Metadata is a foreign concept? Whaaaat?!? (Part Two) – A guest post by Greg Bak

[This is the second guest post by Greg Bak, Archival Studies, Department of History, University of Manitoba.]

Thanks again to Kate for agreeing to publish my presentation on her blog. In Part One of this guest post I discuss some of the Twitter reaction to my talk; in Part Two I include the slides and speaking notes from my talk.

This second part of my guest post is to set out a bit of the context for my presentation, and to provide the slides and speaking notes, which you can access here: Bak_SAA13_s701.

Update, December 18, 2013: At Greg’s request his slides are no longer accessible as he is expanding on his ideas for a lengthier discussion in a journal article. If you would like a copy of the slides, please contact him at

My presentation was the third of three in SAA 2013 session 701. The session was titled “It’s All About the Items: Digital Objects and Aggregations in Archival Description and Access.” My co-presenters, Kelcy Shepherd of Amherst College and Kat Timms of Library and Archives Canada, had just spoken to the challenges posed by item-level metadata within archival theory and practice.

I chose to build on Kelcy’s and Kat’s talks while providing a different conceptual framework. Following their talks meant that I didn’t have to get into the question of why archives must manage item-level metadata: Kelcy had just discussed this with reference to made-digital records, and Kat had done so with reference to born-digital records.

This is a point that I have addressed in an earlier article (Bak and Armstrong 2008). Digital preservation and digital management require that archives create or capture item-level metadata. My presentation is in no way intended to ignore this basic fact of digital archiving. Instead, I focused on the nature of items and aggregations within archival theory in contrast with bibliographic theory.

The presentation was to take only 20 minutes. It lacks the nuance and depth of evidence that I will include in the manuscript that I submit for peer review.  Additionally, I was not able to build upon the basic foundations laid out in this presentation to examine how reconceptualizing archival data could allow us to reimagine not just the description, discovery and access of archival records, but other archival functions as well, including appraisal, preservation and outreach.  Some of these implications are addressed in a piece I published in Archival Science in 2012, while others emerged in the panel discussion after the talk.

As I revise the presentation for publication I welcome your thoughts and comments about the ideas included here. Please feel free either to comment below this blog post or to contact me by email:


Bak G (2012) Continuous classification: capturing dynamic relationships among digital information resources. Archival Science 12.3:287-318.

Bak G, Armstrong P (2008) Points of convergence: seamless long-term access to digital publications and archival records at Library and Archives Canada. Archival Science 8.4:279-293.


Be Sociable, Share!

3 thoughts on “Metadata is a foreign concept? Whaaaat?!? (Part Two) – A guest post by Greg Bak”

  1. Kate:
    Thanks for posting Greg Bak’s slide show on your blog. His point that libraries manage items and that archive manage relationships helps us understand why archivists should be careful about copying models created by librarians.
    I also liked his discussion of metadata. Since my experience was primarily in preserving databases (i.e. data organized in tables) I have understood metadata to be the record layout and code definitions to the raw data. Only reluctantly did I come to accept metadata to also include descriptive data. Because metadata can mean different things to different people I am always concerned when I see the terms being used without definition. Bak has contributed to the awareness that we need to define the term to avoid confusion.

  2. Hi Greg,

    This is a really important point, and one that I overlooked in my presentation. “Metadata” remains a necessary concept, particularly when defined as code definitions and layout for structured data. Everything these days is a database. Nonetheless, perhaps as a hangover from our analog past, we tend to be object-centric in how we imagine information resources.

    When information specialists (librarians, archivists, whatever) consider only the digital object to be “the content” (or “the record”), we have excluded a wealth of additional data from also being “content.” If we consider everything contained in the data fields to be content (or, as in my presentation, data) – objects or characters, it’s all just code! – we could return to the original definition of metadata as being, truly, “data about data” – i.e. the layout, code and field definitions that give a database its meaning.

    I agree wholeheartedly that this would bring greater clarity to how we conceptualize information and information resources.


  3. Very interesting, and it was nice to read some theory during my working day.
    I’m still processing these ideas, but my first thought is that ‘following library models’ is a bad idea, indeed. See archival collections attempting to be put into Institutional Repositories, which were envisioned and built for monographs. The librarians I work with have no concept of aggregations, and the context these provide, or this differentiation between data from the materials and metadata added by the cataloger. Perhaps as libraries begin working with data sets, we’ll find more common ground.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.