you know, the big State of the Union and everything, I'd

00:10:33   be like, Oh my God, I can't wait to jump in. And, you know, sometimes I would, you know,

00:10:37   I'd download the betas that afternoon and start trying to get stuff going like right

00:10:41   during lunch as we're eating the box lunch and everything. So yeah, I guess in practice,

00:10:47   you are right that there is a lot more value in person than I initially thought.

00:10:51   And I think it's definitely going to be a year that I think, well, you'd like the perspective

00:10:55   of someone who says, maybe not this year. I totally respect that. And I think it makes

00:10:59   a ton of sense. And I think for all of us, we're going to have to be finding this line

00:11:02   and this this year, and then it's like, if they keep doing the same kind of thing, and

00:11:05   potentially WBC becomes one day in person, and the rest online, like over, if that sort

00:11:11   of expands in scope and style, like, I could see everyone's year having to sort of recalibrate

00:11:18   themselves about what makes sense for them. But like for me for this year, I'm like,

00:11:22   I'm in but I totally respect anyone who's sort of choices like, hey, maybe not this

00:11:26   year, you know, you're not quite there yet.

00:11:28   Yeah, maybe. Well, we are brought to you this time by Sourcegraph. These days, every company

00:11:35   needs to be a tech company because there's so much code to deal with. It's hard to

00:11:39   have any kind of online presence or digital service without copious amounts of code. And

00:11:43   code affects everyone. It's the foundation of the technology we use every day. But

00:11:46   handling big code bases can be tricky, especially when you think about the volume

00:11:50   and variety of code. So thankfully, Sourcegraph is here to solve all your company's

00:11:55   code related headaches. They're tackling big code. More than ever, companies need

00:12:00   resources to increase developer velocity, to give people what they need to be agile,

00:12:04   innovative and productive. With Sourcegraph, developers can find and fix security

00:12:08   vulnerabilities, stay on top of technical debt, onboard new developers faster and so

00:12:13   much more. Sourcegraph is the future of how companies are built. And it's powering

00:12:18   the infrastructure of today's most successful high growth businesses, including three

00:12:22   out of five of the top tech companies. Sourcegraph's current customers include

00:12:26   PayPal, Uber, Plaid, Reddit, Atlassian and so many more. Visit about.sourcegraph.com

00:12:33   to learn more. That's about.sourcegraph.com to find out why some of the biggest tech

00:12:38   companies in the world use Sourcegraph and to see what it can do for yours. Our

00:12:43   thanks to Sourcegraph for their support of this show and Relay FM.

00:12:47   So our next kind of, we have a couple of mini topics I think to talk about. And this

00:12:52   first one's coming out from an interaction that you and I had regarding the launch

00:12:57   of the big Overcast redesign. And specifically I was asked, there were some server

00:13:02   side sort of load issues that you were having to navigate as part of this because

00:13:07   it's part of the new redesign. Anytime someone, as soon as someone installed the new

00:13:11   redesign, it had to make a call, I think it was to save the colors that they had set for their playlists or something.

00:13:16   Yeah, it was like on first launch of the new version, every copy

00:13:21   out there was going to generate playlist colors and then have to sync those to the server.

00:13:26   And so I was very concerned with this new launch that as that update rolled out to everybody

00:13:31   that it would just overrun my servers with sync requests.

00:13:36   And so, to which I was like, well, did you use the option to do the slow rollout

00:13:41   in App Store Connect? And you said that you hadn't because you were concerned that

00:13:46   you were hoping for and expecting a fair variety of press for this.

00:13:51   And so you wanted to make sure that people were, if they saw, they read the article

00:13:56   on the verge that says, hey, there's this great new update and they go and download the update. You want to make sure that they

00:14:01   could actually see the update and not just be getting the old version. Right. Which

00:14:06   was interesting to me because this is a recent discovery of my own, was that

00:14:11   that's not something we actually have to worry about. And specifically, this was just, because

00:14:16   this is coming mostly from my experience with A/B testing recently, that I have a much deeper kind of understanding of the way

00:14:21   that the slow rollout inside of App Store Connect works. And so it's just like,

00:14:26   I shared something with you that I think was a surprise to you and so it seemed like a good place to share it on the show. And essentially

00:14:31   it's this, that in App Store Connect, when we do an app update and we choose to release it,

00:14:36   we can release it all at once or we can choose to do the slow rollout. And this works

00:14:41   essentially by every day, it increases, an increasing percentage of your users

00:14:46   will be offered the app as an automatic update. And so this starts off super slow, it's like

00:14:51   0.1% the first day and then half a percent the next day and maybe 1% and then it sort of

00:14:56   ramps up from there. But the thing that is sort of key to keep in your mind when you're thinking of this

00:15:01   is it isn't that that's the percent of people who will get the app.

