34: Made The Dot Smaller

0:00:00I took the bold step, the very first time I think I've ever done it, and marked that

0:00:04review as not helpful.

0:00:05What's up?

0:00:06Oh, no, I totally marked them as not helpful if they're not helpful.

0:00:12I want to talk about TV for a minute and not Breaking Bad.

0:00:16And this is going to have a point.

0:00:18I was watching Agents of SHIELD earlier tonight with Aaron.

0:00:23I'm sorry.

0:00:24Well, it's okay.

0:00:26It's got promise, but it's not very good at the moment.

0:00:30Well, the reason I bring this up is because there was a portion of the episode, and this

0:00:36is not particularly important if you haven't seen it yet, but somebody was going behind

0:00:41enemy lines, if you will, and they had like, you know, an invisible, almost invisible secret

0:00:47agent earpiece thing.

0:00:49And this person was trying to schmooze people they didn't know at a party.

0:00:53And so what they did was they had the other agents like up in the magical plane thing

0:00:59telling the person in the field, "Oh, well this is so-and-so and he has twin kids,"

0:01:06and ask him about this and that and the other thing so this person could schmooze and not

0:01:09look like an outsider.

0:01:10Well they walked away from this conversation where they had to schmooze with people they

0:01:14didn't know and they said something to the effect of, "Wow, it's really awesome

0:01:20having you guys here.

0:01:21I could really get used to this."

0:01:22And then they said, and I'm quoting, "It's like Siri if it worked."

0:01:27And so this is a national TV show, and granted America is not the be-all-end-all, but I mean,

0:01:33this is Apple taking potshots on national television, which I thought was a little bit

0:01:38surprising.

0:01:39Well, to be fair, Siri has never been that reliable.

0:01:42No, it has.

0:01:43I mean, it's not unreasonable.

0:01:44I just thought it was surprising that it's become part of the—I don't know if vernacular

0:01:49is the right word, but it's become something that everyone recognizes as an issue.

0:01:52That's the AI trap. Anything you do that, to a layperson, seems like it should be like

0:01:57another person talking to you, until we get actual real, you know, whatever the term is,

0:02:03strong AI, is not going to be like that, and there's going to be an expectation gap. I

0:02:08talk to a thing, I want it to respond to me like a person, it's not going to, we all know

0:02:12it's not going to, but it doesn't matter, because once I start talking to it like a

0:02:14person it damn well better work like a person.

0:02:18And that combined with just reliability.

0:02:20Sometimes it just says, "Sorry, I couldn't do it.

0:02:21Couldn't reach the servers," or "It's not available now," or whatever.

0:02:23But even when it responds to you, it's fun to play with, and then you quickly realize

0:02:27it's not like talking to a person and you're disappointed.

0:02:29And that's never going to go away.

0:02:31Google Now, Siri, these things are going to get better and better and just be leaps and

0:02:37bounds over where they are today, and people will still be like...

0:02:39There'll still be jokes about them on late night TV because they're aren't they so stupid, you know

0:02:45My dog is smarter than Siri like it doesn't understand me doesn't understand and it won't it doesn't understand

0:02:49It won't understand for you know years decades our lifetime. Who knows how long it will take to get actual

0:02:55Intelligence on the level of a human being at the other end of a computer thing

0:02:59And up until that point it's going to be the butt of jokes

0:03:02It's hard to tell it on the agent of shield episode if they were making fun of the reliability

0:03:08or the intelligence or both. Because like the reliability that you know that

0:03:14Apple could slam right now, well we could make this stupid thing reliable at the

0:03:18very least and then you're just complaining about all the responses

0:03:20aren't very smart but when it says oh sorry it's not available right now series

0:03:24now whatever the heck it says when when you know the servers don't send a

0:03:27response that is something that Apple should be ashamed of now but everything

0:03:31else if you're doing anything that you talk to that talks back you're just

0:03:34going to have to take your lumps. But you know as far as as far as popular culture

0:03:39is concerned and as far as regular people are concerned, hell even as far as

0:03:43geeks are concerned, it doesn't really matter whether the server failure

0:03:47happens or whether it does the wrong thing or thinks the wrong thing about

0:03:52what you said. Either way it's a failure and and all it takes is a few failures

0:03:56where you know after that you just decide you just forget about using it

0:04:00you can stop using it. Right, and so it seems weird to me, well I don't know if weird is

0:04:06the right word, but I can't help but wonder, this got me to thinking that I can't help

0:04:09but wonder at what point is Apple going to say enough is enough and let's properly fix

0:04:14this. And I know that we've talked about this a lot with like iCloud and Core Data for example,

0:04:21and we've talked a lot about this with Siri as well, but at some point you have to think

0:04:26that they're going to get together and say, "Guys and girls, we really, really, really

0:04:30have to fix this." Like, is that time ever coming? Am I waiting for a train that's

0:04:34just not going to show up? Well, it was kind of a victory for Apple

0:04:37to be mentioned because the worst thing for Apple would be for them to make that same

0:04:40joke about Google Now or something like, "Siri has the mind share as that thing on your phone

0:04:44that you talk to." And there is value in having that mind share, even if it comes along

0:04:48with all that other baggage and everything, just because, like, you know, not that you're

0:04:54you're going to become the Kleenex or whatever of the thing, or whatever it is, genericized

0:04:57brand name. But that's what they went to for the joke, because they figure most people

0:05:01know of that, they'll know what we're talking about, they'll get the joke. Whereas Google

0:05:06Now or any of those other things that you talk to are more reliable, yes, but again,

0:05:11I don't think the joke was about reliability. I think it's like, if it worked, as in, if

0:05:15you could actually ask Siri things and she would give you answers, versus just saying,

0:05:20"I'm sorry, I don't know what that is," or doing a Google search for it, which is what

0:05:22Siri does when you try to talk to it like a person. I don't know, but I think it's good

0:05:28that the word is in the public consciousness, and I think Apple continues to work on the

0:05:33reliability and fail for the same reason they fail to make all their online services reliable,

0:05:38but it's also working on the intelligence part of it. At this point, I think it's just

0:05:42the cost of doing business. You want to be in the phone market, you've got to have some

0:05:45sort of thing that you talk to that does real-time intelligence searches from multiple sources.

0:05:49So it's never going to go away.

0:05:50They're going to keep trying to make it better.

0:05:52And you know, think about online services.

0:05:55Like, oh, when are they going to fix this whole online thing?

0:05:57Like, you know, .Mac and iTools and MobileMe.

0:06:01Was iCloud the one where they fixed everything?

0:06:03No, not really.

0:06:04They just keep trying, I guess, right?

0:06:07Yeah.

0:06:07I mean, and you have to start wondering,

0:06:11what is it about Apple that makes them, quote,

0:06:14"not good at web services?"

0:06:16Like, we all say that.

0:06:17We all write that.

0:06:18We all think that, "Oh, Apple is not good at web services."

0:06:21But what's different?

0:06:23What's going to be different a year from now compared to now in that area?

0:06:28What steps are they taking or what steps could they even take to meaningfully change that?

0:06:34What is it about the company that makes them not good at web services?

0:06:40I don't really see from the outside any evidence that meaningful change is happening

0:06:45there.

0:06:46you know, as we've discussed before, it's probably a problem of engineering resources

0:06:51and priorities. And up until this time, Apple has clearly put some priority on web services,

0:07:00but they are still a very small company with their engineering resources. And it doesn't

0:07:07really ever seem like that's going to take a massive turn for the better, where suddenly

0:07:14their web services are going to have tons more staff on them, tons more resources, and

0:07:18be a much higher priority in the company. I don't see that happening.

0:07:21Yeah, I don't know either. And it seems weird. Not everything is web objects, right?

0:07:28I know the iTunes stories, but we have no reason to believe everything else is web objects,

0:07:32do we?

0:07:33Right. And I really don't think web objects, the technology, has anything substantial to

0:07:38do with why Apple is not "good at web services." I mean, you can pick on any language or platform

0:07:44and say, "Oh, well, that doesn't scale," or "That's old," or whatever. The fact is, that's

0:07:48not the problem. With proper administration and proper coding, you can make anything scale.

0:07:54You can make anything work. The platform is rarely the problem. The platform is--and by

0:08:01the way, there's nothing saying Apple has to be using WebObjects. Maybe they're not.

0:08:05Maybe they're using it for part of the stuff. Maybe they're using it for just a front-end

0:08:08somewhere and using big Oracle stuff behind it, who knows? But I would not put blame on

0:08:15the fact that they occasionally have WebObjects URLs that we're looking at from the front.

0:08:20I don't think that has a lot to do with it.

0:08:21Well, but the hard thing about WebObjects is that how do you hire for someone that can

0:08:26do WebObjects? Obviously, you can teach any competent programmer just about anything,

0:08:31but if you want a WebObjects guru, there are like, what, four of them in the world and

0:08:36and they're all probably on Apple's payroll already.

0:08:38So yeah, in and of itself,

0:08:39Web objects may not be the problem in the sense of

0:08:43it functions and with, like you said, good coding,

0:08:46it'll continue to function,

0:08:47but it's hard to hire into that role

0:08:50if you wanted to throw people at the problem,

0:08:52which may or may not even be the solution.

