Abstract
This article addresses the intersection between platforms, their sociotechnical process of aging and memory practice, by focusing on local social platform OakdaleTalk and its use in reflecting on September 11, 2001. Founded in 1997, OakdaleTalk serves the town of Oakdale in New Jersey, a place significantly affected by the attack on the World Trade Center in 2001. Situated in literature on digital memory work, this analysis draws on interviews with 15 OakdaleTalk members who have used the platform to reflect on September 11th. Asking how users maintain particular meanings of this date while negotiating sociotechnical changes to the site over time – including the loss of posts from 2001 – it discusses how users perpetuate hyperlocal interpretation and describes how community members grapple with lost content. The article concludes by proposing a ‘preservation paradox’, an emerging memory practice under platformized media conditions describing a contradiction in user preservation attitudes and behaviors toward posts of memory-related significance.
Oakdale is a town of 25,000 in New Jersey, USA. 1 As a commuter suburb of New York City, Oakdale community members were significantly affected by the events of September 11, 2001. Three Oakdale residents died in the World Trade Center. 2 Many other lives were upended in ways large and small: residents witnessed the towers being struck, collapsing or smoldering; they entered and then fled Manhattan; they knew colleagues who passed away in the towers or the planes; they lost jobs and they changed their daily routines. On September 11th and in the weeks that followed, Oakdale community members came together to exchange information and to cope, both in person and through varying media forms. Residents swapped emails with far-flung friends, watched round-the-clock broadcasts of witness video and attempted to get through to family on jammed phone lines.
In the weeks after the attack, Oakdale residents also engaged in what Lohmeier and Pentzold (2014) call ‘mediated memory work’, or purposive memory-related practices like recording, storing and reminiscing that draw on media. Residents, for instance, left messages on posters at a local park overlooking Manhattan; they read the New York Times’ series ‘Portraits of Grief’, which chronicled the lives of those who died in the World Trade Center; they took pictures of vigils held in town. For Oakdale community members, mediated memory work also occurred through nascent digital channels, as many flocked to local social platform OakdaleTalk. The platform was launched in 1997 by Christopher and Mark Smith, two Oakdale brothers in their twenties. They organized it as a discussion forum-based site with one central forum, multiple topic threads and a screenname and avatar for each user. As evidence of its success, by December of 2001, the channel had over 1700 registered users and 55,000 messages, and was popular for discussing local issues. Perhaps unsurprisingly, OakdaleTalk became a gathering place for residents in September 11th’s aftermath. First, it was used to exchange practical information, whether neighbors’ whereabouts or how to help, and then became a site to cope, as users wrote their experiences of that day and visited to read others’ accounts.
While the emergency information on OakdaleTalk’s homepage would come down, and three permanent memorials would be established within the town’s limits in the ensuing years, OakdaleTalk would continue to be used as a site for digitally mediated memory work around September 11th. Among other threads, whether a recent town referendum or the prospects of the Mets’ baseball season, users post threads like ‘Remember our 9/11 victims today’, or ‘Where were you on September 11, 2001, 10 years ago’ on the attack’s anniversary. As the 15th anniversary of September 11th approached in 2016, one user also attempted to find the posts from September 11, 2001, to reflect on them with the OakdaleTalk community. But a software update had deleted all posts before 2007. As users soon found out, the posts from 2001 were gone.
Because of its use for reflection on an event of great memorial significance like September 11th, and its continued use over more than two decades, the case of OakdaleTalk offers a salient junction to understand mediated memory work on platforms over time. Although memory practice is essentially related to the passage of time, this phenomenon has not been empirically considered in the growing scholarly work on memory and platforms. This article argues that a time-sensitive approach to this subject is needed, in part because neither memory nor digital platforms are static entities. OakdaleTalk is a case in point. Some practices and features associated with the site have stayed the same: one of the founding brothers, Christopher, still serves as the site’s webmaster; it has continued to be a hub for diverse community discussion, whether local politics, lost dogs and what TV shows people are watching; and it remains a site where community members reflect on September 11th. But it has also significantly changed, whether in its user composition, as some move to other networks like Facebook or Nextdoor, or in its technical infrastructure, as it requires upgrades that alter the information available on the site. Moreover, because technical updates necessitated the removal of posts from early years of the platform, including posts from September 11, 2001, OakdaleTalk provides a unique glimpse into memory work on platforms when content is lost. As I will discuss later on, content loss is a common phenomenon for platforms as they age but a nearly unexplored scholarly topic in research on these sites and services.
By focusing on OakdaleTalk and its use in reflecting on September 11, 2001, this article addresses the intersection between platforms, their sociotechnical process of aging, and memory work. Situated in literature on digitally mediated memory while drawing attention to the need for time-sensitive study of platforms, this analysis is based on semi-structured interviews conducted in the summer of 2017 with 15 OakdaleTalk members who have used the platform to reflect on September 11th. Asking about the ways that users maintain particular meanings while negotiating sociotechnical changes to the site over time, I show how a core group of members uses the persistent presence of the platform to maintain local meaning of September 11th, meaning that OakdaleTalk users feel is not embraced in other mediated forms, whether Facebook or the New York Times. At the same time, this article shows the ways in which users grapple with the aging components of the site, including the loss of content from previous years. As will be discussed, users do not routinely engage with previous September 11th content but nonetheless identify this content as important for maintaining a community-level interpretation of the past. Observing this tension, this article concludes by describing a preservation paradox, an emerging memory practice under platformized media conditions.