00:15:06   That is only the percent of people who will be offered automatic app updates.

00:15:11   That's sort of the ramp for that. A hundred percent of new users get the latest version

00:15:16   and a hundred percent of people who manually update the apps in their App Store app

00:15:21   will get the new version. Right. And that's something that I didn't realize. That's the key differentiator right there.

00:15:26   And so what that means, so a hundred percent of new users, that's great for press and

00:15:31   for any kind of promotion you're doing. And then, so when you say automatic update, what that means is

00:15:36   that if anybody goes to the App Store app on their phone and goes to the updates tab, wherever it's buried this year,

00:15:41   it'll be there from day one, but it won't automatically

00:15:46   update itself. So they could tap update in the list, it'll be there on day one, but it won't

00:15:51   automatically update until their number is drawn basically? Yeah. Okay.

00:15:56   And I think that difference changes a lot of the dynamics because A, like you were saying,

00:16:01   once you hit, you know, ready for sale in App Store Connect, you release it to the App Store,

00:16:06   no person will ever get a previous version of that app

00:16:11   again. So it's like you've created this clear dividing line. So like from that point on, any download

00:16:16   of that app that is happening is going to be the new version. So you don't need to worry from a press perspective or something that

00:16:21   anyone's ever going to get the old version. And so that's a good thing from a press perspective.

00:16:26   And it's a good thing from your server's perspective, because in this case, you would expect the vast

00:16:31   majority of users won't immediately run and go and download the app

00:16:36   and the number of new daily users, while you would hope that having press, people in

00:16:41   publications talking about your application, you would get a spike in downloads,

00:16:46   that spike is likely not going to overwhelm the size of the existing user base. And so it'll be

00:16:51   gentler on your servers. And so that was just something that was an interesting,

00:16:56   because I think I used to think of it in the same way that I think you were thinking of, that it's like if you did the phased

00:17:01   rollout, that you still could have people who were downloading the old version. But that is

00:17:06   in fact not the case, doesn't seem to be how it works. And so we're safe to, essentially, I do all my rollouts

00:17:11   as slow rollouts now, because there's very little downside to doing it that way. I think the only

00:17:16   time I've ever done a fast rollout recently was when it was a bug fix

00:17:21   release that was fixing a really bad bug that I didn't want

00:17:26   out in the world anymore. And so it's sooner,

00:17:31   it was a defensive thing, but as long as it's not that case, if it isn't the emergency bug fix,

00:17:36   the old version is accidentally deleting people's profiles or something horrible,

00:17:41   it's safe and good and works pretty well as a result.

00:17:46   And do I remember correctly, I did it once a long time ago, do I remember correctly that

00:17:51   if a few days in, you're like, you know what, everything's fine, there's a button you can hit that just releases it to everybody instantly?

00:17:56   Yes, and there are actually two buttons. So there's one that you can say just release it to everybody,

00:18:01   and from that point on, it'll sort of, at the usual rate that whatever it is, Apple offers

00:18:06   automatic updates, everyone will get it, and the other option you can do is to pause the rollout.

00:18:11   And that will sort of cap the percent at whatever it had gotten

00:18:16   to. So if you say you're two days into it and it's gone, the percent for the release is

00:18:21   now at 2% of people will be offered automatic downloads, you hit pause, it'll just stay at 2%

00:18:26   indefinitely or up, I think it's not indefinitely indefinitely, it's maybe, oh, you can do that for up to two weeks or something

00:18:31   along those lines. And so you also have the option then, from a bug perspective, like, I've definitely pushed

00:18:36   that button once or twice where I release an update, I discover a little bug, it's something I want to

00:18:41   fix, it's not like this catastrophic thing, but I can pause the rollout in the meantime while I fix it, while it

00:18:46   goes through app review, and then before it gets replaced. So that bug won't propagate

00:18:51   nearly as far. It'll still be unfortunately shown to all my new users, and it will be shown to, in this case, you know, 2%

00:18:56   of existing users, but it will be much slower and lower than it would be if I'd sort of done a big rollout.

00:19:01   Yeah, that's really interesting. And so what that means is, so if

00:19:06   you're like, you know, in the middle of a rollout, whether it's paused or not, you can still upload new versions

00:19:11   and have them approved and everything and it works? Yep, it works totally fine. And it's essentially once you

00:19:16   have a version that's ready for sale, that's the ongoing version regardless of how its

00:19:21   rollout status is, and then you can just add new versions like you would typically. You can have them go through app

00:19:26   review and then you'll have the option to release the new version or, you know, hold it for pending

00:19:31   developer release, depending on what makes sense for you. That's really good to know. Now

00:19:36   that you tell me more about how this works, now I think I'm with you that like, that probably makes sense to do

00:19:41   for almost every update. There are very few soundlists. So at this point, the only downside

