Interview With Mark Zuckerberg On The "Facebook Phone"

Over the weekend I wrote an article about Facebook’s mobile plans, particularly that they are working on a branded mobile phone. Facebook said the story was innacurate, despite CNET, Alley Insider and others writing independent stories confirming that Facebook was working on a phone. We responded aggressively.

Facebook reached out to us on Monday to clear the air, and invited us to Facebook HQ in Palo Alto, California to speak to CEO Mark Zuckerberg directly about Facebook’s mobile plans. Jason Kincaid and I spoke with Mark for about half an hour. The transcription of that interview is below.

For now this is just the raw footage, so to speak. We’ll parse this in a later post. There are some really fascinating things in this interview though.

Update — Our posts based on the interview:


Mark Zuckerberg: Before we get into this, I want to apologize for the miscommunication. I think our folks were trying to reach out to you and weren’t able to reach you before we issued our statement.

At the end of the day, when people say “building a phone” they actually can mean very different things. Internally, the way we talk about our strategy, it’s like the opposite of that. Our whole strategy is not to build any specific device or integration or anything like that. Because we’re not trying to compete with Apple or the Droid or any other hardware manufacturer for that matter.

Our strategy is very horizontal. We’re trying to build a social layer for everything. Basically we’re trying to make it so that every app everywhere can be social whether it’s on the web, or mobile, or other devices. So inherently our whole approach has to be a breadth-first approach rather than a depth-first one. And we work on all of these different things at the same time, so I’m sure whatever leak you got was probably accurate for whatever the person said. But it was probably just one part of what we are doing. Anyhow. I just wanted to give that context.

Michael Arrington: Ok. So let’s talk about how you see, what is Facebook’s role in the mobile world over the long run?

Mark Zuckerberg: So I guess, we view it primarily as a platform. Our role is to be a platform for making all of these apps more social, and it’s kind of an extension of what we see happening on the web, with the exception of mobile, which I think will be even more important than the web in a few years – maybe even sooner. But the basic thing that we’ve found from building social apps and this platform ourselves is that almost any experience or app can be better if it’s social and it has your friends with you. And we just expect there to be really tremendous disruption over the next five years.

We’ve seen this with a bunch of the apps that we’ve built. Stuff like photos and events and groups — we’ve built pretty basic versions of those apps to start but they ended up being so much more used because of their social integrations.On Platform we’re now really starting to see that with games through the four big gaming companies: Zynga, Playdom, Playfish and Crowdstar, andit’s really interesting looking at Zynga and traditional gaming companies. Zynga is just so much more efficient as a company right now. That’s what disruption looks like, when your model is changing.

It seems like games are often an early indicator of a platform spreading to other verticals. You kind of saw that with the iPad or iPhone, probably more iPhone than iPad. Even early PCs. So our view is that over the next few years we should expect in all of these different industries for there to be a lot of disruption…by either the incumbent in this space or some new entrepreneur coming in and building a social version of whatever it is, whether it’s commerce or media. It could be a number of these different verticals, and Mobile will really just be an extension of that and will eventually get to a larger scale than the web. The web is only at one and a half billion people whereas everyone is going to have a phone and all the phones are going to be smartphones. So our strategy is that we want to go wherever people are building apps so we can make all of those apps social if they want that.

Michael Arrington: With the tools that you’ve given them on the web. Tools like Connect for example..?

Mark Zuckerberg: I think it’s different in different places. For example, take Instant Personalization. Our goal is to make it so there’s as little friction as possible to having a social experience. So you go to some apps, take Rotten Tomatoes, which we just launched last week. If people had to click this blue button to Connect, then some percent of them would, but it would be the minority because you don’t know exactly what you’re going to get before you click it.If you had to put up some modal dialog then that would be crazy from a UX perspective. But the fact that they can do that instant integration for the users that want it means that everyone has a good experience as soon as they get there.