0:08:54- Well, but really, looking at a major code base

0:08:59for a major web service under heavy traffic

0:09:02that's very high profile,

0:09:03I mean, really, does it matter what language it's written in?

0:09:07Any new hire is going to have to go through a lot of training and a lot of time just becoming

0:09:11familiar with this code base and becoming useful in working with it.

0:09:15I don't think, even if it was written in Java, which everyone either knows or can be taught

0:09:20very quickly, I don't think that would really make a difference.

0:09:23I mean, really, I think if WebObjects really was the big problem that they're having, either

0:09:29a giant failure of leadership. Or, well actually no, that's it. It's a giant failure of leadership.

0:09:37That's it. That's the reason. If they're being held back because they're using WebObjects,

0:09:43that's a really stupid reason to be held back, and there's no reason to continue it. These

0:09:47services are all pretty new. If that was really the problem, they could rewrite them. It would

0:09:53be a big undertaking, but it wouldn't be insurmountable. The fact is, I don't think that's really

0:09:57the problem. And if it is really the problem, then I'm still correct that it's a problem

0:10:03with something about high-level leadership and priorities, rather than "this technology

0:10:09can't do this." WebObjects is actually, not now, but back in

0:10:14the day, was actually a tiny example of Apple sort of doing the right thing. And I've done

0:10:19this rant several times on many other shows, and you two should be able to recite it by

0:10:24now.

0:10:25But if you're going to do web services or anything online, at the scale Apple does it.

0:10:33You go into a different realm.

0:10:36And what Marco said about you can make any web service scale on any platform is true,

0:10:39but once you start getting into Apple scale or Google scale, things like that do matter

0:10:44a little bit more than they do.

0:10:47There's a threshold through which you pass, and it's like, OK, now any old platform won't

0:10:51do.

0:10:52platform actually does matter because any tiny inefficiency is multiplied by the hodgillions

0:10:57of servers that we have, or maybe a particular architecture dictated by a particular platform

0:11:01doesn't allow us to be in 8 million different data centers around the world in a synchronized

0:11:05manner, and all these other things that come into play for like seven people in the world,

0:11:09for like maybe Amazon, Microsoft, Google, anybody who's got a worldwide online presence

0:11:15who, you know, has huge servers with just millions and millions of customers.

0:11:20And the good thing that WebObjects had going for it is that, well, they didn't make it

0:11:25themselves, but it was in-house.

0:11:26I mean, it came with Next, obviously.

0:11:28Like half of the technology they're using now all came from Next, right?

0:11:31And that's what you have to do with this scale.

0:11:33You have to take ownership of your online platform.

0:11:36You can't just use sort of off-the-shelf stuff and buy experts and have them hook them all

0:11:40up to each other.

0:11:42Once you pass a certain threshold of scale, you've got to do stuff yourself.

0:11:45Amazon does, Microsoft does, Google does practically everything themselves.

0:11:50down to, you know, specking out their own hardware and everything.

0:11:53And Apple seems to do so much less of it.

0:11:56And Apple is at that scale now.

0:11:58Hundreds of millions of people using iOS devices, connecting to iCloud, like, they're there.

0:12:02They're at that scale.

0:12:04They can't be the only person doing stuff off the shelf.

0:12:06They need to take ownership of their online platform technology.

0:12:09And I don't understand the leadership gap here too, because it's clear that the leadership

0:12:13Gap doesn't exist for the client-side stuff because the organization that does

0:12:19a Mac OS X, iOS, and all this other stuff so clearly, whether this comes from the top or not,

0:12:25but so clearly understands that it needs to take complete ownership of its platform.

0:12:31We have to be responsible in making sure we have good tools, good compiler, good language,

0:12:36and we're going to not just do them once and just say, "Okay, we're done. Coco is awesome.

0:12:39Project builder, we're all set. We've got our own tools. We're great."

0:12:43They're going to be saying, "No, we're not satisfied with that.

0:12:46We need a better compiler.

0:12:47We need to ditch Project Builder.

0:12:48We need to make a thing called Xcode.

0:12:50We need to keep making Xcode better and better.

0:12:51We need to switch out our debugger for LDB.

0:12:54We need to rev the Objective-C runtime."

0:12:57They just take such incredible ownership of their platform,

0:13:00and they know they can't just let it sit there.

0:13:01And they're not relying on some other vendor

0:13:04or some other platform to solve their problems

0:13:06and just throw it at a bunch of people and go,

0:13:07"Here you go. Here's some pieces."

0:13:09They totally take control of their client side.

0:13:11And the service side is just as important.

0:13:14They need to be doing all those same things.

0:13:16Where is the team that has been working on-- this

0:13:18is the technology that we're going

0:13:19to use inside Apple to deploy online services, starting

0:13:2310 years ago and continually revising it.

0:13:2610 years ago, they had WebObjects,

0:13:28which even then was kind of weird and dated.

0:13:29And they just didn't keep it up to date and modernize it

0:13:33and history passed it by.

0:13:35They didn't let that happen on the client side.

0:13:37They've been racing ahead as fast as they can,

0:13:39again, with the exception of the file system.

0:13:41But in most other aspects, they're taking ownership of their platform there, and on

0:13:45the server side they're not.

0:13:46They're going to third-party vendors, and that's an untenable strategy.

0:13:49They need to be more like Google and Microsoft and Amazon and have their own platforms with

0:13:55their own announced technology and dedicate those kind of resources to it.

0:13:58And I don't understand why one half of the company can do that and the other half can't,

0:14:02because it seems like it's the same leadership, right?

0:14:04Well, yeah, but why do you say that WebObjects isn't getting better?

0:14:08I mean, from the outside, there's no indication it's getting better.

0:14:11But who's to say that it isn't getting better on the inside, and they're just not letting

0:14:14anyone see it?

0:14:15The model of the way it works.

0:14:17The model of the way WebObject works is the whole idea of having object transparency and

0:14:23just kind of working like cocoa on the web and all these conveniences that you have.

0:14:26That's not what massive online service is about these days anymore.

0:14:29They're about infrastructure pieces to manage storage and data in ways that are totally

0:14:33unlike the WebObject stack in terms of where the state is and how it all fits together.

0:14:40Just look at Google's Spanner thing that they use for their database stuff and then all

0:14:45of its predecessors like GFS and what was the... Someone in the chat room told me what

0:14:50was the thing that predated... Oh, Bigtable.

0:14:52Bigtable.

0:14:53Yeah. And all these infrastructure projects that have come and gone and MapReduce and

0:14:57all those things that Google is constantly revising, Apple hasn't gone through that revolution.

0:15:01revolution. I mean, Google started out, its first things of just GFS and MapReduce were already ahead of

0:15:06where Apple was with WebObjects in terms of doing things at scale.

0:15:10And Google is constantly throwing away its old ones and replacing with new ones over and over and over again

0:15:15for its service. And same thing with Amazon and all the stuff that it's using to run its services.

0:15:20Apple

0:15:21started with WebObjects, which was already like sort of the old model. And you know, look,

0:15:26it's much more convenient for developers to do, you know, like,

0:15:28Google has proven that

0:15:31you can make things annoying for developers.

0:15:33It's certainly using Bigtable is super annoying,

0:15:35which is one of the reasons Spanner exists for developers.

0:15:38But it was like, you know, scaling is king.

0:15:40And even though it's going to be annoying

0:15:42for you to do this stuff at the application level,

0:15:43scaling is more important

0:15:45and will work out the other things later.

0:15:46Whereas Apple was like, oh, we want this to be all nice.

0:15:49And it's kind of like working with objects

0:15:50and it's real convenient

0:15:51and everything is magically objects and persistent.

0:15:53Sounds kind of like core data, I guess.

0:15:55(laughing)

0:15:56But online, and isn't that nice?

0:15:57And that is nice and everything,

0:15:58But your hands are tied behind your back in terms of how do you scale this to 17 data

0:16:03centers with redundant hardware and all these other things.

0:16:08We'll know the day has come when Apple has finally sort of joined the modern age when

0:16:12they don't have to bring the store down before they introduce new products.

0:16:18The reason they do that is not so much like, "Oh, we have to take the store down to add

0:16:22new products."

0:16:23They don't have to take the store down to add new products.

0:16:24They have to take the store down—this is my theory—they have to take the store down

0:16:28to add new products that appear to customers in a deterministic manner.

0:16:32Because they can add new products without taking the store down, but they have no idea

0:16:36when or where they'll appear for people because their architecture doesn't allow them to have a

0:16:41way to say, "Okay, now this is available for the entire world." It's just like they can add it,

0:16:45and then it trickles out through their whatever system they have going, combined with their CDNs

0:16:49and other stuff. And they don't want it to spoil the surprise. So instead, they bring the whole

0:16:53thing down, rev the whole thing, put all the new stuff in, and then just wait for the moment.

0:16:57say, "Okay, and go bring it back up." And then you're sure that nobody sees it ahead

0:17:01of time accidentally, right? And you're sure that when you do turn it on, everybody sees

0:17:04the new thing because presumably you've had time for it to propagate during, you know.

0:17:08That's my theory of why they take the store down, but that's not how the web works. You

0:17:12can't take your store down when you don't want to spoil a surprise, right?