Digital archives and memory maintenance
September 11th 2001, occurred at a time when the United States was embracing the possibilities of widespread internet use. As such, in the aftermath of the attack, websites, email and other channels mediated by internet infrastructure were used for both information access and emotional connection (Harmon, 2001). The possibilities of internet-based technologies were likewise explored for the significant memorial reflection that followed the attacks. The ability for anyone with an internet connection to circulate their experience of that day seemed to signal a paradigm shift in which individuals possessed newfound agency to shape the historical record (Collins, 2007). In the wake of September 11th, citizen-made digital content and submission of this content using the internet was a major focus of September 11th archival projects, whether the site wherewereyou.org, which chronicled people’s experiences when they first learned of the attacks, or the web-based September 11 Digital Archive now maintained by the Library of Congress. In comparison to the quickly ossified narrative of the attacks that privileged US nationalism in service of justifying military response (Sturken, 2018), digital repositories with open submission policies promised multivocality, or the representation of multiple experiences in the mediated mourning and historical memory of the event (Haskins, 2007; Jarvis, 2011). Moreover, these digital repositories stood in contrast with non-digital, non-networked media like television, the newspaper or personal photographs and diaries, which offered limited perspectives as well as limited public accessibility. Yet, it remained an open question in scholarly literature how the seeming benefits of networked digital archives could be maintained over time. How would the diverse yet hugely abundant amount of digital material be made into a ‘usable past’ by both historians and communities into the future (Haskins, 2007)? How could the multivocality engendered by open submission and open accessibility be productively maintained?
The question of how accumulated digital material would be maintained and productively used in the future is one that deserves revisiting, both in the broader setting of digital memory as well as in the context of OakdaleTalk. But first, digital memory deserves some introduction. Scholarship on digital memory explores the various ways that conceptions of the past are interwoven with digital media technology use (e.g. De Kosnik, 2016; Hoskins, 2018; Reading, 2014). Work on this concept generally argues that our relation to the past is transformed through networked digital media. In the past, mediated memory work interfaced with traditional mnemonic media, whether a diary or photographs, newspapers or home videos. Today, however, information about the past is seemingly abundant, accessible, multiple and shared amongst different social groups (Hoskins, 2011). Perhaps, somewhat paradoxically, networked digital media transforms our relation to the past because an ever increasing amount of digital information is produced in the present. With smartphone cameras and photo albums, automatically archived cloud email accounts, and social network activity preserved in shared feeds, information about the present moment is constantly in production–information which inevitably goes on to describe the past (ibid). In the context of important global events like September 11th, these digital devices serve as a means of communication and coping in the present while at the same time producing a historically and personally significant record about the past (Tandoc and Takahashi, 2017).
Because of these patterns of information production, storage and organization, digital memory scholarship has often analogized the internet–especially social media platforms – as archival, creating ‘living’ ‘digital’ archives (Ernst, 2012; Hoskins, 2011; Smit et al., 2017). However, metaphors of the archive in scholarship on digital memory often fail to address a key function of traditional archives: the long-term maintenance and use of this digital information. To revisit the question posed earlier in the context of September 11th archives, how will this accumulated digital material – this archive – be maintained over time? How will it be productively used in the future, whether for individuals, communities or historians? Moreover, as Kaun and Stiernstedt (2014) have noted, while many platforms might display archival features, they also differ from traditional archives by favoring recent content, pushing what is new or newly relevant to the top of a constantly refreshed page. How are digital technologies used to maintain particular meanings of the past when they insistently favor engagement with what is new? Finally, observations about the digital archives of September 11 noted that, compared to traditional media forms, digitality democratized the collective memory of the tragedy (Collins, 2007). At the same time, scholars have since argued that there is more interaction between digital and non-digital mediated memory than is otherwise assumed (Keightley and Pickering, 2014; Niemeyer and Keightley, 2020). How do diverse forms of mediation interact in the maintenance of memory?
In focusing on digital memory maintenance, I take a cue from recent science and technology studies scholarship, which has argued that analysts should shift their attention from the typically studied cycles of technological production and innovation to the more common but rarely discussed work of maintenance, repair and breakdown (Forelle, 2021; Russell and Vinsel, 2018). These scholars argue that this lens helps surface otherwise obfuscated relations while engendering a more holistic view of technological systems in general. Perhaps, the same could be said for the study of digital memory: a more holistic understanding of digital memory can be reached by studying not just the perpetual creation of this archive through digital media, but instead examining the ways that memory is maintained, repaired or forgotten using digital media. In this article, I explore this juncture through OakdaleTalk with the first of two research nodes: R1) How and why do OakdaleTalk users maintain September 11th memories on OakdaleTalk? Which memories are maintained?
Digital memory maintenance on platforms
The question of digital memory maintenance is especially important as social media platforms become the dominant mode by which the internet is experienced (Plantin et al., 2018). Emerging research on digital memory argues that these sites and services that host, organize and circulate user generated content (Gillespie, 2010) are increasingly involved in shaping and recalling the past. 3 Work in this realm builds on observations of the internet’s ability to act as an archive by showing the ways these archives are shaped by user practices like uploading, tagging and commenting, as well as by the platform’s algorithmic sorting of this posted content (Lee, 2020; Smit et al., 2017). Jacobsen and Beer (2021) phrase this as the ‘automatic’ sorting of memory, in which platform systems both define and resurface ‘memory’, or data posted in the past. This can be seen through numerous memory-specific features that have been incorporated on these sites, including the multi-platform Timehop, Facebook ‘Memories’, Instagram ‘Throwback’ and Snapchat ‘Memories’. To put it another way, this automatic sorting describes how platforms have built the maintenance of user memory into their DNA through a constant return to stored data: in essence, memory maintenance made algorithmic and automatic.
