August 2000 Public Policy Computer Graphics Column

Introduction

Bob Ellis

As usual, the column is multi-topic. It starts with a brief review of the SIGGRAPH Public Policy Program. This is followed with some comments (and my responses) on the program raised by an interested SIGGRAPH member. Once again, Myles Losch presents recent information on digital copy protection, including its potential application to free terrestrial broadcast digital TV. We provide a review of the 2000 edition of one of the most important conferences at the intersection of computing and public policy, the ACM sponsored Computers, Freedom and Privacy Conference (CFP). And finally, we describe the availability of our third on-line survey.

The SIGGRAPH Public Policy Program

Bob Ellis

The mission of the SIGGRAPH Public Policy Program is to provide visibility of relevant public policy information, such as proposed legislation, to SIGGRAPH members and others in the computer graphics technical community and to provide information on the implications of technology to the non-technical community, including policy makers and funding agencies. We do so by this column, our web site (http://www.siggraph.org/pub-policy), serving as a catalyst for studies on computer graphics research topics, participation in conferences addressing policy issues, conducting issues surveys, writing white papers and working with ACM public policy groups, such as USACM.

I have some excellent assistance. Myles Losch provides us with expertise in the areas of telecommunications policy and digital aspects of intellectual property issues. David Nelson is our volunteer webmaster. Laurie Reinhart provides editorial support and assists David with the website.

We have developed special emphasis on several issues. In our most far-reaching project, we have been working with the computer graphics research community on defining a study of computer graphics research topics.

We have placed special emphasis on the interaction of computer graphics and the Internet. Our first (and only to date) white paper was on computer graphics and the Internet; copies may be found on our web site. Believing that broadband Internet access is a necessity for successful computer graphics on the Internet, we have commented extensively on the technology and issues associated with services such as Digital Subscriber Line (DSL) and cable modems (http://www.siggraph.org/pub-policy/CGColumn-0599.html). Long before the popular press commented on the security and privacy aspects of these services, we were alerting SIGGRAPH members (http://www.siggraph.org/pub-policy/CGColumn-1199.html).

Computer graphics and intellectual property (IP) issues have also been important. In particular, digital copy protection schemes in use with DVD technology and proposed for digital TV, strengthen the rights of owners of IP and may limit the rights of consumers under fair use concepts and home recording rights.

A Member Comments

Bob Ellis

I was asked recently by a SIGGRAPH member why we weren't working on the digital divide issue. The digital divide is a phrase coined to highlight the situation where certain groups of people are underrepresented as computer and Internet users. Many dimensions have been proposed for the digital divide: economic, racial, education, gender, location and usability are some of them.

The primary reason I have not spent time on this issue is that all the dimensions, except usability, are primarily social in nature and there is little to be added to the discussion by technical experts operating as technical experts. Another reason for not working on this one is that it is well covered by the general press and SIGGRAPH members can certainly access that information and respond as citizens.

On the usability issue, I have been working with the organizers of the ACM Conference on Universal Usability (http://www.acm.org/sigs/sigchi/cuu/) which will specifically include a policy component. The conference is in cooperation with SIGGRAPH. Quoting from the Call for Participation:

"The start of the next millennium can be characterized by progress in computing and communications technology. Cheaper and faster processors, storage, and networks combined with better user interfaces have now spawned the incredible growth of the Internet and related services. But too often system complexity, incompatible software or file formats, poorly crafted interfaces, and inadequate attention to diverse users leads to confusion, frustration, and failure. It's time to address this challenge."

"We invite submissions for the first ACM Conference on Universal Usability, to be held in Washington, D.C., November 16 and 17, 2000. We seek work whose aim is to enable the widest range of users to succeed in their use of technology for information, communications, entertainment, education, e-commerce, and community and government services. Challenges include the diversity of users (experts & novices, old & young, educated & illiterate, disabled, forgotten, those in ill health, etc.); the wide range of technology (e.g., 100:1 ratios in processor and network speeds); and the gap between what users know and what they need to know."

"We are interested in research, new systems and technologies, empirical evaluations, policy suggestions, and systems that support community activities. The diverse set of participants will include researchers, technologists, policy makers, advocates, and users."

For readers further interested in the digital divide, there are a number of on-line resources. Among these are (http://www.digitaldividenetwork.org/). In reviewing a recent Stanford study (http://www.stanford.edu/group/siqss/Press_Release/press_detail.html), Jakob Neilsen wrote (http://www.useit.com/alertbox/20000220.html):

"The study's analysis of the digital divide is credible because issues like race, education, and age are precisely defined and can be reported very accurately in a survey as long as the respondent feels comfortable that the survey is being administered by a credible institution (Stanford would certainly count here) and that the answers will be treated anonymously."

