WWP blog post and dissertation data display site

I want to plug a couple of (relatively) recent happenings around my dissertation research. First, I was asked join a cohort of Pedagogical Development Consultants, who work with the Women Writers Project to pilot assignments that use TEI or XML (or that use WWP resources). As part of that partnership, I’ve written a blog post, “A New(ish) Approach to Markup in the Undergraduate Classroom.” It is a pretty succinct framing of the pedagogical approach and rationale for using XML in undergraduate writing courses, complete with pictures and links to one of my assignment prompts.

Second, I am making the anonymized XML data and schemas generated in my XML-based writing course available. You can check out the GitHub repo if you are interested in playing with the data, or you can view the student writing on this display site I built.

Keystone DH script/slides

Since I am looking for feedback, I thought I would post my slides and the script from my talk earlier today at the Keystone DH conference.


Hello, my name is Kevin Smith and the title of my talk is (Re)orienting XML/TEI in composition.

I’d like to start with a story: last summer, I worked every day in Northeastern’s digital scholarship commons, which is a collaborative working space for faculty and grad students working on digital projects. While I was plugging away on my own work, the project team for the Early Caribbean Digital Archive was also meeting almost everyday in that space. For those not familiar, the ECDA is a wonderful archival project that, among other things, tries to remix/deform early caribbean literature to foreground the voices of slaves that are mediated and interpolated in white authors’ texts. During that summer the project team was working on customizing a TEI schema to encode their texts in ways that were more in line with their decolonial activist project. And as they went about this, I was overhearing these amazing conversations they were having about the meanings and applications of certain aspects of their TEI schema. How should they tag an embedded slave narrative? What about unnamed slaves? And commodities?

And I’m listening to these conversations and realizing that it is precisely because they are encoding the texts in TEI that the act of encoding literally inscribes these texts with interpretation, that’s why these conversations are even happening; and they’re important conversations! About how we represent our objects of inquiry in the humanities. About the ethics of data representation.

Unsurprisingly, I’m not the first person to have realized this.

Julia Flanders has called this the “productive unease of 21st century digital scholarship.” Flanders is drawing on and responding to on John Unsworth’s definition humanities computing as “a practice of representation, a form of modeling or […] mimicry.” The key, for Unsworth, is that digital representation forces us to grapple, very self consciously, with the fact that our representations are merely surrogates.

So it is this self consciousness that Julia Flanders calls “the productive unease of digital scholarship.” Importantly, for Unsworth, a representation, once expressed in a formal grammar (like XML), is “subject to verification—for internal consistency, and especially for consistency with the rules we have stated” (Unsworth n.p.), I want to say that again because Flanders responds directly to this phrase: our representations are “subject to verification—for internal consistency, and especially for consistency with the rules we have stated.” Flanders writes,

The word verification stands out here, sounding very cut and dried, threateningly technical, a mental straitjacket, but in fact the key phrase there is “the rules we have stated”: it is the act of stating rules that requires the discipline of methodological self-scrutiny … [A]s our tools for manipulating digital models improve, the model stops marking loss and takes on a clearer role as a strategic representation, one which deliberately omits and exaggerates and distorts the scale so that we can work with the parts that matter to us. (para. 23)

So for Flanders, formalization is not an end, but is rather an important step in what she calls “that same oscillating, dialectical pulsation that is the scholarly mind at work” (para. 27). I love that phrase.

And this is what I was witnessing when I was overhearing the ECDA meetings. I was witnessing the oscillating, dialectical pulsation of a whole group of scholarly minds at work. The requirements of formalization meant that discussions could not be tabled, they had to be hashed out and a solution had to be implemented. These solutions, and the encoded texts, are representations, situated and interpretive, inscribed by the agents of their creation.

I was struck by these conversations. I wanted my students to have those kinds of engaging, difficult conversations about writing. I want my students to feel this productive unease.

This actually taps into a larger conversation in the digital humanities around building or making as a way of knowing, or what here I will call critical making. Stephen Ramsay has called building “a new kind of hermeneutic,” and Bethany Nowviskie uses the phrase “productive resistance in the materials” to characterize this way of knowing through map-making in Neatline. She writes:

Neatline sees humanities visualization not as a result but as a process: as an interpretive act that will itself – inevitably – be changed by its own particular and unique course of creation. Knowing that every algorithmic data visualization process is inherently interpretive is different from feeling it, as a productive resistance in the materials of digital data visualization. (“Neatline & Visualization” n.p.)

