Everything you've read about Vista DRM is wrong (Part 2)
Windows Vista includes a new set of features that allow playback software to work with protected media, especially high-definition content. This DRM infrastructure is bitterly controversial, and it's given rise to an enormous amount of misinformation. In part 2 of this three-part series, I continue my detailed examination of the errors, dostortions, and untruths in the most widely quoted paper on the subject, written by New Zealand researcher Peter Gutmann. I was stunned to find that some of the assertions in his epic paper are directly contradicted by his own sources, and that two of his key stats are literally made up. See the proof for yourself.
Windows Vista includes a new set of features that allow playback software to work with protected media. This DRM infrastructure is bitterly controversial, and it's given rise to an enormous amount of misinformation. No one has been more active (or successful) in spreading FUD and misinformation about this technology than Peter Gutmann, a researcher from the University of Auckland in New Zealand.
In Part 1 of this three-part series, I discussed some of the technical errors in Gutmann's paper that illustrate his lack of hands-on experience with the technology he's trying to cover and his fundamental confusion over how Windows Vista content protection features work. (You'll find more examples in Part 3.) If you think I'm nitpicking over these details, you miss the point completely. Gutmann is an academic researcher, and the way scientists have worked since the end of the Dark Ages has been with a rigorous set of principles: You start with a thesis, you design experiments that test that thesis, and using those experimental results as well as those of your peers, you assemble evidence that proves or disproves your thesis. Then you publish.
As I noted last month, Gutmann has completely skipped the "experimental" portion of this time-tested process. He has literally no firsthand evidence to support most of the outrageous claims he makes, and much of the secondhand anecdotal evidence he has assembled is either taken out of context or is of questionable relevance. As I show later in this post, some of his evidence is just plain made up. When someone who claims to be a scientist publishes a paper filled with provably wrong facts, that person's competence is called into question. When all of those errors are in one direction, that person's honesty, objectivity, and devotion to the truth are called into question as well.
In this part, I'm going to drill down into the more controversial parts of his paper that deal with the PC as a platform for digital media. It starts with an amazing political statement.
ERROR #5: FIRST STEP ON THE ROAD TO TOTALITARIANISM?
I include this example because it's a near-perfect illustration of Gutmann's willingness to link to a supporting article that has nothing to do with the point he's trying to make. He appears to be betting that his audience won't actually follow all of those links and notice the disconnect. According to Gutmann, Bill Gates himself is behind the Microsoft conspiracy to deny all of our digital rights, and Windows Vista is part of his evil plan to take over distribution of all digital media. His "proof" is in this statement:
Microsoft have been saying for some years now that they'd really like the PC to go away, to turn into a kind of media platform and content-distribution center for consumers. This was a major theme of Bill Gates' world promotional tour for Vista in early 2007, and in particular something he went into in some detail at the World Economic Forum in Davos, Switzerland.
If it were all just about lining Bill's pockets (and those of the long-suffering Microsoft shareholders), this argument might be tenable, if a bit cliched. But Gutmann takes this conspiracy theory farther than anyone. From that simple paragraph he segues into a short discussion of Microsoft's evil master plan "to lock out any competitors... [And] because they will then represent the only available distribution channel they'll be able to dictate terms back to the content providers whose needs they are nominally serving in the same way that Apple has already dictated terms back to the music industry." It ends, in hyperbolic fashion, with a link to an online web site devoted to the history of the Soviet secret police (NKVD), with lurid details about forced labor camps and warnings about "a continual extension of the security apparatus and an ongoing escalation of repressiveness by the enforcers." Gutmann concludes:
The many examples given in the rest of this writeup are an indication that Windows is already well down this path.
I am not making this stuff up or exaggerating. Peter Gutmann wants you to believe that Microsoft's goal with Windows Vista is the elimination of the PC, which in turn is the first step on the road to forced labor camps and secret police. So, once again, I followed the link he supplied at the start of this section, which leads to a secondhand report at Download Squad, drawn from a Reuters report that is no longer available online. Here's the entire Bill Gates quote from that story:
Certain things like elections or the Olympics really point out how TV is terrible. You have to wait for the guy to talk about the thing you care about or you miss the event and want to go back and see it.
How do you get from here to "Microsoft [would] really like the PC to go away" and then to Soviet-era concentration camps? I'm not sure either. A slightly more detailed report of Gates's remarks includes this additional quote: “I’m stunned how people aren’t seeing that with TV, in five years from now, people will laugh at what we’ve had.” The prediction that delivery of TV programs over the Internet will increase in the next five years isn't exactly earth-shattering (see iTunes, YouTube, Amazon Unbox for the first glimmerings of this trend). It certainly doesn't predict the death of the PC.
