Subscription book app Oyster has come a long way since launching just over a year ago. The service has more than quadrupled its library of books to 500,000, landed big publishers like Disney, and released apps on several new platforms like the web. This week, the company debuted a new feature called Book Lists that’s like GoodReads — except this time, the place to find new books is also the place to read them with one tap.

It seemed like a good time to catch up with Willem Van Lancker, the creative co-founder of Oyster, who also happens to be one of the guys who designed Apple’s iconic iOS emoji. If Van Lancker has his way, his legacy will lie in changing the way we read, but the odds are against him: in the world of six-second Vines and disappearing snaps, keeping someone’s attention for more than a minute is a challenge — and Amazon has a competing service of its own called Kindle Unlimited. He took a few moments to speak to The Verge about how things have been going.

This interview has been condensed and edited for clarity and brevity.

Even my friends with Kindles don’t seem to read a ton anymore. There are just so many distractions that seem to deliver quicker fixes of entertainment. How do you compete as an app sitting right next to Instagram and YouTube?

It’s funny — if an investor isn’t a big reader, they don’t see a market for this. But we’ve seen that for young people especially, reading has taken off and become a bigger form of entertainment than music or movies in certain situations. When you’re on the subway or laying in bed, what we want to do is make reaching for books as easy as possible. And when you see the YouTube logo and Oyster logo next to each other, that you can feel that you can dive into that book really easily. We want to be the premium entertainment source for that content.

It’s important to understand that the reading population in the US is massive. The books market, in terms of total books sold worldwide, is bigger than both movies and music. If you include educational books, it gets even bigger. Book reading is a huge attention-driving activity in the world. How many hours do you have to allocate per day for books, movies, etc? What we’ve found is that there’s a massive appetite for this.

Oyster’s new Book Lists feature in action

Once somebody’s signed up, how do you get them to read more books (and thus realize they’re getting value from Oyster)?

Netflix is unbelievable at this. Netflix’s long term plan is that when people are coming home and deciding if they want to watch cable TV, read a book, or watch Netflix, that they pick Netflix and reach this “moment of magic.” It’s not that they know they want to go watch Skyfall or Orange Is The New Black, it’s knowing that they just want to watch Netflix, and I’ll let Netflix tell me what I should watch. And we want Oyster to do the same thing.

Netflix’s selection can be pretty hot and cold. There are the big hits, and the D-list horror movies that just fill space. How is Oyster’s selection?

Of the top 250 books ever written, Oyster has about 150. And that’s stuff that’s in copyright, it isn’t including Huck Finn and those kinds of books. We have those too obviously. Netflix, last time we checked, had 40 of the top movies. There’s already a parity, if not a greater selection of books on Oyster. The challenge with books is that people don’t just come to them for one reason. You can come to them for entertainment, or for enrichment, or because you want to learn something. We cover all those use cases, so the use case for Oyster is pretty broad.

We understand what our user base likes, and how to acquire more of those same kinds of content, so we’ll continue to sign publishers. But we’ll also be more targeted about the kinds of content we really want. It’s similar to Netflix in how they created their own serieses to reach that, and they just signed a four-movie deal with Adam Sandler. You might not think Netflix = Adam Sandler from a brand perspective, but they must know that everybody watches Happy Gilmore, or some of his older stuff.

Would you ever create books and other content?

It’s not something that we’re thinking about right now. We want to be the best partner to the publishers as we can, and that’s about giving them the right information so they can give us the best content.

Is your selection part of package deals, or individually vetted?

We work with six of the top ten publishing houses. Those are the guys that when we sign with them, it adds thousands of books to the library, and that spans everything from literary fiction to sci-fi and the many imprints within them. On the other end of the spectrum we work with some of the best independents in the world, and we’ve been very targeted towards finding independents that maybe only have 20 or 30 ebooks, but those books are the caliber of the top titles on Penguin or Simon and Schuster, and those are like McSweeney’s, Melville House, Chronicle, Disney, Nat Geo, it’s a lot of big name brands with libraries that aren’t quite as large. On that end of the spectrum we’ve signed pretty much everyone meaningful and have deals with all of them. There are only a couple publishers that we still haven’t formed deals with.

How are people finding books on Oyster?

80 percent of all of our books read come from Discovery. What we’re most excited about is creating demand. How can we merchandise, contextualize, and present the titles in a way where we add value? And we do that through a number of ways. One is algorithmic personalization, similar to what Netflix does, but the benefit being that there’s a lot more low-hanging fruit in books for signals because it’s all just text. Netflix has gotten to the point where they can do sentiment analysis within the videos, but scanning text is much easier.

So are you looking for keywords or what?

Yeah, keywords, thematic pairings, and we’ve built a data science team of three people that are building this as a core of the business — that personalized recommendation — and they’re also creating pairings of books. So the similarity between this book and this book might not just be that they’re in the same genre. Maybe they’re not. But we can actually come to understand and cluster books in really interesting ways. What we do with that data is once we have these clusters and their relationships formed, we editorially add a layer of human-ness to it.

“80 percent of all of our books read come from Discovery.”

