Archives for posts with tag: Travis Alber

BookGlutton Logo

Some startups are driven by money or market control, but BookGlutton was created out of a passion to help people appreciate what they read. Its mission has always been to make reading books more like the way we read, share, and discuss other content.

Since 2007, BookGlutton has delivered an innovative social reading experience. For many years, it was the only way people could discuss a book right from the page. Shared commenting has always been a core part of the experience.

For 7 years now, I’ve devoted countless personal hours striving to iterate on the BookGlutton experience and technology. It’s been an “interesting” ride with many ups and downs. Both my co-founder, Aaron Miller, and I have learned a lot along the way.

Unfortunately, we now find ourselves unable to fund further business operations and continue to devote our attention to innovating on the site. As of September 7, 2013 all of BookGlutton’s operations and associated services will be discontinued. This includes the website, catalog, unbound reader, widgets, APIs, and EPUB conversion tools.

The site has represented a great innovation in reading and publishing, and we’ve seen it inspire other entrepreneurs and visionaries. I would like to personally thank all of the users who’ve used our products.

For those looking for a replacement for BookGlutton, try ReadUps.com, a new reading experience we’ve just launched. Like BookGlutton, it’s also a social reading system that allows paragraph-level comments and real-time messaging. ReadUps, as a platform, is designed for people to “meet up” inside a book / url / personal writing sample. It’s the evolution of BookGlutton, in a way. But unlike BookGlutton, every ReadUp is part of a single group event, with a set duration, after which the content expires. Inviting people to a ReadUp is as easy as just sharing a URL – and it’s the details like url-based sharing that we expanded on from our days with BookGlutton.

If you’d like to know more, I’ve posted a brief post mortem below, as well as some personal notes about the experience of running this publishing startup.

Sincerely,
Travis Alber, Founder
BookGlutton.com / ReadUps.com

—————

POST MORTEM

 

HOW IT STARTED

We came up with the idea for BookGlutton in a bar, when we were sitting around complaining about PDFs and the difficulty of actually getting and reading an ebook. (Isn’t this what everyone talks about it bars?) The crux of the argument: “why can’t you share and discuss a book like you can a movie or webpage?” We wanted to read a book in a browser, like we read news, email, and blogs. We wanted to leave comments for each other, and, knowing full well that many of these might be snarky, we wanted the ability to shield that from the masses. Ideally we’d even be able to talk in real time if we were in the book together, especially with friends far away, in NY or LA.

In truth, the idea had been building for some time. Aaron Miller and I had lived in Krakow, Poland in 2005. There’s a cool English language bookstore there, but we didn’t have access to many of the things we wanted to read. We also didn’t know many English speakers. We were craving conversations with people about real things (not just how we felt about Polish beer!). We wanted to work our brains a bit.

But the concept began even earlier. Aaron’s master’s thesis had been about the ramifications and logistics of  publishing a novel on the web. Part of it was a self-publishing experiment using a homegrown Perl script. This was way back in 1998.

It didn’t dawn on us until much later, not even as we sat at that bar, drawing little boxes on a napkin, that both of us had been thinking about the enormous possibility of  books in browsers for most of our adult lives.

As we sat talking about what kind of reading experience we’d want, Aaron said if we can’t find what we want on the Web now, we should just build it. That sounded like a good idea to me.

 

BUILDING BOOKGLUTTON

At first BookGlutton was going to be a small side project in relation to our other freelance work.

BookGlutton Site Circa 2009

Core features included a reading system built using JavaScript that could parse ebooks in EPUB format. For social features we decided the paragraph was a good unit to attach shared comments, and we also decided to include in-book chat capabilities, limited by chapter.

2448467484_e43daf5ff3_z(1)

As we began to build out core features we realized we’d ignored group dynamics. We’d need to have multiple levels of public / private / group reading, and those settings would need to extend to notes as well as the reading system. Moreover, people needed more things to read, so we built a library and started pursuing publishers to get even more content. We built a free converter to help people get their content into EPUB (and our system). We built a widget so people could embed the reading system in their blog (and still log in to limit it to group members). We built in the ability to skin the reading system, create collections and reading lists, and an  API to get notes out. Eventually we built an entire publishing platform that ingested ONIX files (metadata catalog files) from publishers. We built in the ability to track usage to get statistics on group growth and other key indicators.

