One of the interesting aspects at the 2015 Global Digital Humanities Conference in Sydney was learning about the struggles in DH, which you don’t often hear about in academia. Because DH tends toward experimentation, people seem to be more open about discussing when aspects of a project didn’t work, or when a grand design had to be scrapped or modified due to unforeseen issues.

Here are some highlights of the sessions I attended over three days:

Organizational Practices in Digital Humanities Centers (Smiljana Antonijevic)
Having a separate DH center needs to be carefully planned because people may not use the space in the way you’re intending. One center built a nice computer lab but then undergraduates ended up moving in and going on Facebook all the time. Having a room with gadgets may not be enough to encourage people to get interested in DH. You need support (in accessible language, not “scholarese” or “programese”) to help train people. Otherwise, you can end up with an empty center.

The Question of the Luminary: Building a Resilient Campus DH Culture (Paige Courtney Morgan, Dale Askey)
There aren’t enough DH luminaries to go around so try to avoid relying on one person to carry a program and broaden the base by making local leaders (like library staff, IT staff, supercomputer labs). Even if you have a luminary, it can end up narrowing what people think DH is because they will associate DH with only that person’s projects. The best advice is to win the graduate students and the faculty will follow, and then the undergraduates will eventually follow. Having seminars that discuss the successes and failures is also important.

Psst! An Informal Approach to Expanding the Linguistic Range of the Digital Humanities (Elika Ortega, Alex Gill, Daniel Paul O’Donnell)
This was a really interesting session because it brought up an issue I had never considered that should be addressed not just at DH conferences, but any conference. Gill discussed the need for live translation so people who are comfortable in other languages can still engage and participate fully at the conference. He proposed a badge system that lists the languages you speak so people feel comfortable asking about them. He also called for the work of translating to become more prestigious so that academics will do it. This issue has come up again and again with other postgraduates I know who cannot find translations or transcriptions of commonly used references because that kind of work doesn’t count toward peer-reviewed publication quotas.

Publish: Whatever the Price? A French Study on Structuration of Costs during Publishing Process in Digital Humanities (Emmanuelle Corne, Anne-Solweig Gremillet, Odile Contat)
Apparently the main cost of publishing a journal is still the salary for the copy editor. The average time for an editor per journal/year is 10.5 months at a cost of 42,000 Euros. This means that digital publishing does not actually make much of a difference cost-wise. There is also the issue that paper versions still appear to present more legitimacy.

The Old Familiar Faces: On the Consumption of Digital Scholarship (Daniel Paul O’Donnell, Gurpreet Singh, Roberto Rosselli Del Turco)
Despite the increasing number of digital editions, scholars do not use them very much even though they know of and have access to them. There is a disconnect between interest in electronic resources and their reported use. On average, it takes around eight years for an edition to be cited after publication. There is still resistance to citing new editions even when they are being used.

Disciplinary Impact: The Effect of Digital Editing (Elena Pierazzo)
Humanities scholars must decide whether they are okay with the field splitting like Computational Linguistics did and cutting the relationship with the traditional discipline. Now, there is no interchange between them and mainstream linguists. The DiXiT (Digital Scholarly Edition Initial Training Network) meets every six months and is an intensive training camp populated largely by PhD students from a traditional scholarly background. The number of digital papers at disciplinary conferences has stayed around 14% since 2010. Could this be because there is no clear career path for in-betweeners in most countries? Pierazzo wants the discipline to innovate from within, rather than creating a new discipline.

Pedagogical Hermeneutics and Teaching DH in a Liberal Arts Context (Diane Katherine Jakacki and Katherine Mary Faull)
They discussed teaching DH to first and second year students in a Comparative Humanities program. They exposed students to different methodologies, like distance and close reading, and network and spatial visualization. Students had to think critically about what each method revealed about the subject matter. It was a high risk/high reward learning model. The course evaluations were overwhelmingly positive and even non-participating students eventually got engaged. Best quote: anyone who says you can’t teach DH to 18-year-olds…I challenge you.

Remembering Books: A Within-book Topic Mapping Technique (Peter Organisciak, Loretta Auvil, J. Stephen Downie)
The HathiTrust Research Center has 4.8 million books scanned. Why do we use the page as the unit of analysis and will this have to change with the changes in book publishing? Different editions of books might give different results in research projects.

