GSJ 525 / DH 530 Data, Power, Feminism
The Serendipity Project
A moment of reflection…
"I want to stay with the trouble, and the only way I know to do that is in generative joy, terror, and collective thinking."
— Donna J. Haraway, Staying with the Trouble
And so begins this project, a collision of nerddoms, in partial fulfillment of the requirements of GSJ 525 / DH 530 Data, Power, Feminism and a degree in Library and Information Studies, but also, as all good courses should do, fulfillment and stoking of a larger sense of curiosity.
This course was about learning new data research tools, about the underlying research approaches they were built upon and also designed to support, and applying those tools to real world projects.
Since this “project” is in large part simply a reflection on that course, it is written with a very small, very specific audience in mind, but perhaps if you come along for the ride you too might… serendipitously… pick up a few things…
The research platform
First let me (re)introduce you to HuNI – the Humanities Networks Infrastructure research and discovery platform.
Conceived of in 2010 and first developed in 2011, HuNI provides a virtual lab space in which to experiment in digital humanities research on a large and growing pool of interoperable data. And while ‘experimental’ is definitely the operable word, using the platform does make real many of the concepts and ideas around digital humanities research, such as Social Network Analysis techniques and feminist approaches to studying relational data.
Coming from a background working in graphic design, web development, and relational databases, I encountered HuNI with a reaction that became the first theme of the course for me: that it was remarkably familiar and yet utterly new. In terms of familiarity, it was not simply the mind-map style data visualization, but also the underlying idea of connecting different types of data points together in coherent systems. Collections as data made perfect sense to me.
But there was something distinctly new in how this software allowed me to make those connections. Unlike most other data systems I’ve worked with, in particular donor management software and library cataloguing systems, the complete lack of ‘controlled vocabularies’ or ‘relationship tables’ in HuNI was downright upsetting to – second theme of the course – what turns out to be a deeply ingrained and previously unconscious bias for seeing the world through a librarian’s lens.
That said, the basic schema of ‘records,’ ‘collections,’ and ‘links’ – the “nodes” and “edges” of this network representation – was relatively intuitive perhaps because this platform, at least to me, reflects how the world actually works… but we’ll come back to that in a moment…
Diving in…
(with the following images linked to the actual HuNI source they reference, should you want to dive in as well!)
The “Juvenilia Press” Collection
So upon first jumping into HuNI, I made my first discoveries about the system by not quite exactly following the class assignment (possibly another theme of the course, or my whole degree). Instead of working with existing items and links related to my actual assigned class project, I wandered off and created a different collection because I recognized some other unrelated records.
In particular among the many ‘collisions of nerddoms’ was seeing a record for the Juvenilia Press, a small academic publisher with whom I have a long history, already in HuNI, but ‘orphaned’ and unlinked to anything. And so I made my very first HuNI collection and links:
Figure 1: The Juvenilia Press Collection in HuNI
The “Jan Reimer - Edmonton's Recycling Mayor” Collection
Then came the actual project: to load HuNI with new records based on work with the City of Edmonton Archives – for our group in particular a set of oral history recordings of former-Edmonton Mayor Jan Reimer, part 1 and part 2 – and then create links that told a story:
Figure 2: The “Jan Reimer - Edmonton's Recycling Mayor” Collection in HuNI
What’s interesting here is that, at first glance, the Juvenilia Press ‘web’ looks to have ‘more’ to it, just based on a count of nodes. But both of these images reflect only what I chose to include in a particular collection. Click on the eponymous node within each collection and this is what appears…
Figure 3: The Juvenilia Press as the selected node
Figure 4: Jan Reimer as the selected node
And here’s the same Jan Reimer visualization if you then click on another node…
Figure 5: Jan Reimer and Edmonton as the selected nodes
The result is a fascinating, if somewhat chaotic, visualization of connections between people, concepts, organizations, places, events, and works.
And yet this too is an oversimplification of the actual power of the HuNI model…
Because the lines aren’t just lines…
Because links are not just lines. And neither are relationships.
What isn’t evident from the visualizations above is that what’s represented by the lines between nodes aren’t all the same. In being presented as identical lines, the visualizations mask the quality and nature of the links and relationships.
One of those lines could simply represent “was home to”… as in “[The University of Alberta] was home to [the Juvenilia Press].”
But that same line could also represent “was instrumental as then-Mayor in the establishment of the City of Edmonton's implementation of the”… as in “[Jan Reimer] was instrumental as then-Mayor in the establishment of the City of Edmonton's implementation of [the Blue Box Recycling Program (BBRS)].”
And a line could also represent a negative or non-relationship, as in “[John Smith] is not [John Smith]” if one knew two records actually corresponded to two different humans who go by that same name.
