Before the “How?”of Digital Preservation Are the Questions of “What?” and “Why?”

Before taking this class, I thought of digital preservation only as a skill set I needed to learn. I wanted to know–how do I accomplish this thing that so many job ads expect me to be able to do? I realize now that by focusing exclusively on the “how?” of digital preservation, I had skipped over the important steps of first determining the “what?” and the “why?”. When these questions are taken into consideration, it becomes clear that digital preservation isn’t just a checklist of methods or tools students need to learn ,  it’s also a set of theories about what preservation means. The readings this week stressed that before creating a plan for preservation, it is important to first ask–what do we want to preserve? And why do we want to preserve it (what aspects of it are the worth saving)? Specific preservation goals will make it so that what we deem most valuable is saved, and (to paraphrase Professor Owens) if compromises must be made for practical reasons, having clear intentions will at least ensure that we are making those decisions deliberately.

Owens writes that the “future of digital preservation does not lie in the maintenance of old computing systems and hardware…[but in] making copies and moving those copies forward to new systems.” This may, upon first reading, seem simple enough, and yet it raises a multitude of questions about what it means to say something is the “same” as something else or that it is an “authentic” copy. Increasingly, archivists are coming to see that preservation judgments are (in the words of Herb Stovel) “necessarily relative and contextual.”

Geoffrey Yeo interrogates the idea of “sameness” in his article, arguing that what makes something the “same” as something else depends upon our definition of “sameness,” describing scenarios that roughly correlate to the ideas of the artifactual and informational frameworks we first encountered in Professor Owen’s book. As the title “Nothing is the same as something else” suggests, however, Yeo is ultimately skeptical of any claim to “sameness” with copies, because “attempts to determine the significant characteristics of records are problematic, not least because judgements about significance will vary from one user community to another.”

I was particularly interested in the comparisons that Yeo draws between historical manuscripts and digital objects. Yeo suggests that scholars may one day scrutinize digital objects in the same way that they now examine handwritten manuscripts for clues about the author’s intent in their handwriting, the spacing of the words, or smudges on the paper. We must, therefore, be mindful of what gets preserved with digital copies since,  according to Yeo, there is no detail so inconsequential that we can assume it will never be of value to some researcher. He also proposes that we may one day fetishize old forms of media like CD-ROMS or floppy disks in the same way we do other objects in museums or special collections. For this reason, Yeo proposes saving both the original and a copy (or copies) when possible, but if not, we need to at  least give careful consideration to who may be using the digital objects in the future and what their needs or expectations will be.

In a way that parallels Yeo’s questioning of “sameness,” Edward M. Bruner uses the example of Lincoln’s New Salem, a historic site in Illinois, to examine different definitions of “authenticity.” He proposes four main types–verisimilitude (the appearance of authenticity), genuineness (historically accurate authenticity), originality (the original and not a copy), and authority (an authorized replica). In New Salem, Bruner sees a mixture of the different types, identifying aspects that correspond to each of the four markers of authenticity. He also notices that the desire for authenticity can be very idiosyncratic and selective, pointing out all the different ways in which New Salem has been deliberately made inauthentic (such as paving the roads or adding bathrooms or gutters) and no one seems to give these details much notice. Bruner argues for transcending the inauthentic/authentic dichotomy and recognizing that tourists are not really there as judges of authenticity, but to “construct a past that is meaningful to them and that relates to their lives and experiences.” The lesson that a digital preservationist might take from Bruner’s article is that what makes a recreation or a copy meaningful is often highly personal and will vary from person to person. Unlike Yeo’s article, which emphasized that importance of details, Bruner seems to suggest that in some instances the details aren’t as important as recreating an experience that people can relate to. This reminded me of video game simulations–getting the game’s exact code correct may be less important than other, more personal factors that people associate with the game.

When it comes to what makes something the “same” or “authentic,” both Yeo and Bruner would say that it depends on the person. This connects to a second theme of this week’s readings–the importance of participatory or community-based archiving, especially for marginalized communities. While it is generally acknowledged that archives need to become more diverse in their representation, Katie Shilton and Ramesh Srinivasan write that archives “have appropriated the histories of marginalized communities, creating archives about rather than of the communities,” which can create distorted narratives. Shilton and Srinivasan propose changing the way archivists acquire and process new collections so that they are involving community members throughout the process of appraisal, arrangement, and description, ensuring that the archives represent marginalized communities the way that they want to be understood.