0:17:16Do you really think taking the store down is still necessary, or do you think they're

0:17:20doing it only for the theatrical element? It's because if they didn't, they wouldn't

0:17:24have control over when people saw things.

0:17:26They want it to be visible to everybody all at once as much as possible, but only starting

0:17:32at a given point.

0:17:33I feel like this, again, is just speculation.

0:17:36If they just put it up in the store now, it would either slowly trickle out to people,

0:17:39which would be kind of annoying because you want everybody to see it when you announce,

0:17:42or if you start it early, some people might see it early.

0:17:45So I think it's about making it so that everyone sees the thing simultaneously.

0:17:50It could be purely for theatrics, as in they don't have to do it at all, but I think they

0:17:55would have stopped that by now.

0:17:56So many things that they've done for theatrics have sort of come and gone, but my guess is

0:18:00that it has to do with content propagation and being in control of when it appears to

0:18:05the first person and getting it to appear to the most people as soon as possible.

0:18:10But there's no technical reason why it has to take that long.

0:18:16They don't have to take the store down for an hour and a half

0:18:20to update something.

0:18:21- How long does it take for the new content

0:18:24to propagate through their worldwide network of CDNs?

0:18:26- It depends how they do it.

0:18:28- I know, but I'm saying maybe it actually takes them

0:18:30like two hours to be sure that all the new content

0:18:33is propagated to all the CDNs.

0:18:35- But that's a choice they make in implementation.

0:18:36I mean, they can do it within a few seconds

0:18:39if they wanted to.

0:18:40- I mean, who knows?

0:18:41This is just my guess of why they would do it.

0:18:44And you're right, it's down for a long time.

0:18:45It's not like it's down for five seconds then comes back up.

0:18:48It's down for like an hour during the whole keynote practically.

0:18:51Yeah, actually, it usually goes down before the keynote, and so it's down for like three

0:18:54hours total.

0:18:55Yeah.

0:18:56And the only other thing I think of is maybe they're trying to prevent people from accidentally

0:18:59buying the old products while they're announcing the new ones.

0:19:01I don't know.

0:19:03But you can still go into an Apple store and buy them.

0:19:05You can still buy them like that morning.

0:19:07You can still, you know, it doesn't...

0:19:09That can't be the reason either.

0:19:10Yeah, I wonder if Apple store employees wave you off.

0:19:13Like if you go in and the keynote is going on, surely everyone at Apple Store knows the

0:19:16keynote is going on.

0:19:17If you go in on the October 22nd iPad event and you go in while someone is on stage introducing

0:19:22a new iPad and you try to buy an old one, you would think the Apple Store guy is going

0:19:26to go, "You know they're announcing new ones now.

0:19:29You can still buy this.

0:19:30Here you go.

0:19:31It's fine.

0:19:32You can have it now.

0:19:33But just in case, you might want to know.

0:19:34Maybe you don't know they're announcing new ones right now."

0:19:35I wonder if they tell you that.

0:19:36I think you might be overestimating the geekiness and attentiveness of both the staff and the

0:19:41customers in an Apple Store.

0:19:43I'll be depressed if they don't even know that it's going on.

0:19:48So do you think, thinking about some more of this Apple service stuff before we go to

0:19:53a different topic, do you think Apple's really feeling pain from this? From their stuff being

0:20:02the status quo of working most of the time, but not being up to the standards service-wise,

0:20:08wise, uptime wise, reliability wise of Google services, Amazon services, Facebook services,

0:20:15the other big giants. Do you think Apple's really feeling the pain from that? Do you

0:20:19think this is really hurting them? Because with Google, they had to scale ridiculously

0:20:24well because A, not a lot of other people in their business were doing that. B, they

0:20:30were scaling way past what everyone else was doing. And C, that's their entire business.

0:20:36where everything comes from. If Google doesn't serve an ad, they lose money. Whereas Apple

0:20:41is selling all this hardware regardless of whether iMessage is down this morning. You

0:20:46know, it doesn't really hurt them directly and severely where there's one or two instances

0:20:52of downtime here and there. How much do you think, how much pain do you think they're

0:20:56feeling? Because it kind of seems like they're not feeling enough to do anything drastic.

0:21:00Well, maybe. But let me answer your question by asking you a question. Do you think Apple

0:21:06is proud. And I think it's pretty clear to me that they're a very proud company.

0:21:11And I can't imagine, to come kind of full circle, I can't imagine that they like hearing these

0:21:16potshots taken at them during Agents of Shield. Does that make sense?

0:21:20That's the question, though. The question is, it's like a Bulmer-type question. Are they in denial?

0:21:25If you were to ask them, like, you know, off the record, I'm not going to report this and

0:21:29make a story out of it, just that you bump into Tim Cook in an elevator or whatever, and it's like,

0:21:33Do you think Apple does online stuff as well as Google, Amazon, or Microsoft?"

0:21:40This could explain the leadership gap where they're like, maybe they're in denial and

0:21:46they think, "Everyone has troubles every once in a while.

0:21:49We're kind of in the mix.

0:21:50We're kind of pretty much almost as good as Google and Amazon.

0:21:53Maybe some days worse, some days better.

0:21:54Everyone has their ups and downs."

0:21:56Or do they really realize what the gap is?

0:22:00The gap is not, if you were to put it on a little graph or something, it's only a couple

0:22:04of percentage, but it's like that last couple of percentage, it's like uptime, going from

0:22:08like three nines to seven nines is just astronomically hard.

0:22:12So much harder than going from 90% uptime to 99, you know what I mean?

0:22:16The last little bits of the part that matter.

0:22:18I wonder, that could explain why they haven't sort of gotten religion on this and dedicated

0:22:23themselves to doing the service side, taking ownership of the service side tech the same

0:22:28way they do on the client side, is that they think they're not that bad.

0:22:33Maybe they really actually think, "Yeah, we have room for improvement.

0:22:37We're not satisfied with where we are, but it's not that.

0:22:39We're close."

0:22:40I would say, "No, you're not.

0:22:42It seems like you're close, but really, those last few inches on the graph make all the

0:22:46difference.

0:22:47You are not close."

0:22:48There is nobody, there is not a single person who has any technical clue who would say,

0:22:51"Do you think at any day of any month of any year that Apple does online stuff better than

0:22:55Google?"

0:22:56It's never happened.

0:22:57has ever had that opinion. It's just 100% "Okay, Google is better. How much are they

0:23:01better? Are they a little bit better?" But nobody believes that we're better. Nobody.

0:23:05And I have a feeling that if you ask someone on the iTunes Music Store, "Well, of course

0:23:08we're better. Look how many billions of songs we sold. Look how many apps we give people.

0:23:11Google doesn't do stuff like that. Their store sucks. We do so much better."

0:23:16Because they excel in a few areas of read-mostly distribution of static data to people that

0:23:21they think, "We're an online services company and we're awesome," and it's just not the

0:23:25the same as an interactive thing. Do you think anyone's ever returned an iPhone

0:23:28because iMessage was down for 20 minutes a month?

0:23:31Yeah, no, you're right about them not feeling it. It would be better if they felt they have

0:23:34more of a cushion than Google. They just do it because there's so many other interesting

0:23:37things you can do with the device that don't matter.

0:23:41You can use Google services, for example. You can use Google Maps, another area where

0:23:45they have little trouble. You can use Google Now. It's not their whole business. It's just

0:23:49part of their business. And so, yeah, no one's going to return the thing because Siri is

0:23:53wonky every once in a while. Like it doesn't hurt them as much and it's almost kind of a shame because if it hurt them more

0:23:58maybe you know like jobs I think knew they weren't you know the Google does stuff better

0:24:02Which is why I kept yelling at like the the mobile me team and having all those meetings and trying to do iCloud like at

0:24:08Least he I think he understood

0:24:10We're not good at this. We should get better. He didn't know how to make that happen apparently, but he sure tried

0:24:16So we'll see if there ever comes a time where like I mean I guess Tim Cook

0:24:21maybe did that about maps like apologizing for maps and we need to do better and some understanding that maps but maybe he's just sees

0:24:27That as a data problem and not a server problem

0:24:29I don't know like I mean the thing about maps is if I was Tim Cook could be like, okay

0:24:33Well, no, we just need a better map that it's like, okay Tim. So you'd haven't forgotten about the fleet of cars

0:24:39They're gonna drive over every single road in the entire world and take pictures

0:24:42Did you forget about that part because you don't get that for free if you just get better map data, right?

0:24:46Google did that that's a crazy project. It's huge and it's ongoing

0:24:50Right, and they don't stop doing it. And if you don't have an answer for that, you're not gonna have Street View

0:24:55You're not gonna have like let me fly 3d through the middle of Fog Creek's offices like you're so far from that

0:25:00So how are you ever going to compete and it could be that he's like, okay

0:25:03Well, we're never gonna do that

0:25:04We just need maps to say we have maps

0:25:06And they can just continue to use Google Maps from our things and Google Maps are always gonna be better

0:25:10But that's not an Apple kind of attitude, you know

0:25:13and and it's not like when I say this is like a

0:25:18substantial problem in the company that probably

0:25:20would never change.

0:25:21Like imagine, to take another example in the industry,

0:25:24imagine if some Microsoft CEO comes in,

0:25:29which I guess is plausible,

0:25:32some new Microsoft CEO comes in,

0:25:34and they say, you know what, our devices aren't cool.