A line could also represent two vastly differently qualities of relationships running in opposite directions, for example (very hypothetically), “[John Smith] was an ardent admirer of the work of [Jane Doe]” paired with “[Jane Doe] required a restraining order against the unwanted attentions of [John Smith].” And there’s nothing stopping a researcher from adding such a relationship to HuNI… but we’ll also return to that.
In effect, the visualizations become more of a finding aid than actual data – a hint that there is something more – and until you start digging deeper into the connections, the real story remains hidden.
Here is some of what you’d actually find, a sampling of relationships and connections from just my few collections and links…
Juvenilia Press is directed by Christine Alexander. Christine Alexander is the director of Juvenilia Press. University of Alberta was home to Juvenilia Press. Juvenilia Press was located at University of Alberta. University of New South Wales is home to Juvenilia Press Juvenilia Press is located at the University of New South Wales. Christine Alexander is a professor at University of New South Wales. University of New South Wales has on faculty Christine Alexander
Juvenilia Press is directed by Christine Alexander. Christine Alexander is the director of Juvenilia Press. University of Alberta was home to Juvenilia Press. Juvenilia Press was located at University of Alberta. University of New South Wales is home to Juvenilia Press Juvenilia Press is located at the University of New South Wales. Christine Alexander is a professor at University of New South Wales. University of New South Wales has on faculty Christine Alexander
The Edmonton Regional Mayors and Reeves included Jan Reimer. Jan Reimer was a member of the Edmonton Regional Mayors and Reeves. Jan Reimer was an outspoken and recognized advocate of policies in Edmonton, Alberta, Canada supporting Recycling. Recycling was a concept actively promoted and supported in Edmonton, Alberta, Canada by its then-Mayor Jan Reimer. Jan Reimer was instrumental as then-Mayor in the establishment of the City of Edmonton's network of Eco Stations. Eco Stations in the City of Edmonton were established under the leadership of then-Mayor Jan Reimer. Jan Reimer oversaw the establishment and construction, as then-Mayor, of a series of City of Edmonton Eco Stations. City of Edmonton Eco Stations were established and constructed under the oversight of then-Mayor Jan Reimer.
The Edmonton Regional Mayors and Reeves included Jan Reimer. Jan Reimer was a member of the Edmonton Regional Mayors and Reeves. Jan Reimer was an outspoken and recognized advocate of policies in Edmonton, Alberta, Canada supporting Recycling. Recycling was a concept actively promoted and supported in Edmonton, Alberta, Canada by its then-Mayor Jan Reimer. Jan Reimer was instrumental as then-Mayor in the establishment of the City of Edmonton's network of Eco Stations. Eco Stations in the City of Edmonton were established under the leadership of then-Mayor Jan Reimer. Jan Reimer oversaw the establishment and construction, as then-Mayor, of a series of City of Edmonton Eco Stations. City of Edmonton Eco Stations were established and constructed under the oversight of then-Mayor Jan Reimer.
The Blue Box Recycling System (bbrs) is a practical municipal program level example of Recycling. Recycling can be implemented at a municipal program level through the Blue Box Recycling System (bbrs). City of Edmonton Eco Stations are an active and ongoing implementation of the concept of Eco Stations. Eco Stations are being operated on an active and ongoing basis in the City of Edmonton. Edmonton is the same place as Edmonton
The Blue Box Recycling System (bbrs) is a practical municipal program level example of Recycling. Recycling can be implemented at a municipal program level through the Blue Box Recycling System (bbrs). City of Edmonton Eco Stations are an active and ongoing implementation of the concept of Eco Stations. Eco Stations are being operated on an active and ongoing basis in the City of Edmonton. Edmonton is the same place as Edmonton
A brief critical interjection
In reviewing HuNI and searching its contents for this reflection, I will note an issue with its search engine that I hadn’t noticed before: it seems like the search engine only searches against content within the nodes but not within the edges, or links. So if in fact the edges are the real ‘meat’ for this platform, it seems problematic if I have no way to search their content? For example, how do I look up how many things have been identified as “not the same as”?
So where does this all lead?
Well, building on the underlying themes of this course being “remarkably familiar and yet utterly new” and of challenging my “deeply ingrained and previously unconscious bias for seeing the world through a librarian’s lens,” three concepts come to the forefront, upon which I reflect with yet again a sense of “not quite exactly following the class assignment” as well a bit of the same sense of chaos that has been yet another theme of this course…
Wrestling with the tensions in Data Representation
The importance of Unlearning
Serendipity as an applied research approach and a field of study
Wrestling with the tensions in Data Representation
(in which I pin nothing down and mostly continue to flail with ideas that keep fighting back…)
“A network is not a graph…
…but sometimes it can be useful to represent a network as a graph.”
And yet as discussed just a few scrolls up before here, that graph can mask its true meaning.
So then how simplified can it become to be interoperable and interpretable, and yet be complex enough to be accurate?