"When splitting out the effect of the various variables, the study finds the following three main effects on Internet access:

1. Education (having a college degree): +49%
2. Age (older people compared with 18-25 years' olds): -43%
3. Income (having high income): +21%"

"My interpretation of this finding is that the digital divide is a usability problem. The politicians are targeting the wrong part of the problem when they treat the digital divide as an economic issue. True, there is a (smaller) problem due to the expense of computers, but this third-level problem is rapidly vanishing and will be completely gone in a few years when computers will cost the same as donuts." [Also tending to shrink the problem is better access via libraries and schools, both to computers and to gain experience in their use. - B.E.]

"Old people will not go away. In fact, people who are currently in their 40s and 50s will be around for a long time to come. We can't simply write them off just because kids have fewer problems using computers. The same is true for people without a college education. We can't force them all to go back to school for four years simply in order to participate in society."

The Stanford study itself says:

"21 percent of differences in Internet access can be explained by demographic factors. By far the most important factors facilitating or inhibiting Internet access are education and age, and not income - nor race/ethnicity or gender, each of which account for less than 5 percent change in rates of access and are statistically insignificant. By contrast, a college education boosts rates of Internet access by well over 40 percentage points compared to the least educated group, while people over 65 show a more than 40 percentage point drop in their rates of Internet access compared to those under 25. Age really reflects generational differences, and thus shows what to expect in the future."

Another columnist, Neal Pierce, also claims that the Digital Divide is an issue of usability (http://www.postwritersgroup.com/archives/peir0221.htm).

Another take on the issue comes from The Center for the New West (http://www.newwest.org/). Their study shows that location is a major factor with rural areas, small towns, inner cities and some suburban areas having distinctly less than first class Internet access.

Perhaps one of the most perceptive comments I've heard recently came from the producer of the PBS series on the digital divide. He said, in effect, that it was a complex, multi-dimensional problem, where for example, race was an important factor at lower income levels, but was essentially not an issue at higher income levels.

This SIGGRAPH member also asked about our apparent lack of activity in influencing policy makers. While we are in the process of defining a project with the National Research Council in computer graphics research, we have indeed not been very active in this area. Primarily we've worked with others by participating in last year's USACM Congressional briefing and organizing sessions at policy oriented conferences such as the CFP (see below).

There are a couple of reasons. First, I believe that it is inappropriate to take positions on issues on behalf of SIGGRAPH members; therefore our role is primarily supplying information on the technical implications of policy decisions. Policy makers are open to information on a subject, but the timing of opportunities for this are typically less than obvious and often precede open public discussion. Timing is everything and that requires active staff work monitoring the activities of policy makers. SIGGRAPH has not committed any human or financial resources for this task and even ACM has a very small presence in Washington DC and none in any other capital.

Second, my experience is that policy makers typically want to hear from constituents, contributors and representatives of organized groups. The first two do not include us, except as individuals and the last requires the kind of staff work and presence which we don't have as described above. Just like efficiency dictates that we have a representative form of government, it also means that our policy makers need to hear from representatives. At CFP2000, one U.S. government official mentioned that a recent call for comments resulted in over 60,000 email responses, several dozen of which were over 100 pages long. While these responses represent a wealth of information, even the most well staffed organization cannot do justice to such a massive amount of information.

A few years ago, several of us organized a briefing on computer graphics at one of the Los Angeles SIGGRAPH conferences. We were realistic in our expectations and concentrated our invitations on U.S. government representatives with strong business, scientific and California connections, California State government people and non-United States scientific liaisons in Los Angeles consulates. Out of these hundreds of people, many of whom received follow-up telephone calls, we had less than 5 acceptances and cancelled the briefing. The problem probably was that at that time, very few of our invitees had immediate need for our information. Again, timing is everything. It is of course possible to get policy makers to listen to you at any time, but doing so when the topic is not a current agenda item, requires a lot of work, again as described above.

Digital Copy Protection and Terrestrial Broadcast Digital TV

Myles Losch

Legal and policy disputes over protection of copyrighted works continue to slow the growth of digital television services, an important potential market for computer-generated imagery. Our last column noted a sharp increase in copyright lawsuits by music and movie suppliers, as this trend continues.

Many of the issues are the same for music as for visual content, and seem likely to be addressed sooner for audio, as we explain below. Graphics specialists would thus do well to follow closely the current "music wars."

Contributing to music's importance as a policy trendsetter are several factors:

1) Compressed audio can be transmitted and stored (at acceptable quality) using far fewer bits than video of equal duration.

2) Most popular music consists of brief songs, much shorter than typical video and filmed entertainment. This further reduces the technical difficulty and cost of handling such audio content.

