I live in the South now, and there is what is known as the "come to Jesus" moment when one has seen the light. I had my digital humanities "come to Jesus" moment today in my Introduction to TEI (Text Encoding Initiative) course at the Digital Humanities Summer Institute  (#DHSI2012 on Twitter if you're interested in following along) in Victoria, BC, Canada. Now, I have been working to learn as much as I could about digital humanities, and I knew that the work being done in the digital humanities was solid academic work, if unconventional, at least compared to the "publish-or-perish" culture in which I came through. But it wasn't until today that I actually understood how something we take for granted, like encoding a document, is a deeply subjective and critical academic exercise.
I think the bias stems from two places: the first place, for me, is that I learned long, long ago how to encode in HTML. There are reasons why we use other programming languages other than HTML (and we have HTML5) to create websites: original HTML is rudimentary and a little simplistic (at least when I learned it in 1996). People kept telling me that TEI was a little (and some would tell me a lot) like HTML. My memories of HTML was that it was a simple and rudimentary mark-up language that made text look like something online. The other bias I think comes from the idea that these tools and languages were developed in the computer sciences. "Science" implies impartiality, strick guidelines and rules, and just less flexibility and subjectivity.
Of course, I should have known better. I do know better, as I teach my students to be critical consumers of "science" when it comes to being blinded by numbers, statistics, and other easily manipulated datasets (I'll make a pitch for programming, but I also think that critical literacy for the 21st century should involve a good stats course). But TEI, like any language, is just as subjective and just as malleable as the person using it wants it to be. When marking up a document, you have to decide what pieces to tag, to emphasize, to make "visible" to the computer and thus to any future user or reader. Repeatedly we were told that what we decided to mark up and how would be dictated by what we wanted to focus on, highlight, and imprint on the texts. And as with any traditional research project, there is a finite amount of analysis that we can do as one person within the project. For example, if I wanted to create a critical digital edition of a Dany Laferriere book, I could in fact embed through TEI all of the literary and other artistic allusions he makes in the text. With the translations of Anne Hebert's poetry, I can point it back to the original French, as well as different versions of the same poem by different translators.
The text that the person on the other side of the computer screen eventually will read and interact with will be entirely mediated by the decisions I make as an encoder. I can fundamentally modify and manipulate how you, the reader, experiences the text. My decisions, as a scholar, will dictate your reading of the text, even if I make it appear as interactive as possible. To those of your reading this who have been doing DH for a long time, feel free to pat me on the head, or lecture me on how this is what you've been saying all along.
Have a little sympathy; I've finally seen the light. And it's overwhelming. I came here repeating that I didn't even know what I didn't know. Now I know what I don't know, and it's really a lot to take in.