Tuesday, August 16, 2005

Technological Protection Measures

The DMCA is not the first provision in title 17 to mandate technological protection measures (TPMs). The first time I experienced this was in 1992, as copyright counsel to the House IP subcommittee. Hayden Gregory, the chief counsel of the subcommittee and I were charged with making was referred to jokingly as "DART-Lite", watering down the industry's version of what became the Audio Home Recording Act of 1992. Industry wanted the Serial Copy Management System (SCMS) to be mandated. We balked at mandating a private technology in public legislation, and so, Section 1002(a) refers to SCMS and to "a system that has the same functional characteristics as" SCMS.

DART, by the way, is a classic case of Congress being sold a phony bill of technological threats. We were told that the docks were loaded with these new fangled ditigal machines that would produce endless perfect copies from other copies. I saw no market for the digital audio cassettes contemplated by DART: adults would want CDs since they were hardly any more expensive, while kids would prefer analog cassettes given the much lower price.
And as for the "perfect copy claim," that claim was very powerful among members of Congress in the 1990s. Hayden and I actually insisted that the proponents send over from Japan a technician to demonstrate whether that was true; it turned out that after the 40th copy, there was an appreciable degradation.

Two other points, the utility of DART turned out to be of very short, limited value, while Section 1008, which was intended to take care of home copying once and for all, did no such thing. I wished we had killed the bill.

The issue of TPMs and whether the DMCA has fulfilled its purpose is being discussed on Randy Picker's mob. The mob is based on an article by Fred van Lohmann on the infamous Microsoft darknet paper. Here's the link to the discussion group: http://picker.typepad.com/picker_mobblog/


billb said...

I'm curious about what was demonstrated to you that showed degredation by the 40th copy. It could not have been an all digital process with no or lossless compression.

William Patry said...

The demonstration was of a "master" digital audio cassette containing content that was then copied on to one blank cassette,; that cassette was then copied onto another, etc. for 40 times. We assumed that it had something to do with the nature of the tape medium.

billb said...

Interesting. That's certainly not the case anymore. Now, I can copy a CD to my harddrive, and then to different drives (or other computers over the network), as many times as I like with out a single bit being changed.

So if perfect copying still has any pull with Congress, we may be in for some stricter laws.

William Patry said...

I have been reading a lot of stuff by cognitive linguists about metaphors. Metaphors (of all species) are very very powerful in Congress. Methaphors are seductive because they are a way for people to process complex subjects simply (usually too simply), and in Congress members rarely spend much time on any one subject, so metaphors are quite useful as a pretend-way to signal comprehension. Metaphors are used as if they actually explained the problem, with the audience left with the impresison that the member is actually thinking through the problem: the metaphor is supposedly evidence of that mental process. Instead, the opposite occurs: invoking the metpahor is a shield against thinking.
One IP subocommittee chair was particularly found of "everything is a balance." But the digital perfect copy metaphor as a metaphor for unchecked, dangerous, market-supplanting copying that must, absolutely must be stopped, was the most powerful of all and still gets lots of mileage.

billb said...

I think that it's entirely true what you say about perfect copies (well, except for the "must be stopped" part), but it's not a metaphor. It is the reality of the times. Perfect copies are here, now, and they aren't going away. As soon as the industry realizes it, the sooner we can move on.

It's time that authors found a new (or returned to an old) way of getting paid for being creative. Effortless identical copying is a sine qua non of general-purpose computing devices, and, since the latter isn't going away, we need to get used to the former as quickly as possible.

Fred von Lohmann said...

It is not clear to me that making "perfect digital copies" is as easy as many assume. The Red Book CD audio format, as I understand it, does not include the kind of checksum mechanisms that computer data formats generally do. In fact, software like Exact Audio Copy attempts to overcome this limitation by doing multiple passes over CD audio data, in hopes of discerning the "right" bits when ripping. As a result, I find it quite plausible that multiple generations would result in degradation. I'd love to see an empirical test using CD-R.

billb said...

Not so fast, Fred! Once the copy is sucessfully extracted from the media to my hard drive, I can make as many generations of copies as I want from that point on.

Another point is that once extracted from the media, one doesn't usually keep a RedBook format of the data. RedBook is one way of storing 16-bit 2-channel 44.1 kHz-sampled PCM audio. The tradtional .WAV format is another. All CD ripping programs I know of, after verifying the veracity of the read using the sorts of error detection/correction you mention, have the PCM data, and they choose to write out .WAV formatted audio (for a perfect copy) or something (like MP3, Ogg Vorbis, et al.) for lossily-compressed copies. The audio data is the same (bit-for-bit) between RedBook and .WAV formatted versions even if the containing files are not. This .WAV-formatted data can certainly be copied perfectly by computers (or even loslessly compressed via the Shorten or FLAC formats) and sent around the world in thousands of generations without _any_ loss.

Once the audio data is on a harddrive in an appropriate lossless format, perfect copies are makeable.

joshua wattles said...

As members of Congress become more familiar with personal computers, they should become less affected by the notion of "perfect copies" and come to realize that a perfect copy is a positive thing - - a necessary corollary to the delivery of information from one point to another. Metaphors are best met by metaphors. Here's one that the tech community might borrow from the community hoodwinking Congress with the fear of perfection - - "Beam me up." It should be obvious that we will never reach the capability of sending people by beam from one place to another without first mastering the transmission of "perfect copies." Whiz bang technology always trumps content which is why the content industries spend so much and fight so hard. They have to do so just to keep their heads above water.