Home | Proceedings


AT&T Chat 'N Talk: Getting to Know You without
Getting to Know All About You

Helen A. Fairbrother, Elizabeth A. Hohne, Steven Todd
AT&T Labs
200 Laurel Avenue
Middletown, NJ 07748

hfairbrother@att.com
elizabeth@att.com
steventodd@att.com

Abstract

Interpersonal communication is one of the driving forces behind the success of the Internet and the World Wide Web (WWW). The popularity of e-mail, bulletin board systems, chat rooms, and instant messaging suggest that the exchange of information "online" is here to stay. The variety of communication options and the nature of the Web support direct communication between people with on-going relationships as well as providing opportunities for strangers to establish relationships. One of the primary means of meeting on the Web is via chat rooms. AT&T Chat 'N Talk is an innovative application that provides Internet chat room users with a way to connect by phone without revealing their phone number or identity. This allows users to take the next step toward an ongoing relationship while maintaining a level of anonymity and privacy. The call flow of AT&T Chat 'N Talk violates the typical model of establishing communication events. This leads to some limitations in usability and usefulness, for which solutions are discussed.

Introduction

Communication is one of the driving forces behind the success of the Internet and the World Wide Web (WWW). Communication is broadly defined as the exchange of thoughts, messages, or information that supports relationships that are new, old, business-related or personal. E-mail, bulletin boards, chat rooms and instant messaging provide the main mechanisms for interpersonal communication on the Web today. Taken together, these new methods of communication have fundamentally changed how we communicate, with whom we communicate, and when we communicate.
Web sites and electronic commerce support the exchange of information 24 hours a day. E-mail has revived the art of letter writing and transformed the way we work in global organizations. Bulletin board systems and news groups support communities of interest in which the success of the community rest on our ability to contribute to conversations spanning time and space. MUDs and MOOs create contexts for fantasy and play behind a mask of anonymity and disguise. Chat and Instant Messaging support our need for immediacy in our communications, while Buddy List systems diminish our isolation by letting us know that our "friends" are out there somewhere. The data surrounding the affects of Internet relationships maintained via these communication methods are not all in (Kraut, et al 1998; Hamman 1999; Katz, et al 1997; Rheingold, 1998) , but the phenomenal popularity of online communication services suggests that whatever changes are being wrought are here to stay.
There is considerable overlap in the use of these communication methods in a given situation. For example, many chat programs allow a person to send a private "instant message" to other participants in the room; a sequence of instant messages between "buddies" may lead to a more extended e-mail conversation, or even a phone call. The boundaries between these "on-line" forms of communication and other forms such as paging, voice mail, and telephony are becoming more fluid as users become more familiar with each, and as the underlying technology migrates toward an IP (Internet Protocol) platform. For example, it is now possible to get paged when an e-mail message is received, and to access that e-mail message over the phone. As the boundaries blur, the usability of these communications methods will depend on how well integrated the different capabilities appear to the user.
At the crux of Internet communications are the issues of privacy (Schneier & Banisar, 1997; Agre & Rotenberg, 1997; Kling, 1996) and establishing trust (Rheingold 1993; Dibbell, 1983, 1999; Hafner, 1997). The discussion of privacy on the Internet (or in an Internet-based world) is wide-ranging and complex. The main issues center on the ability of users to send information in ways that prevent eavesdropping--i.e., encryption--(Hoffman & Carreiro, 1997) and the ability of corporations, the government, or individuals to collect and (ab)use information about users, including demographic and identifying information supplied by the user, as well as information collected via "cookies" that track the activities of users as they browse the WWW (Cranor & Reagle, 1997; Lin & Loui, 1998). The privacy of e-mail has received a great deal of attention during the Microsoft anti-trust trial (Meeks, 1999). Issues of trust surface at both the corporate level (Cheskin Research & Studio Archetype/Sapient, 1999) and the personal (Dibbell, 1999). Privacy issues have added relevance to a service like AT&T Chat 'N Talk, because of the social context in which such a service is likely to be used.
Both privacy issues and communication integration issues are critical to the AT&T Chat 'N Talk service, a call-back service that allows chat room participants to shift from a text interaction to a phone call without exchanging their phone numbers. AT&T Chat 'N Talk is an interesting example of how communication services are being developed for the Internet in the late 1990's. The social context and the underlying technology have implications for the usability and usefulness of the service. We will discuss the way in which we have addressed those issues.

