Notes on Ensuring the Longevity of Digital

Documents on obsolete digital technology could be lost

DD are not readable without machinery and the "language" might change as well. Also, data is often "compressed" or encrypted. And software embeds special info that is only meaningful to that software and often only that version of the software. "Bootstraps" might be needed

What to do?
Ongoing updated efforts--migration but data is often lost or changed 
Translating, but we lose info in that process
copying
extend the longevity of stystems that read the documents
Create a standardized format--but even standardized database models have special features that differentiate them from one another. Moreover, relational modeling might be replaced by something like object oriented database
Bootstraping will require annoted notes to explain how the document was created and how to "read" it

Updated information from other sources
see Howard Besser
DPE Digital presrevation Europe
DELOS Digital preservation cluster of the DELOS NoE
Digital curation center

Ross, S. "Digital Preservation, Archival Science and methodological foundations" 2012

Ross argues that not much progress has been made in the last 20 years. This essay outlines what needs to be done to ensure that materials are preserved over time
Definition of DL: “the infrastructure, policies and procedures and organizational, political and economic mechanisms necessary to enable access to and preservation of digital content” (taken from the National Library of New Zealand, quoted in Ross, 2012, p. 49)


"one wonders what the digital analogue of the photograph of the conventional library will be" (p. 44)
Ross states that digital preservation is "about maintaining the semantic meaning of the digital object and its content, about maintaining its provenance and authenticity, about retaining its 'interrelatedness' and about securing information about the context of its creation and use" (p. 45)
Libraries must balance their objectives with their capabilities
Options include
  1. retaining the content
  2. retain the environment and context of creation and use
  3. reproduce the experience of use to ensure the right semantic representation and information
UNESCO guidelines for digital heritage (2003) indicate that libraries need to establish "verisimilitude" (See Lars Calusen 2007)
Ross notes that access to DL objects is correlated to their recurring use for "business" purposes and here he defines this as "evidential, information or commercial" value (p. 46). Objects that go out of use periodically are subject to benign neglect which is not good for Digital objects.
Ross notes that there is not a widespread recognition of and action about preservation for multiple reasons: lack of organizational strategic plan, no standardized practices across or within organizations; benefits to such preservation were not clear to institutions; value of the item did not seem to be worth preserving; organizations were awaiting technological solutions (p. 47)

Ross does not feel the library community has stepped up and made this problem widely known (p. 48) nor has it developed sufficient preservation measures or the data to help understand the problem
What makes a good DL?
Good technology
Organizational and cultural apparatus that makes the operations work (Ross, 2012, p. 50)
Establish the trust of the community (Ross, 2012, p. 50)
Ross argues further that the DL must ensure its users that it will not lose its content (Ross, 2012, p. 50)
DL should make it clear “where the digital materials came from, who created them, why they were created, where they were created, how they were created, how they came to be deposited, how they were ingested. . . how the digital objects was maintained” (p. 51). Or to put it another way they need to ensure "authenticity, integrity, interpretability, and context. . . across systems, time and context" (p. 55)
Diplomatics assists in determining a digital object's provenance and that helps to verify and contextualize the object (p. 52)
Digital preservation ensures the maintenance of the value of digital objects (p. 53) ensuring that they are not altered and are complete given that digital objects are easier to alter. Need a well documented chain of custody
Authenticity is important. user needs to know the object is authentic and needs to be able to validate the authenticity on their own with tools provided by the library (what?). Library should "have documented the process of digital entity ingest, management and delivery" (p. 54)

