Many Voices: A Storytelling Toolkit Proposal

Background

Oral history projects are a powerful tool for recording and preserving the perspectives of the actual people who have lived throughout history. Oral history has been used as a tool worldwide to serve many disparate purposes, from fostering a more holistic understanding of community health among medical professionals (Hernandez, Genkova, Castañeda, Alexander, & Hebert-Beirne 2017) to assessing the impact of interinstitutional historical preservation partnerships (Stieglitz & Nyitray 2017).

In the cases of communities that have undergone trauma, the recording and dissemination of oral histories can help these communities regenerate a sense of agency and foster empathy in those unfamiliar with the history being preserved (Field 2012). Oral histories in the digital age have helped democratize history, providing a valuable tool for studying communities by listening to the actual voices of those being studied (Nyhan & Flinn 2018), furthering social justice and equity through the development of shared historical authority (Shopes & Starecheski 2017).

Though oral histories offer near-limitless potential for community growth and healing, sharing the resources collected by oral history projects in museum or exhibit contexts can prove challenging, both financially and practically (Blatt 2016). This project aims to lower the buy-in cost of exhibiting oral history by producing a modular toolkit for use by oral history projects of any size and scope, allowing them to create exhibit kiosks using a ready-to-go software package and modular kiosk designs with a total material cost of less than five hundred dollars.

Overview

This project will utilize low-cost and open-source technologies such as the Raspberry Pi single-board computer and near-field communication to create a modular interactive exhibit for use by oral history projects. Inspired by other community-based oral history kiosks (Penn Program in Environmental Humanities 2018), this project will expand on their work to create a ready-to-install software package with accompanying modular kiosk designs so that any oral history or public history organization can create an interactive exhibit that features their work with very little monetary investment.

The kiosk will be designed with equity in mind, from wheelchair accessibility and transcripts, to braille and sensory considerations. Its modular design would allow for a degree of mobility to assist with outreach at libraries, schools or conferences. As part of our prioritization of accessibility, we will aim to lower the technical knowledge requirements necessary to reproduce and maintain this model, creating a software package that is not only easy to access, but easy to use. The digital package will be made freely available online, including the core Raspberry Pi operating system image, plans for a kiosk and a manual.

This will be the first phase of a larger project planned at the Burke Museum of Natural History and Culture, exploring new ways to share narratives surrounding the Hanford Nuclear Development Site and peoples affected by plutonium processing, including radiation exposure during production, storage, and purposeful releases into the atmosphere. For decades, communities across Washington and worldwide have been affected by the radioactive materials processed at Hanford. From the downwind farmers, Indigenous tribes of Eastern Washington and Hanford Site workers, to those affected by bombs containing plutonium refined at Hanford that were dropped in the Marshall Islands and Japan, the stories of peoples affected by radiation provide valuable lessons about ethical science, environmental conservation, and public health crises. Creating new venues to amplify these voices can help promote healing and understanding, as well as the missions and educational goals of the Burke Museum and the University of Washington.

We are seeking funding for the initial phase of this larger project. This funding would go entirely towards the creation and sharing of prototype designs for use by oral history projects and community history organizations, with the intent of using the designs to create kiosks at museums, libraries, and community centers throughout Washington about the Hanford Site and affected communities. The enclosed budget would cover the labor and materials required to produce a working prototype, manual, and software package over the course of ten weeks in the fall of 2019. During this design phase, we aim to hold focus groups and seek design input from community groups in order to make the design accessible to the disabled, blind, and hard of hearing communities.

Goals

1. Create a working prototype for an oral history kiosk by the end of fall quarter 2019. The prototype will be considered “working” when all technical elements (RFID/NFC reader, biographical binders with NFC tags, audio playback) are working consistently and securely housed within the physical kiosk.
2. Share designs and software freely online with the public. Upon completion, final prototype designs, a manual, and applicable software will be shared online through GitHub or a similar public software repository service. GitHub automatically tracks download statistics, providing useful data on public engagement with the final toolkit. In addition to usage analytics, GitHub has forums for public comment on posted projects, allowing us to engage with the global open-source community to refine our designs and code during and after the initial design process.

Resilience, Compassion, and Sustainability Through Storytelling

Creating connection through sharing stories is a deeply human practice dating back to our earliest days. It is through the power of storytelling that young children first understand our common humanity and remains the primary way we bond with one another throughout our lives. Oral histories can foster rich empathetic connections between audience and subject, and this project aims to make it easier for community organizations and oral history collections to share the stories they collect. This project would aid the sustainability of these organizations and allow them to share their stories with a wider audience with little financial risk.

