Warning: 100% geeky librarian content

You've probably heard of ISBNs and know what it's used for. You might know what the acronym stands for and that there are ISBN-10 and ISBN-13 numbers. Or maybe even some of ISBN history, such as which came first. That's basically all I knew - up until today, that is. Now I can be some kind of a genie and determine what type of book you have if you just give me the ISBN.

Let's pick an example from the book I have right here:

978-0-06-147901-4: This is an ISBN-13 number. Because there are thirteen digits.

978: In the U.S., the first three are pretty much always going to be 978 with a possibility (that will grow in the future) of it being 979. Apparently 978s will "run out" in other countries before the U.S.

Another way of looking at barcodes (barcodes without 978 in front): "Every EAN begins with a 2 or 3 digit prefix, which indicates the country of origin. EAN’s for companies registered in France, for example, might begin with the prefix 34; Japan’s prefix is 49. Since the book industry produces so many products, it has been designated as a country unto itself and has been assigned its own EAN prefix. That prefix is 978 and it signifies Bookland, that wonderful, fictitious country from where all books come." link

0: Published in an English-speaking country.

06: Published by the conglomeration known as Harper & Row / Collins / HarperCollins

147901: specific ID for the specific book and its specific edition, which needs a site like isbndb.com or even just a library catalog (preferably WorldCat since few libraries have a good chance of having your edition of your book)

This might not seem like a lot of interesting information to some people (like Colin, who says "Just the country and publisher?"), but for someone who's typed up and played with tons of ISBNs (okay, not technically tons, probably just a mere one or two tons) it's an amazing wealth of information that was staring me right in the face and I didn't know about. Sure, I picked up a couple things along the way.


First, I realized I could tell if it was a newer ISBN-13 if there was 978 in front of it. No need to count the digits. ISBN-10s are the older version, first created in 1970. However, in 2005 they started implementing a longer number because they were running out of ISBN-10s. But, as you might have cleverly noticed, the first 3 digits are currently all the same. "So why don't they just stick them to the beginning of each previous ISBN-10 and call it good?" Well, that's pretty much what happened. (Except for the last digit. Keep reading to find out about that magic last digit.) It seems the big reason for implementing ISBN-13 was so that it wouldn't be a hassle to start using 979. This is a lazy but proactive way to double the amount of numbers. 


A little tangent: Have you ever wondered why there seems to be two barcodes right next to each other on the same book? I have. Now I know: the little short one is the currency and the recommended price. I don't need to look it up to see that "51899" means "18.99 USD." Why? Because right underneath is says "USA $18.99." But on books that lack this helpful redundant human-readable tag, this is a sneaky way to see if a bookstore is ripping you off. Personally, I wouldn't have paid 18.99 for this or almost any other book, which is why it's borrowed from the library.


A micro-tangent to the little tangent: Excited to use my new-found powers, I looked at the back of some other books. One says "90000." Eh? Well, that just means the publisher never chose a price for that book. If they did officially make a new price, they are required to make a new barcode reflecting the change.


Second, I wrongly assumed that the very last digit, which stands all by itself, was to distinguish between different editions. I purposely ignored the last digit until now (again, you might have noticed this). Because I discovered what I think is pretty nifty - that digit is a checksum. I recognized that from playing with linux; it's a separate file you download when installing some programs and it is highly recommended that you compare the very long number to the number in the checksum file. If they match up, great - the install went well. If not - groan and start all over again.


However, the idea with having a checksum in an ISBN is to correct human mistakes and to make sure all the other digits are correct. Wikipedia explains the mathy bit here. Basically you multiply each digit by 1 or 3 alternately, add them up, divide the total by 10 then whatever the remainder is, subtract from 10 and you have the checksum.
9x1 + 7x3 + 8x1 + 0x3 + 0x1 + 6x3 + 1x1 + 4x3 + 7x1 + 9x3 + 0x1 + 1x3 =
9 + 21 + 8 + 0 + 0 + 18 + 1 + 12 + 7 + 27 + 0 + 3 =
106/10=10 with remainder of 6
10-6=4
"ISBN is 97800161479014. 4=4. Yay!"


Now I'm aware of the fact that it can be much more complicated than this - all the different codes within the ISBN can be different lengths, yet it still works out if you omit the dashes. If you care to see why, go ahead and browse the source links:
http://en.wikipedia.org/wiki/International_Standard_Book_Number
http://en.wikipedia.org/wiki/List_of_group-0_ISBN_publisher_codes
http://www.isbn-international.org/faqs/view/5
http://www.barcode-us.com/isbn/isbnSymbols.html

To me, and probably some other fellow lib sci majors and grads, this is quite intriguing if not totally awesome. But I do know that this blog is more of a personal, catch-all blog than purely for librarians, so I will spare the rest of the population who doesn't get excited about ISBNs from expanding an already long post. Plus, it's taken me a fair part of the day (on and off) to get this written up. Watch for the next post on what made me keep wandering away from blogging today ...

Brownie points for those who find out what book I'm currently reading!