0:25:38Everything we make, our hardware is not cool,

0:25:41our software is not cool, nobody thinks our stuff is cool.

0:25:45Let's hire somebody in charge of keeping things cool,

0:25:48or let's increase the funding to our cool department by 50% this year. Do you think

0:25:55that's really going to change it? It's not that easy.

0:25:58I feel like looking at Apple and saying, "How can they address the services issue, that

0:26:05their services aren't that good traditionally and continue to be that way?" I don't think

0:26:10there is an answer. I don't think that's the kind of thing that can change in a large

0:26:15I think either it's a priority from the start or it's not.

0:26:19And the elements that Google has that enabled them to produce these kind of services that

0:26:26scale very well and that prioritize all these things, Apple just doesn't have those elements.

0:26:32And in the same way that Google is never going to make something client-side that has the

0:26:36kind of quality and taste of an Apple client-side platform and software, I don't think Apple

0:26:43will ever have what it takes to make Google quality services.

0:26:47But the good thing in Apple's favor is that the thing that Apple has that

0:26:51Microsoft and Google seem not to have is it's kind of like it's one of those

0:26:54things you label as intangible in scare quotes because it's not intangible but

0:26:57it's like it's more mysterious whereas the things that Google has are

0:27:02tangible and a great example of it is Google was a company that made web

0:27:05search and they indexed the entire web which is an amazing technical feat,

0:27:08right? But they decided they wanted to have a client-side OS and they

0:27:13did that by making a client-side platform that they completely took ownership of.

0:27:17They have their own, you know, don't call it Java Dalvik VM, their own API, their, you

0:27:23know, their own IDE, their own store.

0:27:25Like they understood if we want to have a client-side mobile platform, we need to own

0:27:30it.

0:27:30We need to own the technology from top to bottom.

0:27:32We're going to like define the VM.

0:27:33We're going to define the language sort of, you know, it's Java, whatever, with native

0:27:37client and all of those stuff.

0:27:38And we're going to have the ID.

0:27:39They understood that, Hey, we were a server-side company, but we want to get into

0:27:42client side and we can't do it by licensing a bunch of software from someone else. We

0:27:47have to take ownership of it. So that is a tangible thing. And I think that's proof that

0:27:51Apple could, if it wanted to, say, "We need to get into server side," which they have,

0:27:56but also say, they forgot to say, "And we need to take ownership of it the same way

0:28:00we take ownership of the client side." The thing that's harder to transfer is taste and

0:28:05culture and coolness. That is much harder to do because there is no coolness department

0:28:09at Microsoft. You can't increase funding and 50% to something that doesn't exist, right?

0:28:14And maybe it's just like, "Oh, well, designers are tangible. We could just steal them all

0:28:17from Apple." But that's hard to do because they don't want to leave Apple and go to—Microsoft

0:28:21has good designers too in their Metro stuff.

0:28:23Well, and that wouldn't work because the structure around them is different there.

0:28:28Right. But I think the tangible things of a server-side company proved that it can get

0:28:32into client-side and take ownership of it. I think there's no reason that a client-side

0:28:35company can't prove that it can get into server-side and take ownership of the tech

0:28:38It just didn't do it.

0:28:39Whereas the intangibles about coolness and taste and design,

0:28:44that is much harder to do.

0:28:45Like more rarely do you see a company like Google or Microsoft

0:28:49saying, we're not cool and stylish.

0:28:50We're going to put some effort in.

0:28:51And Google has tried to increase its style, and so has Microsoft.

0:28:54But none of them-- they're still not reaching the heights

0:28:58that Apple gets in terms of taste and design.

0:29:00They've both made efforts in that area,

0:29:02but it seems harder to do.

0:29:03Whereas I would say technology-wise,

0:29:05Google has as much control over, you know,

0:29:08pending whatever the lawsuit is with Oracle,

0:29:10but as much control over and like determining

0:29:13its own destiny for all of its core tech,

0:29:16all of its server-side tech, all of its client-side tech,

0:29:18hell, they even took WebKit back and they have Blink now.

0:29:20Like, they have taken ownership of their tech stack

0:29:23and so has Microsoft.

0:29:24Microsoft has always had ownership

0:29:25because they don't like to use anything from anybody else.

0:29:27And same thing with Amazon.

0:29:29You know, even they took Android

0:29:30and they took ownership of it.

0:29:31You know, we're not gonna even call it Android.

0:29:33it's going to be whatever the hell the Kindle OS is, right?

0:29:36- Sh*t OS?

0:29:37- Yeah.

0:29:38(laughing)

0:29:39I think it is totally within the realm of possibility.

0:29:41You do have to change how the company works,

0:29:43like the structure of it and stuff like that,

0:29:45but not in any more radical way than Google had to change

0:29:48when it made Android, you know?

0:29:50- This episode is brought to you in part

0:29:53by our friends at Squarespace.

0:29:55Squarespace is, as usual, the all-in-one platform

0:29:59that makes it fast and easy to create

0:30:01own professional website or portfolio. For a free trial and 10% off, go to squarespace.com

0:30:07and use offer code ATP10 for this show in the month of October. Code ATP10 at checkout.

0:30:1310% off Squarespace.

0:30:15So we've been telling you all this time about Squarespace's awesome platform. They're constantly

0:30:19improving it. You can have new features, new designs, better support all the time. 70 people

0:30:25dedicated to support and growing right here in New York City. And Squarespace is awesome.

0:30:30It starts at just $8 a month, and that includes a free domain name if you sign up for a whole

0:30:34year up front.

0:30:35And they have responsive design, they have SU optimized templates, everything in Squarespace

0:30:40is all set, ready to go.

0:30:42However, what they've also added recently is this entire flexible commerce platform.

0:30:48Squarespace commerce is currently available in the US, UK, Canada, Australia, Belgium,

0:30:52France, Germany, Ireland, Netherlands, and Spain.

0:30:55You can sell any kind of product using a very simple single interface.

0:30:59Digital goods, physical goods, whatever you want you can sell it through Squarespace Commerce.

0:31:04They partnered with our friends at Stripe and Stripe is awesome and everyone loves them

0:31:07for a very good reason.

0:31:08Stripe handles everything so Squarespace doesn't charge any fees on top.

0:31:12Stripe takes 2.9% plus 30 cents per transaction.

0:31:15That's it.

0:31:16Everything else is included in your Squarespace plan.

0:31:18Squarespace Commerce also offers an express checkout mode which bypasses the shopping

0:31:22cart interface if you want to.

0:31:24The order management interface lets you track outstanding orders, resend customer update

0:31:28emails, you can print packing slips, all from a single easy to use interface.

0:31:33They even integrate with all the regulations so that they handle tax, shipping rules, everything

0:31:39that you need to know to be able to sell physical or digital goods.

0:31:43They're PCI compliant, everything is secure via SSL, everything you'd want in an online

0:31:48store Squarespace Commerce offers.

0:31:50So start a trial with Squarespace today, whether you're doing commerce or just a site for yourself,

0:31:55or a blog, or a portfolio, or a site for your business that doesn't involve commerce.

0:32:00Whatever the case may be, you can do it on Squarespace.

0:32:04So they take care of the hosting so you don't have to.

0:32:07You can start a free trial today with no credit card required.

0:32:10A real free trial.

0:32:11Just go to squarespace.com, start building your site, see if you like it.

0:32:15If you like it, use coupon code ATP10 and check out for 10% off.

0:32:19Thanks a lot to Squarespace for sponsoring ATP.

0:32:22They are everything you need

0:32:22to create an exceptional website.

0:32:25- It's ads are making me depressed

0:32:26because I've been wandering websites recently

0:32:29that look like they were designed in 1993 by some person

0:32:31that they probably paid tons of money.

0:32:32I'm looking at home improvement type,

0:32:34construction companies and stuff like that, their websites,

0:32:38they look like they probably spent $15,000

0:32:41to give to some teenager in 1994 to make their website.

0:32:44- It's a Merlin's old company.

0:32:45- Yeah, or like in simple things

0:32:47where you could make appointments or buy things,

0:32:48and it's like, look, if you just spent like $8 a month

0:32:51on Squarespace and put up a Stripe for it,

0:32:53like that experience in all ways, it's cheaper for you,

0:32:57it's nicer for your customers, you'll make more money,

0:32:59it's so cheap, especially when you look at the prices

0:33:02for the work they're gonna do, like you guys have the money,

0:33:04you have all the money in the world,

0:33:05you know the prices you charge, it's so cheap,

0:33:07just get a Squarespace, and I bet they're so proud

0:33:10of their sites.

0:33:11- Oh yeah, they invested so much money early on.

0:33:13- Right, it was like, ugh, 15 grand site in 1994,

0:33:15"No, tear it down. Pay $8 a month." God. And I was one of those people building those

0:33:21sites for people, and everything I built was terrible compared to what any modern web CMS

0:33:27offers you out of the box. Everything I did back then was horrible, and I charged people so much

0:33:31money because it was all one-offs. But now you can just go to these big platforms.

0:33:35I wonder, all these restaurants that have these flash-only sites,

0:33:39don't get me started. They have to be losing more than $8 a month in sales

0:33:45to keep having their stupid Flash site.

