True to their word, the ARDG co-chairs have not engaged in any "editorializing", nor in any attempt to select, recommend, or endorse any proposed technology. They simply collected the views of the various participants, recounted the history of ARDG's meetings, stapled them together, and called it a day. The report does not reach any conclusions or make any recommendations.

The ARDG co-chairs have received 10 expressions of interest. Four of these are VBI-based (that is, they provide data in a publicly-documented place and a publicly-documented format in the video signal), and the other six appear to be watermark-based.

The proposers and proposals are:

1. Consumer Electronics Association: CGMS-A (VBI)


2. VEIL Interactive Technologies: CGMS-A+ (VBI plus visible mark)


3. Visionaire Technology Corporation: a "patent-pending" line 21 VBI technology


4. Macrovision: an unspecified VBI technology that also uses the horizontal blanking interval


5. Philips: two video watermark proposals based on detection by display devices


6. Kenwood Corporation: a video watermark proposal


7. Dwight Cavendish Systems: RightsMaster (a video watermark proposal)


8. Toshiba: HIDE (a video watermark proposal previously submitted to DVD CCA)


9. VWM Companies: a video watermark proposal

These technologies will be the subject of presentations at the next ARDG meeting on Oct. 22 and 23 in Los Angeles.

The ARDG held a meeting in the Washington, D.C., area on July 30. At the meeting, the "analysis matrix" -- now a kind of questionnaire for proposal submitters -- was completed, along with the "request for information" (RFI) seeking technological schemes. Although the RFI is complete, it hasn't been published on the ARDG web site yet.

ARDG will seek technology proposals until October, and will hold a two-day meeting in Los Angeles late in October at which proponents may make presentations and be subject to questions. That meeting will allocate just 90 minutes per proposal to presentations and questions and answers.

As we noted in an earlier post, EFF's proposal to include language in the RFI to protect independent reviewers was rejected at the meeting. It is still not clear how the ARDG will deal with the fact that many claims made by vendors will simply be wrong. This ARDG meeting reiterated that ARDG does not plan to perform any kind of technology review process. What, then, will set the final report apart from the results of printing out a sample of watermark vendors' web sites, or binding their glossy brochures together?

In his presentation to the ARDG, Scott Craver concluded that "the state of the art [in watermarking] favors analysis" (that is, attacks). Thus, collecting vendor claims about watermarks without either performing or providing for review seems irresponsible.

As Richard Feynman wrote in Appendix F to the Rogers Commission report, "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled."

The vast majority of the meeting was spent dickering over the precise working of various questions in the “analysis matrix.” (For more about what this “matrix” is all about, read the entry prior to this one.) The ARDG co-chairs do not want the matrix published until it is finalized. Imagine roughly 30 or 40 questions like this one:

9.2 Are the contractual requirements affecting the technical aspects of the proposal, if any, completely described a priori in non-confidential documents?

The ARDG co-chairs hope to complete the matrix and invite responses from vendors by early August. Vendors will be invited to respond in writing by September 15 and invited to attend and answer questions at the October 1 ARDG meeting in Washington DC.

So, what will the Final Report look like? Well, it hasn’t been finally decided yet (we’ll discuss it at the next meeting, set for July 30 in DC). It will, at minimum, include the self-reported responses of various technology vendors in response to the questions posed by the analysis matrix. Of course, that will amount to little more than a stapled collection of spec sheets and marketing literature.

Michael Epstein (Philips) suggested that there would be an opportunity to ask vendors to clarify any incomplete or ambiguous responses at the October meeting. He quickly stressed, however, that the ARDG is not empowered to “grade” the technologies against each other. So there will not be any probing evaluation and testing of the technologies. But there may be something a bit beyond the self-reported information, if only “comments and clarifications.”

I’m still not at all sure what this final report will be used for. Perhaps it will be submitted to policy-makers, so that various partisans can advocate for technology mandates premised on technologies described in the report? (“Mr. Commissioner, we strongly believe that you should issue regulations requiring everyone in the marketplace to implement technologies 12, 32, or 46 in the ARDG Report.”) I just don’t know. Stay tuned.

The ARDG is in the process of drafting an “analysis matrix” to assemble information about available technologies that might be used to address the “analog hole.” The June 17 meeting (and probably the next several upcoming meetings, as well) was largely devoted to defining the “scope” of the issues to be addressed in the “matrix,” then bickering over the precise language for the items in the “matrix.”

