In hindsight I was extremely lucky to work with NPTN when I did, since it filed for Chapter 7 bankruptcy on September 17, 1996. At the time I was there it was not yet certain whether the organization would continue to exist for months or years, and so a major part of my job involved making sure the RIN program could continue in the indefinite future without my presence. This type of planning was invaluable in shaping my perspective on the community-networking movement: if it is to succeed, it must not depend on any specific individual or organization.
research for this paper
At the same time, of course, I had to do independent research for a paper. I was originally going to study the effects of the changes we were going to make on some of those big Free-Nets. Clearly that was no longer an option. As I familiarized myself with the RIN systems and prepared them for shipment, I started to realize that no one at NPTN -- or anywhere else, for that matter -- had really figured out what it is that makes some Free-Nets phenomenally successful and others ... less so. The RINs I was preparing did not seem ideally suited for success; the ones that had already shipped using the same hardware and software weren't doing very well. I knew it wasn't from a lack of good intention at NPTN, so it must have just been a lack of knowledge of what would make the systems really work.My research question, then, was, "To what extent do Free-Nets succeed in improving the communities they serve, and why?" I tentatively hypothesized that the differences between systems are due to differences in age, size, hardware and software, and management style.
By that time I had already gleaned the Cleveland Public Library and the World Wide Web for all they had to offer on the topic of community networks. (Since the Web is constantly expanding, additional useful documents are probably available now that were not there when I did my research.) Tim Connors helped considerably by lending me his copy of New Community Networks: Wired for Change by Douglas Schuler, which became the most-quoted source in this paper.
I decided the next step was to choose some Free-Nets to study. The Cleveland Free-Net (CFN) is the oldest and largest Free-Net, using FreePort software under the UNIX operating system. The Great Lakes Free-Net (GLFN), located in Battle Creek, Michigan, is middle-aged and middle-sized, one of the first RINs and the very first Free-Net to use a graphical user interface and a Macintosh computer as its server. Finally, the Worth County - Sylvester, Georgia Free-Net (WC-SGFN) is less than two years old and very tiny, and its hardware and software are nearly identical to those in the new RINs I was preparing. Between the three of them, I felt I had a complete sample of the hardware and software setups used by Free-Nets, as well as a cross-section by age and size.
I thought it would be important to collect some sort of quantitative data, so I went through the Free-Nets and mapped their hierarchies of menus. This task would have been nearly impossible on the titanic CFN, but thankfully it has the ability to automatically generate such a map on demand. On the two smaller systems, I also measured the rate of "traffic" (messages per day) in all the message forums. This information is available in the appendices. Most of it turned out to be irrelevant to this paper, but it may yet be useful for future research or entertainment.
Meanwhile, I put together a list of questions that I wanted to ask users of the Free-Nets. Nearly all of them were taken directly from or inspired by things I had read in my earlier research. Rather than phrase them as questions, I left them as opinion statements to which respondents could react. I assembled them into a list in an order which I said was no order at all, but I actually took care to separate similar opinions from each other and threw in a few other opinions which I hadn't actually encountered in print and which weren't directly related to my research but about which I had been personally wondering. The idea was to keep the respondents from figuring out what exactly I was researching, so that they would respond to each statement in isolation. A copy of the survey that was sent to CFN users can be found in the appendices.
The next step was to find some subjects. On CFN I posted a few of the statements to message forums and invited users to react. On the basis of these and other discussions that were already in progress, I assembled a list of people I wanted to question. By the time I got to GLFN and WC-SGFN, I was running too short on time to seed the conversation, so I chose subjects only on the basis of existing discussions. I deliberately did not select any of the systems' sysops.
I then sent the surveys to these people by e-mail. I was careful to not send them from my account at NPTN, since I thought the knowledge of my affiliation might influence their answers. Eleven of the thirty recipients at CFN wrote back with completed surveys, four of eight at GLFN, and only one of four at WC-SGFN.
It is interesting to note that I intended to send out ten surveys at WC-SGFN, but because the software at that system uses separate usernames for e-mail and message forums, I was unable to figure out six of their e-mail addresses! I should also note that even with only one respondent from that system, I still surveyed a much larger percentage of its total usership than at either of the other two Free-Nets (0.53% as opposed to 0.01% at CFN or 0.04% at GLFN).
My nondisclosure of who I was or why I was conducting research disturbed at least one respondent: "I kinda wish you'd say a little more about yourself, like research for who. Is this a paper for school or do you work for freenet or what? I'm not normally paranoid but..." (Twenty-Seven) However, I still feel that giving away that information would have altered some of the users' responses.
Since I was living within the Cleveland calling area, I called four of the CFN users who hadn't responded by e-mail and interviewed them over the phone. I also interviewed Tim Connors and John Kurilec, both of NPTN, in person. All interviews were recorded, with the subjects' permission.
The survey contains a promise that the users' names will not appear in this paper, and I made the same promise to the users I interviewed. All the names of these users, along with those of people who posted messages to Free-Net message forums, were replaced by "Anonymous One," "Anonymous Two," and so on. In some cases the same person was assigned more than one "anonymous" name because s/he provided information from multiple sources. I am currently in the process of contacting these individuals to ask permission to use their names beside their quotes. Several formerly anonymous sources are now given appropriate credit.
breakdown of the sample
Clearly, I did not survey a random or otherwise representative sample of Free-Net users. I chose my subjects on the basis of whether I thought they would give me useful quotes. The following graphs indicate the sample's composition (with the exception of Connors and Kurilec, who do not represent any specific Free-Net) in terms of sex, age, location, years of Free-Net use, years of using online services, and years of computer use.
a word about quotations
Throughout this paper I have reproduced quotations in a form as close to the original as possible. This means that some excerpts from e-mail messages may appear with nonstandard capitalization, punctuation, and even spelling, including varied spellings of the service mark "Free-Net". Quotes from interviews appear as spoken. My intent is not to make anyone look silly but rather to capture the style and character of the various forms of communication. I have corrected a few typing errors which would otherwise have been distracting.Table of Contents
Backward to Background Information
Forward to Greater Cleveland, Ohio, Calhoun County, Michigan, or Worth County, Georgia
Bibliography