These ideas have become especially relevant recently in discussions of how to archive Black Lives Matters and other protest movements. Jarrett M. Drake argues that participatory archiving may not be enough, and that perhaps traditional archivists should avoid archiving the movement altogether and instead allow independent, community-based archives to do the job. Drake does, however, offer advice to traditional archives who still want to be involved. He tells traditional archivists to first look at their existing holdings to “see whether or not black lives matter there,” and to “confront [their own] complicity in white supremacy, patriarchy, and other structural inequalities that the movement is seeking to eradicate.” Moreover, they need to build trust among the “people, communities, and organizations around whose lives the movement is centered, a trust they should pursue not under the guise of collection development but under the practice of allyship.”

I’m curious–what did you think of Yeo’s and Bruner’s discussions of “sameness” and “authenticity”? Do you forsee a future in which digital objects are scrutinized for minute details in the same way manuscripts are today? What about the ideas presented by Drake–do you think that traditional archivists should have a role in documenting marginalized communities? If so, what steps do you think they should take to ensure they are doing so responsibly?





Preservation intent without the racism

Preservation intent

This week’s readings all build off the concept of preservation intent. In Chapter 5, Owens raises the two key questions of preservation intent: What about the object do you want to preserve? What do you need to do to preserve this aspect of the object? These questions should not be asked just at the beginning of the process, but continually during the processing of the content.

Some files will not be kept or preserved after considering preservation intent. The Rose Manuscript, Archives, and Rare Book Library of Emory University opted to remove such unnecessary files after acquiring Salman Rushdie’s laptop. After realizing preservation intent, Owens notes that archivists may choose not to preserve the object itself, but documentations of the object’s use. This was the case with Nicole Contaxis of the National Library of Medicine choosing to preserve a how-to on using the Grateful Med database, rather than the entire database itself. Owens also uses the example of preserving a screenshot for the Form Art website, rather than an emulation. The National Library of Australia (NLA) opted to use this preservation method for their PANDORA web archive as well.

This week’s readings also introduced the concept of significant properties, which are the aspects of digital objects that must be preserved if the object is to have continued significance. Webb, Pearsen, and Koerbin note that this concept can be more hindering than helpful for approaches to digital preservation. The authors assert that preservation intent should be declared before determining what the significant properties are, meaning the significant properties would be subjective rather than objective and universal. Yeo affirms this notion that there are no objective measures for value. There is no single method for digital preservation and preservation approaches should depend on the situation.

Archives and marginalized communities

Community participation is important for the preservation of objects from marginalized communities. If archival repositories intend to preserve the events and movements surrounding marginalized communities, such participation should be considered when creating the preservation intent.

Shilton, Srinivasan, Jules, and Drake note the foundation of white supremacy instilled in traditional archives – archival processing has historically been done without the input of marginalized communities, leading to finding aids that lack proper context and collections without diverse perspectives.

Shilton and Srinivasan promote participatory archival processing, which means seeking and listening to input from the actual community creators of the objects. Having “narrative and thick descriptions” from the community lead to contextual knowledge for the archival collection. Shilton and Srinivasan refer to these as empowered narratives, where the community is no longer being spoken for by archivists.

Jules addresses the significance of social media as it relates to protesting and causes. Social media helps promote awareness, allows news to spread faster, and helps movements grow. The preservation of social media is an obvious choice if an archive is interested in such topics, but it also leads to ethical and legal issues. Jules raises the concerns of protecting people in the social media collection, such as from repercussions from the police, and potentially being sued by the social media platform.

Drake promotes building allyship with marginalized communities should archives want to preserve their memory. Allyship involves genuinely wanting to learn about the lived experience of people in these communities and aid them, not just using them for collection development. Drake suggests that archivists help build up community archives and use their existing collections “to host dialogues, programs, and exhibits” around the issues faced by the community.

Affordances of media

In Chapter 3, Owens asserts that we can easily make quality informational copies with computing, but that artifactual qualities of these objects are lost. This is due to the platform layers involved with digital objects, à la the cake analogy used last week by Margaret Rose. Owens provides the example of the Rent script, where only the final version was visible in Word 5.1, while Jonathon Larson’s edits were visible in other text editors. According to Owens, the digital preservation field has essentially given up on preserving these additional layers and artificial qualities.

With a focus on informational preservation and ease of use, Arms and Fleischhauer detail the sustainability, quality, and functionality factors that the Library of Congress considers when selecting digital formats for preservation. The 7 sustainability factors are disclosure (documentation of the format), adoption (how widely used it is), transparency, self-documentation, external dependencies and how much work it will be to preserve them, impact of patents, and technical protection mechanisms like encryption.