3) A typical consumer's in-home access to the Internet will, for some years to come, be through relatively slow dial-up phone modems. Thus, online transmission is today more attractive (in general) for music than for visual entertainment.

One technically interesting byproduct of the litigation noted above, is a new generation of 'serverless' software for Internet distribution of audiovisual and other content. These decentralized systems (e.g. Gnutella) seem designed to present few attractive targets for potential copyright lawsuits.

But given the widespread infringement attributed to such products, the effectiveness of this strategy remains to be seen. And as for the ongoing court cases, few have yet produced clear policy results.

One possibly relevant exception, though from cases unrelated to copyrights or other intellectual property, is the view of a growing number of U.S. federal appeals judges that computer software is a form of free expression entitled to strong constitutional protection. If upheld in future decisions, this result could impede attempts to suppress controversial programs, or to mandate the features of software (especially freeware).

For more policy materials on digital copy protection (particularly of music, but also relevant to visual works), readers may find these online analyses helpful:

a) (From an Internet news service) http://news.cnet.com/news/0-1005-201-1757865-0.html
b) (From Newsweek magazine) http://www.msnbc.com/news/413376.asp
c) (by Harvard law professor William "Terry" Fisher) http://www.law.harvard.edu/Academic_Affairs/coursepages/tfisher/Music.html

We turn now to the regulatory arena and the FCC's efforts (detailed in some of our past columns) to aid the growth of digital television in the U.S. Despite modest progress toward resolving voluntarily the copy protection dispute between movie studios and makers of TV sets and VCRs (whose position is at http://www.hrrc.org), the Commission continued preparations to impose technical standards for connecting digital TVs to digital cable services. (More on this in future issues.)

Terrestrial Broadcast Digital TV (DTV)

The FCC's industry advisory panel on DTV (which developed the technical standards now in force) decided to revisit the issue of radio signal formats for terrestrial DTV broadcasts. This step, while not implying any future change, keeps the door open for such a result should field experience refute equipment makers' claims to have solved the problems previously noted in this space.

CFP 2000

Bob Ellis
Myles Losch

We have previously reviewed CFP conferences and our role in them (http://www.siggraph.org/pub-policy/CGColumn-0598.html and http://www.siggraph.org/pub-policy/CGColumn-0899.html). We had a major presence at this year's conference jointly proposing two sessions. Ellis led the proposal for and chaired a plenary session on Security and Privacy in Broadband Internet Services, which is reviewed below. Panelists included representatives from DSL and cable Internet providers, consumer ISP activities and research. Because the conference was held in Toronto this year and to highlight the special situation in Canada (cable Internet providers must provide open access to all ISPs), two of the panelists were Canadian.

Myles Losch proposed a session on the open access cable Internet issue and then combined it with another session proposal to provide an all Canadian session on The Broadband Internet and Free Speech (more fully described below). Speakers were asked to address the rationale for, and free-speech implications of, Canada's open (or equal) access policy for cable modem service; and the speech inequalities that broadband Internet service can create, especially for small or poorly funded speakers. A University of Toronto professor chaired the panel, which included an expert faculty colleague; key regulatory officials from the telephone and cable television industries; and a top government regulator.

The CFP2000 website (http://www.cfp2000.org) has an excellent description of the program. Many of the participants supplied papers that are also there. Finally, there is a good set of references to press coverage before and after the conference. Because of this we will focus our comments here on issues we feel are specifically of interest to readers involved in computer graphics.

The session Security and Privacy in Broadband Internet Services is well reviewed by (IDG) and technophilly (http://www.technophilly.com/NEWS/N-0108-broadband.html). While generally agreeing that the broadband services present particular problems, the speakers preferred to speak of the issue in the broader context saying that it was part of a much bigger issue. Users have some control by limiting the information they provide and limiting risky behavior such as providing private and confidential information at publicly accessible terminals and workstations.

Users of broadband services are also particularly vulnerable because they tend to use it for a range of activities people previously did offline. Several panelists said that user education was very important, both for protection and to receive better products. Service providers can also provide assistance such as blocking netBIOS protocols. Most speakers also said that security and privacy are already market differentiators, such as the ability to provide virtual private networks (VPNs). However, one speaker said that the generally unsophisticated user community means that it is problematic to leave security wholly up to the marketplace. The session is available on audio and videocassette tapes through the conference website.

The session on broadband free speech was expertly chaired by Prof. Andrew Clement of the University of Toronto, and featured a wide-ranging, interactive review of numerous policy issues affecting Canada, the U.S., Europe, and other regions. Panelists included Prof. Clement's colleague Liss Jeffrey (an expert on Marshall McLuhan's work); Bell Canada's chief regulatory officer, Sheridan Scott; Christopher Taylor, the top legal and regulatory executive at Canada's cable television trade association; and David Colville, Canada's senior regulator of non- broadcast telecommunications (at the CRTC in Ottawa).

