Evidence of My Research Process with Data About Data

In thinking through how I plan to document and preserve my research process, I first thought about the different sites at which it takes place and comes into being: at the point of the artifacts that are being examined (the museum, its museum components, archival planning records, active or semi-active planning and/or promotional documents etc.); in and through the landscape as cultural text that is in perpetual motion (by examining it’s morphology and documents that speak to the social interactions that constitute its transformation); while undertaking the iterative process of deploying my methodology; in potentially triangulating my primary methodology with socio-economic datasets… there are many tangible and less-tangible sites where research information writ-large needs to be preserved in order to be accessed over-time.

Much like Quill and Kara, my mind also travelled to the digital preservation discussions that have emerged in the archival literature. Archival praxis is being drawn upon to ensure the authenticity of digital object across domains through the creation and application of standards of description and the documentation of record treatment that aims to capture the context, content and structure of records over-time. While all archival objects and their custodial treatment needs to be described in order to preserve the records’ identity and integrity, the digital preservation literature emphasizes the particular preservation challenges of digital records. A digital record “has a divided existence” because it comprises both a digital representation and the perceptible form produced from it; it can be both tangible and ephemeral, which makes it difficult to preserve (Duff INF1330, 2014 citing Levy 2001). Digital preservation is the proactive protection of the accessibility and authenticity of digital records overtime (Duff INF1330, 2014). It's different from traditional preservation for several reasons (Forde & Rhys-Lewis 2007, p. 28):

• Preservation of the information medium is not sufficient.

• Digital information is in a technical format which may become obsolete in a few years, thus creating an urgency for decision and action at point of creation.

• The data can deteriorate or disappear without warning, especially if no management techniques have been implemented.

• Issues of authenticity and authorship are more complex than for traditional materials, as the text can be modified and widely copied.

• The surrounding metadata is needed for the data to be fully functional, adding an extra dimension to preservation management.

• Digital materials can serve many functions, but only if proper preservation techniques have been employed.

Forde and Rhys-Lewis also emphasize that building preservation into digital projects at the outset is easier, more efficient, and less expensive than attempting to preserve (i.e. capture metadata and apply schemas) or re-create the material at a later stage (ibid., pp. 28-9). 

Haynes (2004 cited by Troselius & Sundqvist, p. 10) defines metadata in relation to the functions that it could potentially fulfill in given context or domain with respect to identifying and documenting the integrity of items or groups of items: 1. Resource description; 2. Information retrieval; 3. Management of information resources; 4. Documenting ownership and authenticity of digital resources and 5. To facilitate interoperability between different software platforms. Moreover, metadata plays a fundamental role in authenticating records (or meeting the functional authenticity requirements for records) in diverse contexts (across digital, analog or hybrid sites) to support their evidentiary power.  

So, we can definitely conceptualize the research process as a record generating process that requires the application of metadata for preservation. That being the case, I starting thinking about what I need to do in order to be a good "research process archivist" and the authenticity requirements for all project stakeholders e.g. for me to be able to write up my research findings and disseminate my research; to internal funding agencies; the external funding agencies etc. What trace of evidence do these stakeholders need for their respective functional requirements and how is capturing the trace complicated by the fact that I will most likely not be recording all of my methodological process notes, findings and drafts in analog form? While there is the usual discussion for wanting to preserve digital-born data in an electronic repository like DataVerse or DSpace type platforms etc., all of this in fact made me think more about the "data about data" or metadata that one would need for documenting either analog or digital research process work because in both cases, the work is unfinished and ephemeral to a certain degree, and not self-explanatory i.e. is somewhat easier to preserve something with clear bibliographic information and descriptive attributes. It then dawned on me that the units of analysis and coding framework for the textual analysis research methodology that I wish to deploy can be thought of as the metadata about my project. I will be creating both semantic as well as the syntactic or contextual units of analysis and codes to understand how my findings will come together as a relational framework to inform analysis. I may choose to initially document these units of analysis and the iterative process of adjusting them as I stumble upon new findings in a notebook and then translate some of these notes to a word processing document. Ideally, I would then want to migrate those notes to an electronic repository. 

As an important side note, in INF2122 Digital Preservation and Curation I learned from one my peers' presentations about digital preservation tools in the academic domain that external funding agencies are increasingly moving towards having applicants explain their data preservation strategies as one of the funding criteria. In fact, the UTL has a working group underway that is look at how data management planning tools such as DMPonline can be used to help researchers at this institution fulfill their funding requirements for research preservation. U of T developed a LibGuide to help researchers organize their data that has links on that page to DMPonline tools (that are in use at the University of Alberta) that further assist them in drafting a plan. So, all of this is to say, that there more and more reasons to be a good record-keeper from point of record creation onwards in the research domain.

Forde, H. & Rhys-Lewis, J. (2013) “Managing Digital Preservation.” In Preserving

Archives. 2nd ed. London: Facet.

Troselius, N., & Sundqvist, A. (2012). A comparative case study on metadata schemes at

Swedish governmental agencies. Records Management Journal, 22(1), 7–19. doi:

10.1108/09565691211222063