The readings this week discuss the issues surrounding historical writing and data usage, and how to represent data from sources.  The second reading seems more technical and advanced, because I haven’t learned any of the coding language(s) yet.

Reading Highlights

Gibbs/Owens

This article points out that even though historians are using digitized and born-digital sources in their research, they are still writing in traditional forms.  They call for a “new level of methodological transparency in history writing” and say it may be time to de-emphasize “narrative in favor of illustrating the rich complexities between an argument and the data that supports it.”  I think the sentiment behind the last statement is controversial.  I am not sure I would be comfortable explaining my use of digital methods, as part of my historical writing.

At present, I wouldn’t write about my experience in the archive, or my failures at finding sources, or how I came up with an insight – in a scholarly article about a historical topic.   I am not sure what that kind of transparency has to do with a scholarly argument or why they should be woven together, necessarily.  I do believe it is essential to combine digital methodologies with traditional history writing methodologies to get a more complex and rich understanding of our sources. Yes, the argument is affected by this mixture of richness and complexity.  The historical argument isn’t about mechanics, it’s more about contextualization.  Anyway, this idea is a source of professional tension that I can relate to.

Now, if this point is more about teaching other historians by example, then I understand it. Blogging, workshops, conferences, classes like this one are definitely about digital methodology and are by nature transparent.   These are part of the individual historian’s experience and should be shared, as we are doing in this blog.

Spaeth

In the Lab readings we have been learning about the differences between source-oriented and data-oriented approaches to database design.  Spaeth seems to be advocating a source-oriented database approach that is more flexible because it addresses all the various kinds of information a textual source, such as an inventory list, contains.  In this discussion Spaeth tries to provide a solution to the ordered hierarchy of content overlaps that occur when you have “fragmented texts, implied and ambiguous data, and cross-references” that violate hierarchical assumptions. I have to admit that all of this is difficult for me to understand.  I realize he decides to ‘encode multiple hierarchies’ to  try to represent how items and rooms relate to each other,  in the inventory list.  I know XML is involved because it is more flexible and allows for consistent analysis (61).  Exactly how XML technically does that is what I still have to learn.