The Context for AT&T Chat 'N Talk

A Brief Description

Imagine that you are in an Internet chat room where the topic of conversation is genealogy. You participate in the ensuing conversation by typing in your comments and replies. Based on the conversation, you determine that one of the people you have been chatting with may possess information you have been looking for that would help fill in some of the missing pieces of your family tree. At this point in the conversation you feel like you want to talk to the person on the phone but you are reluctant to give your unlisted phone number to your new Internet friend. You click a button on your chat screen labeled "Phone" which provides a way for you to set up a regular telephone call without either party exposing their telephone numbers.
Figure 1 is a cartoon depiction of the AT&T Chat 'N Talk Service. While in a chat room, chat participants decide that they would like to talk on the phone. Each participant enters in his or her own phone number via the Internet and triggers the phone connection. The phone calls are carried over the traditional, high-quality voice network. This occurs when an element called the "Call Broker" causes a conference bridge in the traditional phone network to place calls to both endpoints. Caller ID is blocked and a "dummy" phone number appears on the bill. Thus, the phone numbers of the participants are never revealed.
Cartoon depiction of AT&T Chat 'N Talk Service
Figure 1: Cartoon depiction of AT&T Chat 'N Talk Service.

The Social Context

As the example indicates, the social context for the use of a service like AT&T Chat 'N Talk is a chat room, and what would compel a chat participant to use the service is that he or she wants to increase the level of intimacy with another participant. In many cases, the level of discussion in chat rooms is impoverished rich enough to indicate that you d like to know more about a person, but not rich enough to really follow through with acquiring that knowledge. The medium of Chat hides (or allows one to hide) a lot of important information about a person: gender, age, etc. Much of this information becomes apparent when the person's voice is heard.
Two privacy issues are particularly important in the chat-to-talk context. One issue concerns the fact that many chat rooms are devoted to salacious topics, as well as to self-help or group support topics. Users may find it particularly desirable to protect their anonymity in such situations. In this case, the user would prefer to protect his or her privacy at the system level. A second, and more personal, privacy issue is the one that drives the need for a service like AT&T Chat 'N Talk: Privacy about who you are and how to get in touch with you is very important to people, especially in a world where anyone who knows your phone number can use it to discover your real name and find a map to your house. The level of privacy that AT&T Chat 'N Talk ultimately supports is the latter. By providing a way for each user to enter his or her own phone number into the system, the participants do not obtain the means to contact each other outside of this particular context. If the phone-based interaction is not satisfactory for some reason, it can end there. Nobody will be required to move to another state or change his or her name. (Note that another way people may increase the level of intimacy somewhat while protecting their identities is by signing up for one of the many available free email services using a false name. Email keeps the communication in the text domain, but allows for more extended conversations, while still protecting the user s "real" e-mail address).
For legal reasons, the actual information about the phone numbers associated with a particular call is store in the system database. This precaution is necessary in case there is a need to subpoena telephone records. The fact that each user must actually provide his or her phone number to the system creates an interesting paradox. A recent survey of Internet users' privacy concerns (Cranor, Reagle, and Ackerman, 1999) found that people are especially reluctant to reveal their telephone numbers (11% would feel comfortable revealing phone number, as compared to 76% for email address.) It is an open question whether people feel more comfortable giving their phone number directly to another person (via an instant message, for example) or to the system via the AT&T Chat 'N Talk set-up screens.
Another variable that contributes to users' willingness to supply information in an e-commerce context is the "trustworthiness" of the site. (This is an e-commerce context, since users must enter credit card information to initiate the call.) Trustworthiness is related to the navigability of a site (Cheskin Research & Studio Archetype/Sapient, 1999). This turns out to be an issue for services like AT&T Chat 'N Talk, which could either be offered as a general-purpose communications program that could be used with any chat room, or could be built in to particular chat room software. It turns out that there are significant usability advantages in integrating the functionality into the chat software, which, broadly speaking, are related to "navigating" the application. These are explored below.

