Blog Post

On Publishing Electronic Editions: Doug Reside @ HASTAC2010

Doug Reside's HASTAC10 talk on "A Technical Framework for Publishing Electronic Editions" begins with a "common anxiety": libraries tend to be underfunded; what money they do have is eaten up by buying and storing print materials; and they simply don't have the budge to deal with the development and long-term maintenence of digital tools. In the talk, Reside proposes a new multilayered model for producing and electronically publishing digital scholarly tools and materials.

First Level: Content

As Reside points out, while software goes out-of-date rapidly, text and image files have "remained relatively stable" for the last twenty years. Libraries still tend to adopt a "protectionist approach," guarding content that "can only be accessed easily through the interface that the library provides." Not much is gained by tying content to the interface, Reside points out, and "potentially everything is lost."

Projects like the Archimedes Palimpsest and Shakespeare Quartos Archive have "taken a more liberal approach to distribution" by providing stable URIs for all content. Taking inspiration from these projects, Reside proposes a collaboratory of web-based archives, with participants agreeing to publish all content at open and stable URIs for 10-15 years (in-copyright materials exempted).

Second Level: Transcription and Encoding

Once digitized, all content should be put online quickly. However, contrary to the policies of most projects right now, in which content and metadata are generated at the same level and by the same people, Reside proposes that the only level-one metadata attached to an object would be its filename. At higher levels, multiple metadata sets could be attached to objects, each one equal to the others in organization. In this way, the "community can decide what metadata to trust."

Third Level: Cloud of Widgets

Interfaces are expensive to develop and tend to become obsolete quickly. Code modularity, then, is key. Whereas large infrastructural projects like SEASR provide a model for developing interoperable digital humanities tools, Reside proposes a similar infrastructure with its scope limited to web-based editions and archives. This would act much like the WordPress CMS, providing a set of core files for developing modular plugins implemented in JavaScript (jQuery).

Reside also proposes an interesting way of deciding which plugins get developed. Proposals could be submitted then voted on by the community of scholars; developers would then bid on winning proposals. It seems this would faciliate communciation between scholars and developers; to a certain extent, it would also decentralize the process of funding by having each project go through multiple voting rounds within different communities. If you only have a few minutes to spend with this talk, scroll to this section -- it's a great proposal!

Level Four: User-Generated Data

Now, users can annotate many editions; however, the servers storing annotations are quickly filled with spam. Users have difficulty separating out good annotations from the bad. In Reside's model, the fourth level would be open data stores run by organizations like university presses. This would act as a "sandbox space" for playing with the content at level one while producing new knowledge for level four. He could imagine a fee for accessing content in these data stores. As Reside concludes, "publication in this walled garden might be a basis for promotion and tenure," much like printed, peer-reviewed publication is today.

Thanks to Doug Reside for a fascinating talk, and for all the HASTACers for their participation!

Further exploration:


No comments