At the Publishing Hackathon at BookExpo America on Friday afternoon, six book discovery startups will pitch their ideas to a panel of six judges, including Reddit cofounder Alexis Ohanian, Silver Lake Capital director Stephen Evans and Perseus Books Group CEO David Steinberger. The winner will get $10,000 and a meeting with William Morris Endeavor cofounder Ari Emanuel.
The Hackathon caps off three days of digital publishing events at BEA. I’ve been here all week, covering topics like new trends in ebook pricing, the new ebook publishers and the biggest difference between book publishers and Amazon.
The event starts at 3 PM ET, and I’ll be live-blogging it here.
That’s it! Thanks for reading along. We’ll be sure to keep an eye on Evoke and see what happens with them from here.
The winner: Evoke. They get $10,000 and lunch (breakfast?) with Ari Emanuel.
Zuckerberg: “You guys will be winning $2500 from William Morris.”
Second place: Captiv.
Randi Zuckerberg: “We only have one trophy, but we decided to hack the rules a bit ourselves and we actually have two winners. William Morris has gracefully stepped up to provide another prize for second place.” (No mention of the dollar value!)
David Steinberger and Jennifer Rudolph Walsh agree that this was “amazing.”
Stephen Evans: “As a technology investor, it was great to see how people are using technology…we find it very hard to make a decision because the quality of how you guys are using technology was awesome. I apologize in advance to the losers.”
Doug Rushkoff: “This means a lot to me, to see this salvation of our industry. The one thing I would recommend as we move forward is to think about more than one user pathway…when you look at a bookstore, there’s a thousand ways that a bookstore works. I’m really into buildings that learn and apps that learn and websites that learn.”
Randi Zuckerberg: “This was a really high quality of projects we saw today.”
Alexis Ohanian: “As a founder, investor, publisher, and soon-to-be-author, it is so fucking awesome to see a hackathon about publishing.” Wild(-ish) applause. “To see this kind of enthusiasm in an industry like publishing…”
Rick Joyce: “We have a decision and we’re going to wrap this up with a comment from each judge, and the judge’s decision, and celebration.”
Judges are done deliberating — early!
In the meantime, people are hastily swilling wine after three days of entrapment in the Javits Center.
They’re going to announce the Hackathon winner in 20 minutes, so stay tuned.
Rick Joyce: “Two things are going to happen now. The judges are going to talk to each other and sift through these excellent options and decide on a winner. You’re going to go over to the drink cart over there, where there’s beer, and we’re going to serve you.”
Randi Zuckerberg asks if users could add Foursquare-like tips. LibraryAtlas says yes, in the future.
“We were really excited to find a project that Google was working on back in 2007 called Google Book Search…we’ve reached out to the guy who’s responsible for that and he’s already offered to come have coffee with us…we think there might already be an algorithm that could do that for us. If not, we’re happy to build one.”
“Where do you get the data about what locations exist in what books?” David Steinberger asks. Answer: “We did it the old-fashioned way. We read.” David Steinberger is going to hate this because it doesn’t scale. Yup: “How do you scale this?”
Answer: No, not yet.
Rushkoff: “Do you see a use for this where people can turn off the push” and just explore a city?
“We incentivize exploring a book not just through its content, but through the city itself.”
A map will pull up nearby bookstores so that you can go and buy the book.
Then when you’re out and walk by a Tiffany’s, you get a push notification and “as you’re looking at the jewelry store, take the phone out of your pocket and read the quote out loud to your friends,” they will love this. The book is “Breakfast at Tiffany’s,” duh.
“After that point, you take your phone and you literally just put it in your pocket and you forget about it.”
LibraryAtlas is a geolocated book and quote discovery iOS app. You check in by book, author and so on, “building a preference page, telling the app what you like.”
It’s 4 PM now and the winner is supposed to be announced at five. So are the judges going to deliberate for an hour while we sit here? Asking for a friend.
Final finalist: LibraryAtlas. “Some of the questions that judges have had for previous teams are also a propos for this team, so that will be interesting,” Rick Joyce says.
Evans asks about reluctance to download a plug-in.
Answer: “That’s why we created the widget.” They also say there’s a revenue-share model when someone buys a book through a widget.
Zuckerberg asks what happens if all of the people in a household are sharing one computer. The answer is that people can just click the browser plug-in when they’re on a given web page and get recommendations based on only that page.
Answer: “There are all kinds of books in the back there” (the back of the plugin?)
