Applying Textual Analysis and Digital Visualization to a Poetic Body of Work

Project Personnel

Ray Henry, organizer [resume linked here]
Stacy Waters, consultant

Project Description

A review and selection of both working manuscripts and completed works. Working manuscripts will be scanned to present a visual transformation from an initial moment of creation through redaction and revision to the completed work. Completed works (all extant completed works exist digitally) will also be transformed in two ways. First, a textual analysis of word frequency for the presented set of works, presented visually (i.e., not as a table, but as a tag cloud or something similar). Second, every work will be broken apart into lines which will be recombined (through particular constraints) into new "machine works," which can be interactively generated. These visualizations will be united on a single website.

• Collect works and select the set of materials to be included.
• Develop structure/organization of website.
• Set up skeleton website based on developed structure.
• Scan working manuscripts.
• Compile completed works and apply textual analysis tools.
• Research and develop tag cloud.
• Break completed works into lines--enter into a database or spreadsheet structure?
• Develop and implement "machine" to generate new poetry.
• Add all elements to the site.
• test/evaluate (standards compliance, accessibility, does everything work, etc.).

Updates shown below in red .

Project Work Plan

TITLE: Develop Presentation Website For Visualizing Poetic Analysis - Work Plan
1. DESCRIPTION: tbd
2. TASK LIST
   1. TASK: Preplanning (6-8 hours)
      1. Make paper mock-up of website structure complete
   2. TASK: Skeleton website (12-15 hours)
      1. Research website accessibility and interoperability
      2. Create placeholder pages
   3. TASK: Collect and select works (15-20 hours, not including travel time to Portland)
      1. Gather all potential work - complete
      2. Identify significant work/examples of work
         1. Identify files to be digitized and setup naming system - complete
         2. Attach filename to each non-digital file complete
         3. Change non-compliant digital filenames complete
   4. TASK: Scan ~10 pages of working manuscripts (~3+ hours)
      1. Scan page
      2. Save archival TIFF image
      3. Adjust and add copyright information
      4. Save JPEG version complete
   5. TASK: Analyze completed works - word frequency (10 hours)
      1. Convert MS Word documents to plain text complete
      2. Run analysis tool complete
      3. Record results complete
   6. TASK: Build analysis visualization (10 hours)
      1. Identify visualization scheme and tool complete
      2. Apply data to tool complete
      3. Archive/link results complete
   7. TASK: Break apart existing works (5 hours)
      1. Break works into lines and compile TEI-compliant XML tagging begun
      2. Add lines to database/spreadsheet complete
   8. TASK: Build "poetry machine" (20 hours)
      1. Research mechanism to randomly pull lines from spreadsheet/database complete
      2. Research mechanism to recombine lines - perl - complete
      3. Research mechanism to display poems - complete
      4. Create "machine" complete
   9. TASK: Add all elements to skeleton site (15 hours)
      1. Add selected scans of manuscript pages
      2. Add text analysis visualization complete
      3. Add "poetry machine" complete
   10. TASK: Test and evaluate site (10 hours)
       1. Check and evaluate functionality
       2. Check and evaluate accessiblity
       3. Check and evaluate adherence to standards
       4. Check and evaluate interoperability
3. TIME LINE
4. TOOLS/RESOURCES
   1. Equipment
      • CPU
      • Scanner
      • Archive/back-up media
   2. Software
      • Spreadsheet/database
      • Image editing software
      • Text editor
      • Code development environment (possibly)
      • Website development tools (possibly)

Total time: ~130 hours (~16 hours/week for 8 weeks)

Back to the main page