These conceptions of critical making, have started a discussion about how this way of knowing can inform our pedagogical approaches, how we can leverage DH methods to engage students in this work. Mark Sample describes critical making as a way to mitigate the disconnect between the traditional essay and the tasks that most students will undertake in their professional and academic lives. Instead of writing, Sample asks students to “weave—to build, to fabricate, to design” (“What’s Wrong” 404). We can see in the quote here that Sample emphasizes the complex interplay between thinking and making:

As my students build—both collaboratively and creatively—they are also reshaping, and that very reshaping is an interpretative process. It is not writing, or at least not only writing. And it is certainly not only thinking. (“Building and Sharing” n.p.)

And while Jentery Sayers instead uses “tinkering” as his term for defining a pedagogical approach, his thrust is the same (279). This pedagogically-focused literature, important as it is, remains somewhat on the fringes of digital humanities discourse. Informal discussions of classroom approaches do circulate, however, the digital humanities has done little to engage the undergraduate classroom as a site of formalized research studies. Sayers recognizes this gap in scholarship when he writes, “More formal in situ research needs to be conducted in order to determine—more concretely—how humanities pedagogy can benefit from tinkering” (295).

Here, finally, is where I would like to transition to my project: a teacher research study of a first-year writing course in which students use a custom XML schema of their own creation as their primary method of composition. This project seeks to enter the conversation around critical making in the classroom, while taking a step toward the gap in formal research that Jentery Sayers identifies through the methodology of teacher research. Let’s bracket, for this talk, teacher research and take on the critical making in the classroom. How is it that students composing in XML will be critically making?

Actually in a few ways. As in the story with which I opened, encoding is thus an interpretive act of representation. First, composing in XML refigures the often audience-less and disembodied writing of composition classes as inscription. Students must explicitly represent their understanding of their own writing by encoding their writing in XML.

We can see this explicitness if we look at the the building blocks of XML/TEI: the element, the attribute, and the attribute value. In this example from the TEI consortium, we can see clearly the human readability of XML: the element “seg” marks an arbitrary segment of text, ending with the backslash “seg” highlighted in red, called a closing tag. The attribute, “type” tells us the that this arbitrary segment is of a certain type, and the attribute value, “stutter” tells us that the segment, I-I-I, has been identified as a stutter by the encoder.

But it also tells us something about the values of the encoder or project. Specifically, it tells us that the encoder is interested in identifying stutters. This may give us a view into the values and goals of the project—it could be focused on broadly on language use, or dialogue in novels; it could be a project that focuses on speech impediments specifically, or any other number of goals. The point is that the act of encoding forces a choice, in this case, the choice is whether or not we care about stutters. And this choice is inscribed in the encoded document and can thus be challenged, critiqued, complicated, etc. It requires the encoder to scrutinize and reflect on the goals of encoding, and their methods for working toward those goals. This is why TEI has been used in undergraduate literature classrooms, often as a method of close reading. It forces students to decide and explicitly identify the elements of the text that they care about.

As we shift to thinking about students as authors, I hope we can also see the potential of this self scrutiny, of this explicitness, for increasing the metacognitive awareness of students in their own writing.

Trey Conatser refers to an increase in metacognitive awareness as one of the primary benefits of his own XML-based composition course. Conatser’s project is a great example of a committed teacher taking on a huge task and, by all accounts, succeeding in the learning goals which he set out. My point of contention with Conatser is that he created a schema for each writing assignment which identified the salient elements students would use to fulfill the goals of the assignment.

For example, in his primary source analysis, Conatser requires students to make the following formulation at least once in their documents:

“In at least one paragraph match evidence (details) from the primary source with the interpretations you draw from them:

<seg type=”ev_interp”>

I would call this a top-down approach to markup. That is, the schema is created and then the applied to the documents. In a writing class, this seems unnecessarily prescriptive.

So the second, and, to my mind, more powerful kind of critical making in this course is the making involved in building a collaborative, customized schema. This sees students as designers and knowledge makers, trying to answer a deceptively simple question: what can we say, together, about writing.

