Navigating the Oral Histories of Immigration and Innovation Collection Online

As she recounts in her oral history interview, Chinese-born scientist Hao Wu attended an international scientific meeting in the 1980s, “one of the few international meetings at the time” to be held in China. At the conference, she listened to physicist, mathematician, and microbiologist Michael G. Rossmann discuss his use of X-ray crystallography in his research on viruses. Interested in his work, Wu spoke with him after his lecture and inquired about studying in the United States. She went on to conduct her PhD research in biochemistry in his lab at Purdue University, jumpstarting her scientific career and beginning her life in the United States.

Until recently, you might have had a hard time finding Wu’s tale of the events that catalyzed her journey to the U.S. You would have either had to listen to a multi-hour oral history interview or search for words or phrases in the transcript. In fact, the words “immigration” and “immigrate” appear a total of only three times in the transcript, so you might have passed by this oral history entirely in your search for material about scientists who immigrated to the U.S.

That’s why we have been working hard on a project, which is made possible through a generous grant from the National Historical Publications and Records Commission, to make 70 oral history interviews with immigrant scientists and engineers easier to access and navigate online. I am going to give you a glimpse of our behind-the-scenes work on this collection, explain how to navigate these rich oral histories, and share upcoming opportunities to learn more about scientists’ and engineers’ immigration experiences.

In the Center for Oral History, we conduct in-depth, life history interviews that highlight scientists’ and engineers’ early lives, educational backgrounds, careers, and even interests outside of work. We often conduct an interview with one person over multiple sessions, which can sometimes result in a completed interview that is six or even ten hours long! Because there is so much material, we want to make it easier for you to access these interviews and engage with the stories and reflections that our narrators so meaningfully share.

So, how do we break down a long interview into more manageable segments? My colleague Rachel Lane and I use a tool called the Oral History Metadata Synchronizer (OHMS), which was developed under the leadership of Doug Boyd at the Louie B. Nunn Center for Oral History at the University of Kentucky Libraries.

With this tool, we are able to create a digital Table of Contents for each interview, which gives viewers the opportunity to browse the topics discussed in the interview, and to see summaries and keywords associated with each segment of the interview. You can also search for words or phrases in the Table of Contents and the transcript itself. Using OHMS also allows us to synchronize the audio recording of the interview with the written transcript, such that, if you find a spot in the transcript you want to learn more about, you can jump right to that spot (or a spot close to it) in the audio recording.

How can you use these features? If you were curious about how Hao Wu runs her lab, you could glance through the digital Table of Contents, select the segment titled “Lab management style,” read a summary of that section, view associated keywords, and click “Play segment” to jump to that part of the recording. Or, if you were conducting research about the Cultural Revolution in China, you could search for “Cultural Revolution” and find places in both the Table of Contents and transcript where Wu discusses growing up during the Cultural Revolution.

If you are enthralled by a particular segment of an interview and want to share it with your research collaborators, add it to a lesson plan, or e-mail it to a friend, you can copy and share a link that goes directly to that part of the interview. Learn more about these and other navigation capabilities in the video below.

When looking at an interview’s digital Table of Contents, you might wonder, “Why did they create a new segment there?” Or you might view some keywords and think, “Why is the scientist’s name formatted like that?” Good questions! While it might seem easy to break up an interview and synchronize its audio and transcript, it takes a surprising amount of thought to do so. Creating metadata—descriptive information about an interview that is intended to make it searchable and accessible—is a subjective process, and we use guidelines from experts within and outside of our organization to inform our decision-making.

We seek to split up the interview into manageable segments and title them in informative and easy-to-read ways, but breaking up an interview into segments is sometimes an imperfect science, given how long a narrator focuses on one topic or the non-sequential ways that topics ebb and flow over the course of an oral history interview. We also add keywords to each segment of the interview to highlight, and make searchable, key concepts and terms mentioned in each segment. After much consideration and discussion, we decided to use a standardized vocabulary, FAST (Faceted Application of Subject Terminology), for our keywords. This provides us with a standardized way of formatting names, places, and terms across interviews (so yes, there is a method to our madness!).

We also have to decide what to do when the audio and transcript for an interview do not match perfectly—a situation that can arise as a result of edits made during the transcript review process, or because of past transcription practices that differ from our current standards. A mismatch between the audio and transcript poses a challenge to synchronizing the audio recordings and transcripts through the OHMS process.

For example, if someone discusses a topic early in the interview but it is found much later in the written transcript, the timestamps in the synchronized transcript will be off. When this occurs, we do our best to synchronize the transcript as accurately as possible and add contextual notes for users explaining why the audio and transcripts do not match. There is also work done before an interview even undergoes the OHMS process to make sure that the audio recording and transcript are formatted properly for the indexing and synchronization process.

As you can see, a great deal of thought and strategy goes into the many decisions we make about how to add context to, and enhance the usability of, interviews in our collection. I have learned that writing metadata is not for the faint of heart—and that the expertise of knowledgeable colleagues is crucial!

In addition to making individual interviews easier to access and navigate, we’ve created a dedicated collection page where you can access all the interviews in the Immigration and Innovation project. The landing page for the project allows you to browse by such criteria as the date of the interview, interviewer name, and birth country of the narrator; you can also search across interviews in the collection—for instance, you can search “Ping-Pong” and see how many interviews mention the topic. (I did just that, with fascinating results!)  The following video explains how you can navigate the collection landing page.

Ready to learn more about the immigration journeys of scientists and engineers, their work inside and outside of the lab, and their scientific contributions? Visit our new ExhibitLab, Migrating Science: Stories of Immigration and Innovation, now on display in our free museum, which is open Wednesday through Saturday from 10am to 5pm, and during special events like Curious Histories Fest and our July First Friday (stay tuned for more information).

You can also delve into the oral histories wherever you are via our Digital Collections. And, if there are ways that we can make our collections easier for you to access and navigate, please send us an email at digital@sciencehistory.org with your feedback and suggestions.

The Oral Histories of Immigration and Innovation project is made possible through the generous support of the National Historical Publications and Records Commission (NHPRC).

Thank you to the many staff members whose work has made it possible for these interviews to be available online and searchable, particularly the digital tech team, which includes Annabel Pinkney, Jonathan Rochkind, and Eddie Rubeiz, and Center for Oral History staff member Rachel Lane.