The University of Iowa
The University of Iowa came to Leepfrog Technologies with a problem very common to higher education: the process used to create, update and manage their course catalog was very time-consuming, involving many people touching the process, and a lot of data provided from different individuals. The university was interested in improving the efficiency of the process, getting more control back into the Registrar’s office, and gaining confidence that the information provided in the catalog was accurate. Beginning in October 2008, Leepfrog and the team from the University of Iowa worked to implement a solution that greatly increased efficiency, control over the process, and accuracy in creating and managing their academic catalog – and brought the project in under budget. Here’s how.
The legacy process at the University of Iowa was very similar to many other campuses: human-driven and labor intensive. A four-person team – three from the Registrar’s office and one from public relations – was responsible for shepherding the content through the various departments and on to publication. Changes to the catalog drove many elements of academic life, including updates to degree audit and receiving college-level approval for department-level programming changes. The catalog process involves every college, department, and program on campus.
The catalog contains 211 distinct sections, each with its own author and workflow. Each section’s content existed primarily in a Word document, but it was also co-mingled with typesetting commands. To prevent altering of the typesetting, edits were done on paper printouts of the catalog. The process used a “colored pen protocol,” which meant each editor used a different colored pen to make their changes, so that subsequent editors could see who authored each edit. Distribution was through campus mail, and the Registrar’s office was responsible for contacting and encouraging the participants to get their copy updated.
The end result was an accurate catalog with a high-quality print component, highly customized for each department on campus. While the final deliverable was of great quality, the effort involved was quite large and was repeated each year. Many elements of the process were known to only a single person within the chain, and much of the accuracy of the catalog depended on sharp-eyed proofreading.
Leepfrog went through the catalog, reviewing the existing formatting and working with the editors to determine rules for when each set of formatting was used. Lists of rules and the exceptions were created and added to the system, ranging from the obvious (such as combining the three History sub-departments into one section while keeping the three math sub-departments separate) to the subtle (differences between departments on the proper plural form of “emeritus”).
Where possible, Leepfrog added automatic checks to the process, making sure that the information on the page was as accurate as possible even before human review. Many checks on the consistency of data – catching missing or duplicated courses, incorrect numbers of semester hours, missing cross-reference courses, and so on – were added to the page logic, requiring the authors and editors to refine the content.
To function effectively, CourseLeaf needed to be fully integrated into the university’s environment. Departmental authors would use the system only once a year. The existing campus single-sign on (SSO) login name and password (called a HawkID) could be used to log into CourseLeaf, and the existing email systems were used to drive workflow where possible. Both of these factors were critical. Because email addresses could not be gleaned from the usernames, a web service to perform routing queries was added to the system.
The catalog references a great deal of information within the course inventory. During deployment, the University of Iowa changed from a legacy SIS to its new, modern SIS (MAUI). CourseLeaf’s deployment needed to work with both systems, and it had to be prepared for the change in the university’s course numbers in a few years as deployment of MAUI continued.
The catalog lists faculty in many places. Some of the locations use the official name and title stored in the human resources systems; other locations require specific titles for each of the faculty. The software was customized to allow authors to explicitly set the faculty list, but to assist in eliminating common input errors, such as automatically using the plural form of titles.
After the framework of the solution had been deployed, Leepfrog copied the content from the existing catalog into the new software. This process was done by a combination of automated migration based on the embedded typesetting codes, manual tagging of content to be replaced with database-driven content, and several passes of proofreading. When the authors began their yearly review and updating of the catalog, all of the existing content was already in place in the new system, ready for modification.
Each department was invited to suggest the ideal workflow for the new process. Common elements were pulled out of each workflow and given standard names. A total of 114 unique workflows were created, most of which shared one of four common sequences. The longest workflow was twelve steps.
A series of training sessions for the catalog’s 318 authors and editors were held over a four-day period. Because CourseLeaf does not have a limit on the number of simultaneous users, group sessions of 25 authors at a time gathered in a single lab for training, which allowed for an efficient environment. Most of the authors brought their changes to the catalog for the upcoming edition to the training session, resulting in hands-on training that boosted learning retention. Most authors left early, comfortable with the experience that the workshop gave them.
The editorial process used automatic email reminders to advance content through the process. Since the prior year’s catalog was brought in as a baseline, every change from the prior year’s catalog was highlighted in colored mark-up. Later editors could view all changes at once, or drill down and see changes by a specific person at a specific point in workflow. Many tweaks were made to specific content’s workflow, dynamically modifying the sign-off requirements even while the content was in the process of review.
Many sections of the catalog had groups of pages that had the same author and workflow for every page in that section. These pages were all grouped together into a single set for the purposes of workflow, greatly reducing the review time of each of the people involved. This was particularly useful when reviewing long boilerplate sections of policy, where changes are particularly important for collaborative review.
The legacy solution used embedded typesetting commands to generate two versions of the catalog: a complete general catalog and a smaller department-specific version of the catalog for each department. A second process was used to generate HTML pages based on the typesetting codes.
The CourseLeaf solution involved the authors in the creation of the HTML content from the very beginning. The authors viewed their progress and changes directly in the web format. A new design for the web pages was introduced just before publication, and the existing, approved content was re-generated in the new format.
A high-end, professionally laid-out PDF version of the catalog was created with many features not supported in HTML, including multiple columns, folio headers, proper hyphenation, indexes and tables of contents, and other standard print-specific features. This version was used to generate print-on-demand versions of the catalog, including a hardcover college-specific version with 300dpi photographs. A minimally-formatted, inexpensive “bulk” PDF version of the catalog was created for archival and regulatory purposes. The solution generated several other variations of the catalog, including a version for the just-released Amazon Kindle.
The effort that was spent creating the 2009-10 catalog was used as the starting point for the 2010-11 edition. The processes, workflows, and structure were there from the beginning, drastically reducing the time that authors and editors take to maintain the catalog. Tweaks to the process will continue to be added every year, improving the overall efficiency of the process and the quality of the final products.