How to process bound items for CME
1. Sequence
Consult the master list of CME items at
http://www-personal.umich.edu/~pfs/med/cmeitems.html#bound
Process items in the sequence given there, if feasible. This
list will also provide the NOTIS id# for each item (when one
is created); the instruction-set two-letter code (e.g. "NP");
a short title; and a more or less up-to-date progress chart.
2. Collate
- Using the Master List, identify and locate the appropriate volume
for the next item (e.g., Englische Studien, vol.12).
- There should be a short printed description tucked into the volume
itself (if there is more than one item in the volume, make sure
you pick the right description). Use this to locate the correct
pages in the volume. Some items appeared in parts and may therefore
appear contain more than one range of pages in that volume, or
may be spread over two or more volumes. The description in that
case will be in the first volume.
- Check for missing pages and physical damage (holes, rips, ink blots,
handwritten notes). If any are found, either correct them on the
spot (erase pencil marks, for example); or mark them with a sticky
note and correct them later. Some wear and tear is acceptable:
most tears will not show up on the scan; ragged page edges that
involve no lost text are normal; and some ink marks that do not
obscure the text can be lived with. Fix what you can. Missing or
obscured text can usually be repaired with a photocopy, or cut-up
bit of photocopy, made from a second copy of the volume: many of these
volumes exist in multiple (especially reprint) copies in the library.
Sometimes the second copy may be in the stacks; more often it is in Buhr
or at the Middle English Dictionary offices.
- Create a 'map' that associates the page sequence (and therefore the image
sequence) with the page numbers physically printed on the page. The map
for each item is a separate 'sheet' in an Excel spreadsheet located here:
library_hti\work\users\shared\cme\cme.bound.xls
- See if a sheet has been named for your item yet. If so, it will be
named either with the NOTIS ID# for the item (e.g. "ACK1432") or,
if no NOTIS ID has yet been created, with a temporary name based
on the instruction-set code (e.g. "NP-temp"). As NOTIS ID numbers
become available, we will be replacing the "temp" series of names
with them.
- If it hasn't been created, make a copy of the "template" sheet (which
should be the last sheet in the workbook), place it last-but-one
in the workbook, and rename it with the NOTIS ID of the item, or
with the "temp" name, as described in (i).
Method: (a) click on tab
of the sheet named 'template'; (b) pull down EDIT-MOVE OR COPY SHEET;
(c) click the box "create a copy"; (d) scroll down the list of sheets
and highlight "template" (which should be next to last in the list);
click "OK". (e) when the new sheet appears, it will be named "template(3)":
double click on the tab with this name, till the name appears in inverse
video; (f) enter the new name for this sheet.
- Look at the brief description stuck in the book; make a guestimate of the
total number of pages in the item. Fill in that many rows of the leftmost
column of the sheet with page-image numbers, like this:
| Page Imgs |
|---|
| P.00000001 |
| P.00000002 |
| P.00000003 |
| P.00000004 |
| P.00000005 |
| P.00000006 |
| P.00000007 |
| P.00000008 |
| P.00000009 |
| P.00000010 |
| P.00000011 |
Use the Excel auto-fill function to fill in the numbers.
Method: type the first two or three of these numbers manually, highlight the
block of cells that you just filled in, then click and drag the lower right
corner of the highlighted block down the column; a little pop-up window will
show you the numbers incrementing till you reach your guestimated total.
- Page through the book, recording in the second column of the sheet
all the printed page numbers in sequence. Leave a blank cell for every
page without a printed page number, even if the page is completely blank.
Your sheet will now look look something like this:
| Page Imgs | Pg.no. in bk. |
|---|
| P.00000001 | |
| P.00000002 | ii |
| P.00000003 | iii |
| P.00000004 | iv |
| P.00000005 | v |
| P.00000006 | |
| P.00000007 | 2 |
| P.00000008 | 3 |
| P.00000009 | 4 |
| P.00000010 | 5 |
| P.00000011 | 6 |
- Record certain features in the third column: title pages (very rare or
nonexistent in the journal items; recorded with a "TPG" code in the third
column), and plates, fold-outs, and other illustrations that will potentially
require special treatment (record illustrations or plates with "ILL"; foldouts
with "FLD" and color foldouts with "CFLD").
| Page Imgs | Pg.no. in bk. | Features |
|---|
| P.00000001 | | |
| P.00000002 | ii | |
| P.00000003 | iii | ILL |
| P.00000004 | iv | |
| P.00000005 | v | |
| P.00000006 | | |
| P.00000007 | 2 | |
| P.00000008 | 3 | |
| P.00000009 | 4 | |
| P.00000010 | 5 | |
| P.00000011 | 6 | |
- Save the file.
