(next) or
(previous)
Table of Contents
The architecture articulates a philosophy and set of principles to establish a set of basic patterns for designing and implementing systems to support current and reengineered processes. This set of principles is based upon an assessment of information systems currently in use at the University and the information system requirements presented by users as part of the analysis phase of the project. The principles specify a distributed computing environment, an open systems environment, sharing of information, adherence to standards, use of modeling techniques, making information visible, utilizing consistent interfaces based on human factors, distributed systems management, client/server computing, process oriented application program structure and a focus on end-user computing.
The philosophy and related principles provide a framework for the information architecture by articulating the objectives and quality characteristics that the architecture should follow. These, in turn, are intended to guide the analysis, design and decisions made relative to all aspects of information systems and processes at the University. The quality characteristics stated in the philosophy determine the technological approach taken in defining components of the architecture and how they must operate. The principles are meant to provide a set of patterns by which information system design decisions can be made.
One of the primary goals of the project was to establish a unifying philosophy for future information services at the University. The philosophy encompasses both technology (information architecture) and practice (process innovation).
The discussion below describes the objectives for the new University Information System (UIS), a system concept for satisfying those objectives, and a set of principles for guiding implementation of the UIS.
An information system is designed to satisfy specific requirements. The objectives are the high-level requirements that have the most direct impact on the system. The objectives for the UIS are put forth as goals for UIS activities and applications. An activity is a use of the UIS to accomplish a specific outcome. Examples include paying for a purchase, determining the implications of a change in strategy, registering a student, and preparing a mailing to a specific population. An application supports a set of related activities. Examples include a general ledger system, a decision support system, a student information system, and a human resources system.
Focus on Activities: Users should be able to focus on the activity they are performing, rather than on the system they are using to perform the activities. The UIS will allow users to focus on activities by:
1. Organizing applications around activities;
2. Delivering the information needed to perform the activity, and filtering out information not needed;
3. Providing on-line help and explanations of policies governing the activity;
4. Standardizing the look and feel of user interfaces across applications; and,
5. Standardizing the information processing tasks that make up an activity.
Sharing of Information: Local units (operating units, schools, and campuses) should be able to share information. The UIS will promote sharing of information by:
1. Aggregating data needed by individual units with information central to the University, without undue constraints to the individual units;
2. Distributing applications independent of how data is distributed;
3. Providing information about the data available through the UIS in a way that facilitates viewing, retrieval, reporting, and analysis; and,
4. Providing means for controlling access to information as privacy considerations dictate.
Quality of Information: Information must be timely, reliable, and accurate. The UIS will assure the quality of information by:
1. Providing the means of establishing ownership of processes and their associated data, and the means for assigning authority to enter or modify data;
2. Checking the consistency and completeness of data at its entry point, and assuring the consistency and completeness of data throughout its lifetime; and,
3. Eliminating single points of hardware failure.
Reduction of Effort: The UIS will reduce the manual effort required to perform information processing activities, including archival activities, and will reduce the use of paper to store and distribute information. The UIS will allow the University to reduce manual effort and paperwork by:
1. Eliminating the keying of data available elsewhere in electronic form;
2. Providing a user interface that facilitates the entry and modification of data;
3. Providing the means to approve transactions electronically;
4. Providing systematic archiving of data to inexpensive storage media, and the means to access archived data for review; and,
5. Capturing records of activities performed, and making the records available for subsequent analysis.
Flexibility: The UIS should be easy to adapt to changes in policy, to incremental improvements in processes, to specific needs of local units, and to advances in technology. The UIS will be flexible by:
1. Allowing the addition of new data fields without requiring changes to programs that don't use the new data;
2. Facilitating changes in where activities are performed, when activities are performed, how activities are performed, and who is authorized to perform activities;
3. Facilitating changes to processing details such as changes in course prerequisites, redesign of report formats, and extensions to the data produced by an activity;
4. Reducing dependence on specific hardware and operating systems;
5. Allowing economical, incremental additions of network and computing capacity; and,
6. Integrating autonomous local systems with the overall UIS.
A system concept is a high-level description of the design of a system, specific enough to communicate the nature of the system to users. It establishes the strategies and framework that will serve as a reference point for subsequent design decisions.
Architectural Overview: To its users, the UIS will appear as a single set of applications automating the information processing activities the user performs. All activities will involve a familiar set of information processing tasks, each with a standard interface. The system will create the illusion that all data is stored and processed at the user's location.
