Maybe I'm the only one that needs to do this, but it works for me.
Fold3.com
.requires users to download images from record sets, such as Civil War pensions, one document at a time. It takes a while. I usually save all these documents in one large folder and use the "page x" file name that
Fold3.com
.assigns to each image--just to help me keep track. The name of the folder contains my "source" information.
Then I open up Microsoft Word and import all the images in at once to a Word document. Then I set a footer with the type of file and page number. The page number corresponds to the image number. I can take notes on the pages, print them out, etc.
No comments:
Post a Comment