by R. Douglas Kirsch and Thomas E. Woods
Return to Table of Contents
At 10:00AM the doors to the Nothin' But FreeNet Web site.)
All of the kids and adults seemed to start off with the same question: "What can I do on the Internet?" We replied, "What do you want to do?" The children asked for games, information about their favorite sports teams and musical groups; the adults wanted to find information about job opportunities and read a newspaper from their home country in South America. With a little bit of coaching to get them started, they all found what they were looking for. They did not stop there, they kept exploring, talking to each other across the table and the room about the new things they were discovering. Before too long, even the most reticent participants were engaged and excited.
There are a lot of reasons for their reaction: the novelty of having computers available, the satisfaction of their curiosity about the much-hyped Internet, the sheer scope and breadth of the resources available through it. But we think the most basic and powerful sources of their excitement were choice and control. The Internet offered them a staggeringly wide range of choices, and gave them control to go straight to the resources of their choice. These are values that transcend technology: people want options and the power to exercise them. This is in many respects the very basis of freedom.
Choice and control are two dimensions of liberty: choice refers to the right of citizens to exercise their liberties with judgment and due concern for the rights and well-being of others and of society; control refers to the means or the tools that citizens need to exercise choice. In the context of information technology, choice refers to users' right to decide what information and services (or content) are relevant and valuable to them; control refers to their ability to access any such content, as well as their ability not to access content that they find harmful or damaging.
These principles apply to content that users create as well as content that they access or use, the place where they come to use technology as well as the technical platform. Users should be able to exercise choice in what content they can develop for dissemination over a universal network, and control over how that content is produced, presented, and made available to other users.
Return to Table of Contents
These values apply to all users of information technology -- including both individual citizens and the communities they live in. As explained in the section on "Community Development", we believe that a sphere exists for community activity on information networks, just as there are spheres for individual enrichment and commerce. Communities should therefore be able to exercise the same choice and control as individuals and businesses, operating on the network on the same terms as other users.
The dimensions of information technology over which users should be able to exercise choice and control include:
Users (both individuals and communities), should have the choice and control to determine how they use the network, how they interface with others over the network, and with whom they interface over the network.
1. Training to Develop Information Literacy
Return to Table of Contents
One of the early participants in our project, Yolonda Thomas, has translated her training on how to use the Internet into a new career. She received technical training from Austin Free Net (AFN) as an Americorps volunteer at the Booker T. Washington Family Learning Center, one of the first public access sites in the city. In addition to her technical training, Yolonda says, her exposure to the Web expanded her horizons and made her aware of opportunities she had never known existed. She credits this exposure with helping her land a job in the City government and leave public housing to move to a better neighborhood.(1)
Why was Yolonda's experience so successful? She credits both the quality and quantity of training that she received from AFN, and the fact that she was trained in a community location that was already known to her -- a comfortable, familiar environment. The Learning Center already served as the focal point for a variety of education and family support activities in her community; when the public access site was created there, it fit naturally into the Center's mission, atmosphere and dynamic.
Unfortunately, the Family Learning Center where Yolonda received her training is now closed. (One of the sponsors of the Learning Center, the Autin Learning Academy, is currently searching for a new location in which to resume the education and family support activities that used to take place at Booker T.) At this writing, there are only three places in East Austin where residents can get public access to the Internet: the Carver and Dove Springs library branches, and the Montopolis Community Center. These public access sites, particularly the one at Carver Library, receive considerable use; but all are in dire need of more on-site training and assistance for users. This shortage of on-site assistance makes it much more difficult for someone like Yolonda to develop meaningful expertise at one of these existing public access sites, compared to the learning environment created at the Learning Center.
Background: One of the largest barriers to meaningful choice and control among the people we've worked with is the above mentioned lack of training resources to help them learn what content is available, how they can access it, and how they can produce their own content to meet the needs of their community and their neighbors. This issue goes beyond learning how to use Web browsers or write an HTML program. It's about developing information literacy:
In order to use the Internet effectively -- or simply to become productive citizens in an information-driven economy and society -- users must possess these skills. No organization structures or sophisticated screening tools can effectively substitute for individual responsibility and aptitude.
Most users of the Internet are well-educated and relatively affluent, and have already developed high levels of information literacy by the time they encounter the Internet. New users from low-income communities rarely enjoy this advantage. Most come from low-quality schools, have comparatively low educational attainment, and have work histories in fields that don't require very much information literacy. The need for training is therefore especially acute for this population.
