00. Introduction

Books are significant objects of our culture, and they are capable in a very direct sense: For thousands of years, they have been capable of carrying knowledge and wisdom, conveying ideals and opinions, influencing thought and action.

Still, the more recent innovations of digital media have introduced considerable temptations in trying to go beyond the static form of ink on paper, to augment print media with the more dynamic and malleable forms of the digital. It is not a coincidence that one of the first demos made for the groundbreaking augmented reality system Chameleon in the early 1990s was a library browser where superimposed digital information showed the way to a sought-for book and provided suggestions for related books. Nor is it a coincidence that overlaid book metadata was an important part of the demo when Sixth Sense, the first consumer-grade augmented reality system, was presented to a general-interest audience in 2009. It is clearly tempting to experiment with augmenting books; to create more capable books, as it were.

A common approach today is to print QR codes on book pages to provide the portals into digital media spaces. This, though, requires the juggling of a smart phone or tablet to activate the connection. The Sony Wonderbook is an example of a book-shaped controller for PS3 where 3D graphics are shown on screen overlaid the video of the user's hands and the controller. These approaches, much like the augmented reality systems illustrated above, tend to foreground the digital.

My interest lies in another direction, where the physical book remains at center stage and the digital is more backgrounded, more ambient. And the way I hope to do that is by combining classical bookbinding with physical computing.

00b. A book that knows which page it is on?

Updated on May 27, 2015.

In order to make a book that is allowed to remain the focus of attention, yet provides ambient digital augmentation, it seems foundational to me to make a book that can transmit which page it is open to. This would open the door to all kinds of corner-of-the-eye augmentations; some illustrative examples could include introducing additional online learning resources for optional perusal in the context of a textbook, peripheral awareness of other readers’ impressions of the poem currently being read in an anthology, or ambient soundscapes for a cityscape photobook.

To clarify, what I have in mind here is a reader reclining on a couch under a pool of light, book in hand and possibly teacup on armrest – but situated in a room that is saturated with digital infrastructures including large and small screens as well as audio playback systems. The first challenge then becomes to pick up the value of the currentPage variable from the book and pass it on to the digital infrastructures.

I can see a number of potentially elegant ways of doing this.

I. A miniature camera in the top of the book’s spine, capturing a video stream of the top edge of the textblock which carries a visual pattern. Image processing is used to determine how much of the book is in the left and the right part, respectively. If this could be done based on traditional top edge marbling, it would be particularly elegant, but a pattern designed specifically for the purpose would also be acceptable.

II. Measuring the flex of the back of the book, inside the spine. But it is unclear to me how this can be done with any precision, and moreover, traditional stitching yields a back that does not have a unique point of flexing for every spread of the book.

III. Apply color to the top edge of the book, much like traditional gilding, but using conductive paint. Then measure the resistance between a point on the inside of the left cover, close to the top corner, and the head band. The idea is that the resistance of a conductive surface depends on the area of the surface – and the area of the left part of the top edge in an opened book is proportional to how many pages it consists of, i.e., which spread the book is opened to.

IV. Similar to III, but leave a gap between top edge conductive paint and headband, then measure capacitance between the left part of the top edge and the headband. This is assuming that the capacitance varies with the area of one of the conductors.

V. Create a set of circuits through the book using conductive ink on the faces of the book pages, in such a fashion that every spread has a unique combination of open and closed circuits. This approach was indicated in 2012 in Elektrobiblioteka, a very impressive diploma thesis project in fine arts by Waldyk Wegrzyn at Katowice, Poland.

VI. If the top and the bottom edges are conductive and contacts are placed on the inside of the cover at the top as well as the bottom, then a conductive bookmark would create a circuit from top to bottom when the book is closed. If the resistance of that circuit varies with the number of pages involved, then the current spread of the bookmark could be measured while the book was closed and reported at least once every time the book was opened.

VII. A book is open to a spread if light falls on that particular spread. Phototransistors are commonly used to sense light. It is conceptually straightforward to imagine a book with printed-electronics phototransistors on every spread. I need to look more into paper electronics to determine whether phototransistors can be printed on paper.

VIII. Exploit the limited range of RFID sensing by having an RFID reader on the inside left cover and a tag per page. The idea is that a tag can be detected when its page is in the left half of the open book, but not in the right half. This may be a quick way of making augmented books with a limited number of spreads, even though it doesn't scale too well.

There are also other ways, such as ambient cameras OCR-ing the page numbers of the open book. I find these less interesting, however, since I want to engage with the physical construction of the book. Ideally, there should also be clear references to classical bookbinding techniques.

Next: The first sketch, using a conductive top edge.