EXECUTIVE SUMMARY --- PENN OFFICE SYSTEMS;
PLANNING A NEW ARCHITECTURE FOR
PERSONAL AND WORKGROUP PRODUCTIVITY SYSTEMS
VERSION 1.1a
University of Pennsylvania Electronic Mail Task Force Working
Group on Office Systems
December 1993 (HTML version last changed on 13 January 1994)
- Introduction
- Findings
- Functions
- Environmental Requirements
- Technology Gap
- Recommendations
- Conclusion
The University of Pennsylvania Electronic Mail Task Force's Working
Group on Office Systems was created and charged by the Vice Provost
for Information Systems and Computing with charting the course of a
new generation of Office Systems. The overarching goal is to combine
the responsiveness and flexibility of personal computers with networked
systems to support personal and workgroup productivity in a secure,
manageable, and reliable fashion.
The Penn Office Systems document addresses the needs of faculty, staff,
and students, many of whose needs have not been addressed by historical
Office Automation systems. In spite of this, we label the architecture
"Office Systems" for brevity, as well as to convey the importance of this
architecture in office environments. The term also emphasizes the
increasingly electronic nature of many of our most basic organizational
tasks --- messaging and e-mail, file management, scheduling, and above
all information sharing.
The first step for the Working Group on Office Systems was to
document the electronic communication functions needed now and in the
future in PENN OFFICE SYSTEMS; Planning a New Architecture
for Personal and Workgroup Productivity Systems. This document
provides a discussion of important findings, detailed outlines of all of the
functional and environmental requirements, and a technical discussion of
standards and protocols.
Although we foresee no "point-solution" fulfilling Penn's Office Systems
requirements, we expect the document's "vision" of Penn's Office Systems
will be relatively stable and long-lived, despite expected changes in
product recommendations due to the emergence of superior applications
and technologies. The document will serve as a "blueprint" for the
selection of standards, equipment, and software.
The most important findings concerning the business and technical issues
surrounding the transition to a new generation of Office Systems are as
follows:
Over the past several years electronic communication has become a
natural part of our daily work experience. Many people frequently
exchange electronic messages with friends and colleagues. Some use
electronic calendars to maintain both personal and office-wide schedules
that are accessible to others. Many people also need the ability to
communicate from home and while on the road.
Penn's Office Systems are currently fragmented, lack important
functions, and fail to address the needs of large segments of the Penn
community. For example, most electronic mail users on campus have no
means of exchanging formatted word-processor documents,
spreadsheets, or programs, while others can only exchange such files
with users of the same e-mail system.
All members of the Penn community, including faculty, staff, and
students, must be able to participate.
Adherence to common, underlying technical standards and protocols is
prerequisite for campus and world-wide communications.
Obsolete desktop computer systems and asynchronous network
connections need to be upgraded to support the next generation of Office
Systems.
The document also details extensive lists of functions and environmental
requirements for current and future Office Systems. Solutions fulfilling
some of these requirements exist today, while others represent industry
direction.
The Working Group has developed a list of the general functions that are
needed or desired in an Office System.
Electronic messaging has been promoted as a useful communication tool
that is convenient, reliable, and easy to use. However, as technology has
progressed and the business needs have changed, the features of our
major electronic messaging programs have failed to keep pace with our
more fully featured e-mail needs. Incompatible electronic mail packages
that are both more useful and less costly are being deployed,
emphasizing the need for a new electronic messaging architecture that
can accommodate our diversity, while supporting our current and future
business requirements.
In pursuing new electronic mail applications, we are looking for options
that provide the maximum quantity and quality of functions available,
while remaining open and flexible enough to meet the varying needs of
the different departments and Schools within the University.
The image of electronic mail as used for simple, short textural memos is
changing. Messages about meeting notifications or personal scheduling
are still prevalent, but increasingly people at Penn and elsewhere are
using e-mail to share more complicated documents, with varying
success. Sending traditional, text-only electronic messaging is not
particularly difficult, but sending almost anything else such as a
formatted word-processing document often requires painful trial and
error, deep technical expertise, and a technical assistant.
