A Small School Ventures Into the World of the CWIS Copyright CAUSE 1994. This paper was presented at the 1994 CAUSE Annual Conference held in Orlando, FL, November 29- December 2, and is part of the conference proceedings published by CAUSE. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, that the CAUSE copyright notice and the title and authors of the publication and its date appear, and that notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To copy or disseminate otherwise, or to republish in any form, requires written permission from CAUSE. For further information: CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301; 303-449-4430; e-mail info@cause.colorado.edu A SMALL SCHOOL VENTURES INTO THE WORLD OF THE CWIS Bev Actis Kenyon College Gambier, OH 43022 In late 1992, Kenyon College in Gambier, Ohio, had just completed its campus network infrastructure. This new environment provided an institutional "integration" that enabled Kenyon to consider the possibility of a campus-wide information system, or CWIS. Kenyon began its planning by investigating many other CWIS on the Internet. In the process, the most important issues to be addressed in launching a successful CWIS were identified. This paper discusses those issues and provides a guide to other small schools in getting started. A SMALL SCHOOL VENTURES INTO THE WORLD OF THE CWIS By the fall of 1992, Kenyon College, a liberal arts school of 1500 students in Gambier, Ohio, had just completed its campus network infrastructure, which provided network access to virtually all campus buildings. The campus community was communicating via email and beginning to share much information electronically. The time was ripe for Kenyon to consider how to utilize this new environment to more fully integrate its wide variety of information resources into a campus wide information system, or CWIS. At that time, a task force from Information & Computing Services (ICS) began looking, via the Internet, at other campuses with a CWIS. Since Kenyon's financial situation allowed very limited personnel and financial resources for such a project, ICS had to find a way to build a CWIS at the "right price." At that same time "gopher" was beginning to become a hot topic on the Internet. It was the gopher software developed by the University of Minnesota that gave us the breakthrough we needed to proceed with our CWIS, since gopher was not only free, but also enabled us to provide both local and Internet resources to the campus. We began planning for KCInfo (as we called our CWIS) by collecting from the Internet all the information we could find about other CWIS, so that we could learn from their strengths as well as their weaknesses. We contacted many CWIS managers for ideas on how to go about getting started. By sifting through all this information, we were able to identify a dozen issues that had to be addressed by ICS before launching KCInfo. These issues included: * Needs & Resource Assessment * Ownership Issues * Access & Privacy * Personnel Needs * Content Standards * Policies & Policy Boards * Main Menu Design * Marketing Strategies * The Coordinator's Role * Sponsors & Information Providers * Ongoing Development * Evaluation The following discussion of these issues is meant to provide a general "game plan" to help other small schools considering a CWIS to take those first steps in getting started. DO WE REALLY NEED A CWIS? The first question we asked ourselves was, "Do we really need a CWIS?" In our present situation of budget restrictions, would the service a CWIS provided to the campus really be worth the money, time, and work required to develop it? What compensating benefits could it offer? Our conclusion was that KCInfo would greatly improve access to a broad range of college information as well as Internet resources. It would organize in one central location information that would be available anytime and from any workstation connected through the campus network or via a dialup modem. As an alternative to printed information, KCInfo could significantly reduce the use of paper on campus. Although it would not completely replace paper, hardcopies of electronic documents would be printed only at the user's discretion; thus they would not be wasted on users who had no real need or desire for them. NEEDS AND RESOURCES: FINDING A MATCH Our next task was to assess our current resources to determine what we needed to make KCInfo a reality. Kenyon's all-campus network was our biggest asset, since it provided universal access to computing resources. The network was connected to a cluster of DEC VAXes connected to workstations, half of which were terminals and half microcomputers. With half the campus still using terminals, we realized that the software to run the CWIS would be have to be installed on the VAX mainframe to be accessible to all. In the future, as we evolved into a microcomputer-based campus, we could migrate to a distributed system with connections to a dedicated machine set up as our gopher server. Using a timeshare environment for our gopher server was not ideal, since it might put a strain on the VAX cluster when usage became heavy. A separate, dedicated machine would have been preferable, but was not possible at the time, so we plunged forward on the VAX to get started. Should system overhead present a problem due to the growth of the CWIS, relatively cheap add-on memory and disk storage could be purchased as needed. Because of budget and personnel limitations, we couldn't afford to allocate staff time for developing our own CWIS software. Given these limitations, the increasingly popular "gopher" software seemed to be the perfect solution for Kenyon, since the software itself and its maintenance would cost us nothing. Gopher's simplicity of design and protocol made it very appealing, and it had already become a standard on the Internet. It offered an easy-to-use hierarchical menu structure and a minimum of keystrokes for the user to learn. Information retrieval was simple because the user wasn't required to know the location of the information. It could easily be expanded by simply linking up to another "server," either locally or remotely. Gopher had other important features that made it seem the right choice for Kenyon. It provided access to all kinds of Internet