I hope to accomplish this through a method of bottom-up schema building through exploratory markup suggested by Wendall Piez in his article “Beyond the ‘descriptive vs. procedural’ Distinction.”

What do I mean by exploratory markup? By that I mean beginning with a very constrained schema, perhaps describing only structural aspects of typical writing: this is a title, this is the author, this is a paragraph, etc. Handing this over to students and having them compose writing assignments while identifying (through tags) what it is that their compositions are doing. Then coming together and collaboratively hashing out the differences until we have one, usable schema we’ve authored together. Then, introducing the students to new writing situations—new genres, new rhetorical situations—and iteratively adjusting the schema throughout the semester. Here you can see prospective workflow where the key element is analysis, which proceeds from the markup corpus either to schema design or back to the markup to make adjustments. The key here is that it is always recursive—new markup always necessitating new analysis and change to the schema or markup.

The idea here is to capture that magic I witnessed in the ECDA meetings—in the tension, the productive unease, of having to take their disparate, exploratory tagged documents, and design (and redesign, and redesign) a schema for describing writing.

Students can actually start doing this very quickly through <seg> tags. Like the example here from a sample document that I marked up in this manner to build the proof of concept. We can see I identify a segment of type “summary” and attribute a quote Tara McPherson. This is very preliminary, but the idea is that the segments of text that the students begin marking up with different attribute values may become elements in their own rights, so there may be a summary element that has a constrained set of attribute values depending on the types of summary the students find useful.

But the XML is useful for more than just prompting schema creation. And this is something I’m only just exploring, but I would like to create a whole visualization engine for the course. I would like to end my talk today by looking at just a few examples of potential visualizations that can be useful both to students and researchers. We can take the document and format and color code it based on XML tags. We can apply rollovers and other methods to render the markup visible in ways that can be useful to other users of the schema, or to those who are not versed in XML. This can be useful both for students reflecting on their own work and comparing their work with peers.

Finally, we can do things like what I call a faceted approach, where you can toggle between different versions of texts. The third example, the markup without the contet, could be a powerful comparative feature for students to compare their texts to other students or texts composed for other assignments; on a larger scale, it could be used to visualize and compare an entire corpus of student work. Employing what Lev Manovich has called large-scale “direct visualization without sampling.”

And that is where I’d like to end, looking forward to the, as-of-now, unforeseen possibilities of this project. thinking about what may be possible when we understand students as builders and knowledge makers in our classrooms. Thank you.

 

Works Cited:

Conatser, Trey. “Changing Medium, Transforming Composition.” Journal of Digital Humanities. 2.2 (2013): n. pag. Web. 15 July 2015.

Flanders, Julia. “The Productive Unease of 21st-Century Digital Scholarship.” 3.3 (2009): n. pag. Digital Humanities Quarterly. Web. 20 July 2015.

Manovich, Lev. “What Is Visualization?” paj:The Journal of the Initiative for Digital Humanities, Media, and Culture 2.1 (2010): n. pag. Web. 18 July 2015.

Nowviskie, Bethany. “Neatline & Visualization as Interpretation.” Web blog post. Bethany Nowviskie. 2 Nov. 2014. Web. 18 July 2015.

Piez, Wendell. “Beyond the ‘Descriptive vs. Procedural’ Distinction.” Markup Lang. 3.2 (2001): 141–172. ACM Digital Library. Web. 18 July 2015.

Ramsay, Stephen. “On Building.” Web blog post. Stephen Ramsay. 11 Jan. 2011. Web. 18 Jan. 2015.

Sample, Mark. “Building and Sharing (When You’re Supposed to Be Teaching).” Journal of Digital Humanities. N.p., 9 Mar. 2012. Web. 18 July 2015.

—. “What’s Wrong With Writing Essays.” Gold, Matthew K. ed. Debates in the Digital Humanities. Minneapolis: University of Minnesota, 2012. 404-405. Print.

Sayers, Jentery. “Tinker-Centric Pedagogy in Literature and Language Classrooms. Collaborative Approaches to the Digital in English Studies. Ed. Laura McGrath. Logan, Utah: Utah State University Press : Computers and Composition Digital Press, 2011. 279-300. Web. 18 July 2015.

Unsworth, John. “What is Humanities Computing, and What is Not?” Jahrbuch für Computerphilologie 4 (2002): n. pag. Web. 17 July 2015.