0:33:47Yeah, some sites do get a clue, but they still, like, you can tell they paid someone in the

0:33:51last 10 years to do it, but it's still gross. It's still like, you know, just, please, just

0:33:56use Stripe. Every time I can now recognize, like, the Stripe form, like, the fact that

0:33:59it's all client-side and everything, it's like so beautiful. It's just like, I feel

0:34:03kind of bad because, you know, I made e-commerce sites too, and it used to be such a pain to

0:34:07make e-commerce sites. And I was like, if we had Stripe back then, it would have been

0:34:10done in an afternoon.

0:34:11Oh, yeah.

0:34:12It has all the features that I'd spent months implementing manually.

0:34:16There was no companies that you could outsource this stuff to.

0:34:19No one knew what they were doing.

0:34:20We were all just like, "I guess we'll take money from people over the computer?"

0:34:23It seemed like a crime.

0:34:24We were getting away with something.

0:34:27And Stripe really isn't that old.

0:34:29When it came out, what was it, maybe three years ago?

0:34:32It's not that old.

0:34:34I just remember even then, even three years ago, taking money online was still a hassle.

0:34:39basically had the PayPal API, which is a disaster in every possible way. Having run Instapaper

0:34:47subscriptions off the PayPal API, I cannot tell anybody enough how much they should not

0:34:51use PayPal for anything. Even if you ignore all of the crazy stories about how PayPal

0:34:59locks your account and takes all your money, that's bad enough. But even when everything's

0:35:03working as intended. It's terrible. It's absolutely the worst thing in the universe.

0:35:10And for managing recurring subscriptions, there's no way to get a list of subscribers.

0:35:16Still, there is still no way to get a list. You just have to take in all those messages

0:35:20saying "hey, new person" and keep track of them yourself.

0:35:23Well, it's like Newsstand, only you can't even do that in Newsstand.

0:35:26That's true, I think. Yeah, although Newsstand has the weird thing where you can go into

0:35:31iTunes Connect and download everybody's zip code. I don't know. But anyway, yeah.

0:35:34PayPal is the worst. When Stripe came out a few years ago, whenever that was,

0:35:39the reason why it made such a splash was on their page it was like, "Here's a cURL command,

0:35:45and here's a block of JSON that's equivalent. Here's how you charge somebody's card."

0:35:49And it's like these four lines of JSON, there's one cURL command. You're like, "Oh my god.

0:35:53That's so much better than everything else I've ever seen to this point."

0:35:57I don't have to form a soap message.

0:35:58[laughter]

0:35:59Nice.

0:36:00There was the Amazon payments story in the news this week, too, right?

0:36:07Speaking of the death of PayPal, PayPal is the experts exchange for some of the other

0:36:12businesses.

0:36:13Stack Overflow is the experts exchange as, insert the blank, is to PayPal.

0:36:17And insert the blank is Amazon payments, Stripe, and all the other companies that finally recognize

0:36:21that despite PayPal being the 800-pound gorilla, everybody hates it.

0:36:24It's terrible.

0:36:25needs to die and all it takes is some new services that can nip away at it.

0:36:31Dwalla's weird. Have you seen Dwalla?

0:36:33I've heard of it.

0:36:34Is that the interest-free lending thing?

0:36:37That's Kiva. I think you're right.

0:36:38Oh, no. That's the one where you pay a quarter and you can send money to anybody?

0:36:43Yeah. It's basically—I'm not sure how it works. I think it might be based on ACH

0:36:48or something. But somehow, it doesn't use the credit card network to move money around.

0:36:53So they charge only 25 cents, but then there's no fraud protection.

0:36:58It works a lot like cash, something like that.

0:37:02I don't know all the details, but somehow they're able to only charge 25 cents for

0:37:08pretty much any size money transfer.

0:37:10It sounds like a complicated money laundering scheme that you are an unwitting participant

0:37:15in.

0:37:16What was weird is, I signed up for that a few months ago because that's how we were

0:37:20doing some of our ad payments. And it feels wrong. It feels suspicious to me to only pay

0:37:3125 cents to move a giant chunk of money somewhere. It made me uneasy.

0:37:36But it shouldn't. It shouldn't, though, because the fact that it costs money to transfer

0:37:41money is an artificial construct, mostly artificial construct based on the old world. Now that

0:37:46we have all our computers are connected together. But the problem is there's no sort of secure

0:37:51standard for, there's no common secure standard for transferring money other than I guess Bitcoin

0:37:57or whatever. Like really secure. And everything is kind of this strange game of like trust and

0:38:03parties are sufficiently trustworthy and you assume that they're not, and then they communicate

0:38:10over this terrible protocol that it's like a check where you just need like the account number and

0:38:15and routing number and somehow you can send money

0:38:17into people's accounts.

0:38:18The entire banking system has not kept up

0:38:20with current technology and it's all kind of a house of cards

0:38:24that were just kind of saying, all right, everybody,

0:38:27let's just not blow this here.

0:38:28So things like Walla come in and it does seem crazy,

0:38:31but you're like, you know what?

0:38:33We shouldn't be able to transfer money seamlessly

0:38:35from account to account.

0:38:37Why would we need a middleman for that?

0:38:38And it's because, well, we don't have any real secure protocols.

0:38:41And if we did the NSA, he's dropping on them anyway.

0:38:44Well, you know what kind of along the same lines is a Square Cash, which I think it was

0:38:49Andre Arco sent, it might've been him, maybe not.

0:38:53Somebody sent me a few dollars of Square Cash just so I could try it out.

0:38:56And basically what's the way Square Cash works is you send an email to whoever, whomever,

0:39:01whatever is supposed to receive the money.

0:39:05You CC their email address, the Square Cash email address, and you put the dollar amount

0:39:12subject line and then Square will send an email to the person receiving the cash saying,

0:39:18"We think somebody is about to pay you.

0:39:20Hold on."

0:39:21They send an email to the person that sent it.

0:39:23So they send it to you saying, "Hey, man," or girl, "Are you really sure you want to

0:39:27send this money?"

0:39:29And then what happens is the person receiving the money just inputs their debit card account

0:39:34number and all the money is transferred and I think it's 50 cents a transfer.

0:39:38I'm sure there's a limit in terms of how much you can send, but I don't know what it is.

0:39:43So it must be pretty high.

0:39:44And it works flawlessly.

0:39:47You did that with us.

0:39:48I think you sent me a dollar or something to try it out.

0:39:50That was pretty neat.

0:39:51And people in the chat room are saying that they have a system in Europe that works with

0:39:53this.

0:39:54It does not surprise me.

0:39:55Europe is the new Japan.

0:39:56Remember, it used to be that in Japan, they had flying cars and hoverboards, and all we

0:40:02had in the '80s was Donkey Kong and Casio keyboards, but they had the cool stuff.

0:40:06And then there are economy tank.

0:40:07But now it's like in Europe they have socialized medicine and unemployment.

0:40:15And Switzerland was passing a law for the minimum monthly income to be the equivalent

0:40:20of $2,800 for all citizens.

0:40:23Yeah.

0:40:24We're going to get so much email from this from people who are like, "I can't believe

0:40:28how terrible your system in the US is.

0:40:30Don't you know how much better this is in Europe?"

0:40:32And the answer is yes, we know.

0:40:34We are very aware of how terrible our system is.

0:40:36Please, you don't have to tell us.

0:40:38Although I think, like, I mean, they're saying, you know, they just need two numbers and they

0:40:41can transfer money to each other.

0:40:42Like, they still don't have, like, Bitcoin is actually at the forefront of this technology,

0:40:46as sad as it is with their, you know, crazy thing that they have going on there.

0:40:49At least they have sort of an attempt to make a secure, you know, mathematical foundation

0:40:54for distributed middleman-less transfers that everyone could be assured are, you know, happening

0:41:00correctly.

0:41:01You know what else Europe does right?

0:41:04and I bring this up only briefly because I just watched a video from Mythbusters about

0:41:07this is roundabouts.

0:41:09So the Mythbusters, which are clearly the bastion of all things good about science,

0:41:13and they never ever flub anything ever, they did a test with a four-way stop versus a roundabout,

0:41:18and the roundabout crushed the four-way stop in terms of throughput.

0:41:21I just thought that was interesting.

0:41:22Oh yeah, I don't even think that's up for debate.

0:41:24I think almost every study has always proven that, that roundabouts really are way better

0:41:28for throughput.

0:41:29Yeah, no, we have a ton of them around.

0:41:31I don't know if you have them down where you guys are, but yeah, Massachusetts is the land of the roundabouts. We have plenty of them

0:41:36Yeah, no, they're all over the place here. Well, they're becoming popular here

0:41:39And it's funny watching everyone try to navigate them because nobody knows what they're doing. Yeah

0:41:45John while I'm thinking of it, how's the review going?

0:41:50Do you feel like you have a release date based on the what is it?

0:41:53October 22nd iPad event that everyone's kind of assuming will be Mavericks as well

0:41:58And we got the GM two hours after the last show.

0:42:01Oh yeah, that's right.

0:42:02That's right.

0:42:03I think it turned out that Apple Insider story that I was poo-pooing, they had it right.

0:42:05It was just that they were talking about the people who get the early seeds.