Rushkoff: Does the range of selections over time become narrower as it learns more about you? “Is there any way to mitigate that so that the unexpected book can fall off the shelf into your life?”
They say you have to opt into sharing your browser history. Ohanian: “Great. Excellent. It’s not that obvious, trust me.”
Yep. Ohanian just asked how you get used to someone “looking over your shoulder when you’re browsing online.”
So who is going to ask about porn first?
You can try it out at KooBrowser.com.
“This widget can provide analytics to publishers, like what kind of articles go best with their books, and so on.”
“We use machine learning inside of KooBrowser to learn about your interests.” I think that each startup now has mentioned machine learning as a component.
“Book readers and non book readers alike are reading in their web browser more and more.” A browser plugin will deliver book recommendations to them based on the articles they are reading.
The fifth contestant is KooBrowser, which I was particularly intrigued by when I saw it at the hackathon a couple weeks ago. It delivers book recommendations based on web browser history.
Fact: Silver Lake’s Evans only claps by slapping his knee, he does not ever put his hands together.
RZ: “My son is obviously too young to be getting into this…I wonder if you might be considering putting a maturity rating on this? Arya from Game of Thrones is a really cool character but there’s obviously so much sex and violence and stuff like that.”
That was the second question that Steinberger — the only book publisher on the panel of judges — had asked about how a model could actually scale.
“It’s really a combination of web scraping — things we find on literary sites that review the books — and it’s part of the snippet process…when you take a look at data frequencies of how people are categorizing that character.”
David Steinberger asks, “How are you going to get profiles of all these characters? There are books published every day that are filled with characters.”
“That just leaves me with two questions. The first is, who would you like to meet today? And the second is, may we introduce you?”
They say they’ll eventually add environments (not just characters), syncing with e-readers and recommendations that increase with character complexity.
The snippets they create are then added to that character’s overall Evoke page. We use “machine learning to create a web of increasingly reliable relationships that obviously relate to the end user.”
Readers look at a character map and create a “snippet” about how they relate to a particular character, like Katniss Everdeen from the Hunger Games.
Evoke’s Lisa Maione says the site “empowers readers and characters to find each other” based on “emotional resonance.”
Fourth team is Evoke. The team members met at the Hackathon.
Fankhauser: “I think there’s ways that somebody who’s not going through a big publisher could find some great art for their book, and that works.”
Uh-oh. Stephen Evans says he’s “a big believer in not judging a book by its cover.” Also, does this model favor big publishing houses that have larger art budgets?
Fankhauser says she really likes the comments that people make on Birchbox and hopes that people do the same on Coverlist.
RZ: What if you like a book but you actually don’t like its cover? Will there be comments or a way to let people let the concept of a book, as opposed to a cover, “as I think that might be an important distinction down the road.”
Randi Zuckerberg: “Again, congratulations on this!”
Fankhauser: “I want people who want to read to love to hang out on this site, so obviously we might add additional features.”
Ohanian: “We see so much of the Internet now helping guide design decisions. Crowdsourced gets overused…[but] do you see a future of” a publisher putting up a bunch of different possible cover ideas and letting users choose?
Coverlist has a partner already: Recovering the Classics, which I wrote about here (http://paidcontent.org/2013/05/28/liked-jane-eyre-hated-the-cover-now-designers-can-sell-classic-books-with-new-jackets/).
“If a writer unfortunately passes away, you could think, oh I really did want to read that book, and you’d find it on Coverlist.”
Frankhauser says that the books shown on Coverlist are updated often, including both classics and bestsellers. “One way we’ll choose is seasonally. During December, maybe we’ll have a book that has a really strong Christmas scene in it.”
“Obviously, it’s a list of books.”
When you click on a book’s cover, there are “reading notes,” similar to “wine notes” in a wine store. “To me a cover doesn’t tell you what a book’s about, it tells you what the book is going to make you feel,” Frankhauser says.
Coverlist is a book discovery site where “you’re asked to judge books by their cover, as we all do.” It’s basically just scrolling down cover images. coverlist.co
Third startup up: Coverlist. Dani Frankhauser, assistant editor of campaigns at Mashable, is presenting.
The woman next to me gave them a little “woohoo.”
“We think a quote is a perfect mirror image to something like a tweet.”
Alexis Ohanian: “I totally agree that the model of showing a book cover on a screen doesn’t make any sense” (did anyone say this?) “Do you think quotes are the hook? What do you think are the new browsing hooks?”
Captiv: “We think quotes are the hook and images are the hook.”