resources, regardless of their type or location. These resources included: WAIS and Veronica search tools; Archie and anon ftp archives; gateways to online library catalogs; multimedia using RFC1341 MIME email extensions; SQL interfaces to databases; and links to thousands of other gopher servers worldwide. Besides hardware/software considerations, we took into account the intangible yet real time savings and productivity increases that a CWIS could offer. Anyone seeking college- related information could find it more readily. College departments were able to update and distribute their information more efficiently. OWNERSHIP ISSUES: WHOSE CWIS IS THIS ANYWAY? In talking to individuals who had implemented a CWIS at other campuses, we learned that information "ownership" had been a big issue. Because of possible legal implications for Kenyon, it was important to make a clear distinction between the role of the institution and those who sponsor the information in the CWIS. The college simply provided a vehicle (KCInfo) that acted as a "collector and storer" of the information, similar to a library. The role of the sponsor of the information is that of "author/editor/publisher." By not editing the information stored in KCInfo, Kenyon would not be responsible for information that might later be found to be illegal, such as copyright and privacy infringements, libelous or derogatory information, etc. Therefore, as "owners" of the information, the sponsors assumed full responsibility for the accuracy, quality, legality, and appropriateness of their information. ICS insisted that each sponsoring group create its own internal process of review and approval of any documents provided to KCInfo. The sponsors were charged with selecting a representative to become their "information provider," who would be responsible for the editorial management of the sponsor's documents (which covered writing, formatting, proofreading, posting, and deleting documents). To identify ownership, every document in KCInfo was required to have a header at its beginning, containing the name of a contact person responsible for the information so that questions and problems could be directed to the proper source. ACCESS AND PRIVACY A gopher-based CWIS is available not only to campus users, but to anyone with Internet access. It is important that sponsors understand this concept of open access before they begin to plan their information. There are different ways in which gopher software can restrict access to certain files and menus (and also by IP address). But since KCInfo was set up to improve information sharing among campus members, it was not to be used for confidential or restricted information. Regarding the issue of privacy, ICS had to address two concerns: user privacy and data privacy. We had to be careful, in any analysis of usage statistics, that the data would not be used to identify individual access of documents. Libraries have long been aware of the need for user privacy in the area of book selection, and Kenyon had to be sensitive to this as well in regard to electronic documents. Data privacy considerations in light of the FERPA act require that personal information (such as an individual's home address and phone number), not be made available for public access without giving that individual an opportunity to withhold it. This had to be discussed with potential sponsors who might inadvertently include inappropriate personal information in their KCInfo documents. PERSONNEL NEEDS: WHO'S GOING TO DO THE WORK? The financial climate at Kenyon was such that there was no chance of hiring additional personnel for KCInfo, so the time involved in implementing and maintaining it had to be carved out of ICS staff time. It was imperative that ICS train those who would be responsible for providing the information to be as self-sufficient as possible. After contacting several other schools about their CWIS personnel needs, we mapped out a tentative personnel "duties" list. (The actual time spent on KCInfo development and maintenance is also included here as well): * System administrator: Manages the CWIS software and hardware; performs upgrades, migrations; customizes as necessary. Estimated time per week: after the initial installation/test period, 1-2 hours ongoing. Upgrades and any special streamlining "tools" would take additional time periodically. * Coordinator: Marketing and public relations work; design of CWIS main menu; day-to-day management of CWIS development; training & support of information providers; recordkeeping. Estimated time per week: about half-time for first several months of development; then 5-7 hours ongoing. * Sponsors: Responsible for information content; define and organize information; develop internal review and approval process; select information provider(s) to post and maintain their information. After initial planning process, additional time would be required only as new menus and documents were planned. * Information Providers: Responsible for editorial management of documents; after initial learning period, the time needed for preparing documents for KCInfo should not be more than required for most paper-based documents. CONTENT: MEAT AND POTATOES OF THE CWIS Although it may seem obvious, the selection and organization of the information (content) is THE most important consideration if the CWIS is to be well-used. It should contain information that is informative and even entertaining because it has to motivate the user to want to explore it further. The menu structure must be organized in such a way that the novice user can find the needed information intuitively. One of gopher's nice features is that the same information can be accessed from multiple locations in the menu structure, thus providing users with more than one logical "path" to follow. Defining and organizing information requires time, and it cannot be rushed. Getting sponsors to understand the importance of this planning phase is the key to the success of the CWIS. Information that is not well-written or menus that are poorly organized will simply not be read, and the CWIS will be bypassed. By making the sponsors and information providers responsible for all aspects of the "content," ICS did not have to spend staff time on proofreading, formatting, or posting the information