Gates is describing something that the rest of us, those who follow the media industry and are neither professional nor amateur paranoids, have been noticing for some time. As bandwidth increases and the penetration of high-speed Internet connections reaches ever-higher levels, the Internet becomes a practical way to deliver movies, music, TV, and more. That has profound implications on the traditional media industry, which has historically relied on cable, satellite, Blockbuster Video, and other controlled channels to reach its customers. Internet-based media delivery is a disruptive technology, and it's only natural that both Microsoft and Apple would both want to be part of this changing landscape. But to sketch a path from content protection technology to the Gulag Archipelago is truly depraved.
In fact, this assertion is one of the foundation arguments in Gutmann's paper. He believes that "content protection [is] Microsoft's number one priority for Vista." Which leads us to...
ERROR #6: NO, MICROSOFT HAS NOT PLACED CONTENT PROTECTION ABOVE ALL OTHER REQUIREMENTS
Gutmann's speculation about Microsoft's motives is fueled by this incendiary paragraph:
You can get an idea of just how important content protection is to Microsoft by looking at the Windows Vista logo requirements. The primary requirement for graphics devices in the Windows Vista Logo Program isn't, as would be expected, the ability to handle a high-resolution display or display a rich palette of colours. It isn't the presence of a good quantity of memory and powerful graphics rendering. It isn't even the ability to handle Vista's much-touted Aero interface, arguably the primary reason for running Vista. Instead, the number one requirement for Windows Vista graphics device certification, “GRAPHICS-0001” in the specification, is “Display adapter supports output connectors with content protection features and provides control via PVP and COPP DDIs”. It's only the follow-on “GRAPHICS-0002” that requires that “Display subsystem meets GPU, memory, resolution, and bandwidth requirements for a premium Windows experience”. This is a pretty amazing admission, because it means that Microsoft is placing content protection above all other requirements for Vista, even the ability to handle Vista's primary feature, the Aero interface. [emphasis in original]
There is so much wrong with this paragraph that it's hard to know where to begin. First of all, Gutmann links to the original Windows Logo Program Device Requirements document (v. 3.09, 321 pages in PDF format). Did he even read this document? The "Graphics Devices" section that he discusses here leads off on page 180 with these two paragraphs:
The basic requirements in this section focus support for minimum capabilities for timing crates, gamma correction, and monitor detection, plus the driver‘s ability to correctly report supported capabilities. For the Windows Vista logo, the adapter or chipset must have a WDDM driver and support GPU, memory, bandwidth, and other requirements supporting Aero capabilities. [emphasis added]
At a minimum, display devices must support DDC/CI standards for communications between a monitor and the communication bus, in addition to supporting timing standards and sleep states. For the Windows Vista Logo, LCD and plasma displays have additional requirements to contain display characterization data and ensure color quality. Displays and monitors must meet the applicable requirements in the "Device Fundamentals" section of this document.
Gutmann claims in his paper that this requirement gives short shrift to "the ability to handle Vista's much-touted Aero interface." And yet "supporting Aero capabilities" is mentioned in the first paragraph as one of the spec's "basic requirements." Further down the page is a detailed table that describes the Windows Vista Aero desktop experience and contrasts it with the other Vista interface options. Did you see any mention of content protection in that opening description? Me neither. In fact, there's not one word about content protection on the entire page. Here, see for yourself:
Gutmann asserts that content protection is the "primary requirement" for graphics devices that want to use the Windows logo. The basis for this argument is the fact that the new output protection spec is numbered 0001 out of 69 specific requirements in this category. He misses the fact that there's no hierarchy at all in these requirements. Each of the items on the list is labeled R (Required) or I (If-implemented). To qualify for the use of the Windows Vista logo, a device manufacturer must satisfy all specs labeled R, regardless of what number Microsoft used in the spec.
Finally, Gutmann misses the I (If-implemented) identifier on the GRAPHICS-0001 spec. On page 10 of Microsoft's specifications, under the heading "If-Implemented Requirements," you can read the following:
[I]f a functionality that is not part of the mandatory requirements is implemented in a system or device, additional requirements might be defined for that functionality to assure that the implementation works well with Windows. Because these additional requirements apply only if the relevant functionality is implemented, they are referred to as "if-implemented" requirements. [emphasis added]
In the original v3.09 spec he cites, output protection is only a requirement (R) if you want the Windows Vista Premium logo. For an entry-level video card, manufacturers are free to ignore this spec and still qualify for the Basic Windows Vista logo. By contrast, GRAPHICS-0013 ("Display adapter or chipset supports standard VGA and can be reset to standard VGA") is listed as R for both the Basic and Premium Windows logos.
And in the most recent update to the Windows Logo Program spec (version 3.10, published August 30, 2007), support for content protection is no longer required even for a Vista Premium logo. Again, see for yourself:
Gutmann uses this ridiculous, forced argument to support his contention that content protection is the single most important aspect of the Windows Vista video subsystem. The reality is exactly the opposite, as proven by the very Microsoft documents he cites.
ERROR #7: HP IS DROPPING MEDIA CENTER PCs? NOPE.
