I attended the Visual Resource Association conference for the first time, and this year it was hosted in Los Angeles. I only became aware of the organization in the last year or so prior, which is odd given how closely the focus and goals fit with topics I am interested in professionally! I am glad to now be more connected to this group, and to have had the chance to meet so many folks through their annual meeting.
My first day was spent attending an intensive course in using the application OpenRefine. This is a free, open-source “tool for working with messy data: cleaning it; transforming it from one format into another; and extending it with web services and external data” (OpenRefine, n.d.). Two members of VRA who had practical experience using the application in their own respective professional settings, led the group in the basics. Using sample spreadsheets they provided, we loaded data, got an overview of the interface, performed basic sorting and filtering functions, and learned how to use functions to automate data cleanup. Most of the actions we took were fairly simple, but it is clear how useful even these basic commands could be. I wanted to take the class to better understand how we might approach our data cleanup project at work, where we face the merger of data from multiple different sources (homegrown databases, spreadsheets) into a new digital asset management system. The workshop helped me to think about how we could leverage tools like this one to work smartly, and how we will need to think creatively in order to cluster data to facilitate automation. In some cases, this may mean we need to perform several actions in order to transform information fully. Though this prospect still seems a little daunting, it was exciting to dig in and try my hand at some of this work.
The next days followed with more traditional conference sessions. The first I attended focused on the digital asset management environment at LACMA and the institution’s use of ResourceSpace. The museum’s Digital Assets Specialist discussed where DAMs is located within the organizational structure of the institution, and how that informs their goals. The Collections Information & Digital Assets (CIDA) team takes a user-centered approach to their work, and they are often acting as bridges between the museum’s IT department, external support services for their DAMs system, and staff. She talked about how this work is akin to translation, communicating between different groups in order to make sure all needs and dependencies are met. A big part of this work is also user education and training, to ensure expectations are realistic. With roughly 300 staff using the system, and over 350,000 assets managed within ResourceSpace, she also stressed the importance of management and governance, and integration with other existing systems (collections management database). As my museum pursues a new DAMs system, this presentation gave me a lot to think about - not just in terms of the solution we pick, but how we think about assets as information and how we can better serve staff.
DAMs management overview at LACMA.
Next up were a series of lighting talks on various projects and programs VRA professionals have recently undertaken. All of these presentations were interesting and showed the broad scope of work folks are involved in in this field. An MLIS candidate spoke about the ‘Information Worlds of Art Museum Staff,’ which appealed to me given my recent coursework in information communities. She set about her research project asking whether art museums’ staff information needs were being met by their current information resources, and what information worlds art museum staff contain. Through surveys sent to participating institutions, she learned that: museum websites are the most common information sources for all participants (intranet, public website) and all staff reported problems using this resource; and that online surrogates for the art object or primary text source were most commonly used as opposed to original resources. This has important implications for institutions like my own, and these findings could be leveraged to create better information resources for communities like these.
Another interesting lightning presentation was given by the Hirshhorn Collection Documentation and Research Manager. She discussed the role of photography in documenting artwork, especially with installation and contemporary artwork. While her institution has staff photographers who create this type of documentation, she posed the question “should visitor created photography documentation be collected and archived?” These type of images reveal how visitors interact with, experience, and perhaps even understand works of art. They can also document how a piece changes over time. While the Hirshhorn does not currently have any policies or protocol in place for the active collecting or archiving of visitor documentation, the museum does apply value to interaction on social media, and they are investigating the topic further. It is interesting to think about the possibilities of allowing more documenters - aside from institution-approved photographers - to contribute to the record of artwork.
The Image Resources Curator at UC Santa Barbara gave the final lightning presentation. She talked about how the UCSB Visual Resource Center underwent a physical transformation as they transitioned away from traditional slide library functionality after a renovation. They adopted a modular design to the new space to meet the needs of their community - professors and students. This included the installation of configurable furniture, a projector, multifunctional walls, multiple new workstations, and a large monitor. They also created space to view and temporarily store special collections, archival, museum collections to be used for instruction at the VRC. The physical space thus became much more closely aligned with the mission of serving the evolving needs of the community.
Kyoto Gardens at the conference venue, a lovely spot to take a break between sessions.
The next session I attended was focused on tools for analyzing art, texts, and films. The panel session looked at the development of various technologies from a VRA perspective: the development of a database that used content-based query for films, a comparison of existing OCR tools, and the application of deep learning to analyze the significance of image library slides. While much was presented as a work-in-progress, these technologies hold the potential to unlock information held within a range of resources. There is potential for VRA professionals to be able to more effectively describe and make accessible their collections, to provide new means for discovery, and to empower users to better analyze collections. It is likely that these tools will only improve with time - as is already evident in the advances in OCR.