The format selection is also dependent on who the primary audience is and how will they use the object. This determines what quality and functionality factors are given priority when choosing a digital format. The authors provide an example of the factors for the preservation of a still image: normal rendering (on-screen viewing and printing), clarity (high resolution), color maintenance, support for graphic effects and typography (filters, shadows, etc.), and functionality beyond normal rendering (layers and 3-D modeling).

Arms and Fleischhauer’s article offers technical insight on the outcome of considering media affordance and preservation intent. I found the article more difficult to understand but appreciated its technicality.

The media affordance available to the NLA led to hurdles with their web archiving. They deal with incorrect renderings due to their software, difficulty in preserving websites with many file types, unpredictable issues during batch preservation actions due to the idiosyncratic structures of websites, and some access issues due to file obsolescence. For the NLA, preserving the website “content, connections, and context” are of primary importance, while the preservation of the site is secondary. The master copy is preserved to the bit level, but the display copy, which undergoes preservation actions such as migration for long-term access, is of greater importance. The snapshots that the NLA preserve also only retain limited functionality of the original websites.

Discussion Questions

Have you noticed a lack of diversity or evidence of white supremacy in collections or records that you have worked with? How do you intend to address these issues when working as an information professional?

How can information professionals effectively assist digital community archives, when digital preservation is permeated with quick obsolescence and continual need for migration, documentation, auditing, and so on?

When creating their preservation intent, the National Library of Australia digital preservation team consider what “adequate access” for the digital objects are and how long that access needs to be maintained. This is a new concept for me because I generally think “forever” or “as long as possible.” I suppose an object only needs to be kept as long as it’s useful, but how do we determine what will be useful 25, 50, or 100 years from now? How do we know when to deaccession a digital object?

What matters most and how do you make it last?

I found this week’s readings overwhelming. This is primarily because it not only drew on a lot of the themes we’ve covered in class so far, but for me, is really at the heart of what it means to be an information professional. The subject was preservation intent, authenticity and selection, which quite honestly, seemed like everything to me. It turns out this is all interrelated.

What does it mean to be authentic?

Bruner describes four meanings of authenticity – verisimilitude, genuineness, originality, and authority – using New Salem Historic Site, a reconstruction of the village where Abraham Lincoln lived in his 20s. I couldn’t help but think of our discussion of artifactual identity in one of our earlier classes since it referenced another historic site, Mount Vernon (Owens, 15-17). Using Bruner’s terminology, the Mount Vernon mansion is authentic because it is the original. According to its website, “restoration efforts aim to represent the estate as it appeared in 1799, the last year of George Washington’s life and the culmination of his designs for Mount Vernon.” This description conforms to Bruner’s idea of genuineness because the idea is that someone from the same period could believe it to be from that period.

Verisimilitude seems one step removed from genuineness. It may pass as believable for visitors, but it isn’t picture perfect. Bruner’s description of New Salem and Mount Vernon’s website include descriptions of modern-day conveniences for tourists and upkeep. Bruner describes gutters on the log cabins that would not have existed at the time, and Mount Vernon has accessible pathways for wheelchairs. Presumably both have bathrooms somewhere on the grounds.

Bruner’s last meaning refers to an authority that certifies something as authentic. For example, the State of Illinois has authority to approve New Salem Village as the official reconstructed site. 

So what’s an authentic digital object?

We’ve learned in class that “digital information is material.” (Owens, 34) Just like words on a page in a book, it’s written on something like a hard drive. However, as a storage medium, hard drives are much less reliable than books. In order to preserve a digital object, you have to transfer it to something stable and be ready to do it again before the storage conditions fail. This concept is outlined in the storage component of the Levels of  Digital Preservation which we read about in our first week of class. The idea then of an authentic digital object precludes Bruner’s third meaning of “original” because we won’t be able to open and use files under the exact same conditions on the same hardware forever.

So how close can we get to the original and what does that even mean? Last week we learned about platforms layers. Digital objects are constructed within a certain context related to several factors such as software, operating system, file formats, etc. When this context changes, it affects how the object appears to us if it can at all. In order then to recreate the object we have to consider what’s important about it and what parts of the object we need to hold onto that ideal. This brings us to the idea of preservation intent.

Owens presents several examples in our readings of preservation which speaks more to creating an authentic experience of the object, but in order to get there, you have to think about the aspects worth preserving. In one case, it might be the appearance recreated through a screen shot; in another, it might be worthwhile to emulate the platforms that were originally used in order to present an interactive experience.

However, you don’t necessarily have control over all aspects to recreate the experience faithfully. Owens uses one example in Grateful Med, a software interface for searching medical information. In order to recreate the experience of using Grateful Med, you would need to emulate all the platform layers required to run the software as well as preserve the external medical databases used. Because of all the variations in platforms involved, this approach was considered impracticable. Instead of preserving the software, preserving the tutorial  served to fulfill the preservation intent which was to captured how the software worked.

