[NOTE: The graphs that accompany the word-processed version of this 
report (available at http://www.cosn.org/EPIE.html) have been "translated"
in most cases into "ASCII art", but in one case into a purely textual 
rendition, because the figure was too complex to do any other way in 
ASCII text.  Graphs were translated by Stanton McCandlish, mech@eff.org.]


EPIE Institute




Creating Learning Communities:
Practical, Universal Networking
for Learning in Schools and Homes 

February 1996


******************


A Report for School and Community 
 
Technology Planners 

and Policymakers 


 By

P. Kenneth Komoski                   
EPIE Institute                  
Project Director

W. Curtiss Priest
Center for Information, Technology & Society
Associate Project Director
                                 

Published by


The Educational Products Information Exchange (EPIE) Institute
Hampton Bays, New York


With Support From

The John D. and Catherine T. MacArthur Foundation




TABLE of CONTENTS


SECTION I	Introduction, Rationale and Overview Ñ Page 1


SECTION II	Achieving Practical Networking: "Do's, Don'ts, Maybe's and
Lessons"

	Creating Practical Networks within Schools, in Local Communities
(School-Home), and for Worldwide Connectivity Ñ Page 4

SECTION III	"Home Is Where the Time Is ... and the Payoff for
Improving School Learning" 

	Increasing the Quantity and Enhancing the Quality of At-home
Learning and Improving Parent-Teacher-Student Communication via
Equitable School-Home-Community Networking Ñ Page 72

SECTION IV	K-12 Networking Ñ Models and Benefits

	"Do the Benefits Justify the Costs?"   Modeling the Potential Impact
of Networking on the Development of a Learning Community Ñ Page 81


SECTION V	"An Achievable Vision" 

	Making Tomorrow Happen With Today's Technology Ñ Page 109

APPENDICES

	Appendix A - School-Home Computer Survey Form Ñ Page 122
	Appendix B - Bibliography Ñ Page 123

SURVEY RESULTS

	Community Networks Reaching Out to Schools Ñ Page 22
	Practical Networking Suggestions from CoSN ÑPage 63

Other Reports from Available from this Project:

	Priest (1995) Findings of the MacArthur Funded July 7th Workshop on
Cost-Effective K-12 School and School/Home Networking
	Priest & Risley (1995) Advances in Modelling School and School-Home
Networking
	Priest (1995) Results of Survey of Community Networking Activities
that Included Schools
	Priest (1995) Responses to Questionnaire Posted to the CoSNDisc
Discussion List on K-12 Networking
	Priest & Komoski (1995) Role for a Public Hand in Telecommunications
	Komoski (1995) Preventing the Virtual Ghetto: Electronic Equity via
Community Telecomputing

Project Partners:

EPIE Institute, 103-3 W. Montauk Highway, Hampton Bays, NY  11946.  Tel:
516-728-9100
Center for Information, Technology & Society, 466 Pleasant Street,
Melrose, MA  02176.  Tel: 617-662-4044

Sources and Acknowledgements



This report is based on information distilled from a wide range of
sources: the 670 print and electronic resources (referenced in Appendix
B), a workshop/conference attended by experts in school, community and
global networking (see Project Reports above), two online surveys of
schools and community-wide networks, and follow-up online and telephone
interviews with school technology coordinators, school administrators,
commercial vendors, networking specialists from schools, community
networks, businesses, universities and government agencies. 

All of the above generously shared their experiences, strategies,
techniques, criticisms, advice and vision about networking and its role
in enhancing in-school and at-home, community-wide learning.  We
gratefully acknowledge their individual and collective input which has
enabled us to base what follows on the realities faced by educational
networking advocates who are working within today's cost-cutting school
climate. 


We are particularly grateful for input from developers of nonprofit
grassroots community-wide networks and local educators and parents who
share a common vision of school-home connectivity as integral to the
development of learning communities.  We applaud all educators willing
to work with local community networks to achieve practical networking
economies and the benefits of school-home connectivity.  As such
cooperation becomes more common, learning communities of the sort
advocated in this report are likely to proliferate. 

We wish to express our gratitude to the John D. and Catherine T.
MacArthur Foundation for the generous support that has enabled the
Educational Products Information Exchange (EPIE) Institute to carry out
the work that has produced this report.  We particularly appreciate the
vision of Peter Gerber who encouraged us to undertake the work
accomplished with the aid of the Foundation's support.

SECTION I: Introduction

"We have access to grants, but we don't have access to communities" 
School Principal, Barry Vitcov, who is "...hoping to find money for an
Internet account for the school before the end of the academic year, but
he worries about the cost at a time when there are other concerns, like
fixing the school's roof which leaks during downpours." (The New York
Times, Jan. 29, 1996)


Introduction and Rationale 

This report appears at a time when most schools and their communities
are operating under increasing fiscal constraints.  This is underscored
by a recent federal (GAO, 1995) report alerting the nation to the eroded
physical condition of its public school infrastructure.  However, this
is also a time when national political and business leaders are urging
that by 2000 all the nation's classrooms and libraries be networked to
the National Information Infrastructure (NII). 

Ensuring networked connectivity of schools and libraries to the NII via
the Internet is seen as essential for preparing America's youth to share
in the anticipated educational, economic and social benefits of life in
the Information Age.  Yet many schools and communities may feel that
they are being asked to invest in a technology with unknown costs, and
largely undemonstrated benefits.  Others may feel that the goal, though
laudable, is impractical, given the constrains under which a their
schools are having to function. 

This report is designed to be a useful tool for those school and
community decisionmakers who may agree with the goal of networking, but
who are concerned about the practicality of achieving it for their
schools and communities.  The most obvious of these concerns are related
to technical, fiscal, and timetable decisions.  But other, ultimately
more important concerns relate to broader and deeper educational and
social dimensions and decisions.  

Among these concerns, there are two that are given particular emphasis
in this report: 

1.  How can networking be effectively used to strengthen and increase
(a) at-home learning for students, (b) parents' support of that learning
and (c) teachers' support of parents involvement in their children's
learning?

2.  How can at-home access to these networked benefits and enhancements
of at-home learning be accessible equitably to all students and parents
within all schools in a community?

The report attempts to relate specific fiscal and technical decisions to
such broader educational/social dimensions and decisions implicit in
these two emphases.  

We have tried to do this in ways that relate the networking of schools
to local and global opportunities to improve communication and learning
among teachers, parents and others who are major players within the
overall "learning ecology" within which students learn (Niebuhr, 1993;
Komoski, 1994).

Our goal is to help school and community decisionmakers develop
practical, equitable and sustainable telecommunication networks capable
of:

1.  Enhancing learning for all learners by contributing to the
development of local learning communities with school-wide,
community-wide/world-wide connectivity;

2.  Improving parent-teacher communication and cooperation via
school-home e-mail and/or voice mail communication networking and
supporting parents' interest in improving children's learning both at
home and in school via homework help, student progress reports,
home-school conferencing and access online learning resources;

3.  Encouraging students to devote significant amounts of out-of-school
time Ñ at home or at networked community learning centers in libraries
etc. Ñ engaging in computer mediated communication activities such as
information exploration, cooperative learning as well as through the use
of curriculum-related instructional resources;

4.  Involving the local community in positively supporting and
strengthening all elements of a community's overall "learning ecology"
(e.g., schools, libraries, families, peers, community centers, media,
neighborhood organizations, youth groups, health, social and civic
agencies, etc. (Niebuhr, 1993);

5.  Responding to the changing needs of learners, teachers, parents and
adapting to changing technologies and pedagogies;

6.  Being supportive of Ñ and being supported by Ñ the local community
as the infrastructure for an evolving learning community.

Answers to the many practical, technical, educational and fiscal
questions to be dealt with to achieve the outcomes implicit in the above
statements will seldom, if ever, be the same for any two schools and
their communities.  What is practical and affordable in one school
community may be judged not feasible by another.  What may be affordable
initially, because of an available grant, gift, or business partnership,
may prove otherwise once such funding has ended.  Some schools and
communities may be granted a "free launch" into cyberspace, but once
launched, is the momentum sustainable? 

The costs of sustaining the technology and ongoing staff development for
networked schools with local and global connectivity are largely
unknown.  Unfortunately,they will remain so until more experience is
gained and the results of that experience are accessible to schools and
communities, via the very networking capabilities addressed in this
report.  Nevertheless, this report is intended to be a practical, useful
work for those school technology planners and policy makers who agree
that the networking of schools and homes can facilitate the building of
learning communities in which equitable, universal access to electronic
learning can be achieved.

Through school-home networking there are those who envision (1) students
and their parents better understanding what curriculum goals students
are to achieve, and (2) students learning how to access the resources
they need to achieve both school's and their own personal goals.  Others
are questioning whether enough teachers are prepared to shift from
traditional chalk-talk-textbook instruction to mentoring, monitoring,
and verifying a student's understanding and progress as a result her/his
exploring and constructing knowledge via inquiry-based learning?  While
some teachers may be ready for such a transition, others are reticent. 
Some of their reticence may be justified, in that effective instruction,
no matter how it is mediated, always has educational value.  Thus, in
time, we feel that teachers and schools will find learner-appropriate
uses of both instruction and exploration learning (see EPIE/CITS Model
in Section IV of this report).

If and when the above-mentioned shift occurs in a particular school it
will depend on many factors.  Not the least of these is when and how
teachers, school administrators and school boards begin focusing on what
is going on around them in the overall learning ecology of the larger
local and global community of which schools are but a part.

The Dynamics of a Complex Undertaking

At whatever level school and community decisionmakers proceed with
networking, they are faced with a complex set of interrelated
educational, technical and fiscal decisions that are both difficult and
important.  They are particularly difficult because of four interacting
dynamics: 

1.  The dynamic of technological change, innovation and competing
networking solutions; 

2.  The dynamic interaction between new technologies of networking and
new approaches to teaching and learning, especially those emanating for
research in cognitive science and a tension between long-standing
pedagogy of teacher-directed instruction and an emergent pedagogy of
self-directed, knowledge construction and exploration learning; 

3.  The dynamic tension between the promise of the new networking
technologies and the many practical and fiscal questions that must be
answered with great specificity before that promise can be realized;

4.  The dynamic growth of computers and connectivity in the homes of
increasing numbers of students and the social tension resulting from the
absence of this technology in the homes of those who cannot afford it. 
Proposed solutions to this at-home access disparity are addressed.

In light of these four dynamics, the report is designed to be a useful
technical and educational resource of practical advice and information
on school/community networking, as well as a guide to other relevant
sources of help on a topic of great educational and social importance.

Overview of the Following Sections 

Strategies, suggestions, technologies and recommendations to consider
when addressing the four dynamics mentioned above are contained in
Sections II-V, as follows:

SECTION II deals with the "Do's and Don'ts" of creating affordable
in-school networks as well as school-home-local and global network
connectivity.

It describes issues and options related to developing practical,
affordable telecommunication networks that provide school-wide,
community-wide and world-wide connectivity.  It focuses on the "do's"
and "don'ts," the "hows," "whys," and lessons that can help schools and
their communities develop networks designed to enhance and extend
learning opportunities.  It emphasizes the importance of staff
development and the need for teachers, students and parents to have
ready access to information about relevant local, national and global
resources.  It also discusses the implications of newer technologies
such as wireless and DVD (Digital Video Disc) on networking.

SECTION III focuses on how networking technologies may be effectively
used to enhance student learning at home, and to improve parent-teacher
communication and cooperation focused on improving children's learning
at home and in school.  It describes:

-  The implications of the growing nationwide disparity between access
to computers in the homes of many students and in schools;

-  The capability of school-home networking to expand students' at-home
learning time and opportunities, and to facilitate parents' involvement
with their children's learning, homework and their communication with
teachers; 

-  The challenge of extending expanded at-home learning opportunities to
those students whose parents cannot afford a home computer.

SECTION IV examines more fully the purposes, promise, and benefits of
networking and telecommunications described in the first three sections.
 While acknowledging that the overall process of education defies simple
analysis, this section describes the directions in which networking
technology is capable of taking teaching and learning on the threshold
of a new century. This is done with reference to a multi-dimensional
dynamic model of K-12 Schooling and Networking and factors related to
learner effectiveness. 

SECTION V describes practical steps schools and communities can take
today with available networking technologies to develop an effective
learning community and to strengthen all aspects of the overall learning
ecology of school and community. In addition, practical strategies for
achieving equitable access to computers, connectivity, and training for
students and parents from low-income households are addressed.

