The first iteration of the wicked problem OHD_MMP_0156
The eighth property of wicked problems is: ‘every wicked problem can be considered to be a symptom of another problem.’1 My theory is that the limited reuse of oral history is a symptom of the limited maintenance of access to oral history. Or at least the maintenance of oral history access is one of the contributing factors to the limited reuse of oral histories. Similarly, the limited maintenance of access to oral history is the symptom of a variety of other wicked problems.
However, the formulation of each version of the wicked problem of maintaining access to oral history is critically ‘essentially unique.2 This was evident in the different organisations that kindly hosted me.
For example, The British Library can trace its roots to the middle of the eighteenth century, and the National Trust has its origins at the end of the nineteenth century. In contrast, Archives at NCBS in Bangalore, India celebrated its fourth birthday while I was there on placement in 2023. The older institutions had more deep-rooted traditions, customs, and habits which seem to make it challenging for them to acclimatise to the digital era. Archives at NCBS, however, was born in the digital era and therefore is less hampered than the older institutions by customs and preexisting standards and curated collections that had been maintained as strictly analogue.
From a different perspective The British Library and Archives at NCBS are similar because they are both archives with oral history collections, while the National Trust is a heritage charity which has collected oral history that it then has deposited with the British Library.
In addition, each organisation has a different funding streams and staffing set-up all off which will, as I discuss later, contribute to how the access to oral history can be maintained. And if we look at the unique situation of Seaton Delaval Hall we get a different picture due to the Hall being a visitor, or tourist, site, and therefore the staff also need to manage a wider infrastructure including, a cafe, toilets, a garden, a shop, a carpark, and a wider voluntary workforce. I recognise that certain spaces in The British Library are also open to the public, but unlike the Hall the General Manager and the Collections staff are not pointing visitors to the correct parking spots. The Library has particular staff to carry out these jobs.
There are many things which contribute to the wicked problem of maintaining access to oral histories. This is evident in two similar talks I gave, where I explained the different areas of the system that grants access to oral histories.3 One was presented in the summer of 2023 and the other in the spring of 2024. I decided to drop ‘value’ from this graphic because value was something that was very specific to my work with the National Trust, due to their collection policy not covering the recording of oral histories or any other type of intangible heritage. However, what remained were the two areas – the maintenance of the expectations of access and the barriers to maintenance – which I believe can work as general starting points to push people to consider the maintenance of oral history more.
Doug Boyd finishes his chapter in the book, Oral history and Digital Humanities, with an anecdote about his daughter coming across a box of cassettes and not recognising the analogue media. After Boyd explains what the tapes are and how they were used, she simply replies ‘I just want to click on it to listen.’4 This is a useful insight into how developments in technology have affected expectations of access. Where before you had to go to a record shop to buy a CD, cassette, or Vinyl; now, nearly all music is available to you via the comfort of your phone. Similarly, in previous decades you had to travel to an archive to access recordings, now oral histories and other archival material are expected to be available online (especially after the Covid-19 pandemic). Digital technologies have created the idea that access to data, information, and creative content should be instant, which has also resulted in the structure of oral history repositories moving from shelves to servers.
This move from shelves to servers is a form of adaptive maintenance. Adaptive maintenance is one of the four basic types of maintenance you find in software engineering, with the others being: corrective, perfective, and preventive.5 T These are mainly used within the field of software engineering but I have found it somewhat applies to all systems including storage and retrieval systems like archives. Adaptive maintenance is changing a structure to make it fit an environment which is continuously changing.6 The move from shelves to server within the context of oral history repositories is adaptive maintenance because it is adapting to meet the society’s new expectations of access – “I just want to click on it to listen.” The most common adaptive maintenance is the digitisation of oral histories.
The digitisation of oral history consists of making a digital copy of analogue recordings. It is however, not a one time thing as my audit of National Trust recordings which had not yet been catalogue by the British Library demonstrated.7 The audit revealed a wide array of formats including CDs, cassettes, MiniDiscs, and a digital file type I did not recognise, which might have to be converted in order to be accessed. Sadly, the large digitisation project at the British Library, Unlocking our Sound History or UOSH, had just finished,The digitisation of oral history consists of making a digital copy of analogue recordings. It is however, not a one time thing as my audit of National Trust recordings which had not yet been catalogue by the British Library demonstrated.7 The audit revealed a wide array of formats including CDs, cassettes, MiniDiscs, and a digital file type I did not recognise, which might have to be converted in order to be accessed. Sadly, the large digitisation project at the British Library, Unlocking our Sound History or UOSH, had just finished, and there was no longer capacity in that project to address the challenges of converting the range of mixed media that is the Trust’s audio collection.9 and there was no longer capacity in that project to address the challenges of converting the range of mixed media that is the Trust’s audio collection.
