marginalia

Working for the William Blake Archive has been exceptionally exciting this semester. Two major project teams are striving to arrive at a better understanding of how to encode some of Blake’s least audience-friendly works: The Four Zoas and his marginalia. The process of approaching these works has required patience, and for every successful moment there have been multiple failures. But these failures are not meaningless, or at least I like to think so. My recent encoding attempts of Blake’s marginalia have not been used by the team as a model of what to do. Quite the opposite, my encoding attempts have consistently been used by the team as examples of what we want to avoid, and I think that’s useful.

Since we are still in the exploratory stages of creating an encoding schema for Blake’s marginalia, every couple weeks the team chooses a couple marginalia objects, and each team member attempts to encode those same objects using the ideas and new approaches we discuss in our meetings. This allows everyone to test our ideas and see what works and what doesn’t. For the past few weeks, I have found myself working in the office alongside other members of the marginalia team. All of us are usually working on the same object, so we usually end up asking each other questions as we work. Sometimes this causes us to work similarly, but at other times, it exposes aspects of the team’s schema that are not clear or that are not adequate to problems posed by marginalia documents.

Sometimes the best way to expose those problems in the scheme is for those who are in the office to intentionally encode an object differently from one another in order to demonstrate the array of possible encoding options. Lately, my work has produced some clunky but schematically precise XML files, and as a result, the team has recognized why certain aspects of the schema need to be abandoned and new ones created. Did I fail to find the schema? Yes. Did I create aspects of a new schema to help our project succeed? No. But my “failed” attempts have exposed flaws in our encoding approaches and helped clarify for the team what we hope our final schema will be able to achieve. I have to believe that contribution means something because it feels a lot like the desired result of teamwork.