Asking about how digital memory is maintained on platforms also sheds light on the interwoven social and technical elements that shape the production and recall of the past. Crucially, in contrast with metaphors of the archive from digital memory scholarship, the lapse in the maintenance of stored data and the resultant data loss that OakdaleTalk experienced is not a rare phenomenon. Platforms shutter and lose data with an understated regularity (Corry, 2021a). This has included sites from the mid-1990s, like GeoCities; networks popular in the early 2000s, like Friendster; and more recent mobile platforms like Vine. Other sites have shorn off user content as their platforms have changed, due to technical, economic or legal pressure. MySpace lost 13 years of content in a botched server migration in 2019. Flickr deleted photos from free user accounts that contained more than one thousand images in 2019. Tumblr deleted all blogs containing adult content in 2018. Despite the ubiquity of platform data loss, and the recurring use of ‘the archive’ to describe digital environments, there is almost no literature on data loss’ impact on memory nor on the material work that keeps these memory repositories running. Furthermore, as Good (2013: 570) has pointed out in an analysis of Facebook’s use for personal memory, it is unknown how aware platform users are about the potential lapse in digital archival maintenance on platforms, and whether a ‘lack of public awareness exists about social media sites’ terms of agreement and how they might affect the future accessibility of the personal digital archive’. As I discuss in the next section, these questions about digital memory maintenance are addressed through a time-sensitive approach to OakdaleTalk.
Studying platforms over time
While this article focuses on memory maintenance and platforms, it also argues for the value in studying platforms as they age, obsolesce and evolve over time. It posits that time-sensitive platform study brings novel questions to research on platforms and memory and contributes to an understanding of the relationships between platforms and public life more generally. Emerging work has shown the promise of tracking platforms through time, whether Twitter (Baym and Burgess, 2020) or Craigslist (Lingel, 2020). These ‘platform biographies’, as Baym and Burgess (2020) call them, help analysts better understand the co-evolution, or the mutual shaping, of platforms and social relations just as they offer a ‘fixed point for considering the current online norm of constant flux and change’ (Lingel, 2020: 3). Moreover, work on online communities, many of which are hosted on platform infrastructure, has long argued that a time-sensitive lens needs to be applied to their study. As Kendall (2011, p. 321) has written in a summation of online community research, there are no ‘longitudinal studies of online communities, and there are only a few studies that look at very long-term users, except to determine if they engage in different kinds of online activities than less experienced users’. She argues that ‘future studies of Internet and community need to close the gap […] and analyze people’s Internet participation over time’. Since this observation, longitudinal studies have remained rare (see Bergstrom and Poor, 2021; Warren et al., 2012).
While this study engages with online community as far as it relates to mediated memory practice, Kendall’s (2011) observation foregrounds the need for broad engagement with the element of time and online sociality. Considering that the study of memory is always concerned with the relation between the past and the present and taking into account the ubiquity of platform use in today’s media environment, time-sensitive platform study is a needed lens for thinking about the relationship between media, memory and its maintenance today. This involves thinking both socially and materially about how platforms evolve, age and obsolesce and how these processes may come to bear on memory practices – and vice versa. For instance, this time-sensitive lens prompts questions like: What happens to an online community based around memory practices if platform engagement declines? Even if a robust community persists on a platform, sites may delete user content due to economic, technical or legal constraints. Indeed, they may close altogether, leaving little trace of their previously hosted content (De Kosnik, 2016). If memory is maintained in part because of its mediated persistence, what happens to it in the absence of these digital traces?
By focusing on September 11th memory practices on OakdaleTalk, this paper begins to answer these questions through a time-sensitive, maintenance-oriented approach to platform study, taking into account the shifting nature of a platform over time, exploring the construction of the past in an ever-evolving and ever-aging platform environment. Drawing on scholarly observations about the mutual shaping of media and memory, and the need for the time-sensitive study of memory practices on platforms, I therefore also explore the following research question through this case: R2) How do OakdaleTalk members understand the platform as changing over time? How does the aging of the platform, including the loss of posts, affect user memory practices?
Methodology
This study is based on 15 semi-structured interviews conducted over the summer of 2017 with OakdaleTalk users who engaged with September 11th reflection on OakdaleTalk. Because digitally mediated memory work might include creating media or simply engaging with media around an event (Lohmeier and Pentzold, 2014), I recruited those who posted or had read posts about September 11. To recruit those who posted, I searched for terms related to September 11th (e.g. 9/11, World Trade Center), and then messaged members who had participated in relevant threads. I also recruited those who reflected through reading by posting recruitment messages in the OakdaleTalk general forum. I was made aware of OakdaleTalk through an extended family member who was a former resident of Oakdale, though she was not otherwise involved in the study.
Face-to-face interviews were conducted in English in a variety of local settings over the summer of 2017, with one interview conducted over Skype; they ranged from 30 min to 2 hours in length. Two interviews turned into visits to local memorials with the participant. While demographic information was not collected, the large majority of participants presented as white, with eight women and seven men who ranged from middle-age adulthood (e.g. 45) through late adulthood (e.g. 75). Interviews were semi-structured, with questions addressing participant’s use of OakdaleTalk; their experience of September 11th; their memory practices and use of OakdaleTalk for reflecting on September 11th; and the loss of OakdaleTalk posts related to September 11th. By the time I left Oakdale, data saturation had been reached, with significant redundancy in themes of interest. Given this saturation, as well as that sampling was from a diminished number of users on a local community platform, 15 interviews were deemed an appropriate amount for the analysis. Interviews and platform posts were coded using QDA software NVivo. A thematic analysis employing a hybrid inductive-deductive coding schema (Braun and Clarke, 2006; Fereday and Muir-Cochrane, 2006) was conducted, in which codes were developed first based on research on digital memory and platforms (e.g. mediated memory work), with subthemes developed based on the content of the data (e.g. opposition to national narratives).