OHD_COL_0262
Alongside the digitisation of analogue material the entire structure surrounding the material, the catalogue and the storage location, also needs to become digital, leaving carbon paper and index cards behind. Rob Perks writes about the overall challenges involved in his paper, ‘Messiah with the Microphone’. During his time at National Life Stories, the oral history charity that is situated within the British Library, Perks notes the ‘painful and protracted’ move from analogue to digital at the British Library. He writes about the acquisition of a digital management system, the development of a new accessioning system, and the training needs required to ensure staff gained sufficient skills and knowledge to implement the new digital content correctly.10
OHD_COL_0272 Collection of photographs of the National Trust Collections’ forms OHD_COL_0271 Collection of photographs of The British Library
The adaptive maintenance does not remove the existence of analogue material or systems. Digitisation rarely replaces analogue systems completely and the rate of adoption will differ radically between institutions and organisations. This can be referred to as the ‘digital divide.’ In the early two-thousands the digital divide primarily referred to the access to hardware, software, and the internet, more recently it has been extended to also consider people’s digital skills, so the ability to use technology, and the frequency they use it, how much it is integrated into their lives.11
This uneven adoption of digital systems often leads to a hybrid system consisting of analogue and digital components, as I saw in the office of National Life Stories.12 Nor does this adaptive maintenance eradicate issues experienced with analogue material such as ‘folios fallen behind books stack.’13 New digital issues are created instead following a basic rule of invention and development – ‘To invent the sailing ship or steamer is to invent the shipwreck.’14 Therefore, after digitisation archivists have to deal with fallen analogue folios and the digital equivalent – dead links.15
In addition to dead links during my research period there were also multiple IT incidents at the National Trust,16 a cyber attack at The British Library,17 the occasionally websites that were just not working,18 and technology becoming obsolete.19
This is not me arguing against the move from shelves to servers, after all my cat also weed on my PhD archive, causing me to throw some archival material away.20 This was not a great loss though because I had already digitised the material. I show these examples of breakdowns to highlight the perils of any structure digital or analogue and how these need to be considered equally. Digital technology is not magic but simply another form of a structure and all structures need maintenance.
The change in society’s expectation of access – “I just want to click on it to listen” – has led to a mass transformation of the oral history structures into hybrid systems with new additional forms of maintenance work. In addition, we also need to consider how this change in expectations has not only influenced the systems which create access but also the feelings we have around access. As Almut Leh writes, ‘the digital representation of original documents has considerably promoted their dissemination and thus their visibility’ and thus are more easily available and can be ingested and interpreted by machines on a mass scale.21 Outside of oral history, this increase in access has had significant consequences on the ethics, laws, and guidelines which surround data management, which in turn have affected how oral histories can be accessed.
However, these laws were not explicitly made for oral histories and so how they should be applied is a little confusing. Those who maintain the access to oral histories now also need to rethink how they classify and handle material. Through my research, specifically the work I did during my placements, I identified three areas which have been affected by the changing expectations of access: ownership, personal information, and sensitive content.
Ownership
OHD_COL_0262
The ownership of oral history and materials in general has changed a lot over the decades as the internet has made sharing and copying material a lot easier. My copyright audit of the National Trust’s sound collection clearly illustrated a history of oral history and copyright.22
For the time being, in the UK and most of the West an oral history is currently understood as a recorded performance, and therefore the performance, i.e. those speaking on the recording hold copyright over their voices. If a recording is to be used those recorded must sign their copyright over to the relevant party. Importantly, this rule of copyright goes for every recorded oral history, even those that were recorded before this law came in, which is why I did the copyright audit at the British Library and why they then created a PhD placement for someone to help the Trust develop a workflow that could be used to obtain copyright or confirm the oral history is an orphan work.
A work is considered ‘orphaned’ when the original creator is not contactable or not known. In the case of oral history a recording becomes an orphan work when the speaker was born 120 years ago (some organisations, like the British Library, the cut-off point is 100 years instead of 120.) In some cases however, it might be preferable to try to contact the next of kin in cases where third parties are mentioned in the recording.
The collection of personal information has become increasingly controversial with the rise of social media. The Facebook-Cambridge Analytica data scandal in 2016 led to the creation of the General Data Protection Regulation (GDPR), which was originally implemented by the EU in 2018 and later adopted by the UK after its formal exit from the EU in 2020. Many of the other data protection laws around the world have either been modelled on GDPR or the EU’s Data Protection Directive of 1995, such as the Personal Information Protection and Electronic Documents Act in Canada. There are data protection laws, such as the Digital Personal Data Protection Act in India, which only cover digital data. GDPR however applies to all data which includes oral history.
Oral histories generally collect two forms of personal data; the personal information of the interviewee on the permission forms, and the interview recording itself. The former is easy to handle under GDPR because what data is collected is clearly understood as personal data: name, address, contact information etc. When I was designing the Research Room at Seaton Delaval Hall all the forms I designed had to follow the same GDPR guidelines.
However, data protection laws like GDPR were not designed for oral history or any other archival material which can be considered an ego-document. This material therefore does not fit GDPR’s framework, as what is considered ‘personal data’ and what is considered important historical knowledge of public interest is subjective. Data protection laws do however give guidelines on how to handle personal information even if what you are handling does not necessarily fit with the particulars of the regulation.
OHD_RPT_0213
One of the outputs created during my final Masters project with Seaton Delaval Hall was a collection of stories my group found on Facebook.23 Now I know this is a blatant data protection violation and I made sure the copy we gave to the Hall was destroyed.
