@dheadshot @timhutton Ran both parser.py and Pandoc on my extracted Twitter archive (which is ~3.8GB by the way). output.md is 18.9M, output.html is 40.5M.

Note that I joined Twitter in June 2009, which should explain the size.