This project aligns with many of the United Nations Sustainable Development Goals. Not only do the proposed kiosks help foster dialogue about the health and pollution issues related to the Hanford Site, but encourage emotional healing as well.  The low cost of these kiosks also helps promote sustainable community action, providing a powerful tool for organizations to share their community’s stories in their own words.

Ethical Review

Developing an oral history project around the Hanford nuclear development site and accompanying kiosk is a fascinating challenge.  While researching other oral history projects in conjunction with examining the ethical guidelines for the Burke Museum, American Alliance of Museums, and the Oral History Association, I have come across several key ethical issues I wish to prioritize over the coming months. While executing this project, I hope to forefront issues of access, equity, accessibility, and reciprocity so that the project not only benefits the Burke, but the public at large—especially nuclear-affected communities whose stories we are seeking to collect.  

Both the American Alliance of Museums and Burke Museum codes of ethics place a strong emphasis on the nature of public trust in museum practice.  I believe that the museum’s responsibility to uphold public trust goes far beyond collecting and preserving as mentioned in these codes.  In addition to the care and preservation of objects and artifacts, the museum must also make those items accessible to the public they serve in order to maintain the public’s trust.  While it is easy to simply make collections available upon request, I believe a much more thorough approach to accessibility is the most ethical practice.  This entails not just making items available to the public, but making them easily accessible to the public without the need to jump through bureaucratic hoops.  Open access collections strengthen public trust and hold the museum accountable to the public they serve.  They also provide an avenue for the public to respond not only to what the museum is collecting, but what is absent from the museum as well.  This creates new avenues for the museum to work in tandem with the public, further strengthening public trust and the museum’s role in the community.  Not only is this best practice, it aligns neatly with the new Burke facility’s design and culture of public facing collections work.  As we collect stories from nuclear-affected people for this project, I want to ensure that the recordings are not only accessible on exhibit via our kiosk, but easily accessible through a digital repository.  The Burke Museum has several digital collections search tools on its website, but none list any audio files.  Nor does the Burke mention digital collections or digital access in its code of ethics.  As part of this project, I hope to help the Burke develop a digital collections policy that is ethical and provides access to not only the stories collected in this project but any digital resources the museum collects in the future.

Creating access to collections doesn’t stop at making digital resources publically available.  Access also means creating materials surrounding the collection that can be accessed by members of the public of different linguistic and physical ability.  While designing our kiosk, it is important to me that we design for inclusivity.  When creating instructions on how to use the kiosk, using a graphical language rather than written words allows anyone to use the kiosk regardless of their linguistic ability.  Likewise, any obligately written materials such as biographical information in the kiosk binders about our narrators should include translations of that material and the oral histories being played by the kiosk.  Designing for accessibility also means ensuring that written materials are available not just in other spoken languages, but in Braille for the vision impaired.  Likewise, any audio or video elements we incorporate should be well subtitled or have a visual counterpart for those who are deaf or hard of hearing.  Ethical best practice for accessibility also requires us to design our kiosk so that members of the public with health and physical ability limitations can access its materials.  This includes, but is certainly not limited to, providing a seat or bench for those who cannot stand for long periods of time and not placing materials at heights that cannot be seen or reached by those in a wheelchair.  These considerations of physical ability are especially important in this project as the nuclear-affected people whose stories we are seeking to record are often disabled or chronically ill themselves.

In addition to the ethics of our final exhibit design, we must take into consideration best ethical practices surrounding the collecting of our oral history stories.  To help develop a framework for this, I have been reviewing the ethical guidelines of the Oral History Association.  These guidelines emphasize the importance of preparation in collecting oral histories.  Drawing from Shawn Wilson’s indigenous research practice and my own belief in collaboration as best museum practice, I hope to consult other local oral history projects in the area such as Densho or the University of Washington Library Special Collections to gain insight from their experience.  If there is potential for creating collaborative partnerships with these organizations—not only for this project but for the Burke as a whole—I hope to help facilitate those as well.  As I have no formal training in history or anthropology, working with these organizations will be an important part of my training in how to best record oral histories.  As I do not simply wish to extract knowledge from these organizations as a resource without providing something in return, I hope to spend part of the preparatory phase of this project doing work for these organizations as both a learning experience for myself and to provide them with something to them in gratitude for their expertise and insights.

