a16z Podcast - a16z Podcast: The Apple WWDC 2014 Deep Dive

Episode Date: June 3, 2014

Benedict Evans is a veteran of Apple’s big events and puts all the announcements and demos into one of three categories: 1) all the cool incremental improvements to the Apple operating systems; 2) the tent-pole features that Apple likes to build marketing campaigns around; 3) and finally, the fundamental strategic moves by Apple that serve to push the company ahead of the competition. With help from Andreessen Horowitz technical talent partner Dave Jagoda, this segment picks apart all three categories.

Welcome to the A16Z podcast. I'm here with Benedict Evans, who's just fought traffic to get back down here to the home office from WWDC. Well, what would do code as it now known as? And we are also joined by Dave Jagoda, who's on our technical talent team, and we will be leaning on his technical talent for part of this discussion. So Benedict, let's start. with you. You were just up there in Moscone Center in the scrum, and you've gone to plenty of

these events, and you have a way of sort of mapping them out. Explain that to us, and then let's just jump into it. So I tend to think about these kinds of events as splitting into three categories. So first of all, there's the stuff that is a nice incremental improvement if you're used to the platform, like something cool happens in the calendar, something cool happens in the way the browser works or something. And that's all, that's great. just kind of happens as the kind of the general evolution of the product. Right. And there's a bunch of stuff in the iOS,

and there'll be a bunch of stuff at Google I.O. in a couple of weeks. I think then there's two more categories. One is stuff that's kind of consumer-facing features that you can go out and build marketing campaigns around, sort of ten-pole features. And the other are fundamental strategic moves by the platform, and sometimes those are the same thing. And sometimes they're not.

Sometimes they're kind of distinct. And so looking at what Apple announced, I think there's a bunch of things that sort of fit into some or both of those categories. So they, and some of them are sort of in the first one. So they did a major revamp of the photos app so that now all of your photos are stored in the cloud. You don't even have to think about storage on your phone. Right. So you could call that one or you could call that two.

They might market it. They might not. More interesting from a kind of industry point of view is that both the wearable story and the home Internet of Things story turned out to be true. So Apple is using Bluetooth LA. and its local discovery stack and a sort of a central API story to store all the stuff

from your health-related wearables in a central data store that any app on your phone can access with permission and that any app on the phone has an incentive to provide data into as a permission. So today they announced just specifically on the health side

and also on the home. So health kit, yes, both for health and for home. So there's a stack for a set of stuff for wearable health things where you have an app on your phone that collects all the data from all the different wearable health things that you have, and that then provides that as a sort of an accessible data store to a Nike app or any other app that might want to use those.

And secondly, they've done something similar, although we haven't seen in so much detail, around managing internet connected devices in your home. So lock the garage door opener, the thermostat, and they put up a big slide with a list of 30 different companies' logos, except, of course, for Nest and Google. right and so those are both I mean they're both a symptom of what I've talked about before which is that you see Apple because it controls the whole widget moving down the stack

and leveraging Bluetooth Ellie in particular but leveraging the fact that it knows exactly what's on the device and exactly how it's going to work so that stuff can actually be done in a predictable way the other thing that they've done on that in that line incidentally is add an API to the fingerprint scanner and they gave a number that I think it was 83% of all iPhone 5S users use the fingerprint scanner and so they're now going to have a kind of a secure API. So any app on your phone can ask the question, is this Michael, yes, no, and get a response. Which could be used for banks, could be used for anything.

It's just as an alternative to entering your password. So even opening my door at home, right? Yeah, potentially, yes. Now, it's not interestingly an identity platform because a third-party app can't use that data to see that you're Michael. Oh, so it doesn't see my actual fingerprint. No, it doesn't see your fingerprint. It doesn't know that you're Michael.

All it does is sort of, ask, can you please verify that this fingerprint relates to this login, yes or no. So it's not as though you could get a new phone, install an entirely new social app, scan your fingerprint in and therefore get your profile ID. Because the third party doesn't get that information. They just get a yes, no response from the API. It's like iTunes. You know, the app doesn't get an app doesn't get your iTunes password.