Then we built a store. And a mobile site. And with every new browser that came out, and new devices like the iPad, we redeployed. Although we were smart about how we built it, it was huge.

We were grappling with typical startup growing pains, namely Feature Creep. We had a good process: functionality docs, wireframes, strategy docs, full-on designed templates. But, like many people building the future, we totally overbuilt.

 

TIME AND MONEY

Three years passed, and we self-funded it. To bootstrap that long meant making some serious personal cutbacks. We whittled down all our personal spending, cashed in stocks we’d accumulated during the dot-com boom, and drained our bank accounts. We moved to a pretty seedy part of town (with drug dealing neighbors) to save rent money and put it toward hosting costs. A guy lived in a van in the driveway next door.

There were many points along the way we should have pulled the plug. But we had plenty of encouragement too. Teachers at NYU and Yale using it in classrooms, NPR interviews, Wired magazine articles, Webby Awards. Partnerships with Random House and O’Reilly Media. Investors calling, other entrepreneurs asking to license our technology. One particularly large company from Seattle cold-called us to suss out our plans, and said we had “deep wells of optimism.” It was probably not meant to be a compliment, but it only spurred us on.

Too many options led to a lack of focus. As the money continued to dwindle, we hooked up with someone who’d come out of a large media company’s M&A department. He believed strongly in the ideas behind BookGlutton, and we brought him on board to spearhead a fundraising round. It was hard to tell if we should raise Angel or Series A. We had only 150,000 uniques a month and a tiny trickle of revenue. But we had just built the first DRM-free, EPUB-only social bookstore, and feeling like that was worth something, we pounded the pavement. We hit both Sand Hill Road in Palo Alto and a number of firms in NYC, making contact with over 200 investors overall. It wasn’t hard to get meetings: the idea was exciting.

But the responses we got broke down into three categories:

  • We don’t invest in books – publishing is not a high-growth market
  • We’d love to talk to you when you top 1M users or when you’re revenue positive
  • Come back when it works on the Kindle.

We never closed a round.

 

WHAT WENT WRONG

For most startups, there are plenty of factors that work together for you to shut you down. For BookGlutton it included:

  • Being too early (we launched six months after Twitter, and two months after the Kindle came out)
  • Having a small market size
  • Running into difficult content acquisition (publishers couldn’t use our system if they wanted digital rights management, but it’s hard to lock a webpage)
  • Realizing complex user expectations (for both content and devices)
  • Money (to build awareness or acquire content)

Most of those factors go far beyond our control as founders. After all, we had a great vision for something we’d want to use, and it did inherently have use to thousands of people. But without a strong product/market fit, or money to drive conversion, it’s very difficult to bridge the gap.

 

NEITHER SUCCESS NOR FAILURE

That brings me to today, and the concept of what it means to not fail and not succeed at the same time. When you’re running a startup, everyone says it’s okay to fail.

“Fail early and often!”

Having done this a few times, I can safely say that 80% of the people who say this are posturing. What people really mean is that you can’t tread water forever. Don’t be afraid to change. Sometimes it’s also hard to define failure. If people still use it, did it actually fail? How much cash do you need to blow through to define failure? Failure is the absolute hardest way to learn a lesson, but you don’t easily forget the mistakes you make running a startup. The ramifications are huge, and often quite personal.

The time for us to move on from BookGlutton was probably two years ago. At that time BookGlutton was 5 years old – ancient for a website. We’d relaunched it twice, but now the chat functionality was overloading the servers – we were going to need to rewrite the entire way chat was handled inside the book, and that was weeks of work. The discussion we had was heated. After years of work, how could we shut it down now, with people still using it? On the other hand, how could we find the time to fix it? By then we’d moved to NYC and had to pay rent.

We talked about joining an incubator and relaunching, but felt our product was too far along. Also, places like TechStars and Y-Combinator require all founders to be on site full time. Since we were married, we were already taking on external work to pay for health insurance and pay our bills. At least one of us needed to have a job. Without time or money, we decided to scrap the chat. In its place we added in Facebook chat – a huge compromise. It didn’t quite do the same thing, and felt different, looked different. It was at that point we started thinking it was time to shut BookGlutton down.