0:42:08As soon as I saw that story, I went to the Apple developer website and saw nothing there.

0:42:12I'm like, "Oh, maybe it's just bogus."

0:42:13But it was just a staged rollout.

0:42:15The super-duper Apple seed program, people got it, and then a couple hours later, we

0:42:21got it.

0:42:22So the update on the review is, "Yay, I've got a GM."

0:42:24Boo, I had to redo a whole bunch of screenshots.

0:42:27God, the things they changed are just insane.

0:42:30They changed functionality too, so I had to rewrite an entire section because everything

0:42:34I had written about it and all the screenshots that I'd taken were no longer there.

0:42:37And the worst thing about the section I had to rewrite is I don't understand one of the

0:42:40things that they changed.

0:42:42They changed the pop-up menus, where they are, what the choices are, and I understand

0:42:47all the choices except for one.

0:42:49And so right now in my review I have written that I could not figure out what this meant.

0:42:54I tried.

0:42:55I tried so many.

0:42:56Maybe it means this.

0:42:57Let me try it.

0:42:58Nope.

0:42:59Maybe it means this.

0:43:00Let me try it.

0:43:01Like, I could not figure it out.

0:43:02And of course, I asked Apple.

0:43:03Like, I have so many questions into Apple, but they, you know, ignore me.

0:43:04So maybe I'll hear from them.

0:43:05Maybe I won't.

0:43:06But having a GM build is good, so I did a lot of work revising things.

0:43:09By the end, I was getting pissed because it's like, I mean, this, we've all seen the screen

0:43:14chest.

0:43:15You know, you know the labels or the tag things or whatever in the Finder?

0:43:18You've seen the screen chest.

0:43:19They put, like, a colored dot next to the file name if you give it, like, a red label.

0:43:23They change the size of those dots by, like, a pixel.

0:43:25I was like, "Come on, guys."

0:43:26They did that just to troll you.

0:43:28Every single screenshot that had a freaking dot.

0:43:30Yeah, I was like, "And I like the old size better."

0:43:32They made them slightly smaller.

0:43:34I'm like, "Oh, come on."

0:43:35So now it's just like...

0:43:36If I had to redo a screenshot because they totally changed the way something works, fine.

0:43:41But I had to redo it because you made the dot smaller, that's just cruel.

0:43:44And then the one I tweeted was that one pop-up menu was two pixels farther away from the

0:43:49label.

0:43:52We have to have some very high ranking design manager on OS X has to be a listener of this

0:43:59show.

0:44:00I was just about to say that, and if that is the truth, can you imagine, or even if

0:44:03it's just whoever's in charge of that particular screen just thinking to themselves, "You

0:44:07know what?

0:44:08I'm gonna troll John Syracuse."

0:44:10The reason I moved the one with the pop-up menu, that one I kind of gave them a pass

0:44:13for it because it was misaligned in the pre-release builds.

0:44:17You know what a dialog box looks like.

0:44:18It has a bunch of pop-up menus.

0:44:19It's supposed to be all kind of left.

0:44:20Their left edges all line up.

0:44:22And one of them wasn't.

0:44:23The top one was sticking out more than it should have.

0:44:26They were simply correcting.

0:44:27Like this is how you can tell it's a GM.

0:44:28Someone did a once over on every single screen and when someone in interface builder didn't

0:44:32drag the little thingy to be lined.

0:44:34You know what I mean?

0:44:35So they realigned it.

0:44:36And it's like, "I gotta retake the screenshot."

0:44:39At that point I was like, "Really?

0:44:41Is every screen changed in some small way?"

0:44:45People are asking me on Twitter, "Is there anything that I'll let slide if it's not off?"

0:44:50That pop-up menu is off by a pichlerite.

0:44:52Once you see that it's off by a pixel, you can't unsee it.

0:44:54I'm not going to leave the pre-release screenshot in there, not just because it's pre-release,

0:44:57but because it's like, "Hey, the top pop-up menu is in a line.

0:45:00Someone screwed up an interface build it."

0:45:01So I'm going to fix that one.

0:45:04In every review, there's at least two or three shots that are not from GM and different ways

0:45:10that only I would notice, and I've never been called on it.

0:45:13So I shouldn't even say this, because now someone's going to go sit there and graphically

0:45:16diff every single thing and find the one that...

0:45:18But rest assured, there will be shots in there that were taken not on the GM, but in immaterial

0:45:25ways of like, you know, a pixel here, a pixel there, that no one will ever notice.

0:45:30And I'm okay with that, because seriously, I'm not going to redo every single one of

0:45:33these screenshots.

0:45:34The only person on Earth who would notice is you.

0:45:37Or like the person who, you know, the graphic designer, who's like, "I changed that.

0:45:40That's a different color now.

0:45:41That just, you know..."

0:45:42Nope, not even them.

0:45:43Well, maybe.

0:45:44Maybe.

0:45:45Like, we'll see.

0:45:46Because it gets cumbersome redoing it, especially when, like, if it was just changing screenshots,

0:45:50fine.

0:45:51But when I had to redo a whole section because they totally changed the functionality, it's

0:45:53like, you wait till the GM build to massively change its functionality.

0:45:58Oh, well.

0:45:59And do you think it's ready to be called a GM?

0:46:02Remember when they started doing this?

0:46:04They'd be like, "Here's the GM seed."

0:46:06And we're like, "What does that mean?

0:46:07What is a GM seed?

0:46:08Does this mean you are seeding us the GM?

0:46:11Or does this mean it's like, will we ever see a release where it says GM seed 2?

0:46:17Because that's the big fear.

0:46:19It's like, what?

0:46:20GM seed 2?

0:46:21That shows that you were misinterpreting the previous title.

0:46:22The previous title did not mean it was a seed of the GM.

0:46:25It meant this is, it's like release candidate.

0:46:28They used to use that terminology of like, maybe this is GM, here you go.

0:46:33And so I really hope there will not be another one that says GM seed 2.

0:46:36And I don't think there will.

0:46:37I think they could ship this and it would be fine.

0:46:40I need a price and a date.

0:46:41I need a price and a date.

0:46:42I have right now the text in my review says I don't know what the price is.

0:46:47Because now I can press the button and like publish it and you know I could submit it

0:46:51to an e-book store and it would be valid assuming they don't change the GM.

0:46:56But I also have two more different versions based on two different guesses of what the

0:47:00price might be.

0:47:01So I need a price, I need a date.

0:47:03I'm waiting patiently.

0:47:05What else is going on?

0:47:06How's the overcast going?

0:47:08This episode is also brought to you in part by Audible. Audible is the leading provider

0:47:12of downloadable audiobooks with over 150,000 titles in virtually every genre. Their catalog

0:47:19is huge and it grows constantly. Only a few months ago we were saying 100,000 titles.

0:47:24Now it's 150,000. They're huge and always growing. If you want to listen to it, Audible

0:47:29has it. You can listen to audiobooks anytime, anywhere. They support iPhones, iPads, computers,

0:47:35Kindles, even old iPods, if you're one of those people still carrying an old iPod, or

0:47:39if you use one to exercise or whatever the case may be, you can even play audiobooks

0:47:43there.

0:47:45Audible is offering ATP listeners a free audiobook along with a 30-day trial.

0:47:49Go to audiblepodcast.com/ATP to take advantage of this special offer.

0:47:56So John, from what I understand, you actually have a book recommendation.

0:48:01Yeah, this is not a new book recommendation because anyone who's listened to my past podcast

0:48:07knows that I talk about the same three books over and over again.

0:48:10This is one of them, but it's in different contexts this time.

0:48:12It's in the context of an audiobook.

0:48:14This book is ... I don't read a lot of biographies, but this is my favorite biography that I've

0:48:18ever read.

0:48:19It also won a Pulitzer Prize, so it has some pedigree to it.

0:48:22It is also, interestingly, a book that most people will never, ever read on their own

0:48:28in paper form, because it's like over 1,000 pages.

0:48:31And truth be told, it's like, if you're not

0:48:34into the subject matter, that's a lot of pages

0:48:36to read about one person, no matter how interesting they

0:48:38were.

0:48:38Now, I am super into this one person,

0:48:40because the biography is of Robert Moses.

0:48:44I knew this was going to be.

0:48:45And I grew up on Long Island, and I

0:48:47went to all the beaches and parks

0:48:48that they talk about in here.

0:48:49And it's just an amazing experience for me.

0:48:51I'm such a crazy big fan of Long Island

0:48:55to read about how all these things that I enjoyed

0:48:57my youth came to be in their sort of tortured history, and the interesting man that was

0:49:01behind them.

0:49:02And it really is an amazing book, but I recognize when I recommend it to people, they're like,

0:49:06"Yeah, I'm going to read this phone book," or whatever, "So get the audiobook."

0:49:10And this, unlike I think in the past episode where Margaret said he liked to bridge things,

0:49:13this is an unabridged audiobook of a thousand-page Pulitzer Prize-winning biography.

0:49:17Do they have an abridged version?

0:49:19Out of curiosity.

0:49:20No, you don't want the abridged version.

0:49:21You want it to be like this, like, "Admit to yourself that you're never going to read

0:49:24it, right?

0:49:25And instead, just use it as an audiobook."