00:19:46   that I'm really aware of was, this relates to some of the A/B testing stuff we've talked about in the last few weeks,

00:19:51   and that is because of that, if you add new behavior into your

00:19:56   application that you're going to be A/B testing, that behavior will be

00:20:01   dominated by new users, new downloads of the app for the first,

00:20:06   you know, whatever, the first week of the app update being out there. Because

00:20:11   that functionality, 100% of new users have it, and you know whatever, 2% of

00:20:16   existing users have it. And so, depending on the ratio of those

00:20:21   numbers, you could potentially have this kind of skew. And so I would have these things where

00:20:26   I was running an A/B test and I thought, "Oh wow, this new paywall is performing

00:20:31   amazingly." And it turns out paywalls perform better for new users than for existing users,

00:20:36   which is sort of, in some ways, notionally makes sense to me, that a new user is much more excited and much more likely to

00:20:41   convert just than someone else. They're exploring around, it's a

00:20:46   much higher sort of conversion activity. And so I roll out this new paywall, the first few days it performs amazingly.

00:20:51   I'm like, "Yay, this is great! Everything's going to be awesome." And then it just sort of immediately plateaus

00:20:56   as in the exact shape of the rollout curve. And it's like, "Oh,

00:21:01   right." So I haven't actually somehow revolutionized my paywall, I've made a marginal

00:21:06   improvement, and it's just I'm observing this. And so if you are running any kind of testing behavior

00:21:11   or if you're seeing weird statistics, it's just something to keep in the back of your mind that if you're doing a slow rollout

00:21:16   the first week or so, it's going to be very skewed towards new users who have access

00:21:21   to that new feature, and so be careful about the conclusions that you draw.

00:21:26   That's really interesting. I am so happy that you exist and share your knowledge. This is

00:21:31   stuff that I would never have thought to try most of these things on my own, let alone

00:21:36   realize things like that statistical anomaly of new users over-weighting things

00:21:41   in the first week. I would never have thought of that. That's very good to know.

00:21:46   I'm delighted to share it. And the other little mini-topic that I wanted to talk about

00:21:51   here was just a realization that I had recently that is one of these things of

00:21:56   the "everything old is new again" kind of a thing, where almost every

00:22:01   Mac app that I've ever used in its menu bar

00:22:06   will be the rightmost menu. What is that called? Help. Help!

00:22:11   It's called Help. In almost every Mac app, there's a section called Help, and

00:22:16   that's there for a reason. That's there to help users explore and understand how to use your

00:22:21   application. Yet I think almost every iOS application has no

00:22:26   help whatsoever. It isn't something that we do, it isn't something we think about, it isn't

00:22:31   something in Apple's own apps that, like, the phone app doesn't

00:22:36   have help in it. There's the, what is it, the Hints

00:22:41   app or something like that? That Apple ships with tips.

00:22:46   Hints. You know because tips has itself opted in to notifications. So on new

00:22:51   setups you always get notifications from tips. Yeah, which is like sort of

00:22:56   a way to do this, but anyway, it is something that I recently, like the next build of Widgetsmith,

00:23:01   I'm elevating Help into its own tab. So the app currently has four tabs,

00:23:06   it is going to have five, and the fifth one is going to be Help. And I'm kind of excited to

00:23:11   try this out and to see where this goes. This is coming from an experience I had

00:23:16   with Widgetsmith where I, in doing my customer support

00:23:21   and having frequently asked questions, and in terms of instrumenting my frequently asked questions that I can see

00:23:26   which questions most people are having trouble with, in terms of which are in my frequently asked questions, when you tap to

00:23:31   expand the answer, I record that and so I can get a sense of proportionally which

00:23:36   questions are most interested. And it was kind of sad to me that the most popular questions were like

00:23:41   "How do I configure a widget? How do I add a widget to my home screen?" Which

00:23:46   is fundamental, like that is all Widgetsmith is. If you don't know how to do those two things, the

00:23:51   app is completely useless to you. So I've completely failed you if you don't know how to do that.

00:23:56   And so I added a new sort of, it's not quite onboarding

00:24:01   in the sense of like a walkthrough, but I've restructured the main screen of Widgetsmith now so

00:24:06   that when you first install the app, until you've configured your first widget, there's this very

00:24:11   prominent banner at the top that says "Learn how to configure your widget." And so you tap that and it gives

00:24:16   you this animated walkthrough. And as soon as you've configured it, it goes away and is replaced by "How to

00:24:21   add widgets to your home screen." And once you've added it to your widget, I can detect that and I can make the banner go away.

00:24:26   And so I've kind of like trying to walk you through this experience.

00:24:31   And that's actually been really, really effective that ever since I rolled that out,

00:24:36   this was probably about two weeks ago, I've seen a precipitous drop in A, the number of people who

00:24:41   are opening the frequently asked questions, so fewer people have questions, and then of the questions that people are asking

00:24:46   are now the questions that I kind of want them to be asking. The kind of more power user, nuanced

00:24:51   things where it's a bit complicated and there's nothing I can do, or it's a choice that I've made, or things like that.