The Business and Technical Context

AT&T Chat 'N Talk was designed as a part of a suite of products called AT&T Interactive Communications, which also includes AT&T Click2Dial Conferencing, an internet-based conference call service. From a business standpoint, these services are important for several reasons:
  1. These services give AT&T an initial presence as a communications provider within a new "distribution" channel: AT&T is in a position similar to Barnes & Noble in the on-line bookstore world. AT&T is changing its image as a traditional telecommunications company. For that reason, it is important for it to invent new services that work in the context of the Internet. These services also provide another opportunity for AT&T to generate minutes of use on existing infrastructure (the voice network).
  2. These services introduce a new population of users to conferencing services: AT&T Click2Dial Conferencing was conceived as a consumer service, and has been promoted for use by consumers for personal communications on holidays and for special occasions. AT&T Chat 'N Talk also provides users with conferencing capacity.
  3. These services contribute to a long-term vision of integrated communications services that could include universal messaging services, instant messaging, directory services, and the like. These integrated communications services will involve voice telephony as well as Internet-based communication modes.
The technology that supports AT&T Chat 'N Talk is similar to what international call-back services use. With those services, the users call in to a special number, hang up, and are immediately called back by the service, at which point they will be connected to the party they want to call. The main difference between such services and the AT&T Interactive Communications services is that the Internet replaces that initial phone call as the means by which caller location and called party information are supplied. The Internet can then be used to collect billing information and provide the user with call status information. In addition, the use of the Internet enables capabilities like creating a private text-chat session only for people who are connected to the call, or supporting "social surfing" for those who are connected (a "leader" browses to a web site; the site is displayed on all the other users' browsers).
Whereas services like AT&T Click2Dial Conferencing are, at the simplest level, examples of services that extend what can be done without involving the Internet (all major telecommunications companies have offered conference calling services for years); services like AT&T Chat 'N Talk are uniquely suited to the world of the Internet.
Chat is an Internet phenomenon. Of course, there are 900-number services where people can have related experiences of talking to people they haven't met before, and there are probably ways to have private "side conferences" with many of them. But the range of topics is much smaller, and the cost of participating is much greater. In addition, the initial level of intimacy required is greater, since voice communication is involved. In contrast to 900-number services, a large variety of chat topics is offered on most portal sites, and users may participate for free (over the cost of an Internet connection), at whatever level they want.
According to some estimates, between 6.5% and 21% of the on-line population are regular users of chat rooms; up to 32% use chat occasionally (Card, McAteer, and Rubin, 1998; IDC, 1998). This indicates that there is a relatively small market for services like AT&T Chat 'N Talk at this time. But this application is one that could readily be extended to other contexts, such as Instant Messaging. One of the challenges of developing services for the Internet is guessing right in a fast-moving environment. At the time AT&T Chat 'N Talk was conceived and developed, Chat had a well-established base compared to Instant Message services, which were just beginning to appear on the Internet.
One of the lessons we have learned from our experience with AT&T Chat 'N Talk is the importance of integrating the call functionality into any other applications where it might be used. The remainder of this paper discusses the details of how a context-independent version of the service differs from an integrated one from the user's point of view.