0:49:2766 hours. It's like more than two and a half days of audio. Say you're going on a cross-country

0:49:34drive or something. This is what you want. Think about the value you're getting for your

0:49:40money for this. Use this as just an unbelievable value, enriching your life in a way that you

0:49:45would never do on your own because you have to admit to yourself that you're never going

0:49:48to read this book, but you will just stick it in your iPod and listen to it on your drive

0:49:53to and from your colleges across the country or whatever. So highly recommended. The Power

0:49:57Broker, Pulitzer Prize winning by Robert A. Caro, a man with an amazing accent if you

0:50:02Google him and find some YouTube videos, all about Long Island and Robert Moses, two things

0:50:06near and dear to my heart.

0:50:08That is awesome.

0:50:09Although not Robert Moses himself because he was a terrible person. But anyway, read

0:50:13the book!

0:50:14All right, thanks a lot. That is so like the typical John Siracusa pic.

0:50:21Yes, it is. Yes, it is.

0:50:22Thanks a lot to Audible for sponsoring ATP. Go to audiblepodcast.com/ATP to take advantage

0:50:28of our free 30-day trial. Thanks a lot, Audible.

0:50:32So was that your hint that you don't want to share what's going on with Overcast?

0:50:37There's not a whole lot to share. I mean, another week went by. I'm working on it. I've

0:50:43been working on it. I'm doing a little more with the sync engine. Today I was writing

0:50:49playlist sync and a couple of other preference sync type of things. Nothing really exciting.

0:50:57Just optimizations, fixes, and just moving forward with the feature set. Adding podcasts,

0:51:03removing podcasts, that's all still in the works. There's still a whole lot missing.

0:51:09But yeah, that's about it. Nutty Gamer in the chat wants me to talk about pricing for

0:51:16the app. And pricing is interesting, but honestly, I have not really made a decision yet. And

0:51:25I'm not just saying that to be coy. I really haven't decided what I'm going to do yet.

0:51:30There's a lot of options. I really don't think I'm going to do the paid-up-front thing.

0:51:37There's a lot of value to that. And I could, with the new iOS 7 receipt validation API,

0:51:44I could launch as paid up front, see how it goes, and then move to free within app purchase

0:51:50and simply migrate those users over who bought it. You can migrate them over and you basically

0:51:57know if they bought the paid version or not when you make the same app free. So you could

0:52:02move from paid to free within app purchase easily for the first time ever, whereas before

0:52:07iOS 7 you could not do that. There was no good way to do it and you just anger everybody.

0:52:13So I could do that, and I've certainly thought about doing that.

0:52:17I'm leaning towards not, but the reality is also I'm probably still at least three

0:52:24to four months away from release.

0:52:27At least.

0:52:28That's why I said I'm going to try to get it out this year.

0:52:32In reality I think January or February is more realistic and more likely.

0:52:39So all these pricing dynamics could be different then.

0:52:43I might change my mind on the business model in the last month.

0:52:47I really don't know.

0:52:51That's about it, I guess.

0:52:52BD Fortin in the chat asked a good question, which I addressed on another show.

0:52:57I don't think I addressed it here.

0:53:00What about an iTunes match style subscription to remove ads and give the money to the podcasters?

0:53:05Basically the readability model.

0:53:07Collecting money from people and then distributing it to what you listen to.

0:53:11There's a number of practical problems to that, most of which is what Readability faced,

0:53:17which is if you default to collecting money for everyone without them claiming it, it's

0:53:21kind of weird and there's a lot of issues with that. I could do something like integrate

0:53:26with Flatter. Flatter is a decent service. It's not really my style, but it's a good

0:53:31service, and they have good intentions and stuff like that.

0:53:34I think the biggest problem with the podcast app or platform collecting money for everybody

0:53:40and distributing it out is that I don't think you could get any number of podcasts

0:53:48to really agree on how they want to do that, how they want to message that, how they want

0:53:52to receive that money. A lot of podcasts already collect money directly through themselves and

0:53:57wouldn't want the competition. A lot of them, it would cause confusion as to who the people

0:54:01should be paying. Yeah, the advertisers. I mean, it's not good for advertisers either,

0:54:05because that hybrid thing makes nobody happy. You can either have a broadcast that's less than or

0:54:09supported or you can apologize as ad supported. But when you try to do both, it's like, well,

0:54:13the advertisers are pissed that they're not getting those people who are paying to skip

0:54:16their ads. And so you're advertising to fewer people, and then some people are pissed because

0:54:22they feel like they have to pay for it or should pay for it so they can skip the ads

0:54:25because they'll know that some people are skipping the ads and they're not. So it's

0:54:27much cleaner to say, "Look, it's free. It's supported by ads." Or it's you pay for it

0:54:32and they give it to you like the magazine type model where there's no ads, you just

0:54:35pay money. Those are so clean and understandable and you don't have this confusion. A hybrid

0:54:39solutions, especially hybrid solutions that you impose on people that they haven't chosen

0:54:45to do, or just sadness all around, I think.

0:54:48Exactly. And John, it wasn't one of the last hypercritical episodes where you talked about

0:54:52how advertisers almost always outbid the listeners or audience directly. That's very true. Almost

0:55:00always, a podcast can make more through ads than through direct payments. But anyway,

0:55:05I don't think it's the platform or app's responsibility to monetize podcasts.

0:55:11I think every show is going to have a different audience with different needs and different

0:55:15priorities, and I think I should just leave it up to the shows and their producers how

0:55:21they want to monetize and where they want to do that.

0:55:24Also, from a practical point of view, didn't Instacast have flatter integration for a while

0:55:30and it caused tons of problems with app review?

0:55:33Anytime you collect money in an app, either not through Apple or if you're collecting

0:55:39money to distribute in some other weird way, you know, and telling people that in the app,

0:55:44you're running a very big risk of being rejected for any update you try to make or being kicked

0:55:48out of the store when you're already in it.

0:55:50And because that's right tiptoeing along the line of what Apple will allow with in-app

0:55:54purchase rules.

0:55:55And it is just--oh, here's the link.

0:55:58Thank you, underscore David Smith.

0:56:00it is just not worth even risking that.

0:56:03And to build a major feature around depending on that

0:56:06is not wise.

0:56:07So not only do I think it wouldn't really

0:56:10get past Apple very reliably, but I also

0:56:12don't think it's a very good idea for one particular podcast

0:56:18app or even any group of them to try to create and enforce

0:56:22a new universal podcast monetization model where

0:56:25every show is going to want to do something different.

0:56:29you know, even the language around collecting the money.

0:56:31Like, I've, I talked forever ago on Build and Analyze

0:56:35about Flatter and TipJoy, I think,

0:56:38and a couple other things like that,

0:56:40where I was saying, I don't like that,

0:56:43I don't like the idea of having like a tip jar on my site.

0:56:47You know, like the, just socially,

0:56:49that's kind of a weird thing to me,

0:56:52and I would want to very carefully control

0:56:55any language or implication or pressure

0:56:58around asking people to give me money for something.

0:57:01And everyone's gonna have different opinions

0:57:03on what that is for them,

0:57:05and what they wanna present to people,

0:57:08what they wanna ask people to give or to do,

0:57:10and in what context and with what language.

0:57:12So there isn't one solution that's gonna please everybody

0:57:16who cares about this stuff as much as I do.

0:57:18- You know, another thing that we've gotten

0:57:20a lot of feedback on,

0:57:22and I don't know if there's really anything you can

0:57:24or do have to say about this,

0:57:25but a lot of people seem to have taken offense

0:57:28you saying that the now playing screen is the only thing that matters. Do you have anything

0:57:32that you'd like to clear the air about? And maybe not. I don't know. I mean, I didn't,

0:57:36I don't, I can see both sides of the story, so I don't know. I don't have any particularly

0:57:40strong opinions about it, but I didn't know if you had anything to say.

0:57:43Yeah. I mean, I don't want to take too long on this cause I don't want to make it too

0:57:46boring. But, um, yeah, last episode I, I threw off the comment that I was focusing a lot

0:57:52of my design effort on the now playing screen and that the rest of the app could just be

0:57:56a bunch of table views and it wouldn't really matter that much because you spend the most

0:58:00time navigating the NowPlayingScreen.

0:58:03Whatever exactly I said there, we got a lot of email about it from people saying, "That's

0:58:07wrong.

0:58:08As soon as I start playing a show, I turn the screen off, put the phone in my pocket,

0:58:11and that's it."

0:58:12So I'm interacting more with the rest of the app.

0:58:15So what I was getting at, whether I said it or not, who knows if I messed up, oh well,

0:58:21That's the reality of talking for two hours every week and unrehearsed and with no preparation

0:58:26What I what I mean is

0:58:30The now playing screen has very frequent

0:58:32interaction whether you're skipping a section or using the scrubber or

0:58:37Playing and pausing because so, you know, you got to listen to something somewhere else very frequent interaction

0:58:42Some of that you can do with the remote with a clicker in some cases some of it

0:58:45You can't some of it you're doing directly. So

0:58:48That screen to me like I'm everything else about the app

0:58:52it matters a lot less how it's designed because

0:58:57No matter how you design it

0:58:59You're probably scrolling through some kind of list or collection of shows

0:59:02That each within it has a list of episodes or maybe you have playlists that have a list of episodes within them

0:59:08Whatever the case is those are pretty straightforward designs like yeah, you can you can tweak it here and there

0:59:14You can add little flourish here and there the you know, there's there's a lot of little decisions

0:59:18you can make differently, but structurally,

0:59:20navigationally, the rest of the app

0:59:22is not that hard to design.

