As genealogists, we often search pictures of texts, for example when we use newspaper websites like Delpher. Often, automatic character recognition techniques (OCR) have been used to convert the images to searchable text. These techniques aren’t perfect, especially with poor quality ink, old fonts or digitized microfilms.
The mistakes that OCR techniques make are somewhat predictable: an m gets recognized as rn, an e as a c, and an l as a 1. By searching for the versions with and without a predictable OCR error, you may find surprising results.
For example, when searching for Hamerstein , also try Harnerstein to search for cases where the m got mistaken for rn. In the case of Delpher, Harnerstein will give you just one hit. But if that’s an article about your ancestor, wouldn’t you be glad you found it?