Where Have all the Data Gone? Issues in Web Site Design
Mary S. Carroll
Publishing archaeological data and information on the World Wide Web presents a unique set of challenges. Along with technical issues such as data standards and file formats come the often-overlooked issues of Web site accessibility and design. Users may never get to the information available if they can’t easily find their way around the site — or if the structure of the site cripples downloading capabilities. This paper will discuss the process of developing useful, accessible Web sites that deliver substantive information and will use the National Center for Preservation Technology and Training’s Web site design project as an example.
Delivering archaeological information via the World Wide Web presents a unique set of challenges. Along with technical issues such as data standards and file formats come the often- overlooked and related issues of Web site accessibility and design. As the Web expands and technological capabilities grow, users expect more substantive content from the sites they visit and Web site managers attempt to deliver that content. But even though there may be a wealth of archeological information present on the Web,1 archeologists searching for raw data, research reports, bibliographies and other resources may never get to the information available because they can’t easily navigate the site.
Much has been written in the Web design world about how to develop cutting edge Web sites, both creatively and technologically.2 In this paper we will briefly discuss the process of developing useful, accessible Web sites that deliver substantive information — and will use the National Center for Preservation Technology and Training’s Web site design project as an example.
Even though the term “electronic publication” often is applied to the Web, well-designed Web sites are not publications — they are organized spaces that connect people with information and with each other. In fact, good Web site design has more in common with good exhibit design than with preparing a publication. Both Web sites and exhibits must entice their visitors, provide clear paths of exploration, keep visitors’ attention and connect visitors with what they are looking for — from interactive experiences to in-depth informational resources.
Developing a well-designed Web site can be divided into five phases. Project planning — during which the site’s objectives and information design are established — consists of the first two phases, definition and architecture. Project development, during which the site is designed and constructed, takes place in the latter three phases — design, implementation and integration. While jumping directly to the development phases may be tempting, the ultimate success of a Web site depends on careful planning prior to design and implementation.
Planning – Definition
Because there are many reasons for developing a Web site, the first step in the planning process is to define clearly the primary objectives, target audiences and project scope. Although Web sites may be developed to create a presence on the Web, this reason alone is not sufficient basis for an effective site. Effective Web sites work because they achieve substantial and well-defined objectives. Taking time to clarify a site’ s objectives, to analyze its audiences and to develop strategies for appealing to each audience is necessary for the success of the whole project.
To begin, a one- or two-sentence Web site mission statement should be written that summarizes the goals of the site and its desired effect on the audiences served. With a general mission defined, the next step is to develop specific strategies for achieving these goals. This may be as simple as a list of the three things that the site should accomplish, or it may involve more detail that outlines numerous objectives. Either way, it is important to keep the objectives specific and realistic. Determining the measurement by which achieving objectives will be assessed also is crucial. Will it be determined by the number of visitors? By which sections are accessed most? By critical acclaim? The answer is best decided by each project.
In the definition phase of NCPTT’s Web site redesign, five objectives were formulated. The NCPTT Web site is intended to — 1) fulfill goals and objectives outlined in NCPTT’s mission and long-range strategic plan; 2) serve as a clearinghouse and delivery mechanism for information sponsored, collected and developed by NCPTT; 3) establish and promote NCPTT’s role in the conservation and preservation community; 4) promote online communication for individuals and organizations involved in preservation and conservation; and 5) provide a means to measure the impact and reach of NCPTT’s work.
Another vital issue to be addressed in the definition phase is a Web site’ s audiences. Web sites typically address five potential audiences — two public, two private, and one semi-private (Figure 1). A site’s public audiences include a target audience and the general browsing public. Private audiences include internal users of the site who may have access to proprietary sections — such as an organization’s staff — and administrative users who are responsible for maintaining the site and its content. Finally, a site may also have a semi-private section for special users, such as an area available only to registered users. The distribution of audience sizes will vary among projects. Some sites may have a very wide target audience and a fully developed internal Web site for staff; other sites may have a very narrow target public and an administrative section used by only a few people.
