Mike Montour (En anglais seulement)

Information archivée dans le Web

Information identifiée comme étant archivée dans le Web à des fins de consultation, de recherche ou de tenue de documents. Elle n’a pas été modifiée ni mise à jour depuis la date de son archivage. Les pages Web qui sont archivées dans le Web ne sont pas assujetties aux normes applicables au Web du gouvernement du Canada.

Mike Montour (En anglais seulement)



Les documents reçus seront affichés dans la langue officielle dans laquelle ils auront été soumis. Toutes les suggestions sont affichées comme elles ont été reçues par les ministères; toutefois, toutes les informations sur les adresses ont été enlevées.

Suggestion de Mike Montour reçue le 16 septembre 2001 par courriel

Objet : Comments on Canadian Copyright Reform Process

Mike Montour
(Adresse enlevée) September 15, 2001
Comments - Government of Canada Copyright Reform
c/o Intellectual Property Policy Directorate
Industry Canada
235 Queen Street
5th Floor West
Ottawa, Ontario

Comments on Canadian Copyright Reform Process

1. Introduction

I appreciate the opportunity to present my comments regarding the Canadian Copyright Reform Process. I speak as a citizen of Canada, as a developer of software and Internet technology, and as a consumer of commercially produced content (listed in decreasing order of importance). I shall begin with some general statements of principles, then some background information about the modern technological landscape. I shall then use this context to discuss possible changes to the Copyright Act. I apologize for the length of the background sections, however I feel that this material is important in order to make informed judgments about the relevant matters.

I shall use the term "content" to refer to the information comprising a work controlled under the copyright act (for example a piece of music, a movie, or a novel), excluding the physical medium (if any) on which that information is stored.

I shall use the term "publisher" to refer to a person or organization that sells or broadcasts content on a commercial basis.

I shall use the term "user" to refer to any person who interacts with published content, OR with any technological device that can be used for the storage, creation, manipulation, or distribution of content.

2. General Principles

I feel that the Copyright Act should be amended to reflect the existence of new technologies such as the networked personal computer. However, I strongly oppose many of the modifications which have been proposed or implemented in recent years.

Measures such as the Canadian levy on blank recording media and the American DMCA serve the interests of large international corporations, but do so at the expense of the rights of ordinary citizens. Such measures also spill over and affect areas of society that have nothing to do with the Copyright Act.

I strongly urge the Government of Canada to develop sensible copyright regulations that take into account the technological realities of the modern world, to ensure that such regulations consider the rights of individual citizens as more important than corporate financial interests, to reject copyright regulations that have detrimental side-effects on other areas of society, and to take a leading role in asserting these principles at the international level.

3. Motives

When considering any changes to copyright regulations, it is very important to consider the source of those changes. In the modern world, large and influential corporations own the rights to a large fraction of the copyrighted content that is distributed commercially. Consider how the citizens of Canada will appear to such a corporation. Some will be shareholders and employees, some will be the creators and original rights-holders of content, but the majority will be nothing more than potential consumers. The company's goal is to act on behalf of its shareholders, to collect a maximum amount of revenue from the consumers while paying a minimum amount out (for example, to purchase the rights from the creators of original content). Note that I do not criticize this business model; I merely state it as a "law of nature".

The success of this business model depends critically upon government regulations such as the Copyright Act. Therefore, it is to be expected that corporations will campaign strongly for changes to those regulations in order to maximize their revenues.

As a Canadian citizen, I assert that my government has a fundamental obligation to me and my fellow citizens, to judge any proposed changes based on their net benefit (or cost) to society. If viewed in this light, it will be clear that many of the proposed changes to the Copyright Act must be rejected. I include in this category regulatory changes that have been introduced at the international level (e.g. WIPO or Free Trade treaties); I assert the government has an obligation to re-negotiate or withdraw from any international agreements that are not in the overall best interests of Canadian citizens.

4. Technological Realities

The introduction of modern technology such as the networked personal computer has fundamentally changed the social and economic environment, as has happened in the past with the introduction of technologies like the automobile or the printing press. In each case, businesses that had been profitable before the introduction of the new technology were not guaranteed to stay profitable after its introduction. However, those who studied and understood the new technologies were able to stay or become profitable.