to KCInfo. As mentioned earlier, each sponsoring department or organization determined its own internal review and approval process for information targeted for KCInfo. Questions about the information were directed to the contact person identified in the document header and so they did not take up ICS staff time. If we had not distributed responsibility for content, ICS would never have had the staff resources to handle the CWIS implementation or its ongoing management. We trained the sponsors and information providers to be as self-sufficient as possible so that we could then concentrate on the system management responsibilities that only ICS could perform. Early in the project certain content and formatting standards were established before any information was allowed to be posted to the CWIS. Some of the standards set for the Kenyon environment were: * Appropriateness: Appropriate information was defined as information that may be of general interest to members of the Kenyon College community. Advocacy, commercial advertising and sales, libelous or derogatory information, and confidential information were deemed inappropriate. * Currency: Information providers must review and update information regularly; otherwise, it could be deleted after notification to the provider. * Accuracy and Quality: Documents were to be error-free in terms of spelling and grammar. The quality of workmanship should be the same as that of paper-based information. * Copyright: Existing copyright and privacy laws must be honored; sponsors were required, if necessary, to obtain copyright approval from the author. * Format: Standards were set for converting wordprocessed documents to ascii format, which was required by gopher; conversion procedures (from WordPerfect to ascii) were established. When these standards were decided upon, the coordinator included them in the Information Providers Handbook which was used in the training workshop. (This handbook can be found in KCInfo at gopher.kenyon.edu by following the menu path: About KCInfo/Interested in Posting Information to KCInfo?). POLICIES AND POLICY BOARD: GUIDE, WATCHDOG AND REFEREE Other campuses had experienced many problems when setting up their CWIS because they had no clear guidelines to follow and no means of settling the inevitable conflicts that arose because of differing opinions. Examples of these issues were: disagreements on the design of the main menu; personnel responsibilities; copyright, privacy, and other legal issues; ownership issues; lack of content standards; access issues, etc. ICS decided that a policy document and a broadly-based forum to oversee KCInfo would be good "preventive maintenance." We set up an ad hoc committee to write a policy document that would be in place before KCInfo was opened up to receive information. The policy document defined the membership of the policy board, which represented the various college constituencies. This board would oversee the development of KCInfo, enforce the policies set by the policy document, and settle conflicts. (A copy of the KCInfo Policy Document is available in KCInfo at gopher.kenyon.edu under the main menu topic "About KCInfo"). DESIGNING THE MAIN MENU Before any information could be posted to KCInfo, the main menu had to be designed. Deciding upon the main menu topics was critical because they had to be broad enough in order to include the wide range of information to be made available to the campus. Looking at many other institutions' main menus and deciding which categories were most appropriate for ours was a helpful start. Listing all the various departments and services that the college offered also gave us a good idea of what kinds of information we had to work with and how we might best categorize them. Once the topics were set for the main menu, we set up a "prototype" menu. We wanted feedback about the appropriateness of the topics. Were they broad enough, inclusive enough to cover the information to be posted by the many and varied campus groups? The main menu remained flexible during this test phase, but after a limited time, it was then "frozen," ready for posting information. This prototype menu was skeletal, with a very limited amount of information in it, but what it had was useful or entertaining enough to draw the reader into browsing through it. The local information that was put in already existed, either on paper or online somewhere else: campus directories, newsletters, calendars of events, computing and library information, announcements, course listings, etc. Putting this information into KCInfo made it much easier to find, which served to illustrate its potential as a "one-stop" information resource. Since gopher enables one to browse the Internet, we included several Internet resources as enticements, such as the U.S. Weather Service, White House Press Releases, the Internet Hunt, and other interesting gopher sites. MARKETING STRATEGIES For a project of this kind to be successful, we had to market the idea to several audiences: first, to the staff of ICS, since they would be responsible for a new resource that required much cooperation and support among the staff to get it started; second, to the upper management of the college, since their support was necessary to motivate the sponsors to get involved; and third, to the campus in general, so that they could see the value of KCInfo in providing them with readily available, easy-to-find information. We used the KCInfo Policy Document itself as an important marketing tool. To obtain approval for the document and the proposed KCInfo Policy Board, we had to go before the senior governing board of the college to present our "vision" of KCInfo, answer their concerns, and ultimately gain their support. The policy document was a symbolic statement that we saw this resource as an invaluable tool for all constituencies of the college and intended it to be developed in a purposeful, responsible manner. In the beginning it was important to create publicity and interest in whatever way possible: an initial college-wide email news release with an invitation to attend a demonstration of KCInfo; introductory meetings and workshops for potential sponsors and information providers; articles in college newsletters describing its features; monthly email messages describing