Progress continues on accessioning and depositing our first publication’s archive into our Stanford Library’s digital repository. While the project still lives in its original format online after its initial release almost two years ago, it’s never too soon to start safeguarding against the inevitability of digital decay. In fact, there’s a lot to be said for archiving in tandem with publication. The general assumption remains that a website will only last 2-5 years without significant updates. Even if a digital publication lives for several years beyond that without minor breakages as code and browser standards change, it’s a good idea to maintain an archived version of the work so that it remains part of the scholarly record.
We’re applying a few different strategies for ensuring the longevity of the interactive scholarly works we’re publishing. The first is providing authors with recommendations and guidelines for creating sustainable projects. Included in these guidelines is the requirement that authors submit documentation on their projects. Another approach we’re taking is capturing each project as a web archive by using tools like Webrecorder. The resulting WARC file can then be delivered through a browser with minimal perceived loss of fidelity. In addition to web-archiving, we’ve also begun investigating the potentials of emulation and virtualization. While some of these technologies still pose challenges to the specific kind of work we’re doing, and they come with their own dependencies and maintenance requirements, one sure way we can at least keep the files constituting a publication safe is through a repository collection.
Creating our first project’s repository collection has taken a lot of coordinated work and collaboration between the Press and Stanford Library. And even though we aren’t quite finished with the first run-through, it’s already proven to be a very educational process. It’s meant planning a structure for the collection, learning the repository’s accessioning platform, building relational spreadsheets, generating metadata describing each of the objects and how they relate to each other, and agreeing on permission and access levels. Each of those steps has involved quite a bit of conversation and coordination, and the various technological components especially have presented learning curves of different degrees. But what has already started emerging is a much clearer idea of a workflow for next time.
Once the archive is finalized, we’ll post a more thorough report and explore the possibilities of its implementation. But for the next couple weeks, at least, work continues. Stay tuned!
Jasmine Mulliken is Digital Production Associate at Stanford University Press. She coordinates the production and workflow of born-digital projects, including recommending platforms and coding standards to authors, consulting with authors on projects’ technical attributes, and evaluating best practices for archiving and preservation.