What, you may ask, is this “analysis matrix”? (Please, no red pill/blue pill quips.) Well, as those of you following along are well aware, the MPAA’s stated goal in ARDG is to find some technological measure that will embed “rights signaling” into analog signals. This “rights signaling” information will then be detected by devices upon reconversion of the content into digital form (which, in turn, will presumably enable the re-inscription of digital content protection on the new digital copy). In other words, the MPAA wants to find some way of preserving content protection states when digital content is converted to analog and passed over analog links.

The “analysis matrix” consists of several dozen questions regarding the technical capabilities of available “rights signaling” technologies. Once the “matrix” of questions is completed and embraced by ARDG, technology vendors will be invited to respond to the questions, thus vetting their own “rights signaling schemes.” The outcome will look like those big product review charts in computer magazines, comparing products by various performance criteria.

Interestingly, the major industry groups at ARDG appear to agree that this collection of self-reported information will constitute the final ARDG product. At least that’s what I thought I heard Brad Hunt (MPAA), Jim Burger (Computer Industry Group) and Seth Greenstein (5C Companies) agree to in the meeting. No consensus recommendation of any particular technology, no probing evaluation by the ARDG of any of the technology claims made by vendors.

Just a stapled together collection of vendor-provided marketing hype about the characteristics of various watermarking schemes. Doesn’t seem like much, does it? It remains unclear to me who this ARDG final report will be sent to, or for what purpose.

The May 7 ARDG meeting in Washington, D.C., featured a presentation from Scott Craver, a PhD candidate in the Electrical Engineering department at Princeton University. Scott discussed the state of academic research on embedding and attacking watermarks, discussed the process of research, and mentioned several specific statistical attacks. He also noted that there are general attacks on watermarking systems which do not require attacking the watermark itself, and mentioned several attacks which attempt to conceal the presence of a watermark from a detector without removing the watermark itself. He emphasized the need for peer review of watermarking technologies, and concluded that the current situation "favors analysis".

Alan Davidson of CDT gave a presentation about consumer requirements in which he suggested that consumers' rights must not be down-rezzed (here, a joking term for "rounded down"). This led to an interesting exchange about the "rounding problem".

Suppose a user receives a movie to which a movie studio has attached a policy such as "View for 48 Hours". If an analog marking system can only express four policies -- "Copy Never", "Copy Once", "Copy No More", and "Copy Freely" -- there is no policy which faithfully reflects the movie studio's original policy.

One could then ask which of those four states is the most faithful to the original policy. "Copy Never" is somewhat more restrictive than the original, and "Copy Freely" is somewhat less restrictive. Alan argued that it was most appropriate to round up in order to ensure that consumers never lost out as a result of an analog conversion. In effect, motion picture industry representatives who opposed him argued that the reverse was true -- that it was most appropriate to round down in order to ensure that movie studios never lost out. I was troubled to hear that the movie studios appeared to believe that a policy of rounding down was somehow logically necessary or logically preferable to a policy of rounding up. As far as I could tell, "Copy Never" (more restrictive) is just as inaccurate a paraphrase of "Copy for 48 Hours" as is "Copy Once" (less restrictive). It's simply that one choice is more protective of publishers and one is more protective of users.

Alan presented another example: if the original policy is "Copy 3 Times", two possible paraphrases are "Copy Freely" (less restrictive) and "Copy Once" (more restrictive). If the policy is paraphrased as "Copy Freely", the new policy permits things which would previously have been prohibited; if it's paraphrased as "Copy Once", the new policy prohibits things which would previously have been permitted. Again, I see no strictly logical basis upon which one paraphrase could be preferred to the other. Both are inaccurate.

It would be unfortunate indeed if MPAA presented the situation as though there were no choice to be made between alternatives like these when paraphrasing policies. Nobody at the ARDG has presented -- in the context of MPAA's view on the policies which ought to be expressible in an analog signal -- a way to avoid having to make such a decision.

In a subsequent conversation with Brad Hunt, I pointed out that, outside the ARDG, MPAA tends to point to the existence of the analog hole when DRM technologies are criticized for thwarting fair uses. For instance, in the 2600 litigation and in the Library of Congress DMCA rulemakings, movie studios suggested that people who needed to make fair uses which were prevented by CSS could use analog to make those uses. If MPAA hopes to do something which closes the analog hole, will that particular safety valve evaporate?