Results
Maintaining memories through OakdaleTalk
How, and why, do OakdaleTalk users maintain September 11th memories on OakdaleTalk?
In interviews, OakdaleTalk users described how they engaged in memory work around September 11, 2001, using the platform, whether shortly after the attacks or through the time of interviews. For those I spoke to, one way to maintain community-level meaning was through posting, and reading others’ posts, on OakdaleTalk. This typically happened on the anniversary of September 11th, when one user would initiate a memorial thread and others would respond to it. Users noted that posting on the forum was a way to recognize the shared trauma of the event. For instance, in 2009, Passionflower – a 60-year-old woman who now lived outside the US – responded to a thread recognizing the anniversary by posting a picture of the local train station on the morning of September 11th of that year. The station, which had a small memorial, was a symbolic site for the community. It represented the space from which some people had left Oakdale to go into Manhattan in 2001. She noted that in capturing and posting that image, she was able to acknowledge the shared pain of the attack years later: [Posting] was a way of communicating my memory, sharing my concern over what had happened, of wanting to communicate with this group of people on OakdaleTalk about something that had happened to all of us. I was just one person, and there were many people out there that had different experiences, but we all went through something very traumatic together.
Others posted on OakdaleTalk as a means of revivifying a feeling of shared social cohesion. As NJneighbor, a 75-year-old retired male resident, noted: September 11th is kind of one of those tragic holidays–well, not a holiday, but a memory of the date. You post something like [what I posted] and say, “Hey, let’s think about being human and helping others,” because you did see a huge amount of help for man when that happened. Help came from all over the country.
While users posted and read posts on OakdaleTalk to acknowledge the continued importance of that day for the community, still others stopped posting to protect themselves from that day’s trauma. User KMK, a 45-year-old lifelong resident of Oakdale who lost a loved one in the World Trade Center, had often posted a picture of him on September 11th. In recent years, though, she refrained, “Every year we would post, [but] I stopped. It’s easier to just… It’s like why dig into the wound.”
OakdaleTalk was used to maintain memories of September 11th in part because it was recognized as a place that had precedent for discussing the event. As participants observed, September 11th was an event that brought Oakdale, and OakdaleTalk, together as a community. OakdaleTalk webmaster and cofounder Christopher noted that the site ‘definitely brought people together. It was a place to go and console each other and really figure out what to do and how to help people’. Moreover, many users identified the site’s structure as offering affordances that fostered this memorial culture: in particular, features that allowed for information exchange and honest reflection immediately following the attacks. For instance, OakdaleTalk allowed for anonymous, asynchronous and geographically dispersed communication, while still maintaining a community focus. LindaJones, a 75-year-old retired woman actively involved in community events, for instance, would note that people could: … be at home doing whatever they had to do, and at the same time they could reach out. They could gain information. They could find out how they could help. The mayor would come onto OakdaleTalk and give us updates as to what the town was doing, how you could get home.
Users cited similar reasons for why OakdaleTalk continues to be used for reflection on September 11th and used in general as a community site. As Bookworm, a 60-year-old librarian in Oakdale, would note: A lot of people post [on OakdaleTalk] because they don’t have physical capability. I know a couple of people that this is their lifeline to community. Again, it might be too painful to come out. There’s a wide variety of reasons people use that in addition to in real life.
Christopher, the webmaster, explained that the site was used to maintain memories of September 11th because it was one of the only venues where emotionally recognizing the event felt acceptable. Furthermore, in contrast to platforms like Facebook that require posting under your real identity, he interpreted OakdaleTalk’s emotional openness to be due in part to the ability to choose an anonymous handle. How many other venues are there to really talk about [9/11] or mention it or bring it up? There is something unique about it, like the mix of real names and usernames and stuff like that. It might be just easier to bring it up there.
Users also recognized the site as exemplifying the interactive affordances of the internet more generally, and juxtaposed OakdaleTalk with the perceived restrictions of mass media. For instance, I spoke to Passionflower via video on 12 September 2017, a day after the 16th anniversary of the attacks. Now living outside the US, she described how watching TV the day before helped her feel emotionally connected, though it lacked the interactivity of OakdaleTalk and the internet. As she noted: I wanted to be watching the television, and when the “Star Spangled Banner” played, I felt a little bit emotional. I felt connected. I imagine it’s the same connection that I had seven years earlier [when I posted], with the internet. Except that here on the television, you can’t communicate, you’re just receiving information. But with the phone, and the internet, you can communicate your feelings to others, and record them.
In sum, users chose OakdaleTalk to maintain September 11th memory because they recognized the platform as having technical affordances, like interactivity, chosen anonymity and the ability to be accessed at a distance – along with a social tradition of memorial reflection.
Which memories are maintained?
While OakdaleTalk users acknowledged the ways that the platform was used for reflection, they routinely noted that they did not engage with OakdaleTalk in order to have an accurate, mnemonic remembrance of September 11th. As LindaJones would state, ‘I was in my office window facing the Twin Towers and saw the impact of the first plane and of the second plane. I still have those images in my mind, so I’m not looking to commemorate or revisit’.