Several existing community access projects place a strong emphasis on training. Two of the most impressive efforts are the following:
Here in Austin, AFN has focused in its first year on developing public access stations and providing training for community nonprofit organizations. AFN's director, Sue Beckwith, appreciates the need for more community-based training resources, and is working with businesses, individual volunteers, access sites, and other community groups to develop them. Sue hopes to assemble teams of volunteers that will be tied to each public access location; and over the next several months such locations will be opened at every branch of the Austin Public Library. The magnitude of this effort is considerable, however, and may be beyond AFN's capacity with existing resources. If you can help AFN achieve its vision of public access in Austin, please contact Sue Beckwith.
What our community partners have taught us: We have found a strong demand for training resources that people can use in community settings that are convenient and comfortable for them. In planning meetings with our community partners, the most frequently cited need is for a comprehensive training site in East Austin. Our partners believe that such a facility could promote ongoing public access efforts by (1) giving people inside the community a single place to turn for help in achieving information literacy, (2) giving volunteers and donors outside of the community one place to focus their efforts to improve information literacy on East Austin, and (3) producing information-literate East Side residents who can share their skills with others, building the capacity of the whole community to capitalize on the information economy.
Recommendations: We believe that the primary responsibility for providing this kind of training resides with community organizations: they have the locations, the existing relationships with people in the community, and the responsiveness to the particular needs of their neighborhoods. Both government and businesses should support training efforts by such organizations by providing funding, in-kind assistance, and technical support.
We envision a network of partnerships among community groups, businesses, local government, and concerned individuals from across Austin. The overall goal of this partnership would be increasing information literacy on the East Side and helping to ensure that both East and West Austin benefit from the city's burgeoning high-tech economy. The first major project of this partnership would be establishing a comprehensive community networking center, charged with the following tasks:
2. Organizing Information on the Network
Return to Table of Contents
We met one new user at the Carver Library public access station. Formerly homeless and currently a volunteer for a local homeless shelter, this man was searching for information about Christian faith-based programs for the homeless. In his first encounter with the Web a few days before our encounter, he had performed a keyword search through one of the major search engines and came across a reference to a program in California. When he performed the same search on his second encounter with the Web, the reference did not appear. One of our class members spent about 30 minutes trying to help the man locate the reference by further tailoring the keyword search. But while the search turned up several other potentially relevant URLs, they failed to locate the desired reference. The man observed that a reliance on keyword searches is both a strength and a weakness of the Web: while he was frustrated by his inability to track down the information he originally sought, he was glad to have browsed through some Web sites that he may otherwise never have seen. Still, he concluded, "It seems like there ought to be some way to pull all those things (i.e., URLs related to faith-based programs for the homeless) together somewhere."
Background: The World Wide Web is rapidly evolving into the predominant interface with the Internet. Through Web browsers like Netscape and Mosaic, users can locate, view, and download text and graphical documents designed for the Web; download text, graphics, video, multimedia, and other types of materials stored in gopher and file transfer protocol (FTP) formats; participate in Usenet resources such as newsgroups; and send electronic mail. Yet for all of these capabilities, the Web is simple to use and is based on a straightforward, text based set of codes called Hypertext Markup Language (HTML).
However, the Web lacks some of the organizational qualities of private subscriber systems such as America OnLine (AOL). AOL provides access to their own internal content which is very well organized, and soon they will give all of their subscribers Netscape. Also, you can access AOL anywhere. On the other hand, AOL tends to be expensive compared to Internet Service Providers (ISP). Their low cost, is made less compelling here in Austin by the availability of public access sites and e-mail accounts for low-income families through Austin Free Net. These advantages have made the Web the focus of much of our group's work with our partners in the East Austin community. One problem that our partners have cited with the Web, however, is the difficulty of finding information.
Search Engines Research on Web usage indicates that a large share of the traffic online is centered on search engines like Webcrawler, Lycos, Yahoo!, and Alta Vista(5) search engines are commonly used features of the Web for locating information or other content available on the Internet. Each search engine is set up to respond to user inquiries for information on a particular topic. For the most part, they rely on keyword searches. Some, like Yahoo! and Webcrawler, supplement the keyword searches by grouping links in broad topic categories; others, particularly Alta Vista, rely almost exclusively on keywords. These tools have the strengths of providing users with a wide variety of potentially relevant uniform resource locators (URLs) to choose from, and allowing users to conduct searches according to their own criteria rather than pre-selected criteria.
But existing methods also lack coherence and consistency. The same search parameters yield sharply different results when used in different search engines, and sometimes at different times on the same search engine. Even the best-defined keyword search can miss some relevant URLs -- and find some completely irrelevant ones -- simply because keyword searches can not put words in context. For example, a search for information on "mustard" could just as easily turn up information on chemical weapons as on cooking. As the volume and variety of materials available over the Web increase, the limitations of existing search engines will only become more apparent, and the need for some sort of network-wide organizing criteria or protocols will only become more evident.
