Input data

TCP data:

• 25,xxxx texts now
• 4,000 more "soon"
• 5,000 more somewhere further up the pipeline
• if EEBO has 125,000 texts, then we're on a path to have > 30% of them.

Example markup

Encomia-related counts:

• 665 (of 25,xxx) texts have encomia.
• 2,335 encomia
• ~2,400 names

Problems w/the TCP data:

• Encomia signatures are not always marked (e.g., in Latin encomia).
• Encomia names are not standardized
• I don't understand the TCP's attribution of authorship (the "Thomas Cecil" problem)./li>

Conversion process

Workflow:

• Automated SGML-to-XML conversion (not to full TEI)
• Automated extraction of encomia and loading of our encomia database.
• Automated merging of TCP metadata (dates, authors and titles) and our encomia database./li>
• Limited automatic correction/spelling standardization of encomia signatures.
• Lots and lots of manual review and correction/spelling standardization.
• Extract data from the database, then build the visualization.

The underlying data

Problems:

• Some character-related issues (long s's) in the SGML-to-XML conversion.
• Some slop in the correction/spelling standardization of encomia signatures.
• Problems with encomia signed by initials, or unsigned, or signed in some non-typical (Latin) way.

Visualization

• The relationships are complicated and tangled, so the visualization is too.
• The tool contains the best way to visualize these relationships.
• We're at the edge of what is possible in terms of interactivity in the browser.
• The circle sizes are suggestive (bigger means more), but not mathematical.

The visualization tool.

Needed Enhancements:

• Google-map style zoom and pan.
• Better/sharper colors.
• Don't display people with only one connection (i.e., simplify the graph).

Next?

• Clean up some issues in the programming (visualization enhancements, conversion issues, database design).
• Another brief round of correction/spelling standardization.
• Explore the data for other kinds of possibilities (types of signatures, book subject matter, etc).
• Staff a summer 2011 project. Focus on fixing encomia signatures, on evaluating the data against paper bibliographies, and on using the data to make preliminary arguments.