What is Cataloging?

The title to today’s post has been a question floating around in my mind for the past several months. As a matter of fact, I wrote an answer to that question a while back, and when I re-read it, I was very much disappointed with my answer. As you probably know, I am in two cataloging courses right now, and to that skill and field is very much on my mind this semester.

When I wrote that original post, I took a very narrow view of cataloging. Really, I discussed what I knew at the time, but due to what I have learned from class and from fellow catalogers I feel better equipped to answer the above question. That is not to say, though, that I won’t try and answer it again.

Cataloging is a specialized term librarians use to describe how they create records for items in the library’s collection to be used in that library’s catalog. That’s a pretty narrow definition, though. Let’s take a step back and look at this outside of just the library, and their traditional collections of books and music.

My preferred definition is that cataloging is the generation of metadata about an object through the use of accepted standards, and that metadata is then presented in a standardized framework.

Wow, that’s pretty jargon-heavy, isn’t it? Allow me to break that down a bit.

Metadata is commonly defined as data about data. Still confusing, I think. It certainly was for me when I started learning about all of this. To use an example, metadata is like the song information you have in iTunes. The title, album, artwork, etc. are all data that describes that item (song). How about another example - when you upload a photo to say, flickr, you create “tags” that describe, to you, that photo. So, a photo of a cat might be tagged with “cat” or the cat’s name, or whatever. When you tag, or create information about a “thing” that then describes that thing, you are generating metadata.



Look there, Metadata on flickr!

Feeling better about what metadata is? I hope so, because it’s only going to get more complicated from here.

Much of the metadata generated today is created through the use of standards. This ensures that metadata is generated in a consistent format with standardized grammar, sources of information, and language. Would you like another example? Well, let’s think about iTunes again. A metadata standard would help you determine what the accepted sources for the metadata are. Where does the album title come from - the spine, the cover art, the CD itself? Standards help us have a consistent spot to look for this information. These standards also help us determine how to phrase the metadata in a consistent manner. Once again, it would help us determine how to present the album title, and what is (and is not) capitalized or punctuated.

There are many metadata standards out there, and at this point the one which is used by many library catalogers in the United States is AACR2r, or the Anglo-American Cataloging Rules, 2nd revised edition. Here’s a photo of my copy, sitting on the desk:



It came loose-leaf, and it is well thought out, but very long. Of course, there are many other metadata standards, including ISBD (International Standards for Bibliographic Description). These standards are different for different items being described. AACR2 just happens to be the one right now for library items, but it will soon be replaced by RDA.

The metadata generated through these standards needs to be displayed in some form of standardized format. Think about the iTunes example again - we have all this metadata, but it is really of no use unless we can present it (as we see in the iTunes interface) in some usable manner. That’s where metadata structures come in. Amber Billey does a great job of talking about the most commonly used (except for MARC) metadata structures here. Wait. MARC? Well, MARC is an acronym (another one, imagine that) for MAchine Readable Cataloging. MARC is the common structure used in libraries, but some use Dublin Core and other formats. MARC presents the metadata in a standard format, so that library systems can read the metadata. What does that mean? Well, it takes something that looks like this:



And displays it like this:



Really, that is the point of all of this - so that you, the end user of the “thing” about which this metadata is generated can look at that metadata and determine if that thing is for you without having to look at that thing itself. Connecting people with the information they need.

Well, I hope I didn’t confuse you all too much! Please let me hear your comments or questions in the comments below!