SECTION II:  Achieving Practical Networking


I am not saying that a World Encyclopedia will in itself solve any
single one of the vast problems that must be solved if man is to escape
from his present dangers and distresses and enter upon a more hopeful
phase of history; what I am saying Ñ and saying with their utmost of
conviction Ñ is this, that without a World Encyclopedia to hold men's
minds together in something like a common interpretation of reality,
there is no hope whatsoever of anything but an accidental and transitory
alleviation of any of our world troubles.  As mankind is, so it will
remain, until it pulls its mind together.  And if it does not pull its
mind together then I do not see how it can help but decline.
H.G. Wells, World Brain, 1938, pp. 24-25


When this study was originally conceived, we imagined creating a matrix
of teaching objectives and comparing those objectives with networking
technology.  This goal was challenged in many ways:

-  Some proponents of networking technology are actually advocates of
technology-aided school reform.  How does one assess a moving target
where the objectives of curriculum are being challenged and changed by
the technology itself?

-  It became apparent that networking involved more than the conveyance
of information Ñ it involves communication, monitoring, and feedback. 
Out of this realization we expanded the scope of the project to include
motivational effects of networks such as feedback for improved student
self-esteem.

-  Schooling has traditionally been under local and state control.  This
complicates matters because each state and, often, each school district
defines its own curriculum objectives.  Different objectives imply the
need for different priorities among networking technologies and
networked resources.

-  Almost weekly there is yet another important technological
development that pertains to networking.  For example, we have been
tracking the developments of very high storage CD-ROM discs.  Called DVD
(digital video disc), these discs can contain nearly 20 gigabytes of
data, about 30 times the storage of current CD-ROM discs.  How does one
provide practical advice for K-12 schools when the relative costs of
local versus distant archival of knowledge can change so dramatically?

-  The costs of networking are also dependent on the Telecommunications
Act of 1996 which was passed by Congress on February 1st, 1996.  The
bill provides for affordable telecommunications for K-12 schools.  Yet
there is no guidance in the bill as to what determines affordability. 
The decision is left to State utility commissions and the FCC.  This
results in cost uncertainties about school costs for telecommunications
(see Rate Reduction tips below).

-  Some well regarded observers of education and educational technology
note that technologies that conform to how teachers teach in classrooms
are accepted, while those that provide other promises are consistently
rejected, decade after decade.  For example, it is often stated that the
blackboard and chalk has been one of the most pervasive technologies in
schools.  This raises the specter of digital network connections in
classrooms going unused.

-  In contrast, there are students and teachers building web pages in
cyberspace, some electronic portfolios are already available for both
employers and colleges to peruse, and some teachers are acceding their
control of the classroom.

-  There is not one network to consider but several.  There are existing
networks of people and practices. There are digital networks and there
are telephone networks (both within school and from home to school). 
And, there are video networks Ñ some for distance learning, some for
cable programming, and some for locally, but centrally, concentrated
resources such as VCR's.  There are proponents of a single network for
all these purposes pitted against the reality of network devices that
only work on cables designed specifically for those devices.

-  There are many players.  Media specialists bring one point of view,
librarians another, computer lab supervisors another, innovative
teachers another, Internet surfers another, and several thousand
software and content providers provide yet other points of view.  School
administrators determine the standards by which teachers are judged and
how funds can be spent.

-  Other respected observers note that the process of public schooling
is bound by several complex roles including "conserving cultural
values," encouraging creativity, acting as custodians, and credentialing
each student.  To these observers the relationship of networking
technology to all of these tasks is not clear.

-  Many teachers and students are eager for the excitement that
exploring the Internet brings inside the walls of schools.  Can learning
always be fun?

-  Roles of parents are in flux.  Parental involvement in every day
schooling has become technological more feasible.  Will enough parents
have the time and interest to make a difference in improving children's
learning?

-  The U.S. economy has entered a period of extremely high debt levels
at the personal, municipal, state and federal levels.  The federal
government hands off more welfare and health care activities to states
and localities, these demands will compete for scarce local funds. Yet,
despite these concerns many schools are eying bond issues to pay for
technology.

-  Eventual costs of telecommunications are largely unknown.  While the
costs of low volume traffic such as e-mail will remain trivial, the
costs of higher end applications such as video conferencing are yet to
be understood.  Also current congestion through many Internet providers
is paid for by waiting for responses causing some to say WWW stands for
"World Wide Wait."  What can schools expect for response times?

-  Most materials on the Internet are free of additional costs.  But the
development costs of high quality content, especially documentary
quality videos, are very expensive.  Schools do not know what charges
they will eventually see for networked resources when quality materials,
especially designed for teaching, are made available.

-  Large uncertainties exist around issues of digital copyrights. 
Industry has been quite vocal about protecting copyright and this will
restrict a school's ability to use these materials.  Legislation is in
preparation that looks especially onerous to users and educators' rights
under "Fair Use" may be abridged.  (Hearings on legislation are being
held as this chapter is being edited.)

-  In addition to many sources of uncertain costs and availability, the
value of networking is also vague and uncertain.  Already there are two
camps forming: 1.)  "Netniks" extoll the wonders of telecommunication
based projects as sources of "authentic learning" while, 2.) 
"Stollniks" (after Clifford Stoll, author of Silicon Snake Oil) fret
about the loss of real teachers and true hands-on learning experiences. 
As one of them put it, "Internet pictures of frogs are not real frogs
(Brown, Unplug, 1995)."

-  Many parents look to schools to provide highly structured,
disciplined environments.  Self-directed learning is low on their list
of priorities.

-  Many other parents do look to schools to provide opportunities for
independence and creativity, and welcome technologies that encourage
this.

-  In an era where 47% of today's jobs require computer experience
(McKinsey, 1995, p. 7) parents are encouraging students to gain this
experience.

-  But as noted in Section Three (Grunwald Associates, 1995) the
purchase of a home computer to aid a child's learning is now the
dominant driver of home computer purchases.  As also noted, home
purchases of computer software and Internet connectivity greatly exceed
purchases made by schools.  What role do public schools have in
redressing inequities in access between families that can afford these
tools and those that cannot?  And what role do states have to address
possible "pauper technology schools" Ñ similar to the state crusades
against pauper-schools such as occurred in Pennsylvania in 1834 and New
Jersey in 1838 (Cubberley, 1920).

-  Where does the time come from for teacher learning and adaptation? 
The recent National Education Commission on Time & Learning (Kane, 1994)
found teachers being asked to do more in an environment already time
deprived and "imprisoned" by time.

-  There is actually too much information, too many stories for any one
teacher to read.  Magazines, journals and books recount stories of
Internet projects, and networking examples.  The Internet, itself, is
increasingly populated by accounts of technology plans (e.g.
http://www2.msstate.edu:80/~lsa1/nctp) or school case studies (e.g.
http://www.ednet.apple.com or http://www.lloyd.com).  Yet there is too
little wisdom, perhaps because of the newness of the task before us.  In
reality many technology plans borrow from too few resources, fettered by
a limited access to counsel or expertise.

-  Where is the line between education and edutainment?  As parents we
witness how mesmerizing video games are to a child.  In recent months
several vendors have offered enhancements to MOO/MUD environments to
extend them towards places of virtual reality where students can become
virtual characters and interact with other students and fabricated
features of the environment.  Virtual online universities do exist (e.g.
Athena University, http://www.athena.edu) and very serious teaching and
learning takes place in these environments.  What is the relationship of
these virtual worlds to the world of the classroom, or to the
"classroom" of living in a community?

-  Where does the current technology for learning and collaboration lie
in relation to our educational technology needs?  The Web page, while
graphically alluring and more intuitive than earlier Internet tools, is
a morass of uneven, disjoint educational resources (Priest, 1995,
Primer).  Databases of quality and depth such as by Dialog, EPIC, and
Homework Helper are only available via the Internet for a fee (often
substantial).  And collaborative tools such as joint-authoring
softwares, web-based computer teleconferencing, threaded newsreaders,
and listserve exchanges are still at an awkward stage of development
where hours of interaction can produce little learning.  Information
"when needed, where needed, as needed" is still an elusive goal for the
developers of groupware and collaborative tools. (Perhaps one of the
most advanced so far is CSILE (in Means, 1995, p. 14).)

-  There is the Myth of the Online Expert and other resource fallacies. 
Some fortunate students have been privileged to partake in networking
experiences which have brought them in contact with scientists and other
experts.  In one study parents were reported to be envious of their
child's access to famous authors (Teles, 1991, p. 69).  We must
recognize that such opportunities cannot be common to all students. 
Some accounts of learning scenarios (in Fulton, 1995, and elsewhere)
promise unrealistic amounts of access to scarce resources.  Nonetheless,
we will have improved access to "secondary-experts" Ñ librarians and
teachers who specialize in particular knowledge areas and can be
contacted via networks.  And we certainly should continue to encourage
scientists and others to selectively identify blossoming student talent
that would benefit enormously from direct contact.

-  And the Everest Syndrome (Gallo, 1994, p. 17 Ñ "computers should be
implemented into the school curriculum 'because they are there.'" 
Indeed, when Honey (1993, Case Studies) identified the allure of the
Internet to teachers and students she used phrases such as "like an
ocean," "exploration, discovery, murkiness, mystery" and "uncharted
territory."  The thrill and challenge of such quests beckon many to the
Internet, but this exploratory call should not be equated with everyday
matters of curriculum and learning.

-  Some (e.g. Riel, 1995) depict significantly restructured schools
where teachers become just one of many resources at the tips of
student's fingers.  The description of this new kind of school is simply
vibrant.  No longer is there just teacher, but there are four levels Ñ
learning guides, entry teacher, mentor teacher, and master teacher. 
Made possible by the Charter movement, the students benefit from
Learning Centers built around themes, such as the "Ocean Center." 
Technological aids and tools are all about.

-  Stepping back, how readily will any one school system transform
itself into Riel's vision or something like it?  Will the Charter school
movement provide the necessary push?   Or will Charter schools stumble
as described in "Charter School's Hopes Collide With Reality (Avenoso,
1995)."

-  Will the business community be the impetus?  Will the National
Information Infrastructure Advisory Committee, along with Messrs. Gore
and Clinton provide a vision?  Will Richard Riley's leadership for
parental involvement and challenge grants to the states provide the
platform for change?  Will local superintendents and principals pave the
way?

It is in this bewildering torrent of complexities that this report
addresses practical school networking.  A common theme in the literature
is that there is no "one blueprint" for school networking.  Indeed, even
if everyone could agree on Ethernet vs token-ring, or CATV vs T-3
digital TV, the diversity of choices that relate to the specific
character of the state, localities, and personalities produces a myriad
of different choices for each school and each district.

There is no one right networking answer for all schools and communities.
 There are, however, a number of guidelines that help schools from
making mistakes.  We have distilled from the literature and from our
surveys and research a snapshot of do's and don'ts to serve as basic
guidelines for school and community technology planners.

Some of these suggestions are fundamental while others are relative.  An
illustration will help.  A fundamental do is "standardize."  Stick to
one type of wire, one type of LAN card, one brand of LAN card, one type
of PC, one brand of PC, etc.  Resist buying different brands at
different times because of changing prices or "close-out" sales.  You
standardize to reduce inefficiencies when you install equipment, when
you maintain equipment, and when you upgrade parts of the system.  The
few dollars savings on a "deal" is not worth the hours of aggravation
when a new "driver" is incompatible with an odd-ball PC.  If you are
going to encounter a conflict, you want the time to solve the conflict
to apply to all of your equipment.

Another do is to run more than one wire type when you wire a school. 
But this do is more relative to your circumstances and relative to when
you wire.  In the distant future one wire will suffice for all wiring
needs.  This single wire will provide electrical power, telephone
service, video service, and digital-network service.  Today, however,
each of these services have distinct and different wiring requirements. 
The cost of wiring is mostly the labor.  Category 3 wire is 4 cents a
foot, category 5 wire is 10 cents a foot, 1 fiber optic cable is 36
cents per foot, and 4 fiber optic cable is $1.02 a foot.

Technical aside:  there are many technical guides cited in our
bibliography.  The prices above are from MilesTek (1995) which is not
only a catalogue of wiring products but also contains wiring tips. The
choice of wiring relates to the speed of the signal passing through the
cable.  Category 3 wire is fine for telephone and can often serve for
digital data up to 10 Mbps (megabits per second).  Category 5 wire is
more common for data communications and can be used to about 100 Mbps. 
Fiber optic cable has an upward limit of 100 Gbps (gigabits per second)
(Chapman, 1995, Personal Computer).  CATV is often run using coaxial
cable RG-59U.  Digital data can also be run using RG58A/U or RG58C/U. 
Sometimes digital data is run over CATV, especially if there is CATV
cable already in place.  For affordable routers, terminations, and
interfaces the common practice today is 10 Mbps using Category 5 wire. 
While fiber cable, itself, is relatively inexpensive, the hardware
required to use its higher bandwidth is not.  (SAMS publishers sell a
guide Understanding Fiber Optics which contains 456 pages on all aspects
of fiber optics.)