A presentation on a technology called CGMS/A+ (supposed to be a next generation version of CGMS/A) was met with a wide variety of criticisms. CGMS/A+ would add a visible "rights assertion mark" -- a logo in the corner of a viewer's screen -- to programming which had previously been marked with CGMS/A (a non-watermark label in a well-known portion of the signal which is not normally viewable, nearby any closed-captioning information which might be present). The claim was made that this visible mark would be difficult to remove, and devices might make certain assumptions about CGMS/A data based on the presence of a rights assertion mark.

The reasoning behind the proposal to add the rights assertion mark is that it is fairly easy to remove (and somewhat easy to alter) CGMS/A data. The mark might function as a signal that CGMS/A information had originally been present and had subsequently been removed by somebody. However, it did not appear to many participants that CGMS/A+ would be broadly useful. One reason is simply that many important devices like computers typically don't respond to CGMS/A in any way, and aren't legally required to. So trying to preserve CGMS/A marking can't do anything to affect how these devices behave.

Jim Burger presented an interesting set of requirements for discussion, prompting others vigorously to discuss them.

We hope Jim's document will be made available to the public soon.

In both the broadcast flag and ARDG contexts, there has been wrangling about just how "robust" a compliant product must be against "tampering" (i.e., modification by users). The standard proposed by the MPAA in Section X.17 of their proposed broadcast flag mandate is that content protection systems should be implemented so that they cannot be defeated by using general purpose tools widely available at a reasonable price.

According to presenter Andrew "Bunnie" Huang, achieving this level of security in mass-market hardware devices is very difficult, if not impossible. Hobbyist hardware hackers have an impressive array of general purpose, low-cost tools at their disposal. Consequently, staying "robust" against low-cost tools requires, at a minimum, that all cryptographic secrets (and all code or data from which they could be derived) must be kept inside custom-designed chips, or any bus across which such data could travel must be encrypted. If a secret is available in the clear on any pin or trace on a circuit board, it can be extracted using general purpose tools for, in Bunnie's words, "much, much less than $1000, often less than $100, and frequently for free."

So just how much unnecessary cost is "robustness" going to add to consumer products? Quite a bit, if "robust" means immune from hobbyist hardware hacking.

CPTWG (the parent organization for the ARDG) is administered by three co-chairmen, one each to represent the consumer electronics (CE), information technology(IT), and motion picture industries. Decisions regarding the organization and procedures of CPTWG are made by these co-chairs, as most recently illustrated at the April 10, 2003 meeting of CPTWG, when the three co-chairs unanimously rejected a proposal by public interest and consumer groups to permit members of the press at meetings of the ARDG.

The events of the April 10 meeting make it clear that public interest and consumer groups are not adequately represented in CPTWG or its subgroups. Public interest and consumer groups have been participating in the CPTWG process on a regular basis for over a year, and anticipate being actively engaged in the future.

The issues discussed at CPTWG meetings are of critical importance to consumers and the public at large. It is literally the future of consumer media devices that is being negotiated at these events. Unfortunately, the ability of the public to engage in the debate is hampered by the expense of attending these meetings, their technical nature, and the exclusion of the press.

In light of these limitations, it is particularly important that groups representing consumers and the public interest be afforded the ability fully to participate and contribute to the CPTWG process. Having equal representation in CPTWG decision making is one aspect of fulfilling that goal.

The co-chairs indicated that the EFF proposal will be addressed at a future meeting of CPTWG.

The work of the ARDG is to develop a fix for the "analog hole" -- the fact that anything that emerges from an analog output can easily be redigitized. Technologists would see this as an opportunity for innovation, not a "problem": that which was once analog becomes digital, and infinitely more useful in the process.

The MPAA sees things differently. In their view, since all digital outputs will be hobbled by "content protection" technology, we must plug the "analog hole" to somehow make sure that the analog output stays at least as hobbled as the digital. The MPAA newspeak term for this is "Equivalence".

The MPAA recognizes that the original digital file's DRM may be capable of expressing complex rights and permissions -- things like, "allow copying within the home," or "allow retransmission for 30 days" -- that will be impossible to express in the crude "Me Tarzan/You Jane" pidgin of analog watermarks.

In an effort to address this inevitable "equivalence gap," Hunt's presentation set out the minimum set of "states" that the MPAA demands in any analog signaling (read: watermarking) system. They are:

1. Copy Never [and no redistribution]
2. Copy once (similar to the SCMS system used in DAT) [and no redistribution]
3. Copy no more (the condition of a second-generation "Copy once" copy) [and no redistribution]
4. Copy freely, but no redistribution [redistribution controlled by encryption]
5. Copy and redistribute freely (could be implemented either by absence of a signal, or by specific "copy freely" signal)

Furthermore, if a device "re-digitized" analog content that showed any of the above states (except the last), it would be required to encrypt the content. [It went without saying that all devices would have to be "tamper-resistant," as well.]