This reminded me our readings last week on Documenting Dancebecause it showed how an experience can be documented without being strictly representational. You don’t have to make a direct copy. You just have to drill down to what you think is important to remember.

Who decides what’s authentic?

Bruner’s last meaning of authenticity dealt with authority. I think this idea was captured in two of our readings – Preserving Social Media Records of Activism (Jules, 2015) and Expanding #ArchivesForBlackLives: Building a Community Archives of Police Violence in Cleveland (Drake, 2016). Both of these articles have to do with social media, but it’s also about who has historically had the authority to save or neglect the history of marginalized people? Drake especially tackles this head on and describes how alienating the archival profession has been for black people. Archivists don’t have a place in preserving this story unless they acknowledge complicity in maintaining the white patriarchal structure.

Social media has been described as a way to give voice to people to tell their own story, but it’s complicated by issues of privacy and ownership as well as a means to capture “authentically” an experience from what may amount to millions of different perspectives.

I have to digress a little here because my feelings about social media are complicated. As an information processional, it’s not my place to direct how the public creates records. It’s even questionable as whether it’s my place to preserve them. I have to say as my own opinion, I question the value of social media as a way of authentically preserving an experience. Jules acknowledges the limitations of Twitter, but I think there’s a suggestion that these limitations can be overcome, and I’m not sure I believe that. The essence of Twitter is after all a means of surveillance, not sneaky government surveillance but marketing. Owens gave the example of Documenting the Now, an effort to ethically collect and preserve social media content. I have to hope that if smart people are putting their heads together to ethically preserve this, then maybe they can come up with a better alternative to current social media platforms all together.

There was so much more in our readings with week so I’ll look forward to reading your impressions.

“To Thine Own Preservation Decisions Be True”; or Deepening the Impact of Intentionality

Our readings this week shifted us from a purely theoretical approach to digital objects and their preservation to a more practical application, albeit in a generalized sense. Much like how you must learn to walk before you can run, one must learn to appreciate the digital forest before attending to individual trees. An underlying principle, inherent in all good preservation practices but particularly emphasized here, is to interrogate and articulate the reasons behind your digital preservation aims. This idea has been percolating in our readings and discussions since day one, but this week’s readings reinforce the need for an agreed-upon sense of purpose before any action commences. As Owens highlights in his fifth chapter, this starts with understanding what is significant about the object you are trying to preserve and how it relates to your institutional aims. We saw how this might play out in practice in the article on the National Library of Australia’s preservation intent statements. The NLA defines its institutional purpose through accessibility and it leverages its preservation intent statements to make explicit decisions regarding its digital collections in those terms. As demonstrated in the article, writing a preservation statement can yield consensus and clarity to stakeholders as well as direct suitable preservation actions based on the digital collection’s unique significance and needs. In essence, a preservation statement of intent is the basis for a road map for good preservation and a key starting point for effective policies.

The need for intentionality in digital preservation cuts to a deeper theme in this week’s reading: what it means to render something as “authentic.” The articles this week sought to upend any assumptions we might hold as to what makes something authentic or inauthentic by pulling the binary rug out from under us. As Bruner’s article on New Salem highlights, authenticity is largely a product of cultural values which not only shift over time but build upon one another. Thus, what makes something authentic is a matter of perspective and cultural context. The multiple iterations and applications of authenticity in Stovel’s article likewise emphasize its contested nature. Stovel’s deep dive on the Nara document and subsequent declarations on authenticity from the world heritage sector cast authenticity as a conditional process and one with an evolving set of definitions and applications.

By deconstructing assumptions about authenticity, Stovel and Bruner’s articles force a reconceptualization of how authenticity might relate to other values like originality. To hijack one of Professor Owens’ examples, the multiple iterations of the game, Oregon Trail, demonstrates the variety of ways a digital object can claim authenticity. For me, an authentic version is the second edition run on a Windows 95 system. It’s the version I played most often as a kid on the operating system my parents had which makes it the one I immediately think of when someone mentions Oregon Trail. The fact that there are earlier and later versions of the game that run on different systems doesn’t make my version inauthentic (though some may argue that); it merely demonstrates the constructed nature of authenticity and the various ways it can manifest across digital systems.