Throughout this project data protection never took a central position in the work I did during my placements, but it was always present. This continuous awareness of data protection, as annoying and restrictive as it sometimes is, did affect my designing in a positive manner.
The general advice I started giving people when asked about data protection, is admittedly very basic but I believe it captured the general aim of the regulation.
Make sure the data is kept in a safe place and access to this storage is monitor and recorded.
Make sure to be transparent with the owner of the data, in terms of where it will be storage and who will use it. And make sure it easy for them to ask questions about their data and have it removed if necessary.
GDPR and other data protection regulations made me consider how the collected personal data requires a particular storage location. This realisation then affected how I understood the wider processes of storage and access systems, like the Research Room. The requirement that things should be transparent changed the entire tone of the structures I was building. They had to be open and importantly easy to understand.
Sensitive content
The third area of ethics is the management of sensitive content. This is not particularly linked to the rise of the internet as morality outcries and the management of morally dubious or harmful content has always existed. However, due to the ease of access through the internet there are people who worry about children or vulnerable adults coming across upsetting, inappropriate or ‘triggering’ material. It has therefore become common to conduct sensitivity checks on archival material.
The particular sensitivity check I made for Archives at NCBS was specifically designed to be quick. Archives at NCBS generally follow the archival principle of ‘More Product, Less Process’ outlined by Mark Greene and Dennis Meissner, so speed was key.24 I therefore created a green light/red light system which could be used to quickly show what is categorised as sensitive or not. There were two perspectives on sensitivity: material which could harm the donor, their kin, or close associates, and material which could harm the researcher accessing the material. For the former I created a flowchart and for the latter I simply had a list of sensitive and non-sensitive material.
The two versions of the "Sensitivity Check Workflow" document contain both similarities and key differences. Below is a comparison:
### Common Elements: 1. **Purpose and Definitions**: - Both documents emphasize that the sensitivity check is designed to identify sensitive material in archival collections. - Definitions of sensitive and non-sensitive material are present in both versions, including red and green label guidelines.
2. **Examples of Sensitive and Non-Sensitive Content**: - Both versions classify sensitive material under categories like personal information, medical records, slander, and cultural sensitivity. - Non-sensitive materials include public information, certain academic records, and older data (30+ years).
3. **Cultural Sensitivity**: - Both documents stress the importance of contextualizing culturally insensitive materials.
---
### Key Differences: 1. **Version Naming**: - **V1 (February 16, 2023)** is simpler and less structured than **V2 (February 28, 2023)**.
2. **Flowchart in V2**: - V2 introduces a **flowchart** to guide decisions on sensitivity labeling, making the process more visual and systematic. V1 lacks this tool.
3. **Additional Context in V2**: - V2 provides more detail on the steps for contextualizing material during sensitivity checks, including advisories and catalog descriptions for sensitive items.
4. **Expanded Categories in V2**: - V2 introduces a **second part** to the sensitivity check, focusing on content that could harm visitors (e.g., offensive or triggering materials). - New examples of sensitive materials in V2 include **explicit sexual content**, **graphic violence**, and more nuanced rules for cultural insensitivity.
5. **Inclusion of Misinformation in V2**: - V2 explicitly mentions misinformation and provides guidelines for its contextualization, while V1 does not address this.
6. **Terminology and Philosophy**: - V2 describes the sensitivity check as "prescriptive" and distinct from the Archive's broader philosophy on access and sensitive material handling. - It emphasizes providing advisories and detailed catalog descriptions, even for embargoed material.
---
### Recommendations: - **Flowchart Addition**: V2's flowchart provides a clear decision-making aid and could be helpful if added to earlier workflows. - **Misinformation Context**: Including misinformation handling, as in V2, could enhance sensitivity processes. - **Part Two Sensitivity Checks**: V2 expands the scope to visitor safety, which could be valuable for archives focused on inclusivity. - **Consistency**: If you prefer the concise nature of V1, consider integrating V2's updates selectively to avoid complexity.
Let me know if you'd like further analysis or edits!
The purpose of this quick sensitivity check was to get material available quickly and then later do a more thorough sensitivity check. This is quite different to the approach the staff at Seaton Delaval Hall took when considering how they would manage the sensitivity of their oral histories. They decided to create a group of staff members that would work together as a form of ethics board. This again shows how different versions of the wicked problem will require different approaches.
Out of the three areas I have discuss sensitivity is the most nebulous and dynamic, as it is mostly dictated by the opinions of society rather than any rules or regulations, although laws such as the Online Safety bill in the UK might change this. Nevertheless all three areas of ethics might look very different in the future and there might be even new areas of ethics that will influence the access to oral histories – you never know what the future has in store.
This is why such documents and processes, like a take-down policy, are so very important. A take down policy exists exactly because things change. The take-down policy I made for Archives at NCBS emphasises change and underlines the importance of taking things on a case-by-case basis.25
ChatGPT Comparison of OHD_RPT_0249 - Archives at NCBS Takedown policy [generated 06 / 01 / 2025]
Here are the key differences between the three versions of the "Archives at NCBS Takedown Policy" documents:
---
### **1. Focus and Scope**: - **Version 1 (V1)**: - Primarily describes the "Notice and Takedown" policy for handling complaints about online material. - Focuses on specific scenarios like copyright, personal data, defamation, or errors in publishing. - Emphasizes removal, negotiation, and possible outcomes (removal, replacement, or unchanged return).
