Skip to main content

IBM Cambridge Research Center

  Project: Carlisle Community Center

Researchers: John Patterson

A Collaborative User Experience Project:

Project Information: Click on a topic to go to the entry.

by John F. Patterson

The idea of the Carlisle Community Center was born at a meeting on Human-Computer Interaction (HCI) in March, 1998. Dr. Robert Putnam, a Harvard political science professor, gave a talk on the importance of local community groups for achieving effective governance. He argued that all the myriad casual contacts and small decisions required to make a club run build up the social capital that is needed to make bigger decisions. Dr. Putnam went further to document the decline of social capital within the United States over the last 15 to 20 years. In essence, there are fewer people involved in clubs and less time is spent on these social activities. The challenge posed to the assembled HCI professionals was to harness computer technology, in general, and the World Wide Web, in particular, in the service of building up social capital.

Following his talk, I chatted with Dr. Putnam and others about the idea of creating a community server for my town, Carlisle, MA. We discussed the attributes of an online service that would promote the formation of social capital. The goal was not one of substituting online for face-to-face interactions, but of fostering and encouraging them. By making it easier to keep track of a club's activities and concerns, members with busy work schedules might join in when possible rather than drop out from frustration at being unable to keep up. By presenting school schedules online rather than relying solely on student delivery, parents might be better able to keep track of school activities.

In our discussions, three principles emerged as potentially important. The first is to ensure that support is available for clubs, libraries, schools, and volunteer organizations in preference to political and governmental organizations. One of the attendees at Dr. Putnam's talk told about his city's web site for discussion of city issues. The online discussion was contentious and occasionally filled with invective. He viewed the web site as negative and unpleasant to visit. I believe that social capital is created in the low-risk conversations and decisions required to keep alive an avocation; to some degree, it is spent in political and governmental activities. This does not mean that political and governmental organizations should be excluded. They simply should not be the exclusive or even central offering.

The second principle for a web site fostering social capital is to eliminate anonymity and pseudonyms. Every action that I take on the web site should reinforce your knowledge about me as a neighbor and fellow towns person. Moreover, I should be willing to stand by whatever I say. Online discussions are simply another way of conversing. When we meet on the street, you should be able to say, "Oh that's right, you're the one who would prefer to meet Wednesdays." The conversation simply continues from one medium to another. When anonymity and pseudonyms are permitted, there are two bad effects. First, the continuing conversation is broken; no one can determine who said what. Second, the conversation can become less civilized, since there are no enduring social consequences to bad or even ill-considered behavior. Social capital is built upon what we know about each other, for which we are willing to be held accountable.

The third principle for our community web site is restricted access. Access should be restricted to townspeople and employees of the town. As a purely practical matter, restricted access is needed to eliminate anonymity, but there are other reasons to want it. People will feel more comfortable making contributions if they know that the audience is limited. They will feel more comfortable sharing private information -- phone numbers, email addresses, and names of their children -- if they know that the distribution is limited. And, parents will feel more comfortable permitting their children to participate in chats, online games, and other activities if they know that the other participants are from their own community.

I continued to think about the Carlisle Community Center. As a researcher, I want to know whether the web can be used to promote the development of social capital. I want to know whether the three principles mentioned above are the critical ones. I want to know what townspeople will expect from a community web site. As a Lotus employee, I am interested in knowing how our technologies can and should be used. I am interested in discovering new services to deliver over the Internet. Finally, as a resident of Carlisle, I am interested in obtaining more ways to understand my town and the people in it.

Service Description

The web site for the Carlisle Community Server is like a building -- a Community Center -- composed of many different rooms for engaging in different activities. Upon entering the building we encounter a lobby through which we can travel along corridors to a variety of meeting rooms. Each room is designed for a different purpose.


We prefer the term room because it brings to mind several characteristics not always associated with web pages. Primary among these is the notion of controlled access. One can lock the door to a room. One can post a note indicating who may enter a meeting. Or, one can fling the door wide open and simply see who walks in. Rooms are a physical technology for managing meeting and group membership. We want to carry that notion into our online applications.

The rooms in the Carlisle Community Center are organized as a Hierarchy. At the top is the Lobby. This acts as the entryway into the Center. Off the Lobby are four corridors:

    1. a school corridor,
    2. a town corridor,
    3. an organizations corridor, and
    4. a clubs corridor.
These corridors help organize the many rooms into meaningful groupings.

Off the Lobby and the corridors are rooms designed for a specific purpose. This might include a class room for teachers and parents to meet. A committee room for town committees to keep documents and schedules. Or, a club room for a book club to discuss the latest reading. The room is set up by its members to accommodate a particular group engaging in a particular set of tasks.