Rather, users engaged in memory practices that they felt could maintain a particular, hyperlocal meaning of the event that recognized shared trauma within the community. Users frequently noted that OakdaleTalk offered a space of discourse where community members felt its appropriate meaning could be embraced. They juxtaposed these descriptions with examples of what users understood to be inappropriate, albeit dominant, interpretations of the past. As 43-year-old user Midwife80 would note, the platform offered a space in which community members could come together around shared experience: It felt important to share experiences with people who had parallel, some worse, experiences. Because everybody’s had the experiences of like, your relatives come to town and they want to go to the memorial, or they wanted to go see the site when it was still a smoldering pile.
In other words, those who experienced September 11th at a distance memorialized the event in ways that were anathema to OakdaleTalk users’ reflection.
OakdaleTalk users contrasted their hyperlocal memory with various narratives that were understood as totalizing, applying one dominant narrative to the event at the expense of recognizing individual experience. As users swapped memories of what happened on September 11th in the OakdaleTalk forums, they would often describe the sensory experience of the event, or otherwise enumerate narratives defined by proximity. Like Midwife80, interviewees often compared what the event meant to those who experienced it locally to those who experienced the event at a mass-mediated distance. As LindaJones, who worked in a building across from the World Trade Center, noted: Most of the people who lived here and worked in Manhattan worked in lower Manhattan. They worked right near the Trade Center or in the Trade Center. That’s a very different experience from looking on television and seeing the buildings fall. When you are on that ferry, and the debris was coming down around you and you get back into town on a train, and the fire department is there to hose you down, that’s very different from reading about it in the newspaper.
As LindaJones describes, those who had mediated instead of proximal experience understood the event differently and could forgot its meaning in ways that Oakdale community members would not.
OakdaleTalk users also deemed narratives totalizing when they did not pay adequate attention to the individual lives lost in the attacks. Indeed, individual narratives have been part of the US national narrative of the attacks. Each year on September 11th, the names of those who died in the towers are read in a ceremony, just as their names are etched into the borders of the official monument. In memorial threads, OakdaleTalk users would often post the names of those who were lost in the community. At the same time, users noted that these individual narratives were no longer maintained on a national level or in public discourse. User BKind, 50 at the time of our interview, for instance, compared what she thought was the dominant narrative about that event to narratives framed around individual loss: ‘I think it’s way too easy to just gloss over a big event like that and just say, “Oh, that was a terrible tragedy”. Well, it was, but those were also very specific individuals and each of their lives meant something’.
The memory maintenance that was performed on the platform was not separate from other memorial practices that supported these hyperlocal interpretations. Rather, offline memorial practices were part of the memorial meaning embraced and maintained on OakdaleTalk. For instance, users’ willingness to engage with local memorials – participating if memorial ceremonies were held or acknowledging their presence as they passed by them – strongly contrasted with users’ willingness to see the official 9/11 Memorial Museum in Manhattan. While nearly all users I spoke with reported interacting with local memorials in some way, and most visited and had enjoyed the memorial outside the Museum that listed the names of those who died in the towers, none reported going to visit the 9/11 Memorial Museum which had opened in 2014. Rather, respondents either had a marked ambivalence toward the 9/11 Memorial Museum or a strong reaction against it. Reasons varied for not wanting to visit: it would be too emotional; one respondent feared she would see something belonging to a loved one who passed away there; others said they identified New York City as a place of life, and when they would visit it was for enjoyable events; still others thought the museum was trivializing, a tourist attraction instead of a space of mourning. Regardless of the reasons, the museum was framed as highly narrativizing in such a way that did not align with OakdaleTalk users collectively maintained local memory. Like the use of OakdaleTalk, the use of memorials tracked with the upholding of local interpretations and individual experience in contrast to what were considered dominant or totalizing interpretations.
Memory practices on an aging platform
How do OakdaleTalk members understand the platform as changing over time?
Through the two-decade span in which OakdaleTalk existed at the time of interviews, the platform had undergone changes in design, usage and audience. In conversations about users’ September 11th-related memory practices on OakdaleTalk, participants readily brought up the ways the platform had changed during their membership. In this section, I briefly show how users conceptualized the platform as changing over time to establish a foundation for understanding how these changes affected user memory practices around September 11th as explored in the next section.
Users’ longitudinal experience of the platform was often structured by descriptions of the major underlying software updates that OakdaleTalk had gone through. Estimated as having occurred three times over 20 years, these updates caused tumult among the users. As Midwife80 noted, ‘It was a very simplistic platform initially, and every couple of years there’s a platform change and everybody freaks out’. The platform updates were announced to users beforehand, and sometimes required OakdaleTalk members to sign up again with new usernames, at times making for renewed anonymity. As LindaJones noted, ‘Since the software keeps shifting and people have trouble using their existing usernames on the new software, some of them will change their usernames. It’s hard to tell [who they are] unless you happen to know the people’. Moreover, updates often meant that content from earlier in the platform’s tenure – especially content hosted and formatted for previous software packages – would be removed from the site. At the time of interviews, content was viewable from 2007 onward, with the first decade of content no longer accessible, including posts made on or around September 11, 2001. In addition to lost posts and new usernames, software changes would result in a slightly altered user experience, with a newly configured discussion forum or the adding or removal of certain features, whether live chat or user blogs.
Aside from updating the site’s feature set, OakdaleTalk’s cofounder and webmaster Christopher noted that these updates were often made in response to a software package’s obsolescence. Over its tenure, he described how OakdaleTalk moved from one third-party software package to another before finally migrating to their own custom-made script. These software changes were prompted by either technical or corporate obsolescence, compounded by issues of technical maintenance. OakdaleTalk would undergo these updates, Christopher noted, when ‘eventually either a company is going out of business, or else I tweaked the script so much that I couldn’t upgrade it properly’.