It might be something not too interesting like “fast paced,” “short,” and “written by a woman in the 1990s.” That doesn’t actually connect with readers, so how can we add an editorial layer on top of that and name that set well? That’s also similar to the merchandising of books in a bookstore. The table doesn’t just say “Fiction and Literature,” it says “Books for a rainy day.” We’ve hired people from McNally Jackson, and Kevin [Nguyen, from Amazon] who have made a really strong push to build that editorial side.

Book recommendations, on the hierarchy of recommendations, tend to mean a lot and people really seek them out. If I’m telling you to read a book, I know it’s going to take you seven or eight hours. It’s not like “Go watch that Adam Sandler movie for an hour and a half.” That’s something we really want to unlock the potential of the streaming model for that, creating connections between users and a deeper sense of community around books.

Amazon doesn’t release much info about the data it gleans from its many readers. What are you guys seeing?

The biggest thing is that our engagement is really off the charts. Our average user uses Oyster for over an hour every day. That’s on the level of Facebook. People are staring at Oyster, which counts for a lot in the mobile economy in a weird way. And most of the time that’s in a book. We get people to books really quickly. On average people tend to open about four books to each book they end up going on to read, so there’s that sense of sampling and browsing. We have the fastest downloads of any other ebook platform. The idea is that you tap the book and you’re on the first page, which really lowers the barrier to reading.

In terms of funny trends, romance books get read faster, especially towards the end. There’s a lot more highlighting in non-fiction. The movement around books is interesting, because with fiction it’s always just a straight line. And there are some books that people just don’t finish. Gatsby is started constantly and never finished.

“Gatsby is started constantly and never finished.”

Where are people reading more, tablets, phones, or on the web?

We’ve always been really big believers that the device of the future for books is the phone. That’s the first thing we went to publishers with when we started talking about the differentiation of Oyster, that we can provide the best possible mobile experience.

It’s hard to get the data on this with Android, because, what is a tablet? But between iPhone and iPad, it’s a 50 / 50 split. It might even be higher on the phone in recent months over the iPad. This is an app that people use on their phone constantly, and we see the actual activity spiking during the week at lunchtime, and through the evening and peaks around midnight, and on the weekends it’s pretty sustained. Unlike a lot of products, our biggest days are Saturdays and Sundays, but when we added the web reader, you see it spiking on weekdays because people are reading during work.

We thought about making a button you could hit that would make Oyster look like Microsoft Word like they do for March Madness. It would be funny to bring that to books.

Why did your gut tell you that people are going to be reading on phones in the future?

It was my own behavior. Even when I’m in bed at night, I have an iPad mini with Retina and I still use my phone. And I have an iPhone 6 now, which is even better.

Are you seeing any meaningful innovations in the ways we actually read?

One of the first and most noticeable thing we do is that the pages default to being vertically oriented. On Kindle, your pages slide sideways, on Oyster they slide vertically. You can pull up and read between pages, then it snaps into place, so you maintain that organization of a page, which is so important with a book, because you think about it as a unit of measurement of a whole book. There’s this number of pages left in a book. But then also what happens when you pull to the side is your eye breaks the line and you have to go back up to the top.

“How can we embed books within other pieces of information?”

We allow for horizontal swiping, but the vertical swiping to me is how you read on the go. Why hold on to some of these tired metaphors of curly page turns and leather bookmarks? We also condensed all the reader settings into themes which feels much more like Instagram’s filters or something, but for books. That’s because the average user doesn’t know the difference between Palatino and Times and Courier, they just like the feel of some things. So we pulled font, type size, background color, texture into themes.

In terms of other technologies, definitely the download and speed of use one. We also have done a lot working with ePub, the standard reflowable ebook technology, but as we develop we want to keep pushing the limits of that. As we think of Oyster and our books as less just on the iPhone or iPad or web, it becomes interesting — how can we show different parts of books, how can we embed books within other pieces of information, there’s a lot of ways books can be broken up. Whereas traditionally books have been thought of as objects you buy from Amazon, it downloads to my device, and then it stops communicating with the outside world. It’s about having books work more like the rest of the internet, and less like a piece of paper.

“We’re not trying to build the next Snapchat — we’re trying to build the next WordPress.”

These aren’t the words you expect to hear from the guy building Facebook’s next big app. Facebook has spent the last two years cloning Snapchat, trying to buy Snapchat, and eventually creating a pseudo-Snapchat. None of these plans have worked, so now it’s building… a blogging platform?

Not exactly. Today, Facebook is launching Rooms, an iPhone app that lets you create tiny message boards for posting text, photos, and videos. In each room you can create your own username and identity, and post or comment with friends or strangers about anything from minimalist furniture to Kendama or Destiny. Like on conventional message boards, you can set moderators, pin posts, set age restrictions, type out some ground rules for posting, and boot bad members. You can set a wallpaper and theme, and even alter the “Like” button that appears below posts to be another verb. But Rooms has no connection to Facebook or your Facebook friends in any way.

