"Lee" == Lee Passey <lee@novomail.net> writes:
Lee> _Pride and Prejudice_ was part of the "Harvard Classics" Lee> collection, and is so indicated by the Gutenberg Bibrec Lee> (which I would still like to see copied to readingroo.ms). Lee> I would suggest you re-run your experiment against Lee> http://www.archive.org/details/harvardclassicss03elio or Lee> http://books.google.com/books?id=7dgNAAAAYAAJ. It might be a different revision of the Harvard Classics, since there are a few differences not due to OCR. Some systematic: [-dare say-] {+daresay+} 29 times and some occasional, like: [-quite-] {+very+} {+you+} (added/removed) [-may-] {+might+} [-daughters-] {+nieces+} {+very+} [-that-] {+so+} {+had+} [-hear-] {+overhear+} {+all+} [-out-] {+very+} [-making themselves-] {+being+} [-more pleasant-] {+pleasanter+} [-is-] {+was+} [-would-] {+will+} (I stopped at 5%) and several punctuation differences. The PG version seems very clean, but a bit more "modern".