Matias Duarte is the Director of Android User Experience at Google, which means he's the artist who pretties up the green robot's gears. We got some one-on-one time with him at Google I/O, and he opened up about the Nexus 7, Jelly Bean, and why we shouldn't be so huffy about Android fragmentation.
Giz: What is your overarching vision for where Android is heading in terms of UI and design?
MD: Well, I have a particular style in the way that I design things. Fundamentally, I believe that people want to touch and interact with things like they do with things in the real world, and that every time of interaction experience can be a delightful experience as well as an easy experience. A lot of people see those two things as being at odds, but I really like to try to find design solutions that blend the two.
I like to say that things are more fun than buttons. And you can solve most design problems by turning them into tangible, physical objects, without necessarily making them skeuomorphic copies of real world objects, but still tapping into those parts of the human brain and the human heart that are like, "I've got hands. I was built for picking up and moving things and swiping things around." So I want to transform the types of interactions we have with computers that are today really all about hunting and pecking and picking and menus, into an experience that is a much more gestural, physical, emotional experience. And transform the entire OS like that.
We've made huge strides. We started in Ice Cream Sandwich [ed. Android 4.0], I think, to kind of set that vision to make an operation system that was really beautiful, and easy to use, and powerful. In Jelly Bean [ed. Android 4.1] we've taken one tiny next step, but I'm very happy with the improvements we've made in performance because that helps create that entire illusion.
Giz: Ice Cream Sandwich was very well received; what were the first changes you wanted to make in the update to Jelly Bean?
MD: Well, one of the biggest things was Project Butter, where we declared a war on jankiness and crustiness and graininess, because that's so important. If you really want to create something that's beautiful and emotionally engaging and feels like a real thing, you need to be fast enough. Right? You just need to get to the point where you're fooling the eye and fooling the finger. So that was one of the top priorities right from the very beginning.
And then of course we wanted to continue to improve all of the other aspects of the OS. Things that seem small, like the expanded notifications, have a big impact because now all of a sudden every application can reach out and talk to you in a a way that is less intrusive. When I see I have mail now, I know who's actually mailing me. Can I deal with it later, or do I actually have to switch apps? There's a big cost to switching apps. We make multitasking really easy on Android, but still it's much easier to just peek open that shade, be like, "Nah, I don't need to deal with that," and swipe it closed again.
Giz: The version of Jelly Bean that we have right now is a developer preview, right?
MD: Yes, it's a developer preview so it still has a few bugs that we know about, and obviously a few that we don't know about, but when the devices actually begin shipping in a few weeks you'll see a lot of additional polish go in.
Giz: Is there anything that wasn't announced that's going to be going in when it launches?
MD: Surprise features?
MD: Well, if we tell you now then it wouldn't be a surprise.
Giz: Fair enough. Going back to Ice Cream Sandwich for a minute. It was a significant leap forward for Android in terms of intuitiveness and design. Does it kind of drive you crazy when you look at the data on how many devices are running it and it's such a small percentage?
MD: Oh no, I think we're actually doing pretty good in terms of adoption. It's one of those things where any time you release a new OS there's a gradual adoption curve at first and then it picks up pretty quickly. Like right now, everybody's running Gingerbread [ed. Android 2.3]. Well Gingerbread was the first version I released of Android—it came out a few months after I arrived (at Google), but at the time people were saying the same thing. "Oh my gosh, Gingerbread is only out there on a fraction of devices!" Well yes, you release it, and then it takes some time to roll out.
One of the cool things about the way that we're able to release capabilities is that we don't just release capabilities with the OS updates. So, yes, a lot of awesome features like Google Now are tied to Jelly Bean, but there's other things that we roll out through our applications. All of new the improvements and content types to Google Play, you can still be running Ice Cream Sandwich. In fact many of them go back to Gingerbread. You can get all the latest content types and you don't need to be on the latest OS version. So it's not one monolithic thing.
Giz: You mentioned Google Now, which is one of the banner features of Jelly Bean. Would you talk a little more about what it is?
MD: One of the most exciting things for me about Google Now is that it's not just a product—it's not just a think about flights or a thing about traffic—it really is a platform for Google. So the same way that search was a platform for Google to deliver information, and we started by giving you just ten blue links, and then we were like well, we'll give you weather in one box, and translation in one box, and we'll give you this knowledge sidebar, image search. All of a sudden, all of these things show up in Google as part of search. Google Now is the same kind of platform.
What you see today are the first use cases that Google rolls out, but it's going to grow in that same organic, magical way. Day after day, new things are going to start showing up. So it gets better not just from starting to understand you and understand your patterns, but also because Google is building better and better cloud in the background, with even more and more capabilities.
Giz: It seems like it's pulling a lot of information from other apps, but does it—or is it going to—have the capability to push information to other apps. For example, would it be able to add an appointment to Google Calendar through a voice command?
MD: Well we have some integration for that already. You can ask for navigation, you can place a phone call through Google Now, you can even set reminders. It doesn't integrate fully with Calendar events yet, but those capabilities are built-in. Those APIs exist. Like I said, the idea for Google Now is that it's a platform, and all of your Google services are going to get better and better over time.
Giz: Are there plans to bring it to the desktop side through either Chrome or Google+ integration?
MD: Well, we don't have any announcements about that yet.
Giz: Going back to the design of Android as a whole, most hardware manufacturers put a skin on top of Android, sometimes to its detriment. In ICS any pre-installed app can be disabled. Have you toyed with the idea of requiring that third-party skins would be disable-able, so the consumer could choose to run stock Android if they wanted to?