If you decide that you do want telephone, CATV, and data in the near
future and the design of your school requires a high level of labor Ñ
say fifteen minutes per foot, at $45 per hour the labor-cost per foot is
$11.  In comparison, even 4 fiber optic cable is about 10% of the labor
cost.  In such as case you would seriously consider running fiber for
future use.  So running 4 cables at one time is often practical. (You
would leave the fiber unterminated as the cost of termination is
significant and the technology will improve over time).  If, however,
you have easily accessible conduits and can leave readily accessible
cords to pull fiber through in the future, then you might neglect fiber
now.  There is also the possibility that high speed transmission cable
(such as fiber) will change in the next ten years, but this doesn't
appear likely.

Also, one popular approach to wiring schools is with volunteer help. 
The temptation with volunteer help is to minimize all costs.  This can
mean that the coordinators are not thinking about multiple wiring needs.
 From a strict economic standpoint the labor cost is still the dominant
cost regardless of how it is provided (volunteer wiring of schools
simply is a contribution to the school in lieu of higher school taxes). 
Thus, rationally, the fullest complement of cables should be pulled. 
Nonetheless it is somewhat more difficult to pull multiple cables and
this additional burden may seem to be to much to ask of volunteers. 
(Often the volunteers have a "notion" that they are connecting their
school to the Internet, and the presence of other cables may feel
superfluous.)

Thus the pulling of multiple wires is a relative issue.  It depends on
many factors particular to each site, decisions about different
networking needs, and the source of labor.

Background on Networking

Networking choices for schools, for the many reasons outlined above, are
not been made with very clear objectives.  Rather, they are largely made
by imitating the low-end networking that businesses install.  This
occurs for three reasons.  Many schools work with a local networking
company.  Such companies have sprung up in every state and region. 
These companies spend their days networking businesses and have a good
idea of what connectivity seems to satisfy their business customers.  So
the first reason schools put in a certain level of networking comes from
the familiarity of these networking companies with "typical
installations."  (Some of these companies specialize in school networks
and have a better sensitivity to school needs.)  The second reason
schools tend to put in a certain level of networking is a result of the
current costs of equipment.  A 10 Mbps Ethernet card costs around
$30-$40.  Anything else costs more.  A 100 Mbps card costs a lot more. 
The same is true for the other components -- cabling, hubs, routers,
servers, etc.  The third reason is a little more idiosyncratic.  Video
teleconferencing via CuSeeMe (currently a slowly changing black and
white picture that occupies about 1/8 of the screen) is seen as a
benchmark for connectivity.  If you can't establish one such session,
you can't "show off" the network.  Anything more doesn't seem important,
yet.

Often there is already an existing MIS (Management Information System)
network in the district office of a school district.  This is most
likely a Novell network running an Ethernet LAN (Local Area Network). 
The purpose of this network is administrative.  It is actually the
"business end of the school."  In many schools there are labs running
ILS's (Instructional Learning Systems) and these use their own
proprietary networks (Sherry, EPIE Institute, 1990).  There are also
LAN's which permit schools to provide student computers which run
networked software.  And there are often small networks in libraries for
the purposes of sharing CD-ROM resources.

In our review of K-12 networking we found very little information about
how schools are combining and integrating these existing networks.  We
conducted a survey of members of the Consortium for School Networking,
selected networking companies and were able to find very little about
integration efforts.  Since many of the existing networks run Novell
Netware as their networking software we contacted a Novell SE (software
engineer) who discussed levels of integration.  

* Information passing along a digital network is comprised of data
packets.  A software program communicates to the network card via a
driver and/or stack.  The format of the packets is dependent on the
"protocol."  Novell has used IPX for a number of years.  The Internet
uses TCP/IP.  There are other protocols, e.g., DEC uses LAT.  There are
three popular hardwares associated with networks: Ethernet, token-ring,
and arcnet.  Ethernet is the most popular.  Token-ring is by IBM and
schools that want to standardize to IBM have stayed with token-ring. 
There are strengths and weaknesses of the three "architectures."  Apple
supports both Ethernet and localTalk.  LocalTalk (also called AppleTalk)
is popular because Macintoshes have been "network ready" for over ten
years by supporting the LocalTalk protocol which requires no extra
hardware to interconnect Macintoshes.  Many of the first school networks
were achieved using LocalTalk for the price of connectors.  LocalTalk,
however, is limited to 230.4 Kbps which is about 40 times slower than an
Ethernet at 10 Mbps.  (LocalTalk is somewhat slower than today's fastest
modems at 28.8 Kbps.)

Schools may integrate the different physical networks.  Both IPX and
TCP/IP packets can run simultaneously on the same Ethernet network. 
Bridges can be used to go between different network architectures, e.g.
a group of Macintoshes can be connected via LocalTalk to a bridge and
then onto an Ethernet.  For a school administrator to run, say Netscape
under Windows, there must be a second stack for TCP/IP in addition to
the IPX stack for Netware.  Alternatively the administrator can change
over from IPX to TCP/IP entirely using Netware IP.  For UNIX systems,
using Netware for NFS Services, both disk drives and printer
transparency can be achieved.  One third party vendor, FireFox, makes a
product that runs both IPX and TCP/IP as a single stack (saving memory
on the DOS machine).

More difficult to integrate are DOS (non-windows) based CD-ROM networks
in libraries.  McQueen (1990) reports on softwares that can be used to
provide telnet access to CD-ROM networks.  This permits students at
other telnet capable terminals to access these resources either from
within school or from home.  The number of simultaneous users in this
configuration, however, is very limited.  As libraries upgrade to
servers that can be placed on the Internet the technical problems of
remote access become easier.  Unfortunately there remain site license
problems since vendors of CD-ROM's do not wish the entire Internet to
access a disc licensed to one particular library.  Today there is no
effective way of determining the geographic location of someone on the
Internet.  While many domains can be matched with a geographical area,
domains such as aol.com and compuserve.com can be anywhere in the world.
 Access must then be confined using registration, ID's and passwords. 
An alternative is to require that a user have an account on a gateway
machine to the resource.  This is commonly employed by OCLC for access
to their FirstSearch databases.  Everyone who has an account on the
gateway machine (often a campus wide machine for students and faculty)
is site licensed to access FirstSearch.

Copyright and license problems will be an increasingly important factor,
and perhaps a major hurdle to affordable school networking (as discussed
above).

Networking using TCP/IP can be done quite frugally.  Since the protocol
has been the subject of research and development both inside and outside
of government and universities, there are many pieces of an Internet
network available for free or as shareware.

The choice of server is important.  UNIX servers are popular because
many of the first TCP/IP connections were among UNIX machines. 
Unfortunately many report that UNIX systems are pesky in terms of
administration.  The UNIX shell was originally developed with the
programmer in mind.  As a result UNIX never penetrated into businesses
and the same may be true for schools.  Nonetheless, there is an
excellent public domain UNIX operating system called Linux.  Linux is
available for both Intel based machines (the ones that often run DOS)
and for most other popular machines. (One consultant recommends the
Debian distribution of Linux over the Slackware version.  The same
consultant recommends Cyclades as a low-cost, multi-io card to run with
the system.)

Companies including BBN (1995) and Lloyd (1995) have entered the market
to provide Internet servers specially designed for schools.  Further,
Microsoft has announced they are working on a package of software tools
to facilitate school and home-school connections based on their NT
server (Microsoft, 1995).  The software tools will be donated to schools
to encourage them to adopt NT servers.  Microsoft also donates NT server
software to schools.

One of the features of the BBN server is a software component called
FrontDoorª which allows school staff to customize and administer the
server without having to access the server's underlying UNIX operating
system.  "This makes it easy for library and school staff to manage the
server Ñ regardless of their technical expertise." (BBN, 1995, Boston
Schools)

* Unlike other technologies such as photocopy machines, networks are
comprised of many separable pieces.  These pieces require a common
language and infrastructure to interconnect them, usually TCP/IP and
Ethernet, but the network can be comprised of many different machines
and architectures, even within a school district.  While the
standardization argument above encourages schools to adopt uniform
hardware and softwares across the district there are some good reasons
to break with this rule:

	1.  Administrative systems, such as MIS's will continue to have
network capabilities that cannot be attained via Internet tools.  For
example, client-server tools for accessing record databases may be only
available under certain operating systems and servers.

	2.  Library systems, especially catalogue systems, may be only
available under certain operating systems and servers.

	3.  For economic reasons parts of the network might be either:
	A.  Earlier DOS-based machines running DOS TCP/IP tools such as NCSA
telnet, University of Kansas DosLynx (web browser, Boutell, 1995),
University of Minnesota Minuet (graphical web browser, Boutell, 1995,
Gonsalez, 1995), or KA9Q, Phil Karn,'s TCP/IP DOS telnet program
(Horzepa, 1989, Karns, 1991, Snow, 1995)

	B.  Yet earlier machines that run public domain/shareware
telecommunications software to appear as VT-100's and be connected via
serial ports to a local terminal server (note: this is an excellent
solution to utilizing early Macintosh machines and Apple II's both
within schools and for connecting low-income, have-not, homes with
schools)  It provides these machines with text-based access to the
Internet and their local disk drives can be used to store e-mail and
other materials for students to work on with word processing tools that
run on the particular machine.  In particular, we have identified the
Coker Terminal Server software which can connect these machines to an
OS/2 server (OS/2 is an excellent multi-tasking operating system that
can run on a low-cost 386 or better PC).  The OS/2 server then connects
the terminals to a Linux Server with Internet access.  The terminal
server software is priced (including educational discount) as follows:

		Number of Lines	Price (U.S. dollars)
		4	$60
		8	$105
		16	$150
		32	$260

	Thus it would cost $8 per VT-100 machine to provide an Internet lab
of 32 machines.  Availability Russell John Coker,
rjc@snoopy.apana.org.au. (per e-mail, Priest, 1/7/96)  A single copy of
OS/2 for the server, at an educational discount, is $72.  Linux is
shareware.

4.  Inclusion of a Community BBS system to enhance home-school
connectivity.  To date the features of a BBS software such as Major One
by Galacticomm provides a better environment for community activities. 
These BBS's have been refined over a decade to provide features
including forums and local chat.  Nonetheless features of the Internet
are able to capture some of the flavor of a BBS by using local
newsgroups (and a news reader), IRC (chat), and interactive web pages
(forums).  Even major online services such as AOL (Internet World, 1995)
are conducting surveys to understand how they can provide a more "local
presence" to their users.  One major community BBS system, the National
Public Telecomputing Network (NPTN, 1994) Free-Nets, has been described
as "fraying" (Toronto Star, 1995).  With 50,000 active users the
Cleveland Free-Net's 75 public phone lines are "jammed day and night."
"Those who do get through often just use it as a local-call access to
Free-Nets in Buffalo or Tallahassee that are easier to use and have more
features.  And many have abandoned it in favor of the growing commercial
on-line Internet services. (Toronto Star, 1995)"

Fundamentally we are witnessing a struggle between maintaining a sense
of local community and the lure of "virtual communities" on the
Internet.  The LINCT Coalition (Komoski, 1995, LINCT) encourages local
community building efforts around a local BBS.  The LINCT program
includes volunteerism and encourages low-income families to earn donated
computers by learning to use them and by volunteering to help others in
the community.  An adaptation of the Time Dollar system (Burdick, 1994
and Cahn, 1994) called Community Service Credits is made an integral
part of the BBS.  For one hour of volunteer work anyone can earn 1 CSC
Credit.  LINCT is currently studying how its goals can be achieved using
Internet based tools to simulate the look and feel of a local BBS.

As discussed further in Section Five, the features of BBS's are becoming
more attainable using Internet tools.  We anticipate that within the
next couple of years the sense and feel of a BBS will be achieved using
Internet machines, causing many BBS's to become variant Internet
servers.  We are witnessing this transformation with Galacticomm's Major
One BBS.  Thus our advice is for schools and communities to work with
and link to existing BBS's.  For new efforts we suggest working with
Internet tools as described in Section Five.

