Doesn’t reading feel like a creative process? When you read, isn’t it like you’re making something new in your head? As you browse the Web, going from link to link, isn’t that trail itself a creative object?
Those were the kinds of questions that got me interested in the Web in the first place, and in 2002, after running Atomic for a couple of years, I decided to explore the idea of reading as a creative experience.
Read on for another of the many origin stories of the Atomic Cinematic Universe :)
With my experience from building websites for Random House, Fodor’s Travel, Pearson, and other publishers, and inspiration from George Landow’s 1992 book Hypertext (https://archive.org/details/hypertextconverg0000land), I got to work.
What could a creative reading experience look like? How would you capture it? In the new world of the Web, how could my reading connect to your reading?
The first thing I needed was something to read. I reached for text that was public domain and had already been used as a kind of source code in people’s brains and on the stage for centuries: Shakespeare’s 37 plays.
I liked that Shakespeare’s plays are highly structured data: 5 acts, divided into scenes, divided into characters’ speeches, stage directions, asides, and more. This appealed to me as a programmer but also presented an interesting challenge: all of the available public domain sources represented the content of the plays as plain text, either literal text files or PDFs. So, part of the project was to build tools to convert the text into a structured database. I did that in Java with the help of JPedal, and called my library JavAtom. That experience became the foundation of many projects to come for clients, including the first-ever publisher-owned search-inside-the-book system, Random House’s Insight.
Back to 2002: The reason I wanted the plays broken down "atomically" (!) was so that a person could chart a course through the plays at a word and passage level, stringing the bits of text together in a way that they wanted. Once I had the text parsed, I designed a database schema, implemented it in Postgres to hold it all, and automated the process of populating the database from the files.
With my new Shakespeare database ready, I could build the interface. I had my own custom Atomic tag library for JSPs ready to roll (this also became very useful later for every site I built in the next three years). The tags were able to pull from the database and reassemble all the tiny pieces of the text into web pages, no coding required.
That meant I could build any interface, but what? This moment in software development is something I talk about with our partners: the hardest part of development is knowing what to build; how to build it is relatively pretty easy, and that's where Atomic comes in. In this case, it was all Atomic, including the hard part, but I had some thoughts.
First, whatever this product became, I wanted it to be simple. I wanted my parents and my 15 year old brother to be able to use it to make something, even if they didn’t care about Shakespeare at all. That established my user focus – something that has always been extremely important at Atomic, kind of a 25 year obsession.
Second, I wanted people to create entirely from the raw material of the original source text. Third, I wanted the individual creations of all the users to stand on their own and to fit together into an interesting whole.
With those design constraints, and some help from a talented designer who was willing to donate her time (and whom I married a year later, not only for her web design skills), I built Atomic Shakespeare.
I launched it and was pretty happy with it. I tested with some friends and family. They were maybe puzzled but I noticed they kept using it.
I emailed a bunch of newspapers, and one day the New York Times called me up and it was Pamela Licalzi O’Connell on the phone, who wrote the weekly Thursday “Online Diary” column in the paper, the actual literal New York Times physical newspaper! Holy smokes! Atomic Shakespeare was written up in the New York Times (https://www.nytimes.com/2003/07/31/technology/online-diary.html), which felt like being an actual rock star. After that, newspapers all over the world picked it up.
This year, I decided to see if I could make the antique 20-year-old code run again, and sure enough, it did -- even the applets!
So, for everyone interested in the kind of weird internet hypertext theory stuff that was happening right after the dot-com bubble burst, I give you Atomic Shakespeare: the author’s text is raw material; use it to build something new.
Blog posts are still to come on how and why I got it working again 20 years later, and also the project that came right after: Word, a search engine for scripture.
Comments