Laboratory encoding: what matters for the model?
Goals
- Gain experience with laboratory materials
- Reflect on your encoding decisions
Resources
Hammersmith ghost story
The edition IS the model IS the encoding.
Markup depends entirely on your research interests, or possibly the research interests of your audience. Let’s imagine some alternative editions together.
Sometimes the encoded data is the model (ie, you want to directly transform your TEI to HTML), but more often you don’t want your whole TEI document– you want bits and pieces, you want transformed and formatted information, you want to count things or run statistics. In this way the data informs the model, which in turn becomes the edition.
This is all a very complicated way of saying: if it’s not in your encoding, don’t expect to surface it in your edition.
Research approaches
- You’re a Hammersmith cultural historian, interested in social gatherings in and around the historic Black Lion Lane
- You’re a social scientist researching the history of mass hysteria events
- You’re a law scholar interested in the history of self-defense cases
- You’re a literature researcher tracing the roots of ghost stories in early mass media print culture
- Other ideas?
Document analysis
What information should we definitely record?
What’s missing from the text files you were provided?
What kinds of information should we leave out for a given research approach?
How can we accommodate competing research approaches in more collaborative projects? How can we accommodate more ambiguous research approaches for a “general public” type audience?
Auxiliary data
What should live within the document markup, and what should we separate? How should we structure such a document?
How can we avoid repetition and keep references consistent?
View laboratory encoding
What did you include that I didn’t?
What did I include that you did not, or would not include?
(Feel free to critique my TEI here too! or file an issue!)
Preparing our first feature task: create a title list
Working backwards from the feature, how do we think we’re going to build a list of titles? What is step 1?
How do we divide the roles of project manager, researcher, and developer when we think about building this feature up?