Library Protocols Large research and university libraries are currently developing protocols like Standard Generalized Markup Language (SGML)(6) that attach identifying labels to electronic content. These labels allow information to be organized by topic, author, genre, format (i.e., text, video, graphical, interactive), bibliographic information (i.e., place and date of original publication), and several other criteria. These protocols, in combination with search tools that can accommodate multiple search criteria at once, make searches in library systems (like the UTCat here at the University of Texas) much more focused and rewarding than searches over the Web engines. But attaching such labels to content is very time- and labor-intensive, and multi-level keyword searches can be very complicated and cumbersome to conduct over a network. The approaches being developed by research libraries represent a significant advance in the rational organization of electronic content, but are probably too cumbersome to be applied to the Web as a whole.
Index Pages Individual users and organizations have attempted to address the dispersion of information through the Web by creating "index" pages that provide lists of links to URLs that they consider useful. These informal indexes represent the Internet equivalent of word-of-mouth: they are helpful pointers for visitors to a site from the people who "live" there. Everyone involved in our project has come across valuable URLs through such indexes, and we dearly hope that they remain a feature of the Web and whatever technologies supersede it. But their ad hoc, informal nature is a weakness as well as a strength.
What our community partners have taught us: Our contact with individuals in low-income communities who are using the Internet for the first time taught us that content needs to be organized more coherently online. Our community partners have been uniformly impressed by the scope and value of information available through the Web, but they have also expressed frustration with the limited tools available to help them track down information that they need. A few examples may help to illustrate their frustrations:
We believe that the difficulty of locating relevant information online presents not just an inconvenience, but a barrier to achieving the goal of real universal access. User-directed searches and user-assembled indexes are and should continue to be essential tools for locating content over large electronic networks like the Internet. But as more and more types of content emerge on such networks, and as the diversity of users, interests, and needs increase, it will become less and less effective to continue to rely on these tools alone. More comprehensive, coherent, and consistent protocols must be developed to supplement and strengthen existing search engines.
Recommendations: Individual users should be held responsible for attaining information literacy. (See the section on Training.) Even the best-designed search tools and protocols for organizing information can not substitute for users who know what information they're looking for and where to look for it. Community organizations, government, and business should support the efforts of individuals to develop their information literacy by supporting community networking initiatives and training programs. Secondly, there is a need for software that allows casual users of public access sites ways to store e-mail addresses, URLs, FTP sites, or other information they want to record, on diskette. This would allow people to bring in a floppy disk with an organized directory of Internet resources, and then just double-click on a particular item and have it launch the appropriate program. For those facing restricted lengths of computer use at public access sites, a disk like this could save them immense amounts of time. Ideally, this program would be in the public domain of the Internet. If you have any ideas or information about how such a program could be developed, please contact us!
Government could also work with software developers, content providers, librarians, and users to develop protocols for categorizing Web-based content. Such protocols should be voluntary, and almost certainly would not be adopted universally right away. But it should be possible to develop an Internet-wide standard to make information easier to locate and use.
3. Protecting the Privacy of Users
Return to Table of Contents
Our partners have told us that churches and church-sponsored community organizations in East Austin are hesitant to incorporate the Internet into their programming for fear that it could expose their members, especially young children, to objectionable materials. Of course, the leaders of these church groups have seen the stories in the media about pornography being easily available through the Web. But of equal or greater concern to them are materials that promote or glorify violence, racism, or anti-religious attitudes. We believe that grass-roots community organizations like these should play a leadership role in networking low-income communities (see our paper on "Community Development"). Their anxieties therefore present a challenge to creating true universal access.
Background: User privacy encompasses both the materials accessed over a network and the contacts made with other network users. It involves protecting children from exposure to materials their parents deem inappropriate; protecting the confidentiality of private communications and personal information; and protecting both children and adults against unwanted communication or harassment from other users.
Protecting children The availability of pornography over the Internet, particularly the danger that children could be exposed to it unwittingly or without their parents' knowledge or consent, has been the subject of much media attention. While this particular problem is not really of the magnitude suggested by the media, many parents are also concerned about materials that contain racist or violent images or opinions. To help parents restrict or monitor their children's access to such materials, software companies have developed several applications that screen incoming files for keywords and deny access to those files that appear to be objectionable. Programs like Surf Watch, CyberGuard, and Net Nanny have the advantages of adaptability (parents can select which words to screen for), context (some words will only trigger the screening mechanism if they appear repeatedly or along with other words), and simplicity (they are easy to install on a single computer). They have the disadvantages, however, of being cumbersome (taking up large amounts of computer memory and slowing the rate of transmission to the computer) and somewhat vulnerable (since the program resides on the individual computer, an enterprising child could remove, deactivate, or subvert the screening program.