The UIS architecture will be distributed and layered. Eventually, all applications will be constructed and integrated using foundation software, including a data management system, common utilities, a user interface library, and network software. Each will conform with emerging industry standards for distributed information systems. Such standards facilitate the use of common tools such as spreadsheets and statistical packages, facilitate electronic data interchange with organizations outside the University, and promote independence from individual vendors. In the interim, existing systems and commercial packages will be evaluated on their ability to meet functional needs, their compatibility with UIS data management and network standards, and the ease of integrating them with the UIS interface library and common utilities.
Data Management System: The UIS data management system will standardize the description, storage, and retrieval of all UIS data. All applications will access data through services provided by the UIS data management system. Furthermore, applications will use the data management system to determine whether a user has privileges to perform activities. The data management system will:
1. Implement a natural model of the data and will minimize constraints on physical characteristics such as the size of fields and the number of links among records;
2. Integrate central data with local data (data of interest only to individual campuses, operating units, and schools) without requiring users to differentiate between the two;
3. Allow distributed access to data and provide mechanisms for purposeful control of access to data;
4. Provide user-friendly navigation of UIS data, and a straightforward language to define subsets of data for decision support tools such as spreadsheets and statistical packages;
5. Allow both real-time and batch updates of data;
6. Provide archiving services; and,
7. Allow replication and partitioning of data.
Common Utilities: The UIS common utilities will provide applications with consistent services common to most UIS applications. Use of the common utilities by all UIS applications will help standardize the way users perceive and perform activities, and will reduce the effort required to create and integrate applications. The UIS common utilities will include software that:
1. Provides on-line help and on-line access to policy, both in context of the activity being performed;
2. Checks the completeness and consistency of inputs, allowing exceptions according to policy;
3. Allows a user to select an activity to be performed;
4. Notifies users when an activity has been triggered;
5. Logs descriptions of activities; and,
6. Schedules the distribution of results of activities.
User Interface Library: The UIS user interface library will allow an application to interact with a variety of interface devices. Applications will obtain interactive input and present interactive output through routines in the user-interface library. Consistent use of library routines will standardize the look and feel across UIS applications. The user-interface library will:
1. Be based on window environments featuring the use of menus, control buttons, dialog boxes, command keys, and click-and-drag and cut-and-paste operations to initiate and perform activities; and,
2. Support the variety of access devices in use at the University, with lower-capability devices emulating higher-capability devices within limits imposed by the lower-capability devices.
Network Software: The UIS network software will tie together the configuration of computer processors, access devices, data storage devices, and other hardware employed by the UIS. The network software also will connect the UIS to information systems used by University suppliers, contractors, and customers. The UIS network software will:
1. Support client-server technology as the basic paradigm for providing information services;
2. Provide a bridge allowing electronic interchange of data with outside systems;
3. Support single sign-on security; and,
4. Support network management.
The following architectural principles are intended to provide a framework for implementing the information architecture. These principles should guide the analysis, design and decisions made relative to all aspects of information systems and processes at the University. They are meant to provide a set of patterns by which information system design decisions can be made.
Regard information as a University asset: Although it is a commonly held view that information is an important organizational asset, the costs of creating, acquiring, maintaining, and managing information has a high degree of uncertainty. If every item of data and every document in the University is considered an asset, then all data and documents can be treated equally. However, data and documents are not all of equal value and some criteria must be formulated to determine what data and documents will be treated as an asset to the enterprise and which will not. This is not unlike other items created and acquired by the University.
Treat information technology and infrastructure as a University asset: The University's computing hardware, software and networking investment is an asset and will be treated as such. This requires managing all information technology from the desktop to the mainframe in a manner similar to other University assets.
Use functional capability and cost effectiveness as investment criteria: Investments in information technology must provide the functional capabilities required by users and support the services demanded by customers. Information technology must be shown to be cost effective and fit the architecture before an investment will be made.
Guarantee choice via an Open Systems Environment: An Open Systems Environment (OSE) consists of a computing infrastructure that facilitates the acquisition and development of application software having the following characteristics:
Can be executed on any vendor's platform;
Can use any vendor's operating system;
Can access any vendor's database;
Can communicate and interoperate over any network;
Are manageable;
Are secure; and,
Interact with users through a common human/computer interface.