- **Version 2 (V2)**: - Adds historical context about "Notice and Takedown" policies and their evolution, referencing laws like the DMCA and "Right to Be Forgotten." - Discusses more detailed review processes for different complaint types, including copyright, personal data, and culturally sensitive material. - Introduces the concept of negotiating balances between public access and individual rights, recommending a panel review for complaints.
- **Version 3 (V3)**: - Expands the policy to include alterations, such as metadata edits, and internal staff requests for redaction or changes. - Formalizes a structured, multi-stage process: "Request and Receipt," "Review and Report," "Negotiate and Amend," and "Agree and Action." - Addresses internal accountability and archiving of all change requests and resolutions.
---
### **2. Detail and Complexity**: - **V1** is simpler and focuses on immediate actions for handling takedown requests. - **V2** introduces detailed sections on different complaint types (e.g., copyright, defamation) and historical/legal background. - **V3** includes more granular procedural steps and additional types of requests (e.g., metadata changes, internal staff flags).
---
### **3. Handling of Specific Cases**: - **V1**: - Describes removing material online temporarily during review and emphasizes swift resolutions. - No mention of internal requests or metadata changes.
- **V2**: - Begins discussing ethical considerations (e.g., historical preservation vs. privacy). - Suggests negotiating outcomes but does not formalize a multi-step resolution framework.
- **V3**: - Introduces new solutions like redactions, Google search removals, and partial removal. - Outlines separate approaches for different requesters (e.g., associated individuals, staff, concerned citizens). - Encourages thorough documentation and archiving of the resolution process for transparency.
---
### **4. Tone and Guiding Principles**: - **V1** has a reactive tone, focusing on removing material until issues are resolved. - **V2** balances proactive and reactive measures, embedding ethical and historical considerations. - **V3** adopts a proactive, systematic, and inclusive tone, emphasizing negotiation, transparency, and detailed procedural guidance.
---
### **5. Structural Changes**: - **V3** introduces a more formalized four-stage process: - Receipt of requests, - Review and generation of a report, - Negotiation (if applicable), - Final action and documentation.
- **V2** only outlines the need for case-by-case handling without explicitly structuring the process.
---
### Summary: - **V1** is a straightforward guide to takedown complaints. - **V2** expands scope, adding historical context, and ethical considerations. - **V3** evolves into a comprehensive "Takedown and Alterations" policy, emphasizing procedural rigor, inclusivity, and accountability.
Let me know if you'd like more detailed comparisons or additional analysis!
Things will change. Even during the four years of my project technology has changed a lot. In November 2020 I wrote a blog post about a computer being able to write my PhD by ingesting the archive I was creating.26 Two years later I could actually test this theory with ChatGPT. I do not wish to test it again now, although I did have a play around with summarising interviews.27 AI has also complicated copyright and data protection with many of the large tech companies developing AI being accused of scraping material from the internet without asking permission of the creators. As a result the tech companies are facing several class action lawsuits and the outcome of the lawsuits could again affect how we share information. However, at this point in time many of the archives will have already been scraped to be used in large datasets for training large language models.
Me testing if an AI chat bot could write my PhD. It could not.
digital ; AI ; ChatGPT
OHD_Archive
Index entry for OHD_WRT_0241
OHD_GRP_0176 My PhD in a graphic. I did a seminar based off this image (OHD_WRT_0202)
In a writing workshop, I was assigned the task of visually representing my PhD research.28 My resulting graphic illustrated the changes that occur after an oral history interview is conducted. Although there are certain parts missing from this graphic, such as data scandals, it shows the core challenge of maintaining access to oral history is ensuring the system fits within society’s changing expectations of access. This necessitates ongoing updates and adjustments – a form of continuous maintenance. However, as is demonstrated by the dead links, the missing copyright, and the existence of take-down policies, things go wrong and this maintenance is not always carried out, thereby jeopardising the long-term accessibility of oral histories.
Every formulation of a wicked problem is unique, even if the topic is the same. Each version of the wicked problem of maintaining access to oral history will differ in what type of adaptive maintenance is required: digitising material, acquiring copyright, updating the catalogue, doing a sensitivity check. It will also differ in what hinders the completion of this adaptive maintenance and maintenance in general, causing the access to oral histories to fail due to deadlinks or incomplete paperwork etc.. However, there are recurring barriers – the usual suspects – which appear across versions of this wicked problem obstructing the completion of maintenance (adaptive or corrective): limited access, limited time, and limited financial means.
Limited Access
I did an interview with an archivist who had built the archive at the contemporary art gallery BALTIC in Gateshead when it first opened in the early two-thousands.30 He talked me through the various iterations of the online archive and the processes behind them. It is a story I became increasingly familiar with throughout my research.