Multipurpose mail is defined as a system that allows composition and
sending, via the traditional e-mail delivery infrastructure, anything that can
be created on one computer so the recipient on a different computer can
successfully use it. We seek applications that easily allow the sending and
receiving of text, formatted documents, images and sound.
The ability of University faculty, staff and students to
manage their time
effectively is essential. The two main types of electronic calendaring are
personal and group calendaring. Personal calendaring allows individuals
to electronically manage their own calendars. Group calendaring
involves multiple individuals in a workgroup or organization managing
their personal calendars using software that also facilitates the
scheduling of meetings and appointments.
In finding a software solution for calendaring and time management,
personal scheduling and group scheduling features must be balanced. For
example, we seek a scheduling package that is easy to use, prints in a
variety of formats, handles repeating events, helps resolve conflicts, and
integrates well with to-do-list and project-management software. Users
should be able to control who has access to their calendars, and the extent
of that access.
Today, electronic messages, word-processing documents, calendars and
paper files are all managed separately. Gathering complete information
on any topic requires sorting through various filing systems that store
information differently. Some are stored by keyword, by date, by
subject, or by author. Gathering this information is time consuming and
difficult.
An integrated Office System facilitates seamless integration among all
sources of information upon which an office depends for its functioning.
The specific applications that comprise Office Systems should facilitate
filing and records management for senders and recipients, as well as others
who may be granted custodianship of messages and appointments.
If our electronic communications community were self-contained,
relatively small, and resided on a single system, addressing messages
and scheduling requests would be as simple as recalling a name.
However, to achieve universal electronic communications between office
at Penn and the world beyond Penn, we need services, not unlike
telephone directories that simplify addressing messages to users on
different e-mail systems. There are significant benefits for a directory
service that provides an easy way to obtain e-mail addresses and to
incorporate those addresses automatically when sending mail or
scheduling meetings. Integrated directory services can help remove the
barriers to broader campus use of electronic communications. E-mail and
calendars will be more useful when it is less difficult to obtain addresses,
and easy-to-use addressing will allow information to be shared in new
and valuable ways.
As we developed the list of functions we seek in our Office Systems, we
also identified four major issues within which to frame the approaches to
implementing our Office Systems environment.
As with any computer application, the user interface of an Office
Systems program determines not only what users can do, but what they
will choose to do. Currently, most of our office systems provide
primitive user interfaces, typically pre-dating the innovations of personal
computers and Graphical User Interfaces (GUIs). These systems have the
advantage of requiring only "lowest common denominator"
technologies. However, because they are not integrated with the local
desktop environment, they do not support multimedia, allow printing on
locally attached printers, nor have the same look-and-feel of other
applications running on the user's Macintosh or PC running Windows.
One of our goals is to incorporate the power, functionality, and ease-of-use
of desktop computers and their GUI's into our Office Systems. Where
possible, applications should be both similar across different desktops and
similar to other applications on the platform on which they are run. By
doing this, barriers to using the Office System applications are lowered,
allowing users to focus on accomplishing their work.
Members of the University increasingly need to use computers and
electronic communications wherever they go. Network access makes
interaction between remote users and campus offices not only a
possibility, but a necessity. On-campus use of the network and its
resources is also changing, as the network augments the telephone and
conventional mail as a vehicle for business communications.
The purpose of Office Systems is to assist members of Penn's
community to accomplish business tasks like sharing documents,
calendar management, accessing data to facilitate decision making and
sending messages to colleagues, whether they are in the office or out of
the country. Office systems must therefore function anywhere, anytime,
and in a consistent and reliable manner.
Today, electronic communication is available to a growing audience, and
individual Office System security needs vary. For those who use e-mail
to schedule their own meetings and to correspond with a small group of
peers on professional matters, security measures beyond the usual may
be of little concern. For someone in a visible, public position who deals
with sensitive matters, security may be a major concern and extra
precautions may be in order.