David Steinberger: “You said you’ve got some kind of algorithm that’s going to match the quotes…there’s going to be some kind of streamlined process?” Captiv: “It’s all done automatically.”
Jennifer Rudolph Walsh asks what happens if users don’t like the books they’re recommended. Captiv says it’ll eventually track click-throughs to improve recommendations.
They also have a “strong design philosophy” and say that the best way to draw in reluctant readers is with “a really pithy quote.”
But he notes that there are already a lot of companies using social recommendations. Captiv: “Our engine is real-time book recommendations. THere are a lot of sites that plug into your social media…but a lot of people who don’t like books won’t even visit those sites in the first place….you don’t have to visit any external site.”
“I love your passion, congratulations,” Silver Lake’s Stephen Evans says.
“The versatility of our platform means we have a variety of ways to touch users in the moments when they’re most likely to want to read more.”
It also works for world events. If you tweet about running the New York marathon, you get a quote from “Run to Overcome.” It seems as if it might work better with nonfiction than fiction.
And you are recommended George R.R. Martin’s “A Dance with Dragons,” because this book is (apparently) about “the joy of reading.”
“Let’s say your child just learned to read her first book…you’re excited about this personal milestone, so you go on Twitter and type in a tweet about it.”
Because everyone loves Nicholas Sparks!
For example, if you said you’re excited about getting married, you’ll get a book by Nicholas Sparks.
Captiv “seeks to uncover the semantic meaning, the emotion, the context, behind things like your latest tweets.” They then analyze the meaning and suggest books to you.
Apologies: It’s going to be tough for me to get the names of each of the hackers for each of the startups. I’m going to refer to them by the startup name unless otherwise noted.
Rick Joyce notes that the next team up, Captiv, “has a lot of data scientists on it…they spend most of their time at a place in the Meatpacking District called Think Coffee.”
The people using the site are “going to be interested in anywhere they’re going,” BookCity’s Vincent Trivett says.
RZ: Most people dont’ spend the majority of their time traveling. “How do you engage someone who maybe is only going on one vacation a year, or one vacation every few years?”
Randi Zuckerberg: “First of all, congratulations. I know you guys pulled this together in two weeks.”
Doug Rushkoff wants more geolocation.
The founders note that travel is a good time for book discovery. “We could partner, perhaps, with people that make guidebooks…or language books…to sort of provide the whole gamut of what everyone would need when they go on a trip.”
“This is especially good for cities that don’t have a really famous literary culture.”
You can check it out at bookciti.es. There are maps, book excerpts, “recommendations from local authorities and novelists — these obviously are not real.”
“The idea of BookCities is that the best time to start a book and discover a new novel is when you’re in the city that it takes place….the best time to read Mrs. Dalloway by Virginia Woolf would be on your trip to London.”
First up is BookCities: “Discover the best novel for your trip.”
Rick Joyce reminds us that a Hackathon is about “starting it. Getting on it…We did not ask these folks to come up with a business model.” So don’t ask them if their ideas are actually going to work, I guess.
Randi Zuckerberg has a new book, which may or may not explain why she agreed to be a judge today. It’s called “Dot Complicated.”
The judges are being introduced. David Steinberger of the Perseus Books Group, Jennifer Rudolph Walsh from lit agency William Morris Endeavor, author Doug Rushkoff, Stephen Evans of Silver Lake Capital, Randi Zuckerberg and Reddit’s Alexis Ohanian.
NYC coworking space AlleyNYC hosted the original hacking day two weeks ago. Joyce gives AlleyNYC Jason Saltzman a shout-out.
Today’s six finalists, narrowed down from 30 two weeks ago, are all pitching startups that aim to improve book discovery. Here’s our previous coverage of them, with more to come (obviously). http://paidcontent.org/2013/05/21/six-finalists-in-the-book-discovery-publishing-hackathon-winner-to-be-announced-at-bea/
Rick Joyce says that a lot of people in the digital community aren’t really sure what the point of book publishers is. This hackathon is “an exciting and necessary conversation.”
Okay, we’re getting started. “You’re here, you’re part of the future of our industry,” says Perseus Book Group’s CMO Rick Joyce.
By the way, pro tip for next year’s BEA: Did you know that there is an American Express lounge with free Wi-Fi, snacks and outlets directly across from the press room? You need an AmEx business card to get in.
Hey guys, I’m here at the Publishing Hackathon. Should be getting started in a few minutes, and Wi-Fi willing, I’ll be here.