Alright, Josh. People keep sending this link to me from your recent newsletter blast, so looks like I’d better respond. My response has gotten to be almost as long as your original post, so Kate has generously agreed to post my response on her blog, ArchivesNext.
I agree that archival audiovisual media require more intervention than most types of archival records, and I agree that traditional processing workflows can ignore their needs. Without special attention to AV, you often get poorly described media, inaccessible both intellectually and physically, and therefore more at risk and hidden as their collections get checked off as “processed.” But I think there are lots of ways to approach this problem, and for us and any repository with an active EAD implementation as part of their workflow, EAD finding aids have a big role to play.
I think we may be dealing with a problem of talking across communities of practice, here. Media-specific archives whose systems are friendlier towards item-level cataloging (or even sub-item level access) don’t see the utility of EAD, and archival repositories that are not specialized in audiovisual media, but who have AV mixed in their collections, tend to look to their existing processes for the solution.
At the Archives of American Art, we do both the things you and Adam are talking about in the comments here – we create item records when we digitize, either for access or for preservation, and we create finding aids for collections which may or may not have digitized content. The digitization queue and the processing queue are driven by different factors, but truth be told, for description, the processing queue is much more efficient and effective than the digitization queue, as active as it is. Just looking at the numbers, over the last 6 years or so, we’ve managed to digitize/preserve about 1500 AV items from our collection, which contains about 15K AV objects. 10%. In the EAD finding aids we’ve done in the same time period, at least twice that many AV records have been described.
And along the way, we’ve also created a detailed inventory of 15,000 media objects in 800 collections for preservation and collection management. We don’t put that data in our finding aids because the researcher has no need for it. Description for researchers at a minimum needs to say what it is, how it relates to the other stuff in the collection, and how to access it. By getting some basic form and content information into the title, and by including a little bit about format, researchers know if they want to play it as part of their research, and our reference staff knows what we can and can’t do with it. And meantime, the processing staff doesn’t get bogged down in detailed, item-level work. Instead, we collect the detailed information we need to support preservation and collection management on accession of a collection with media, and the database we use to capture that information is used to generate our preservation queue, which is used to create grant projects and support collection-wide activities like planning for cold storage, for instance.
And whether or not you’re familiar with the standard, EAD is here to stay. A recent pre-survey of moving image catalogers, done as part of the update to the AMIA cataloging practice compendium, showed that 31% of moving image catalogers worked at institutions that use EAD. EAD has wide international use and free tools for implementation and the support of the archival community. For us, and I suspect for many collections that are not solely comprised of audiovisual recordings, online finding aids are going to continue to be the primary descriptive tool for our collections. So media in collections are going to get described in EAD, whether or not they get described at the item level as well as a result of digitization.
The real problem in my opinion is that the EAD as written doesn’t provide any guidance for how to describe audiovisual media, so media in collections described in EAD often aren’t described accurately or adequately. The revision coming out this year, EAD3, looks promising for improvement in that area (and a study group is forming that seems like just the right forum for improving media description as the new version is implemented). In the meantime, we’ve been working on guidelines for our processing archivists that are standards compliant and allow us to describe AV in our collections to what I believe is a minimally acceptable level, and to do it consistently and in a standards-compliant way. Last year those guidelines were adopted Smithsonian-wide via the implementation of Archivist’s Toolkit. I routinely get asked for them by other repositories. (Here’s a link to a draft if anyone’s interested).
A couple of things I wanted to point out, too, in response to your blog post. Most archives that create finding aids have collection-level MARC records in their catalogs that link out to the finding aid, so the main catalog does provide access to finding aids, generally speaking. Also, many repositories have found ways to make their finding aids discoverable via web searches, including ours. Try searching for Ad Reinhardt, for example. As of this writing, Google gives our finding aid as the third result. It takes a little doing to make that happen, but it’s do-able, and it’s do-able because EAD is such a portable structure for descriptive metadata.
And with developments like EAC-CPF and research going on around EAD content discoverability, it’s only going to get better. A recent webinar hosted by OCLC Research came with a follow-up e-mail noting a lot of smart people experimenting to see what you can do with EAD structured data. Check out ArchivesGrid, SNAC, findingaids.priceton.edu, etc. Not to mention the awesome regional portals, OAC, NWDA, MWDL, among others (see the appendix of this document, for example). There’s lots of evidence that people like finding aids, too. An Ithaka S+R user study of historians last year found that the number one thing historians want more of from archives is online finding aids.
In light of this widespread adoption and appreciation for their utility by researchers, I think it makes sense for moving image and sound archivists to improve the standard to make sure it serves our collections better than it has in the past.
There are lots of ways to solve the AV access and preservation problem, and different solutions will work in different contexts. In our experience, that is, in a manuscript repository with large, mixed-media collections, EAD is the most efficient descriptive tool for making media in collections intellectually and physically accessible.