5.  High School or community-center Computer Shop activities that permit
students to build "toaster nets" out of most any machines the shop can
obtain.  (Barry Kort describes the building of toaster nets Ñ a
collection of surplus machines tied together using public domain
software Ñ at the end of this Section.)  Recalling high school car shops
and A/V (audio-visual) clubs in schools, some schools are offering a
place where students can learn to take computers apart and put them back
together as part of a local network.  With the availability of LINUX,
the computer shop can take many common computers from IBM PC's to DEC
workstations and put environments on them such as MOO (Multidimensional
Object Orientations).  A MOO permits the creation of a virtual place
which can include features of a game, such as Dungeons and Dragons, or
of a learning environment with virtual classrooms, virtual slide
projectors, and places for students to assemble with online teachers. 
We suggest that one potential contribution of the U.S. Tech Corps could
be to help establish and advise such Computer Shops in schools and
community centers.

Authentic learning with computers involves the hands-on experience that
a computer shop provides a student.  It is this kind of experience that
will produce the next generation of engineers that this country needs if
it is to compete the global marketplace.

Working with donated computers and public domain softwares, students can
learn to "hack."  While the word hacker has gained some negative
connotation through association with unauthorized computer use, in the
computer world, being called a "computer hacker" is a compliment.  The
computer hacker goes beneath the surface of the computer and gets to
know it at its various levels of software interfaces.  The imagination
of a hacker should be recognized as a tribute to the inquiring mind that
is able to breach supposedly impenetrable software walls.  (School
administrators will be pleased to know that they can erect "firewalls"
to sensitive records, preventing breaches.  More anxious administrators
can insist on keeping administrative record systems (and their networks)
physically disconnected from other school networks and the Internet.)

Networking Models

Several studies have used "networking models" to help schools and
policy-makers make decisions about different "levels" of networking. 
The seminal work was by Newman (1992, Models) at BBN.  In his report
Newman sketches out the various ways in which schools were providing
connectivity including phone lines, local area networks, and various
forms of Internet connectivity.  While 1992 was not long ago, the ways
schools can achieve Internet connectivity have changed dramatically.  In
the late 80's and early 90's it was common for schools to connect
through a university.  But in the last couple of years various Internet
providers have offered affordable connections to both institutions and
homes.

In 1994, Warner (1994) and others developed a most useful extension of
Newman's work addressing the school decisionmaker.  One section shows an
overview of connection models features and benefits, and other sections
show how to extend existing networks to enhanced networks.

Also in 1994, Rothstein (1995) took Newman's models and extracted five
"cost models of K-12 Networking."  These were:

- Single PC in each schools, with modems (14.4), dial-up model, district
office with file server and 56 Kbps line to the Internet

- School with LAN and shared modem (14.4) to district office with file
server, LAN, Router, and 56 Kbps line to the Internet

- School with LAN, Router, and 56 Kbps line to the district office, and
district office with file server, LAN, router, and 56 Kbps line to the
Internet

-  School with LAN, file server, router and 56 Kbps line to the district
office, and district office with file server, LAN, router and 1.5 Mbps
(T-1) connection to the Internet

-  School with LAN, file server, router and 1.5 Mbps (T-1) connection to
the district office, and district office with file server, LAN, router,
and 1.5 Mbps (T-1) connection to the Internet

Rothstein computed a low and high cost of establishing and maintaining
each model for the all public schools expressed as a total and as a cost
per student.  For the last model the total U.S. one-time cost ranged
from $51 billion to $113 billion, and the annual costs ranged from $4
billion to $10 billion.

In 1995, McKinsey used an alternative "model of infrastructure
deployment."  They describe 4 models:

- Single room school Lab with 25 computers, Ethernet LAN in the lab and
10 telephone lines

- The above plus a computer and modem per teacher

- Partial Classroom model where half the classrooms have 1 computer per
5 students, Ethernet LAN across and within all classrooms, and a T-1
connection (1.5 Mbps)

- Classroom model with all above plus in all classrooms 1 computer per 5
students

The total U.S. cost for the last model is $47 billion with an annual
maintenance/operating cost of $14 billion.

McKinsey also presents the costs as a percentage of K-12 spending in a
year.  This is expressed as a percentage of "Public K-12 Spending in the
Final Year."  They take the deployment period (5 years for the first 3
models, and 10 years for the last), annualize the capital expenditures,
and add the operating/maintenance expense in the last year.  Since the
operating expenses are highest in the last year of deployment, this
figure represents a peak annual budget load.  For the 4 models, this
percentage is 1.5, 3.0, 3.4, and 3.9 respectively.  Also, the average
total cost of the fourth model, per school, is shown as $555,000 with an
average annual cost of $165,000, and the average total cost of the
fourth model, per student, is shown as $965 with an average annual cost
of $275.


Missing from these Models

Implicit in these models is a series of machines that are connected to
the Internet.  What these models do not show is any relationship to
other forms of teaching resources be they video tape, educational
software program, or local access to DVD CD-ROM material.

There is the sense that whatever is provided via the network will come
via telecommunications and from the outside.

We question this assumption for the following reasons:

- One of the "benefits" expressed about the Internet is that it is
always changing Ñ there is always new information.  We question whether
teachers want their curriculum materials to be "always changing" on
them.  While exploring the Internet can be a facet of education,
teachers and learners will also require resources that are more like
books Ñ stable, organized, and reliably available.

- As mentioned above, we are on verge of a dramatic increase in the
local storage of digital materials Ñ the 9/18 gigabyte DVD CD-ROM.  Ten
of these disks (at an estimated production cost of  $1.00 each) will
contain most of the materials residing on the Internet that any teacher
may wish to use.  The players for these disks are priced around $500 in
the first year of production.  A $700 carousel player containing 100
such disks would provide access to 1,800 gigabytes of material.  At 30K
bytes per jpeg image, this could contain 1,600,000,000 images.  If a
teacher decided to selectively view 1/10th of all these images for 1
minute each the task would take 270,000 hours looking at images or or
almost 200 years based on a standard school day.

There would be enough space for text material to keep students reading
for centuries.  Only as we might wish to provide full motion video does
this archive start to look small.  But even for all the video resources
one school might require this mechanism may be the most cost-effective
over the next decade.

- Steven Hodas and Gary Warren of NASA have demonstrated (Hodas &
Warren, 1995) their HorizonNet system for lower-cost school and
community networking in which they place a large caching disk on the
Internet server.  In the Yorktown Elementary School Demonstration the
cache fulfilled 92% of the information requests by students without
reaccessing the Internet.  This greatly reduced the need for high-speed
Internet access.

- If we combine the CD-ROM technology above with the NASA caching system
the resulting load on the Internet by all U.S. schools and libraries
would be almost nil.  A DVD archive could be established to respond to
the same URL codes constructed as Universal Resource Locators.  Thus,
instead of using megabytes of telecommunications resources to access
NASA photographs, the photographs would be pulled from the DVD archive. 
Already there is software that scans popular sites and alerts users to
updated materials.  This same software would ensure that any recent
additions to a site would be recognized and provided via the Internet. 
(Already very popular sites on the Internet must be "mirrored" because
of congestion.  DVD archives would simply extend the concept of
"mirroring" to the district level.)

EPIE Institute/CITS Networking Model of Resource Access

In developing the following resource access model we draw from the
insights of Larry Cuban (1995) and Margaret Riel (1995) presented at the
Office of Technology Assessment Technology Futures Workshop (Fulton,
Futures, 1995).

Cuban observed that elementary school teaching is much different than
high school teaching.  He notes that in many ways high school teaching
is modelled on college teaching.  He further comments that people aren't
running about asking why there are not computers in college classrooms!

The common pedagogy of high school, as of college, revolves about an
ongoing exchange of knowledge, ideas and questions between teacher and
students.  The technologies that fit that culture are those that provide
the teacher with the ability to illustrate, present, and explain
content.

In contrast, elementary school activities tend to be more fluid and
students often break into groups for various activities.  Cuban, thus,
suggests that the exploratory aspects of network computing will be
assimilated in elementary schools and rejected in high schools. 
Depending on teaching styles of individual teachers, middle schools will
tend one way or the other.

Consequently we see educational technology dividing into two broad
categories:

- High-end presentation machines that can draw on networked educational
software, CD-ROM, and the Internet in conducting teacher/class
interactive presentations.  The computer would have both high capacity
CD-ROM player, access to a local area network and school CD-ROM/Video
archive, access to the Internet, and high-quality projection
capabilities (Color LCD projectors or large high definition monitors).

- Exploration machines.  Ideally these should be high-end PC's, but,
depending on a schools budget, these could be donated older machines
with text web browsers, or the forthcoming HollowPC (new, low cost
machines primarily intended for looking at materials on networks). 
These machines would have wordprocessing capabilities, etc., but no disk
drives Ñ all materials would be stored more economically and safely on
the server.  ("Oracle plans to ship its first Internet PC in March:
Oracle Corp. is putting the finishing touches on its first Internet PC,
which it says will ship in March, and is working with Acorn Computer
Group on a second version.  The new device will meet the $500 price
goal, but will come without a monitor.  Instead, consumers will use
cable to hook it up to their television sets or PCs.  The manufacturing
cost, according to Oracle's VP of network computing, is under $200 Ñ
$100 for four megabytes of RAM, $30 for an ARM 7500 microprocessor, and
the rest for the keyboard, mouse and network connections (Wall Street
Journal, Jan. 11, 1996, p. B2).)

In most school districts (given Cuban's analysis) we would expect to
find the high-end, presentation machines most frequently opted for by
teachers in high school with exploration machines either outside the
classroom Ñ in convenient areas where students could access them and/or
at home where they have the time to make use of them.

We would also find the exploration machines in the classrooms of
elementary schools where students would be engaged in multiple
activities using them, either alone or in groups.  The teacher would
move about more as a coach and guide as in the scenarios described by
Margaret Riel (1995).

Further, we expect that facilities such as SmartSystem (Dukane, 1995)
will flourish in proportion to the relative costs of providing
resources, physically, in the classroom, via media server within the
school or school district, and via the Internet.  SmartSystem involves a
high quality monitor in the classroom with a remote-control; the
materials being called upon are located at a central room and the
facilities include VCR tape machines, CD-ROM's, and other media players.
 In this era, the video material is still provided via coaxial cable
using analog CATV.  In the next decade we see those installations giving
way to digital high-definition monitors and video digital data
transport.

* While most of college/university teaching occurs as Cuban describes,
there are various "distance learning" approaches that have succeeded. 
Some are conducted "asynchronously" using various computer
teleconferencing softwares.  Others use video links with audio return
links.  Another successful approach has been by Athena University
(described above) where a "virtual classroom" is created using the MOO
environment tool.  MOO enhancements can provide the user with a
"graphical appearing" classroom environment where a class room appears
on the screen and the members of the class "appear" on the computer
screen.  A related use of computers in a "teacher centric" classroom is
a system by Discourse Technologies (Discourse, 1995).  Students are
given small LCD display keyboards at their desks and the teacher has a
computer monitor that shows what each student is typing on a single line
of the monitor Ñ one line per student.  The teacher can glance across
the rows and know how active each student is and what they are "saying."
[A video tape is available called "Teachers See the Difference Not the
Technology."]

What is the relationship of these systems to both high school and
college/university teaching?  Will "virtual universities" put
traditional universities out of business?  The answer is not clear.  We
do know that many colleges and universities are scrambling to improve
their "distance learning" capabilities.]

In Addition

We also see an increasing use of telephone networks both within schools
and between schools and homes.  In our review of practical networks Ñ
actually in use Ñ we found many schools opting for telephone and
voice-mail systems.  In a U.S. Department of Education funded study,
Priest (1973) found that lower income parents in Albuquerque, NM had as
their highest goals for their high school children, better communication
among teachers, students and themselves.  In the Benefits Section (IV)
of this report we examine three dimensions of the effectiveness of
networks.  While voice-telephone does not provide a conduit for learning
resources, it rates highly along an inspirational (affective) dimension
and a discipline/monitoring dimension (see, further, Section Four).

With voice-mail, teachers are better able to receive and send messages
both within the school and outside.  The literature is filled with
disparaging remarks about the lack of telephones in classrooms,
describing teachers waiting in line for pay telephones with students.

The potential trade offs and complementarities between voice-mail and
e-mail systems is still unknown (an e-mail network provides better
access to learning resources).  While some parents have regular access
to e-mail through their work or through online services, most do not.  A
recent Rand study (Anderson, 1995) recommends universal e-mail access. 
Telephones are a convenient device; requiring little time to access and
use.  In contrast, computers have start-up time, connection time, etc. 
While services such as ISDN promise connection times of less than a
second, the rates by the Baby Bells have been too high for all but the
most serious Internet user (e.g. USWest proposed flat rate ISDN
residential service of $184/month, tap-info@essential.org, 12/27/95, see
rates below).

Community Expanded Model

Figure A illustrates one of many possible community-expanded models. 
Within this model, the school networks are part of a larger set of
networks serving the community.  In this illustration, the base of the
community server is in a local library.  In other communities it could
be an extension of the school server or an extension of an online
service providing "local presence."