On phones we can actually do something better. We can do a single sign-on if we do a good integration with a phone, rather than just doing something where you go to an app and it’s automatically social or having to sign into each app individually. Those are the two options on the web. Why not for mobile? Just make it so that you log into your phone once, and then everything that you do on your phone is social.

Michael Arrington: You’re turning on a layer…

Mark Zuckerberg: That’s what we’re trying to do. The reason I just gave that example is that some things, like the implementation is different on mobile.

It’s different on mobile than it would be on the web simply because it is not really possible. I guess maybe Google or Microsoft could have you log into the browser, but we can’t because we don’t build a browser – but, that is the basic strategy.

And then, we invest differently in different platforms depending on how big they are, and how many users are there. So, iPhone is the one we’re investing in the most now, and Android increasingly. If Windows Phone 7 takes off, then I’m sure we’ll put resources on that. We invest at all different layers of the stack too, so we have people who are working on the lowest common denominator HTML5 stuff that works across all systems. So maybe we’re not building a lot of specific stuff for RIM and Blackberry, but the HTML5 stuff that we’re doing will work there.

And maybe we’ll build specific apps for iPhone and Android. And then, for something that is as important as iPhone or Android, we’ll also build integration into the operating system. So for iPhone, we built in contact syncing, and for Android we integrated and did contact syncing pretty seamlessly. The question is – what could we do if we also started hacking at a deeper level, and that is a lot of the stuff that we’re thinking about.

And then the question is: different people come to us with different ideas all the time, and we mentioned the example of the INQ phone in the past, and I think you appropriately said that it isn’t some massive big thing, but it is cool and actually a lot of people bought it. But, people might want to take different cuts of different operating systems, or build different feature phones and integrate Facebook in different ways. The mobile market is just so fragmented that there is a lot of experimentation that can happen now too, so we’re figuring out exactly what the optimum level of the stack is to be at. The reality is it will probably be different things to different phones. For some devices that we’re not going to do a lot of specific work for, it will be HTML5.

For platforms that are really important, but are hard to penetrate, like iPhone, we’ll just do as much as we can. For Android, we can customize it a bit more. Other folks are going to want to work with us on specific things. But, our goal is not to build a phone that competes with the iPhone or anything like that.

And this is the thing that was potentially really damaging about having that meme our there is: if all of the people that are our partners, who are the main people that we’re trying to work with to make everything social, think that we’re trying to compete with them, that makes them not want to work with us.

So, it is really important for us that people understand what the strategy is and that the real approach is to make everything social, not to build a vertical approach.

Michael Arrington: So, are you working on a ground-up operating system for mobile phones today?

Mark Zuckerberg: An Operating System? No.

When you get into the definition of what building a phone is, there are all these different things, which I think is the grey area, where, actually, sometimes even people internally refer to stuff as ‘Facebook phones’.

There’s the Apple approach of really designing all the hardware – I don’t think that they manufacture it themselves, but they probably work very closely with Foxconn – but they do have chip design and all of that in-house, I’ll bet we’ll never do anything like that. That is different from what we do.

And just from the 6 years of experience we’ve had building Facebook, which is a pretty complex system, you can’t just wake up one day and decide you want to do this stuff. Even building a whole operating system is, like years and years of work. So that is what Google does.

And Google has been in a bunch of different places. Nexus One, I think people would say they built that phone a bit more, but the Droid, I don’t think you could consider that is a phone that Google built, but some people would. I think what you’ll find is that all of this stuff that we do, is lighter than all of those things.

I mean, who knows, 10 years down the road, maybe we’ll build our own operating system or something, but who knows. That is more history than we’ve had so far with the company, so it is really hard to predict that far out. But for now, I think, everything is going to be shades of integration, rather than starting from the ground up and building a whole system.

Jason Kincaid: So, it sounds like a customized version of Android seems like the best thing, the deepest integration you could do without building your own operating system.

Mark Zuckerberg: Yeah, and I mean…

Michael Arrington: Is that putting words in your mouth?