Available on audio cassette through the conference website, the well-attended session covered Canada's cable television open-access policy (for independent ISPs), end-user pricing models, speed asymmetries between up- and downstream traffic, exclusive content deals, parallels between European, North- and South American regulatory debates, and many other policy issues affecting broadband Internet service.

One other session we feel may be of particular interest to computer graphics professionals is the one on the circumvention of digital copy protection mechanisms in products such as DVD players, set top boxes and digital television sets. This panel was organized by the Electronic Frontier Foundation (http://www.eff.org), which is funding the defense against the movie industry's anti-DeCSS lawsuits. Presenters including law professor Paul Schwartz, computer journalist Declan McCullagh and ACLU official Barry Steinhardt analyzed the anti-circumvention provisions of the 1998 Digital Millennium Copyright Act (DMCA). Joining with other legal and technology experts in the audience, they identified many vague and/or problematic features of the DMCA, and in a lively debate dissected the cases that have been initiated under the new law.

Opinions of the DMCA heard in the session were strongly negative, indicating that long and intense conflicts lie ahead in this area of law and policy. The National Research Council study reviewed in our May 2000 column provides helpful background information on this subject, and is recommended for readers wishing to learn more. Related CFP2000 sessions held earlier on the same day (and referenced on the conference website) were an intellectual property panel led by Prof. Pamela Samuelson, and a luncheon address by Prof. Jessica Litman on "The Demonization of [copyright] Piracy."

Illustrating the variety of other interesting sessions at CFP2000 were a highly polarized panel debate on the Bertelsmann Foundation's Internet content rating and filtering proposal and a review of criminal justice topics related to information technology. On the latter panel, Jim Dempsey of the Center for Democracy and Technology in Washington, DC (http://www.cdt.org) reviewed in depth new U.S. Justice Department surveillance proposals, including a major expansion of investigative wiretapping.

A novelty at CFP this year was the daylong Workshop on Freedom and Privacy by Design. Over a hundred experts, believing that the Internet's open, democratizing character is grounded in its technical architecture, gathered in hopes of technologically reinforcing the 'Net against perceived threats from political, commercial and other interests.

Since details of the day are on the conference website, only a summary assessment is offered here. Of the three main workshop themes, the least focused was that of encouraging responsible online business practices (though interesting ideas were presented by several researchers).

More promising was the half-day review of alternatives to the domain-name system (DNS) for online navigation. Since it is an optional overlay to the 'Net's basic numerical addresses, some workshop participants thought that efforts (e.g. by trademark holders) to restrict disfavored speech by using DNS as a chokepoint will only spur the growth of alternatives. If (as some think) tiny wireless devices become a popular means of 'Net access, non-DNS navigation may be further encouraged by the ergonomic limitations of such hardware.

The final workshop theme covered cash-like alternative payment methods for consumer e-commerce. Here, the profit motive will spur growth of new systems, as e-merchants reach out to groups whose use of credit cards is low, e.g. teenagers, the poor, and non-US customers. Some payment options will likely use non-personalized digital bearer instruments, in what has been called the vending-machine model of online retailing. As with physical vending machines, customer privacy would be enhanced.

Third On-Line Survey on Public Policy Issues in Computer Graphics

Bob Ellis
David Nelson
Laurie Reinhart

As part of the ongoing mission for the SIGGRAPH Public Policy Program we are hosting a new online survey regarding general policy issues. We would like to gather as much information and opinions as possible from SIGGRAPH members and other professionals in Computer Graphics and Interactive Techniques about what public policy issues are most important to them and the industry as a whole.

The survey should only take 5 minutes to complete and is located at: http://www.siggraph.org/cgi-bin/pp/survey.cgi (survey closed 03/2001).

Our first two online surveys asked questions relating to technical aspects of computer graphics (results are published in Computer Graphics November 1999 issue, available on our website, http://www.siggraph.org/pub-policy/). These first two surveys were sort of test cases for our real purpose, which is to ask questions related to public policy.

This survey asks questions related directly to public policy, specifically those issues that affect the Computer Graphics community as well as questions which pertain to the public policy activities of ACM and SIGGRAPH. As part of these surveys we are also asking for optional information from our participants about their profession and location in order to gauge our data more accurately. We respect our respondents' privacy, so these questions are optional.

The results will be tabulated and posted on our website and subsequently in the February 2001 Computer Graphics column. We value your time and appreciate your interest and participation in our activities and surveys.