Make no mistake: this isn’t an anonymous chat or discussions app, as some speculated Facebook was building. It turns out that Snapchat and Secret aren’t the only apps Facebook’s eyeing as it grows its portfolio of social experiences. Rooms is all about building identity, but just outside the context of the world’s largest real identity service (Facebook). Rooms is perhaps most like Reddit, the web’s town square for discussing specific interests. But Rooms forces you to create a different identity for each room you’re in, and offers no front page or ranking system — yet, at least. For now, Rooms have chronological feeds, just like Instagram and Facebook.

Rooms is all about building identity, but outside of Facebook

The app was built by Josh Miller, the co-founder of web discussions site Branch which Facebook acquired nearly a year ago. After joining Facebook, Miller pitched Mark Zuckerberg on an old idea. Everybody wants to talk about their favorite stuff, but nobody wants to spam their friends who don’t care about it. There’s Facebook Groups for that kind of thing, but Miller emphasized that each user should be able to cultivate different identities for different spaces. Thus, the age-old concept of a message board — a place to build your own persona and talk to people with mutual interests .

But don’t countless message board communities already exist? Yep, but not on mobile, Miller argued. Now Zuck was interested.

“I would love to impress that this was not our idea.”

“I would love to impress that this was not our idea,” admits Miller. “The early web seemed infinite, like what else is out there? You just type in a URL. But today you don’t get that. I only have a few apps on my home screen.” He argues that the majority of Facebook’s increasingly mobile, 1.3 billion users likely haven’t used any of these sites on their phone. Until this week, Reddit, which boasts 175 million monthly users, didn’t even have a mobile app of its own — and even now Reddit’s app is built largely for consumption, not creation. There’s no way to create new subreddits (topic-based rooms) or post videos on the fly. And this is setting aside the fact that message boards like Reddit can be awkward, unfriendly, or downright impenetrable to the average person unless you know exactly where to go. Additionally, most interest-based boards besides Reddit all live on their own websites, built on different systems.

“Mark said that in the early days of Facebook, the site was getting big enough where other colleges were interested, but his inclination was to let UT and Dartmouth have their own Facebooks,” says Miller. “But Dustin Moskovitz said, “No, we can’t do that. Maybe in the future, the idea that [Facebook is] one network is what’s gonna be most powerful about it.’” Moskovitz was right. You do need a platform, not disparate silos, to have real network effects.

But unlike Reddit and the other message board sites of yore, there are no discovery features to speak of — an almost unholy feature omission from a “Facebook app.” The only way to join a room is if you’ve been invited to one. This happens in one of two ways — both involving QR codes. Before you pass any judgments based on QR codes, a notoriously terrible mechanism for sharing URLs, hear me out.

To invite you to a room, I tap “invite,” which generates a QR code image that looks like a square movie ticket. Then, I text you the image. You simply save the image to your camera roll, and when you open Rooms, the app adds you to the room automatically. How? Rooms, like every other social app, asks for access to your camera roll. Each time you open the app, Rooms scans your recent photos for QR code invites, then automatically adds you to the corresponding rooms. If you’d rather do things manually, you can always tap Use Invite in the app and choose the QR code image, or even take a photo of a QR code you found in the real world. Miller hopes that everyone from yoga instructors to concert venues will print and post QR codes for people to grab and join rooms.

Man Utd fan? Join my room! pic.twitter.com/ZVlltx92xB

— Oliver Cameron (@olivercameron) October 23, 2014

It’s a fanciful, unique mechanism that Miller says feels much more native to phones than typing in URLs or searching with text. Nevertheless, the mechanic is a barrier, and virtually guarantees that all growth on Rooms will be organic — a real sign that Miller and Facebook aren’t bent on finding day-one viral success. Everything from MetaFilter to Reddit to FilePile took years to “go viral,” Miller muses, so he approached the founders of all his favorite internet communities and asked how they did it. He heard one thing over and over again: you have to empower creators, and then a larger audience will follow. “We think this will be like forums with the 1-9-90 rule, where the vast majority of people are gonna lurk,” says Miller, because that’s just the way interest-based online communities work. They have to be built from scratch.

Rooms won’t have much content on day one, and the team behind it refuses to leverage any of Facebook’s massive distribution network, or even your phone’s address book. “You can’t connect a Facebook account. We don’t ask for address book access. We don’t ask for names. The only thing that’s optional is connecting email in case you lose your account,” says Miller. “Netscape didn’t need a bunch of info from you to let you create and visit websites,” he adds as he grins. It’s quite the comparison to draw — between a launch-day startup and the first web browser — but Miller clearly believes it.

“Communities are just as addictive as things like Facebook.”

If Rooms works, it will be because Miller and co. built a great tool for creating discussions. “We are a tool for making things, so your room brand should be ahead of ours,” he says. “There’s no Facebook icon or name here. If Apple let us, we’d let each room be a separate app.” And none of these apps would bear any Facebook iconography whatsoever. So why is Facebook putting some of its best engineers — including Alan Cannistraro, the guy responsible for many of the iPhone’s first apps — on a product that has nothing to do with Facebook?

“Communities are just as addictive as things like Facebook,” Miller says. If Rooms succeeds at building a more personal, accessible Reddit for phones, it might be well on its way towards building the next Facebook.