Inventory for handling scanned documents (was: Better indexing on bitsavers)

[Al Kossow]

 > what do you do with "fold out" pages within documents

The IS520 will scan fold out pages directly in most cases.
For the REALLY long pages, I scan them in overlapping sections
and append a letter to the page number. The first piece is the
left side of the page, so as you page through the document the
overlapping sections scan across the long sheet left to right.

Variable width documents add another level to the scan directory
A convention I came up with a long time ago was to add '_number'
to the file name. Changing scanning size is fairly easy between
sections (changing the scan width, normally) in the program. So,
if you scan a doc that has an 11x17 page, you'll end up with the
following directories

foo_f
foo_b
foo_f_1l   (the 'l' is a marker that let's me know it's a long page)
foo_b_1l
foo_f_2
foo_b_2
..and on an on for all the alternating page sizes

Then I have to flip the filters around for all the differing page sizes
and orientations

Eventually, this produces sequentially numbered tiffs of differing sizes
which are then 'tumbled'

..also forgot that I rotate 8x11 pages to 11x8 if the page has horizontal
data on it.

DEC tech manuals are the worst.. they will often have a single 11x17 page
so I'll end up with dozens of directories that need to be manipulated.