Such an environment supports portable, scaleable, and inter-operable applications through the use of standard services, interfaces, data formats, document formats and protocols. Openness means commercially open, technologically open, operationally open, and user open.
Allow shared access to information via adherence to database standards: The most important capability demanded by users is access to data that is required to operate and manage their responsibilities. The adherence to database standards for storing, processing and accessing data and documents is necessary to guarantee sharing of data by multiple users.
Base information processing on policies, standards, and models: The acquisition of data and its subsequent processing and use will be guided by a set of information policies and procedures which reflect the adherence to industry, national and international standards. Information processing will be based upon models such as an enterprise data model, a client/server processing model, a hierarchical, data warehouse storage model, a distributed computing model and process models, among others.
Establish data ownership and stewardship policies: The ownership of information is an issue for all organizations that strive to manage and eliminate the duplication of information. The main questions are who owns the information and who should have the responsibility for its creation, modification, accuracy, reliability, consistency, security, and timeliness. In many cases, information is duplicated because multiple units in the enterprise believe they have ownership of information and are responsible for it in attempting to service their customers (students, faculty, funding agencies, etc.).
The owners of data and documents should have the major voice in its identification, creation, indexing, storage, retrieval, reporting, and security as long as it is congruent with enterprise-wide information policy. A data and document administration function will be responsible for ownership and stewardship policy and procedures.
Along with ownership goes a set of responsibilities which the owners must assume. The owners are responsible for the accuracy, integrity, security, reliability, versions and timeliness of their data and documents. Although the actual implementation of these responsibilities may be fully or partially relegated to the stewards of the information, the ultimate responsibility lies with the owners.
Ensure data and document integrity via a data administration function: A data and document administration function is responsible for proposing and implementing policies and procedures related to all aspects of the data and document management activity. Primary among these responsibilities are:
Maintaining enterprise-wide data and document standards;
Identifying and defining data and document sources;
Analyzing documents and defining Document Type Definitions (DTD);
Standardizing naming conventions;
Defining a data model in conjunction with users;
Resolving data and document ownership issues;
Tracking the content of the University databases and data repositories;
Designing and implementing training and informational programs; and,
Evaluating and recommending data and document management hardware and software.
Make data and documents visible: Provide a means for easily determining and identifying from any geographical location what data, documents, services, people, equipment, buildings and other resources are available across the University enterprise using information processing technologies. The establishment of an information resources repository based upon an inventory of all information will accomplish this objective.
Provide flexibility via a client/server architecture: Client/server is an alternative information system architecture to the mainframe-centric architectures that have dominated computing in large organizations for the last 30 years. The client/server model provides information services that attempt to maximize the benefits of low-cost, high-performance, open systems computing devices without risking critical mainframe-based applications and without wasting the investment in mainframe application and information system development skills.
A client is a computer attached to a network that a user needs to access network resources such as other computers, printers, FAX's, or databases. A server is a machine that provides client machines with services and resources such disk storage, databases, application programs, or network connections. Servers can be mainframes, minicomputers, workstations, powerful microcomputers or other specially-designed networked devices.
Establish ease of use via an intuitive graphical user interface (GUI): An easy-to-use and easy-to-learn graphical user interface will allow end users to access information, technology and systems in such a manner that the majority of their effort will be spent in performing the intended task without the technology and systems interfering with this objective.
Integrate applications and workflow via a process orientation: By viewing the University enterprise as comprised of a series of processes supported by functional units working together rather than individual functional units in an assembly line, work flow and integration of applications can be achieved. The end result is higher quality and more efficient service for customers.
Make seamless interconnections: The ability to access, send and receive data and documents without regard to physical location will be possible via networking capabilities and a highly integrated information system environment.
Reduce paper by capturing data one time, at its source: Acquiring data in a digital form at the time it is made available and storing it in a standard database that can be accessed by all who are authorized to utilize it will reduce the creation of paper documents and eliminate the physical flow of paper within the organization.
Provide workflow assistance: The implementation of workflow software will assist information workers in their work through the use of work flow knowledge, automatic alerting, electronic forms routing, and electronic authorization.
Ensure effective use of information technology via education and training: Investing in technology without investing in the people who use the technology is not cost effective. The University will educate and train its users to enable them to become the most technologically literate University staff of its size in the United States.
Introduction
Framework for Building Information Systems
Philosophy and Principles
Architectural Overview
Summary
or (Return
to Reshaping the Enterprise: An Overview)