| April 15, 1997 Secretary Re: Consumer Privacy 1997 -- W3C REQUEST TO PARTICIPATE P954807 Question 2.14 & 3.14 regarding Technological Developments The World Wide Consortium (W3C[1]) requests to participate in "Session Two: Consumer Online Privacy" and "Session Three: Children's Online Privacy". We would like to present the structure of our Platform for Privacy Preferences (P3) project. This project is currently under member review (see attached briefing package.) In our efforts to prepare for this project we have been working with W3C members, such as the DMA, and speaking with policy makers so as to understand the relevant issues at hand. We have also been working with the Internet Privacy Working Group (IPWG[2]) to develop a grammar and vocabulary for describing privacy practices.[See REQUEST TO PARTICIPATE P954807 AS IPWG MEMBER] There is a constant tension in commercial media between a desire by the public for privacy and the desire for information about the viewing audience. As the Web continues to mature as a commercial media, it is particularly prone to this dissonance because of its interactive nature. Users often wish to provide information so as to customize their experience without forfeiting all privacy. Services wish to understand the characteristics of their users, oblige the users' preferences, and comply with relevant regulations. To address this tension, the W3C has recently proposed a Platform for Privacy Preferences (P3) project to its membership. A component of this project will be work W3C staff have participated in as members of IPWG. This work will enable computer users to make choices about the collection and use of their personal information on the Web. IPWG has been working to create a privacy "vocabulary" and "grammar" for expressing site practices and user preferences. The W3C plans to create a machine encodable and readable mechanism (like PICS) that will allow sites and users to communicate their privacy practices and preferences respectively. Other aspects of the W3C project will include the ability to exchange privacy practices and preferences in a flexible ("negotiated") manner; and addressing issues related to http, cache, and cookie technologies and their impact on user privacy. We are including a copy of the W3C-P3 project briefing package that provides further context and details on the scope of our proposed activity. Sincerely, Jim Miller Joseph M. Reagle Jr. Ralph Swick [1] The World Wide Web Consortium exists to realize the full potential of the Web. The W3C was founded in 1994 to develop common protocols for the evolution of the World Wide Web. We are an international industry consortium, jointly hosted by the Massachusetts Institute of Technology Laboratory for Computer Science [MIT/LCS] in the United States; the Institut National de Recherche en Informatique et en Automatique [INRIA] in Europe; and the Keio University Shonan Fujisawa Campus in Asia. [2] IPWG is an informal organization of public interest organizations and private industry engaged in commerce and communication on the Internet. It is coordinated by the Center for Democracy and Technology Attachment: W3C Briefing package for Platform for Privacy Preferences (P3). Briefing Package for Platform for Privacy Preferences (P3) Project 2 April, 1997 1. Executive Summary The proposed P3 Project will address the twin goals of meeting the data privacy expectations of consumers on the Web while assuring that the medium remains available and productive for electronic commerce. Following the principle of providing consumers notice of site privacy polices and choices, and allowing users to express and act upon their privacy preferences, one goal enhances the success of the other. The potential for electronic commerce is immense, some predict the results will be revolutionary. However, commerce on the Web is at risk of regulatory restrictions for lack of technologies that effectively address the personal privacy concerns raised by consumer advocates and in some cases already embodied in national or local law. Hence it is essential that we enable an environment in which consumers feel comfortable about their data privacy. More practically, governments also understand the potential benefits and concerns inherent in ubiquitous on-line transactions. Through this project, the W3C has the ability to offer solutions in this domain. This in turn will remove a major impediment to users using the Web for sophisticated transactions. Users will understand the data protection policies of a service provider and will make informed choices about the release and dissemination of their personal data. This will help users feel secure in their transactions, and enable selective, customized, inexpensive, or otherwise improved services for the user. Those conducting electronic commerce on the Web will have access to better information with which to make more competitive and relevant offers. The P3 project will (1) create a vocabulary and encoding method for expressing privacy policies; (2) define transport and negotiation protocols for such policies; and (3) address the discrepancy in current proxy and cache technology between network efficiency and the collection of accurate Web site usage information. This briefing package details the context of the P3 project, what we hope to achieve, as well as the expected planning and management of the project with respect to time, resource, and external funding. 2. Background The Web enables information to be simultaneously dynamic and personalized; a welcome turn away from info-glut. However, for information to be customized, personal data and preference information need to be communicated by the user, and understood and honored by those conducting electronic commerce within a global framework. We observe a correlation between customized service delivery and targeted marketing: as the value of customized information services increases for the user, so does the value of targeted offers for the offeror. One indicator of this correlation is that many services have been offered without charge and users have been given enhanced capabilities without additional fees. For instance, in the past year we have seen a number of services appear. Juno (http://www.juno.com) offered free e-mail to its users, and Cyberfreeway (http://www.cyberfreeway.net/) offered free unlimited Internet services; both chose the US commercial television revenue model based on advertising and take advantage of the ability to collect and use detailed personal profiles completed by their subscribers. Additionally, Freeride (http://www.freeride.com/) credited customers' accounts at participating ISPs for selected purchases and for direct interactions with advertisers. However, as the value of customized information services increases for the user, so does the potential for the abuse of the associated personally identifiable data. As a result, policy makers have expressed interest and concern about the new media, particularly with respect to children. The Bureau of Consumer Protection of the US Federal Trade Commission convened a workshop (http://www.ftc.gov/bcp/privacy/privacy.htm) on Consumer Privacy on the Global Information Infrastructure in June 1996. The European Commission has issued a directive on "the protection of individuals with regard to the processing of personal data and on the free movement of such data" [95/46/EC, Eur. O.J. L281/32 (Nov. 23, 1995)]. In the report of the June FTC workshop (bcp/conline/pubs/privacy/privacy1.htm) it is noted that
