Access and preservation are considered throughout the process of
digitization. We have trained our students to capture and store master
TIFF files locally on our production computers, each is backed up with
a RAIDed data storage unit that protects all short-term work. Master
images are then cropped and de-skewed, and color profiles are
assigned. Students embed pertinent collection and keyword
information to master TIFFs using Adobe PhotoXMP. When this is
done, NISO Z39.87 technical metadata is extracted using the JHOVE
API ( JHOVE - JSTOR/Harvard Object Validation Environment,
2009). JPG derivatives are generated and moved to the CONTENTdm
production server; master TIFF images are copied to our SAN
(Storage Area Network) for long-term storage. If working
collaboratively with partners at a geographic distance, JPGs are often
stored where the partners can access them via the Web for metadata
generation and quality control.