These changes, especially recent updates, were described by users as spurring on an already ongoing process whereby community members engaged with OakdaleTalk with less intensity and with local Facebook groups with more intensity. Indeed, OakdaleTalk used to be ‘the only game in town’, as Bookworm phrased it, but eventually had to compete with other corporate social platforms like Nextdoor and Facebook. Turtle00, a mid-50s woman who worked in IT, summarized the links between the site’s updates, its decreasing usage and the rise of dominant platforms like Facebook for local community engagement: Two or three times [OakdaleTalk] had major upgrades that required them to move to new platforms, and it sometimes required everyone to change their password or do some work to get back on. Then there’d be a lot of discussion about that. The last time they did that, they were working really hard to improve the platform, and I think they did, but they had some stumbling blocks in getting back on. Meanwhile, there were some factions that were really angry with the group [of other users, due to political discussion]. And Facebook was bursting on the scene. It was after that, it’s like it never really came back to where it was. I mean it’s still there, I still use it, but it isn’t the way it used to be.
YourWindowDr, a 50-year-old user who now lived in a nearby town, noted similarly that usage dropped off after the last interface change. Previously, he observed, ‘you’d see someone posted three minutes ago on 20 different threads. Now, you go down five or six threads, and the last post was made two hours ago’.
Moreover, in addition to long-time OakdaleTalk users joining and using Facebook primarily, or using both Facebook and OakdaleTalk, OakdaleTalk community members described how the site had been affected by a generational shift in the Oakdale community. Notably, all but one interviewee had joined the platform in its earliest years, between 1997 and 2000. BKind, whose children were now in their 20s, noted that OakdaleTalk’s decreasing activity may be associated with changing modes of communication for the town’s incoming families: ‘This is the nature of the critical mass of discourse in town. It’s a bedroom town. People live there when their kids are in schools, and then my generation, they’re starting to leave’. Instead, interviewees described OakdaleTalk as having been whittled down to a core group of users. As Bookworm noted, ‘OT is down to a few die-hards. We all sort of know each other. There are plenty of people I don’t know, but it’s not like it was in 2001 or even 2011’.
How does the aging of the OakdaleTalk platform affect user memory practices?
As users of OakdaleTalk described, the platform had changed both socially and technically over its 20-year tenure. The surrounding digital environment had evolved since 1997, offering alternatives for online connection. Moreover, software updates caused either by technical obsolescence or corporate obsolescence resulted in user information and previously posted content to be removed from the platform, while also spurring on the transition from OakdaleTalk to corporate platforms like Facebook. Just as these changes over time came to bear on users’ everyday online experience, these changes also affected OakdaleTalk users’ memory practices, especially in the context of September 11th.
While OakdaleTalk members noted that the audience and user engagement had decreased on OakdaleTalk, members did not discuss this loss of community as removing the ability to reflect on September 11th on the platform. As Christopher had noted, OakdaleTalk continued to provide a rare online space in which to bring up this event, its impact on the local community, and residents’ proximal experience. The continued presence of the platform, and the ‘die-hard’ community of OakdaleTalk users, many of whom used the platform on September 11, 2001, provided a community in which reflection could continue. Instead, memory work was most impacted by the loss of posts after software updates. This section thus dwells explicitly on the ways that OakdaleTalk users encountered the loss of posts from September 11, 2001, and how they conceptualized this loss in relation to their broader September 11th memory practices.
In response to the 15th anniversary of September 11th in 2016, LindaJones had gone in search of the posts that were originally created around September 11, 2001. She posted this request on the OakdaleTalk forum and engaged in conversation with webmaster Christopher about whether these threads could be found. ‘In preparation for Sunday [September 11th], I tried to locate the original 9/11 threads that captured what was going on in Oakdale on that day 15 years ago’, she wrote. ‘Advanced Search did not return any threads older than 8 years ago. Are these older threads still available? If so, how would one locate them today?’ Some users responded supporting the request, though the threads were never located. Out of the users I spoke to, aside from LindaJones, Bookworm had also attempted to find them using the Internet Archive’s Wayback Machine, though to no avail. Indeed, it was a rare occurrence that other OakdaleTalk interviewees had thought to look back for these specific threads. Rather, many users, when asked whether they had ever looked back at old September 11th posts, were surprised they were gone at all. As I referenced LindaJones’s search for these posts, KMK asked ‘Are they gone?’ while Turtle00 asked, ‘They’re irretrievable? I mean, do you have those threads? I’d love to look at some of them’. Most users, after realizing that they were in fact gone, referenced the software updates that had occurred on the platform.
Most interviewees had faith that these posts could be retrieved by talking to the webmaster, Christopher, a known presence in Oakdale and on OakdaleTalk. ‘I wonder though if Christopher doesn’t have it someplace’, or ‘Christopher probably has all that stuff’ or ‘If you ask Christopher he might have them in the archives, right?’ In my conversation with Christopher, however, he acknowledged the difficulty of maintaining decades of user content, and the reality that these posts were likely lost: ‘I wish I had the [threads], but yeah’. When Christopher wanted to look back at old iterations of OakdaleTalk, he would pull up the Wayback Machine. He maintained some individual files of his own that related to September 11th, including photography of an impromptu memorial at a park that overlooked Manhattan. These photographs had at one point been on OakdaleTalk, in a special forum on September 11th where users could post their photos, but were removed in one of the updates. ‘We had a thing where people could submit their pictures’, Christopher noted. ‘I still have the pictures, I just don’t have the pages’.
