Searching the New Jersey Digital Legal Library

The full-text searching option available for most of our NJDLL collections currently runs on an index which consists of uncorrected OCR (Optical Character Recognition Output) text taken from each of the scanned documents in the collection. OCR text is automatically generated by a computer program that tries to guess what the words that appear on a scanned image are. While this process makes older collections accessible to full-text searching programs it does sometimes result in errors. We currently use the ABBYY Finereader OCR Program, which is recognized as one of the most accurate OCR tools on the market. It usually achieves accuracy rates of 97 to 99 percent of a document's text.

If you run across any nonsensical or unrecognizable words in your search results these are the result of the imperfect OCR process. If you have any confusion about the contents of a document please refer to the DjVu or PDF copy of each document that is available. These two formats contain an exact copy of the original print document so the text displayed to you in these two formats does not suffer from OCR errors. If you are unfamiliar with DjVu, which is a document display format with properties similar to Adobe's PDF format, you can find more information about it on our DjVu Help page.