Be able to download text constructed from the original OCR

[vvasuki]

I wish I could get a diff between the original ocr and the proofread version, so that corrections could be carried over to preexisting copies stored in significantly different formats.

Also, would help calculating/ analyzing the net impact of proofreading, OCR defect patterns etc..

So, it would be wonderful if one could download a text constructed from the original OCR.