GUIDELINES FOR
DONATING BOOKS
First, we are thrilled when someone has an antique
needlework pattern and wants to make a scan for the library!
Here
are some guidelines:
1. Has your book or booklet
or magazine already been scanned?
|
Check the catalog to
see if your book, booklet, magazine or article has already been
scanned by someone else. Also
check our work-in-process
– this is a very rough list of books that have been scanned but not yet photo edited or otherwise processed for upload. |
2. Is your book in the public domain?
|
Check the copyright information of your
book. This information is usually on
the title page or on the inside of the cover.
If your book was published in the |
|
|
|
If your book was published in the |
|
**If your book is not in the public domain,
then it is illegal to make copies of it, and we can’t use it for the library.** |
3. Scanning
books is a lot of work! So that we don’t duplicate work and waste effort:
|
If you have issues of Needlecraft ( |
|
If you have any of the wonderful books by Flora Klickmann that haven’t
been uploaded to the library yet, please contact Helen - enfys (at) dsl.pipex.com
, to coordinate. |
|
If you want to work on typing or editing text (so that we can make
more efficient html files of some of the books) please contact Sonja
Henderson heysonja (at) gmail.com, who is
organizing the image-to-text project. |
|
If you have another book, booklet, or magazine that you would like to
scan for the library, please contact Judith Combs - nornsneedlework
(at) gmail.com |
|
**If you donate a book scan, we will list
your name as the donor at the Antique Pattern Library website in the book
list entry, on the donor’s list (with the URL to your website, too, if you
want), and on the cover page for each book.** |
NOTES ON SCANNING:
|
In many countries, the act of making a scan
creates a new work with its own rights.
We are using the licensing language provided by Creative Commons to try to keep
the scans free for everyone. We also
do what we can to keep our scans from being “lifted” and copies sold at the
online bookstores or eBay. |
|
Scanners and other computer equipment vary a lot, as do our skill
levels. Old books tend to scan as
faded text on yellowish brown background, with foxing, creases, and other
damage. Right now most of the books
are scanned at 300 dpi in full color, then the pages
are cleaned up in PhotoShop and usually converted to grayscale. Sometimes we can then reduce the pages to
150 dpi, and/or process with OCR (optical character recognition) software. The pages are then
assembled into pdf files with a cover page, footnotes, watermarks and
security. |
|
Scans are not usable unless they can be printed
legibly enough for a needle worker to use.
That means (after one of us is done editing)
black text on white background with illustrations as clear as the originals,
if possible. In many old pattern books the illustrations are more important than the
instructions, and some of the old books have almost no text at all. |
|
Right now the best ways are to scan the pages of the books into tif, png, or jpg format in full
color at 300 dpi, or use OCR software to convert the images of the letters to
text characters: |
|
JPG files are reasonably
good for either photo editing or OCR scanning later on, and compress well
enough to email. They are “lossy”, which means they lose a little resolution every
time they are saved, and the resolution is fuzzier when they are converted to
PDF format. |
|
TIF files keep their resolution and can be
read by OCR software, but are usually too big to email, so must be sent via
FTP uploads, or snail mail. |
|
PNG files keep their resolution the best
after conversion to PDF format, and are small enough to email, but don’t read in OCR software. They are the best choice for filet,
cross-stitch, and beading charts. |
|
OCR scans (usually converted to DOC files)
need editing to retype mis-read characters, but the
text stays crisp throughout the pdf conversion process, and the file size is
much smaller. The book illustrations
usually still need to be cleaned up in a photo
editor. Some old typefaces can’t be read at all by OCR software. |
Tips for Better OCR Results
|
When using OCR software, recognition
quality depends greatly on the quality of the scanned image. The image quality may be
improved by altering the main scanning parameters: resolution, scan
mode and brightness. |
|
Resolution. Use 300 dpi
for regular text (font size 10 pts. or greater) and 400-600 dpi for text in
smaller font sizes (9 pts. or smaller). |
|
Scan mode: Scanning documents in the grayscale
mode produces the best text recognition accuracy, as brightness is usually adjusted automatically. If you wish to retain color in the images,
text or backgrounds of your electronic documents, use the color
scanning mode. A medium
brightness value of around 50% should be good enough in most cases. |
Some of the pdf files in the library were assembled with pages made from several methods - for
example, TIF for pages with some text and several illustrations, PNG for filet charts,
OCR for pages with no illustrations. We’re flexible. If
you can make a legible scan, one of us will find a way of making a downloadable
book from it.
Scanning and cleaning
up images of old and damaged pages can be time consuming! And addictive. Some of the books we put up in the beginning
are awful copies, but we're gradually rescanning and
photo editing those, and the newer files are better. We're saving and storing the raw and edited
scans - as our skills and software improve over time, we'll revisit some
issues.
|
Site Map and other links |
jcombs last updated
10may2007