These "states" are familiar to old DRM hands, as they mirror states currently supported by the "5C" family of content protection systems, which include DTCP (1.3MB PDF). The question of redistribution control also came up, with Brad initially suggesting that redistribution is implicitly forbidden for all states, except the last. Under fire, he backed down, suggesting that the issue of redistribution control was open to further negotiation.

Two things stood out about Brad's presentation.

First, I was struck by the MPAA's view, bluntly expressed by Brad, that the role for analog outputs was to support "legacy" equipment. Apparently, all future innovation will have to rely on the DRM-hobbled digital outputs. This is a major change from the status quo that has given us innovations like the VCR, where analog outputs not only supported legacy devices, but were generally available to support innovative products not yet imagined.

If you think analog is only good for supporting legacy devices, then your view of the "analog hole problem" is likely to be very different from someone who thinks of analog outputs as an innovation opportunity.

Second, it is clear that a huge fight is brewing on the "rounding" problem. On the one hand, Brad has repeatedly emphasized the importance of "equivalence" (i.e., that analog be hobbled so that it's no more capable than DRM-protected digital outputs). On the other hand, Brad admits that analog watermarks will not be able to express the "full richness" of DRM states expressed in digital content.

Hence, the "rounding" problem: where analog signaling systems can't adequately express a digital DRM state, do you "round down" to a more permissive analog state, or do you "round up" to a more restrictive analog state? So, for example, if a new digital DRM-system supports "copy once, but keep for only 14 days," how does that get expressed at the analog output? Does it become "Copy Never" (rounded up) or "Copy Once" (rounded down)?

If you round up, then you've made analog artificially unattractive, by making it more restrictive than the digital outputs. If you round down, you have the same problem in reverse. You have to choose, there is no "neutral" alternative.

I know where I come down. Stay tuned to find out where the CE and IT industries come out. ]]> Meetings 2003-04-18T13:22:59-08:00 http://analog.blogs.eff.org/archives/000273.html At the April 10 meeting of CPTWG, after a brief debate, the three co-chairs of the group (one each... 273@http://analog.blogs.eff.org/ At the April 10 meeting of CPTWG, after a brief debate, the three co-chairs of the group (one each representing the motion picture, consumer electronics and information technology industries) unanimously rejected a request by EFF, Public Knowledge, Consumers Union and DigitalConsumer.org that members of the press be permitted to attend the otherwise public meetings of the ARDG.

EFF then proposed that the three co-chairs be supplemented by the addition of a fourth co-chair, representing the views of public interest and consumer groups. The co-chairs indicated that they would consider the request at a future meeting.

The ARDG is a sub-group of the CPTWG and is governed by its press policies. CPTWG has long had a policy of excluding members of the press, while simultaneously representing their meetings as open to the public. The official press policy was recently codified and posted to the CPTWG website after The National Journal wrote a letter to the co-chairs protesting the CPTWG policy.

EFF, Public Knowledge, Consumers Union, and DigitalConsumer.org had asked the co-chairs to reconsider their policy, at least as applied to the ARDG, in light of the importance to consumers and the public of the technical issues being discussed by the group. The Center for Democracy and Technology also spoke in favor of allowing press to attend.]]> Meetings 2003-04-17T16:58:38-08:00 http://analog.blogs.eff.org/archives/000267.html Princeton computer science professor Edward Felten has posted several interesting thoughts regarding the ARDG press ban on his excellent weblog,... 267@http://analog.blogs.eff.org/ Princeton computer science professor Edward Felten has posted several interesting thoughts regarding the ARDG press ban on his excellent weblog, Freedom To Tinker.

Some within ARDG have expressed the view that the ban on press attendance is necessary because many who attend ARDG meetings are not authorized to speak to the press on behalf of their companies. As a result, so the argument goes, many who would otherwise contribute to the ARDG discussions would be forced to remain silent if press were in the room.

The more you think about it, the less this looks like a valid argument and the more like circular reasoning. After all, any chill felt by attendees is occassioned by their companies' policies regarding the press. It seems that these policies need to be justified, rather than being relied upon to justify exclusion of the press.

But all of this may overstate the import of the issue. So, rather than questioning the company policies, I'd like to make a few observations regarding the context in which these concerns arise.