SDH_AI_001 Interview with BALTIC archivist
Like the archivist at BALTIC, when oral historians first started to develop technologies to improve the access to oral histories at the end of the twentieth century and the start of the twenty-first, there was no software that could do what the oral historians wanted it to do so they made their own.31 However, the oral historians had not considered how their technologies would fail and therefore did not maintain them leading to their eventual demise. Oral historians like many others started to look outside of their field, outsourcing and purchasing off-the-peg material.
The consequences of outsourcing were sadly demonstrated by the cyber attack on The British Library. In a blog post Simon Bowie, who had previously worked at the British Library, writes how the Library was not replacing IT staff when they left and continuously outsourcing parts of the IT systems.32 This led to the IT maintenance being overstretched, leaving the system vulnerable to attacks.33
When software is outsourced or brought off-the-peg the system is a black-box which obstructs the ability to maintain. Maintenance is then no longer in the hands of the archivists or those with similar roles, but instead with middle men who might not be as invested. The capacity to maintain is therefore restricted due to limited access.
Google Gemini Summary of OHD_WRT_0135 - The Journey of Oral History Technologies [generated 06 / 01 / 2025]
This text explores the challenges of developing and implementing digital oral history technologies, drawing parallels to the inaccurate predictions of future technologies in popular culture (like Back to the Future Part II).
Key Arguments:
Early Optimism and Subsequent Challenges: Early projects like Project Jukebox demonstrated initial optimism about the potential of technology to revolutionize oral history. However, these projects often faced unforeseen challenges, including:
- Technological Limitations: Rapid technological advancements quickly rendered early solutions obsolete, requiring constant maintenance and updates. - Funding Constraints: Limited resources often hindered the development and maintenance of sophisticated technologies. - Design Flaws: Overemphasis on aesthetics or specific user groups can lead to exclusion and limited usability.
The Importance of Human Factors: The text emphasizes the importance of human factors, such as:
- Digital Divide: Acknowledging and addressing the digital divide in terms of access, skills, and usage. - User-Centered Design: Considering the diverse needs and preferences of different user groups, including oral historians, archivists, and the general public. - Ethical Considerations: Addressing issues of data privacy, digital inequality, and the environmental impact of digital technologies.
The Need for a More Human-Centered Approach: The text argues for a more human-centered approach to digital archives, emphasizing the importance of:
- Collaboration and Communication: Fostering effective communication and collaboration between oral historians, archivists, technologists, and community members. - Learning from Past Failures: Analyzing past projects to identify and address recurring challenges. - Embracing a "DIY" Approach: Exploring more pragmatic and sustainable solutions that leverage existing tools and skills.
Overall:
The text provides a critical analysis of the challenges and opportunities presented by digital technologies in the context of oral history archiving. It emphasizes the need for a nuanced and collaborative approach that acknowledges the limitations of technology and prioritizes human needs and values.
Google Gemini Summary of OHD_WRT_0172 - Chpt. 01 History of oral history tech [generated 06 / 01 / 2025]
This chapter explores the attempts to find a technological solution to the problem of archived oral history recordings. It discusses the challenges faced by oral historians and sound archivists in preserving oral histories and the failures of past technological solutions like TAPE, Project Jukebox, VOAHA, Interclipper, and Stories Matters. The chapter also analyses the reasons for these failures, including a focus on aesthetics over functionality, a lack of consideration for long-term preservation, and a reliance on new technologies that are not always sustainable.
Here are the key points from the chapter:
Oral historians and sound archivists have different needs for technology. Oral historians want technology that helps them focus on the orality of recordings and contextualize them, while sound archivists prioritize long-term preservation.
Many past attempts at creating oral history technology have failed because they did not address the needs of both groups. For example, they focused on creating beautiful interfaces that were difficult to maintain or used technologies that became obsolete.
There is a need for a more sustainable approach to oral history technology that considers both functionality and long-term preservation.
The chapter concludes by calling for a more DIY approach to oral history technology, using established technologies and focusing on creating solutions that are easy to maintain and will last over time.
Limited Time
‘Between indulging in delicious food and gandering around the stunning campus of the National Centre of Biological Sciences (NCBS), I sit at a hot desk in the basement where the Archives at NCBS is housed. The term ‘basement’ is slightly misleading as the sun shines through windows which face a sunken outdoor amphitheatre, where I can watch paradise birds flirt with each other in the trees. When not distracted by birds or food, my attention might be drawn away from my work by one of the eleven other people working in the Archive. It is the loudest archive I have ever been in, even when I discount the constant humming of the air conditioning. It is also the most welcoming workplace I have ever worked in. The Archives at NCBS is a hub of multidisciplinary folk, all coming together to build this archive, which is still very much in its infancy, celebrating its fourth birthday on this month. Therefore, a lot of the work is focused on growing the archive, with some team members creating a digital catalogue, others expanding the collection, and many involved in developing the various work flows necessary to keep an archive running.’
