ARK de Triumph: Linking Finding Aids & Digital Libraries Using a Low-Tech Approach

Editor's Notes

1. Please pardon our badly punned title, we are butchering the beautiful French language for a good cause. We want to tell you a little bit about the digital linking process we developed in our archives. I’ll start us off with a little bit of background about our project and then turn it over to Sara, our metadata and EAD specialist for the archives and Andrea, our metadata Librarian to tell you more about the process.
2. We are from Utah State University’s Merrill-Cazier Library, which is an academic library in the west, serving a student body of about 28,000 students. Our library includes a Special Collections and Archives division that houses our Archives, Special Collections, Digital Initiatives, and Cataloging/Metadata services. About two years ago, we formed a small working group to look at streamlining some of the metadata workflows in the division. Prior to that point, our physical archives were inventoried, described, and housed in Special Collections and then were passed on for digitization to our Digital Initiatives Unit and then further passed onto our Cataloging and Metadata Services unit for description and loading into our content management system. We found that there was a lot redundancy in our workflows that could be streamlined if we all worked together. So, we formed the Special Collections and Archives Digitization group. You can see the outline of the group members here. At the time, we had three people who were working as floaters between all three units, we called them the “SWAT” team because their sole job was to strategically tackle the backlogs in all three departments. The remainder of the 6 person group was composed of one person from each department.
3. In working on streamlining the workflows, we tried to look at how our material appeared to patrons and noticed a significant problem. We had worked for years to create digital collections out of our archived collections but there was nothing in our finding aids that told our patrons the item was available online for viewing. This proved to be a problem for our staff, as well. Our research desk in special Collections is staffed by a rotation of archivists, staff, and students, who had an inconsistent understanding of what collections had been digitized and even where to look to check if something had a digital surrogate. How can we simplify our infrastructure to make access to items simplistic and obvious. We had to come to terms with the fact that our systems didn’t interact and likely wouldn’t interact anytime in the near future. There would need to be human intervention.
4. LW But, while we needed human interaction, there was the ability to create a more seamless process. Our team also noticed that many times we recreated metadata instead of reusing existing metadata. So, we focused on developing ways to map information out of existing finding aids or other records instead of starting from scratch. As we developed these workflows, we discovered that these processes aren’t just helpful for converting legacy collections into digital collection, they would also be helpful for increasing our turn around time for getting born digital collections online. In fact, we could put born digital collections online at the same time we put up the finding aid for the collection. This collection – the Cache Valley Refugee Oral History project is one example of an early collection we were able to put online at the same time we put the finding aid online. From start to finish (including the longest and most difficult task – gathering the oral histories – this collection took three weeks to put online. This includes a digital collection, digital exhibit, and finding aid.) Sara and Andrea will be telling you all about that process. But before I turn this presentation over to them, let me set the stage just a bit by telling you about some of the tools we used and why we intentionally kept them low-tech.
5. LW Originally our SCA-D Team had been training on XML transformations, hoping to create a small specially trained group that could handle transformations from one standard to another. As we went along, we realized that such training, while incredibly valuable, was also time consuming and building any workflows dependent on intense training was likely to see problems in the long run. One of the key reasons is that we, like many institutions, see a lot of turn over. Constant retraining on in-depth coding would be time intensive and cemented our workflows around a few key people, rather than making them flexible. So we aimed to see if we could create streamlined workflows using common tools. Now, we deliberately do not call this “No Tech” process because we do utilize some techy tools. But we aimed to create and document processes with a very low technological barrier that could be outlined in a series of steps and would not need a lot of training to complete.
6. LW Here are a list of the tools we use in our process. Most of them are very common (such as Excel and Word). We do use an XML editor, but it is very minimally and relies on templates so that knowledge of XML code isn’t necessary for anything other than troubleshooting potential errors, which is a step we could leave for the few that have had the training. We also use an ARK generator called EZID, which assigns ARKs to digital objects. I’ll talk more about that in just a minute.
7. LW Just so you know, our workflows were built with certain continengies. At the time we developed the workflows, and even currently, we do not have ArchivesSpace for our finding aids. We hand code all of our EADs – or rather, we use a template that is filled in by our archivists and students. Our content management system is CONTENTdm, so we needed to keep in mind the capabilities and processes for both ingesting and downloading content from this system. And we needed to keep in mind the standards for our two consortium – ArchivesWest, which is were we submit our finding aids. And Mountain West Digital Library. This consortium harvests our digital objects. We needed to abide by the metadata application profiles and best practices for each consortium.
8. LW Which created a complex intersection of rules, issues, best practices, and needs.
9. One last thing to explain before Sara and Andrea talk about the actual process is the ARKs themselves. While we wanted to streamline workflows and link together our digital object with the finding aid, one thing we needed to consider was the longevity of our content management systems. Once we created hundreds and thousands of links between collection items and their digital surrogates, how we would we manage those links and prevent link rot. A key part of our process was to connect finding aids to digital objects using ARKs (hence our bad pun from the title of this presentation.) ARK stands for Archival Resource Key. ARKs are persistent identifiers that can be assigned to a specific digital object. Its like a barcode for a file, but it is “actionable” which means it starts with HTTP protocol. The key to using ARKs is that they are system agnostic – and not assigned by the content management system in use, in our case this is CONTENTdm. While the ARK still uses the CONTENTdm reference URL as a pointer to the digital object, using EZID supplied ARKs allows us to maintain those links in a completely separate database. We have the ability to and batch update those URLs to new URLs should we ever move to a new CONTENT management system.
10. Creating an ARK is as easy as filling out a small form just like this – sayig what the item is, who created it, when it was created, and where it can be found. Because we were doing these in large batches, we had our IT department create a portal to EZID where we could batch upload a spreadsheet with this information and have it spit out ARKs.
11. Here is an example of an EZID ARK record. These are created by registering items in the EZID systems. There is short metadata record along with the Reference URL. And there is some information about the ownership of the item. Now Sara is going to tell you about how used these ARKs