Tagging vs. Cataloging: What It's All About

by Chiara Fox

November 30, 2006

Tags have taken the internet by storm. Where once the question was “what are they,” now all people want to know is whether a given site offers them. But what are the actual benefits of tags? What motivates millions of Flickr, del.icio.us and blog users to add tags to their photos and posts? And what is it about tags and tagging that gets information architects and user experience professionals so excited?

A Tag Is a Tag Is a Tag

To begin with, the concept of tags and tagging is not a new one, no matter what the Web 2.0 kids say. Tags are essentially just metadata, and librarians and database builders have been using metadata (and controlled vocabularies to populate said metadata) for years. The basic concept behind both metadata and tags is essentially the same: one or more descriptive words are assigned to an asset, whether it’s a photo, web page, article, person, book or piece of data. And by distilling the description of an asset down to a few words, it’s easier to find and understand that asset in the future.

Because librarians have been doing this for so long, albeit in the form of cataloging and MARC records, the idea that tags are new and will revolutionize the world has created some backlash. And rightly so. Yet even though the concept of describing an item so it can be found later isn’t a new one, what is new is the way in which that description comes into being.

Librarians have traditionally lived in a controlled world filled with the ultimate of “keyword pick-lists”: the Library of Congress Subject Headings, the Medical Subject Headings (MESH) or even the Dewey Decimal System. A cataloger looks at an item to be cataloged, determines its “aboutness” and then finds the word (or words) from the list of accepted subject headings that best describes it.

These keyword pick-lists aren’t perfect, nor are they all-inclusive. But they do help control the terms that are used to describe items, which, as any cataloger worth her salt will tell you, is of the utmost importance. Without a way to account for the ambiguities of language and meaning, there is chaos. Terms and values must be controlled in order for the system to maintain a high level of recall and precision in results. This is something that every cataloger is taught in library school. Tagging, however, throws this whole concept right out the window.

How Tagging Is Different

Tagging differs from traditional cataloging in a number of ways. First, tagging no longer belongs solely to the world of librarians and indexers: now anyone can tag and describe assets. And not only is it possible for any user to apply a tag, but in some systems (such as Flickr), users can even add tags to other peoples’ assets.

Second, tagging does not require users to draw on a controlled list of values, rather they can describe an asset using whatever words they choose. There is no approval process; any word will do. This is a very bottom-up approach, and it allows the body of tags in use to grow and shift as users’ vocabularies grow and shift. This can clearly be seen on Flickr, where unusual tags such as “think of the children” and “the escalating id photo contest” can be found among more common tags such as “birthday” or “vacation.” Thomas Vander Wal named this phenomenon of bottom-up term generation “folksonomy.” Proponents of folksonomies argue that they lead to better retrieval because they rely on common language, which is immediate and accessible, versus pre-determined values, which can sometimes be limited and rigid.

Tagging systems also differ from traditional classification and indexing when it comes to the way terms are applied to assets. In traditional classification, a user chooses a term from the controlled vocabulary then follows strict rules such as the Anglo-American Cataloging Rules (AACR2) to determine how the term is entered into the system. Even in the early days of hand-coded web pages, users had to include a cumbersome <meta name=”keyword” value=”tag name”> tag in the header of their HTML, a significant enough barrier of entry that most common web authors opted not to tag their assets.

With the birth of Flickr, del.icio.us and Ajax, this barrier disappeared. Flickr makes the process easy by displaying a simple text box where users can type in their terms and automatically add them to the list of tags. Del.icio.us takes things a step further by employing a type-ahead feature that suggests similar tags a user might be interested in using. This not only helps people who have trouble deciding on which terms to use, but it also brings about some level of standardization among all the terms, the lack of which being one of the biggest criticisms of folksonomies.

What Have You Done For Me Lately

Tagging has more than just its ease of use to thank for its popularity; it’s also a powerful tool for making connections between assets, and its effects are easy to see immediately. Adding keywords to content assets in the traditional way often feels like working in a black box. Big, enterprise-level content management systems are often involved, and it can be hard to see the direct result of the classification.

For example, adding metadata can improve search, but the algorithms that Google and Yahoo! use have become so complex, and keywords are such a small factor in determining relevance, they often don’t seem to have much of an impact. Certainly there is rarely an immediate effect: Assuming keywords do make a difference, it can still take weeks or even months for a webpage to see improved rankings in web search engine results. It’s hard to see the direct effect of applying keywords in search results.

But with tagging, the results are instantaneous. A user can create a tag in Flickr, click on that tag, and immediately find any and all photos that share the tag. Tags are also a navigational tool: On LiveJournal, users can move among the posts of all their friends based solely upon the tags those friends have applied. Certain circles even feel peer pressure to not only tag their items, but to come up with the most creative tags possible.

So, tagging has brought metadata to the masses, thanks to both its ease of use as well as its clear value. But is there a hidden cost? Tune in to my next essay, “How Close Is Good Enough,” in which I’ll be taking a look at the darker side of tagging.

Chiara Fox is a senior information architect for Adaptive Path. She’s worked with Fortune 100 and 500 companies such as PeopleSoft, AT&T, Business Objects, L.L. Bean, and Hewlett-Packard.