Archives at NCBS is in a completely different position than the other organisations I did placements with because it is so young – only five years at the time of writing. As I wrote in a blog post on my first month there ‘a lot of the work is focused on growing the archive, with some team members creating a digital catalogue, others expanding the collection, and many involved in developing the various work flows necessary to keep an archive running.’34 Although I spent most of my time at Archives at NCBS developing their takedown policy and their sensitivity check, I also observed the work environment. I was given an excellent opportunity to do this when I was asked to facilitate their annual away-day. The larger questions to be explored during the away day were:
The second activity of the day the staff had to brainstorm around the four activities of the archive: collect, educate, research, and engage. I think the final outcome of this activity demonstrates how much happens in a space like Archives at NCBS.36 However, within these different sections there is even more work as shown in the workflow I created, which maps out the process of collecting and then making something publicly accessible.37 This workflow does notably not show the maintenance tasks that need to occur when something is finally in the archive.
As part of my final output I brought together my observations of the work environment at Archives at NCBS, using the framework Mierle Laderman Ukeles offers in her, Manifesto for Maintenance Art, 1969!, that the world is made of a development system and a maintenance system.38 I concluded there are two systems that operate under the roof of Archives at NCBS.
OHD_GRP_0261 First half of the graphic ‘What is Archives at NCBS?’
My solutions to how this might be handled was partially based on the final activity of the away day which was an exercise where the staff had to write what they would like to stop, start, and continue in the workplace.39 Additional ideas came from my experience of working at Seaton Delaval Hall, because I found there to be some similarities in their respective work environments. The Hall is a functioning heritage site and so there are other activities to do besides managing the collection and acquiring new material, similar to Archives at NCBS. They also have a similar number of staff, so it felt appropriate to bring some activities across, such as the updates of individual activities during the weekly meeting.
OHD_GRP_0261 Second half of the graphic ‘What is Archives at NCBS?’
What my experience at Archives at NCBS and my other placements revealed was how much work is needed to keep these organisations operational. On a foundational level, keeping the organisation working is important to maintaining access to oral histories, however on a more detailed level the smaller tasks included in maintaining access to oral histories, such as updating copyright permissions, risk being sidelined to create time for more foundational activities to the organisation’s wider existence. This is what I mean by limited time obstructing the capacity to maintain access to oral histories, like the necessary adaptive maintenance discussed in the previous section.
Limited Funding
Chasing funding
We went to three different heritage sites of varying status and every single one of them mentioned funding many, many, many times.
Every organisation will have different funding sources. In the UK the largest funding body is the National Lottery Heritage Fund (NLHF). The old archivist at BALTIC mentioned they had to wait for funding to update their system, which eventually came from NLHF.40 Seaton Delaval Hall got funding from NLHF for their The Curtain Rises project, which contained a collaboration with Northumbria University called Rising Stars. This collaboration included the final project for my Masters in Multidisciplinary Innovation which led to the development of this PhD.41
NLHF does seem to have a preference for particular projects. In a screen recording I used during a talk, I searched the NLHF website for projects that were doing ‘development,’ ‘restoration,’ or ‘maintenance.’42 This search revealed a focus on restoration and development. Restoration is only necessary when maintenance fails. In How Buildings Learn Stuart Brand quotes John Ruskin, ‘Take proper care of your monuments, and you will not need to restore them. Watch an old building with an anxious care, guard as best you may, and at any cost, from every influence of dilapidation.’43
When I did a casual audit of oral history projects funded by the NLHF, there was little or no evidence of the recordings being archived suggesting energy and money was put towards recording over maintenance.44
There are cases of maintenance activity being funded but this is primarily digitisation. The British Library received funding for, Unlocking our Sound Heritage, which was also partially funded by NLHF.45 Archives at NCBS received a grant for Arcadia, a family charitable foundation, ‘to collect, preserve and make available online endangered cultural artefacts.’46
Funding will vary from organisation to organisation, and I have not been able to do a full investigation into every funding body in the world to fully prove whether maintenance is definitively funded less than development and restoration. However, maintenance is an undervalued activity in society, which among many things pushed Ukeles to write her Manifesto for Maintenance Art 1969!47 A lot of this undervaluing comes from the invisible nature of maintenance48 which I discussed in a talk I did – ‘The Unseen: Maintenance Labour on Heritage Sites.’49
Google Gemini Summary of OHD_PRS_0120 - The Unseen: Maintenance Labour on Heritage Sites [generated 06 / 01 / 2025]
Key Arguments:
The Invisibility of Maintenance: Maintenance work is often invisible and undervalued, taken for granted until things go wrong.
The Impact of This Invisibility: - Underfunding: Maintenance projects receive less funding than development or restoration projects, as they lack the "wow factor" and are seen as less impactful. - Exploitation of Volunteer Labor: The undervaluation of maintenance work leads to the increased reliance on unpaid volunteers, potentially displacing paid professionals. - Unsustainable Practices: A focus on development and restoration can lead to unsustainable practices that require constant and costly interventions.
The Importance of a Maintenance Perspective: Recognizing the crucial role of maintenance in ensuring the long-term sustainability of heritage sites. Designing and implementing solutions that are adaptable, maintainable, and integrate seamlessly into existing maintenance routines.
