Today was the first day of the 2012 Best Practices Exchange (BPE), an annual event that brings together archivists, librarians, IT professionals, and other people interested in preserving born-digital state government information. The BPE is my favorite professional event, in no part because it encourages presenters to discuss not only their successes but also the ways in which unexpected developments led them to change direction, the obstacles that proved insurmountable, and the lessons they learned along the way.
As I explained last year, those of us who blog and tweet about the BPE are obliged to use a little tact and discretion when making information about the BPE available online. Moreover, in some instances, what's said is more important than who said it. As a result, I'm going to refrain from tying some of the information that appears in this and subsequent posts re: the BPE to specific attendees.
Doug Robinson, executive director of the National Association of Chief Information Officers (NASCIO) was this morning's plenary speaker, and he made a couple of really interesting points:
- CIOs are juggling a lot of competing priorities. They're concerned about records management and digital preservation, but, as a rule, they're not worried enough to devote substantial attention or resources to improving records management or addressing preservation issues.
- Cloud computing is now the number one concern of state CIOs, and CIOs are starting to think of themselves less as providers of hardware and software than as providers of services. Moreover, the cloud is attractive because it reduces diversity and complexity, which drive up IT costs. Robinson suspects that most states will eventually develop private cloud environments. Moreover, a recent NASCIO survey indicates that 31 percent of states have moved or plan to move digital archives and records management into the cloud.
- CIOs are really struggling with Bring Your Own Devices issues and mobile technology, and the speed with which mobile technology changes is frustrating their efforts to come to grips with the situation. Citizens want to interact with state government via mobile apps, but the demand for app programmers is such that states can't retain employees who can create apps; at present, only one state has app programmers on its permanent payroll.
- Cybersecurity is an increasingly pressing problem. States collect and create a wealth of data about citizens, and criminals (organized and disorganized) and hacktivists are increasingly interested in exploiting it. Spam, phishing, hacking, and network probe attempts are increasingly frequent. Governors don't always grasp the gravity of the threats or the extent to which their own reputations will be damaged if a large-scale breach occurs. Moreover, states aren't looking for ways to redirect existing resources to protecting critical information technology infrastructure or training staff.
- Most states allocate less than two percent of their annual budgets to IT. Most large corporations devote approximately five percent of their annual budgets to IT.
I was really struck by Mary Beth's explanation of the cost savings Oregon achieved by moving to the cloud. In 2007, the Oregon State Archives was able to develop an HP Trim-based electronic records management system for its parent agency, the Office of Secretary of State. It wanted to expand this system, which it maintained in-house, to all state agencies and local governments, but it couldn't find a way to push the cost of doing so below $100 per user per month. However, the State Archives found a data center vendor in a small Oregon town that would host the system at a cost of $37 per user per month. When the total number of users reaches 20,000 users, the cost will drop to $10 per user per month.
Bryan made a couple of really intriguing points about the challenges of serving as a preservation repository for records created by multiple states. First, partners who don't maintain technical infrastructure don't always realize that problems may be lurking within their digital content. Washington recently led a National Digital Information Infrastructure Preservation Program (NDIIPP) grant project that explored whether its Digital Archives infrastructure could support creation of regional digital repository, and the problems that Digital Archives staff encountered when attempting to ingest data submitted by partner states led to the creation of tools that enable partners to verify the integrity of their data and address any hidden problems lurking within their files and accompanying metadata prior to ingest.
Second, the NDIIPP project and the current Washington-Oregon project really underscored the importance of developing common metadata standards. The records created in one state may differ in important ways from similar records created in another state, but describing records similarly lowers system complexity and increases general accessibility. Encoding metadata in XML makes it easier to massage metadata as needed and gives creators the option of supplying more than the required minimum of elements.
I'm going to wrap up this post by sharing a couple of unattributed tidbits:
- One veteran archivist has discovered that the best way to address state agency electronic records issues is to approach the agency's CIO first, then speak with the agency's head, and then talk to the agency's records management officer. In essence, this archivist is focusing first on the person who has the biggest headache and then on the person who is most concerned about saving money -- and thinking in terms of business process, not records management.
- "If you're not at the table, you're going to be on the menu."