Finally, I hope to collect oral histories for this project in an ethical, collaborative manner.  Learning about Hanford and other nuclear stories this quarter has shown me how much pain nuclear-affected people have gone through, and my hope is this project provides them with even a modicum of healing.  As we develop our interview practices for this project, I want to make sure the interviews we are conducting and collecting are thoughtful, well-researched, and balanced.  Should we decide to conduct interviews remotely, ensuring easy access to our interview process will be paramount.  This should include providing copies of the interview for narrators in either a physical or digital format that they choose.  In the interests of equity and diversity, we should also seek narrators that provide a diverse and complete picture of Hanford, including workers of all backgrounds, downwinders, affected Native peoples, and more.  Allowing these people to tell their story by amplifying their voices instead of speaking for them is the primary goal of this project.  We must also be certain to ensure our narrators are providing clear informed consent and are fully aware that we intend to not only make their stories available in our exhibit but to the public as a whole.  This might include providing narrators with the ability to record their oral histories anonymously or under a pseudonym should they desire.

Telling the stories of Hanford and nuclear-affected people worldwide is a daunting and important challenge.  This class has helped shape my personal and professional ethics, and I hope to continue to develop more skills that will allow me to conduct best museum practice as I work on this project and more.  While the above considerations are important, I do not anticipate they will be the only ethical concerns that arise as this project matures.  As they do arise, however, I plan to address them with the same spirit of collaboration, equity, and inclusiveness that I hope to weave into every aspect of this project.

Technical Specifications

Overview

This project’s goal is to create a Kiosk for exhibition in the new Burke Museum that appropriately fits within the Burke’s new mission of transparency.  This kiosk will be used to allow the dynamic playback of previously recorded audio using modern technologies, such RFID, as well as primarily open-source hardware and software.  Due to the community that this kiosk is seeking to serve, it should be designed to be as accessible as possible by a variety of patrons, with a special focus on transcripts, subtitles and wheelchair accessibility.

Description

A Raspberry Pi Model B+ computer would be used to power an RFID-reader, which will be embedded inside the kiosk.  When an item with an embedded RFID tag is placed above the reader, it will trigger the playback of an audio recording through secondary powered speakers.  

The item with an embedded RFID could potentially take many forms, with the only requirement being the ability to embed an RFID tag.  These RFID tags come in various form factors, such as a plastic card, a coin, or a small microchip.   The RFID-embedded item could be a binder with additional information, a photo album, small 3D printed objects, or more.  

Each RFID-embedded item would be coded differently, meaning that the RFID-reader would be able to differentiate the items and play different recordings accordingly.  Audio recordings would be stored on either a USB flash drive or an external powered hard drive.  

Assembly

Features

The Raspberry Pi, which is used to power the exhibit, could be displayed as a part of the exhibit.  It could be displayed through a cut-away window, or with a specialized case that displays the Raspberry Pi outside of the kiosk.  Interpretation labels could be provided to showcase the open-source nature of the electronics that power the exhibit.

Construction

This kiosk could be constructed from various materials:

• Reclaimed materials from local community businesses, such as Ballard Reuse, or other supplies from similar companies, like ReCreative.  This could further community relations.
• Flat CNC-machined or table-cut plywood.
• Flat CNC-machined or table-cut synthetic, such as Nylon, ABS, HDPE or acrylic.

Design

Due to the small footprint of all the required electronic components, the physical design of the kiosk could take many forms, and each would potentially have benefits and detriments.

Full Kiosk

This exhibit could be designed as a full freestanding kiosk, similar to an arcade gaming cabinet.  

The head of the cabinet could display the title of the exhibit, and could potentially be back lit.

The middle section would act as the shelf for the RFID-embedded items that are required for the exhibit to function.  

The bottom section panel would have a 30° - 45° incline which would be used to prop up the RFID-embedded items, such as an open binder.  Hidden beneath this panel would be the RFID-reader, as well as the speakers.

Depending on how this kiosk will be used (individual vs group), a chair or bench could potentially be provided.  Allowing the chair to be moveable would provide accessibility for clients using a wheelchair.

Partial Kiosk

This exhibit could also potentially be designed as a partial kiosk with a smaller footprint, so that it could be placed in a hallway or foyer.

The RFID-reader could be embedded beneath a surface that is at an angle and the Raspberry Pi could either be hidden in a black box, or displayed in a clear box.

Alternately, the RFID-reader could be the mobile element with stationary RFID tags embedded into the kiosk itself.

A shelf could be added below the angled surface and used to store the RFID-embedded items.

Table-Top Kiosk

This exhibit could be designed as a table-top kiosk with a smaller footprint.

This form factor would allow the kiosk to be placed on a table, which could potentially have related information.

Because it can be placed on a table, this may make it possible to create a display that is more accessible.

