The Digital Archimedes Palimpsest Released

Very exciting news – the complete dataset of the Archimedes Palimpsest project (ten years in the making) has been released today. The official announcement is copied below, but I’d like to point out what I think it is that makes this project so special. It isn’t the object – the manuscript – or the content – although I’m sure the previously unknown texts are quite exciting for scholars. It isn’t even the technology, which includes multispectral imaging used to separate out the palimpsest from the overlying text and the XML transcriptions mapped to those images (although that’s a subject close to my heart).

What’s special about this project is its total dedication to open access principles, and an implied trust in the way it is being released that open access will work. There is no user interface. Instead, all project data is being released under a Creative Commons 3.0 attribution license. Under this license, anyone can take this data and do whatever they want to with it (even sell it), as long as they attribute it to the Archimedes Palimpsest project. The thinking behind this is that, by making the complete project data available, others will step up and build interfaces… create searches… make visualizations… do all kinds of cool stuff with the data that the developers might not even consider.

To be fair, this isn’t the only project I know of that is operating like this; the complete high-resolution photographs and accompanying metadata for manuscripts digitized through the Homer Multitext project are available freely, as the other project data will be when it’s completed, although the HMT as far as I know will also have its own user interface. There may be others as well. But I’m impressed that the project developers are releasing just the data, and trusting that scholars and others will create user environments of their own.

The Stoa was founded on principles of open access. It’s validating to see a high-visibility project such as the Archimedes Palimpsest take those principles seriously.

Ten years ago today, a private American collector purchased the Archimedes Palimpsest. Since that time he has guided and funded the project to conserve, image, and study the manuscript. After ten years of work, involving the expertise and goodwill of an extraordinary number of people working around the world, the Archimedes Palimpsest Project has released its data. It is a historic dataset, revealing new texts from the ancient world. It is an integrated product, weaving registered images in many wavebands of light with XML transcriptions of the Archimedes and Hyperides texts that are spatially mapped to those images. It has pushed boundaries for the imaging of documents, and relied almost exclusively on current international standards. We hope that this dataset will be a persistent digital resource for the decades to come. We also hope it will be helpful as an example for others who are conducting similar work. It published under a Creative Commons 3.0 attribution license, to ensure ease of access and the potential for widespread use. A complete facsimile of the revealed palimpsested texts is available on Googlebooks as “The Archimedes Palimpsest”. It is hoped that this is the first of many uses to which the data will be put.

For information on the Archimedes Palimpsest Project, please visit:

For the dataset, please visit:

We have set up a discussion forum on the Archimedes Palimpsest Project. Any member can invite anybody else to join. If you want to become a member, please email:

I would be grateful if you would circulate this to your friends and colleagues.

Thank you very much

Will Noel
The Walters Art Museum
October 29th, 2008.

This entry was posted in Digital library, Open Source, Projects, Publications, Rights, Standards. Bookmark the permalink.

2 Responses to The Digital Archimedes Palimpsest Released

  1. To echo Greg’s comment, the value of the project goes beyond its rich content and technical accomplishments. Because of its mode of distribution, it stands as an exemplar of genuine open source scholarship, in which primary sources, and not merely syllabi and essays, are shared with the public.

    So often the will to openness is directed toward the finished products of scholarship and not to the raw materials that make such research possible.

    Beyond scholarly communication there are two less overt resources resources that make scholarship possible, and which for the most part remain closed to those who do not belong to elite research organizations (e.g. universities): (1) direct access to experts in the field, and (2) access to raw data and primary sources. The former may perhaps be opened by new forms of collaboration mediated by social software; the latter, however, have remained jealously guarded by the scholars who acquired them, or whose institutions own them. Archaeologists and art historians are familiar with this situation — real research requires keys to collections, keys which are not given to just anyone. These resources are the “Intel inside” for humanistic research, the capital that drives the production of scholarly communication.

    Perhaps the Archimedes Palimpsest project will show the way to a different model for providing access to the scholarly capital. And hopefully projects such as Freebase and dbPedia will have other, richer, deeper resources to work with besides Wikipedia.

  2. Minor correction to the above — I meant Dot Porter, not Greg Crane.

Leave a Reply

Your email address will not be published. Required fields are marked *