It just gets the iTunes store saying, yes, you're allowed to do that. I see. So there's those things which are sort of user face. and you can see resulting in a whole bunch of interesting innovation. And then there's two other things. One of them is what they call extensions, which is a way for third-party apps to hook themselves much more deeply into the system, much more controlled than on Android.

And so there's an obvious response, which all Android's always had that. And the answer was, Android never tried to stop people from doing that, which is a sort of swings aroundabouts point here. So Apple has given a bunch of quite controlled ways that a third-party app can offer services within the system. So you can send a photo to any app at 20 app now. You can send, and any app also can add itself to a menu to provide services.

So the example they give is you're on a web page, you press the services button, and then you've got Bing translate there as an option. So instead of you having to then open the Bing app and then load the page into it or open the Google app and load the page into it. And so the funny thing here is, of course, this is basically giving Google and Facebook a whole list of ways

to get their hooks deeper into Iowa. I was going to say, so this is a good thing for Google and Facebook? Well, you know, there is going to be a share this photo to Facebook button and a share this photo to Instagram button and a, you know, send this link via WhatsApp button and a whole bunch of other things that, of course, are very tightly under the user's control in a way that they aren't necessarily on Android. But are there. And what Apple is also doing is they're allowing third-party widgets onto the home screen

and they're allowing a few other bits and pieces. So, for example, there's a story around photo editing. So third-party apps can pull, a photo app can invoke another photo app in order to edit. And so, you know, again, it's a bit like multitasking that Apple is doing something that was on Android like three years earlier. And they're trying to do it without, or picking a different set of trade-offs. So it's more controlled, but then you don't have the security implications or battery implications. And Dave, chime in here too, but, you know, Apple is famous for having that control. And their argument for a long time had been that, like, without that control, we could.

couldn't make the experience as, you know, wonderful as we liked it to be at Apple. How is it that they, do you sort of describe this a little bit as a loosening of that control and why are they able to do it and still have the experience that they want? It seems like some of the things that Benedict was talking about, I think they're probably share buttons for already, so they're already sort of cooperating or facilitating potential competitors to do stuff. So this just opens it up to other ones, unforeseen competitors. Yeah, I mean, I think there's a couple. of bits here, I think one of my impressions from the event was that Apple has had, I mean, any

company that isn't doing something you think they should do, it's either they've decided not to do it, or they haven't got the resource to do it, or it's on the roadmap, but it's not ready yet. And it feels like what happened to Apple today was there were like five or six major projects all hit the end of their waterfall today. So there was a whole set of questions around what is the sharing in the passing of information between apps look like, and that wasn't finished last year. Right. And you can sort of see looking at iOS 7, there's like placeholders for this stuff that

aren't, that wouldn't be there if they hadn't been planning to do this. And, you know, an API for the fingerprint scanner, well, they'd only launched a fingerprint scanner last year and they wanted to make sure they were getting it, going to get it right. So a bunch of this is just timing. I would suspect, this is kind of more Dave's area than me, that a bunch of this is also kind of Moore's law. Right. You know, the more CPU you've got, the more things you can put inside of sandbox instead

of letting them run on bare metal. Right. And so quite a lot of it may be to do with that. that they just feel like they can control it and have a good experience. Whereas, you know, if you wanted to happen at all five years ago, you'd have to have it completely uncontrolled.

Yeah, which is sort of the Android experience, right? You just go at it and let's see how it works. And then obviously the final thing is the new programming language called Swift. And I think John Gruber said on Twitter during the event that basically a third of the room was sitting going completely quiet,

saying I have absolutely no idea what these people are talking about, which is the category I was in. Oh, new programming language. And the other two-thirds of the room, room are going berserk. Yeah, and the other, and two, the other two is the room and going, wow, wow, wow. We were joking about curly braces and Dave,

this is where we need your help. Swift, as it was described, as objective C, without the objective or without the C, I can't remember which, but tell us from, and granted, they just announced today and it was a surprise to everyone. What it starts to look like and how we see it rolling out. Sure, you know, I've obviously, I've been looking at it for a little a while. I see that, to me, it seems like they've tried to take the very interesting features

from other languages and incorporate them into Apple's development environment. So I, just looking through the documents looked at before, there's features of Python, Ruby, CoffeeScript, JavaScript, Go, and even hints of languages like Java and Lisp that they've incorporated into it, which is sort of, you know, amazing to have touched so many different interesting features. We talked about you guys like teasing about curly braces. Another one also maybe in the punctuation category would be semicolins. You know, traditionally, Objective C has to have semicolon the end of every line. Swift doesn't seem to have that.