Where possible, it is important to determine as much as possible about each audience, such as reasons for visiting the site, anticipated frequency and length of visits, the educational and professional background of the audience and whether the user will be accessing the site from home or office. While it is acknowledged that ascertaining this kind of information is difficult, it can be valuable to the design process if it is available. Other useful demographics are related more to the audience’s hardware and software configurations, such as computer platforms, browser software and speed of Internet connection. If the project involves redesigning an existing Web site, server logs that record information about Web site visitors may provide some of these statistics.
Since NCPTT’s project involved redesigning a Web site that functioned primarily as an interface to NCPTT’s gopher — and since there were no server logs available — we were not able to generate target audience statistics. However, NCPTT’s audiences were easily defined. The target audience is professionals in the fields of archeology, historic architecture, historic landscapes, objects and materials conservation and interpretation. NCPTT staff, a subset of the broader preservation community and the general browsing public were also considered in the planning process.
Planning – Architecture
The second phase of site design — architecture — pairs the objectives and audiences with content. Methods involved in choosing appropriate content, organizing and prioritizing content and creating a clear navigation system through the content are drawn from the growing field of information architecture.3 When designing the information architecture of a site, use the already defined objectives and audiences to decide the nature of the site’s content. Actual content might come from sources such as a digital version of a museum’s collection or a printed history of a preservation organization. Other content may be developed specifically for the site, such as a database of an organization’s members or an online discussion forum.
For NCPTT’s Web site redesign, audience needs were determined to fall into three categories: information about NCPTT for those who may be unfamiliar with the organization, information about programs sponsored by NCPTT and current and archived information on preservation topics. Actual content was drawn from several sources. Information about NCPTT and its programs was adapted from existing materials. A new online version of NCPTT’s newsletter, NCPTT Notes, was adapted from print versions of the publication. Additionally, NCPTT already provided a wide variety of preservation-related information via its existing gopher site or via hardcopy, such as research findings, conferences, job postings, funding opportunities and other online resources. This information continues to be relevant and has been transferred to a searchable database system for enhanced access via NCPTT’s site.
Content organization is the important next step. While there are few universal rules for information design, taking the time to examine three principal factors leads to better results. First, determine natural organizational systems within the content that would help to make the information more accessible. For example, the information presented in a Web site on southwestern archeological projects might be organized geographically or chronologically. Second, determine particular objectives that would require giving certain information priority. For NCPTT’s Web site, disseminating results of PTTGrants and PTTProjects is critical, so that information will be prominently featured and available via various routes. Third, determine what the site’s audience will be looking for when contacting the site.
Working from the three categories of audience needs, NCPTT’s Web site is configured in three “meta” components — “About NCPTT,” with generally static organizational and background information; “NCPTT Components,” a more active section for NCPTT’s three components — research, training and information management; and “Resources,” a searchable database system of preservation-related information that is currently in development.
Visualizing the information architecture of the site is best done with a site schematic — a graphic representation of the site’s structure that shows how the information in the site is organized and connected. Figure 2 is a corner of the very large NCPTT Web site schematic; it is impossible to reproduce it in its entirety without its being indecipherable. The site schematic also is a good tool for planning future areas of the site, since these can be placed into the master scheme even if they are not developed initially. NCPTT’s site schematic encompasses all planned functions of the site. During the phased development process the plan is fine-tuned to deliver substantive information to an appropriate audience.
Upon completion of the planning phases, the three phases of development — design, implementation and integration — can begin. However, elements of the planning process will continue throughout the development phases. We recommend that development takes place in a staged approach to facilitate modifications of the plan and so that each stage can build upon the previous.
Development of NCPTT’s Web site was divided into four phases. Phase one encompassed the design of the graphical user interface and development of static site content, including descriptions of NCPTT, its mission, its program areas and the Preservation Technology and Training Board.
The design of the graphical user interface began by laying out the site’s organization and testing the information architecture with basic navigation elements. NCPTT staff reviewed the navigation process for ease of use. Next, a few elements were put in place (Figure 3) and then sample images and text were added giving it an appearance close to the final product (Figure 4). Figure 5 graphically represents the main concept sections of the core page. The final core page can be viewed at www.ncptt.nps.gov; the center image varies according to the current feature.