Mark Zuckerberg: No, I mean, we’ve looked a bunch of different technical things.

I know if we were going to build an operating system, then we wouldn’t have anything to talk about for 4 years. And I know we didn’t start 4 years ago, so I know we’re nowhere near anything on that. What I can say generally is that our goal is not to build an operating system from scratch, or else not to design hardware from scratch. Our goal is to make it so that we can design the best integrations in the widest variety of phones.

One thing that I think is really important — that I think is context for this, is that I generally think that most other companies now are undervaluing how important social integration is. So even the companies that are starting to come around to thinking, ‘oh maybe we should do some social stuff’, I still think a lot of them are only thinking about it on a surface layer, where it’s like “OK, I have my product, maybe I’ll add two or three social features and we’ll check that box”. That’s not what social is.

Social – you have to design it in from the ground up. These experiences, like what Zynga is doing or what a company like Quora is doing, I think that they have just a really good social integration. They’ve designed their whole product around the idea that your friends will be here with you. Everyone has a real identity for themselves. And those are fundamental building blocks. Now, I don’t know how long it’s going to take to get the mobile environments that you see today to a state where you can build really robust social applications on top of it. So that’s the biggest driving force for us — to try to work with these folks and see how deep we can get on our own to make sure that we can build that plumbing. Our goal is to make it exist.

Michael Arrington: The folks in this case being the manufacturers and the carriers..?

Mark Zuckerberg: And the people building the Operating Systems. It could be anyone. There are people that manufacture phones who come and talk to us. And are like, OK, we have our own OS where we’re working with Android and we want to see if we can build a deep integration with Facebook on this. Interestingly a lot of this stuff, we don’t even do the engineering work.

The INQ phone, I don’t think we had any engineers work on it. And certainly HTC modifies all their own Android stuff — Sense. I think a lot of companies are trying to figure out how to differentiate on that. A bunch of them are interested in talking to us, I think it makes sense. Social is becoming more important to make all these applications better. If we can help them do that, that can potentially be very valuable, but that’s more them. I don’t know — it’s a very decentralized ecosystem and there’s a lot of interesting stuff going on. But I think the main message that I would hope that you guys would come away with from our strategy is that our goal is breadth not depth.

Michael Arrington: I know these seem like small questions compared to the big vision you just put out. Have there been discussions with hardware manufactures to have a Facebook-branded phone some time sort of soon, in the next year or so? This is what CNET reported on yesterday.

Mark Zuckerberg: What does that mean… Facebook-branded phone…

Michael Arrington: You know, like the Nexus One was Google-branded even though it was also HTC-branded. I would imagine we’re talking about physical branding on the phone itself.

Mark Zuckerberg: I think there are probably exploratory conversations around.. on the order of what I just said with all these different companies. But none of them have gotten to that level of detail. Like Apple in their ads has Facebook as a fairly prominent feature. But you wouldn’t call this a Facebook-branded phone. And we want to do as deep an integration as we can with them, because this is an incredibly important device. I don’t know.. whether we help companies out with marketing, I think is almost more of a tactical question.

Apple had to ask us if they could use Facebook in their ads and we said “yes” — we think it’s good for us and good for them and good for the Facebook app on the phone and ultimately for users, and all that. I think that’s kind of the way we’ll think about it for all these other things too. Our goal is really just to make it so that over time there will be deep integrations everywhere.

I think this was what was tricky with your blog post. Actually I didn’t even read the whole blog post so it’s tough for me to even comment on this. But I would imagine that a lot of the facts in there are correct from whoever gave them to you, but I think the framing was not how we think about it. In a way that was potentially really damaging to what we’re trying to actually do. So, do we have any conversation with someone to do deep integration? I’m sure we do. And I’m sure we’re talking to them about marketing.

Michael Arrington: Is another way of saying that the goal will never be to go head-on against the iPhone and Android and their businesses, at least, the strategy as you currently see it?

