Main »

ACFAS 2014

ACFAS or the l’Association francophone pour le savoir is the Francophone equivalent of the Federation for the Humanities and Social Sciences. I was invited to demonstrate some of our tools at a one day session on Repenser le numérique au 21e siècle (Rethinking the digital in the 21st century).

Note that these conference notes are being written live about talks given in French so they will be brief and erratic. If you have corrections send me a note.

"Le Devoir" ran an interview with Michael Sinatra on new technologies (in French).

Ève Paquette-Bigas: (Re)penser les archives de la danse

The first presenter talked about dance archives. Such archives pose interesting problems. At best we can digitize surrogates of dance, not dance itself.

Marcello Vitali Rosati and Michael Sinatra: Amour (en) Numérique

Marcello and Michael have started a project to look a love online - specifically looking philosophically at the relationship between love and dating algorithms. The question they are looking is the question that everyone who uses online dating sites (like OK Cupid) ask themselves:

Can one find love on the Internet?

The idea is that one can find love in real life, but how can one find love in the virtual world guided by algorithm. Or, "what is love on the Internet?"

The way they are tackling this is to look at what is valued in the different sites as they don't have the algorithms. What ideas of love are these algorithms based on? This means looking at the profiles on the dating sites. Some character

Marcello also argued that our idea of love is being changed by the dating sites. Perhaps we shouldn't talk about real-world and online love, but just about a culture of love. That allows them to look at the culture of love over time.

• Love as destiny - an algorithms of fate
• Mystery - machines and randomness

Geneviève Has: Mort au Babel Fish : redéfinir l’image de la traduction dans un contexte numérique​​

Has gave us a great little history of the hype around automatic translation starting with the Babelfish from the The Hitchhiker's Guide to the Galaxy. She then showed a 1954 public relations short movie about how IBM is translating Russian into English. She then asked questions about the relationship between the human craft of translating and the human relationship with text on the one hand, and computer translation. With the rise of the digital, text has become "content" for the digital age. Further, vverything is code for translating - when localizing a game it is more than just the text. She seems to feel that some hybrid of human/computer may be the future. The computer augments the translator. We shouldn't think of the computer as an other apart from translator.

Panel: Le statut de la littérature à l’​​époque du numérique​​​​

We then had a panel about "the state of literature in the digital epoch." They had an interesting discussion about how the experience of fiction changes when you can check things on the internet anywhere. Then a discussion about how the idea of a "work" changes with the digital. Is something a work because a publisher says so?

There was a question about the relationship between Québec literature and electronic literature. Québec literature is a national literature which electronic literature is not. One can ask instead how Québec literature is inscribed on the web?

I learned a new phrase, "Adaptation Studies". This is the study of how a work (like the Lord of the Rings) gets adapted to another medium. Adaptation seems important to the digital which adapts everything into digits. Once digitized, everything can be transcoded.

Jean-Guy Meunier: Le numérique versus le computationnel : enjeux pour les humanités​​

Meunier started by talking about definitions of humanities computing or humanités numériques. He talked about McCarty's ideas about humanities computing as modelling. It is the modelling that is important. He talked about different levels of models. Models of literature, functional models and models that can be run on a computer. He used this idea to then question the importance of the "numeric" compared to the "computable". What matters is not the quantitative, but the formalization and the computability.

I suspect that the "numériques" is meant to be a translation of "digital" not a translation of "numeric."

He then talked about the computable and the Church Turing thesis. If something is computable then one can write an algorithm for it and one can describe it in terms of rules. Computation is not about numbers but about logic.

This led to the question of the computation of cognitive functions. How can we make computable the humanistic practices we care about like interpretation. One needs to be able to render that which is thought about in discrete data. One needs to be able to describe a practice with an algorithm.

He talked about functions that can't be computed which outnumber those that can be computed. The humanities should be interested in what cannot be computed. Many of the things that really interest us could turn out to not be computable.

Luca Paultrinieri: Quel savoir pour les MOOCS?​​

Paltrinieri comes from philosophy and is interested in MOOCs (or CLOM). He commented on the democratic dreams of MOOCs. He mentioned the ideas of Tommaso Labranca who talked about nesculture - culture on demand that can be easily acquired. By comparison the traditional education is more like an expresso machine where you have to be evaporated to then be passed through knowledge and end up coffee.

He talked about knowledge practices and Christian Jacob "Lieux de Savoir." There have been all sorts of different practices for becoming wise in the history of thought. Do MOOCs allow us to try these practices or are they built on one style of knowledge practice. He talked at the end about the practices they have found in the archives of Foucault. Through his archives one can reconstruct various different practices of list making, rewriting, citation and so on.

Dominique Bérubé: Défis de la mise en place et de l’exploitation de corpus en sciences humaines et sociales - le cas Érudit​

Bérubé talked about the challenges of running Erudit an open publishing platform that supports a large number of journals in Quebéc. One of the challenges is that of securing ongoing funding. They also worry that their journals might be lured away by other platforms like JSTOR. They are now looking at ways to provide an experimental environment within Érudit​ for people to try analytical experiments.

Stéfan Sinclair and Geoffrey Rockwell: Demonstrations.

In the final hour Sinclair and I demonstrated some of the tools we have been involved in, including:

• TAPoR - a site for discovering and reviewing text analysis tools
• Voyant - a text analysis suite
• CWRC-Writer - an XML editing environment that can be plugged into large distributed projects
• Visualizations - some prototype visualization tools