Research in digital preservation needs to increase. Current "artisan" practices of preservation do not scale. Automated and scalable pracitces need to be introduced (p. 56)
The Digital preservation Europe group has identified nine themes to bring digital preservation in line with those in the analog world (p. 58-59
  1. Restoration-some computer forensics methods exist but more need to be developed
  2. Conservation--what needs to be done to ensure that the data will still be accessible far down the road (annotation? metadata? migration?)
  3. Collection and repository  management-how should the collection be constructed and maintained
  4. preservation as risk management--need to translate the uncertainties invovled in digital preservation into quantifiable risks that can then be managed
  5. Preserving interpretability and functionality--need to remain semantically meaningful over time, authentic, usable, reliable. How is this to be done?
  6. collection cohesion and interoperability
  7. automation in preservation
  8. preserving the context
  9. storage technologies and methods
Ross imagines some kind of testing facility to see what methods will work (p. 59)
Firm theoretical foundation is necessary for DL (p. 60)

I just read a very interesting essay about digital preservation that relates at least in two ways to the discussion of goals and purposes. First, the article notes that the definition of a DL should include something about preservation. He offers the following definition of a DL from the National Library of New Zealand: “the infrastructure, policies and procedures and organizational, political and economic mechanisms necessary to enable access to and preservation of digital content” (taken from the National Library of New Zealand, quoted in Ross, 2012, p. 49). Second, he argues that DL MUST have a theoretical foundation in order for "research and development in the DL area" to flourish and to "communicate our cultural and scientific knowledge" to the future (Ross, 2012, p. 60). Definition and purpose would certainly be part of a theoretical foundation.

Rothenberg (1999), after recounting some possible losses of digital data, seems to be establishing the parameters for thinking about the coming trend of digital libraries that could be foreseen in 1999. He establishes the nature of the potential problems: we must have machinery to read digital data AND the machinery must be standardized so we don't lose data when the machinery changes. Obviously, the article is dated, but these concerns still seem very relevant (as evidenced by Hart and Liu, 2003; Ross, 2012). Rothenberg suggests some solutions going forward: ongoing updating efforts; translating of data across different programs; copying; extending the longevity of systems that read the documents; creating a standardized format for preservation; and some kind of "bootstrapping".
In the little reading I have done about preservation, the problem of preservation in the digital library persists. Ross (2012) identifies 9 areas of concern that the Digital Preservation Europe group has noted (p. 58-59) which include those Rothenberg forecast but add that preservation also needs to have restoration strategies and the context of the digital objects needs to be maintained.
The point raised by Hart and Liu (2003) that most discussion of preservation has left out the issue of trust between the user and the digital environment adds another dimension to DL preservation. They state, "Unless people have sufficient confidence in digital documents, as they have in paper and monetary currency, they will not trust electronic document management appliances" (p. 94).  Ross (2012) briefly addresses this, stating that the DL should make it clear “where the digital materials came from, who created them, why they were created, where they were created, how they were created, how they came to be deposited, how they were ingested. . . how the digital objects was maintained” (p. 51). Or to put it another way they need to ensure "authenticity, integrity, interpretability, and context. . . across systems, time and context" (p. 55).
Hart and Liu (2003) and Ross (2012) make it clear that preservation is still an issue that demands attention and that solutions need to include not only those noted by Rothenberg but also some way of ensuring institutional trust and integrity of digital data (which will require some kind of record keeping that documents the context of and changes in digital documents) and creating some kind of technical digital restoration strategies.

What to do?
Ongoing updated efforts--migration but data is often lost or changed 
Translating, but we lose info in that process
copying
extend the longevity of stystems that read the documents
Create a standardized format--but even standardized database models have special features that differentiate them from one another. Moreover, relational modeling might be replaced by something like object oriented database
Bootstraping will require annoted notes to explain how the document was created and how to "read" it
Hart, P. E., & Liu, Z. (2003). Trust in the preservation of digital information.
 Communications of the ACM, 
46(6), 93-97.

Ross, S. (2012). Digital preservation, archival science and methodological foundations for digital libraries. New Review of Information Networking, 17(1), 43-68. doi:10.1080/13614576.2012.679446


Rothenberg, J. (1999). Ensuring the longevity of digital information. Council on Library and Information Resources, , June 16, 2016.

Comments

Popular Posts