Industry Canada is floating a proposal to update Canadian copyright law. One of the proposals is a DMCA-like provision banning the possession of technology which circumvents digital copy protection.
Please read the proposal itself; then read the following which I have submitted as a comment to Industry Canada. I welcome feedback.
Industry Canada is considering a proposal to provide sanctions against persons who "engage in activities related to the circumvention" of technological measures against technologies designed to thwart copyright protection.
I believe such sanctions are bad law, bad for technology, bad for scientific research, and bad for society as a whole. In this paper, I will provide many arguments against such sanctions, and provide arguments supporting my contention that existing copyright law is sufficient to protect digital content from unauthorized copying or distribution.
Further, I will show how such sanctions will lead to a chilling effect in certain areas of research, damage Canada's technological competitiveness, and can be used to increase protection of content beyond what is specified by copyright law, to a point which damages society.
Finally, I will demonstrate how a comparable law in the United States, the Digital Millennium Copyright Act, has been used unjustly to block free speech and to disproportionately punish people who have not even been shown to have violated copyright law.
Technological means of content protection are irritating for law-abiding citizens, because they make it difficult or inconvenient to make fair use of copyrighted works. At the same time, technological means of content protection do little to deter content pirates, because any means of content protection can be thwarted, and usually quite easily.
Most technological means of content protection rely on encryption schemes to protect the content. The content can only be unlocked with the appropriate decryption key. The key may be locked to a particular device such as a music player or an electronic book reader, or it may be necessary to download the key over the Internet after you have authenticated yourself.
While it is possible to design strong encryption schemes and fairly secure key distribution mechanisms, these serve no useful purpose in restricting piracy. The following schemes will break any encryption-based content-protection scheme:
If the content being protected is music, a pirate merely needs to purchase one legitimate copy. He can play the music into a high-quality analog recording device, and re-digitize it into the format of his choice. While a few percent of the population may notice some difference between the re-digitized content and the original content, for the vast majority of people, there will be no discernible difference. The pirate who can invest in high-quality analog recording equipment, therefore, is not deterred. The law-abiding citizen who wishes to lend his music to a friend is greatly inconvenienced.
If the content being protected is video, a pirate can similarly re-digitize it on high-quality video equipment. This is much too costly for a law-abiding citizen, but a small cost of doing business for organized pirates.
If the content being protected is text (an electronic book, for example), a pirate can hire a typist to transcribe the book contents into the format of his choice. Again, this is just a business expense for a pirate, but a huge deterrent for law-abiding citizens.
Security experts such as Bruce Schneier  agree that there is no possible technological means of completely protecting content. Thus, all the digital protection schemes simply inconvenience consumers without deterring piracy.
Copy-protection schemes often have serious implications for privacy. Some schemes rely on a user authenticating himself before accessing protected material, or registering before being given access.
This lets publishers track who is accessing their material, and use this information in any way they see fit. Today, if you buy a book about a controversial topic, the publisher has no way of knowing who you are. Tomorrow, if e-books require registration for access to their content, publishers could build up databases matching every person to every book, CD or video he has purchased. This can be used for many purposes, from relatively benign marketing to more sinister exchange of information with government or police agencies.
If technological copy-protection is permitted, the government must strongly regulate the use of any information collected to enable access to protected material. Industry self-regulation is a failure; corporations do not believe in privacy if they can make money by breaking privacy. 
Copyright law provides for fair use of copyrighted works. Thus, for example, it is perfectly legal for you to lend a book you have bought to a friend. You are also allowed to lend CD's and videos to friends. You can even tape music or video off the air for your later private enjoyment. Society recognizes that while copyright holders must benefit from and have control over their works, this must be balanced with the right of citizens to enjoy those works.
Technological means of content-protection can take away the rights we enjoy. They can make it very difficult to exercise the rights that copyright law has granted you. In fact, the only way to exercise those rights might be to circumvent the copy-protection scheme.
Another danger is that content providers could tie their content to specific hardware. This would be akin to having to buy multiple VCR's to watch videos from different producers. Not allowing reverse-engineering of hardware for the purposes of making compatible hardware would greatly reduce your rights.
