• Creates, migrates, and manages metadata for resources in the library’s physical and electronic collections
• Processes and helps manage the physical items in the library’s collections
• Performs batch loads of metadata records for serials, government documents, and purchased collections of information resources; also, helps write code for loader software
• Maintains the data in the Skyline online catalog to keep it current with the latest national standards
• Participates in cooperative, national cataloging organizations, including NACO, SACO, and OCLC Enhance. Our NACO and SACO work concentrates on creating national-level authority records for Colorado people, places and events
• Conducts research related to metadata and information retrieval and advises Library administration on best practices for selecting and implementing optimum information discovery/management systems
• Collaborates with our counterparts at the Health Sciences Library on the Anschutz Medical Campus

There are two categories of searching or “information discovery.” The first is looking for a known item. This includes looking for a particular book, known by its title, or looking for a work by a specific author. For example, a search that involves looking for the book It’s Not about the Bike by Lance Armstrong is a known-item search.  

The second type is looking for an unknown item. This involves looking for information on a particular topic, without knowing the details of any specific books or works by particular authors. For example, someone searching for information on rock climbing in Colorado might do a subject heading search for “Rock climbing—Colorado” in an online catalog. A library catalog makes possible both of these types information seeking.

There are also two classes of searching, defined by how the search engine works: probabilistic searching and deterministic searching. Probabilistic searching is like the searching done in web search engines such as Google. It’s called probabilistic searching because it relies on search algorithms that calculate which resources will most probably match the resources needed as indicated by the words entered in the search.

The other type of search is called deterministic searching. This is the type of search done in an online library catalog. Deterministic searching is much more precise because it uses metadata (structured data that describes other data) to generate search results. So if you search for information about rock climbing in Colorado, you do a subject search, and the metadata records that contain the subject heading “Rock climbing—Colorado” will be retrieved in the search. And because only books that are really about rock climbing in Colorado will come up in this search, the results are precise and not filled with pages that just happen to have “rock climbing Colorado” in them, like in a Google search.

One of the things that makes deterministic searching precise is controlled headings. We use consistent and unique headings for each author and subject in the catalog. The “Red books” of Library of Congress Subject Headings are what we use for subject headings, and this data is now available online at the Library of Congress Authorities web site.  

This site also lists personal name headings. We endeavor to use a unique name heading for every author and for every person who is the subject of a book. This is called name disambiguation, and it’s very helpful in making information retrieval more efficient. It saves time by collocating all books by a given author under a single, unique heading.

Library cataloging is undergoing rapid change. Some are even predicting that library catalogs will not exist in as few as five years. But information discovery tools for print resources will still be needed, and deterministic search systems add great value to information. But for electronic resources, many think that a simple search interface, like Google’s, is all that’s needed. What do you think? Let us know your thoughts.

Contact the cataloging department

back to top