Process

We began our analysis by conducting a thorough inventory of existing homepage elements and nomenclature on all ten UC Library homepages, identifying common elements on each homepage, and recording the nomenclature used for each element. All nomenclature variations were noted and tallied when duplications existed. This process gave us an awareness of the great variation in nomenclature on UC library homepages. For example, the element “VPN,” was found on all ten homepages, but named six different ways:

• connect(ing) from off-campus
• off-Campus access
• remote access
• connect from home
• connecting from off campus (VPN)
• off-campus log-in

Table 1: First Tier Homepage Elements

Nearly all elements had at least four different names; the majority had six variations or more and several elements had more than seven. Seeing these variations emphasized the need for greater consistency in nomenclature across campuses.

After conducting the analysis of the homepages, we examined more than 300 chat transcripts from the UC QuestionPoint 24/7 chat reference service to determine the frequency of questions and terms employed by users. We used this analysis to help us narrow down key elements and nomenclature that should be considered essential on every UC library homepage.

The most difficult part of our charge was determining best practices for homepage layout. Graphical overlays were generated for several key elements to help us analyze the current location of common content. These overlays illustrated the differences in current homepage layouts and designs across campuses. Further analysis and discussion of current best practices for Web design were also a part of our process for determining layout/design recommendations.

Findings and recommendations: Homepage elements and nomenclature

During the analysis phase, we discovered that in some cases, local naming conventions and branding (such as local catalog names, etc.) could make adherence to systemwide nomenclature recommendations problematic. Moreover, we realized that not all campuses need each homepage element, e.g., social networking and FAQs are not used on every campus. To accommodate for these differences while maintaining the goal of consistency throughout homepages, we concluded that it would be best to create two tiers of recommended content. We listed elements deemed essential for all campuses as First Tier elements and included recommendations for nomenclature (see Table 1). To further clarify, we agreed that essential elements would be considered available on the homepage if they existed as direct links on the page or as dropdown menu items that appear when rolled over by a mouse. We listed elements considered useful but optional as Second Tier Elements and made no recommendations for nomenclature (see Table 2).

Table 2: Second Tier Homepage Elements

Findings and recommendations: Layout

Our analysis of homepage layouts revealed a wide array of designs and locations for various elements. For example, the links for hours were displayed in many distinct areas on different sites (see Figure 1).

A number of related design challenges emerged during our discussion on layout.

• Local limitations. On some campuses, library Web page layout may be restricted by campus design mandates, such as header and footer information or the requirement to follow a particular style sheet.
• Content management systems. Different libraries currently use different content management systems (Drupal, Joomla, etc.) that limit design options.
• Audience. While some campuses focus on undergraduates, others may want to target their homepages to a different audience (faculty, graduate students, alumni, etc.). This could have implications in the design/layout of the page.
• Global navigation. In addition to the “what” and “where” of homepage elements, the way those elements follow the user through the Web site is critical. This has implications for homepage menu systems and subpage design.
• QuickSearch. A recent trend on many, but not all, campuses has been to incorporate a prominent search area or federated search box on the homepage. This is desirable but significantly changes the layout of the page, making consistency across campuses difficult.
• Design trends. Web design styles change so quickly that any recommendation made by a taskforce would likely be outdated before it could be implemented.
• Web design as art. The aesthetics of a good Web page require artistic input from programmers, designers, and developers. This is subjective input that cannot be determined by a task force and would also require usability testing.

Figure 1: Composite hours links.

So although we felt confident making best practices recommendations for content elements and nomenclature for UC library homepages, the complexity of Web design and the many technical and political barriers that exist in current UC library environments precluded us from making a recommendation for layout. A uniform layout would only be feasible in the future if all library Web pages shared a common technical infrastructure.

We concluded that resolving these technical issues was beyond our purview but recommended that continued investigation in this area by another group should be considered as the next step toward a more consistent user experience with UC library homepages.

Findings and recommendations: Mobile Web sites

Though it was not included in the original charge, we felt it was also important and timely to include recommendations for mobile Web sites. A study conducted in summer 2010 by the California Digital Library (CDL) found that UC mobile end-users are primarily concerned with information that they need “on the go” rather than full site content or functionality.1 Mobile users want quick snippets of information, such as finding directions, quick e-mail, fact checking, and perusal of already discovered academic resources. The more complicated tasks associated with searching, analyzing, writing, and comparing resources are not well suited to current mobile devices. Based on these findings and an environmental scan2 made of current library mobile site home-page elements the task force made recommendations for mobile Web site content. These content elements were also divided into two tiers:

First tier mobile homepage elements

• Ask a Librarian
• Database search (links out to mobile vendors)
• Directions or maps
• Hours
• News and events
• OPAC search (This element was often-times categorized as a second phase mobile development effort.)
• Staff directory
• View full site option

Second tier mobile homepage elements

• Borrowing information
• Computer availability
• Floor plans
• RefWorks mobile
• Social media (Twitter, Facebook, Flickr, etc.)

Conclusion

In November 2010, UC HOPS enthusiastically accepted the “Best Practices for UC Library Home Pages Report”3 and endorsed our recommendations. Specifically, HOPS called for UC libraries to implement recommended nomenclature for homepage elements within the next six months wherever it is possible to do so.

They further recommended consulting the report prior to any future large-scale Web site revision.

It will likely be several more months before we see our recommendations widely implemented on UC library homepages. Initial feedback from library administrators has been generally favorable; they acknowledge that having these standards is at least a useful first step in looking critically at their current homepages to see what could improve usability. But Web page changes are rarely fast or easy. Buy-in from Web teams takes time, and anyone who has managed a Web page knows that even seemingly small changes can be complex in the execution. But we’re optimistic that with time, each UC library will recognize the value of carrying out these first steps toward standardization.

The recommendations of our taskforce have potential value beyond the ten-campus UC system. Librarians have long accepted the idea of controlled vocabulary for subject headings as a way to streamline information retrieval. Standardizing terminology for at least the main elements on library homepages will improve the user experience and save Web teams countless hours debating nomenclature. Perhaps the time is right for ACRL develop with Web page best practices and standards for North American college and university libraries.

Until then, other library systems may find the UC report a useful reference for recommended nomenclature and/or a guide for the process of coming up with nomenclature standards of their own. Though our recommendations were based on UC data and developed specifically for our libraries, we suspect that parallel analyses by other institutions will replicate much of what we discovered and ultimately result in similar conclusions and recommendations. We look forward to these additional studies and eventually to a set of standards we can all agree on.