The Office System solution seeks a base level of security and password
encryption, with several levels of special facilities available for those who
need them. Such facilities include third-party validation of sender and
recipient authenticity, message integrity checking, and encryption.
Individuals now rely on the ability to communicate electronically with
colleagues and to access resources both inside and outside the Penn
community. However, because of Penn's decentralized computing
environment, individual schools, departments, groups and people are free
to make their own choices when selecting computing solutions. Often,
issues of interoperability and portability have not been prime
considerations in making these decisions.
One of the purposes of Office Systems is to facilitate the exchange of
information through the selection and implementation of protocols and
standards. The choice of protocols and standards should serve to create a
distributed communications environment that operates across Penn's
heterogeneous environment and provides enough functionality and
flexibility to meet Penn's many needs.
Presently, PennŐs Office systems needs are not fully met by any
combination of industry offerings. While second-generation
client/server technologies offer significant enhancements in some areas,
a "technology gap" inhibits our transition because other areas reveal
inferior functionality as compared to our first-generation, host-based
systems.
For example, in the client/server model, we find a significant
improvement in the user interface. Electronic mail in a product like
Eudora looks similar to all other products on the desktop and easily
allows interaction between multiple desktop applications. On the flip
side, dialing-in to your electronic mail in the client/server model is
presently difficult if not impossible. The goal of the Working Group is to
position the community to take advantage of the new technology by
providing products that serve as stepping stones to our Office Systems
vision. It is not the intent of the Working Group to try to meet all of the
needs immediately, but rather to provide the basic functionality and ease
of use.
The "double-S curve" graphs above illustrate the technology gap we
currently face in the user-interface and remote-access examples described
in the preceding paragraph.
The Working Group makes the following recommendations with respect
to Penn Office Systems:
Working Group continue effort
That the Working Group continue its efforts through
September of 1994,
at which time its progress and goals should be reviewed, and its charter
renewed or extended if appropriate;
ISC project management and technical support
That the Vice Provost for Information
Systems and Computing provide
project management and technical support for the on-going efforts of the
Working Group with the assistance of (guided by) the Working Group
Chairs; and
Steps
That the following, temporally ordered and overlapping steps be pursued
by the Group:
Step 1: Build consensus
Present the "Penn Office Systems" document to the
University
community for comments and modify it as necessary.
Step 2: Set priorities, standards and criteria
Prioritize functions, select standards, and
develop product evaluation
criteria, with input from ISC Advisory Committees, Project Cornerstone,
the Academic Computing Task Force, and the PennNet Architecture
Committee.
Step 3: Learn about, evaluate, and request information on products
Learn about and evaluate existing experimental and production second-
generation systems, both at Penn and at peer institutions. Test and
evaluate products through closely controlled pilot projects in conjunction
with established providers of service. Request information from vendors
on their products' abilities to address the needs described in "Penn Office
Systems."
Step 4: Enhance current office systems
Determine where, if anywhere, it is
reasonable to enhance existing
second-, and even first-, generation Office Systems products to facilitate
standards-based interoperability with new products.
Step 5: Recommend products
Recommend products that allow members of the community
to
participate fully in the Office Systems environment.
Step 6: Support conversion
Work with ISC and other members of Penn's community to
develop
appropriate support structures to aid in the conversion process.
Timetable
We are now in the process of eliciting comments, prioritizing
requirements, and selecting standards. We plan to begin evaluating
products during the first quarter of 1994, and hope to be able to make
recommendations by July 1, 1994.
The Working Group looks forward to receiving your comments on
Version 1.1 of "Penn Office Systems," and hearing your thoughts on
selecting standards, and setting priorities. We are eager to begin product
testing, pilot projects, evaluation, and implementation of our vision for
Penn Office Systems. We recognize that we face considerable
challenges and difficulties, but we expect the benefits to the Penn
community will make the effort worthwhile.
To comment or request more information, please contact Dr. Noam H. Arzt, arzt@isc.upenn.edu.
http://www.upenn.edu/open-systems/off-sys/ExecSumm.1.1.html