Looking at Google Newspapers has got me thinking that the same sort of zooming-panning interface that’s used in that, and in Google Maps, could also be used for viewing books. An example of this is shown in the screenshots from videos on Seadragon linked below.

Seadragon is a zooming-panning technology, owned by Microsoft, and used as a component in other tools, such as PhotoZoom, Silverlight, Photosynth, and various Microsoft mapping applications. When it was acquired by Microsoft in 2007 it got attention as a powerful component of other Microsoft applications, but I haven’t seen it featured as a potential interface design tool for ebooks. This is a relatively small part of the videos below, but the screenshots give a feel for it. These are from two different videos, both showing how the system can be used to zoom in on pages from a book.

The sequence above, which is made up of 800 images from a map collection at the Library of Congress, shows how easy it is to zoom in to find pages that have text and pictures together. This video (2:13) is made by the company from which Microsoft bought Seadragon.

The second sequence is from a video (7:42, the first 2:50 on Seadragon) of a talk by Blaise Aguera, the creator of Seadragon. As indicated, it shows zooming in on a large text source.

Both of these videos emphasize the obvious usefulness of Seadragon technology for mapping applications. But they also show that it has potential usefulness for viewing online e-books — So it’s too bad Microsoft dropped out of the Internet Archive digitization project in May, 2008!

4 thoughts on “Seadragon and Ebooks

  1. Way back in the ’80s I used similar systems to create ‘data spaces’ in software development projects. One I ran on behalf of the UK’s CAA used something almost identical to this for document retrieval. It is very effective and can be taken much farther than this. We had a powerful search tool running where each pixel represented a document. Pixels were located, brightened and coloured according to algorithms that worked on the document metadata. You can imagine how much data could be displayed in a small amount of screen real-estate.

    It’s odd that this kind of system (which was in the research literature long before I came across it) is now ‘owned by Microsoft’. I assume they managed to patent it too, despite all the 20-odd-year-old prior art.

  2. Sounds fascinating! – Do you have any links to the projects you describe, or more specific project names that would be searchable? Sounds like it would be worth writing up.

  3. Lol! Links to the projects? Afraid not. This was all many years before the Web was available. Besides, it was all commercially confidential work for paying customers and was never published. The rights to it belonged to the company I worked for but the division that did the work was shut down more than ten years ago and the whole company has merged with a big Dutch company since. I doubt there is any way to trace the records of those projects – if they still even exist.

    The academic research on which it was based will still be available I assume – although maybe not online. Sadly, after 25 years, I’ve forgotten the names of the researchers who did the work. Most things exciting came from Xerox PARC in those days, so that might be a plce to start looking.

  4. Pingback: Seeing the picture » Blog Archive » Keep the Ball Rolling: Twitter & Blogging Together

Comments are closed.