- Check off column "B" in the progress chart on the wall.
3. Evaluate.
Decide if this item can be successfully scanned on the flatbed scanner.
If it can, go on to step 4. If it can't, put the item aside and move on to another, putting
a "d" for "deferred" in column "C" in the progress chart on the wall.
Reasons to reject a book as suitable include: (a) excessively narrow gutter or
tight binding (so that text will be lost to gutter shadow, or distorted by the
curvature of the page); (b) excessive fragility (so that even careful handling
is likely to damage the book seriously).
4. Scan.
Start by printing out the spreadsheet "sheet" for this item. Then...:
- You should find the computer on (the monitor may be off) and the scanner on. If everything is off,
turn on the scanner first, wait for it to go through its warmup exercises, then turn on the computer.
- Start up scantool and specify "00000001.tif" as the first image file (assuming you
are starting a new item); specify D:\Bound\xxx as the directory in which to
place the image files, where "xxx" is the same name you used to to identify the
spreadsheet sheet, i.e., either the NOTIS ID no. or the temporary ID in the form
"NP-temp". The folder should already exist, but if it doesn't, create a new
one in the D:\Bound folder.
- The controls should already be set to 600 dpi / text-and-halftones / image
quality "auto", deskew CHECKED, despeckle NOT-CHECKED, no boxes checked
in the "input document" area.
- Select a page size, remembering that they are listed in the form
height x width. Most of our journals will scan alright with the
size 8.5 x 5.5 inch. Some may require a custom size, e.g. 8.2 x 5.2,
which I used with the squat Archiv volumes. The ideal is to avoid
losing text and to avoid scanning gutter clutter when the book is
placed contiguous with the north and west edges of the scanner plate.
- Scan each page separately. The "top" of the scanner is the western edge, the
one furthest from the operator. Scan alternate pages right-side-up and
upside-down.
- Feel free to move the book around a bit on the bed so as to minimize edge
clutter and so as to center the image as much as possible. Some of the
upside-down scans will require that the book be placed with the bottom
edge of the page overlapping the edge of the scanner bed a little bit,
so as to center the text vertically as accurately as possible.
- Check the image numbers in Windows explorer as you go, comparing the file
number with the file number listed in the spreadsheet for this page.
Catch any mismatches before you've gone too far.
- Check the final file number when you're done, to make sure that you
have the same number of files that you thought you'd have.
- Check image and page quality as you go. Don't be afraid to stop and make
adjustments (rescan if necessary); to erase stray pencil marks; or to
fix physical damage that wasn't caught earlier.
- Check off any item that you've finished in column "C" on the progress chart on the wall, with
the date and your initials.
5. Edit the images
- Copy (do not "move") a directory or two to Novell space in
library_hti/work/users/shared/cme/imgedit
- Using ACDSee survey all the images in the directory, checking especially for
lost text at the margins (this is more important for tight-guttered,
narrow-margined books, and may be skipped with some of the others).
- Rescan any that turn out to be missing text; copy the resultant rescans
so as to overwrite the bad images in both the original directory on
local drive D:\Bound and the copy directory on shared drive
library_hti/work/users/shared/cme/imgedit.
- Using Paint Shop Pro,
- rotate the upside-down pages 180 degrees.
- if it can be done easily, erase edge clutter/shadow on all affected
pages.
- save (overwriting the files in the imgedit directory).
- When you're done with an item, copy the directory (folder) with the edited
images to local directory (folder) D:\rebound on the computer attached
to the scanner.
- Check off column "E" on the progress chart on the wall.