Anonymous Voice Chat Breaks User Models of Initiating Communication Events

Across different communication domains, users experiences with initiating communication events is similar: Using a context-appropriate address, the initiator of the communication contacts the other party. This typical model of initiating communication events holds true in the telephone domain, the email domain, and the chat domain. In the telephone domain, the context-appropriate address is the telephone number. In the email domain, it is the email address; and in chat, the address is the chat handle, which allows a person to address comments to a particular participant within the chat room, or to send a private message. Of course, in each case, there is an implicit assumption that the "called" party has given the "caller" permission to contact him or her, either by explicitly providing the address, or by listing him or herself in a public directory, which allows the caller to find the address using an independent piece of information such as a name.
In this typical model of communication, the initiator is more or less in "control" of the communication session. The caller sends a message to initiate the session, the called party is alerted that the communication is there (via a new mail appearing in the mail box, an private message box popping up on the screen, or by the phone ringing) and may refuse to engage (by not responding to the email or private message, or by not answering the phone). Figure 2 displays the call flow of a typical phone call.
Figure 2:  Initiating A Typical Phone Call
Figure 2: Initiating a Typical Phone Call
In the chat room scenario participants are reluctant to give out their phone numbers. Because of this, a service like AT&T Chat 'N Talk can protect privacy only by violating the typical model of communication events (described above) in the following ways:
  1. The initiator does not have access to the context-appropriate address (phone number), since the "called" party does not wish to provide his or her number, and the caller cannot independently obtain the phone number (since, even though many chat services include directories or the ability to look at another user's profile, phone information is unlikely to be included there.)
  2. The first step in establishing the session is inviting the "called party" to participate.
  3. Each party is required to supply his or her context-appropriate address to an independent "agent" (i.e., the phone network).
  4. Since the addressing takes place indirectly (the middle contacts the ends), another variable is needed to make the agent tie the correct ends together, so a "session identifier" or "access code" must be created and used by both parties to the call. Thus, the access code must be provided to both parties and used to communicate with the service.
  5. By modifying the order of establishing the session (caller requests, recipient agrees, calls established in both directions and bridged), the call flow introduces an element of surprise for the "calling" party: at the end of this sequence of events, both the called and the caller s phone will ring. Figure 3 displays a generic call flow for an AT&T Chat 'N Talk call.
Figure 3:  AT&T Chat 'N Talk Flow
Figure 3: AT&T Chat 'N Talk Flow
Although this call-flow logic is required to establish an AT&T Chat 'N Talk call, the context in which it is initiated can make all the difference from the user perspective. In the version of the service that is independent of a particular chat provider, the user must complete all of the above steps, including obtaining the session identifier and communicating it to the "called" party. Communication about the call set up must take place as messages sent within the chat room (or instant messages associated with the chat client). In a version of the service that integrates more tightly with the chat client, much of the complexity of the flow could be hidden from the users, greatly simplifying the call flows. These issues are discussed in more detail in the next section.
AT&T Chat 'N Talk Usability Challenges
The major usability issues for the AT&T Chat 'N Talk service surround the violations of the typical model of initiating communication events. The contrast between initiating a typical phone call and initiating an AT&T Chat 'N Talk phone call highlights these usability challenges. Solutions to these usability challenges are described below.
Independent AT&T Chat 'N Talk User Experience
When the service is independent of the chat client, a chat room has a link that simply opens a browser to a web page that can trigger the call. The browser runs separately and independently of that chat room. This maximizes the complexity of the call set up, since important information about the call (the session identifier used to bridge the two endpoints) must be communicated between the participants, most likely through an instant message in the chat software. Figure 4 displays the call flow for an AT&T Chat 'N Talk call from both participants perspectives.
  1. The Host accesses a page allowing him to place a call. Access may be via a link within the chat client or via the AT&T Chat 'N Talk home page.
  2. The Host enters & submits his telephone number and billing information (i.e., credit card number and expiration date).
  3. The Host receives an Access Code.
  4. The Host sends this Access Code to his friend (perhaps privately via a private chat room or instant message).
  5. The friend accesses the web site that is independent of the chat site, and selects the option that will allow her to receive a call.
  6. The friend receives the Access Code from the Host and then enters & submits her telephone number and the Access Code.
  7. Both parties are informed that the service is adding the other to the call; the Host s phone rings and he answers; the friend s phone rings and she answers.
Figure4: AT&T Chat 'N Talk call flow for the version of the service that is independent of a particular chat provider.
Figure 4: AT&T Chat 'N Talk call flow for the version of the service that is independent of a particular chat provider.

User Experience for Integrated AT&T Chat 'N Talk

When AT&T Chat 'N Talk is tightly coupled with the chat client, the chat room integrates within the chat environment itself the ability to support both text-based and voice-based anonymous conversations. This integration facilitates the identification of individuals and the initiation of both kinds of conversation, since the person initiating the voice chat can use the other user's chat handle to initiate both the text and phone-based conversations, as described below.
  1. The Host presses a button (labeled "Phone" or "AT&T Chat 'N Talk") on the chat client.
  2. The Host enters his telephone number and billing information (i.e., credit card number and expiration date).
  3. The Host enters the chat name of his friend.
  4. The Host informs his friend to click on the Phone button within her chat client (perhaps privately via a private chat room or instant message).
  5. The friend clicks on the Phone button.
  6. The friend enters & submits her telephone number.
  7. Both parties are informed that the service is adding the other to the call; the Host s phone rings and he answers; the friend s phone rings and she answers.
The tight integration allows the Host to identify his friend by her chat name. He does not need to receive an "access code" from the service and then pass that code to his friend; she, in turn, does not need to receive that access code from the Host and submit it back to the service. Tight integration allows identification via an anonymous alias (the chat name) and maintains the mood of anonymity in the transition from text to voice. Users can do this within a chat room interface no separate browser window is required.