When we recognize that government services, including social services,
are intimately entwined with families and schools, the possibilities for
using a local community network to strengthen the community's overall
"learning ecology" is enormous.  Networking promises to bring parents,
teachers, children, and social services together in responsive and
responsible ways not before achievable (Niebuhr, 1994).  Cook (1995),
however, warns us of how the government services of Washington State
have taken on a darker side as services become intrusive rather than
helpful.  Clearly technology cannot be simply applied without an
underlying model of conduct that respects privacy, acts responsibly, and
actually increases the well-being of the community.

One promising approach promoted by the LINCT Coalition (described above)
rewards members of the community for helping others.  When a community
is comprised of active, involved, concerned citizens, the responsive,
responsible community model becomes far more achievable.



FIGURE A: GUIDING INFRASTRUCTURE MODEL

[text version based on graphic in original document]

INFRASTRUCTURE COMPONENT/USER             INTERCONNECTED WITH...

Library terminal                                Hub, and
                                        Local CD-ROM resources
Homes (Parents & children)                        Hub
Teachers (at home)                                Hub
Hub: Community Communication Server   Library terminals, homes
   and Bulleting Board                (parents, children, teachers)
                                       local CD-ROM resources, and
                                           telephone network
Local CD-ROM resources                 Local learning coordinator
                                      database, library terminals,
                                               Hub, LAN
Telephone network                              Hub, LAN
LAN (Local Area Network)             Telephone network, Internet
                                Local learning coordinator database,
                              classrooms, local CD-ROM resources,
                            State databases, State network resources
                               and Regional resources
Local learning coordinator database        LAN, LAB, ILS, Internet
                                   resources, Local CD-ROM resources
                            Stand-alone software, Regional resources
                          State network resources, State curriculum,
                                    and State datbases
LAB                              Local learning coordinator database
Classrooms                                     LAN
ILS                              Local learning coordinator database
Internet resources          LAN, local learning coordinator database
Stand-alone software             Local learning coordinator database
Regional resources          LAN, Local learning coordinator database
State network resources     LAN, Local learning coordinator database
State curriculum                 Local learning coordinator database
State Databases             LAN, Local learning coordinator database
   (e.g. NY State Gateway Jobs)

Note: An important distinction between data flows and information
flows is made.  Except as indicated below, all interconnections
indicate two-way data flows.  When people are designing networks, it
is the flow of data that is commonly thought about and planned for.
These flows are TELECOMMUNICATIONS FLOWS.

However, more important is the information flow about the 
relationships among learning resources and ways to permit users to
identify learning resources in relation to their learning needs.  It is
these flows that are the heard of our Challenge Grant proposal.
These flows are CURRICULUM-RESOURCE MAPPING FLOWS.

The latter, informational, flows are one way. The relationship between
the Local learning coordinator database and all the resources linked to
it is a one-way (FROM the other resources, TO the LLCD) relationship,
with the sole exception of the LAN (the relationship is present, but
there is ALSO a two-way data flow, as the LLCD is on the LAN).  All
other relationships in the chart, to reiterate, are two-way data
(telecom) flows.

Final note: 1) Web pages, or 2) Search enging for resource search and
alignment may be sited either at school admin. or on Community Resource
server.


[end Figure A.]



Do's and Don'ts of School Networking Ñ Tips

By this point it should be clear that the do's and don'ts relate to any
number of different models of networking.  Thus these guidelines imply
value judgements as well as network configurations.

Some of the tips found in the literature are conflicting.  For example,
some practitioners have found e-mail to pen pals to provide a way of
introducing students to the Internet.  Others disparage pen pal e-mails
as ineffectual and too unstructured to be helpful.  Actually, there are
elements of truth in both positions.

The tips are organized alphabetically by keyword.  The bibliography of
this report has nearly 4,000 keywords assigned to 670 references.  [A
future EPIE/CITS activity is to make these references available for web
browsing by keyword.]

The tips fall into 6 categories and are labeled: _ Project, ) Hardware,
Z Software, * Administration , C Environment, L Resources.  We provide
these categories as a legend in the footer of this part of the report. 
Project {_} related tips pertain to organizing and maintaining network
conversations or communications.  They are often "collaborative
projects."  Hardware {)} and software {Z} tips address the mechanics of
networking.  Tips regarding how to organize people or what people need
to be organized fall under Administration {*}.  Tips relating to the
atmosphere or ambience are labeled Environment{C}.  Finally, connections
to learning resources are labeled Resources{L}. 

_  (Do have an) Active Curator (Bull, 1994, p. 241)

In the area of computer teleconferencing and forums, active
participation depends on a facilitator or "curator."

"Research indicates that is not simply the organizational structure of
the menu but the presence of an active curator that results in active
use of instructional resources."  "The curator can identify and bring
resources to the forum and provide suggestions and ideas for ways of
integrating these resources into the ongoing curriculum."

Lesson: In constructing any forum-like discussion for learning, be
prepared to have an alert, talented individual massage content, prompt
members (of the forum) for responses, and ensure that the discussion
does not ramble.

)Z  (Use) Adequate State-of-the-Art Hardware and Software (Addessio,
1994)

Many uses of the computer are more user-friendly as hardware has
advanced and software has taken advantage of those advances, such as in
graphical interfaces.

We provide two caveats.  This statement came from a project where the
Los Alamos National Laboratory put technology into the Los Alamos Middle
School.  In such projects a school may become the recipient of a level
of technology that it would otherwise not be able to afford, thus
providing an unrealistic, non-scaleable situation.  Second, we note the
advances, described above, in making older machines into exploration
machines. (The exploration machines may be used to explore the CD-ROM
carousel, the cached Internet materials, or the Internet directly, as
described above.)

Lesson:  It may seem that there is an endless progression of computer
improvements and software advances.  According to "Moore's Law," the
density of computer chips quadruples every three years.  This is
expected to continue until the year 2000 (Kozma, 1995, p. 20).  Meindl's
projections are less optimistic, with the density of chips growing at 20
to 35 per cent through the year 2111 (Kozoma, 1995, p. 20).

Computers even today are overpowered for most of their uses such as
e-mail or wordprocessing.  Only in areas such as complex graphics and
artificial intelligence are even today's machines inadequate.

If the price difference is relatively small, always buy the more
"powerful" machine.  Someone will eventually develop an application to
push it to its limit.  However, 32 bit machines with 8-12 megabytes of
memory will meet many needs for a good while.  Even in the past when
"price was no object" there was seldom built a machine past 64 bits.

- A 32 bit machine means that the arithmetic registers and, usually, the
bus (connections) to the processor are comprised of 32 parallel bits. 
The width of the registers determines the complexity of instructions
that can be performed and the width of the bus determines the speed at
which data bits can enter and leave the processor.  The IBM XT was was
an 8 bit processor with a 16 bit bus.  The Apple II was an 8 bit
processor and 8 bit bus.  Most machines today are 32 bits except DEC's
Alpha and some of the new game computers for kids (64 bits).

Lesson:  Buy the most powerful machines you can afford unless you are
certain their use will not require that power.  Remember, even the
earliest PC is more than adequate for word processing and as an
exploration machine. (For example, this section is being written on a 16
bit Macintosh built in 1984.  With Versaterm's Versatilities program
(priced at $40 for educational buyers) this machine is connected to the
Internet running telnet as the client.  With telnet any text-based
Internet tool from a server can be operated including a text web
browser, [Synergy Software, 1995].  In conversation with Paul Reese (a
pioneer with Macintoshes in the classroom, COSN List, 1/11/96) we
learned that he is using Mac Plus computers running System 6.5 which
will run Mac TCP providing the full range of Internet tools including
Eudora mail and Mosaic web browser (there are many Macs that cannot be
upgraded to System 7 out of memory limitations).  Reese uses a TribeStar
bridge to interconnect the LocalTalk network with an Ethernet network. 
He also uses LCIII on the LocalTalk part of the network which run
Netscape.  "The Tribestar was designed to break a localTalk network into
8 separate sections and therefore reduce packet collisions.  It works
very well doing that but has also worked nicely giving TCP to my older
computers.  It is a nice way to speed up a LocalTalk network without
having to invest in Ethertalk cards and wiring." (Mac TCP is not
shareware and many introductory kits provide a copy in the $20-$30
range.)

Comparable telnet capabilities can be achieved on even an 8 bit DOS
machine by running Columbia Kermit (shareware) which provides the
machine, again, with telnet capabilities using a $30 Ethernet card on
the machine [da Cruz, 1995].  And full graphical web browsing can be
achieved on the 8 bit DOS machine using Minuet (shareware) [Gonzales,
1995]).

_  (Be sensitive to) Adjust the Pace (Kimball, 1995)

This pertains to an electronic conference or forum (as above).

When students participate in a conference at different frequencies, they
can become lost if material appears to go by too fast.  "One way to even
out the 'rolling present' is to provide cues that let participants know
which items are hot and active.  You can put this information in an
information file or in the first conference message, or you can send out
periodic e-mail updates (p. 56)."

Lesson:  Asynchronous communication (the ability for people to access an
on-going discussion at different times) is useful but can lose members
who get out of "synch."  One activity of the facilitator is to help
stragglers rejoin the discussion.

*  (Gain) Administration Involvement (Lipman, 1994)

A lone, dedicated teacher can only do so much without a committed
administration and a source of funds.  School-wide systems must have the
full backing of the school administration (and the school board).

Lesson:  Each school and school district is comprised of a unique set of
actors.  While there may be consensus about the networking plan or
changes needed in a current plan, there it may be necessary to shuffle
staff or to hire someone from the outside to get things rolling.

C  (Provide) Ample Time (Peha, 1995)

As the National Education Commission on Time & Learning (Kane, 1994)
emphatically notes, schools run short on time.  Yet networking projects
require time for orientation and time to carry on.

Lesson:  Embark on a student networking project only if you can provide
ample time for it to succeed.

C  Be Organized (Walker, 1995)

"Make a flow chart so students can see how the parts relate to the whole
and so they can understand how their contributions fit in.  For example,
in our weather project, students took daily instrument readings.  Then,
using spreadsheets, they created forecasts to produce long range
studies...(p. 32)"

Lesson:  The Internet is an interesting place to explore but as the
teacher tries to put an Internet project in the context of the
curriculum, the relationship of the Internet activities to a workplan is
essential.

C  Be Realistic (Walker, 1995)

"In planning an effective Internet project, it's important to keep your
feet on the ground.  That is why we chose Grade 6 meteorology as the
basis for our project.  It offered the potential to do real hands-on
field scientific observation.  Yet, it did not require expensive
instruments. (p. 32)"

Lesson:  It is one thing to find data on the Internet and another to
incorporate it into a project that requires that students accomplish
tasks to learn something.

_   Brainstorm Your Project (Walker, 1995)

Lesson: In running an Internet-based project engage others in the school
such as the librarian and a vice principal to think through the project,
break it into manageable sub tasks, and provide a showcase (a "Weather
Wall" for the library) for the results (p. 32).

)  Build New Schools with Networking in Mind (Bryne, 1990)

Lesson:  While existing schools can be very difficult to wire or provide
electrical power for computers, there is no excuse for a new building to
lack adequate conduits, wall plates, and electrical outlets.  (And as
discussed above, run cable for telephone, CATV, category 5 wiring, and
fiber for the future).

)  Cable Plant Standardization (West Ottawa, 1991)

New buildings are not the only places to concentrate cable
standardization efforts.  "Standard raceways of cable trays in the halls
and conduits to each room must be installed during the construction to
facilitate installation, maintenance and evolution of the communications
cable plant in the schools."  "Cabling designed to carry voice, data and
video communications must be installed to every room.  Additionally,
four strands of fiber optic cable will be installed to each room for
future use.(p. 9)"

"Currently fiber optics provides a cost effective solution for
building-to-building communications and backbone installations in larger
facilities.  For this reason, the fiber to the room in each school won't
be used except in a few select cases but it will be available for the
future as the cost of the electronics continues to come down.  In the
larger facilities, such as the high school and the middle school, fiber
will be used in the backbone networks in those buildings. (p. 9)"

Lesson:  As shown above, the labor component of cabling is the dominant
cost.  Wire for the future.

L  (Use) CD-ROM's for Research (Guerette, 1994)

As suggested elsewhere, CD-ROM's as learning resources will be a vital
way for schools to provide students with tools for research.  In this
reference "New Brunswick Leads in Network Technology," the district has
implemented advanced uses of CD-ROM's.

Lesson:  Reduce reliance on telecommunications and improve stability of
resources by using CD-ROM based resources.