Mark Zuckerberg: Yeah, I mean I think we’re trying to do something pretty different. I think those are both, iPhone is clearly a vertical strategy, Android is a bit more horizontal although they’re focusing on specific things now too. I mean like the Droid is vertical, Android I think is horizontal. Ours is definitely more of a horizontal platform. Our goal is to have Facebook be everywhere and everything be social rather than a specific device.

Michael Arrington: And significantly beyond even a native app.

Mark Zuckerberg: It’s interesting, I think a lot of the time there isn’t such a black and white difference between what’s a platform and what’s an app. It’s really just like the most important apps become platforms. It’s like Facebook was an app for a long period time before we created a development platform but the fact that it was the app that was most used by its users kind of gave it the license to be a platform for a lot of other use cases. Right? And I think we’re starting to see that on mobile as well where I’m pretty sure on iPhone, Facebook is by far the biggest app. There are all these stats that come out, I remember there’s one in the UK that’s like Facebook is more than half of all mobile web minutes and I haven’t seen anything that contradicts that in other countries, I only saw this particular study in the UK. So I think it shows that phones are really social devices and all the apps on them should be social as well and I think we want to try to make that happen. But is this our place to build hardware? No. Is our place to manage virtual memory? No.

Michael Arrington: So do you think maybe these rumors–I’m going to ask you to speculate–came out of one of a billion conversations that you guys tend to have all the time with different people?

Mark Zuckerberg: It’s really hard for me to say.

Michael Arrington: Fair enough

Mark Zuckerberg: If I knew who leaked it to you, I would’ve fired them already.

Michael Arrington: One of the interesting arguments is that you’ve got Erick Tseng and Joe Hewitt, who you’ve had around forever, and you have Matthew Papakipos, that’s a pretty serious team of hard core dudes. Is this what they’re focusing on, what you’ve been describing over the last few minutes? These deep integrations across multiple platforms?

Mark Zuckerberg: So they’re actually to date focused on different things. I think Joe and Matt work together on a bunch of stuff. Erick has actually been working on a different part of the strategy. But that makes sense, what I’m describing is a strategy that has a lot of different pieces so you kind of need to spread out the people who can drive these things.

You know one of the things that Joe and Matt have in common is a real passion for web development. Matt did Chrome OS, and Joe complains a lot about app development [laughs], so one of the things that will be interesting for us is a lot of these companies now, that aren’t us, are doing a lot of the heavy lifting on web platforms to evolve them towards HTML5 and I think we may actually be…we’re i hope, helping to push that along as well, but frankly, we’re not the ones who are driving the development in this ecosystem right now, quite the way that Android and iPhone are, but a lot of what I hope that we can do is provide some standardization there.

We’re investing a lot in HTML5 and we want to make that a really good platform and I think that is some of the stuff that they’re doing. So those guys are definitely working on mobile stuff, at least part of the time, I know Matt is. And I know that they both care a lot about the HTML stuff we’re doing and the HTML5 and making that a standard thing.

Michael Arrington: You said Erick’s working on something else, what’s that?

Mark Zuckerberg: Um, oh he’s working on a bunch of stuff.

Michael Arrington: What’s the top secret mobile thing he’s working on though?

Mark Zuckerberg: I actually think that what he’s working on, is like I don’t think that any of the stuff is that top secret.

Michael Arrington: So there’s a theory in talking to my sources that there are…

Mark Zuckerberg: That there are levels of top secretness [laughs]? I mean, we have a pretty open culture.

Michael Arrington: There’s a theory that part of our story is being driven by sources with old information maybe over the summer, the stuff that CNET got into and that part of it is being driven by something else going on here that hasn’t been uncovered yet. And that it’s super secret and it’s Erick’s project and it’s the super secret mobile project. Maybe you don’t want to say what it is, but maybe you want to say there is something or that’s also a red herring or what?

