Open Book 2014

This weekend, a group of about 50 developers met at the historic NY Public Library for the Open Book 2014 Hackathon with the goal of imagining a new future of digital books, as well as advancing the open source and open API building blocks needed for diverse ecosystems of authors, designers, developers, publishers, libraries, booksellers, and readers. A wide range of extremely talented people came from as far as Japan to take part in two days of programming, discussion, and novel thinking around the topics of ebooks, annotation, and libraries. Special thanks should be given to both the Readium Foundation and NYPL labs for organizing an excellent event in such a beautiful environment with so many exciting people and ideas that this humble blog post won’t be able to do justice to how much fun it all was.

There was even a lion!

Open Annotation

From the initial ideas pitch and a short demo, a great amount of excitement developed around the prospects of open annotation.  “Social reading”—the idea that ebooks combined with annotation can be a great venue for online communities of readers—was a theme running through many of the groups. But there was also a great amount of interest in other aspects of annotation, such as improving citations and providing additional context and high quality information to the reader. Great discussions were had with many from NYPL labs on the potential for libraries to use annotation technology and the role they play in the community to curate high quality annotations on books of all kinds. Ben Vershbow pointed me to a project he initiated four years ago called Candide 2.0, which showcased a type of Medium style paragraph level comments and great annotations.

Many of the projects that sprang up chose to use Hypothes.is or Annotator in some fashion, and I walked away with a ton of feedback—most notably the need for better documentation on the API, an authentication API, and a request for a groups API that would allow 3rd party applications to easily create groups for social reading applications. A project called Open Books used Hypothes.is and Epub.js to set up reading groups where people could upload an epub and read/annotate it with their friends.

Additionally, I had a wonderful conversation with the folks at Readium about integrating Hypothes.is and open annotation, which is work we hope to begin very soon. This is especially exciting since the Readium SDK will be used by ereader device makers, making open annotation accessible to readers with e-ink screens.

Paul Beaudoin from the NYPL started work on a semantic tagging plugin for Annotator, using image annotation to crowdsource information about playbills for an upcoming NYPL project. Many were excited about the possibility of using annotation to crowdsource semantic tagging for research/organizing purposes, and I’m very excited about Paul’s work.

Hugh McGuire of PressBooks fame was also present and interested in implementing Hypothes.is with PressBooks, which was incredibly easy thanks to the WordPress plugin. His group then began work on utilizing ideas around the semantic web PressBooks. Using the PressBooks wordpress plugin they edited an ebook with additional metadata that revealed itself in an interface similar to eLife Lens. Indeed, it was very complimentary to Paul’s Annotator work since annotation can provide an way to add this meta-data without knowledge of XML.

Improving the Digital Reading Experience

One of the main themes was improving the digital reading experience. Topics explored included pushing the boundaries of what’s possible with ebooks, aiding book discovery, and improving the checkout process for ebooks from the library.

The Epub.js team was hard at work addressing bugs, refining the interface, and adding a beautiful new infinite scrolling view to the library. While a simple feature, it was a vast improvement for the library since reflowable paginated books are only one type of document that aren’t suited for every type of document or even book—a noted use case beyond fixed layout epubs was coding books, where the reader wouldn’t want a long code example split over multiple pages. It was encouraging to see the open source library being used by many of the projects for rendering books in the browser, and there was a great amount of discussion over how it could make the process of getting an ebook from the library as simple as clicking on a link (know having to download a file, plugin a device, and place the file in the appropriate folder which can be difficult for less computer savvy readers).

There was additional talk on how books in the browser could create a much simpler and more effective DRM for ebooks, thus saving libraries a great amount of money since DRM creates a huge tech burden for lending systems with only expensive proprietary solutions like Overdrive. The Epub.js team showed that streaming books were much simpler and easier to control than models where the reader is given a file. Combined with browser-based local storage it was widely agreed that this could potentially be an effective method for lending ebooks in the future.

There was a great amount of the thinking about how to better capture the wonderful amount of curation libraries do in new interfaces for discovering books. Often when people are searching for ebooks, they are limited to an Amazon style list and given limited search options (relevance, price, user rating). While useful for e-commerce, these types of interfaces don’t preserve the work of librarians in their curation efforts to organize knowledge. A team lead by Cori Allen designed a novel interface to aid book discovery and provided a remarkably fresh idea for browsing through digital books. The team’s interface solved this by preserving semantic information on relationships between content, offering a way to explore a topic in terms of broader or more specific concepts with data from the DPLA. The example they used was the topic of dinosaurs, and their interface gave users the opportunity to dive into specific dinosaurs such as T-Rex, or broader topics with books on the Jurassic Period or Fossils in general. The project showcased how the work of librarians (as organizers of information) could be useful for organizing and curating information, aiding the information discovery process in a very 21st century way.

While there were many cool projects to come out of the hackathon, the winning team See and Read had by far the coolest demo which featured an interactive children’s book read through multiple devices simultaneously. All the team members were given Nexus 7 tablets provided by one of the key event sponsors, Google. Other sponsors included O’Reilly Media, Perseus Books Group, and Hypothes.is.

The Future of Libraries

The last day of the hackathon featured a tour of the majestic 5th Avenue library led by one of the members of the NYPL labs. In a lobby filled with books containing meticulously scanned copies of card catalogs in giant leather bound books, describing to us the system of pneumatic tubes that used to be used to send requests to be filled to seven floors of book storage and librarians working below our wonderful tour guide exclaimed, “This entire building is a giant $&@!ing steampunk database.” While very funny, it highlights the role libraries used to play in society, that has in some ways been usurped by the internet.

Inevitably at such an event, in such a venue the conversation often turned to the future of the library. There was much debate on what that should be (and even more on how that change should unfold). Some focused on the role of libraries in the communities, providing support to teach literacy and how this could be expanded to technical literacy. In an event remarkably well attended by women and minority programmers, some argued for a library that would be a place where people could learn to code or have access to a 3D-printer. Others, argued that those who seek information still should be able to turn to Libraries just as readily as fifty years ago, and that now, more than ever we need access to quality curated information, a stamp of authority and quality in an age where it is in increasingly hard to distinguish between signal and noise. They pointed to the incredible role that libraries can play in evaluating and curating knowledge.

Whatever the future might hold, getting a bunch of bright people to meet and work on these great problems together is a great idea, and I for one hope to keep participating in the discussion.