The Good, the Bad, and the Ugly

It’s already some time now since I’m back from the digital humanities conference double pack, which involved about 3,000 km of rail travel: first DHd 2015 in Graz, then the DH Summit in Berlin. In Berlin several people, who couldn’t make it to Graz, asked me about how it was; here’s basically what I told them.

Please note: As DHd 2015 had four parallel tracks, I could only listen to about a quarter of the talks. The following is solely based on my personal impressions and is thus not only subjective, but also incomplete.

The Good

Graz is a very nice city, and the organization of DHd 2015 was flawless. Even though this is only the second edition of the DHd conference, it has already established itself as the place to be for DH researchers in German-speaking Europe. It was great to meet old friends and to making new connections. DHd (the association) is doing a great job here.

The talk Interdisziplinäre Nutzung von Forschungsdaten mithilfe einer technisch-abstrakten Modellierung (roughly: Interdisciplinary reuse of research data supported by an abstract technical model) by Carolin Odebrecht was not only the best talk I heard at DHd 2015, but also a really good talk in absolute terms. If you don’t read German, I’m sure she’ll soon publish an English-language paper about her work as well.

The Bad

It was announced during the opening that 60 of 200 submissions had been accepted, which gives us a very respectable acceptance rate of 30%. However, I was seriously disappointed with the quality of the research in many of the presentations that I attended, in particular with respect to prior and related work.1 It’s probably not surprising that this especially concerned work done “outside” of the authors’ main disciplines.

Of course it’s a challenge to be aware of relevant work in all involved disciplines when you’re working in an interdisciplinary field. However, there’s no way around it: either you familiarize yourself with the literature, or you bring in people from the “other” fields. The “big tent” metaphor should not be understood to mean that research of any quality is welcome; if the digital humanities want to be taken seriously, they have to enforce quality standards.

I don’t really blame the reviewers; it’s a general problem with “humanities-style” conferences, i.e., where authors don’t submit full papers, but only abstracts. Abstracts are promises, not finished papers, so reviewers can only assess the potential of a talk, not the reality of a paper. As long as this mode is kept, there is very little one can do to address this problem in general, except urging authors to explicitly mention related work in their abstracts.

One could, however, address the specific—but frequent—case of research done in the context of doctoral dissertations by instituting a doctoral consortium. At a doctoral consortium, PhD students in an early or intermediate stage present their thesis proposals in front of a panel of senior researchers (and an audience), who give constructive feedback in a friendly and relaxed atmosphere. A doctoral consortium is a good way for PhD students to identify weaknesses, omissions, or methodological problems and to get help with refining the thesis topic. Given an interdisciplinary panel, this would also be an opportunity to receive guidance with respect to related work in other disciplines.

The Ugly

I was appalled at the fact that you can get spontaneous applause at a DHd conference for polemicizing against computer science by simply reiterating conspiracy theories like, “computer science wants to take over the humanities and make us all redundant.” This is, of course, bullshit, and repeating it won’t make it any less so. Still, it’s disheartening to see that such cheap propaganda apparently still resonates with a non-negligible part of the audience.


There’s a lot more one could say, but this post is not intended as a conference report. Others may have written more about the program of the conference, or check the book of abstracts. I’m not sure I want to draw any other conclusions at this point other than: there was the good, the bad, and the ugly…



As I already mentioned above, I could not attend all parallel sessions, so I may have been simply unlucky, and this is actually not a problem at all.

Cite this blog post
Michael Piotrowski (2015, March 30). The Good, the Bad, and the Ugly. NLP for Historical Texts. Retrieved February 24, 2024, from

Author: Michael Piotrowski

Computational linguist, computer scientist, professor of digital humanities at the University of Lausanne.

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.

Search OpenEdition Search

You will be redirected to OpenEdition Search