Building Models of Connectivity: A 7 layer model

John Sechrest
Oregon State University
Computer Science Outreach Services
Hovland Hall, Rm 8
Corvallis Oregon, 97331
(503) 737-5562
May, 1995

The process of deploying networks and computers into organizations or communities has several layers that need to be addressed to maximize the effectiveness of the network deployment. This outlines 7 layers of issues surrounding the organizational process of connecting to the network. Without addressing everything from network access to organizational change, the deployment of a network may not contribute to the effectiveness of the organization.

As the Internet becomes more popular and continues growing at the current explosive rate, many different types of organizations are beginning to connect to the network. All parts of the community are starting to get connected. From the hospital, the local school, the small business down the street, all are starting to investigate Internet connectivity. However, at the meetings called to discuss connectivity, the typical discussion focuses around network access, hardware and software. Almost never does the discussion relate to the task of the organization and how the network will help the organization be more effective in what they do. The following 7 layer model of the barriers to effective use of computers and networks provides a framework for exploring the issues around a network connection.

Without addressing each of these levels, an organization will limit the opportunities for utilizing computers and networking well. Over the last decade, there has been a large investment in office automation. However, the measured productivity has not risen significantly. This is likely due to a focus on the technology without attention to the issues of content and organizational change.

Over the last few years, OSU CSOS has been working with different organizations around the state of Oregon in an effort to facilitate the deployment of Internet access to the state. In the process of working with different groups, it has become painfully clear how easy it is to deploy a very expensive doorstop. As someone has already said: "getting the network access is the easy part.". In these efforts, some organizations have made very effective use of the network resources that they had available, while others did not fare so well.

It was difficult to understand how different organizations with the same level of resources would make fundamentally different use of those resources. As a technical organization focused on solving networking problems, CSOS would typically respond to the question: "How do I get on the Internet?" with a technical response. This would include outlining the different types of equipment and connections that were possible. And while this seemed to address the question, it did not satisfy the question that was really being asked.

Often the question being asked was: "How can I improve my effectivness by getting on the Internet?" To answer this question requires connecting the process of network access to the end goals.

But there are many steps between the bits needed for network connectivity and the effectivness of the organization. Like the proverbial war that was lost for the want of a nail, the connection between the network and effectivness is a long chain of issues.

Let's look at a typical school asking to be connected to the network. They have read the Wallstreet Journal and now know that they "want" Internet. For this wish for Internet access to have a positive impact on the educational process, the school will need to address some basic questions at each level. By building a model at each layer, we can make it easier to answer some of the basic questions that need to be answered.


The network access question is a typical engineering trade-off. You must trade off cost, bandwidth and services. The chart above helps to outline the issues surrounding this trade off and in the process makes the decision making more effective.


Each organization has a cultural and religious bias towards particular hardware. The typical manifestation of this bias is illustrated by the ongoing PC vs Mac debate that many schools are having. The best approach to solving the hardware issues, is to evaluate the application software needs. From these needs, you can derive the hardware needs. Again this is a typical engineering trade-off. You must optimize cost against different features. However, at any given time, there are typically only a few configurations that make economic sense.


In a networked environment, the software issue is broken into three groups: server software, client software and application software. Each of these will effect the other. However, the real question to answer about the software is what problems are in need of being solved. These problems will outline the services that need to be available.

System Administration

Once the hardware and software is available, who will configure the system to work? Who will plan for the next round of network deployment? If these tasks are not coordinated, then it is very easy to complicate the support process by having a wide variety of versions of hardware and software. This can become very expensive.

User Support

Even when the equipment and network are in place and properly configured, you can not get very far if people don't know what to do with them. User support comes in many flavors, each meeting different needs. The Help Desk provides just-in-time problem solving. The professional development workshop connects the tool to the context in which the tool is effective. The skill training session provides the hands-on experience to make a strong basis for building a basic skill.


Without materials and data sets available for the topic area of interest, it can be hard to make effective use of the system. On the internet, some groups are more active creating materials than others. While there is a lot of computer related material, there is far less business materials available. Fortunately, the Internet is the process of communicating between people. This can be used as the basis for generating new materials in any topic area. But care will need to be taken to ensure that the materials are indexed and organized so that it is possible to use the information easily.

Organizational Change

It is possible to deploy a system and training people how to use it; create application oriented content , and still the use of the system will be ineffective. This can be due to an organizational issue that gets in the way of using the computers and networks.

When your organization deploys a new World Wide Web site, who will be responsible for maintenance of the information? If you are trying to do a hands-on science experiment, how do you fit it into a 45 minute class period? These types of questions illustrate some of the issues related to organizational assumptions. These assumptions can be the limiting factor on the deployment of new technology.

Without addessing the organizational and cultural issues explicitly, there can be some specific traps that can be painfull. An excellent example of a trap is the "Appropriate Use Policies". Many schools have jumped into network access before they have thought about AUP documents and network ettiquette training. Yet others have halted all network access due to the lack of policies. both of these processes illustrates the ways in which organizational change issues can cause limitations in the effectiveness of a network deployment.

One way of addressing these issues is to explicitly model the organizational change process. This can provide a means to understand the issues around organizational change. One resource for this is the Bootstrap Institute which looks at the question of explicitly modeling organizational activities and applying new technology to augment the organizational change process. By exploring organizational change and evolution explicitly, the process of adapting to the rapidly changing technology can be managed more effectively.


By modeling the issues in each of the 7 layers outlined above, we have been able to help organizations make more effective use of the computer and networking resources they were deploying. These different layers provide a framework for discussions of technological deployment.


John Sechrest is the Executive Director of the Oregon State University Computer Science Outreach Services, which is building models of connectivity as a means of understanding how computers and networks can help people be more effective. The CSOS program has many projects including Dial-up access, Information services and K-14 networking.

WWW References:

CSOS activities:
NERO activities:
Bootstrap Activities:
John Sechrest ( --
V1.0 -
Updated: 7/1/95