Mark Zuckerberg: No, I think that one of the ways we break stuff down internally is that we have some people who work on projects and some people who run the whole operational strategy. So for example the highest level version of this — you have Chris Cox and Mike Schroepfer who run the organizations for product and engineering, and then you have people like Bret Taylor who’s CTO who will work on specific projects at a really high level but doesn’t have anyone reporting to him.

The way that the mobile stuff is structured — Erick is really the lead PM for mobile stuff in general. I’m sure there’s specific things in there that aren’t announced yet. But he’s not running a secret project. That’s… I think that’s more like the type of stuff that Joe is working on or that Matt would typically work on.

Michael Arrington: The secret stuff goes through him?

Mark Zuckerberg: Not in general… I don’t think the secret stuff is that secret (Elliot Schrage: There’s some cool things, even if they’re not secret). There’s some stuff that hasn’t been announced, but there’s nothing like a fundamental departure from the strategy that I am laying out.

Jason Kincaid: So when he wrote about a Facebook phone, my initial thought wasn’t about a smartphone, it was about a feature phone. And this isn’t a sexy feature at all, but Facebook Zero, you guys launched last spring.

Mark Zuckerberg: That’s actually going really well.

Jason Kincaid: For me, and I saw that and at the time you had 50 carriers signed up and I’d imagine you have more than that now with free data access to basic functionality. And when I heard Facebook phone, my understanding with the way it works right now, is these people who don’t have great data connections and have pretty.. not great phones, and they don’t have great browsers. And they have to fire up their browser manually and go to And when I heard Facebook phone, I wondered, what if they got a phone that launched directly into Facebook for free for basic functionality like messaging, and your address book. Carriers can charge for other content like media. Am I totally off base here?

Mark Zuckerberg: I need better more updated stats on how Facebook Zero stuff is working. My understanding is that it’s a really big deal in Africa and India and that it’s helping those people get on. Certainly a big thing that we will hopefully eventually be in a position to work on is spreading smart phones and spreading more of this stuff to people who can’t get them today. And that kind of links to the HTML5 stuff that I was talking about. We’re not working on anything specific to what you’re saying today. But just because there’s so many mobile phones today and we’re not on all of them so I think that’s the first order bit — it’s building a mobile platform that can span everything and make it so every app everywhere can be social.

But I think a big goal for this over time will be.. the web is so good for this. I can just speak as a developer, 2004 when I was coding Facebook there was no question in my mind like what I was going to build for. It was, you’re clearly going to build for the web. I’m not going to build software and I’m not going to build for a phone.

And that clarity was so valuable whereas today it’s like, Ok, we want to go build an app. Even a new product that we launch. We’re working on Questions, and it’s like OK. So we build Questions for the web, then we build the “m” site for Questions, then we build the Touch HTML5 version of questions. Then we build the iPhone version of Questions, and then the Android version, and then maybe.. (Elliot Schrage: iPad…) Right, the iPad stuff. And then we don’t work on a RIM version and then a bunch of people are pissed because it’s not available on their phone.

It’s kind of a disaster right now. I really hope that the direction that this stuff goes in is one where there’s more of a standard and again I think we have some people who are pretty good at working on this and hopefully we can capitalize on that because frankly we don’t… we have 4 or 500 engineers at the company, it’s pretty hard for us to build a lot of new products and build them all for these different platforms. So if something like HTML5 becomes a big standard then that would be hugely valuable for us. So we’ll help push that. I imagine that over the long term that will be the solution to this problem that you’re talking about.

The HTML5 stuff I think it’s going to be pretty good. Some of the stuff that we’re working on. Let’s say we build something for the web. And we have an HTML5 version of it. But we haven’t built an iPhone version yet. If we can build a container in so that we can at least make it so that as part of the Facebook app on the iPhone you can get notification hooks and things like that and get an HTML version of the app until we get around to build an iPhone version then that’s cool. We just launched Places a month ago and we still don’t have it done on the Android version yet. And that sucks. If we could build just one HTML5 version then have it everywhere then that would be awesome.