In the end, we’ve spent hundreds of thousands on hosting, development, and lost income. Note to other couples: don’t marry your co-founder – someone needs to bring home the bacon. We postponed starting a family for years, thinking it would be dumb to try to do both. When decisions are being made at that level, failure doesn’t seem like an option. It just means pivot-as-needed. I was willing to work weekend after weekend, month after month. Plenty of people still wanted to use the site, despite the compromises we were making in the product. Four hundred schools used it to study and discuss humanities online. English as a Second language teachers used it for tutoring, in places as far away as Japan. Publishers used it to develop author audiences. Even families were using it to mark up Tom Sawyer together – how cool is that? All the while, large internet companies would check in frequently to see how we were doing. There was the frequent talk of an acquisition. It’s hard to shut something down that people found useful and interesting. Wasn’t that why we built it in the first place?

Today BookGlutton is still only one of a handful of web-based reading systems with extensive social features. So why shut it down?

The answer to that is that we respect our users. If we can’t afford to maintain the technology and plan for future iterations, we shouldn’t run it. People expect something to work, and all the explanations in the world won’t pacify an angry parent whose kid can’t leave a comment because the server’s choking. Also, I personally want to do good work, and if the money and time aren’t there to maintain something, it’s not set up for success.

Some people say what kept us going on it was passion, although close friends called it addiction. In either case, what we’re left with is an affirmation, mainly to ourselves: that whether you build a company or just a site, the beauty of the Web is that you can build what you want and share it with the world.

A lot like a book, really.

Travis Alber & Aaron Miller   |   August 9, 2013

 

We (Aaron Miller and Travis Alber) just contributed a chapter to Book: A Futurist’s Manifesto – Collections from the Bleeding Edge of Publishing. The digital book is being released by O’Reilly in three parts. Our chapter, which appears in Part 2, comes out today!

You can buy a digital copy now, and get an update when the rest is released. You can also check it out and discuss the book in-progress on Hugh McGuire’s site, Pressbooks, right now. The entire book (including a print version) comes out later this Spring.

Check out our chapter in Book: A Futurist's Manifesto

Book: A Futurist's Manifesto

Our chapter “Above the Silos, Reading in the Age of Mechanical Barriers” is part philosophy, part social reading, part internet history, and part technology. We think it’s a good blend of what we built, what we learned about social reading, and where it’s headed next.

Here’s an excerpt from the chapter introduction.

…We think there is a very simple but profound answer to the question of why people read books: people read books to make connections. This can be considered at a cognitive level, through simple, repetitive pattern recognition, or at a conceptual, spiritual level. Either way, the basic work of the reader’s mind is to make connections, and the basic mode of higher thought is to exist both in and out of the physical world for a bit, drawing lines between the two.

In any written work, there is a cognitive process of connection-making which makes up the act of reading itself: glyphs form letters, letters connect into words, words into phrases and sentences, sentences into paragraphs, paragraphs into a sense of semantic completion. As we read, we progress through linear rhythms of pattern recognition even as we gain higher understanding of an author’s argument, a character’s motivation, or a historical event. By connecting very small patterns together into larger ones, we connect concepts back to the real world around us, to real people and places. The pattern-recognition part can be thought of as a linear progression, necessary grunt-work for the brain to get at the concepts. However, the tangential connections we make are the ones that matter to us — and they’re the reward which is so hard to get to for those who have trouble with the mechanical work of processing the words and sentences. We may even make many unintended connections along the way, and sometimes it’s those surprises that keep us going. From a description of a road on a summer day, we might recall a bike ride from our youth. From a listing of facts about milk, we might be startled by a sudden craving for ice cream. Perhaps during an introspective passage about spirituality, we look up from the page to see our future spouse for the first time.

