Digital Publishing Services
As both a distributor and archive of scholarly literature, the Scholarly Publishing Office strives to create a balance between the specific needs of each publication and a desire to build a system that will be maintainable and usable by broad audiences over the long run. Our publications have unique workflows and delivery systems, yet they share many things in common, some of which are requirements of our system architecture that help ensure long-term preservation of digital content and searching across publications.
The most prominent component of our system architecture is DLXS, which includes open-source middleware and a licensed version of XPAT, a powerful search engine for digital text in XML format. Digital resources hosted by SPO are not simply websites created using conventional web development tools: they are interfaces to text and images stored in non-proprietary formats in the University of Michigan's digital library architecture. The University of Michigan has a distinguished history in custodianship of digital resources, successfully maintaining them through changes in technology, and SPO likewise pledges to maintain its publications.
Improving accessibility to our content is an ongoing priority for SPO. Currently, we strive to meet Priority level 1 guidelines for accessibility. If you are experiencing a specific problem that hinders your access to content published by us, please let us know.
Digital text
SPO distributes primarily new works of scholarship that have not been published elsewhere. In most cases, publishing partners give us files in Microsoft Word, PDF, Adobe InDesign, QuarkXpress, or another file format. We convert these files using commercial software, R2Net, and our own custom scripts, to the Text Class XML format used by DLXS.
Publishing partners also supply us with accessory files to the text documents (such as images or other media formats) and provide metadata about the articles, such as titles, authors, and keywords. This information is stored in a database of metadata, from which information is extracted and combined with the converted "bodies" of the articles.
Publications fitting this scenario include the Encyclopedia of Diderot and d'Alembert Collaborative Translation Project and the Journal of Electronic Publishing.
Downloadable versions
SPO has not yet developed an automated way of producing high-quality PDFs from our digital text. If content providers supply PDFs, we will present an option for users to download these files. See, for example, the Proceedings of the Western Society for French History.
Page images
Some publications are concerned about preserving exact formatting of the printed page. For these, we can extract text from PDF files in order to enable searching in DLXS across articles or even publications, but we display images of pages to the user with the option of downloading a full PDF. Philosophers' Imprint is one such publication.
Another reason to use this model is for an extensive publication backfile which cannot easily be reconstructed as digital text, such as with the Bulletin of the American Society of Papyrologists. In this case, print copies are scanned and searchable digital text is created using OCR software.