Gutmann gets downright giddy over any signs of failure in the Windows digital-media ecosystem. Witness this snarky passage that he uses to paint Media Center as a miserable failure:
In fact the entire market's reaction to Windows Media Centre PCs has been a single colossal yawn, with HP, the last remaining major MCE vendor, dropping its entire product line in early 2007. What's left now is a bunch of “niche players making nice products […] but they're not HP or Dell or Gateway or Alienware for that matter. In other words, they're not mass-market products”.
This entire breathtaking assertion (you can almost hear his "Ha ha!" delivered a la Nelson Muntz) is based on a single article published at the CEPro website in March 2007. The link in Gutmann's paper doesn't actually go to the story he's quoting, and in fact it appears that CEPro has tried to scrub the story completely. The only trace I could find is this print-formatted version. It turns out, however, that the reporter got the story wrong. [Update 22-Sep: Julie Jacobson of CEPro provides a current link to the story and says the publication stands behind it. Read her comment in full here, and see my reply here.] Here are the top three models in HP's current lineup of desktop PCs today:
Two of these three models are called Media Center PCs, and the third is in the living-room form factor and specifically identified for use in "home theater entertainment" environments. And if you dig into the text descriptions, there's even more. Here, for example, is the current (September 2007) description of one of HP's current PC lines, the HP Pavilion Media Center m8100y:
This series comes with Vista Home Premium, which includes Windows Aero, Windows Media Center, and instant search functionality. Upgrade to Windows Vista Ultimate on your Pavilion Media Center PC, upgrade the quality of your digital life. You'll notice the difference in the new look, instant search, built-in Windows Media Center, and automated security features. You'll be able to create beautiful family photo albums and home videos, listen to your favorite music, automatically record TV shows and movies onto your PC, and explore a new world of games.
The rumors of their demise has apparently been greatly exaggerated. No wonder CEPro was so anxious to erase the story.
One of Gutmann's favorite techniques in this paper is to make an unsupported assertion and then draw conclusions from it. It's the sort of logical flaw that you're taught to spot in the first week of a freshman rhetoric class: When you start with an unsupported statement, none of the conclusions you draw can be considered proven.
In this case, the unsupported assertion starts with market share numbers pulled out of thin air. Under the heading "Disabling of Functionality," Gutmann writes:
In order to appropriately protect content, Vista will probably have to disable any special device features that it can't directly control. For example many sound cards built on C-Media chipsets (which in practice is the vast majority of them) support Steinberg's ASIO (Audio Stream I/O), a digital audio interface that completely bypasses the Windows audio mixer and other audio-related driver software ... [emphasis added]
See how he slipped that little statement in there to make the problem he's discussing seem like something that will affect "the vast majority" of Windows systems? The trouble is, the vast majority of sound cards are not "built on C-Media chipsets." Don't take my word for it; that's what the company itself says. In reporting on a 2006 deal between C-Media and Asus, DigiTimes quotes a report in the Chinese-language Economic Daily News (EDN)
C-Media anticipates that its market share in the high-end audio IC market will hit 10%, up from the current 1-3%, according to the company...
The last time I looked, 1-3% was a tiny blip, not the "vast majority."
Another example of made-up market statistics is this statement:
...the vast majority of drivers running on PCs today aren't signed, not so much because the developers couldn't be bothered but because the WHQL process that produces the signed drivers is so slow that they're obsolete by the time they've been approved by Microsoft [...]
There's no support for this assertion, and for good reason. It's ridiculously wrong. Gutmann has fallen head first into the Google fallacy. "You can use Google to find endless examples of the use of unsigned drivers," he writes to support this statement. Go ahead, click the link. I looked on the first three pages of Google results and couldn't find a single downloadable driver, only articles about what the unsigned driver dialog box means and why it's important not to install unsigned drivers and how to work around this warning if you encounter it. And yes, this is important information if you plan to install a beta driver or if the manufacturer of one hardware device you use has decided to release a driver without signing it. But it's hardly indicative of market share.
As of mid-2007, Microsoft reported that more than 2 million Windows Vista devices were supported with signed drivers. I can't find any hard numbers about how many Windows drivers are unsigned (and neither can Gutmann, which is why his assertion is unsourced). But I'd be willing to bet a month's pay that the "vast majority" of drivers in use in the Windows ecosystem today are indeed digitally signed. Why do I believe that? Simple. Historically, at least 80% of all copies of Windows are sold with fully assembled computer systems. When you buy a new Windows PC from a big OEM like Dell, HP, or Sony, all drivers included with that new PC must be signed. That's a requirement for use of the Windows logo. When you download drivers from Windows Update, every single one is signed. All current (non-beta) drivers from Nvidia and ATI and Intel, the leading makers of video subsystems for Windows, are digitally signed and WHQL-certified, as are shipping drivers for sound cards from Creative and other leading device makers. In short, it appears that the vast majority of Windows Vista drivers are, in fact, signed.