Again, you might say who cares, but programmers argue about this stuff to no end. So it's interesting. They've added that in another, I think, really interesting productivity feature. I mean, this is clearly definitely in Benedict's third category of fundamental sort of things is the so-called REPL or Read-Evaluate Print Loop, which is really, important when you're developing code instead having to like make your change compile it build it copied overseas to that work or not you can just see what's the effect immediately as you're doing which helps you know trim that development cycle so the example on the keynote was

he was building a game and there was a blimp and you could actually see the changes you were making to the to the app and to the blimp as it right as you were doing it exactly and this is the thing that people you know that love languages like python or ruby love to have they just get to see it as they're building it and also it's useful to try something out without changing things around too much. So one other observation I had about it was it looks very much to me as if this was built, you know, it's a programming language that was built by programmers, obviously, but for programmers.

I think sometimes people try to think, what could we do for our users that would, you know, well, we sort of know best what our users want. Let's build that for them. But it felt to me, at least like they're building this for themselves as well to help Apple's own efforts as opposed to like, here's a kit, you go use this. We'll use the real stuff.

You guys can use this sort of dumbed down. unlimited thing, and it doesn't feel like that at all. They're doing their own dog food. Exactly, totally. Yeah, this is pretty exciting. I know you guys are clearly thinking I'm in that second, two-thirds camp, but pretty exciting. To your point, Benedict, like, you know, the strategic point of it is why.

Like, what does this do for developers? So they already had their own program development language already. I mean, there was a chart that I was looking at where they've got a two-by-two matrix of developer productivity versus app performance. And so you've got, you know, Ruby and Python and JavaScript and so on with high productivity and low performance. and obviously C, and objective C and so on in the other corner of performance and their productivity, and they're claiming that this is in the top right.

That's to see, you've got both the productivity and the performance. Yeah, I think there's something to be said for that. I think another thing, too, is people do, you know, these are developers have preferences, of course, about what they're writing in. So by trying to acknowledge the really good, nice developer features that exist in these other languages by incorporating those, I think it helps win people over. So is this something that is going to be constrained in the Apple world? It just makes it easier than ever for folks to write apps for iOS or OSX.

But are we going to start to see it outside the Apple world, do you think? You asked me a question earlier, and even I've had time to think about it. I don't know. It seems like this is very dependent on Apple's systems, so I don't know if they could easily take it outside the world or even if they'd want to. But it would appeal, certainly would appeal to me if they could let us use this environment for other things other than OSX and iOS. And one suspicion is that once people really pick over it,

well, there'll be a bunch of things around how this fits into Apple's sort of 10, 50 and a roadmap of, you know, because Objective C has been around, what, 15 years. Probably more than that, yeah. And so there may be a bunch of sort of things that it lets them do that they couldn't do before. Right. Yep. And the other interesting developer story is ICloud Drive. I mean, we've all heard the story of Steve Jobs meeting Drew Houston and saying that's not a product, that's a feature. Right.

And, you know, now Apple, after going through a period where they wanted to get away from the file system, It's kind of ironic because Apple basically invented the idea of having a visible, tangible file system where you touch things and move files around. And then the last couple of years, they've tried to get away with that. And iOS obviously didn't have a file system at all because that was going to be much simpler and much easier for people to understand. And up to a point it was, but we all sort of end up. You either use Dropbox or, you know, you start putting PDFs into programs that aren't into one file

and Word documents into another program. Right. And you think, well, you know, it's kind of nice to have just a place where I could just put stuff. like that PDF and those photos and that word document and that spreadsheet and that thing that I need to have. And now they've done it. So now you have iCloud drive and it works, well, I hate to say it works like Dropbox. You know, it's a folder you put stuff in and it's on all of your devices.

