Building an Electronic Records Archive at the National Archives and Records Administration: Recommendations for Initial Development (2003)

Developing and operating the ERA will require technical expertise that NARA now lacks. While NARA recognizes that its growing focus on digital archives will require a culture change and has initiated an associated “change management” program, the need for expertise to launch the project is both urgent and specific. Based on briefings and other interactions with NARA staff, the committee concludes that while there is recognition of the challenges of the ERA program, few NARA staff members appreciate the complexity of building, managing, and operating a production digital archiving system. Multiple paths can help NARA to obtain the necessary expertise.

As the preceding chapters have outlined, development of the ERA represents a significant technical and management challenge well beyond that of a typical IT system procurement. NARA has a few staff in leadership roles who appreciate the significant technical challenges of the ERA program. However, briefings to the committee suggest that NARA today does not have sufficient technical depth to assure success in launching the ERA program—that is, to define and manage the overall architecture, develop the appropriate request for proposals (RFP), evaluate technical responses, negotiate changes in the architecture with vendors, and manage the implementation of the system.

Recognizing the need for additional IT expertise in the ERA program office, NARA has supplemented the office with a small cadre of contractor technical staff who are working on ERA design and technical management issues. The documents and briefings received by the committee to date show some in progress identifying desired features of the ERA and some of its associated technical and organizational challenges. But these documents, although useful early steps in system development, are a long way from the detailed engineering analysis and architecture definition that are critically needed.

NARA needs a small staff that holds complete technical knowledge of the ERA. Even if the

system is built and augmented by contractors, an in-house contract monitoring staff (e.g., the contracting officer’s technical representative) is required that is technically at least as good as the contractor’s people. This staff is essential to developing a basic system design and implementation strategy, preparing an RFP, evaluating proposals, monitoring contract progress, and managing system deployment.

Highly skilled staff will be critical to managing the evolution of the ERA. Chapter 7 recommends an iterative development approach for the ERA designed to reduce technical risk by exposing unforeseen problems and overlooked requirements earlier in the software development process and by providing flexibility in reacting to them. However, because this added flexibility requires an active procurement process and an active management style, it also demands particular management skills. For example, because system components can be contracted at a smaller granularity, contracts must be written to provide flexibility, because requirements and designs for each system component are subject to revision. Not only are there more (but smaller) projects to specify and manage, but the management must also have considerable technical sophistication to recognize technical limitations, to determine new requirements, to identify what should be evolved or replaced and when, etc. The increased flexibility of iterative design helps avoid or correct mistakes but requires deep technical understanding of the system as it evolves. Also, short design cycles require frequent testing, frequent deployment of new systems, frequent comparison of alternatives, frequent tuning of system requirements and goals, and frequent interaction with users.

Since ERA systems and their basic architecture will change over time—owing to both iterative development and improved understanding of the requirements as experience is gained—it will be essential for ERA staff to be expert on their technical properties as they evolve. For this reason, NARA should plan to hire permanent staff having this expertise, although short-term contractors or consultants may also help build a critical mass of expertise to launch the ERA program.

The complexity and novelty of the ERA require that the key IT staff have a breadth of expertise in areas such as the following:

• Developing modern IT systems using networked heterogeneous elements;

• Web-based access techniques, including scaling to meet variable and increasing loads;

• Operating large-scale systems, and a knowledge of elements of their design that allow robust, continuous operation;

• Working with digital media and file formats, including content searching; and

• Security.

Staff with this expertise might be recruited from developers of digital libraries or other online information-retrieval services. To address the kinds of issues laid out in this report, this staff would ideally have an understanding of digital archival issues.

The committee recognizes the challenges facing NARA or any other federal agency in recruiting and retaining IT talent.¹ However, the need for key staff is urgent if NARA is to

stick to its current implementation plan, because their expertise is needed in order to formulate an architecture and RFP(s), evaluate responses, select one or more contractors, and begin development.

NARA should seek ongoing advice from the outside to ensure the success of its ERA development. NARA currently has several advisory committees, but none with a concentration of expertise in electronic records archiving. Indeed, the Advisory Committee on Preservation has only one member who is directly involved in digital preservation. The underlying technologies associated with electronic records and their preservation continue to change rapidly, making it important to seek out high-level strategic advice continually. The appropriate expertise can be found in several areas, including the digital library community, government and commercial operators of online information-retrieval services, and the research community.

Advisors can help chart the course of the ERA. For example, as noted above, external review of the overall ERA architecture would be valuable. Ideally, advisors could also help craft RFPs and evaluate the proposals responding to them. A strong advisory board might also help recruit the necessary senior staff and might spin off more in-depth study groups focusing on particular technical issues.

Operating the electronic records archiving systems, as distinct from specifying and developing them, will require strengthening the IT skills of NARA’s staff. Some of these skills will focus on the following:

• Managing the hardware and software resources of electronic records archiving systems, including setting up procedures for testing new software releases, for auditing the data stored in the archive, for incorporating new hardware (e.g., disks), for scheduling media refresh and consequent decommissioning of old disks, and so forth. The critical preservation responsibilities of NARA place the highest technical demands on staff who manage the system.

• Operating an online service to access electronic collections, including such activities as accommodating varying loads (i.e., scaling up services to respond to public demand) and defending against possible malicious attacks arriving via the public network.

• Wide-ranging problem-solving associated with ingesting records. Skilled people will be needed to solve problems posed by old and incomplete digital data transferred from agencies: extracting useful data, checking its integrity, figuring out in what forms it should be archived, and so forth. Automating robust procedures for processing large numbers of records or record series that are received on a recurring basis will also require skilled IT people to develop the necessary tools and procedures.

As ERA systems are built and deployed, and as digital records become a major part of NARA’s work, much of the staff will need to become savvy and comfortable with digital records and preservation. As the volume of electronic records increases relative to that of conventional paper records, NARA will need an increasing fraction of its archivist staff to

have a combination of technical and archival skills and it will need a larger, highly skilled systems administration and technical support staff. Achieving this shift will require a culture change, as NARA has recognized and reflected in its appointment of a change manager.

A critical aspect of NARA’s new culture will be that IT must be recognized as a “core competence” that must thrive to support digital preservation. NARA management will have to overcome the tendency to look at electronic records as a special problem that can be shunted to an isolated office where only specialists will handle them. NARA will need far more staff who feel equally confident dealing with paper and electronic records and many more specialists with competence in managing and preserving electronic records. The preponderance of training and technical assistance that NARA provides to agencies should also shift its emphasis from paper records to electronic records. Achieving this sort of agency-wide transformation will be difficult.