Despite apparent surprise at the loss of posts originally made in 2001, some users had been aware of the potential for software updates to delete forum conversations and had engaged in acts of preservation to save other significant posts. Midwife80 recounted an emotional, long-lasting thread in which two people conversed about their respective loved one’s battles with cancer. She noted how, as a platform change was announced, someone had printed out copies of these threads to preserve them for the two users. Bookworm noted that she took screenshots of posts when she would want to save them – especially of local politicians – in case they were removed. While most noted they did not typically look back on forum posts, both Midwife80 and Turtle00 noted that they did because it served as a diary-like record for them.
While users were often hopeful that Christopher would be able to retrieve the September 11th posts, they also expressed that it would be a loss to the community if they could not be found. Some noted their historical significance to OakdaleTalk and the Oakdale community, and that they deserved a space for display. As KMK noted, ‘I think Christopher should [find them] just because it’s a capsule in time… he should have a little museum of content’. As LindaJones, who originally went in search for them, described, ‘I felt it was a shame [to lose them] because they are living history. They really captured the moment from multiple perspectives’. Others noted that they would like to look back on them to see what they and others in the community were saying on that day, especially to reflect sensemaking process around the event. For instance, Turtle00 stated: I would want to go back and look at them periodically. You know, from the perspective of all these years later. All the theories that we had about what was happening. What should we do? There was a lot of political debate as to what kind of response there should be. I’d be very interested in that hindsight.
Bookworm noted: I would like [to see the posts], not to remember but to memorialize it and see what we were all saying in the community then, ‘cause that was the hub of the community. We were all running around and talking to people and crying on the street. But this is where the most people could get together in the fastest period of time and post to each other.
While two participants were ambivalent about seeing the posts – as the 60-year-old former local politician JSmith would state, if they became accessible, ‘I would read them. I don’t want to, or not want to. I don’t even think about it, but I would read them just to see what actually most people wrote’, the large majority expressed a desire to see the posts. Some asked if I found them, that I would let them know.
Discussion and conclusion: Memory maintenance on platforms and the ‘preservation paradox’
OakdaleTalk offers a unique case through which to understand the maintenance of digital memory on online platforms over time. Despite OakdaleTalk’s aging – its diminishing engagement and user base, its loss of content as it shifted software programs – its continued existence and the investment of a core group of users nonetheless maintained a space in which September 11th reflection was embraced. Indeed, OakdaleTalk’s longstanding tradition of September 11th reflection shows how platforms’ modes of discourse are not just shaped by the technical affordances of the platform itself, nor just defined by the conversation du jour pushed to the top of the feed. Rather, traditions of discourse become associated with a platform and therefore can be continually reactivated and maintained over time. Even with the advent and use of other hyperlocal platforms by the users I spoke to – whether local Facebook groups or geographically bounded Nextdoor – most chose to continue engaging in September 11th reflection on OakdaleTalk.
This tradition of discourse, the site’s continued accessibility, and the persistent investment of a cadre of users became vectors for the maintenance of September 11th’s hyperlocal meaning over time. Yet, as OakdaleTalk users show, maintained meaning was also supported by personal memory – the vivid individual experiences remembered of that day – in addition to other local memory practices, such as the visiting of local memorials or the embracing of monuments that reflected the loss of individual life. These practices together sustained specific local meaning, one marked by physical proximity to the event, recognition of individual life lost, and appreciation for the shared trauma of the Oakdale community. In so doing, this article confirms and extends the findings of studies like Keightley and Pickering (2014), showing that practices that could be classified as digital memory do not take place in a purely digital environment. At the same time, hyperlocal meaning was supported by the vocal rejection of what were felt to be totalizing narratives, especially that of the 9/11 Memorial Museum and the interpretation of those who experienced the event only at a mass-mediated distance, through newspapers or television. Through digitally networked, personal, and physical practices, OakdaleTalk members reactivated a hyperlocal interpretation of that date, drawing boundaries through both mediated and non-mediated memory practices around what the September 11th attacks meant.
Furthermore, as scholars of cultural memory have shown, the mediation, interpretation and circulation of the past creates a field of cultural contestation in which unified cultural ideas about the past are produced and in which existing dominant ideas can be strengthened or challenged. OakdaleTalk shows the ways that particular, and in this case local, meanings are in part maintained through the persistent presence of a digital platform. While the maintenance of this particular hyperlocal meaning shows how non-dominant meaning can persist through digital channels, it does not necessarily offer the vision of multivocality theorized around digital media publishing and September 11th (Collins, 2007; Haskins, 2007). Rather, OakdaleTalk users created a relatively unified narrative, hyperlocal though it may have been, that could be collectively maintained through mediated circulation and continual reactivation within this particular social group.
This article examines OakdaleTalk as it existed until 2017. What is clear here is that the continued liveness of the platform, and the continued investment of a core user community, preserves one vector of meaning maintenance. What can only be speculated upon in this study are the ways that users would maintain these shared meanings if or when OakdaleTalk folds, or when it is rendered obsolete through an even more significantly reduced user community. Moreover, the usefulness of this forum for Oakdale community members outside of those original, die-hard users who came on in the years before 2001 is unclear. As platforms continue to age and shutter, future studies might look to what happens to a platform-based memory community when a platform sunsets altogether. How does a memory community persist in this case? What do users understand as being lost when a particular network is severed? How is meaning maintained – perhaps on other platforms – in its absence? Importantly, while OakdaleTalk’s content was unstable over time in terms of its persistent presence, the way that content was organized largely remained the same. This stands in contrast to major platforms like Facebook or Twitter, who introduced new ways of organizing content (like Facebook’s Timeline) as well as algorithmic filtering over their existence. Following discussions of algorithmic memory on platforms (e.g. Jacobsen and Beer, 2021), future work might apply the time-sensitive frameworks explored here to ask about the ways that the introduction, assimilation and evolution of platform algorithms has shaped memory maintenance through time.