C  (Use) CD-ROM to Maintain Acceptable Use (EPIE/CITS)

The new DVD CD-ROM not only has the capacity to hold all school-relevant
text and image resources from the Internet but it also passes through an
editorial stage where content is carefully chosen.

Lesson:  Our first amendment for free speech is likely to make the
Internet contain more materials than individual parents or school
systems will want to provide access to.  While there are software
approaches to "blocking" specific information, these approaches are
difficult to implement.  The DVD will provide the ability to chose
content, provide stability in retrieval, and reduce telecommunication
costs.

)  (Use) Central Pool of Modems for Home Connections (California Guide,
1994)

The California Guide is one of the most comprehensive compendiums of
information for K-12 networking.

Lesson:  Maintaining dial-up connections is expensive and requires
effort.  The California Guide recommends placing dial-up connections for
home access in a central pool.  (As an alternative the school might
consider buying a block of accounts from an Internet Provider.)

C   (Recognize that) Change Takes Time (Carlitz, 1994)

Section IV describes how much of teacher training is, in actuality, a
process of changing modes of instruction.  From the perspective of
introducing networking, this process change takes time.

Lesson: Introducing networking has more to do with changing how teachers
teach, than it has to do with learning to use a technology.  It is for
this reason that studies find the dominant "costs" of introducing a
network relate to "professional development" rather than the costs of
the physical network.  Whether these should be called "costs" is
discussed in Section IV.

*  (Networking is actually) Changing Teacher Practice (Means, 1995)

One of the most comprehensive reports about the relationship between
technology and change is Technology's Role in Education Reform sponsored
the Office of Educational Research and Improvement (OERI).

"Placing technology in classrooms does not ensure that it will get used
appropriately, or even that it will get used at all.  Many of use have
visited classrooms with one or two computers in the back covered with a
plastic cover that is rarely removed.  The reformer's vision of 
project-center classrooms with students using technology tools makes
extensive demands on teachers.  Teachers are expected to orchestrate a
classroom in which students pursue different questions, work at
different speeds, use different materials, and work in flexible groups. 
Students will be working with original data sources, often pushing
beyond the limits of the teacher's knowledge, and learning to work
together to produce products that demonstrate what they have learned. 
All of this must be carefully planned and supported by a teacher in such
a way that the students take ownership of their projects and feel
responsible for their own learning, while at the same time ensuring that
the essential content in local, state, or national curriculum standards
in multiple areas are met and that students will perform well on
whatever high-stakes assessments are to be given.  There is no doubt
that the reform agenda calls for fundamental changes in teaching
practices on the part of most teachers.  In some ways, the introduction
of technology only adds another level of complication to what is already
a daunting task.  How does a school get all or almost all of its
teachers on board, particularly when many of those teachers have little
experience with technology? (p. 66)"

Lesson:  There are two basic ways people are proceeding, in accord with
the EPIE/CITS model described above.  If networking tools are simple
extensions of the traditional practice of teaching, little change is
asked of the teacher and the technology will be adopted as each teacher
sees its utility for traditional classroom teaching.  In contrast, if
the exploratory model is emphasized, significant changes in teacher
practices must occur.  This will require such a fundamental change in
the practice of teaching that chaos and uncertainty may fill many
teacher's lives to accommodate the change.

- The above reform report lists five precursors for "success" in
achieving technology-based education reform at a schoolwide perspective
(Means, 1995, pp. 165-166):

	-  Time devoted to developing a schoolwide vision.  A consensus
around instructional goals, and a shared philosophy concerning the kinds
of technology-supported activities that would support those goals.  Such
consensus takes time to achieve, and although it requires instructional
leadership, it also requires the active involvement of teachers. 
Site-based management and grant opportunities appeared to serve as
catalysts for such discussions.

	- Adequate technology access for all students.  To the extent that
there are only a few computers in regular classrooms or computers are
clustered in a few labs in one part of the school, most teachers have
little opportunity to integrate technology into their instruction and
indeed feel little responsibility for doing so.

	- Time for teachers to learn to use technology and to incorporate it
into their own curricular goals.  Particularly after the first initial
hurdles, learning to use a new piece of hardware or software in a
mechanical sense is a fairly short-term activity and can be accomplished
through the typical in-service session.  Thinking about how technology
can support one's own instructional goals, however, and learning how to
orchestrate a class in which students are doing challenging projects,
portions of which are technology based, take much longer.  These kinds
of learning need to occur over time, preferably with opportunities to
observe models, to practice, and to receive feedback on one's section.

	- Easily accessible technical support.  Most schools have a few
teachers who are comfortable with technology and able to do much of
their own troubleshooting.  But most teachers have limited experience in
this area, even if they are comfortable using a technology they have not
completely mastered in front of their students, these teachers will not
be willing to plan around technology use if there is a good chance they
will encounter technical problems that they cannot get fixed for days or
weeks.

	- Rewards and recognition for exemplary technology-supported
activities.  Like the rest of us, teachers are influenced by the reward
structure around them when it comes to deciding where to place their
energies.  Not surprisingly, school leadership that values technology
and education reform activities was associated with more widespread and
sustained emphasis in these areas.

The same report also associates "success" within schools to (pp.
166-168):

	- Good curricular content.  Although in some cases the availability
of new technology inspired a project (e.g., the production of multimedia
materials about local leaders), in all cases the most fully developed
projects had strong curriculum content and many 
components that were not technology based.

	- A structure within which teachers can innovate.  Many of the early
technology enthusiasts dreamed of a "teacher-proof" system embodying
sound principles of teaching and learning and engaging students directly
without the interference of a teacher whose knowledge base might be
incomplete or whose pedagogy might be faulty.  Studies of classroom
implementations of technology have demonstrated that this goal was not
only unrealistic but wrong-headed.  Teachers can subvert practically any
kind of instructional material to their own goals and ways of teaching. 
Thus, in newer conceptions, the teacher is an essential part of the
instructional application of technology.

	- Opportunity for teachers to collaborate with peers.  Just as the
difficulty of what we are asking teachers to do in moving to
project-centered instruction implies an advantage for building on
existing curriculum structure, it also suggests that the support
teachers can get from collaboration with their peers will be important.

	- Teachers and students already comfortable with project-based
learning.  Bringing technology into a classroom and implementing
student-centered projects is much easier if the teacher and students are
not trying to learn about a new technology at the same time they are
struggling with new roles and new structures for organizing classroom
activities.

	- Use of technology across subject matters and classrooms.  There is
a certain amount of "overhead" that goes with learning to use any new
technology.  Students need to acquire keyboarding skills and learn how
to get into programs and files and to store their work in appropriate
ways.  Passwords and Internet search skills require a certain amount of
knowledge that has nothing to do with most curricula and is unlikely to
carry directly over into adult settings for any but perhaps senior high
school students because of rapid changes in technology.  Given this
reality, the more classes and grades over which this "technology
overhead" can be spread, the better.  Teachers in schools that use
technology throughout the school find it easier to use technology
because they do not have to teach all the technology skills themselves. 
Moreover, when technology is used across a broad range of classes, many
more students find enjoyable uses and feel confident about their ability
to learn new technology applications. [EPIE/CITS comment: Many arguments
are made for introducing technology in schools to provide vocational
skills.  The author of this item clearly believes that technology is
changing at such a rate that only imminent graduates will benefit from
this exposure.]

CL  (Integrate a) Community BBS (Priest & Komoski, 1995)

As the expanded EPIE/CITS model suggests, schools are but one facet of
the resource and communication base for students and parents.

Across the country are many community networking projects.  Some of
these are funded by the Corporation for Public Broadcasting and their
CWEIS program; others are funded under the Department of Commerce's
TIIAP program; many others are self-funded (see, for example, discussion
of NPTN above).  A study of 80 community projects that included schools
in their list of activities (Priest, 1995, Community Networking Survey)
showed that very few were successfully integrating their activities with
public schools.  This is be partly because the community networking
movement has preceded the use of networking within and for schools.

- Fifteen of the eighty community networks responded to our community
survey.  We asked respondents to describe how schools were involved in
their projects, cost-effective steps to keep the costs down or make the
process work better, relationships to outside partners and grant
resources, the role of training, and what advice the respondent had "in
making school/home and school networking happen in the next five years."
(Many projects didn't even have e-mail, indicating that networks were
not yet in place in at many sites.)

Many of the respondents were receiving funding from sources such as
TIIAP but expected activities to continue beyond the funding duration. 
Charlotte's Web of North Carolina has been one of the more successful
community networks in involving the schools.  "The school system hired a
trainer with part of their NTIA grant money.  They will retain her with
school funds.  this trainer is running workshops for teachers, creating
training "masters" at each school site to manage the training needs for
each school.  The training has been done in a series of week-long
workshops the teachers and media specialists attended during the summer.
(for public training, we do twice a month HTML training and once a month
unix file manager training.  Each unix session is two hours long with
the class limited to 10.  Each HTML session is limited to 10 and lasts
an hour. (lines 295-302)"

For school involvement, "The school systems had developed its own World
Wide Web pages accessible through Charlotte's Web (see
http://www.charweb.org/project/cms).  These pages contain information
about the schools for the community at large and teachers: calendars,
background reports, a 'Parent Center,' lesson plans, access to the
curriculum Research Center's library catalog, etc. (lines 1632-1644)"

Advice from Charlotte's Web is "Be willing to start small and scale up. 
Whatever you do -- DO something.  Don't wait for final approval of a
technology plan, a budget or PTA support.  Just get started and build
from there.  Plan carefully but know you must be willing to change the
plans radically.  The technology is evolving rapidly; better equipment
is becoming cheaper all the time.  Be willing to address sensitive
content issues up front -- bring them up and get parental and
administrative support in the beginning. (lines 321-330)"  In summary,
"a community network is a distributed network - the work of building it
is shared by many institutions, agencies and individuals.  A special
breed of volunteers seem to be attracted to this type of project and are
willing to contribute their time and talents...'It takes a whole village
to raise a child' ... the school's can't do the job alone. (line 1721)"

In contrast, The Iowa State University of Science and Technology (also
TIIAP recipients) is developing information materials.  Schools in Iowa
are connected either by private providers (often Iowa Network Services,
a cooperative of local independent phone companies) or through the State
of Iowa's Communications Network (a state-developed fiber optic
communications backbone for education and public development).  "We are
providing an information resource to educators and public planners.  At
this stage we are not providing an educational program.  We have
recently received $50,000 to begin building community analysis modules
to be linked with the information resource, providing education on
demand to individuals and groups that want to engage in public analysis
activities on a local basis.  This development is just barely getting
underway. (lines 510-515)"

Efforts in Los Angeles are held back by access to the same systems. 
"The main deterrent to more active participation is the fact that
relatively few members of the school community have LA Free-Net accounts
and only a few classrooms have an Internet connection.  From home, only
those students and staff who have computers and modems can participate.
(Lines 760-765).  Within those confines, "each participant school has a
menu area designed by and administered by staff members of the school
itself.  Characteristically there are menu areas for the Administration,
for curriculum, for parents/community, and for students.  Each of these
menu areas contain at least one interactive discussion area for public
dialog as well as for 
announcements. (lines 753-758)"

At the Greater Columbus, Ohio, Free-Net there are several school related
projects.  1.)  "Each school system has an area where it posts public
information such as telephone numbers, activities, schedules, general
descriptions, newsletters, etc.  In a few districts, teachers post
things about their specific classes.  2.  Several teachers use the
Free-Net to sponsor special, electronic projects for their students, and
3.)  There are several joint curriculum based projects with a couple of
specific schools aimed at integrating technology into an
interdisciplinary curriculum. (lines 987-997)"  Their suggestion for the
next five years, "The schools should try to form consortia across
district and municipal lines to handle a central modem pool and
centralized services for students in a municipal system.  Otherwise,
everyone will need to make parallel investments in equipment and staff
that are very inefficient.  There are economies of scale and scope in
creating shared systems. (lines 1084-1088)"

At a library based effort in Lexington, VA, another TIIAP funded effort
is only beginning.  Their response is like a number of groups contacted
in that they have identified target audiences (in this case libraries,
schools and local government) but specific plans or activities have not
begun.  At a Green Valley High School effort in NV, teachers and
students use Internet connections and the school district's FirstClass
Internet program Ñ but dial-up access from homes is not available and
they hope to receive another grant for that (lines 1334-1338).  A
Greater Knoxville Community Network (KNET) is also only beginning.  "As
KNET develops, we envision Knox County Schools publishing up-to-date
information and announcements within the menu items for citizens to
easily access.  Some mentoring between teachers and University of
Tennessee staff is already occurring. (line 2005)"

Project COnNECT, Denver, CO Free-Net, had plans for 3 educational
components, "1.)  mathenet - an online tutorial for students to post
questions and receive replies from student/teacher volunteers, 2.) 
electronic study groups - students throughout the state would engage in
online workgroups to discuss calculus, geometry, and 3.)  school health
- establish an online site of adolescent health info, with nursing
students responding to students' questions."  Yet "the first two of
these planned projects were never created due to lack of involvement of
the education partners.  The third, school health, is in the process of
being established. (lines 1482-1497)."  Instead, "a more general and
modest math and science online resource was established using the
gopher/Free-Port platform of Denver Free-Net and the www capabilities of
Boulder Community Network.  Contained within are curricular plans,
online projects, e-mail service, and numerous related usenet discussion
groups.  A local discussion group was also created for rural CO teachers
to share their experiences using technology in the classroom. (lines
1499-1504)"  Costs were reduced by using both the Denver Free-Net and
the ACLIN state library system for public access.  The overall project
is CPB supported and continuation is uncertain after the grant ends.