The proposed sanctions would make it illegal for you to exercise your legal rights to fair use of the copyrighted works. Even if your intent is simply to assert your rights, the sanctions could punish you. Sanctions against devices to circumvent copy-protection schemes in effect are sanctions against thought crimes. Even if you have no intention of violating copyright law, the mere possession or dissemination of technology which can circumvent copy-protection schemes becomes illegal.
If there are legal sanctions against circumventing copy-protection, then companies have little incentive to provide high-quality copy-protection. They need only make some effort to prohibit copying, and can rely on legal measures to intimidate reverse-engineering of the copy-protection technology.
In fact, there have been two recent examples illustrating how companies feel they can get away with weak copy protection. The DVD industry uses a method called "Content Scrambling System" (CSS) to encrypt DVD contents. The method was kept secret until it was eventually reverse-engineered. As is the case with most proprietary encryption algorithms, CSS was revealed to be woefully inadequate and virtually trivial to crack. There is an entire gallery of CSS crackers available on-line as computer hobbyists thought of ever more-creative ways to break CSS.
The DVD industry did not invest the time and money in research to create more effective encryption, because they knew that the U.S. Digital Millennium Copyright Act would give them power to punish anyone who revealed the details of CSS, thereby giving them a legal club to beat nosy cryptographers with.
Similarly, Russian programmer Dmitry Sklyarov revealed that Adobe's e-book protection mechanism was woefully inadequate, less effective even than CSS. Rather than being thanked for pointing out the weaknesses in the scheme, Sklyarov was thrown into jail when he visited the U.S.
As these cases show, laws like the DMCA hurt technological development on two fronts: On the one hand, they reduce the incentive for companies to research more effective encryption techniques. It's far more cost-effective to sue or intimidate opponents than to develop encryption algorithms which give them a strong mental challenge. The DMCA therefore gives companies a dangerous sense of complacency -- they are confident in the strengths of encryption algorithms which are not in fact strong -- they are merely unbroken because researchers are unwilling to point out weaknesses in them for fear of retribution.
On the other hand, laws like the DMCA hurt cryptography research. Traditional cryptography research works like this: A cryptographer proposes a new encryption algorithm and publishes it. His peers spend time analyzing it and trying to break it. They publish the results of their analysis. The original cryptographer may improve the algorithm based on the analysis, and eventually this back-and-forth may result in a very good algorithm.
Laws like the DMCA completely disrupt this process. Companies do not publish details of their algorithms, so anyone attempting to analyze them must first reverse-engineer them. Companies can then use the law to silence or intimidate analysts , because publication of how to break the algorithm (or even just details of how it works) could be considered illegal. Thus, cryptography researchers are intimidated into silence, and the normal research cycle is broken.
Techniques like reverse-engineering of circuitry (which are currently legal) can be used to circumvent copy-protection. However, they can also be used for other purposes, and banning the techniques themselves would have very negative consequences.
For example, a former employer of mine makes a business out of reverse-engineering integrated circuits and selling the extracted information. While this information could be used to violate intellectual property laws, it is in fact used to uphold those laws by providing evidence of patent or copyright infringement.
The same reverse-engineering technology developed by my former employer was also used to aid in the extraction of data from a flight computer used on the Swissair jet which crashed off the coast of Nova Scotia.
If techniques which could be used to circumvent copy-protection had been banned, it is very likely that no-one in Canada would have had the technology or expertise to extract the flight computer data, and everyone would have suffered because of the stifling of technology and research.
Deciding which technology can circumvent copy-protection is difficult. Are general-purpose PC's useful for circumventing copy-protection? Sure. Should they be banned? The logical extension of the desire to ban such technology is available in the dystopian essay "The Right to Read" by Richard Stallman (See "Further Reading" at the end of this article.)
Once a large percentage of music, video and text information becomes available only in copy-protected digital format, society will undergo radical and very negative changes. Here are some of the negative effects:
Current copyright law allows for fair use. It also places a time limit on the duration of copyright, after which formerly-copyrighted material reverts to the public domain. Copy-protection technology will subvert both of these legal protections. We've already seen how copy-protection technology interferes with fair use; now consider how it can subvert time limits on copyright.