Usability Issues: Independent vs. Tightly Integrated AT&T Chat 'N Talk

There are two main usability problems with the independent call flow that tight integration resolves:
  1. The cost of leaving the chat environment to enter calling and charging information is high. The user is forced to confront multiple applications and must move information across those applications. Specifically, obtaining and communicating the access code is difficult, since many chat clients have limited ability to accept cut and paste information. Depending on how the host transfers the access code to the participant, they may have an instant message or private chat room visible in addition to the main chat window. In our initial design for the service, we displayed the URL for the web site and the access code on two lines. When the Host tried to copy the information into the messaging context, only one line would be copied, and often the user did not realize that important information had been omitted. Even though we solved this problem by having the host send the other caller to the web site at an earlier point in the process, there were still problems with this step. In addition, the large number of relevant windows (at least 2, and often 3 or 4, depending on the chat client) on the screen made it difficult to manage navigating the task.


  2. This complicated interaction takes quite a bit of time. In usability tests conducted on an earlier version of the software, it was not unusual for the host to take 3-5 minutes to complete the steps between asking the participant if he or she wanted to talk and coming back to the participant with the access code. Participants spontaneously sent messages to determine if the host was still there. Given the tenuousness of the interaction, the fact that the Host was off setting up the call for several minutes while the other party waited led us to predict that the number of actual call completions would be unacceptably small.
Since users may feel tentative about taking this next step toward intimacy, these problems may seriously affect the rate of use of the service.
By keeping the user's focus within the chat application and integrating the technology that allows the two true identities to be hidden and managed by the system, the complexity of establishing the call is significantly reduced.
One other feature has been added to the service to support repeated users. In its initial release, AT&T Chat 'N Talk was viewed as a "per call" service, and required the user to enter a credit card number each time they placed a call. Users may now enter their credit card number and password-protect it, so that each time they place a call, they need enter only the password, not the phone number and credit card.

Conclusions

AT&T Chat 'N Talk is an example of how the Internet stimulates the need for new and different kinds of communication services. It enriches the anonymous chat experience by allowing people to extend it into the voice domain. Users can take a step closer to their chat acquaintances without revealing all. AT&T s decision to work with chat software developers and chat room providers is an important step in creating an integrated, usable service.
In addition to providing a usable service, the success of AT&T Chat 'N Talk rests on the customers trust that their information will remain private. Communication companies like AT&T carry with them a history of trustworthiness. The Caller ID blocking and the lack of any part of the participant or host phone numbers appearing on the credit card bill confirms the users confidence that their calling information will be kept confidential. With services like AT&T Chat 'N Talk in which private information is kept private, users will be encouraged that they can evolve their internet relationships without risking personal safety.
What remains to be seen is whether, once users understand that their phone numbers will be kept private, they will in fact decide to take a step toward the next level of intimacy. The availability of such services make it possible to study the circumstances under which people may want to safely increase their level of intimacy. With services like AT&T Chat 'N Talk, we can begin to understand whether people move to more intimate communications media as soon as they can do so "anonymously" or whether the special characteristics of text-based chat make it, in some situations, the communications medium of choice.
In the age of Internet communications, the issues of privacy and trust have technical and human components. Looking for solutions to these communications issues is one of the most exciting areas of application development on the Internet today.

Acknowledgements

We would like to thank Nick Benimoff and Randy Pilc for their review of this paper and thoughtful comments. We would also like to thank Brian Kurtz for his hard work on the usability evaluation of the service during his summer internship at AT&T Labs. Finally we would like to thank Justin Bird for the colorful cartoon that appears in the paper.

References

Agre, P. & Rotenberg, M. (1997). Technology and Privacy: The New Landscape. Cambridge: MIT Press.
Benimoff, N. I. & Whitten, W.B., II (1993). Human factors issues in multimedia conferencing. In Proceedings of the Human Factors in Telecommunications International Symposium (pp. 227-236). Heidelberg, Germany: R.v. Decker s Verlag.

Card, D., McAteer, R., Rubin, R. (1998) Chat and Instant Messaging: Matching Messaging Technologies to Diverging Site Applications. Jupiter Communications Strategic Planning Services Report.

