Accessibility

Alt-text, short for “alternative text,” is an HTML attribute that provides a short, visual description of an image for users who are not able to see the image for a variety of reasons. Structured as a sentence fragment of around 15 words, it indicates the most important content of an image (for more information on adding alt-text to HTML, see this webpage about alt attributes from W3). Long descriptions, on the other hand, provide more detailed visual information. Often anywhere between a couple of sentences to multiple paragraphs and written in complete sentences, they extensively describe the most important aspects of the image with attention to detail. 

​

For cultural heritage institutions, whose collections so frequently depend on images to communicate information, including visual descriptions is of the upmost importance. The Cooper Hewitt, Smithsonian Design Museum’s Guidelines for Image Description is the most comprehensive and well-respected guide for cultural institutions developing visual descriptions. Developed in collaboration with accessibility leaders and other museums, the guideline presents a guiding structure and detailed recommendations for creating effective descriptions for complex images. One of the most helpful aspects of the guidelines are their recommendations for describing people with respect to physical features, age, gender, ethnicity and skin tone, and identification. Visit the guidelines for full recommendations alongside practical examples.

​

How can a museum implement visual descriptions and alt-text into previously digitized collections, especially when their number might exist in the thousands (or even the millions)? The Cooper Hewitt’s pilot project for developing alt-text and long descriptions – which led to the development of previously mentioned image description guidelines – provides an intriguing example of how a museum might crowdsource staff expertise to increase the accessibility of their website and digital collections. In the words of Ruth Starr, the museum’s Accessibility Manager, “We had a simple premise: a group of staff would join the image description project; each week they would describe some images; and slowly, incrementally, we’d chip away at the 210,000-plus images” lacking visual descriptions. 

​

Within the first three weeks, the team – consisting of staff from the museum’s visitor experience, digital, SHOP, exhibitions, curatorial, education, and library departments – developed 842 descriptions. For a detailed discussion of this project, read this blog post by Ruth Starr. Guiding ideas and lessons learned from their pilot are as follows:

​

• A Core Belief: All staff members, regardless of department or prior experience, bring valuable and unique perspectives to the development of visual descriptions. Careful, thoughtful effort in description development was the most important. 

• Guiding Questions for Description Development: To develop the previously mentioned image description guidelines, the team started with the following guiding questions: 

  • “What is important about the image?”

  • “What word choice best conveys the visual elements?”

  • “How can we use language to best reflect the impression / intentions of the image?”

• Workflow:

  • For each object, staff developed two descriptions: an alt text and a long description. 

  • Staff were welcome to add as much or as little as they could each week without judgment. 

  • Weekly sprints involved all team members selecting images they wanted to work with from a group of strategically prioritized objects, pulled from current and upcoming exhibitions or available through the Smithsonian Open Access initiative.

  • Bi-weekly project meetings provided opportunities to discuss the process, ask questions, and collaborate with others. Collaborative, discussion-focused meetings were important to rich description development; as Ruth Starr notes, “We recognized that descriptions are inherently enriched through discussion—more perspectives offer deeper understanding of images.” During meetings, they discussed difficult questions such as: “What word choice feels right? What are we assuming? How do we describe what is ambiguous to us? How much design vocabulary is meaningful to include? What are we prioritizing in the hierarchy of an image?”

• Keeping it Fun: To invite playfulness and support deeply exploring the Cooper Hewitt’s collections, the team chose one object a week to use in responding to a lighthearted challenge prompt. Some examples:

  • “What will keep you cozy at home on a rainy day?”

  • “I Spy: an object you have at home?”

  • “An object you’d like to have if marooned on a deserted island?”

  • “It was Miss Scarlett, in the dining room, with the candlestick. Something red, dining-room related, one of the 243 results when you search ‘candlestick’ in our collection?”

• Implementation: To ensure newly-developed descriptions were stored and implemented into all of the Cooper Hewitt’s platforms, the team used Coyote, an open-source software project, developed by the Museum of Contemporary Art Chicago, to support the creation and implementation of alt text and other visual descriptions in public museum websites and digitized collections. For more information, visit the Coyote project website. 

 

As of the blog post’s publication in May 2020, the team was excited about the possibility to scale and continue this work as they move into the future. 

​

For another perspective on developing an alt-text creation process, read this blog post from The Block Museum at Northwestern University. Here, they discuss a pilot project for developing alt-text that also had unexpected benefits for connecting staff across departments.