Protecting confidentiality Confidential information and communications over the Internet also need protection. Currently it is possible for unauthorized users, with some expertise and basic information, to read and even alter such information and communications. While this kind of vulnerability is not unique to electronic media and doesn't create problems for the vast majority of users, it is a real problem and serves as yet another deterrent to people in low-income communities from participating in the Internet. Technology offers the possibility of protecting confidentiality with encryption, or encoding information on the network in such a way that only authorized users who possess the code keys can read it. Because encryption could be used to conceal transmittal of illegal information it is a controversial approach; and under current law it is illegal to export encryption software. One increasingly prevalent encryption standard, PGP (for "Pretty Good Privacy"), has been disseminated for free over the Internet, landing its creator Phil Zimmermann in legal trouble. While our project has not addressed the legal issues surrounding encryption, we are interested in the potential of such tools to make communications more secure.
Protection against harassment New users and potential users are also concerned about being able to avoid harassing or offensive e-mail from other people over the network. This is a particular concern in low-income communities and among racial and ethnic minorities. People in these communities have received scorn and hostility from others outside their communities before, and are keenly aware of the risk that the Internet could easily become an avenue for such attacks in the future. We are aware of few existing tools that would be analogous to "call screening" in the realm of telephone service. So far as we're aware, users must protect themselves against such harassment by using caution (e.g., never arrange meetings in a private place with someone you've only encountered via e-mail) and policing their own e-mail (e.g., learn to recognize the addresses of people from whom you've received harassing e-mail and simply delete any transmissions from them without opening them).
What our community partners have taught us: Our partners have told us that concerns about protecting privacy and shielding children from inappropriate content are barriers for individuals and community organizations that deter them from full participation in the Internet. For example, participants in the Austin Learning Academy expressed some anxiety about creating their Web page. They feared that a site devoted to an organization serving predominantly low-income, African American children living in public housing could provide an irresistible target for hate mail from racists or from people who resent public housing residents. These fears have not been realized, but we do not believe that they are unfounded.
Our partners tell us that they want both choice and control: choice in determining what materials are objectionable or inappropriate, and control over their and their children's access to such materials; choice of who to associate with over the network, and control over who can communicate with them. There is a need for tools that put this kind of choice and control in the hands of users, rather than those of network administrators, government officials, or content providers.
Recommendations: Individuals and organizations should be responsible for setting the terms of their own participation in the Internet or any other network. Software makers, access providers, and government regulators should work to provide all users with the necessary tools and training to fulfill that responsibility, and ensure that anyone who violates the legitimate privacy rights of someone else on a network is identified and punished.
The Platform for Internet Content Selection (PICS) encryption system outlined by the World Wide Web Consortium at MIT represents one attractive vision of how encryption could serve these goals.(7) PICS is based on a series of encryption codes that would be attached to all electronic content. The codes would not only encode the materials but provide basic information about the contents (i.e., format, general topic, intended audience). These codes would enable software packages to screen content according to parameters identified by each user. This screening device could reside either on individual computers (where users could easily change or adjust the screening parameters) or on server computers (so that, for example, a religious organization could screen all materials going to church groups in their local area). The PICS system could eventually be expanded to support improved information searches, by adding codes that classify materials according to some accepted standards. (See above section on "Organizing information online".) The PICS approach is certainly worthy of further exploration. Some version of PICS could be tried out on a local level here in Austin when the city-wide broadband network is constructed; CSWC, the contractor that is negotiating with the City to build the network, should adopt such a system and work with community organizations to see that it is implemented in conformance with community needs.
Return to Table of Contents
(1) Meghan Griffiths, "The Net Result of a Free Ride," Austin Chronicle, April 4, 1996. Also see http://www.austinfree.net/press/artchr96.htm Return to body of text.
(2) Plugged In Home Page, http://www.pluggedin.org/PluggedIn.html Return to body of text.
(3)Plugged In Enterprises Home Page, http://www.pluggedin.org/pie/piehome.html Return to body of text.
(4) Somerville Community Computing Center Homwepage, http://www2.wgbh.org/mbcweis/ltc/sccc/sccc.html Return to body of text.
(5) University of Michigan Business School, GVU/Hermes Research Project on Usage and Applications of the Worldwide Web, http://www-personal.umich.edu/~sgupta/hermes/survey3/access.html#table Return to body of text.
(6) Stanford University Libraries and Academic Information Resources, "TEI P3 Chapter 2: A Gentle Introduction to SGML", http://www-sul.stanford.edu/tools/tutorials/html2.0/gentle.html Return to body of text.
(7) Worldwide Web Consortium, PICS Homepage, http://www.w3.org/pub/WWW/PICS/ Return to body of text.