And as an extension to that, therefore Apple is now sort of broken apart the file sandbox a bit because it now means that apps can access documents created in other apps, which is the way it works on computers, on desktop computers, as opposed to the way it works on a traditional. you worked on iOS. So it feels like Apple are kind of backing off a little bit from the absolutism of you. There should not be a user accessible file system at all. Right. And what does that say about sort of Mac and OSX versus the iOS world? I mean, how are the sort of relative importance of those things, you know, being described in what was announced today? Well, we haven't

talked about MacOS at all yet, actually. Maybe that's one clue, right? And, you know, Apple spent the first hour talking about macOS and you know from the kind of the broader technology industry um they've said they've got 80 million and install base of 80 million of which 40 million are on mavericks and that's like a higher percentage in windows eight but there's a lot more windows eight devices out there than there are um max because there's you know something like one and a half billion windows PCs out there right so um you know for apple the mac matters and for developers the mac matters but for and for us because we all use max the mac matters but in the great wide world that of the enormous transformation of the technology industry and the media industry and the whole

industry in general by mobile, the Mac is kind of irrelevant. So, yes, there's a bunch of nice stuff in there, and they've basically revised Mac OS to clean up all the kind of croft that had accrued over the last 10, 15 years in much the same way that they did with iOS last year, with iOS 7. And, you know, they've tied it and they've cleaned and they've systematized and they've made it all kind of make sense again. And they've done a couple of really nice things like, you know, you can now use airdrop to send files wirelessly from your phone to your Mac and or vice versa.

And the kind of the cunning thing is, and this is maybe also a strategic move, is they have this thing called convergence where, you know, if you have an app on your phone, if you have a webpage open on your Mac or an I work document open on your Mac, you have a little icon on your phone on your iPad that relates to that. So you tap on that icon on your iPad and your iPad will open up with that same web page or that same document. and I mean the kind of the really tangible use case is you can highlight a phone number on a web page and select dial and your phone will dial it and we've all been in that room where you're kind of you're looking at the phone number on the web page and dialing it with your hand right going four one five eight two no eight three and they solve that you also can use your phone

your Mac as caught for to dial the phone number to receive a call to use it as a speaker phone get caller ID. You can send not just I message, but SMS from your Mac over your phone. So you're basically turning these things into devices that all work together and kind of understand that they're in the same room and made by the same company, which is the sort of thing that, it's one of those things that feels like it should have done that 15 years. The point is for all of it, for all of this stuff, it's about driving the devices. Right. And you're coming back to the point I made earlier about, you know, Apple are giving Google and Facebook all sorts of hooks into that platform. Well, that's fine, but they're doing it on a $600 device that people have to buy.

Right. So you've said this before. that Apple is a hardware company. Like, their business is not as complicated as people think. And how does this help them as a hardware company, what you saw today? So, okay, so let me give you a specific answer and a general answer. I think the specific answer is that Apple is using cloud and to some extent Bluetooth LE to make rich native apps do stuff that you feel like they ought to have been doing forever. So you take a photo on your iPhone. It's on your Mac. It's on your iPad. You edit it. It's on

your Mac. It's on your iPad. You know, you change an address. You add a phone number. Anything that you're doing on one device is on the other device. And it's all being done through the cloud. But they don't talk about the cloud. So it's like the, it's sort of the mirror image of Google where the cloud, you know, for Google, it's like Mark Andreessen's old phrase about Windows becoming a set of badly debugged device drivers. And for Google, any physical object in a sense, it's like a set of imperfect device drivers that you have to go through to get to the pure glory of the cloud.

Right, right. Whereas for Apple, it's completely the other way. And, you know, also for Apple, the device hardware should disappear, which is the ideal of the Chromebook. The hardware and the software should just disappear, and it shall be in the cloud. And for Apple, it's completely the other way around.

So why the hell should you have to know that it's sitting on a server in rural Oregon or something? You know, it should just take the photo, and the photo should just be there. And you shouldn't have to worry about you've taken 35 gig of photos and you've only got a 16 gig phone, the phone. They're just there. You don't have to think, okay, they're being stored on my Google Plus account, or my ICloud account. It's just there. And that's a very different approach to Google.