[page 1] There are numerous guidelines in existence on fair information practices; see, for example, the OECD "Guidelines On the Protection of Privacy and Transborder Flows of Personal Data", available at http://www.oecd.org/dsti/iccp/legal/priv-en.html. The W3C's role is to help its members create the base technologies on which they and others can offer additional services. Through the P3 project the W3C intends to provide mechanisms for enabling notice and privacy preferences while improving network efficiency in light of how caches and proxies are often used. This project does not expect to be competing with any of the existing stakeholders (see http://www.w3.org/pub/WWW/Privacy/Overview.html) but rather enabling them. All stakeholders have a role in supporting robust and consistent networking protocols and the policies which govern them with respect to control (data protection) and dissemination (collection, auditing). Search engines will benefit from the access to richer metadata. Users can include their privacy preferences in a query which the engine can use when returning potential sites of interest. Services such as eTrust (http://www.etrust.org/) are promoting a market brand as a reputable and trusted third party that assures merchants adhere to their stated privacy practices; such services are complementary to our activities. Marketers wish to provide offers viewed as valuable, timely, and reasonably priced to a segment of the market most likely to be receptive. The P3 project will provide a means for marketers to express their data collection and use practices so consumers will participate. The technology vendors have a direct interest in the work of this project. A critical step to the reduction of risk for commercial use of the Web is a consensus framework in which users feel they can use the Web safely. Hence, it is our analysis that we are not directly competing with any current market stakeholder or standards body, that we will provide a significant benefit to our membership, and that additional companies will consider becoming members to participate in this project. 3. Current W3C Status/Position This is an activity for which we will seek additional funding. MIT held a workshop on the use of Internet survey and data logging technology in January of 1996. This Project is one of the recommendations of that workshop. Joseph Reagle of the W3C staff has been actively involved in the Internet Privacy Working Group, hosted by the Center for Democratic Technology. 4. Proposed P3 Project Introduction and Scope of the Activity We propose a Project to create mechanisms for data privacy assertions, preference expression, flexibility/negotiation between users and sites, and the improvement of caching and proxy technology. The Project itself consists of an Interest Group, a Coordinating Group, and several Working Groups. Structure of Project The P3 project will initially consist of four working groups and one interest group, supervised by a Coordinating Group. The Coordinating Group will initially be staffed by the W3C Project Manager, W3C Policy Analyst and Working Group chairs. Additional Coordinating Group members will be nominated and voted once the project is underway. The Coordinating Group may create or disband working groups according to the W3C Process as the project progresses. Per W3C Process, Advisory Representatives will receive notice of formation of Working Groups and each Working Group will be tasked to generate a status report at intervals no greater than every three months. Membership in the Interest Group is open to any W3C member organization. No specific resource commitment to the Interest Group is required. Membership in each Working Group is restricted to W3C members who make a commitment of resources. The Working Groups will conduct their business by e-mail. Each Working Group is free to schedule face-to-face meetings, not to exceed once a quarter. The W3C Project Manager is charged with scheduling these meetings in a way that minimizes their expense and maximizes their productivity. The initial Working Groups are proposed as follows: Vocabulary and Expression Group This group will describe a vocabulary in which privacy policies and preferences may be expressed and provide a grammar for expressing this vocabulary. Hence, a key activity is to understand what categories (personal, hobby, professional, etc.) or pieces of data (name, address, email, click-stream data, etc.) are required and the actions to be taken on that data (collected, used for internal use, redistributed, etc.) Additionally, the preference vocabulary may include the ability to specify policies about the management of data stored on the client side. This could facilitate the use of pseudonyms and the control of cookies. Technical methods for implementing this will occur in the Networking and Client Side Data Control group. This W3C working group will benefit from examining the Internet Privacy Working Group's previous work on this topic and may work jointly with IPWG. The specific relationship with the IPWG will be decided jointly by W3C and IPWG. Policy Encoding, Transport, and Negotiation Group This protocol-oriented group will create the means of expressing policies in the language developed by the vocabulary group, transporting/distributing those policies over the Web, and ensuring that such policies are machine readable, consider anonymous (or pseudo-anonymous) interactions where appropriate, and enable negotiation or flexibility between users/clients and services/servers. This work will be coordinated with work in the PICS, DSig, and XML working groups. Networking and Client Side Data Control The network behavior group will focus on the relationship of caches and proxies to the collection of access, hit, and impression data for server administrators. They may also investigate issues related to client side data storage and control (collaborative filtering preferences, cookies, etc.). Currently caches and proxies do not fulfill the requirements of administrators/marketers and are consequently being used in ways which degrade network performance. (For instance, even static data is often not cached because of concerns over accurate hit information or concerns of intellectual property protection.) Coordination between this working group and work in the W3C Architecture Domain will be handled by W3C staff with the help of the respective Coordinating Groups. Prototyping The prototyping working group will create proofs of concept to demonstrate the mechanisms produced by the Policy Encoding and Vocabulary and Expression Working Groups. [...] Projected Schedule The following proposed schedule is predicated upon receiving sufficient funding by May 16.
5. Annex Intellectual Property This project does not expect to depend on any pre-existing intellectual property other than related W3C recommendations. Any new specifications produced in the P3 project will be made freely available as W3C documents. Joseph Reagle reagle@w3.org
|