The HTML view and the TEXT view complement each other.
The HTML text is extracted from the Original document. In addition to retaining some of the formatting, it includes metadata such as Track Changes and Comments when available. The TEXT view contains text extracted from the NormalizedPdf. This includes OCR text for any documents that required OCR as part of the processing. The text from both the HTML view and the TEXT view is stored in the dual search index thereby increasing search accuracy.