Where DO all those records on Ancestry.com come from?

I have never really stopped to think about how a document such as the US Census finds its way onto the Ancestry.com web site, but a presentation at the 2009 NGS conference titled “The Journey of an Ancestry Record: How an Image gets from the Archives to the World “ by Laryn Brown really made me appreciate all that goes into such a journey.

Just obtaining the rights to documents sometimes takes several years to negotiate: the length of time from start to finish for some paper documents can take up to 5 years!

Ancestry has several microfilm scanning operations running around the clock in several locations around the world. To index their documents they employ specially trained Chinese typists to do character-recognition typing of documents. They also employ a group of English speaking typists in Uganda to translate more complicated documents that require an understanding of context.

Ancestry makes a great effort to insure that their transcriptions are as accurate as possible. However, 20 to 30% of the images that they deal with now are literally "indeterminate". Even expert, full-time professionals see different things in repeated blind testing because the images are so faint or have been damaged. Ancestry depends on subscriber feedback to correct these problems, only possible when somebody knowledgeable about the context of the information views the image (i.e., you find your grandparents on a census and can identify an otherwise illegible entry as their son or daughters name).

It was interesting to note that Ancestry anticipates exhausting most of the available microfilm based information within the next 10 years or so.