Each room supports five activities. Four of the activities permit room members to examine and add information to the room. These informational activities are:

Bulletin Board
Read and post messages. Room members may create a response to a posted message and even post responses to responses. The entire discussion "thread" or hierarchy originating from an initial topic may be examined.

Look at and add events to the room calendar. Calendar events also support responses, so that a discussion may ensue from a posted event.

File Cabinet
Examine and "upload" documents. The file cabinet is a repository for computer files that room members wish to share among themselves. Like the bulletin board and calendar, room members may also respond to documents in the file cabinet.

Follow and add interesting links to other information on the web.

The fifth activity is the ability to navigate to and create rooms beneath or within a room. While the ability to travel to the inner rooms is generally available, the ability to create new rooms is usually restricted to room members with special privileges, namely the room administrators.

Room Administration

The users of a room are not all equal. There are three types of roles supported by a room:

  • Observer
  • Member
  • Administrator

Generally speaking, Observers may look at the contents of a room, but may not enter new information. Members may both look and enter new information. Administrators are members, who have permission to rearrange the room.

The privileges of the administrator include the following capabilities:

  • the ability to change the name, description, and icon for the room,
  • the ability to change the welcome page of the room,
  • the ability to control who may use the various activities,
  • the ability to determine who may be an observer, a member, or an administrator, and
  • the ability to clean up the contents of the room.

In a sense, the administrators of the room set the policies of the room. It is possible to have a totally egalitarian room in which all members are administrators, an "authoritarian" room in which one benevolent dictator sets the policies for everyone else, or something in between. Each group must decide which approach best meets its needs.


The purpose of the Carlisle Community Center from a research perspective is to examine the interaction between online communication and face-to-face communication, and to test whether or not a "closed" community in which identities are verified makes users more comfortable to participate in online environments.

Data Collection

In order to learn from the Carlisle Community Center project, it will be necessary to collect data about its development, uses, and effects. Data collection should strike a balance between rigor and relevance while always protecting the privacy and confidentiality of individual respondents. On the side of rigor, it will be useful to repeat measures used in previous research, so that Carlisle results can be compared with them. It will also be important to use techniques that minimize demands on users' time such as automatic logging of server use. On the side of relevance, it will be useful to rely upon key informants such as room administrators, and key users of the site to help us identify emerging issues.

We will collect data at three times to assess site use and effects.

Pre-usage Baseline: We will collect basic descriptive data on people's background characteristics (e.g., gender, age), characteristics that might be expected to affect their use of the Community Center (e.g., prior computing experience), and characteristics that using the Center might change (e.g., size of Carlisle social circle, involvement in Carlisle activities).

Ongoing Usage: We will collect data on use of the Center. "Ongoing Usage" is actually not a single moment in time, but will be the average of a person's use data over a specified duration of time like one month or one year.

Post-usage Snapshot: We will collect the baseline measures again to identify potential effects such as involvement in Carlisle activities.

Questionnaire Data Collection

We will send at least two questionnaires to each person who requests a user account, a baseline questionnaire for background information and a follow-up snapshot questionnaire for "effects" information. We may send additional questionnaires to explore particular development or use questions. For example, if we discover that one room seems to be extremely popular, we may develop a special small questionnaire to send to people who visit that room to try to understand its popularity. Or we may send a standard questionnaire to every room administrator to try to understand factors that are beneficial or detrimental to administrators.

Interview Data Collection

We will schedule informal interviews at people's convenience to hear, in their own words, how people describe and assess their experiences with the Community Center.

Automatic Data Collection

To the extent possible we will collect data on how people use the Center by automatically logging behaviors such as how frequently a room is visited and by how many people. Automatic logging minimizes demands on users' time and produces accurate results. Data will be logged for individual users in order to be able to correlate use with other characteristics of people, but logging data will be reported only at aggregate levels to protect individual confidentiality.

Data Confidentiality

To understand how people use the Community Center and with what effects, it will be necessary to collect data describing characteristics of individuals such as their age, gender, level of community involvement, and use of the Center. We will follow procedures for the protection of respondent data developed over time at Boston University and Carnegie Mellon University in previous studies of home and community computer use. Generally speaking, these procedures entail obtaining informed consent from everyone who provides data (including parental permission for respondents under the age of nineteen), assigning a coded identifier to each person's data (whether it be from a questionnaire, interview, or data logging protocol), then stripping any personally-identifying information (such as name or address) from the data. Data will be reported in aggregate form only and no individual identities will be revealed.

Data about individual users is kept confidential and is only available to the Lotus and NYU researchers.

Contact Information

For questions concerning the project in general, please contact:

John Patterson (Lotus Principal Investigator) at Lotus Research: or 617-693-4236.

For questions concerning data collection, subject's rights and any other concerns or questions related to the data collection and analysis, please contact:

Lee Sproull (NYU Principal Investigator) at NYU: or 212-998-0804.