This doesn’t mean that originality can’t exist in digital objects and that that might not have value. This is one of the more perplexing nuances for me. As someone used to thinking of digital objects in an abstract, informational form, it’s difficult to imagine that a singular document or e-mail would claim originality in the same way a manuscript does. This is compounded when you think of the role copies and transfers play in the creation of digital objects. Yet, as Yeo makes clear through his example of the John Poindexter email, “…there may be potentially ‘significant’ properties of originals that are not reproducible or can only survive copying in attenuated form. Originality itself may be significant and cannot be replicated.” We’ve seen that demonstrated in the ways that Salman Rushdie’s laptops revealed contextual information that made his files more informationally rich (not to mention the emotional value of being able to look at a device and say, “This is what was used to create x.”).

Pursuing intentionality in digital preservation also includes more scrupulous attention to transmitting and preserving the community context in which digital objects were generated. Three of our readings took a close look at the ways common archival practices silence or distort the views of marginalized communities and presented ways archivists could better engage with and empower these communities. Bergis’ blog post highlights how social media affords archivists the chance to capture a layer of dialogue surrounding historical events that has often been hidden in the archival record. Similarly, Drake’s post on archiving the Black Lives Matter movement and Shilton and Srinivasan’s article challenge archivists to rethink their approach to basic archival functions within record maintenance and services provided.

I think marrying these principles with the purpose of the preservation intent statement can more fully realize digital preservation. As we’ve learned throughout this semester, preservation can follow a variety of paths within the digital world. Thus, it is necessary to be intentional in mapping a preservation purpose to ensure that things get done and stakeholders are on the same page. Yet, these decisions are often done internally without consulting the community from which these objects came. In doing so, archivists aid in the silencing of minority voices. Critical contextual information about the object and its production is also often lost. By bringing community members into the archival process and leveraging their input, we can pursue a better form of authenticity in digital preservation that grounds the object in its contextual heritage. More importantly, we can help community members’ reclaim their authority over their own records and how those objects are presented and preserved.

I began this summary by stating that our readings this week were beginning to move us towards more practical applications but that the content itself was still generalizable. This, I think, will always be one of the key hang-ups when it comes to writing useful but broadly applicable texts for digital preservation. There seems to be so much variance in what forms a digital object can take and what aspects of it are important to preserve that writing with any specificity can create more blinders than illumination. Yet, perhaps an expanded approach is the best way to think about digital preservation. Imagining digital content beyond one aspect or function allows us to appreciate its potential more fully. In doing so, we can develop a preservation plan that speaks to more than just its surface level expressions but captures its complexity in meaningful ways.

To Preserve, You Must Understand

This week’s readings solidified what I see as a collector in the present day (being any librarian or archivist). You will need to understand digital objects and have the confidence necessary to handle them.

Understanding the objects

The need to understand how a digital object functions in the information is imperative for the current and future relevance of the field. As modern information professionals, we all will work with digital objects: ranging from the catalog record, to more complex digital files such as images, articles, and books. The Owens chapter clearly illustrated the minute mechanics of a digital object. (It reminded me of how large internet companies track your internet browsing and searching. It also made me think that this practice may not be as scary as the news has made it out to be.) He also explained the importance of understanding these details and using them to become better digital curators and managers of digital content. Basically, if we know what makes up a digital object, and how we can optimally organize those objects, we can better preserve our collections and provide access to the general public.

When considering physical collections, space is such a clear consideration. Physical items take up physical space. Digital items take up space in a similar, more abstract way. They take up virtual space. We, as digital information professionals would need to work to ensure that our digital objects take up as little space as possible without losing some of the quality necessary to consider the object to be the same.

Practice of preparing, preserving, and using digital objects

In the Chan article, prestigious design museums have begun cataloging and digitally preserving symbols that have become popularized as a digital form (such as the “@” symbol) as claim-able objects. The authors explain the need for a thought process and concept where something like the “@” symbol can be digitally claimed and preserved: “The larger issue facing design museums is that more and more of the products “made” by design practitioners now lack any form at all.” In traditional media, design objects would be fully physical, or at least more easily claim-able by an institution. Now, however, design objects are created and continue to live in a virtual space. Design creators have become unwilling or unable to fully contribute to the longlasting preservation of their work.

Alternately, this can also come in handy when digital forensics are needed. Kirschenbaum and team explain this process and the necessity for awareness of how digital objects could be used for forensics and larger problem solving. In the information professionals’ world, this is often accessing records on outdated and no-longer-used systems. They lay out three reasonable options for ensuring continued access to materials: migrate the files and save both the original and the manipulated files, retain or obtain the original systems required for the media, and create or use an emulation to show the material on modern systems as if it were on the original system.

Questions for the class

How do you think having an understanding of digital objects will help you when you go to consult with your small institution?

How might we (future information professionals) go about preserving these kinds of digital objects now that we understand their makeup and how they can be saved?