0:59:24It just isn't.

0:59:26If you want to go that route with your podcast app,

0:59:28feel free.

0:59:29But the fact is, it doesn't really

0:59:31matter how I present a list of episodes

0:59:33in the grand scheme of things.

0:59:34I'll try to do it as nice as I can,

0:59:35but I'm not losing sleep over that.

0:59:38The Now Playing screen, there's more variability--

0:59:40although you wouldn't know it based on looking at the apps

0:59:42out there right now-- but there's

0:59:44more variability in how that can be designed.

0:59:48And I think it matters more because maybe you're

0:59:52interacting with it for a split second

0:59:53and putting it back in your pocket.

0:59:55But what if you're jogging?

0:59:56Or what if you're in a car and you can't really

0:59:58look at it safely?

1:00:00There are situations like this where it matters

1:00:02how it's designed.

1:00:03It matters how it's laid out.

1:00:04It matters what's on there and what's not to a level

1:00:08that I think is more nuanced and more important and harder

1:00:10to design for than a list of episodes.

1:00:13That's what I meant.

1:00:14I think I understood what you meant

1:00:16And what I would have said to--

1:00:18my interpretation of it was that that's

1:00:20where you're concentrating your development

1:00:22effort for the 1.0 because you have to pick something.

1:00:26But the one kernel of truth that I think

1:00:29was in all the feedback, at least from my perspective, is

1:00:32when I think of podcast apps, which again I don't use--

1:00:35I'm going to listen on the iPod Shuffle--

1:00:37that the problem that I would like solved by them that is not

1:00:42solved adequately by most of the ones that I use

1:00:45is the particular way that I deal with deciding what I'm going to listen to.

1:00:50When I do it manually, it's this terrible process of using iTunes, and iTunes 11 has

1:00:54gotten even worse about this than me hunting around for the stupid things, dragging them

1:00:58onto my iPod shuffle.

1:01:01If I'm lucky, I can hunt around for all of them and then drag them all at once, but if

1:01:03I'm unlucky, I have to do it in two or three trips.

1:01:05Then I go over to the iPod shuffle, and all those things are down at the bottom, and then

1:01:08I manually drag them up into the order that I want and rearrange them, and then wait for

1:01:11it to sync again.

1:01:14I have an idea what I want to do. I'm like, "Okay, well today I'm going to listen to this,

1:01:17and then I want to listen to that when it comes out, and then I'm going to go on a binge

1:01:20and catch up with this podcast, and when I go on this car trip I want to do these things."

1:01:24That whole idea of managing what I'm going to listen to next is the thing that I just

1:01:29think is not solved by the current crop, certainly not by my stupid iPod Shuffle, and by the

1:01:33current crop of iOS podcast applications. Even though I would be spending most of my

1:01:39time on the now playing screen, the features that are important to me are the ones that

1:01:43It let me sort of set up my queue of like, you know, I don't know if you want to call

1:01:47like a Netflix queue or whatever.

1:01:48Like, here's what I'm going to listen to next.

1:01:49And that queue is just random, arbitrary.

1:01:52It's not like one podcast is higher priority than the other or like there's no automated

1:01:56way to do it.

1:01:57It's like I manually pick what I want to listen to in what order.

1:02:00And that's what I, if I had to pick another part of the UI for you to concentrate on,

1:02:06that would be the second one because yeah, most of the time we'll be on either the now

1:02:09playing screen or the controls on the sleep screen or whatever.

1:02:12to get to that screen at all, I have to sort of have my queue set up, you know what I mean?

1:02:17Makes sense.

1:02:17Yeah, and the other thing about this, I think, is that I'm assuming you're not telling us every

1:02:24single feature that you're going to have in the application, because why would you? You know what

1:02:26I mean? So that's something for people to keep in mind, is they don't realize that you're not going

1:02:30to reveal every single feature that this app will have or that you're planning in the future,

1:02:35because that's just not what you do when you're making an application, because why would you give

1:02:38your competitors the head start and never mind the features that are going to be in version 1.1

1:02:42in 2.0 and so on and so forth. I would keep that in mind.

1:02:45Oh yeah, I mean that's the other thing too. Everything I'm talking about is 1.0. As

1:02:51soon as I release it, there's hopefully going to be a lot of users. And those lot

1:02:56of users are going to give me feedback and are going to use it in ways I didn't expect.

1:03:00And I'm going to see how it works with full scale and how the server stuff works and how

1:03:06the structure of the app works, how navigation works. People are going to report problems

1:03:10or suggest improvements that I haven't thought of. And so all this planning and all this

1:03:17thinking about the design and making these decisions, that's all just for 1.0, and it

1:03:22could all change like a month after I release it.

1:03:24Yeah, think about Instapaper 1.0 and how it compared to, you know, the final version.

1:03:28Oh my god, I don't even want to think about Instapaper 1.0.

1:03:31So yeah, long way to go. So it's a marathon, not a sprint.

1:03:35Some people have nostalgia and positive memories and they look back fondly on the stuff they

1:03:41made in the past. I am not one of those people. I look back on stuff I made like three years

1:03:46ago and was like, "Oh, I'm embarrassed. I don't even want to think about it. I don't

1:03:51want to look at it. I'm just deeply embarrassed by it." Which is probably unhealthy. But

1:03:57I don't know. It keeps moving forward at least.

1:03:59Yeah. All right, we good? Yeah, I think so. Cool. Thanks a lot to our two sponsors this

1:04:07week, Squarespace and Audible, and we will see you next week after our Singleton trip.

1:04:12Indeed.

1:04:13Now the show is over They didn't even mean to begin

1:04:19'Cause it was accidental Oh, it was accidental

1:04:26John didn't do any research, Marco and Casey wouldn't let him

1:04:31'Cause it was accidental (accidental)

1:04:33It was accidental (accidental)

1:04:36And you can find the show notes at ATP.fm

1:04:41And if you're into Twitter, you can follow them

1:04:46@C-A-S-E-Y-L-I-S-S

1:04:50So that's Casey List M-A-R-C-O-A-R-M

1:04:55♪ N-T-M-A-R-C-O-R-M-N-S-I-R-A-C ♪

1:05:00♪ U-S-A-C-R-A-C-U-S-A ♪

1:05:02♪ It's accidental ♪

1:05:04♪ It's accidental ♪

1:05:06♪ They didn't mean to ♪

1:05:08♪ Accidental ♪

1:05:09♪ Accidental ♪

1:05:10♪ Tech podcast ♪

1:05:12♪ So long ♪

1:05:15- This is like the least we've ever gotten

1:05:17through our little notes document

1:05:18that Marco doesn't look at.

1:05:20- Yeah, I know, I know.

1:05:21- One item, this zero follow-up,

1:05:23And Casey put in the Agents of SHIELD item right before the show, and that's the only

1:05:27one we did.

1:05:29This is the first time I've actually gone the whole show without even looking at the

1:05:32document.

1:05:33It is?

1:05:34Yeah, exactly.

1:05:35Well, it doesn't matter.

1:05:36Two out of three, Casey, as long as we stay strong.

1:05:39We have to not compromise with the—don't negotiate with terrorists here, just because

1:05:44he doesn't want to use the document.

1:05:46We have the majority.

1:05:47Two out of three people use the document.

1:05:49The document is official.

1:05:50It's been placed.

1:05:52whether Marco looks at it or not.

1:05:55OK, so did you see that Bimmerpost put out a thing--

1:05:59this is accidental neutral--

1:06:00Bimmer?

1:06:02--Bimmer, whatever, Bimmerpost.

1:06:03Don't even talk about religious arguments.

1:06:05Don't even get into that.

1:06:06Oh, god.

1:06:07Yeah, seriously.

1:06:07Is that a real-- is there actually

1:06:08a controversy about that?

1:06:09There is.

1:06:10Oh, jinx it.

1:06:10Oh, well, do we both-- do we all three of us

1:06:13agree that it's Bimmer?

1:06:13Apparently Casey doesn't.

1:06:14No, it's because it's spelled B-I-M-M-E-R.

1:06:17I know, I know.

1:06:18I understand.

1:06:19Well, anyway--

1:06:20But the nickname for the car is Beemer, right?

1:06:23Yeah, well, that's the religious debate,

1:06:25is that the motorcycles are Bimmer or Beemer,

1:06:28and the car is Bimmer or Beemer.

1:06:32And I think the car is actually supposed to be Bimmer,

1:06:34if you talk to a zealot.

1:06:36And yeah, I don't like it at all.

1:06:38See, this is why I just avoid saying either of them.

1:06:40I don't want to have to deal with that.

1:06:42Porsche, Jaguar, yeah, it's a wrap.

1:06:45Subaru.

1:06:47No, no one says that.

1:06:48Wow.

1:06:49I see that.

1:06:50Now I can't even remember what the hell it was I was trying to talk about.

1:06:53Bimmer post. Controversy.

1:06:55Bimmer post.

1:06:56Uh...

1:06:57(door slams)