At a Santa Clara, CA project one school is involved (K5) and the
participants are networked by e-mail and through the Internet.  "School
administration, teachers, parents and students have formed a group
concerned with promoting computer literacy and access to Internet
information, particularly for those parents/students who otherwise would
have difficulty with online access.  The project was largely
parent-driven, with the active assistance of teachers who instruct in
computers.  The program is entirely extracurricular."  "An integral part
of the project is participation in Even Start, a federal grant family
literacy program that serves families at risk.  Donated equipment and
low-cost non-profit network fees put these families online so that they
can communicate by e-mail with their teachers, with school
administration, and with each other. (lines 1798-1889)"  For cost
savings the effort recommends employing VT-100 terminals with modems in
place of computers (donated or very low cost), use of a non-profit
community based Internet service provider, and use of volunteers as
trainers & shareware/freeware software as available.


_  (Use) Computer as a Communication Tool, Not as CAI (Teles, 1991)

In early application of computers the focus was on Computer Assisted
Instruction (CAI).  With networks, educators are learning that the
communication link can make up for the lack of intelligence of the
computer as a machine.

Lesson:  Computers are not very smart.  People are smarter.  Use
computers to get people to achieve learning by interacting with other
people.

)Z  (Use) Computer Shops (Priest, Interim Report, 1995)

Computer shops provide students with the opportunity to understand both
the hardware and software of computers and communications.

Lesson:  Inquiring students need the opportunity to go beyond just
running programs.  Some of these students will become future engineers
and scientists.

_)Z*CL  (Provide) Computers to Take Home (Means, 1995)

"According to estimates given by both the principal and the technology
coordinator, fewer than 1% of the students enrolled at Nathaniel
Elementary School have access to computers at home.  To enable their
students to compete with students from more affluent homes in future
education and work settings, Nathaniel staff felt that they needed to
try to provide these students with in-home as well as at-school
technology experience.  Nathaniel Elementary decided to purchase 78
Macintosh Classics for use in a special parent/student take-home
computer program.  The take-home program provides families with
computers on a 4-week loan basis.  Participation in the program is
self-selected. (p. 65)"  [EPIE/CITS comment:  While "buddy system"
loaning of computers has been popular the problem with these programs is
the child must return the machine.   In contrast LINCT (Komoski &
Priest, 1995) recommends providing children and their families with an
opportunity to earn a donated computer by learning to use it and earning
their training by doing volunteer work in the community.]

Lesson:  Home is where the time is.  Especially, low income families are
unable to provide their children with the same advantages that higher
income families are providing.  It is part of the mission of public
institutions, especially public schools, to help provide computers to
homes.

_)Z*CL  (Provide) Computers for Teachers to Take Home (Means, 1995)

"One strategy for getting teachers involved with technology that has
been used in many places is to give teachers computers for their
personal use. (p. 68)"

Lesson:  Home is where the time is.  Programs giving teachers computers
for home use give teachers a better idea of what can be done with the
equipment and get them accustomed to using the equipment as a tool for
their own productivity.

_  (Use) Conferencing that is Project Oriented (Levin, 1989)

"As an alternative initial activity to "computer pals", let me recommend
that you join one of the "project" oriented activities currently ongoing
on the network.  Read through the $IDEAS postings for the past month or
so and select the one that is most interesting.  Its much easier to
start out by joining in with an ongoing activity, and then once you've
gotten a feel for how these networks work, suggesting new ideas for
activities. (line 128)"

Lesson:  A task oriented discussion has better longevity and more appeal
than more idle networking chit-chat.


)Z  (Use) Consultant in Establishing a Network (Graham, 1995, Rothstein,
1995)

"Even a few hours with an experienced, professional consultant can be
significant in ensuring that the network is installed properly.
(Rothstein, p. 24)"

Lesson:  Expert advice can be accessed by working with Vendors and/or by
hiring a consultant.  For example, even something simple such as wiring
requires adherence to fire codes and the use of non-flammable cables
through fire walls.  As another example, each connection in a network is
a potential source of reflections.  Improperly made connections, of,
say, the incorrect impedance BNC connector can cause problems that even
sophisticated equipment finds difficult to locate.

)  (Consider) Converting School Intercom System into a Network (Hodas,
undated)

Many schools were wired for school intercoms.  While their cables were
never intended for data communications, they can be tested for their
potential to carry a local area network or to provide, at least,
telephone service.  While the speeds attainable may not be over 2-300
Kbps, this is adequate for Macintosh LocalTalk.  [Note: Also look for
existing telephone wires for the same purpose, see results of CoSN
survey below]

Lesson:  Many Internet and networking applications do not require very
high data rates.  If funds are very scarce, consider starting with
existing wiring, be it old telephone lines, CATV cables, or intercom
lines.  But be careful, investing in low-speed technology may be
wasteful.  Consider Macintoshes that can be used both for LocalTalk now
and Ethernet later.  Look into serial port LAN network software that
doesn't require more than a small license fee and serial cable boosters
that increase the range that data can travel on older lines.  (Perhaps
the thoughtful combination of existing wiring and a newer "backbone"
line would suffice.  The existing wiring reduces the labor going to the
many classrooms but the higher speed backbone provides the connection to
the server.)

*  (Consider) Coordinating Academic and Administration Networks
(Carlitz, 1994, section IV-8)

"Obviously there needs to be a level of interoperability between these
two systems so that teachers can provide administrative data and can
access this data as needed.  On the other hand there is a need for
system security in administrative systems which may go beyond that
required for academic systems. (line 921)"

Lesson:  Plan early on how an academic network will connect with not
only administrative networks, but library networks and community
networks as well.

)  (Use) Correctly Placed Servers (Gargano, 1994)

This advice comes from one of the Internet RFC's (Request for Comment)
documents.  RFC's are an entire catalogue of information about the
Internet. (The location of these is referenced in the bibliography) 
This document, K-12 Internetworking Guidelines, should be in any
networking administrator's top twenty Internet documents.

"Networking servers will be located where they can be managed and
supported, and also provide access paths with adequate bandwidth.  A
system of hierarchical servers should be created in larger school
districts, with automatic transfer of common information from a central
system to the secondary systems each night, or at appropriate intervals.
 Local servers will allow each school to provide on-line information
particular to its programs and community.  This model optimizes use of
network bandwidth as well. (p. 6)"

Lesson:  While an entire school district can run with one single, main
server, this may not make sense either in terms of distributing the
network load or in establishing more local servers that provide
information more relevant to the local users.


)  (Use) Cost-effective and Manageable School Interconnections (Gargano,
1994)

"School interconnect topologies (links) must be both cost effective and
manageable.  Communication between schools, district offices, county
offices of education, and the State Department of Education must be
reliable and of sufficient capacity to support the primary applications
as well as allow development of new applications.  Capacity is measured
both by total data traffic volume and by response time when information
is requested over the network.  Reliability is measured by the
percentage of time that the network is able to transport data. 
Reliability should be well over 99.7%.  Capacity should be such that no
more than 10% of the communications bandwidth is used during a typical
work day.  This is intended to leave adequate capacity for good response
time to short term communication demands. (p. 6)"

Lesson:  There are professionals whose entire job is to project network
load demand and design for capacity and reliability.  To the school
board that is contemplating "running some wires" we remind these
planners that an integrated network, especially for a school district of
any size, is a professional technical undertaking.

C  Create an Ambience (Kimball, 1995)

This is another tip in reference to conducting an online conference. 
"Even though your conference members or list may be part of a larger
network with its own culture, you can give your virtual group its own
flavor.  Think about how the first message or topic will set the tone;
about how you model message formatting; and about how you respond to
comments.  How can you help the group create a mental map of the
environment so that members develop appropriate expectations? (p. 55)"

Lesson:  Most conferences do not take care of themselves.  Facilitation
is important.

)  (Use) Custodians to Help Build Network (Graham in Priest, 1995,
Seaford, NY)

Many schools have wired and installed equipment using their own school
custodians.  These people know the buildings intimately, know existing
conduits and crawlspaces, and come at no added expense to the budget.

Lesson:  Think about spreading the wiring and installation process over
enough time that custodians can work it into their schedules.  With the
right spirit they can become an active part of the new enterprise,
giving them recognition they don't otherwise receive.  Make sure they
receive training.  This may simply involve their working with a
networking technician for one day to learn how to create a reliable
network.  Make sure they can call the technician back (or by phone) for
problems that may arise.

)Z  (Use) Dedicated, Qualified Staff for Networking (Addessio, 1994)

Network installations of some size require two categories of staffing. 
These staff categories are called the NIC (Network Information Center)
and NOC (Network Operations Center).  The NIC staff person(s) should be
able to answer questions ranging from "why does this e-mail bounce back
to me?" to "how do I put this scanned picture on the school web page?" 
The NOC staff person(s) is required to maintain the network and respond
to questions such as "I was looking up a student record and got a
message 'Internal Server fault, request requeued.'"

In a small school these two jobs may be assigned to one person, at
first, perhaps, to the technology coordinator and later to a trouble
shooter and an information specialist.

For a town of about 30,000 people, an administrator can plan on hiring
one NIC and one NOC type of person to handle networks throughout the
schools.  At about $35,000 salary and an overhead of 100%, budget $35K x
2 x 2, or $140,000 per year.

For the nation, this cost is about $1.3 billion to handle the technical
support for the nation's schools.

Lesson:  Plan ahead to keep a network running and to provide the
technical support teachers and students require.  There is no way to
avoid these costs, and provided well, the services will assure that the
investment in networks and teacher professional development bring the
full benefits of networking to the school.

_  (Be sure to) Define Roles in Networking Projects (Kimball, 1995)

"Once the purpose [of the project] is clear, you can begin to ask: Are
the participants peer learners?  Team members? Neighbors?  Is the
moderator expected to provide expert knowledge?  Support and
encouragement?  A guide to other resources? (p. 55)"

Lesson:  Facilitating online networking projects by providing clear
roles for participants will improve the results.

_  (Be sure to use) Designed Online Educational Activities (Teles, 1991)

"The design of the online environment is a key issue.  Online
instructors need to design the educational activities and learners need
to know in detail what they are expected to do online (Fredman, 1989;
Harasim, 1990; Riel, 1990; Rogers, 1989; Teles, 1991)."  "Timeline
considerations must be taken into account: short projects with
well-defined end-points have been successful (Fredman, 1989, p. 6)."

Lesson:  Structure is essential in assuring learning.  While "wandering
the Internet" is sometimes insightful, the educational use of the
Internet requires design and structure.

)  (Use) Disk Caching of Internet Materials (Hodas and Warren, 1995)

NASA plays an important role in aiding K-12 use of the Internet.  One of
the more useful Internet sites with many links to other resources is
http://quest.arc.nasa.gov/OER/.

As described above, one important way to reduce Internet bandwidth needs
is to "cache" recent Internet requests.  Caching as many recognize is a
technique used on today's PC's to speed disk access.  A portion of the
machines fast memory, often 2 to 4 megabytes, is reserved to hold recent
disk accesses.  This greatly enhances machine speed since it is
generally more likely that the most recently used disk materials will be
reaccessed.

The same principle is true with the Internet.  The NASA HorizonNet
project found that students' access of the Internet was repetitive at
any given time.  For example, if one student downloaded an image of
sun-spot activities, other students tended to download the same image. 
In one elementary school, 92% of all Internet requests could be handled
from the cache.

Lesson:  Request that your Internet server be Internet cache enabled. 
Don't assume that because Internet usage is often not charged per item,
that conserving bandwidth doesn't reduce costs.  Indeed, when Internet
access is reduced by 90%, this means that an Internet bandwidth of 1/10
the size can be purchased, saving the difference between, say, a
dedicated T-1 line and more cost-effective Frame Relay connections. 
This can mean thousands of dollars per Internet connection per year. 
Also, as material is placed on the Internet on a per item basis (a trend
beginning with various news databases), a cached item will save a school
in access charges by many times.

