A Real Page Turner
A few years ago, zinc Roe began working with the Toronto Public Library to develop a tool for displaying archival books from their various special collections branches. The library’s inspiration was Turning the Pages; a tool created by the British Library that, at the time, required the Shockwave plug-in (The British Library has recently produced Turning the Pages 2.0 as a Windows Vista / Silverlight application). Turning the Pages was available for license through the British Library, but the Toronto Public Library saw the benefit of developing a tool from the ground up which would share the same basic premise but expand the feature set with the main goal being a well presented source document that could be supplemented with further text, images, audio, and video.
We also knew that we wanted to create both the book assembly and viewing tool in Flash as opposed to Director. The tool is written in AS1 and uses Screentime Media’s MProjector software to create a standalone tool for creating the bundles of assets and text that are loaded into the viewer which works equally well locally or online.
What we see now as a version 1.0 tool has been used successfully by the Toronto Public Library in a public touch-screen kiosk format at the Metro Reference Library and had it’s widest release just recently as the core of a jointly produced project called Ontario Time Machine.
At this year’s SXSWi and the Museums and the Web conferences, the “VirtualBook” or “Page Turner” (it needs a real name) was greeted with interest by quite a few people from libraries and museums and it got us thinking about the possibility of a version 2.0 which would build upon the functionality and strengths of the existing version.
We know some of our basic goals in working towards a new version:
- Author in AS3 (bitmap handling and more structured code just a couple of the reasons)
- Streamline the book assembly process and application (cross-platform desktop application and/or hosted process)
- Make the viewing interface relatively “skinnable”
- Expand the display/export options (Flash viewer, HTML view, text only, PDF, etc.)
- Open the process and the product up to benefit from outside knowledge, criticism, and development
The first step in this is this post. We know there are a number of projects with similar goals out there in varying stages of production and we certainly have our personal opinions of where those tools succeed and fail but we would love some direct public feedback on what we have done with version 1.0 and on an on-going basis as we work towards a new version.
As an aside, here are a few companies and organizations with related goals or technology: Open Library, Google Book Search, Zoomify, and Issuu, Any others?
P.S. Please excuse the extremely punny title of this post. I shall endeavour to keep humour reigned-in during all future internet ramblings. Some may say it has been suitable reigned in that they noticed no humour present herein. To them, point taken.
May 29th, 2008 at 2:34 pm
I’m also wondering how institutions are scanning books for archives and more active public display? The BookSnap system by Atiz seems like a great option but it’s also not a small investment for smaller libraries and museums.
A lot of what I’ve seen as far as raw digitized material has been from flatbed scanners or from a digital photo copy stand. The quality has been mixed and gutter shadows and colour balance across spreads are the biggest issues.
June 3rd, 2008 at 12:28 pm
I’ll just keep commenting on my own post…
Wired Magazine took a look at the Internet Archive’s custom Scribe scanning stations. There’s also a 2006 post from Jared Benedict that talks in a bit more detail about the “scanning” set-up. The Scribe (to which, it seems, the Atiz systems owe some of their design) are essentially dual-camera copy stands using fairly high-end digital SLR camera equipment.
I’ve also been wondering, in terms of our own goals of building an archival document builder/viewer, whether it makes sense to try and work with an existing document collection format like DjVu or whether that would add a lot more overhead? Because we are intending to work with Flash, there would need to be some server processing layer to pull images out of the DjVu archive on the fly so it might make more sense just to store the images in a ZIP archive. I know not of these things so I am just theorizing.
June 4th, 2008 at 3:19 pm
I took a quick look through DjVu, and it says the file format is open. AS3 has low-level access to binary data (thought the ByteArray class), so we could (in theory) write a decoder in AS3 to load and parse the files directly.