Required Electronics

• Raspberry Pi 3 Model B + Heatsink
• 5V 2.5A Power Supply
• UHS-I 32GB SDHC memory card
• VESA Mounting Plate and Nylon Standoffs
• Adafruit PN532 RFID/NFC Reader
• Elegoo EL-CP-004 Ribbon Cable
• Adafruit 13.56Mhz RFID Sticker
• Syba USB 2.0 External Audio Interface
• 1ft USB3.0 extension cable
• Full Range Coaxial Speakers - Set of 2
• Lepy LP-2020A Class-D Audio Component Amplifier
• Durable Power Strip

Estimated Cost: $230

Hardware

Raspberry Pi

The Raspberry Pi is a low-cost System-on-a-Chip computer produced by the Raspberry Pi Foundation.  With the exception of the proprietary processor, the system is open-hardware with complete specifications available through the manufacturer.

The Raspberry Pi comes in multiple varieties with differing performance and hardware specifications, making it easy to adapt into many different applications.  

It uses an ARM processor, which is more similar to the processor found in a cell phone than in a desktop computer.  They produce much less heat and are generally considered more efficient than desktop processors.  As such, it can only run operating systems and programs that were made for ARM processors.  

Because the Raspberry Pi runs its operating system entirely off of an SD card, it is easy reproducible for additional exhibits or maintenance.  Since Raspbian is designed specially for the Raspberry Pi, it means that SD cards are hot-swappable between systems; if one Raspberry Pi is malfunctioning, simply move it to a new system.

The Raspberry Pi can be set to a ‘read-only’ mode which improves security and reliability, because the operating system files cannot be altered or overwritten without physical access to the Pi to disable the ‘read-only’ mode.

Specifications (Raspberry Pi Model B+):

• Broadcom BCM2837B0, Cortex-A53 (ARMv8) 64-bit SoC @ 1.4GHz
• 1GB LPDDR2 SDRAM
• 2.4GHz and 5GHz IEEE 802.11.b/g/n/ac wireless LAN, Bluetooth 4.2, BLE
• Gigabit Ethernet over USB 2.0 (maximum throughput 300 Mbps)
• Extended 40-pin GPIO header
• Full-size HDMI
• 4 USB 2.0 ports
• CSI camera port for connecting a Raspberry Pi camera
• DSI display port for connecting a Raspberry Pi touchscreen display
• 4-pole stereo output and composite video port
• Micro SD port for loading your operating system and storing data
• 5V/2.5A DC power input
• ExPower-over-Ethernet (PoE) support (requires separate PoE HAT)

5V 2.4A Switching Power Supply

This is the main power supply for the Raspberry Pi, which is a MicroUSB.  There are many different viable options, including ones that have a power switch.

If the Raspberry Pi is going to be connected to ethernet, it could be powered via the Ethernet cable, but this would require the purchase of an additional Pi extension card and an ethernet power injector.

8+ GB UHS-I/U3 Micro SDHC Memory Card

Raspbian requires at least 8GB of storage.  If this exhibit were to employ an external USB hard drive or USB flash drive for storage of audio recordings, we would not require more than the minimum required 8GB.

Raspbian recommends a minimum speed of UHS-I, though a speed of UHS-III would improve kiosk performance.

Adafruit PN532 NFC/RFID controller breakout board (v1.6)

This is the RFID-reader board that is required for the kiosk.

Adafruit 13.56MHz RFID/NFC Sticker (1KB)

These are the RFID-tags that are required to be embedded in items for the kiosk to function.

USB 2.0 External Stereo Audio Adapter

Higher-quality USB audio interface for audio playback.  Output is 3.5mm audio.

5.25-Inch Full Range Coaxial Speaker - Set of 2

Coaxial speakers to be attached to the Kiosk.

Audio Component Amplifier

Audio amplifier to power coaxial speakers.  Input is 3.5mm audio, and output is powered speaker cable.

External USB 2.0 HDD

Optional.  This would allow the kiosk to save the audio recordings on an external hard drive.  This would improve reliability because if a Pi ever became non-operational, it could be replaced without compromising the audio recordings.

Software

Raspbian Operating System

The Raspberry Pi Foundation actively maintains an open-source distribution of Linux known as Raspbian, with integrated Linux security updates.  It is based on the widely-supported Debian linux operating system.

Raspbian has access to a collection of Linux programs that are actively curated by the Raspberry Pi Foundation, as well as an active community of Linux and Raspberry Pi users.

This distribution has undergone intense testing by the manufacturer and community. Raspberry Pi’s are exceptionally stable, especially in black-box applications.

BalenaEtcher

This software is used on a Linux, Windows or macOS computer to flash the Raspberry Pi operating systems to microSD cards.

Scripts

Adding and Removing Audio Files

A Python script will be created that can copy all .WAV files from an inserted flash drive to local storage (SD card, USB flash drive, external HDD), while leaving the originals untouched.  

