[post co-authored by Alex Gil and Chris Forster]
Alex: Last week we finally introduced our students to the wonderful world of markup, in our case of the TEI variety. Since I meet with my team on Mondays I got to dip my feet in the cold water first. I’m calling my approach top-down to contrast it with Chris’s approach, which I think we can call bottom-up. I started with the larger structure and then dug my way down to the minutia, eventually arriving at the text itself, while Chris (who will get a chance to tell his side of the story soon enough) started from the text and moved up the TEI.
My approach, if it can be called mine at all, came to me as a natural offshoot of the TEI handout that Chris and I drafted for the class. I opened with the larger theory of mark-up, pointing out that punctuation is an early form of graphical markup and then moving on to the present where mark-up allows our computers to interact more meaningfully with texts. From there I moved to an introduction of TEI through the question of standardization. I wanted my students to understand that TEI is a human agreement on a shared vocabulary more than a precise description of the texts. Then I laid the overall TEI structure on them –I must admit I lost a couple of souls when they saw this:
<teiHeader>[information about the text]</teiHeader>
<text>[the text itself]</text>
I’m aware that for students who have not seen XML syntax, the first encounter can elicit a brain freeze of sorts, and I thought that fanning out the hierarchy gradually would move my students from simple to complex. I used oXygen to open the nodes one by one going from high to low in the hierarchy, eventually arriving at the
Their assignment for next week is simple. They must each take a different version of the same poem by Emily Dickinson, four texts total, and produce a relatively soft TEI encoding of it by Monday. In the end, I think they were relieved to find out we had created a template for them to use. This week we will get our first results from our students, and I will have a better sense of the success or failure of our 1 hour session last week, but after having a chance to see Chris’s approach, I wish I would’ve structured my class a bit differently. I guess I should step aside now, and let him tell it like it is
Chris: Alex’s description of the difference between our approaches as top down and bottom up seems apt. And indeed, my chief interest was trying to help students see the sorts of things that we would be interested in marking up, and what the potential advantages of such markup might be. The long, strange history of markup—the arcane passage from SGML to HTML and its various versions; XML; the misadventure of XHTML and the excitement surrounding HTML5, etc. (for a nice review of this history I recommend this wonderfully brisk Brief History of Markup)—is certainly important, but it can be an overwhelming introduction to a process that I like to simply reduce to “description.” So instead of reviewing TEI and its history, I was interested in introducing students simultaneously to the practice of markup (simplistic reduction: angle brackets) and to the process of thinking about what should be marked up (simplistic reduction: … I don’t think there is one).
To this end I appealed to an example that at this point seems like something of a cliché in discussions of markup: the recipe (I drew inspiration from this page in particular, but one finds similar examples all over the place). I presented students with a recipe and asked them to “mark it up”; what types of information does it contain? How would you describe them? How would you divide it up? Students got the general structure pretty well; recipes have titles, descriptions, lists of ingredients, and then a set of instructions. They nicely noted the distinction between the unordered list of ingredients and the ordered list of instructions (they were thinking in HTML terms here; some of my team has had experience with HTML). They did not catch the distinction between amount of an ingedient (which itself includes the question of what standard of measure we’re using) and the types of ingedient; but that was a helpful opportunity to show a level of markup a little deeper than just basic document structure (inline, rather than block, elements we might say).
I tried to stress what advantages the marked up version of this recipe would have over the plaintext version they had started with. If one had a database of recipes marked up in this way, you could automatically generate a shopping list; you could query your pantry to see what recipes you can make based on the ingredients you have; you could easily double a recipe to make more; and so on. We had already looked at a poem and talked about how we mark it up the previous week, so the transition from describing a recipe to a poem was not as abrupt as it may seem (especially thanks to the great examples at TEI by Example). If we were interested in a poet’s use of rhyme, for example, we might mark rhyme with a high degree of precision, with an eye towards eventually analyzing rhyme data.
This is not an entirely uncontroversial way of explaining things. My focus on what you can do with marked up text appeals to a different set of values than the sheer, rigorous, scholarly description of a textual object. In part this betrays my own inborn tendency towards text analysis; but I think it has pedagogical benefits as well. It helps to illustrate some of the possibilities of markup, and the way in which markup is never (and here, I know, I’m getting a little polemical) simply a matter of objective description.