“A network model is more than a metaphor”
But for someone who has lived and interpreted the world in metaphor, I am still sorting out the ideas of ontologies vs data models vs networks vs graphs, which is where we return to the idea of HuNI, as an example, reflecting how the world actually works… its nuances and complexities, what Edouard Glissant called “the fluctuating complexity of the world” in Poetics of Relation (p 32).
But with also a return to Scott Weingart’s warnings that “When you’re given your first hammer, everything looks like a nail” and “Methodology appropriation is dangerous.”
The validity of the representation determines the quality of the model
And this, I have learned, is determined at so many levels… the network concept, the data actually collected, the question you start with… the abstract conceptualization and the practical representation of the data.
But a sound model does allow us to empirically research relationality, which is, once again, a remarkably familiar and yet utterly new approach for me to such complex networks.
The importance of Unlearning
“The illiterate in the 21st century will be those who cannot learn, unlearn, and relearn”
– Alvin Toffler
In which I just reflect on just how “librarian” I am…
Those tensions in wrestling with data representation are as much a reflection of my own tensions in viewing and processing the world, then of any inherent issues with data representation itself. Where platforms like HuNI, and their social network analysis underpinnings, seek to bring in more complexity and capture a more accurate and nuanced view of the world, my previous favourite course was LIS 502 The Organization of Information, in which we gleefully categorize and subcategorize the world down into neat little boxes that, as we also learned in that class, impose a very specific and limited, and sometimes problematic, ontology onto said world.
So, despite being someone who delighted in discovering the idea of “folksonomies” in LIS 502, I had a fully visceral reaction to seeing a long messy dropdown menu appear in HuNI of suggested possible relationships, sorted alphabetically in what resulted in, for me, an extra jarring and dissonant suggestion that there was any sort of order to be found in the list. And if you scrolled the list to the bitter end, those two terrible words: “Create new…”
Which is exactly the power of HuNI.
Fortunately some of this unlearning is being championed from the inside of the library walls, with Emily Drabinski’s “Teaching The Radical Catalog” (from Radical Cataloging: Essays at the Front edited by K. R. Roberto. Jefferson, North Carolina: McFarland, 2008, 198-205) being a prime example.
And yet this does not hit at the core difference between my inner librarian and my inner digital humanities researcher. In both trying to make better sense of the world, they do so by pulling in opposite directions.
Which brings us to…
Serendipity…
Part of me just wanted to place a link to the Serendipity Society and leave it at that. As they describe themselves, they are “a network intending to promote rigorous research and support for researchers of serendipity of all sorts.”
File this under “unlearning” but the leap from acknowledging serendipity to studying it as a phenomenon was another moment of “unlibrarianing” myself…
And yet I should hardly be surprised. In studying to be a librarian, I have already come across articles like Patrick L. Carr’s Serendipity in the Stacks: Libraries, Information Architecture, and the Problems of Accidental Discovery or Rose-Wiles, Shea, & Kehnemuyi’s Read in or check out: A four-year analysis of circulation and in-house use of print books that look at serendipity.
Maybe the difference lies in actually trying to cause serendipity to happen, not simply studying it as a phenomenon but instead actively trying to apply it as a research approach.
This video about HuNI pretty much captures the difference in both tone and philosophy between the approaches in library studies and digital humanities research…
And so what is the actual “project” part of the Serendipity Project?
Possible next steps…
Some for you, some for me…
-
I don’t know how you got here, but since you are here, maybe there’s something else here you were meant to find. I invite you to pop over to our Visual Explorer and see if something happens to catch your eye…
-
Because maybe what you were meant to find isn’t right here, so here are some additional resources from the course to share:
Data as Relation: Indigenous Data Sovereignty and Ethic of Care, presented by Kayla Lar-Son, February 2, 2021
Data Feminism, by Catherine D'Ignazio and Lauren F. Klein, The MIT Press, 2020
Digital serendipity: be careful what you don't wish for, by Aleks Krotoski, The Guardian, August 21, 2011
-
We’ll see if this one happens, but where LIS 502 resulted in The Catalogue Projects, perhaps GSJ 525 will level things up.
I floated the idea in class of wanting an “empty HuNI” into which I could load and play with my own library catalogue information. The TA’s response was “why not just load all that info into HuNI and play with it there?”
So that’s what we’re exploring next… how to perhaps load data from The Catalogue Projects into HuNI…
A closing thought…
It occurs to me as I close this window of reflection that even the librarianship/digital humanities dichotomy in which I have framed much of this reflection is, ironically, a product of the librarian approach of categorizing, rather than thinking about all of this as a series of lenses through which to filter the world. I even referred to it earlier as a “librarian’s lens” but it’s really more a librarian’s somewhat singular point of view. So not enough to simply switch out or add new lenses, I really need to move my entire self once in a while in order to gain a whole new perspective.
The unlearning continues…
Respectfully submitted, and put out into the Universe, by Winston Pei on December 13, 2023