)  (Install a) District Wide Telephone System (West Ottawa, 1991)

"To make communications as efficient, cost effective and easy to use a
single standardized phone system should be installed district wide. 
Each building will have a full function PBX (phone system.  Where
buildings are on common property a system may be shared by more than one
building ... "  "Using the fiber network, all phone systems will be
connected to a central system and share a common voice mail system at
that location.  All calls between schools will be across the fiber
network.  Long distance calling will be consolidated to gain discounts
based on volume access... (p. 11)"

- The strategic plan for the West Ottawa School District is one of the
most comprehensive and integrated plans reviewed.  The following
objectives show the careful thought behind the plan:

	- Maximize the non-print resources available to every classroom
	The recent proliferation of materials available in non-print format
Ñ- videotapes, videodiscs, educational broadcast programming, computer
software, electronic encyclopedias, etc. Ñ combined with the desire of
teachers and students to make use of these resources leads us to
investigate cost-effective means to bring them into every classroom

	- Put information management tools directly into the hands of
students and teachers
	If computers and related technologies are "down-the-hall" or in
another building or shared with many people, they are less likely to be
used.  Regular access is one of the first keys to successful use.
	- Encourage more active involvement in learning by students
	The most powerful feature of technology is its ability to put the
control of the pace, selection and use of information into the hands of
the student.
	- Enable students to participate in the process of producing
materials generated by technological tools
	Word processing, desktop publishing, video presentations,
interactive displays Ñ these represent common means of presenting
information in the world around us and our students are eager to try
them.  We need to provide alternatives to paper and pencil for student
presentations.
	- Increase the frequency of use of the tools of technology
	An overriding consideration is that the technology implemented must
be easy to use.  A second consideration is to find the right tool for
the right task.   Just because something can be done on a computer, for
example, does not mean it is time-effective or cost-effective to do so.
	- Ensure equity of access for all grade levels and all service areas
to the resources and tools of technology
	Although it might be very difficult to deliver the many technologies
suggested in this proposal, simultaneously, the distribution of
equipment and services should quickly become as equal as possible Ñ
across grade levels and geographically throughout the District.
[EPIE/CITS Note: From the standpoint of equity we encourage equal
access, however, as the EPIE/CITS model suggests, utilization of
networked technology is likely to be higher in the lower grades.]
	- Improved teacher-student-parent communication
	Through improved phone and data services, it becomes possible to
increase the quantity and quality of communication.  This is especially
true when face-to-face meetings are difficult: the ability to inquire
about homework assignments in the evening, for example.
	- Improve internal staff communication
	The use of voice mail, electronic mail, and electronic bulletin
boards give the educational professionals options which can reduce
annoying interruptions during class time, reduce traditional mail, and
eliminate "phone tag."
	- Increase staff efficiency
	Time that is devoted to any teacher task that can be done more
quickly or easily through the use of technology, can be reallocated to
students and instruction.  Reduction in time spent reporting attendance,
preparing lesson plans, record-keeping, and preparing report cards will
all be possible through the use of appropriate 
technology.
	- Improve staff support in the areas of training, curriculum
decisions, management of hardware and software, and troubleshooting and
maintenance.
	If efforts to support staff with training or outside help does not
accompany the installation of new technologies, then the expected
results for improved  instruction, improved communication, or work
efficiency will not occur.  The kind of training necessary varies with
the complexity of the application.  A flexible, but carefully monitored,
training program may best suit teachers' needs.  In many cases, expert
help is required.  This expert help may need to be on staff if the
situation warrants.  (West Ottawa, 1991, p 5-8)

Lesson:   As other parts of this report discusses, the use of telephone
may actually be more important for many tasks associated with schooling,
than digital networks.  Schools should carefully prioritize their
communications needs and ensure that telephone systems including
voice-mail and many other possible features are considered.

)  (Install a) District Wide Video Network Using CATV (West Ottawa,
1991)

"To maximize the district's ability to distribute video information
sources within the schools and across the district, a district wide
video network should be built.  Within each school a distribution system
should be installed to be able to broadcast video information from at
least six sources located in the media center.  At this time the network
should be based on Cable TV (CATV) technology to each room.  Each room
should contain a TV that supports the building network connection and an
in-room video source connection.  To start with, the sources in the
school should be video tape, laser disc, C-TEC CATV, video bulletin
board and the district video network. (p. 10)"

The media resources described include laser-disc, VCR, and live
production.  In anticipation of the Toshiba/Sony/Philips high storage
CD-ROM we suggest that all schools plan to install a DVD CD-ROM 100 disk
carousel player.

Schools will also wish to consider systems such as Dukane's SmartSystem
(Dukane, 1995) which provides automated access to the media center
materials using easily used remote controls in the classroom.

Schools will wish to consider multi-purpose monitors that can serve as
both CATV monitors and computer monitors.  Monitors are becoming the
most expensive item in a computer and/or video display system.  Avoid
purchasing two of these per classroom, when one will suffice.  And
remember, as teachers increasingly use such a monitor for teaching,
there are poor economies in purchasing too cheap a monitor.  The monitor
should be bright enough to be easily viewed throughout a class even on a
bright day.

Lesson:  As discussed elsewhere, the economics of the location of media
players is still uncertain.  The Internet should be seen as yet one more
media source with its attendant costs and uncertainties.  There is a
growing trend to avoid moving media players around in schools and to
concentrate them in media centers, distributing the content via CATV (or
via either CATV or digital network for DVD's)

)  (using) Donated Equipment (Snow, 1995; Komoski & Priest, 1995)

Perhaps no subject raises more debate among educational technology
experts than the use of older PC equipment.  Many in schools have seen
donated equipment sit in storage for lack of appropriate software.  And
many newer pieces of educational software require machines manufactured
in the last few years due to memory requirements, monitor requirements,
and operating system requirements.

Nonetheless, the use of "dumber machines" appears to have come full
circle.  As more content and facility is available via the network, the
computer can be less powerful, again.

Also, these machines should be viewed callously.  Most schools will not
have the requisite talent to make a non-operable one work or to fix one
that breaks.  So treat them as dispensable.  When one breaks, replace it
with another.  Give the broken one to one of a dozen organizations
across the country that can swap parts and fix it (see, e.g., German,
1995).  Alternatively, if you have a successful computer shop, give it
to them.  It makes an ideal puzzle for the inquisitive student.  There
is "nothing better than natural puzzles."

Lesson:  Many earlier machines can be made into both text and graphical
web browsers with software described above.  The VT-100 (addressable
cursor terminal) has become a defacto standard for the computer world
and every computer ever manufactured, including the earlier CP/M
machines are capable of emulating a VT-100.  (The Santa Clara, CA case
study reviewed above recommended obtaining actual VT-100 terminals for
community networking.  These terminals are so populous that they are
given away at most amateur radio "swap-fests.")

While it may take the glitz of Netscape running on a $2000 Windows-based
machine to attract a student to web pages, many learn to shut off images
as a way of getting to the textual information, quicker.  When
scientific photographs are the center of a web-based project, a
graphical browser is absolutely necessary.  However, if the decision is
between providing a student with a text-based computer for home use, or
no computer at all Ñ opt for the text-based computer.  It makes a fine
exploration machine and, because it won't run Myst and other diverting
games, it may actually propel the student to more exploratory learning
than a high-end machine.

- Earlier machines need not be viewed as only VT-100 emulators. 
Geoworks, a graphical based Windows look alike, runs on the original IBM
XT.  LINCT provides Geoworks with shareware software for word
processing, spreadsheets, and telecommunications to low income
households who earn their donated computer by learning to use it.  These
tools are excellent starting points for families that would otherwise
have no tools from which to learn.

) (Use) Drawings and Specification (Lipman, 1994)

"Once ... decisions have been made, involve a network designer and have
a diagram drawn.  It is best to have a formal diagram with
specifications done before you put out your bids. (p. 10)"

Lesson:  The more clear and specific a specification, the lower bidders
will bid.  A hazy specification causes bidders to include costs for
unknowns.

)Z  (Take advantage of) Education Discounts (Crawford, 1992; from
Edupage in Chronicle of Higher Education, Feb. 9, 1996, p. A21)

Many vendors, especially software vendors, provide exceptional low
educational pricing.  Some vendors do this because they realize that
schools find it more difficult to justify costs; others do it because
they wish to introduce prospective customers to their products while
they are students.

"Librarians, Unite -- for Buying Power: College librarians are banding
together to purchase electronic resources for their institutions. 
'We've found that this sort of group purchasing power has really enabled
us to leverage the dollars that we have and to get resources we couldn't
have otherwise,' says an associate librarian at the University of Texas
at Austin, which buys publications through the TexShare consortium.  The
group buying arrangement is also advantageous for publishers, who don't
need to spend as much on marketing: 'As a group, we'll pay the vendor
more money than they can realistically get by slogging it out school by
school.' says the executive director of OhioLINK, which includes more
than 40 colleges and universities.  The president of Britannica Online
figures that more than half of the 233 institutions using his product
are doing so under consortium-based licenses. (Edupage)"

Lesson:  Always ask what the educational discount is for any products or
services you procure.  Form or join buying cooperatives in your state. 
Team up with libraries.  The OhioLINK,see discussion below under (Use)
Online Databases, exemplifies how bulk or consortium buying can reduce
costs.

_  (Encourage) E-mail (Caruso, 1995; Anderson, 1995, RAND)

"Based on Rand's study of on-line communities (like Latino Net in San
Francisco and the Blacksburg Electronic Village in Virginia), and what
it discovered about how citizens use E-mail today, the report concluded
that 'use of electronic mail is valuable for individuals, for
communities, for the practice and spread of democracy, and for the
general development of a viable information infrastructure.' (Caruso, p.
D5)"

Students, teachers, and parents can use e-mail to tighten the loop in
communications (see Section IV).  E-mail also fosters improved writing,
especially in peer-to-peer environments.

Lesson:  While the use of "electronic pen pals" is not high on most
networking educator's lists of learning tools, e-mail is useful in
communicating amongst teachers, students, and parents.

_  E-mail Encourages Writing (Kuttner, 1995)

In The Revival of a Lost Art, Kuttner describes how e-mail "In a fine
reversal, e-mail is the revenge of print on video.  It is producing, of
all things, a generation of writers."  Kuttner, a journalist, recounts
how kids increasingly stay in touch with each other through e-mail, and
how kids at college stay more in touch with their families.  "E-mail not
only lends itself to a lot of correspondence but also to quick response,
hence to whimsey, and also, remarkably enough, to candid revelation. (p.
19)"

Lesson:  In an era of the video, writing is making a comeback.

)  (Use) Ethernet (California Guide, 1994)

To many, Ethernet is synonomous with networking.  It has become the
Kleenex of the networking world.  While IBM can still tell you good
reasons to use token-ring, most schools aren't listening.

Lesson:  When a predominant technology form emerges, prices drop and
features increase.  Not only do schools prefer Ethernet, but they prefer
10BaseT to 10Base2 wiring because the hubs provide more informative
diagnostic information (even though the range of 10BaseT is only 300
feet compared to 10Base2 at 1200 feet.)

- While this report is not a guide to the technology, it is recognized
that certain terms need defining.  In a basic network a cable (for
10BaseT this is 2 sets of 2 twisted wires) connect from a card in the
computer Ñ an Ethernet card Ñ to a device at the other end using a plug
that looks like a telephone plug, only wider Ñ an RJ 45.  In an Ethernet
the wires fan out from points of concentration like a "star."  The wires
must go to a device that boosts the signal every 300 feet.  The device
that both permits multiple Ethernet wires to come together and boosts
the signal is a "hub."  The hub can be smart or dumb.  Smart hubs
provide diagnostic information when there are problems.  The Ethernet
signals are passed from hub to hub to a server.  The server has an
Ethernet card in it, just like each of the computers on the network.  If
the server passes signals onto other destinations, such as the Internet,
it is called a "router."  If the only function of the server is to pass
signals onto another network, it is only referred to as a router, not a
server.  If signals from a different kind of network need to be
connected to the Ethernet, the device at the union is called a "bridge."
 The speed of the Ethernet is commonly 10 Mbps (megabits per second). 
Connections to other buildings or to the Internet are done at different
standard speeds.  Common speeds are 56 Kbps (dedicated phone line), 128
Kbps (ISDN adjusted phone line), 1.5 Mbps (a T-1 line), and 44.7 Mbps (a
T-3 line).  Higher speeds are available through services l