A book and its patterns, and the place we sit reading it, and the person we fall in love with, can become forever tied together. It is at this level that reading interests and addicts us. We think of it as a solitary act, but it’s often the connections we make back to the real world that make it so rewarding. These connections are sometimes even more interesting when made across larger gulfs. Fake worlds, or extinct ones, can interest us more than the one we live in. We’re fascinated by fictional characters when they mimic or reflect real personalities. Even the most outlandish science fiction can be interesting in this way, because of the allegory, or the grand sense of scale that crisply dramatizes contemporary issues, or the parallels we can make between even the most alien worlds and our own. It’s these very large, meaningful connections that are the ultimate goal of reading. It’s the understanding we gain, or at least feel we gain, about the world we live in, and the people we share it with, that are the deepest connections we make when we read. In that sense, it is entirely social.

Stay tuned, this is about as publishing-geeky as we can get. You might have noticed we’ve been gradually open-sourcing parts of the BookGlutton platform as time permits. We want to share some of the tools we’ve built over the last five years to encourage development of reading systems, startup-technology, and, of course, the publishing revolution (underway now!)

Aaron Miller (@vaporbook), who built the technology running BookGlutton and ReadSocial, and who is now working with NetGalley, has open-sourced the PHP ONIX Importer we use on the BookGlutton site.

GET IT HERE
BookGlutton PHP ONIX Importer
https://github.com/Vaporbook/POI-PHP-ONIX-Importer

WHAT IS ONIX?
Most people probably haven’t heard of ONIX (ONline Information eXchange). ONIX uses XML to process metadata for book publishers. If you’re a publisher that wants to deliver all your titles and associated metadata (title, author, publishing date, price, cover image, etc.), you push it out in an ONIX feed for the retailer to pick up. There are a lot of variations on this — every publisher formats their ONIX feed differently, and they change them at will.

WHAT DOES BOOKGLUTTON’S PHP ONIX IMPORTER DO?
The PHP ONIX Importer is an easy way to import any kind of ONIX and make it available as JSON data structures. JSON interfaces well with web applications and can be served directly from Web APIs and consumed by various kinds of Web clients without depending on other libraries. It’s a small tug, but it gets publishing a bit closer to the web, so we can easily use vital metadata about book products.

WHY IS THIS GOOD?
We attended the Books in Browsers 2011 Conference at the Internet Archive and saw that people are speedily moving toward the web for reading experiences, publishing platforms, book catalogs and reading recommendations. This code will help some of those endeavors get a head start. The BookGlutton PHP ONIX Importer moves the conversation forward, because it is:

  • Based on the most widely proliferated and supported languages of web applications: PHP
  • Timely in the age of HTML 5 where JSON-interchange is replacing XML
  • Compatible with widely used CMS systems such as Drupal and WordPress
  • Battle tested in production on the BookGlutton.com site for several years

AARON MILLER’S OTHER OPEN SOURCE TOOLS
You can find some of Aaron’s other work on github under vaporbook. A lot of it has had a good workout on BookGlutton.com. He’s also involved with the National Information Standards Organization (NISO) and its Standards Development for E-Book Annotation Sharing and Social Reading committee. Here’s some of Aaron Miller’s other open source code:

We’ve been doing the ebook thing for five years now. A veritable flash-in-the-pan for the publishing world. But like dog years, web years fly by at an accelerated rate. Five years is a lifetime in web-years. Five years should really get you fifteen years of street cred.

We’ve learned a lot. About building communities. Running destination sites. Integrating with publishing workflows. Reaching out to third party systems. Pulling content in from other sites. Creating online reading systems. Making people happy. This has all been a valuable, powerful, (sometimes painful) learning experience. But when we sat down and thought about what we really know, we know the most about Social, with a capital “S.” We know what people will and won’t use (which is not to say that we can’t be surprised). We know how baggage from other web communities figures in to people’s expectations for a digital reading system. We know what types of behaviors people bring with them from the print world, and what they really miss when they switch back to it from digital. We know about user experience, and the compromises that sometimes need to made of it in terms of schedule and technology. We know all about user-funnels, stickiness, and a whole host of other concepts that figure prominently into the digital publishing world, whether or not publishers realize it. But when it comes right down to it, we decided we should focus on what we really know. What we’re passionate about.

ReadSocial is launching soon.

Photo Attribution: Will Clayton

Happy Birthday, BookGlutton! You were but a glimmer in our eye in Fall of 2006. A few months later, when the two of us started working on you full time (Jan 07), we knew we were doing something exciting – after all, who had heard of social reading then? In the last four years we’ve built a lot. We’ve seen the industry change right before our eyes. We were in private beta when the Kindle came out. The iPhone was brand new. We were early.

