Playing Librarian...

  • Article

Those of you who know me (and my family) from beyond my blog know that among my our many passions, one of the biggest is books.  And we've got a lot of them.

A couple of years ago, Valorie got me a Flic barcode scanner and a copy of the program Book Collector.  I've been using it steadily since then adding the books from the biggest of the 4 different libraries in our house (yeah, we've got 4 separate libraries in the house, I did say we read a lot of books - they are: grown up fiction, kids fiction, non fiction and teaching materials).

Since I was fortunate enough to take the month of December off this year (one of the benefits of working at Microsoft for as long as I have is that I get a lot of vacation time, some of which was due to disappear), I decided to take on the project of working through the books in the big library.

Since I've been working on it pretty much every day off and on for the past 4 weeks, I've looked at a LOT of books recently.

I've developed a pretty good workflow (it could unquestionably be improved, but this one works) for the process of scanning books:

  1. Start Book Collector.
  2. Head into the library from my computer with the Flic scanner in hand.
  3. Start where I last scanned.
  4. Take a book off the shelf: 
    1. If the back cover contains a UPC code, check the barcode - if it starts 978 or 979, scan it and go to step 5.
    2. Open the book to the inside front flap.  If it contains a UPC code, scan it and go to step 5.
    3. Put the book on the "to be scanned manually" pile.
  5. Take the books on the "to be scanned manually" pile back to the computer.
  6. Plug the scanner into the computer, which will cause the UPC codes to be read in and let the program search for the books.
  7. Add the books found by the scanner to the program.
  8. For each book on the "to be scanned manually" pile:
    1. Check for an ISBN number on the back of the book (usually near the UPC code).  If it's there, enter it and go to step 9.
    2. Check the spine of the book.  If one of the numbers there looks like an ISBN number, enter it and go to step 9 (I often combine step 8.1 and 8.2 together).
      1. If the number on the spine looks like an ISBN number but is 1 digit too short, try typing it in but add an "X" for the last digit.
    3. Look at the page after the title page of the book - sometimes there's an ISBN number there, if so, enter it and go to step 9.
    4. If no ISBN number is found, put the book on the "to be entered manually" pile.
  9. Have the program scan for the ISBN numbers you just entered, verifying each one as it's found, then add the books if they're correct.
  10. For each book on the "to be entered manually" pile:
    1. Enter the title and author for the book manually
    2. Search for the book.
    3. Walk through the items found adding the best fit ("best fit" can be subjective, I try to find an entry with accurate cover art or an accurate publishers number - but the only books that fall into that final set of books are typically more than 20 years old, so your ability to find accurate information on those books is spotty).
  11. Pick up the books you took from the library and to back to step 2.

 

That's it.  So far I'm up to just short of 3000 books scanned, and I'm in the middle of the letter "S" in the biggest of the 4 libraries.  I've added more than 2 thousand books to the program this  month (wow).

Some things I've noticed in the process...

  • We have a lot of books.
  • As far as the scanning process goes, books fall into roughly five categories:
  1. Those with ISBN numbers on their UPC code - this includes most graphic novels, hard cover books and trade books.  Many new paperback books have ISBN13 numbers on their UPC code, but some still don't.
  2. Those with ISBN numbers in a bar code on the inside front cover - essentially this category contains all paperback books from some time in the early 1990s to the present.
  3. Those without ISBN numbers in a bar code on the inside front cover, but with UPC codes that include the ISBN number.  Essentially this category contains all books from the mid 1980s to the early 1990s.
  4. Those without a UPC code on the book, but with an ISBN number (or sometimes a SBN number) inside the book.  This includes most (but not all) books from the 1970s.
  5. Those without ISBN numbers at all.  This includes most books before the early 1970s (yeah, I've got paperback books that date from the mid 1960s).

For books that have ISBN numbers, the amount of information available about the book depends highly on how new it is.  For those that post-date Amazon and Barnes&Noble (the primary data sources for Book Collector), the information available is quite good (including reasonably accurate book cover images).  For older books, the information available is spotty, usually dependant on the information that 3rd party sellers provide to the various online retailers.

We have an awful lot of books.

I'm only beginning the process of taming the book collection, and I've not even started thinking about dealing with how to maintain the library going forward (but I've got some ideas).  As I said, my workflow above could be improved (for instance, I should use a laptop and take the computer to the library to deal with the "to be scanned manually" pile instead of schlepping the pile back and forth.

But working through the piles has absolutely been an enjoyable process - I've also appreciated the opportunity to re-discover old friends, which is always a good thing.