Implications for the Author's PhD Research: Acknowledging the realities of maintenance work at Seaton Delaval Hall, including resource constraints and the role of volunteers. Designing a solution that is adaptable and easy to maintain by the Hall's staff. Collaborating closely with Hall staff to understand their needs and integrate the project into their existing workflows.
Key Concepts:
Maintenance as Invisible Labor: The work of maintaining a heritage site is often invisible and undervalued, taken for granted until things break down.
The Development vs. Maintenance Dichotomy: Society tends to prioritize development and restoration over maintenance, leading to underfunding and underappreciation of maintenance work.
The Importance of Sustainability: Designing solutions that are sustainable in the long term requires careful consideration of maintenance needs, including resource constraints and the availability of skilled labor.
What I concluded from my placements is that adaptive maintenance is not done or completed because the capacity for organisations to do is restricted by a variety of factors. This can include: an inability to access the system which needs to be maintained, not enough time in the day, and not enough money.
One of the bonus outputs I created for Archives at NCBS was a document which outlined what needed to be considered when choosing a technology to help access oral history recordings. In a strange manner it can be considered the TL;DR for this chapter as it factors in the various elements explored in this section.
Tech options for making oral history recordings accessible
V1. January 30 2023
Hannah James Louwerse, Archives at NCBS
Making oral history recordings accessible to people has been infamously difficult, with the oral historian Michael Frisch referring to the issue as “oral history’s deep dark secret”. There have been many attempts to solve this problem with some being more successful than others. By analysing the history of oral history technologies one can see how using technology to access to oral history recordings depends on three factors: maintenance, ethics, and user-friendliness. This short report will go through each of these factors bringing examples of oral history technologies to explain what you should look for when seeking a solution to putting oral history recordings online
1. Maintenance
Maintenance is often the biggest killer of solutions to the deep dark secret of oral history. Maintenance depends on a continuous supply of money and labour, which is not always easy to get hold of, especially within grant cycles. It is therefore essential to think about the maintenance necessary to sustain a technology which allows access to oral history recordings. How you do this depends on the source of the technology and how it was developed.
1.1 Tailor-made, in house development and maintenance
Creating your own digital oral history archiving system allows it to be perfectly tailor to your collections needs. However, it also means the maintenance of this system is solely in your hands, which can be very risky, especially when working within grant cycles. Projects like the Visual Oral/Aural History Archive (VOAHA) created by Sherna Berger Gluck at California State University, Long Beach and Civil Rights Movement in Kentucky Oral History Project Digital Media Database developed by Doug Boyd built tailor-made technologies specifically for their existing oral history collections, either developing the technology themselves or hired someone to do it for them. At the time they were the height of technology, but when the money ran out there was none left to maintain the archives/databases. Both VOAHA and Boyd’s Civil Rights Movement in Kentucky Oral History Project Digital Media Database were “digitally abandoned” and left vulnerable to inevitable technical obsolescence and online hackers (Boyd and Larson, 2014, p. 7; Boyd, 2014, p. 90). In the end the two projects were absorbed by their respective universities’ libraries.
1.2 Use existing specialist oral history software
By using specialist oral history software, the maintenance is no longer your responsibility, which is both a risk and a benefit. The benefit is how it is a cheaper option in comparison to hiring someone full time to take care of the technology. But the risk is that the software developer stops maintaining the software, which is what happen in the case of Stories Matter, an oral history software developed by the Centre of Oral History and Digital Storytelling at Concordia University and a software engineer from Kamicode software (High, 2010; Jessee, Zembrycki, and High, 2011). The Kamicode website still has a page on Stories Matter, but the software is not downloadable. The reason for this is unclear, however it is easy to imagine the maintaining of such niche software is unlikely to be a high priority for a software company.
A more successful example of specialist oral history software is Oral History Metadata Synchronizer (OHMS), developed by Doug Boyd after his reflections on Civil Rights Movement in Kentucky Oral History Project Digital Media Database. OHMS has been in existence for some years and is a popular way for oral history projects and archives to organise their oral history metadata and link the video/audio file to a searchable text. Unlike Stories Matter, OHMS is developed and maintained by people who are interested in oral history and use it for their own projects. Maintaining OHMS is therefore in their own interest.
1.3 Use existing mainstream third party platforms
Another cheaper option is using more mainstream platforms such as Soundcloud or Spotify. These are less niche technologies and therefore do not have the benefits more specialised software has, but the maintenance is pretty much guaranteed since these platforms are universally used. Certain projects have created Spotify playlists and other have Soundcloud versions of their recordings alongside the original files in the brick-and-mortar archive.
2. Ethics
The internet is an ethical nightmare and putting someone’s personal story online in an ethical manner is not an easy task. The starting point will always be clear communication to the interviewees on how people will be able to access their recording, and thorough paperwork which accompanies the recording. Following this there are a couple of other things people have done to support the ethical handling of oral history recordings.
2.1 Extracts
The simplest of ethical practices is to only make certain extracts available online. This means you can avoid putting online more sensitive information but still give an example to the archive visitor of the kind of content the oral history holds. If the archive visitor wishes to hear more, they can request the full recording via email. A possible consequence of this might be people only using the online extract and not bother enquiring any further because it is deemed as “too much effort.”