00:24:56   Like, those are the questions that I understand to some degree are always going to be there.

00:25:01   But how to fundamentally use the application should not be a question that anyone has. And because that was so

00:25:06   successful, I've now moved that into its own tab, and now it's a place

00:25:11   that everyone will be able to see and hopefully create less confusion. That it isn't,

00:25:16   if you have questions, there's an obvious place to go and get it, and I can surface this information

00:25:21   to you. And it's a bit too early to report back on exactly

00:25:26   how it's going, but based on my experience with doing the sort of onboarding side, I have a suspicion

00:25:31   this will ultimately be really a positive, good thing that will

00:25:36   help more of my users know what's coming, and know how best

00:25:41   to use the app, and bounce off of it a lot less. Because the last thing I want is, if I've gotten you

00:25:46   all the way through the App Store process, through the marketing process, you've downloaded the app, you open

00:25:51   it, you don't know what to do, and you then close the app and delete it, that is a giant

00:25:56   fail on my part, and I need to do everything I can to keep you in it to at least

00:26:01   try it. And if you try it and you don't use it, that's fine, but if you don't try it because you don't understand it, that's a big problem for me.

00:26:07   Yeah, that's, man, look at, here you are, making more work for the rest of us again, by making

00:26:12   sense and trying things that work, showing us like, oh, we should really be doing X, Y, Z.

00:26:17   No, this is something, I, you almost never

00:26:22   see iOS apps with help. I mean, heck, you hardly ever see Mac apps with help anymore. You know, that's, almost anything

00:26:27   designed or shipped by independent developers or just anything from

00:26:32   recent companies that, you know, they don't really care about this kind of stuff, you almost never see this anymore.

00:26:37   Because making good help, you know, it's a form of documentation, and making documentation is a

00:26:42   ton of work, and to do it well, you really need like, you know,

00:26:47   specialized skills, you need people who are good at technical writing, and it's a lot of work if you ever want to

00:26:52   integrate things like screenshots or any kind of like video even of little clips of how to do stuff.

00:26:57   Like, that's all tons of work, you have to update it over time as your app or

00:27:02   the look of it change over time. Like, it's a huge amount of work to do this well.

00:27:07   And so, I've never even thought about it, and people have emailed me frequently, some, you know, I would

00:27:12   maybe not like, you know, I'm not talking like every day or every week, but maybe once a month or so,

00:27:17   I get an email from somebody saying, "Is there any help available on the app?" Or, you know, "Is there a guide

00:27:22   somewhere?" Or, you know, "Where can I find help?" And I've always thought those requests were

00:27:27   kind of odd, because, you know, in mobile that doesn't really seem to exist in a consistent way.

00:27:32   Or it's just like, you know, "Here's a link to our website knowledge base," which is usually not very helpful.

00:27:37   So, I've never even thought of doing this, but I think, now that you're saying, like, you know, I have had requests for that before,

00:27:45   and, you know, there's probably a lot of people out there who look for it and don't find it and who I don't hear from.

00:27:50   And so, I think you might be onto something here.

00:27:54   Well, and I think the thing that I would say, too, is I think I was intimidated in all the same reasons that you were just,

00:28:01   like, doing it well is difficult, and being comprehensive or updating it and those kinds of things can make it feel really big and scary.

00:28:09   But I think what I found from this experience was that I got a huge benefit from just spending a little time on the, like, the absolute core experience of the application,

00:28:20   and providing help for that, and not worrying necessarily too much about providing a lot of documentation or information for the more esoteric features.

00:28:28   So, like, in Widgets methods, configure a widget, add it to your home screen. If I can communicate that, I'm winning.

00:28:34   In Overcast, it would be like, find a podcast, play a podcast. Like, if you are using the app, like, if you open the app and you don't know how to do that, you're completely stuck.

00:28:45   And so, I think identifying some very core behaviors, some very core user paths, and making sure that there's sort of help information about those,

00:28:55   is going a long way. Because I think the nice thing is, users are, it's like, they're smart people. They can, once they have that basic path, starting to branch off of it,

00:29:05   and doing, you know, the more customizable options, or getting more advanced in their use, that's something they will naturally want to explore.

00:29:12   But the last thing I want is for someone to feel foolish, feel like they don't get it, this, you know, this app's too complicated for them to move on.

00:29:19   And so, optimize for that core experience, I think it will give you huge leverage at a relatively sort of low cost in terms of the amount of time and effort to build and maintain that kind of core help experience.

00:29:31   I hate to admit it, but you're probably right. Thanks everybody for listening, and we'll talk to you in two weeks. Bye.