Looking at things from a startup perspective, early isn’t always positive. In truth, we would have done better to build less and start later – but then we wouldn’t have experimented as much. We spent a lot of time building for laptops, wishing tablets would finally happen. We had to build our own social network from the ground up because Facebook didn’t have an API (and then pivot when it did). And we had very little to base our interface on…so we made most of the user experience up as we went along.

What we built at BookGlutton includes:

BookGlutton grew to become a huge system, and has given us plenty of opportunities to geek out. Our initial plan was clear: we just set out to build a reading system with social features. As we moved through the process we found that, to do this, we needed to build a social network to use it…and then a publisher’s system, a content repository, etc. Not everything we built has been a resounding success, but we have learned about all the different aspects of digital publishing and where it intersects with the web in unique ways. Buy us a beer sometime…we can talk about it for hours!

    Over the years we’ve seen some cool uses of the site:

  • People in Iceland embedding Dracula with BookGlutton’s widget and reading it together.
  • Teachers in Phoenix using BookGlutton to teach English as a Second Language (ESL).
  • Japanese classrooms using it to read Jane Austen.
  • Grandparents forming groups with grandkids and leaving them notes.
  • NYU students logging on at midnight to meet as a class to prepare for class.
  • Authors embedding the BookGlutton widget on their websites and leaving comments inside for their readers.
  • Soldiers using it to read with people back home.

It’s been a good ride. We recently launched a new user-funnel with some social gaming aspects and tight Facebook integration (yes, I should send a newsletter out about it). With ebooks taking off, more people are starting to see things our way. We’re excited to see where that leads us next. Aaron and I have launched a separate endeavor, ReadSocial, which brings what we’ve learned about social reading to other reading systems. BookGlutton still has great things in store…

Thanks to all the people who’ve used and supported BookGlutton over the years!

-Travis
travis at bookglutton dot com

Aaron’s presentation at Books in Browsers had 3 fundamental points, and 2 product announcements, so I’d say he pulled his weight among the heavies of publishing present at the Internet Archive last week. It was an impressive guest list, and in truth we were honored to be presenting. His presentation, The World Wide Web (of Books?), is embedded below. I felt these were the three most important concepts:

  • The future requires a new kind of publisher, the Cloud Publisher, who asks “What else can we charge for?” This is a common way for people with web-development backgrounds to approach new markets – what can we create a market for using new technology? Think Foursquare. Think Twitter. It particularly makes sense when facing the Gorgon of online publishing and distribution.
  • Communities are different than Audiences. The ideal community for a book may not align with its current audience. This explains why some retail chains can’t make the jump to community, and why communities are best built from the ground up. This will be a blog post in itself, but it’s worth mentioning here.
  • There are many layers that will live on top of social books, including the metadata layer, hyperlinked layer, and, you guessed it, the social layer.

-Travis Alber

As a designer, you conceive your design with the core values of a project in mind; you strive to reflect the ideas and feelings behind it. Contrary to that, the first lesson in web development is always separate your design from your code. It’s important that projects be flexible. A myriad number of screen sizes and devices mean the “presentation layer” should be designed to change, particularly when you use web technology. Moreover, partnerships will impact your design.

Aaron and I have been making websites for 15 years, so we get that. Most people don’t know it, but what we’ve built at BookGlutton is flexible in many ways. Easiest to change is the look and feel. Over the years we’ve had a number of conversations about offering our “BookClub in a Can,” the ability to export the social experience to other sites, so they can curate their own book clubs. Business considerations and content deals ultimately kept these projects from launching, and BookGlutton remained a destination site. But it’s fascinating to see how associating the reading experience with a different brand affects your relationship to it.

ANSWERBAG

ANSWERBAG

GOODREADS

Skinning the Reader takes almost no time at all. However, it changes the experience significantly. The Reader takes on the trappings of that community.

TOR

ELLE

All the mockups listed here preserved the buttons and layout, but even that can change. It makes for interesting consideration. Sometimes these mockups were presented in meetings; sometimes the discussion ended prematurely. See more skins, as well as the original BookGlutton design on Flickr.