It's something a bit different. When you're dealing with hundreds of thousands of titles in a big branch library or the millions of titles in a city library system you find out that you can't trust ISBNs or the commercial data that can come (it doesn't always come sometimes) from vendors. ISBNs (and their cousins the ISSNs) are nothing but serial numbers and the way they are applied to books varies greatly from one publisher to another. When you can really trust the ISBN (or it's not there, as is the case for many titles in million-title collections) you often can't trust the book data that can come "free" from the vendor or distributor, because it's too irregular. It's OK for stores with just 10,000 titles but it's not precise enough for a library with 100,000 titles or to permit fast retrieval through that million title database coming from so many different sources. Which author exactly? The vendor writes "John Q. Smith" but is it John Quint Smith or John Quentin Smith? Also, the vendor is irregular in dealing with "problem" names: McCluskey or MacClusky? Then, another vendor is regular but not in the same way as that other regular vendor. And so on again, with the way the titles are noted from one edition to another, or in other circumstances
That's why even though librarians usually have access to "free" book data from their vendors they always have to start from zero when it comes to cataloguing the book, describing it according to the rigorous demands of AACR instead of the nearly free form manner of the publishers and vendors. If they don't they'll end up with a mess of unfindable titles. That's why it can cost such a high price to process a book, apart form the price of the book itself.
Some of the largest libraries split up the work to lower the cost. A library clerk will open the boxes and input the book and its existing data into the system (often with the same scanning pen as in a bookstore) and then check to see the matches with books previously ordered, flagging things as necessary. A library technician (someone with a junior college degree in library techniques) will check the incoming book in the database and match it to existing cataloguing records. If an AACR level description already exists from a vendor (like the OCLC consortium, a huge co-op which does not sell a single book but specializes in selling quality AACR descriptions of titles, sometimes doing the cataloguing themselves, sometimes reselling them from Library of Congress descriptions) the technician will check to see how well it matches the bought title and decide to buy or not that precise description.
About 80% of books bought already have an available description, in the US. They go through what is called "copy cataloguing" because a library person then digitally copies the description that was sold to them by OCLC or another library cataloguing company.
The 20% remaining goes through "original cataloguing". That's when the library technician decides to send the job to one of the library's own cataloguing librarians (someone with a Master's degree in Library and/or Information Science, usually with a specialisation in cataloguing) so that the book can get the exact description it needs, in order to be sure it can be found later.
Paying these persons (or OCLC or others) for the intellectual work of describing a book to high standards (essential for finding anything when dealing with that number of not-always-best-seller books) or for checking to see if a book description is correct accounts for the bulk of the "processing" cost of the book.
That's what I'm talking about when I bring up the extra cost of having to find books in a 100,000 title collection instead of a tiny (for a library) 10,000 title collection. The part where you physically stick a Dewey call number (or alphabet letter(s) for novels) on the spine and the library's own "zebra code" on the book is a relatively minor cost.
no subject
Date: 12 Jan 2009 03:30 am (UTC)That's why even though librarians usually have access to "free" book data from their vendors they always have to start from zero when it comes to cataloguing the book, describing it according to the rigorous demands of AACR instead of the nearly free form manner of the publishers and vendors. If they don't they'll end up with a mess of unfindable titles. That's why it can cost such a high price to process a book, apart form the price of the book itself.
Some of the largest libraries split up the work to lower the cost. A library clerk will open the boxes and input the book and its existing data into the system (often with the same scanning pen as in a bookstore) and then check to see the matches with books previously ordered, flagging things as necessary. A library technician (someone with a junior college degree in library techniques) will check the incoming book in the database and match it to existing cataloguing records. If an AACR level description already exists from a vendor (like the OCLC consortium, a huge co-op which does not sell a single book but specializes in selling quality AACR descriptions of titles, sometimes doing the cataloguing themselves, sometimes reselling them from Library of Congress descriptions) the technician will check to see how well it matches the bought title and decide to buy or not that precise description.
About 80% of books bought already have an available description, in the US. They go through what is called "copy cataloguing" because a library person then digitally copies the description that was sold to them by OCLC or another library cataloguing company.
The 20% remaining goes through "original cataloguing". That's when the library technician decides to send the job to one of the library's own cataloguing librarians (someone with a Master's degree in Library and/or Information Science, usually with a specialisation in cataloguing) so that the book can get the exact description it needs, in order to be sure it can be found later.
Paying these persons (or OCLC or others) for the intellectual work of describing a book to high standards (essential for finding anything when dealing with that number of not-always-best-seller books) or for checking to see if a book description is correct accounts for the bulk of the "processing" cost of the book.
That's what I'm talking about when I bring up the extra cost of having to find books in a 100,000 title collection instead of a tiny (for a library) 10,000 title collection. The part where you physically stick a Dewey call number (or alphabet letter(s) for novels) on the spine and the library's own "zebra code" on the book is a relatively minor cost.