Cheskin Research & Studio Archetype/Sapient (1999) eCommerce Trust Study. Published by Cheskin Research. Available online at http://www.studioarchetype.com/headlines/etrust_frameset.html

Cranor, L. & Reagle, J. (1997, September) Designing a Social Protocol: Lessons Learned from the Platform for Privacy Preferences Project. Proceedings of the Telecommunications Policy Research Conference. Available online at: http://www.research.att.com/~lorrie/pubs/dsp/dsp.html
Cranor, L., Reagle, J., Ackerman, M. (1999). Beyond Concern: Understanding Net Users' Attitudes About Online Privacy. AT&T Labs-Research Technical Report TR 99.4.3.
Davis, G. (1998, March): From POTS to PANS: Transforming Plain Old Telephone Service to Pretty Awesome Network Services in the Age of the Internet. In Proceedings of the Voice on the Net (VON)Conference, San Jose, CA.
Dibbell, J. (1993). A Rape in Cyberspace. Village Voice, 12/23/93. Available online at: http://www.levity.com/julian/bungle.html
Dibbell, J (1999). My Tiny Life: Crime and Passion in a Virtual World, New York: Henry Holt & Company, Incorporated.
Dyson, E. (1997). Release 2.0: A Design for Living in the Digital Age. New York: Broadway Books.
Fussell, S. R. & Benimoff, N. I. (1995). Social and Cognitive Processes in Interpersonal Communication: Implications for Advanced Telecommunications Technologies.
Hafner, K. (1997, May). The Epic Saga of The Well: The World s Most Influential Online Community (And It s Not AOL). Wired, 5.05. Available online at http://www.wired.com/wired/archive/5.05/ff_well_pr.html
Hamman, R. (1999). The Online/Offline Dichotomy: Debunking Some Myths about AOL Users and the Effects of Their Being Online Upon Offline Friendships and Offline Community. Final draft version of MPhil Thesis, Department of Communication Studies, University of Liverpool. Available online at: http://www.socio.demon.co.uk/mphil/index.html
Johnson, S. (1997) Interface Culture: How New Technology Transforms the Way We Create and Communicate. San Francisco: HarperEdge.
Katz, J. & Aspden, P. (1997). Friendship Formation In Cyberspace: Analysis of a National Survey of Users. Project Report for The Markle Foundation. Available online at: http://www.markle.org
Katz, J. & Aspden, P. (1997). Cyberspace and Social Community Development: Internet Use and Its Community Integration Correlates. Project Report for The Markle Foundation. Available online at: http://www.markle.org
Kling, R (1996). Computerization and Controversy: Value Conflicts and Social Choices. San Diego: Academic Press.
Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukophadyay, T. & Scherlis, W. (1998). Internet Paradox: A Social Technology That Reduces Social Involvement and Psychological Well-Being? American Psychologist, 53(9), 1017-1031. Available online at: http://homenet.andrew.cmu.edu/progress/research.html
Kraut, R. E., Scherlis, W., Patterson, M., Kiesler, S. & Mukhopadhyay, T. (1998). Social Impact of the Internet What Does It Mean? Communications of the ACM, 41(12), 21-22. Available online at: http://homenet.andrew.cmu.edu/progress/research.html
Lin, D. & Loui, M. (1998). Taking the byte out of cookies: privacy, consent, and the Web. Computers and Society, 39-51.
Meeks, B. N. Electronic Frontier: The privacy hoax. Communications of the ACM 42, 2 (Feb. 1999), Pages 17 - 19
Rheingold, H. (1993). The Virtual Community: Homesteading on the Electronic Frontier. New York: Addison-Wesley Publishing Company.
Rheingold, H. (1998). Misunderstanding New Media. Feed Magazine (online). Available online at: http://www.feedmag.com/essay/es102lofi.html
Rosenberg, S. (1998). Sad and lonely in cyberspace? Why the new Net-depression study is something to get bummed about. Salon Magazine (online). Available online at: http://www.salonmagazine.com/21st/rose/1998/09/03straight.html
Schneier, B. & Banisar, D. (1997). The Electronic Privacy Papers: Documents on the Battle for Privacy in the Age of Surveillance. New York: John Wiley & Sons.
Schwartz, P. (1996, June) R-Tech: An Interview with Albert Bressand. Wired, 4.06. Available online at: http://www.wired.com:80/wired/archive/4.06/rtech.html
 

"AT&T Chat 'N Talk: Getting to Know You without Getting to Know All About You "
<- Back

Thanks to our conference sponsors:
A T and T Laboratories
ORACLE Corporation  
National Association of Securities Dealers, Inc.

Thanks to our conference event sponsor:

Bell Atlantic


Site Created: Dec. 12, 1998
Last Updated: June 10, 1999
Contact hfweb@nist.gov with corrections.