MD: The way Android is architected right now, that would be really difficult to do, but it's not something that's inconceivable at some point in the future. We realize that the customizability of these skins, while some of them are bad, some of them are very good, and some of them are things the customers actually prefer to stock Android or to other skins. That's one of the things that makes the Android ecosystem work: it provides so much choice. So we would never do anything that would discourage that choice or prevent OEMs from having that kind of flexibility.
With Ice Cream Sandwich one of the things that we did in order to enable more consistency and more ease of developer portability, was it was required that from Ice Cream Sandwich onwards, the core widgets and framework themes were always included, so that developers can target them and have their application look, feel, and work more consistently no matter what skin has been put on it. So we like solutions like that, that put control in the hands of developers or users without taking away any options.
I think there are some terrific skins that've been done out there. I'm really looking forward to the new collaboration between Sharp and Frog Design, and I really, really want more Android OEMs to do that kind of innovation—where it's not just minor tweaks, but really explore the potential of Android. We're building one particular vision for how we think you should have a mobile computer and how we think you should organize your life in a mobile space, but there's millions of different visions out there.
One of things that really frustrates me about desktop computers is that basically everybody got locked into this Xerox Parc world of windows and files and folders, and then it was 20 years and nothing changed! There was really no innovation at all. It wasn't until this mobile era where things are starting to change and people are starting to think about different types of experiences on the desktop like you see with Windows or some of the stuff that Mac OS is doing. I think it would be terrible if everything were locked into one pattern again and operating systems didn't change for another 20 years. So I really want to see a greater variety of skins out there in the Android ecosystem, because I think that's a positive force that keeps that kind of ossification from happening.
Giz: Android is open by its very nature. Do you ever feel paradoxically limited by its openness?
MD: Oh yeah, it's much, much harder to work in an open ecosystem. When I worked on other platforms before where we had this completely vertically integrated type of development environment, you can just simply say, "I don't care about that problem, because I'm not going to build that hardware." In Android you never have that option. When you design something, and when you build it, it has to be robust enough to handle anything that an OEM can throw at it. That takes more time, takes more effort, and it's a harder problem. That's why I like it. That's actually a really interesting challenge.
It's like this whole question of app-sizes: how should an application developer write an app for different sized screens? Well, you could pretend that there are only two screen sizes in the entire world, or that there are only two screen sizes that anybody would want, but that's incredibly limiting. It would be like if on your desktop you could only have two different sizes of windows instead of rescaling them to be exactly the size that you want. So it's a much more challenging design problem, but it's one that I think is really worth doing, because that's what it's going to take to really deliver the next generation of computing.
Giz: If I gave you five bucks would you considering requiring all phone manufacturers to have a physical button for the camera?
MD: (laughs) No, absolutely not. That's actually one of the things that I feel really strongly about: the idea that we should require as little as possible, because I want to have as much innovation as possible out there. For example, two years ago there was a Chinese company that was able to release an Android device that didn't have any buttons at all. Not just on-screen soft-key buttons like we have in Ice Cream Sandwich and now Jelly Bean, not just capacitive buttons, not just not-physical buttons, but no buttons at all! And it supported all of the Android functionality—homescreen, back, etc.—by using gestures, like of like what we did with WebOS. And it was great, because that was compatible with Android, because our requirements are so loose that people can innovate that way. There are other things that people could do, and I don't want to do anything that would stifle that innovation. I want someone to build an Android that's all gestural, I want someone to build an Android that's got no physical buttons, I want someone to build an Android that you run by braille. And I don't want to have to treat those things as exceptions. They're all part of a family.
Giz: Anything else you want everyone to know?
MD: Yeah, one of the things that I spend time evangelizing is the kind of thinking and effort that went into Nexus 7, the design that went into this product. It's not just about making an affordable tablet for Google; I think it's a really interesting class of tablet that people have sort of overlooked. This product was really born out of my frustration with the idea that tablets were done. Everybody thought, "Okay, we've made ten-inch tablets, and they are the answer to media consumption and entertainment, and we're done." And I thought that couldn't possibly be right, because they're not the answer to two types of entertainment that are really important to me: books and games.
When I read a book, I wanna be able to have a cup of coffe. I wanna be able to hold onto the strap in a bus, right? So one-handed usage is really important. You can't do that with a ten-inch tablet—it's ginormous. When I play a game I want something that's about the size of a portable gaming system. I don't want something huge, again, where I'm going to get tired out. The Nexus 7 tablet, you see, has very similar dimensions to a portable gaming system. So this is a class of tablet that suits particular uses cases that are really underserved. We wanted to show that you could build a tablet in that space and it could be, not a wishy-washy cut-rate tablet, or wanting to be a tablet, but a truly complete tablet experience at a size that changes how you interact with it.
Everything about this design is optimized for that size. The curve of this device has been very carefully crafted. Yes it's rounded on the back but it's not sharp on the front. We have this crisp chamfer around there that effectively rounds off the device so it's very comfortable to hold one handed. We've given it these asymmetrical bezels here on the sides, so that when you're holding it in landscape you have more grip because it sticks out longer and you have more weight to counter-balance. And the bezels are great when you're playing games because they give you a place to rest your thumbs without obscuring any part of the screen. We carefully designed the balance of how the bezels work with the screen so that there's always a stronger dark area at the bottoms, just like the matte on a print. Like if you were going to frame some artwork, so it has a perfect visual balance in both dimensions.
All of these things come together in my mind to make Nexus 7 a really new class, a really interesting and exciting class of tablet for entertainment, for gaming, and for all of the digital communication that we do.