new information areas; personal contacts with key department heads to persuade them to participate, hoping that by doing so they would encourage other departments to do likewise. THE COORDINATOR: CHEERLEADER AND TRAINER Once the groundwork was done, the next step was to select the right person to manage the day-to-day development of the CWIS. There were several important qualities to look for in a coordinator, among them excellent organizational skills. This person would assist the sponsors in organizing their information and selecting an appropriate location for it in KCInfo. Especially important was having a "vision" of a fully integrated and operational CWIS, and being able to communicate this vision to the campus. Much of the preliminary work involved marketing and cheerleading, but they were necessary in getting the campus "hooked" on a CWIS. The coordinator had to develop good recordkeeping tools to keep track of the growth of the menu structure and the information providers. Good records were essential in troubleshooting problems effectively. They were also important in developing tools to streamline the operation of the KCInfo. The coordinator was also responsible for the training and support of the information providers. This included: developing a training workshop for them; writing a handbook of guidelines; developing ways of streamlining the preparation and maintenance of documents; and phone consulting. KCInfo itself had to be monitored regularly. This meant spot- checking the documents for inaccurate and out-of-date information. When first beginning, information providers will need reminders to review their documents. This can be done with a monthly email message, which can also be used to request feedback from the information providers. INFORMATION PROVIDERS: THE WORKER BEES As mentioned earlier, the information providers were given responsibility for the editorial management of the information for a particular sponsor. They had to be trained in formatting, updating, and posting documents to KCInfo, and deleting them when they became obsolete. All information providers were required to take an initial training workshop for this purpose. There were NO exceptions to this rule. This workshop covered: job responsibilities; standards; a demo of KCInfo; help in organizing information and setting up a work environment; and instruction in preparing and maintaining documents. The Information Providers Handbook was written for this workshop and contained all the above information for later reference. ONGOING DEVELOPMENT: KEEPING THE WAGONS ROLLING KCInfo passed through some distinct phases in its development. In the preliminary or "evangelistic" phase, the coordinator had to work hard to motivate potential sponsors and to interest the campus in using KCInfo. After some initial training workshops were held, there was a very gradual growth during the next six months, as a few of the key departments came on board. But by the end of its first year, the number of documents in KCInfo had grown to nearly 1,000 and the number of connections to it per week increased to nearly 10,000. As more information was added, the usage increased and as usage increased, more potential sponsors came on board, as they began to realize its potential. It became difficult to keep up with the demand for workshops and menus. It is important to keep in mind that it took about a year for KCInfo to become truly useful, when the point of "critical mass" was reached. This was the point where people began to consider KCInfo as an integral part of campus life. The pendulum had swung, so to speak. During this phase, good recordkeeping became especially critical in order to keep the growth organized and under control. The coordinator must continue to remain "tuned in" to new developments on the Internet affecting gopher and to new resources that might be added to the CWIS. Regularly collecting and organizing information from the CWIS-L listserv and other Internet newsgroups helped us to incorporate important new features into KCInfo. As KCInfo grows, so will the need for new tools that will streamline and even automate posting and file maintenance. Some of these tools have already been developed at other gopher sites. Some examples are: automatic expiration of documents; document conversion into ascii format; automated transfer of information into the CWIS; reports on "what's new" in the CWIS; document access statistics for information providers; programs that identify "dead" links to other servers, full text and keyword search tools. Many of these are available upon request. The UMN gopher or the CWIS-L listserv are good sources of help. EVALUATION: LOOKING FOR FEEDBACK IN ALL THE RIGHT PLACES As with any new project, it is important to get regular feedback in as many ways as possible. This can be done, of course, with periodic surveys and questionnaires. Patterns of growth can also be plotted from gopher usage logs, which also can provide document access statistics to information providers. Some of the most valuable feedback is less formal: suggestions from users via a "comments" box on the main menu; followup with information providers; email from remote viewers; the CWIS "image", revealed subtly in references to it in conversation and campus publications. Even negative feedback is valuable in providing ideas and impetus for improvement. CONCLUSION The explosion of information in our society today requires that we, as information technology experts, find the most efficient tools for accessing that information and presenting it in a meaningful way. KCInfo has become that kind of tool, providing access to a broad range of information at a relatively inexpensive cost to users wherever they are and whenever they may need it. Although KCInfo was intended as a supplement to other forms of information distribution, it has already replaced some kinds of paper-based communication and promises to significantly reduce the use of paper over time. One of its biggest benefits is that it has greatly improved access to information for the campus and has increased communication among all groups: students, faculty, staff, alumni, parents, and friends of the college. And as it has grown, it has been instrumental in promoting a greater "cohesiveness" in the life of the Kenyon community.