This post was written by Lizzy Zarate, Web Archives Student Assistant for NYU Archival Collections Management and Student Member of the Web Archiving Steering Committee. She is currently completing an MA in Archives & Public History at NYU.

Among the technical elements involved in web archiving, it’s easy to neglect the importance of description. It doesn’t take long to notice that an archived website is not playing videos or that the images are missing. It is harder to discern what is missing in a website’s description. Unlike a physical document, most websites are constantly changing. As such, it can be difficult to write descriptions that persist over time. Many of the websites in NYU’s collections were first captured in the 2010s; naturally, society has changed, and descriptive language should evolve with it. This consideration led me to wonder: how can we engage in reparative description work for NYU’s web collections?

In February of 2022, with the help of Web Archivist Nicole Greenhouse, I began researching best practices for inclusive and anti-oppressive description. While the resources I discovered were extremely helpful, I wasn’t able to find much guidance specifically geared towards web archives. Granted, many of the practices from traditional archival description can and should be applied, but there is still the problem of describing materials that can rapidly and drastically change at any time. With this consideration in mind, I utilized resources such as the Digital Transgender Archive Style Guide and Anti-Racist Description Resources to inform my work.¹ As I began to comb through the web collections, it became clear that much of the reparative description would focus on revising languages of exclusion. 

Here’s one example. The Communications Workers of America website has been crawled 324 times since 2007. This is how the website was originally described:

“CWA, America’s largest communications and media union, represents over 700,000 men and women in both private and public sectors.”

The use of “men and women” implicitly erases nonbinary people and other individuals who don’t identify with these categories. To figure out a course of action, I started by visiting the most recent version of the archived website to read their current organizational biography, which referred to members as workers rather than in terms of their gender. Next, I used the history of the website itself to verify that the language I was using was faithful to the organization’s history. In the earliest crawls, the website had also used “men and women” to refer to its members. Using the archive, I was able to determine that this was changed in 2015.

CWA’s webpage in 2008 refers to its members as “men and women”

CWA’s webpage in 2016 refers to its members as “workers”

Because of this change, I felt it was appropriate to broaden the language used in this description. Here is how I revised it:

“CWA, America’s largest communications and media union, represents over 700,000 workers in both private and public sectors.”

This is a small shift in wording, yet it has larger implications for the archive: our descriptive practices should not default to language such as “men and women” when we’re really just talking about a group of people, gender identity irrelevant. The value of archiving CWA’s website is to document the history of labor organizations. In this case, the language that was initially used actually ends up being a distraction from the primary function of the description. Much has been written in the field about archival silences. For web collections, this is present not only in whose websites we choose to collect, but in how we represent them.

Many of the descriptions I flagged belonged to entries in the Student Organizations collection. It appeared that most of the descriptions in this collection were reproduced from the organization’s own pages at the time of their first capture, which raised a few questions about gender-inclusive description. If a club for women referred to itself as an “all-female” group in 2013, what obligation did we have to preserve such language in 2023, if at all? Given that student members had written their own descriptions, what authority did I have to define the stance of their organization? After all, these descriptions were written by students who may have changed their views since then, but I am also a student. What if the work I was doing ended up later being seen as inadequate, the same way I was labeling theirs? I wasn’t entirely sure how to proceed.

In most cases, I tried to look up the club’s current page on the live web. Many had updated their information with trans-inclusive and gender-equitable language, so I could revise the description without qualms. Still, a few websites remained where they had either gone inactive or still retained this language. For these instances, I decided to keep the language as it was while adding quotation marks as needed. As written in Archival Collections Management’s Statement on Harmful Language, “While we have control over description of our collections, we cannot alter the content.”² Making these changes avoided misrepresenting the position of the organization, but clarified that the language used did not necessarily align with the stance of our department. 

Working through this issue forced me to confront the idea that I had less power over the archive as a student worker. The choices that I made would directly impact how users interact with NYU’s web collections. They would also indirectly reflect ACM’s position on these topics. Consequently, I had to take responsibility for the choices I made in reparative description. I did so with the understanding that all description is iterative and no language could ever perfectly represent all the voices of one community.

Reparative description is often discussed in the classroom, but to engage in it practically as a student worker in web archives has helped clarify my own personal ethos as an archivist. The work is ongoing with no clear endpoint, but it is important to make the time and space for it within our daily work. As Dorothy Berry writes in “The House Archives Built”, “Our descriptive systems are often the first interaction patrons have with our institutions, and when the language and systems feel alienating, patrons will take what they need and leave the rest.”³ By repairing harmful descriptions where we see them, we can remove an unnecessary barrier to access for users of web archives.

References:

1. “Anti-Racist Description Resources,” Archives for Black Lives in Philadelphia, Oct 2019. Accessed Dec 2023. https://archivesforblacklives.files.wordpress.com/2019/10/ardr_final.pdf.
   “DTA Style Guide,” Cailin Roles and Eamon Schlotterback, Fall 2020. Accessed 1 Dec 2023. https://docs.google.com/document/d/1qou1h4DLFQEZg4BIvXiEpGy_TI3rDnrJsPXCsRL-Ki8/edit.
2. “Inclusive and Reparative Work”, Archival Collections Management, NYU Libraries, updated 4 Dec 2023. Accessed 4 Dec 2023. https://guides.nyu.edu/archival-collections-management/inclusive. 
3. “The House Archives Built,” Dorothy Berry, 22 June 2021. Accessed 1 Dec 2023. https://www.uproot.space/features/the-house-archives-built. 

Author bio:

This post was written by Lizzy Zarate, the Web Archives Student Assistant for NYU Archival Collections Management. She is currently completing an MA in Archives & Public History at NYU.