The structure of second level pages — NCPTT Programs, About NCPTT, Resources — maintains the same navigational pathways and graphical interface as the core page. See Figure 6 for a graphical representation of the second level page sections. A final second level page can be viewed at www.ncptt.nps.gov/im. (Figure 7) All pages at this level will have the same look and navigational tools; only the details in text and images will vary.
Phase two included designing the databases that will replace the gopher-based Resources section
and implementing one prototype database — Preservation Internet Resources, which replaced the annotated list known as “Internet Resources for Heritage Conservation, Historic Preservation and Archeology”. A thorough and exhaustive needs analysis was undertaken by the Web site designers. NCPTT staff were asked to complete questionnaires for each category of information to be disseminated via the Web databases. Staff were also interviewed in person by the development team. The goal was to address the current scope of the database project and to insure the scalability of the system developed. Design specifications for all of the tables in the system were codified in a 78-page document entitled “Design Documents for the NCPTT Web Site: Phase II Development,” which provided the basis for Phase three.
Phase three involves implementing the other databases in the system. The system includes General Databases (conferences, jobs, grants); Program-related Databases (Training and Education, Analytical and Materials Testing Directory, Materials Research Bibliography); and Grants, Projects and Publications Databases (Grants and Projects Catalog, Publications Management).
Phase 4 is intended to add functionality to the Web site and to target a smaller audience — a PTTCommunity. Not yet through its planning stage, Phase 4 may include a system that will allow users to register to be notified when items matching their interests are added to NCPTT’s Web site and may also allow users to input additions to the database system for consideration by NCPTT staff. In addition, a system to allow online submission of PTTGrants proposals and to allow PTTGrants reviewers to access proposals and submit comments online will be developed.
The planning phases are the real foundation upon which a successful Web site is built. The surface of an archeological site usually gives some indication of what lies below — what could be called the “content” of the site. As we all know, it’s necessary to remove the soil to discover the “content.” But the surface of an archeological site will rarely, if ever, indicate everything that is below. There are often unexpected discoveries — “content” that was not anticipated based on the surface survey or collection. And since archeological sites are created over time through natural processes, obviously no one plans or develops surfaces to show the underlying “content.”
However, we can control what could be called the Web site’s “surface” — the core page of the site. And unlike an archeological site, the full content of a Web site can — and should — be clearly mapped at the “surface.” There should be no surprises to users, especially unanticipated lack of content.
1. Mary S. Carroll, “Internet Offers Many Electronic Journals of Preservation Interest,” NCPTT Notes 15 (1996): 8-9. Mary S. Carroll, “The National Archeological Database,” NCPTT Notes 17 (1997): 6-7.
Mary S. Carroll, “Preservation Internet Discussion Forums,” NCPTT Notes 21 (1997): 6. www.ncptt.nps.gov/notes/21/5_fs.stm.David Carlson, this volume. S. Terry Childs, this volume. John Hoopes, “Avoiding the Driest Dust that Blows: Web Site Reports,” SAA Bulletin. 17.1 (1999): 23, 26- 27, 39. John Hoopes, “You’ve Got News! Archaeology Journalism on the Internet,” SAA Bulletin. 17/2 (1999): 26-28.John Hoopes, “Electronic Quipus for the 21st Century: Andean Archaeology Online,” SAA Bulletin. 16.1 (1998): 20-22. www.anth.ucsb.edu/SAABulletin/16.1/SAA16.html. John Hoopes, “The Online Lab Manual: Reference Collections on the Web,” SAA Bulletin. 16.5 (1998):
17-19, 39. www.anth.ucsb.edu/SAABulletin/16.5/SAA14.html.
2. Bart Marable, “Once Upon a Time: Using New Narratives in Educational Web Sites” (Paper presented at Museums and the Web 1999, New Orleans) www.archimuse.com/mw99/papers/marable/ marable.html.
Bart Marable, “Bringing Stories to Life Online,” Web Techniques. 4.3 (1999): 18-23.
3. Clement Mok, Designing Business: Multiple Media, Multiple Disciplines (San Jose: Adobe Press, 1996). Louis Rosenfeld and Peter Morville, Information Architecture for the World Wide Web (Cambridge: O’Reilly and Associates, Inc., 1998).