Ideally, the script would run the script at the startup of the Raspberry Pi.  

Optionally, the Raspberry Pi can actively convert found audio into the Ogg Vorbis audio format, which is open-source and royalty-free.

Writing RFID tags

The Python script provided by the RFID reader will be altered to program RFID tags following the required identification naming convention.  The only information that will be programmed to the tag will be a numerical representation of 000 - 999.

Linking Audio Files to RFID tags

When files are copied to the Raspberry Pi by script, they will be renamed to follow a basic naming convention, such as 001_audiotitle.ogg.  The three prepended numbers will be used as an identification tag, while any title after the underscore will be ignored and is used only for archival purposes.  The identification number of the audio file must match the identification number programmed on the RFID tag.

Reading RFID tags and Triggering Audio

When the RFID-embedded item is placed on the reader, it will trigger a script that reads the identification number.  The script will then search the designated archival folder to play the audio file that has the same identification number.

Security

• Disable the Serial interface
• Disable the Wireless/Bluetooth radio
• Ideally, the Raspberry Pi would be connected to the internet by hardwire as this would be harder to compromise.
• If the Raspberry Pi is not connected to the internet, it would not necessarily be required to update the operating system if all means of access are disabled (e.g. wireless, bluetooth, ethernet).
• If the Raspberry Pi is connected to the internet, automatic updates could still be disabled as long as the Raspberry Pi disables the Ethernet port unless it actively needs to connect to the internet.
• Disabling automatic system updates on the Pi would decrease security if network access is required, but it would improve reliability because there would be no chance of update or compatibility errors.
• Local SD card Audio backup/fallback system in event of USB hard drive failure.

Questions

Design

• How should the kiosk be physically secured to alleviate health-safety concerns?
• How should the RFID-embedded items be secured to the exhibit?
• What should the RFID-embedded items be?
  • Photo Album
  • Binder
  • Picture Frame
  • Envelope
  • 3D Printed Object
• Should we provide a 3.5mm jack to allow those who are hard of hearing to use headphones as desired?
• Should we provide a manual volume-control knob?
• Depending on the materials used to construct the cabinet, it may be recommended to create a plexiglass cut-out that would allow the RFID tags to work more reliably.
• Should there be a visual indicator that the RFID was scanned using RFID-powered LEDs?  These could also be embedded in the object.

Construction

• What are the museums concerns for archival and construction, such as material outgassing?

Mobility

• Depending on the needs of the museum and the material used for construction, the kiosk could be made to be mobile.
  • Should the kiosk be able to be disassembled and moved?
  • Would allowing the kiosk to be disassembled decrease the overall durability of the kiosk?
  • Would allowing the kiosk to be disassembled pose a health safety risk if re-assembled incorrectly, or incompletely?
  • Would allowing the kiosk to be disassembled compromise the overall integrity of the kiosk in a way that lessens its ability to withstand the potential stressors of repeated client use (e.g. leaning, pulling, standing)?
  • Due to the kiosk being freestanding, how should it be securely placed (e.g. bolted to a wall, floor, weighted bottom, anchored feet)?

Potential Additions

• Web Database for Audio Recordings
  • Front-end website for museum-goer playback of audio recordings
  • Back-end website for management of audio recordings
  • Online database for audio recordings
• Back-of-House Raspberry Pi System
  • Uploading audio recordings to a database
  • Acts as database server?
  • Allows backup copy to flash drive?
  • Allows updating of front-end kiosk
• Audio Recording Update System
  • Nightly Kiosk update to download new audio recordings
• Touch-screen interface
  • Information about open-source technologies and Raspberry Pi
  • Volume control
  • Lighted subtitles/transcript
  • Font-size control
• Scripting
  • Verify that copied audio files aren’t duplicates (libaudiodiff).
  • User interaction log

References

Field, S. (2012). Chapter 9: Beyond "Healing": Oral History, Trauma, and Regeneration. In Oral history, community and displacement: Imagining memories in post-apartheid South Africa (pp. 153-164). New York, NY: Palgrave Macmillan.

Hernandez, S. G., Genkova, A., Castañeda, Y., Alexander, S., & Hebert-Beirne, J. (2017). Oral Histories as Critical Qualitative Inquiry in Community Health Assessment. Health Education & Behavior, 44(5), 705-715. doi:10.1177/1090198117728546

Nyhan, J., & Flinn, A. (2018). Chapter 2 : Why Oral History? In Computation and the Humanities Towards an Oral History of Digital Humanities (pp. 21-34). Cham: Springer International Publishing.