If a corporation has old work which is about to become public-domain, it can simply refuse to allow copying of the work. It can use legal measures to prevent such copying by arguing that any technology which allows copying of public-domain material can also be used to copy material whose copyright has not yet expired. In effect, publishers would have a legal loophole to extend copyright indefinitely.
In my opinion, any work which cannot be copied should not be allowed to be protected by copyright. The whole reason for copyright laws in the first place was the recognition that intellectual property can be copied, and that authors needed legal protection. If publishers insist on technological protection, I believe they should be forced to waive their legal protection, or else we will end up with a situation in which copyright law is subverted for the benefit of publishers and the detriment of society at large.
If most content is in copy-protected digital format, the very existence of institutions like public libraries becomes impossible. This will have enormously negative consequences. Poor people who rely on public libraries for access to information will be deprived of that information. Scientific research will stagnate as researchers find it almost impossible to find scientific papers in a central location. Schools will be unable to provide children with affordable books, and school libraries will shut down.
Consider what would happen if a publisher of copy-protected scientific journals goes out of business. Suddenly, all of that scientific research would be inaccessible to most people, and recovering it all would be a tremendous burden. The danger to science of irretrievably losing information is enormous. This danger is recognized by some scientists, who are boycotting publishers which refuse to make scientific work freely available .
Because of the DMCA, a noted British computer scientist has refused to visit the United States for fear of persecution . If Canada passed a similar law, Canada would become a less attractive destination for researchers because of restrictions on their research.
Copy-protected digital content poses a huge threat to society, especially if it becomes widespread. Any law which encourages the spread of copy-protected digital content is a bad law.
In the United States, the Digital Millennium Copyright Act (DMCA) has already had negative effects. In addition to being used to threaten academics , it has also been used to charge the Russian programmer Dmitri Sklyarov with an offense which carries punishments of up to five years in jail and a $500,000 fine. Note that the alleged crimes took place in Russia, not the United States.
Sklyarov's "crime" was to produce software which breaks encryption on Adobe Systems Inc's e-book product. Because the encryption is so weak , it was quickly broken, and a gallery of Adobe remedies appeared . Rather than fixing the technology, Adobe decided to use the DMCA as a weapon. Sklyarov was arrested after he presented a paper describing his decryption program.
Although Adobe Systems Inc. originally asked for Sklyarov's arrest, they soon backed down after protests and called for his release . In my opinion, however, this is simply a cynical ploy to divert outrage from Adobe. Once Sklyarov is arrested, the U.S. Department of Justice is obligated to prosecute him, and he remains in jail as of this writing.
What is despicable about the Sklyarov case is that no-one has accused Sklyarov of actually violating copyright law. If he had hired a typist to transcribe one of Adobe's e-books, he would face only civil prosecution, and not criminal prosecution. Here, he faces criminal prosecution for actions he performed in Russia, and which do not even break copyright law.
The film, music, software and publishing industries are among the most successful and profitable industries today, even in the face of so-called piracy. Indeed, some industries are using the threat of piracy-prosecution as a marketing tool .
If these industries were suffering, and could demonstrate that piracy of digital content was making the difference between survival and bankruptcy, I would see some urgency in reforming copyright law. But the fact is that all of these industries are succeeding, and are as profitable today as they ever were. I therefore see their call for additional copyright protection as simply a money-making scheme without any socially-redeeming features.
Now it's true that digital content is easier to copy more faithfully than old-style books, tapes or videos. But that is the nature of the technology. If publishers cannot bend digital technology to fit their old business models, they must come up with new business models. Changing the law to force new technology to fit old business models simply hurts technology, hurts the legal system, and hampers business innovation.
Any law which provide sanctions against technology which circumvents copy-protection is bad for the following reasons:
This paper is Copyright 2001 Dianne Skoll. It may be reproduced verbatim, but changing it (except for translating it into another language) is not allowed.