This is reminding me a little bit of Sun, which I would say also a hardware company, where they realize Java might help their hardware business along, but it feels like they didn't take it far enough. Too much of it was left to the developer, left to people to figure how to take advantage of the hardware. And I think Apple's shown you can have good or hardware. I would agree they are a hardware company. And hey, look, it looks great together. You don't need to worry about all those gory details.

It'll basically work. So I think that's smart. That helps cement their hardware position. I think, I mean, to the broader point here is Apple's kind of got a lock on the high end of the phone business. They sell a phones and an average of $600 in an industry that averages $100. average is about 180 and Android average is about 250 and in the high-end space they have as it might be two-thirds to three-quarters of the market and Samsung has most of the rest and a certain number of people switch every year and rather more switch from Android to iPhone and switch back so

there's kind of a ratchet effect going on there that Apple is slowly gaining share in that space and in the US they've got half the market because the subsidy environment means that there's no cheap Androids, even the iPhone 4S is basically free in the U.S. market. So they've got that lock on that high-end market. And what all of this does collectively is as every kind of, is the kind of the, as the same as what happens with an iPhone release is, once a year Apple jump forward and set a new bar. And for the next six to nine months, Android and the Android OEMs catch up. And then for the next three to six months, they're ahead.

Right. And then Apple jumps forward again. there's this kind of zigzag, you know, there's a period of three to six months where the best phone on the market is not an iPhone, and Android does loads of cool stuff that Apple doesn't do, and then it flips back again. But what this doesn't do is change that dynamic. You know, Apple continues to do, amongst the best hardware in the market, and one of the two best operating systems on the market, depending on your personal taste and your personal preferences, what priorities you have. But they make it at $600. And there's a very large number of people in the

world who like a smartphone and actually quite like to do apps and aren't spending $600 on a phone. Now, in the US, the only people buy a lot, and the great majority of, let me phrase it slightly differently, in the US, because the price premium is so small, most of the people who aren't choosing iPhone are doing it because they don't terribly care very much. In the rest of the world, there's an awful lot of people who care quite a lot and can't afford an iPhone. And the only thing that really changes that dynamic is a phone alert price. It's a phone app for the sake of argument $300, at which point you can go and buy a really, really nice product with a really good build quality and a really good screen, nice performance is just not a $600 phone.

You know, you're buying a, you know, you're buying a Ford and not a BMW. And, you know, nobody sits in a Ford and says, gee, this is a crappy car. It's just not a BMW. Right. And that's the kind of the driving dynamic of the phone business for the last two, three, four years. There's a certain number of people who buy premium products and there's a certain number of people who don't.

And amongst the people who buy premium products, all this stuff that Apple does drives their purchase decision. And, you know, as it might be two-thirds, three-quarters of them, prefer the choices that Apple makes to the choices that Google makes and choices that Samsung make. But there's a whole other conversation going on out there that Apple is not really engaging with yet. And so I think that's the kind of the super-high-level point

that Apple is just kind of maintaining the status quo. They're not actually changing the conversation. Well, the rest of the conversation will be happening pretty soon with Google's version of WWC-I-O. Yeah, I mean, and that comes back to my point, that as Apple pushes stuff down into the hardware and the integration and the 64-bit chip in the fingerprint scanner, and, you know, almost, how can I put it?

There's a thing I said in another context that a lot of what's going on now is creating dumb glass. So a TV is dumb glass or a car that display in a car is dumb glass and the smart lives somewhere else. But another part of it is turning the cloud into dumb cloud, dumb storage you know say what Apple is

what Google is doing is they say well the device is dumb glass and all the cool stuff happening in the cloud and what Apple is doing

is saying no no no no no the cloud is just dumb storage and yeah I mean it's a bit more too than that but you know basically the cloud is storage

you don't have to think about that you've got this fantastic device that does all these wonderful things clearly there's there's two worlds

that want it you know dumb storage and those that want you know just beautiful devices and then there's the Android fans

so we'll get that side of the story at I.O. And I know, Benedict, you're going to be there and Dave will bring you back. But I want to thank both of you for breaking down WWDC. And that's the A16Z podcast. Cool. Thank you.

Thank you.