2.2 End user agreement
Archives like Trove and Centre for Brooklyn History have “end user agreements” the archive visitor must agree to before they are allowed access to the oral history recording. These end user agreements contain information on basic copyright and data rights, a disclaimer about the opinions expressed in the recording, and outline the archive user’s obligations. These obligations include correctly citing the recording, adhering copyright law and data protection law. These end user agreements are a way for archives to hold users accountable in case of misuse or rights violations.
3. User-friendliness
People have a low tolerance of bad user-experience design. The software Interclipper, championed by Michael Frisch was reviewed during the development of Stories Matter and VOAHA and was deemed difficult to use in both instances (Jessee, Zembrycki, and High, 2011; Gluck, 2014). It no longer exists. OHMS offers both a backend metadata synchronizer and a viewer, the latter however is often left in favour of an in-house interface design. Project Jukebox developed by the University of Alaska in collaboration with Apple Computers Inc. in the 1990s, is still available online but still looks like it was made in the 90s, even though at the time it was described as “a fantastic jump into space age technology” (Lake, 1991, p. 30). It is therefore important the user experience and interface are updated as fashions and taste evolve across the wider internet.
Boyd, D.A. (2014) ““I Just Want to Click on It to Listen”: Oral History Archives, Orality, and Usability” in Oral History and Digital Humanities. pp. 77-96. Palgrave Macmillan: New York
Boyd, D.A. and Larson, M. (2014) “Introduction” in Oral History and Digital Humanities. pp. 1-16. Palgrave Macmillan: New York
Gluck, S.B. (2014) “Why do we call it oral history? Refocusing on orality/aurality in the digital age” in Oral History and Digital Humanities. pp. 35-52. Palgrave Macmillan, New York.
High, S. (2010) “Telling stories: A reflection on oral history and new media” in Oral History. 38(1), pp.101-112
Jessee, E., Zembrzycki, S. and High, S. (2011) “Stories Matter: Conceptual challenges in the development of oral history database building software” In Forum: Qualitative Social Research. 12(1)
Lake, G.L. (1991) “Project Jukebox: An Innovative Way to Access and Preserve Oral History Records” in Provenance, Journal of the Society of Georgia Archivists. 9(1), pp.24-41
Smith, S. (Oct 1991) “Project Jukebox: ‘We Are Digitizing Our Oral History Collection… and We’re Including a Database.’” at The Church Conference: Finding Our Way in the Communication Age. pp. 16 – 24. Anchorage, AK
Douglas A. Boyd, ‘”I Just Want to Click on It to Listen”: Oral History Archives, Orality, and Usability’ in Oral History and Digital Humanities, eds. Boyd and Larson (New York, Palgrave Macmillan, 2014), p. 77. ↩︎
E. Burton Swanson, ‘The dimensions of maintenance’ in Proceedings of the 2nd international conference on Software engineering, (1976), pp. 492-497. ↩︎
E. Burton Swanson, ‘The dimensions of maintenance’ in Proceedings of the 2nd international conference on Software engineering, (1976), p. 493. ↩︎
The National Archives, ‘British Library,’ on The National Archives, [Accessed 6th January 2025]. ↩︎
Robert Perks, ‘Messiah with the microphone? Oral historians, technology, and sound archives,’ in The Oxford Handbook of Oral History, eds. Ritchie, (Oxford, Oxford University Press, 2011), p. 326. ↩︎
Jan A. G. M. Van Dijk, ‘Digital divide: Impact of access,’ in The international encyclopedia of media effects, ed. P. Rössler, (John Wiley & Sons, 2017), pp.1-11. ↩︎
Almut Leh, ‘”The answer is 42” – When Algorithms Take over Digital Memory. Experiences with Artificial Intelligence in the Archive Deutsches Gedächtnis,’ in Von Menschen und Maschinen: Mensch-Maschine-Interaktionen in digitalen Kulturen, eds. Gerlek, Kissler, Mämecke, and Möbus, 1(1) (Hagen, Hagen University Press, 2022), p. 173. ↩︎
Mark Greene, and Dennis Meissner, ‘More product, less process: Revamping traditional archival processing,’ in The American Archivist, 68(2), (2005), pp. 208-263. ↩︎
The Sound and moving image catalogue was powered by SirsiDynix, a company in the United States which produces integrated library system software and associated services for libraries. ↩︎
Mierle Laderman Ukeles, ‘Maintenance Art Manifesto,’ in 100 Artists’ Manifestos, ed. Alex Danchev(London, Penguin Group, 2011), p. 382. ↩︎
Susan Leigh Star, ‘The ethnography of infrastructure’ in American behavioral scientist, 43(3), (1999), p. 382., Stephen Graham, and Nigel Thrift. ‘Out of order: Understanding repair and maintenance,’ in Theory, culture & society 24(3), (2007), p. 8., Susan Leigh Star and Anselm Strauss, ‘Layers of silence, arenas of voice: The ecology of visible and invisible work’ in Computer supported cooperative work (CSCW), 8(1), (The Netherlands, Kluwer Academic Publishers, 1999), p. 16. ↩︎
