Last summer, I was asked by a client at a small administrative tribunal to help with an interesting project. Over the organization’s 20-year history, it had accumulated a number of “issues files”, which document the evolution of its thinking on a range of questions and problems which had arisen over time. The collection was a valuable store of corporate knowledge and history, but it was difficult to know where to look for a particular piece of information or to know what questions might be answered by using these files. Could we recommend a way of cataloguing the contents of these files in such a way as to make them more searchable?
A number of constraints faced us in this project.
- The solution had to be low or no cost.
- The library does not have access to the client’s servers, and their staff can’t access ours. Whatever solution we find has to be in neutral territory, but protected.
- The solution needs to keep these records private.
- The solution has to be easy to maintain, so that staff can take over the long-term management of the collection.
Clearly the library’s own catalogue was out – it failed on a number of the requirements. The client doesn’t own a document management system or intranet because of its small size, and we had to find a solution that didn’t rely on assistance from the government’s IT department.
Serendipitously, I remembered seeing a student project at an annual fair mounted by the Faculty of Information Studies (now i-School) at the University of Toronto. A student had used Library Thing to catalogue a collection belonging to the Toronto Book Awards. Could this be the tool for us?
Library Thing is similar to Facebook’s Shelfari application – it’s a way to list your collection, and to find others with “eerily similar tastes” (according to the home page). It pulls records from Amazon and over 600 library catalogues from around the world. Canada is well represented with CISTI and Library and Archives Canada records, and the catalogues of a number of notable university and public libraries making the list.
I’ve had a personal (free) account on Library Thing for a couple of years, and have used it to find books by favourite authors, or similar works by other authors. I’d never considered using it to catalogue an organization’s holdings. Upon further investigation I found that sure enough, there are institutional memberships, and you can make your collection private. After consulting with my clients, we purchased a lifetime membership ($15/year), and got to work.
The collection consists of a large number of file folders, and a small number of books. Contents generally include memos, cases, photocopies of news stories or journal articles on a given topic. Some files include legislative history or answers to specific questions. We were able to adapt the fields in the Library Thing interface to our purposes. The ISBN field became the folder title, and the Dewey call number became the collection designation (we distinguished between Library and Issues files, so that staff will know whether they’re looking for a book or a folder). Detailed procedures document the choices that we made, in the hope of ensuring consistency over time.
Using practicum students from Seneca College, we were able to catalogue the entire collection fairly simply. We used the notes fields to list the inventory of each file. Tagging replaces Library of Congress Subject Headings. It will be interesting to watch how this part of the catalogue evolves. The materials catalogued by library staff are fairly simply described, as we don’t “speak” the vernacular of the host organization. Once staff is responsible for the ongoing maintenance of the collection, there may be more subtlety and differentiation in the tag cloud. I plan to monitor the collection to see how it changes.
The cataloguing portion of the project is completed, and now user training begins. The product can be browsed using a half-dozen or so preformatted display styles (sorted by author, title, tags or date – and these can be customized). A simple “Google-like” search window allows you to search quickly from the home page. Fielded searching is also available. “Comments” is a searchable field, and we took advantage of this by putting folder inventories into the comments area. Users will be able to check for a particular memo or decision by searching the Comments field. It is also possible to browse the tags, which can be displayed as an alphabetical list, or as a cloud. The screen grab below shows part of the home page:
I think that this experiment may be interesting to small organizations which don’t have a full-time librarian. Although this product isn’t sufficient for serials management, it could be an excellent stop-gap for a small firm or government organization that simply wants to know what information is available. As I learned, it’s not that difficult to adapt the interface to control internally-generated information as well. The secret is in creating detailed notes, which are searchable.