Every scanned page of the Deddington News has also been subjected to OCR. OCR stands for Optical Character Recognition, where specialised computer software tries to convert the pictures of text in to actual text files - as if they had been typed in by hand. This is an inherently very difficult task, even with the latest OCR software. For example, there a great variety of fonts and much hand written text in the originals.

This means that the resulting OCR text is not directly worth viewing due to the results achieved. Even one or two percent incorrectly recognised letters can make it quite hard to read. Typically one would hand correct the results of OCR, but clearly this is not possible with the large volume in the Deddington News Archive.

However, we are able to offer a search facility based on the OCR text. This presumes that words you might want to search by will be repeated sufficiently that you can often find the page or pages you are interested in. It must be stressed that this process is not infallible, and certainly should not be relied upon to find all occurences within the originals.

 

Deddington News Text Search