While platforms might be understood as networked, ‘living archives’, as much of digital memory scholarship has noted (Ernst, 2012; Haskins, 2011; Smit et al., 2017), most OakdaleTalk users did not engage with the platform explicitly for its archival features – its storage, organization and accessibility of past information. However, its archival features were never useless. Some referred to OakdaleTalk as like a diary, while others noted that writing there was also a way to record memories. Crucially, the surprise at the loss of posts from September 11, 2001 shows the ways in which the archival elements of the site were at once implicitly trusted yet not explicitly considered. Despite widespread awareness of the software changes to the site, OakdaleTalk users voiced astonishment that the posts had not been saved. Indeed, participants’ desire to see posts from September 11, 2001 shows the ways that this content, despite not being explicitly reflected upon, is deemed to have memorial value. Users described the ways that this content represented important local historical information and would offer an understanding of these events within Oakdale and the OakdaleTalk community. In contrast to the 9/11 Memorial Museum or mass media, users were enthusiastic about seeing these old posts, thus showing the ways that this ‘old’ content was valued because it would support and could perpetuate the hyperlocal interpretation of September 11th. In turn, this case signals how platforms’ digital obsolescence may come to bear on memorial culture and historical reflection.
What this juncture presents – the desire to maintain these posts in some way yet the general absence of widespread individual effort to retrieve them – could be called a preservation paradox. Inspired by the language of the privacy paradox (see Barnes, 2006), which observed the ways that users desired digital privacy but rarely went through the steps to ensure it, the preservation paradox shows the ways in which users want meaningful posts to be maintained in some fashion but rarely go through the steps to save them themselves. Crucially, the preservation paradox is not so much a critique of user behavior as it is an articulation of an emerging type of memory practice under the conditions of platformized digital memory.
While more research on this practice is needed, this study suggests that the preservation paradox is influenced by user expectations over information persistence as well as the overarching structure of platforms that favors routine engagement with recent content, as observed by Kaun and Stiernstedt (2014). Where does this expectation come from? For one, the idea that these sites function as archives is a logical extension of their organization and typical ability to host and make accessible previously posted content, as is reflected in digital memory scholarship. Moreover, the growth of social media in the mid-2000s came along with much publicized warnings about the permanence of what is put online (Corry, 2021b) – the idea that ‘the internet is forever’. Most important may be that platforms have always incorporated memory functions into their interfaces, whether photo albums on early platform Friendster, or Facebook’s introduction of the ‘Timeline’ in 2011, which reformatted the user profile as a story of one’s life. Platform-supported memory maintenance may be more expected than ever, as platforms incorporate automated and algorithmically governed memory functions, as shown in studies like Jacobsen and Beer’s (2021). Of course, unlike the mandate of traditional archives, webmaster Christopher had never promised that OakdaleTalk would be a permanent digital record. Like most other platforms, the site’s terms and conditions also stipulate that OakdaleTalk admins have the discretion to remove content at will. Yet, that did not seem to singlehandedly account for users’ surprise at the missing posts. As such, this study responds to Good’s (2013: 570) question about whether the ‘lack of public awareness exists about social media sites’ terms of agreement and how they might affect the future accessibility of the personal digital archive’. A lack of awareness about user agreements could very well go into the preservation paradox, but so too do broader sociotechnical factors like those explored here.
At the same time, members had faith that Christopher – an accessible member of the OakdaleTalk and Oakdale communities – would be able to retrieve these posts. As Christopher attests, however, these previous posts were difficult to maintain over time because of both technical and social, or corporate, obsolescence. Interestingly, the two users who had gone back to look for posts from September 11, 2001 – LindaJones and Bookworm – both were involved in local libraries, either as a volunteer or as a long-time employee. What this might suggest is that local public information institutions like libraries may be natural homes for this type of content, with the ability to maintain memorial meaning without relying on the technical maintenance of individual users or site owners.
Though OakdaleTalk is one data point within the ecosystem of digital platforms, these challenges surface broader issues about the maintenance of user-generated content over time. They reveal the relatively limited power of platform users in determining whether large swaths of data will be removed or maintained, just as they show how deletion mechanisms are centralized under platform ownership. Though today’s dominant corporate platforms may have better technical ability to deal with obsolescence, unlike OakdaleTalk there is rarely an accessible means of contacting a webmaster or testifying to the value of lost content. This is especially true in cases when a site folds entirely or removes large swaths of data, an understudied but common occurrence in the history of online platforms, whether Vine, GeoCities, MySpace or others. As corporate platforms become the dominant means of experiencing the internet, and as these sites age, obsolesce and shutter, we might consider the value of content which is not governed by the principle of newness, the importance of some of these traces for supporting non-dominant meanings of the past, and the responsibility of platform organizations to their users, even in moments of decline. Mockup of OakdaleTalk. Note. An anonymized mockup of OakdaleTalk in 2021.
Footnotes
Acknowledgements
I am grateful to the Annenberg School for Communication & Journalism at the University of Southern California for the Annenberg Research Fellowship that made this project possible. I would also like to thank the users of OakdaleTalk for their generosity and kindness in sharing their experiences with me, as well as the anonymous reviewers for their generative feedback.
