The National Register of Historic Places has begun the task of digitizing the National Register documentation on listed properties and is looking for states, federal agencies, and tribes to partner with in completing this exciting endeavor. If you are interested please contact the National Register at our mailbox . Beginning by commissioning a study by the Cornell Institute for Digital Collections of our own collection, we next scanned the text of multiple covers which are now online. The subsets we are currently scanning are properties owned or managed by the National Park Service and properties owned by GSA which includes all of the nomination text, photographs, and selected USGS maps.

We scan logical subsets of National Register documentation and complete them as funding becomes available. We do not use OCR, or optical character recognition, on the nomination text opting for as faithful a copy of the actual nomination as is feasible. We strike a balance between preservation and access by scanning at sufficient resolution to have an 'archival' copy while at the same time providing derivatives of this information for information sharing over the Web via the commonly used Adobe Acrobat pdf, or portable document format. We do not digitize those portions of the text or any photographs of restricted information on the assumption that if this were electronically stored then it could probably be hacked. We provide access to this information via a Web enabled database with links to two pdf files for each property, one for text and one for photographs. While the Web technology will grow in sophistication and power as we move forward with new technologies, we also allow others to simply download pdf files for incorporation into their own state, federal, tribal or other information systems. We scan maps only in those cases where there is not an adequate verbal boundary description since Internet mapping has evolved to the point where feeding proper coordinates from a database can reasonably plot a property on a topographic map. We keep the pointers to a scanned nomination as only one of a number of other links to resources managed by the National Register Information System. We keep copies of this information at a secure offsite location.

We prepare all nominations for sending out in batches in-house to our scanning contractor. All properties are bar coded, checked out when they leave and checked in when they return. We update our database for missing properties and update a flag when scanning is completed. Prep includes putting the files in the right order and clipping what is to be scanned. We xerox pages that are too fragile for scanning. In cases where there is restricted information we make a copy of the nomination text and blacken out all sensitive information. Items like correspondence are not scanned. Text is delivered to us on CD ROM by a scanning contractor as both individual pages and in multi-page uncompressed TIFFS so, as mistakes are uncovered, or we receive additional documentation, we can simply add a page and remake the multi-page TIFF. Photographs are also delivered as uncompressed TIFFS, however there is no multi-page TIFF made due to the large size of these files. All photographs and all documentation describing photographs is scanned on the assumption that, while expensive, we are only going to get one crack at it. While the photographs are fed through a high speed scanner, some tweaking of the photographs for contrast is done at the contractor as necessary. Captions are scanned without cropping and rotated if askew. Sketch maps and topo maps are scanned as necessary after first reducing to an 8 1/2" by 11" format and xeroxing in grayscale. Topographic maps are only scanned when there is no adequate legal boundary description since we feed UTM coordinates for display on the Topozone web site. The master copy is always a TIFF, although photographs are converted to JPEG images for speedier delivery over the Web. We make all of the Acrobat files in-house to save money, as well as make JPEG images for photos. Images are currently stored on a dual processor server with a dozen 36 gigabyte drives configured with RAID 5 for redundancy. Offsite storage is kept on 100 gigabyte Ultrium tape drives. We scan at 400 dpi bitonal uncompressed TIFF for text, 300 dpi grayscale uncompressed TIFF for photos, and 300 dpi grayscale uncompressed TIFF for maps (only as necessary). Text is served through a single Adobe Acrobat file using the eight digit National Register number with pdf as the extension on the file name (e.g. 66000005.pdf). Photographs are stored similarly except they are stored in a photo directory. Maps scanned in grayscale are stored behind the photographs in the photographic file. Thumbnails are available for previewing in Acrobat and the 'fast web' option is enabled for speedier delivery. We make the original TIFFS available to partners on electronic media.