I'm kidding, there was no Big Bang, really. No "one day I wake up and Bang! next thing I know, I'm doing a PhD." Although, since we're not going to pretend that I started this research blog as soon as I enrolled in a PhD program, I figured that I could dedicate one or two posts to the genesis of the project.

Last night, I was eating at a joint with one of my PhD colleagues and a couple of their friends and, as it often happens when you say you're enrolled in a PhD program, they asked me what my topic was. Which I haven't even told you about yet! Well, here you go: I want to make it possible for my fellow DH researchers to have more solid references on which to base their estimation of how much time and resources they are going to need in order to use automatic transcription of manuscripts in their projects. I will most certainly write about this again later, but let's say that currently the resources available are feedback provided by more or less enlightened users, which means that it is hard to draw general rules from them.

Going back to yesterday's conversation: I explained my topic, and one of the reactions was something like "wow, how the hell did you get the idea of working on this?" (I am sure I am not the first one to hear that, and this won't be the last time either). I think the curiosity of my conversation partner was mostly based on what they thought my background was, which itself was based on what they knew about my colleague. And I mean, they were not completely wrong: after all, I initially got a Master's degree in Art History. But that was 6 years ago. Now, to understand how I ended up working on an advanced Machine Learning technological environment after writing a Master's thesis on early 20th century French proto-cinema, we have to agree that a lot can change in 6 years. Right?

Even in 2016, though, I was toying with the idea of doing a PhD. But I was not decided to do it yet because I was in a complicated relationship with my field of expertise, Art History, and I wanted my work to be more meaningful, as in somewhat useful to the society in general. After graduating, instead of starting a doctorate or finding a job in a cultural institution, I went through two more years as a Master's student at the École nationale des Chartes in Paris, which is probably one of the best decisions I've made. A pretty intensive year of training in Digital Humanities and an internship later, I started working as a research engineer at Inria, surrounded with colleagues doing PhD in Natural Language Processing (something I knew nothing about back in 2018), and being personally involved in projects using... handwritten text recognition.

I worked as an engineer at Inria until November 2021, which is when I officially started my PhD and signed up at the École Pratique des Hautes Études and the University of Montréal. This was already five months ago. In the next post I'll certainly try to summarize what I have been up to since then, but I want to take the time right now to say that even though this experience contributed to making me feel more comfortable with my current project, it also makes it harder to know where to start with this blog. So, hopefully you're still hanging on.

Now going back to my colleague at the joint last night, we sometimes talk about how they feel that their research project is "so unclear even though they started the PhD program in September." But here is the thing: I may have officially started in November, but I have been working on defining my PhD topic ever since I took the final decision to start one. Which was somewhere around the 7th of March 2020 (on a beach, near Montpellier). That's right! Just a few days before COVID turned everything upside down... I won't retrace the evolution of my topic since 2020: it is tied to many many things involving my missions and projects at Inria, the state of the world in general, preliminary searches for corpora, readings and, above all, conversations with people. Most definitely my current topic has barely anything to do with my initial ideas, and I don't think I could have compressed the time it took me to inform my ideas.

Now the very first draft I started for this blog was an attempt to name the people who have accepted to talk about "PhD" with me over the past couple of years. A sort of "thank you" disclaimer well before the final manuscript. No doubt, this list will grow, but to finish this post today, I would like to thank already, for their time, wisdom and encouragements: Anne Baillot, Emmanuel Château-Dutier, Thibault Clérice, Antoine Courtin, Léa Duflos, Yohan Dupont, Antoine Fauchié, Clémentine Fourrier, Hugo Fraslin, Edward Gray, Margot Mellet, Marie Puren, Mathilde Regnault, Charles Riondet, Laurent Romary, Peter Stokes, Daniel Stökl Ben Ezra, Pedro Otiz Suárez, Lionel Tadjou and Stéphane Tison.