I picked the next session largely because I was unclear of the definition of the core topics: digital scholarship and digital humanities. I was relieved when the presenters discussed this collective uncertainty we have about these subjects. They talked about how these disciplines bridge technologies, scholarship, and the humanities to bring VRA collections to life through interpretation. These types of projects range from complex and collaborative, like the digitization and presentation of the American Academy photographic archives in Italy, to more discrete and independent, like the interactive website ‘How to Make a Fake.’ The creation of robust metadata and notations, faceted searching options, use of IIIF, and mapping functionality help transform digitized resources to educate users and tell stories. The last presenter touched on some of the challenges within this domain, especially as it relates to the technical expertise needed to develop and sustain these projects over time. One example that helped illustrate this point was the use of a commercial real estate walk-through application, which was employed in a digital humanities paper to show the recreation of a historic interior in 360 degrees. This experience may need to be migrated to a new tool, should this service ever become defunct, and navigating licensing and copyright limits while attempting to effectively archive and preserve the paper can be difficult. VRA professionals cannot expect themselves to have the answers to all these potential roadblocks, but rather must leverage existing contacts with expertise, and help to bring together potential collaborators. As I have found time and time again in library school - it isn’t necessarily about what you know, it’s about knowing how to find the information you need - which includes our information networks.
The session, ‘Framing Places and Identities: Biographies of Photographic Archives and Their Environments,’ provided multiple perspectives on the complex histories and provenance many visual resource collections. In one case - a collection of photographs of the Salerno ivories at the Kunsthistorisches Institut in Florence - the original physical images, digital surrogates, and new born digital images underwent transformations over time. The newest images were merged with the existing collection, and the collection as a whole was transformed as it became available online and was published widely. A different collection, Foto Arte Minore, has continued to grow over the decades. The intent of the photographer whose work comprises the core of the collection was carried out by the predecessors in his company, and more recently by commissions for similar work by the collecting institution. The relationship between repository and creator has thus become more fluid and active. A third presenter discussed how his institution had to piece together the body of work created by one important local photographer after it was dispersed by his family. This involved scouring sites like Craigslist and Ebay, proactively reaching out to the community, and maintaining lines of communication with potential donors over time. Though not all of the photographs were recovered, the portion of the collection that was reunited helped to tell the everyday story of the Japanese and Japanese-American community in 20th century California. The process also brought surviving subjects of the photographs, researchers, and the repository closer.
Next, a VRA professional from a neighboring art museum here in Chicago talked about ethical image use, or “sure, we can, but should we?” She made the case that as time has gone on, it has become easier to determine copyright but less clear if there are ethical implications for licensing images. Several case studies and their issues were presented: the use of an artist’s portrait rather than their work, overlaying text on artwork in marketing material, using artwork that aligned questionably with new product for the museum gift shop. In all situations, though copyright would permit use, there was the potential for conflict over the choice and context of the use of the images. The presenter stressed the importance of communicating with and educating requestors in these cases, to determine if a compromise or alternative solution could be found. Since she is often dealing with living artists, she also discussed how she often reached out to them to clarify these issues. In response to the question she posed of ‘who should be responsible for making ethical image decisions within an institution?’ she concluded by saying that everyone needs to feel ownership. Hopefully this type of awareness and consciousness is growing in libraries, archives, and museums that license images of and from their collections.
Custom-designed digitization station at the Getty.
The final presentation of the conference took one digitization project from start to final products: ‘Ed Ruscha’s Streets of LA: A Lesson in Digitizing, Organizing and Presenting Visual Information at the Getty.’ The team involved in the project first talked about the origins of the work, namely the access issues due to the format of the materials and the conservation concerns. They worked with a vendor to develop a custom solution to digitize the collection to minimize preservation issues and allow for ventilation, while also ensuring quality capture. For metadata, the team relied on accompanying written notebooks, notations on original housing material, OCR’ed street and business signs, and Google Cloud Vision API to create item-level descriptive data. They were also able to leverage existing GIS location data when they were able to determine street addresses. For both Google Cloud Vision API auto tagging and imported GIS location data, the team had to spend some additional time to perform quality control and assess how accurate and helpful this generated data was in this context. Finally, the team employed IIIF and designed two separate online portals for the anticipated users of the collections: scholarly researchers and the general public interested in the subject of the images (streets of Los Angeles over time). It was helpful to understand all of the work that went into the project like this one, especially as it relates to both expertise and staffing.
Example of auto-tagging suggestions for one of the images digitized as part of Getty’s project.
In between all these sessions, I had the opportunity to connect with other VRA professionals and new friends. We debriefed on what we learned, shared about our work, and commiserated over common challenges. Even though I am new to this community and felt uncertain about my role and fit, I felt welcomed. I will value the personal interactions and connections just as much (if not more!) than all the knowledge I gained through the formal presentations.
References:
OpenRefine. (n.d.). Welcome! Retrieved fromhttp://openrefine.org/