Press F6 to Reload: Games Studies and the Future of the Digital Humanities in India (Padmini Ray Murray, Souvik Mukherjee)
She referenced Tara McPherson’s “Why are DH so white?” and how tools like XML are not neutral. There is cultural swapping that goes on when video games move countries/cultures. For example, rice bowls become doughnuts when they move from Asia to North America. There are no Indians in Call of Duty set in India, just Russians and Americans, as if they wouldn’t be concerned at the violence occurring in their neighborhoods. The postcolonial exotic is fashionable in games. There is a Developers Dilemma where women in India are doing a lot of the grunt work but not the original work of game design and creation. There are implications for those who speak up about issues like GamerGate.

Game of Thrones for All: Model-based Generation of Universe-appropriate Fictional Characters (Matthew Parker, Foaad Khosmood, Grant Picket)
Video games have ever-expanding universes as they develop. Huge worlds need an increasing number of characters to keep the suspension of disbelief and achieve a higher immersion. But producing characters is cost-prohibitive. Audiences are asking for more sophistication. This project is part of a larger one, NPCAgency, to create off-the-shelf characters with unique names and histories who can converse with main characters realistically.

Gender Markers: Distinctive Words in Male and Female Authorship (Sean G. Weidman, James O’Sullivan)
Using textual analysis, they looked at differences in the words female and male authors used in books over time. In the Victorian era, female authors used more words around family and personal pronouns, while male authors used more contractions, words about beer, and sexual language. Male authors’ words were more colloquial and quantitative. Interestingly, the use of personal pronouns is linked to mental health (Secret Life of Pronouns). More recently, female authors used more words of private/micro places (home, church, school), while male authors used words of public/macro places (country, earth). Females used the language of uncertainty (wondering, seemed, believed) and males the language of certainty (exactly, absolutely). But the context is key. Women may still write more about the home, but it is in a different way, can be critical of it.

What Do You Do With a Million Readers? (Roja Bandari, Timothy Roland Tangherlini, Vwani Roychowdhury)
There is research that can be done on crowdsourced literary criticism (like Goodreads or LibraryThing), like examining how people read and remember books. They often focus on events they like or plot summary. This project compared the reviews of books that had over a thousand reviews with information from SparkNotes to see how accurately people remembered the books.

Anatomy of a Drop-Off Reading Curve (Cyril Bornet, Frederic Kaplan)
I haven’t been able to stop thinking about the topics in this session. It was about how ebooks can track precisely how they are read and the analytics can be used for multiple marketing purposes. In a July 2012 Wall Street Journal article “Your E-Book is Reading You” it found that science fiction readers read more quickly and finish most books they start. There seems to be an assumption that slower reading is negative. Will reading analytics be part of a digital novelist’s toolbox? As in, will they consider how people will consume their book and have that affect how they write? If they are paid by how much of the book is read (as Amazon is now doing with some books), will this stifle creativity? There is an ethical dimension to how ebook data is used.

Traces of Lives in Digital Archives: Life Writing, Marginalia, and Google Books (Tully Barnett)
Andrew Norman lost his job for taking pictures of workers leaving Google Books scanning facilities for his project ScanOps. Today’s reader can’t leave a trace like writing in a book or having a stamped library record. There is a loss of corrections, spelling changes, or a comprehensive intervention on the text that some readers have historically made. Google Books scanning has made books stuck in amber. This has also led to the discovery of handwritten notes which are enshrined in digitized form. OCR can pick up some handwriting. Writing in a book leaves traces of the multiple lives it touches which can alter the reading and research of these texts. Barnett also mentioned the Book Traces project.

Modeling Concepts to Improve the Search Capabilities on Ancient Corpora (Muhammad Faisal Cheema, Judith Blumenstein, Gerik Scheuermann)
Doing a simple keyword search on ancient texts is not always helpful when you are looking for concepts. Also, ancient texts have free word order, ancient vocabulary, different ideas, cultures, and practices we’re not aware of. Building in a feedback loop allows scholars to make changes.