Life Cycle of a Digital Resource

To guide this project, we defined the key stages of support of a digital resource beginning with concept development and planning through the activities needed to continue into the future.

We asked faculty and staff to report on these stages as a way to see how faculty were taking care of these steps, and which campus units were helping them. Along the way, we gathered some examples of good practice, which we share below. We hope that this list will prove to be a useful starting point for administrators at colleges and universities who are thinking about the full range of activities needed when developing campus-wide strategies for supporting the digital humanities and its outputs.

Planning

Planning involves the development of budgets and decisions about whether the project leader will build or use existing infrastructure and tools, all while keeping in mind that the resource should be structured in a way that ensures its preservation. These plans should encourage early preparations for the long term by identifying the eventual costs of upkeep as well as what activities and costs may be required to achieve future goals. This is the stage at which to determine which projects really require customization, and which can be done using existing platforms, templates, and other scale solutions.

At Hamilton College’s Digital Humanities Initiative (DHi), a DH center founded in 2010, faculty apply to become fellows and work with the DHi on projects. One of the first conversations DHi staff have with potential faculty partners is about the full life cycle of the resources they wish to create, from what will be needed to realize the faculty member’s vision to a plan for branding, dissemination, and using the final resource in the classroom. Talking through all these stages at the very beginning and then training faculty to create clean documentation and data (assistant director Janet Thomas Simons says that faculty essentially “become metadata librarians”) is the way DHi attempts to ensure its projects against obsolescence.

Content creation

Content creation may include digitization and processing of born-digital assets that will populate the resource or even scholarly work produced by faculty members and crowd-sourced content. When developing the initial content that will populate the resource, it is worth considering what the resource’s future content needs may be, whether that is new material or just updates.

The Digital Collections Center (DCC) at the University of Wisconsin at Madison Library scans the images and produces the other assets that populate UW’s Digital Collections. When time permits, they also assist faculty who need to digitize materials for their own collections. But before production begins, project proposals must spell out their needs and demonstrate that the collections will fit DCC specifications; if the process advances, a cost estimate worksheet is filled out, and a memorandum of understanding is signed. This upfront preparation helps to manage expectations on both sides of the table and to spell out what precisely is needed by the PI and what is possible for the DCC staff to accomplish, so that the latter group can integrate these outside projects into their workflow.

Technical development

Technical development may involve determining what platform will host and serve the content for the resource, as well as how the content will be viewed and retrieved. Development of tools for search, browse, visualizations, and the interface are important matters to consider at early stages of the project. Planning for the employment of technical staff to support the initial building of a project should ideally build in enough time to test a prototype with some users, so that changes can be made as needed.

HyperStudio, a DH lab at the Massachusetts Institute of Technology, has developed software that allows staff supporting digital projects to work from a common platform but that is flexible enough to meet the different needs of researchers. According to their website, the software was created “to allow Hyperstudio the flexibility to pursue . . . customizations, without re-inventing the wheel. Its purpose, in short: to put the individual professorsʼ humanities approach first”. As prototypes are being developed, project leaders meet regularly with HyperStudio staff, and “extensive” end-user feedback is incorporated into the projects plans.

Storage

Digital storage involves the hosting not only of files that need to be retrieved on a regular basis, but also bulk storage for large databases. It requires servers that are stable and secure.

At Indiana University at Bloomington, University Information Technology Services (UITS) has created a massive (42 PB) tape archival storage system called the Scholarly Data Archive (SDA) as well as the smaller Research File System (for more active files), and has made them freely available to all faculty and staff for their digital storage and access needs. UITS believes they can promise it will be available for as long as anyone can foresee, as they back it up in two locations, one in Indianapolis and the other in Texas—a best practice in the field, should disaster strike any one location. The SDA is most pertinent to DH projects, storing their bits while the front end is typically managed by the IU library. Because all storage is centralized, UITS can manage and migrate everything at once as they see fit; the library can also adjust the front end according to their changing needs without having to worry about the back end.

Project management

Ongoing management is not one discrete life-cycle stage, per se, but represents the need to have someone—whether that is the original project leader or someone else—take ownership of the project, to assure its ongoing upkeep. Such management may include time to promote the project and build audience, to prioritize upgrades when needed, and to determine what new functionality may be needed. Most often, this falls to the project leader or faculty member; in some cases, digital humanities centers and libraries have started assigning staff to be ongoing managers of specific projects.

Some examples of units that we saw taking on long-term project management include: Matt Bryant at CESTA supporting many projects, Clara Henderson at IDAH supporting EVIA, and Wallace Hooper supporting The Chymistry of Isaac Newton.

Technical upkeep

Every long-term digital resource will require some degree of technical maintenance. Even those that are built on shared platforms and are drawing their content from a central repository might find that their links to that repository break when a small back-end change is made. More complex projects may need much more attention, from interface upgrades to providing a light “help desk” access for users.

Preservation

Preservation includes all the activities necessary to ensure the long-term accessibility of a resource, from file migration as technologies change to checking for bitrot on every file. Basic digital content such as PDFs, TIFFs, JPEGs, Excel workbooks, etc. is relatively easy to protect, but interactive resources are much harder, given that those applications rely on underlying software that is itself at risk of obsolescence.

For more on how leaders in the field are thinking about digital preservation, see theDigital Preservation Management Workshop and Tutorial, which was originally developed by Cornell University Libraries and is now extended and maintained by MIT Libraries; and Trevor Muñoz’s notes from a presentation he gave in early 2014, entitled “Digital Preservation’s Place in the Future of the Digital Humanities.”

Dissemination

Dissemination includes any outreach and any efforts to make the digital resource available but also to reach out to its potential audiences. This can include posting it in a repository (for access), but also more proactive forms of outreach, including offering demos at conferences and in the classroom, writing articles or blog posts, or engaging with potential users via social media.

Sheila Brennan at the Roy Rosenzweig Center for History and New Media has long thought about dissemination strategies for the resources created at her DH center. A “cheat sheet” on this topic, entitled “The Four Ps of Digital Project Outreach,” is available on her blog.