First, it has been my experience that the majority of engineers who attend CPTWG and BPDG meetings never say a word. To the extent this proves to be the case at ARDG, their employers need not worry about their remarks being attributed to the company.

Second, many who speak at these meetings are authorized by their companies to speak to the press. I have frequently seen Michael Epstein, Chris Cookson, Bob Schwartz, Brad Hunt, Jim Burger, Seth Greenstein, and others quoted in the press on topics relating to CPTWG issues. These fine gentlemen are expert at choosing their words with care and precision. Their employers and clients need not fear for their discretion should a few reporters appear at our meetings.

Third, to the extent some of the most important contributions come in the form of questions, it is not clear to me that companies need worry about a question to someone else being construed as a company position.

Fourth, it has always been my understanding that nothing said in these meetings is confidential, nor are attendees prohibited in any way from talking to press (or anyone else) about what transpires at the meetings. Accordingly, under our current rules, when someone contributes a bon mot to our meetings, that person's employer may see the substance of the remark in the next morning's press coverage.

On the other side of the coin, the recent experience with the BPDG suggests that excluding press makes it much more difficult for interested members of the public to evaluate the material that emerges from these groups. Despite the fact that the "broadcast flag" proposal, if adopted, would materially shape the future of DTV equipment in the U.S., I did not read any detailed analysis of its implications in Sound & Vision, The Perfect Vision, Stereophile Guide to Home Theater, Home Theater, or other consumer press. Hence, the consumers most likely to be affected by the proposal are hardly aware of it, much less able to adequately evaluate it. A similar outcome for the ARDG should be avoided.

So the question is whether the incremental burden imposed by admitting press is outweighed by the incremental benefit.

In a joint letter to the CPTWG co-chairs, the Electronic Frontier Foundation, Consumers Union, DigitalConsumer.org and Public Knowledge expressed concern regarding the exclusion of the press from ARDG meetings.

Citing its policy of excluding members of the press from its otherwise public meetings, CPTWG has extended this policy to include ARDG, which is a CPTWG subgroup. Drew Clark, of National Journal's Tech Daily, was ejected from ARDG when he attempted to attend its inaugural meeting.

Journalists serve a critical role in educating consumers, innovators, researchers and other interested members of the public about the debates taking place at ARDG. Recommendations that emerge from ARDG may become the basis for federal policy, as the final report of the BPDG became the basis for the "broadcast flag" proposal now being considered by the FCC. In the absence of press analysis, members of the public will be at a serious disadvantage in evaluating any policy proposals that may emerge from ARDG.

At the March 5 meeting of the ARDG, Brad Hunt from the MPAA offered a 3-slide presentation aimed at defining the "analog reconversion" problem and a "reference architecture" within which it might be examined. The presentation was not intended as a substantive proposal, but rather an overview of the scope of the problem.

Nevertheless, it introduced several themes that I expect will recur frequently during the ARDG's future meetings.

First, Brad introduced the MPAA's preferred rhetoric for the ARDG's efforts. The challenge is, in Brad's view, guaranteeing "equivalence" between protected digital outputs and analog outputs. At a minimum, I imagine this means that all analog links between digital video devices should offer roughly the same protections as the DTCP-protected FireWire links just now beginning to appear on high-def TVs and cable/satellite set-top boxes.

Of course, this is a bit like choosing the part of the playing field that is the most tilted in your favor, then arguing that all other parts of the field must be "leveled" up to match it.

Second, the PowerPoint slides foreshadowed what I predict will be two recurring battlefields at ARDG: "robustness" and "how many detectors".

It's all in Brad's last slide. The slide expressly recognizes the need for a "trusted environment" for both source and sink devices, by which Brad means to suggest that devices will have to be designed to be "robust" (i.e., with the hood welded shut to prevent user "tampering"). Just how "tamper-resistant" these devices have to be will be a major issue for makers of general purpose computers. It will also, of course, determine whether open source programmers have any ability to participate in the digital video revolution, as "robustness" rules disqualify open source programmers by charaterizing their products as "tamper-friendly" (i.e., user modifiable).

This slide also identifies a variety of places where watermark detection might take place in a sink device, without taking a position on where or how many times the device has to check for the watermark. This is also likely to be a major issue for IT companies (i.e., companies that produce devices intended for use with general purpose computers), whose enthusiasm for watermark detection may be inversely related to the number of places inside a general purpose computer that they have to install detectors.