As noon approached, lunch beckoned and the discussion became atomic, situational, one-on-one. Folks began following each other on Twitter, gathering their thoughts, etc., and just then I heard fellow historian Lincoln Mullen mention casually that as the users who comprised the ‘#codespeak’ hash-tag’s ‘network’ on Twitter began following each other, the diversity of the network actually became _less_ interesting, as what were once distinct branches of ‘follows’ were fused. An incredibly unpackable idea, in the ether with so many others at this one-of-a-kind event. Perhaps good DH was, in a sense, all about mixing it up – just as we were doing here at the conference – keeping the approach fresh, diverse, and dynamic while surfacing the deep humanities questions we were asking.
After this intellectually athletic morning session I was ready for a burger with a few of my new friends, each one fascinating. After the discussion over lunch (which was refreshing to both mind and body) we ambled back happily to the Lab in Alderman Library and settled down to work – and play. Next up was classicist Hugh Cayless, whose own work on TEI and text organizational systems was close to my own (historical text analysis-besotted) heart. Hugh walked us through the many layers of meaning and function in text – and the complicated nature of managing these computationally. Organizing systems for managing and researching text must confront the daunting reality that (thanks John Laudun), ‘texts’ themselves as an object of research can possess context, subtext, and paratext, can be related to one another thru allusion and reference, and can serve any number of purposes, whether narrative and poetical, linguistic or material.
As Hugh plainly stated, “Texts are complicated and the tools are stupid.” True enough. And maybe some of us can help change that (indeed, many of those in attendance have already started).
Many humanists have text-based sources, about which they (we) want to learn and understand a lot more – so a vibrant discussion about what it means to research and work with (text and other) data in a DH framework ensued, a central theme being the nature of ‘mastery’, and what it means for the research and pedagogical relationship. In many developer communities, walls of expertise are erected around those with core knowledge – many times, those who are also best equipped to provide the kind of education in a software tool or code base that a DH scholar might need. The ‘noobie’ has enough trouble just trying to learn to swim in a fast-moving code repository, and having a ‘mentor’ or ‘master’ to help you can be instrumental in providing that ‘tacit’ knowledge.
The last seminar of this whirlwind day was led by Mia Ridge, whose own work and presentation on managing ‘messiness’ in data was incredibly apropos to some of my own work. I was fascinated by the demonstration of her approaches to working with cultural heritage metadata and excited to learn more from the group.
The resultant conversation about decision-making in data management, specifically regarding the uniformity (or incongruity) of data, including error, “edge cases” and ambiguous or contradictory information, had me at once both rapt and raring to go. A meta-theme was, essentially, how to manage the balance between crafting one’s data to fit the tools used and, more importantly, crafting the tools to best work with the data. We talked about some of the error scenarios and strategies many of us have faced in text analysis of library archives (including OCR typos, fuzzy dates, metadata errors and the choices behind when to use ‘stopword lists’ and ‘stemming’). The code and data we assemble is certainly a discussion between technologists and academics, as mentioned by Bill Turkel, but it also even more true that as Jean Bauer commented, the data is as much an argument itself as it is a subject of argument. Even more fascinating was the conversation about ‘visualizing absence’, a blog post about which was quickly disseminated via Twitter as the post-conference conversation continued. From Mia’s seminar (as with the others), I came away with a trove of conceptual and practical ‘nuggets’ I have only just begun to unpack and examine in greater detail.
Last was a building exercise – we separated into groups and defined a project based on the discussion we’d had and the themes, platforms, approaches and goals that had most interested us. I sat down with a few of my colleagues and we started hatching our plan. The topic of engaging people (audiences and scholars) from ‘marginalized’ communities had been a fascinating and recurring theme (ie the varying technological and academic knowledge successfully made available to, and the varying participation of, LGBT individuals, those of other ethnic/class backgrounds, the disabled and others). so we would set about exploring (in addition to the many ideas I detailed above and the many more that I have not), possible solutions to those barriers. Whatever obstacles those who are ‘marginalized’ may face, it is clear there are (and there are in DH) many novel means to overcome and upend them. Given its ongoing revolutionary aspect, Digital Humanities promises to remain a progressive and dynamic locus of study for many scholars of all fields and backgrounds. I am excited to be a part of the field and of the community that has emerged from the Lab.
PS: During dinner that night, we had a poetry slam and my limerick won!
“The secret to DH as racket
Is to download some source code
and hack it
Do whatever it takes
to make sure nothing breaks…
…and never forget that [close bracket]!”
I was given a lovely 3d-printed bracelet courtesy of Bethany, Jeremy and the Scholars’ Lab MakerBot! Awesome ending to the conference.