Here are some modern technical realities:

4.1 - The "meaning" of a collection of digital bits depends on the context in which they are interpreted. Therefore, one collection of bits can have more than one meaning.

4.2 - Any collection of digital bits can be stored, duplicated, and distributed without any need to impose any further "meaning" beyond that of "raw bits".

4.3 - There can be many different collections of digital bits that all have the same "meaning".

4.4 - Storage, duplication, and distribution of "raw bits" is a natural and necessary process in the context of a networked personal computer.

4.5 - In a well-designed cryptographic system, all of the security is in the key and none is in the algorithm.

To illustrate point (4.1), consider the following number (line breaks inserted for ease of reading):

4856570896573978293098418946942861377074420873513579240196520736 6869851340104723744696879743992611751097377770102744752804905883 1384037549709987909653955227011712157025974666993240226834596619 6060348517424977358468518855674570257125474999648219418465571008 4119086259716947970799152004866709975923596061320725973797993618 8606316914473588300245336972781813914797955513399949394882899846 9178361001825987901031601961835034344895687053845208538045842415 6548248893338047475871128339598968522325446084089711197712769412 0795862440547161321005006459820176961771809478113622002723448272 2493232595472346880029277764979061481298404283457201463489685471 6908235473783566197218622496943162271666393905543024156473292485 5248991225739466548627140482117138124388217717602984125524464744 5055834628144883356319027253195904392838737640739168912579240550 1562088978716337599910788708490815909754801928576845198859630532 3823490558092032999603234471140776019847163531611713078576084862 2363702835701049612595681846785965333100770179916146744725492728 3348961600064758591746278121269007351830924153010630289329566584 3662000800476778967984382090797619859493646309380586336721469695 9750279687712057249966669805614533820741203159337703099491527469 1835659376210222006812679827344576093802030447912277498091795593 8387121000588766689258448700470772552497060444652127130404321182 610103591186476662963858495087448497373476861420880529443

This number is expressed in decimal form, however it can trivially be converted into a sequence of binary digits ("bits"). What is the "meaning" of this collection of bits?

One meaning is purely mathematical - this happens to be a 1401-digit prime number (one that cannot be divided evenly by any number smaller than itself). There are infinitely many prime numbers (starting with 2, 3, 5, 7, 11, 13, ...) and this just happens to be a member of that sequence.

However, there is another meaning that can be applied to this collection of bits, and that is that it represents the source code of a computer program, compressed with the "gzip" utility. The sequence of bits representing this prime number can be fed into a "gzip" de-compression program, and the output will be an ASCII text representation of a computer program. Incidentally, this happens to be the famous "DeCSS" program that can render the contents of an encrypted DVD viewable by a user. In other words, another "meaning" of this collection of bits is that it is an illegal "circumvention device" under the American DMCA.

Note: to avoid possible legal liability, I have randomly transposed 4 pairs of digits in the above decimal representation. Thus the number shown is not actually prime, nor does it decode to the "DeCSS" program. More information about the real "illegal prime number" can be found at http://primes.utm.edu/curios/page.php?number_id=953

I hope this example illustrates my point, that a collection of bits does not necessarily have a single, unique meaning. This leads to the very important principle (4.2), because digital storage media like CD-Rs, magnetic tapes, hard disk drives, or flash memory cards, store only collections of bits. There is no longer any direct association between the type of storage medium, and the meaning of the bits stored upon it. Before the digital age, there was such an association - a vinyl record would almost always contain music, while a VHS tape would usually contain a movie. But now, there is no reason to think that a CD-R contains music - it is just as likely to contain a functional software program, or a collection of digital photographs, a table of historical stock-market prices, a combination of these, or absolutely anything at all!

As an aside, it should be clear by now that the current levy on blank recording media (paid to certain members of the music industry) is inappropriate and grossly unfair. Even if one assumes that a fraction of all blank CD-Rs are being used to store content in violation of copyright regulations (which, by the way, is contrary to the general presumption of innocence established in the Charter of Rights and Freedoms), there is no reason to believe that the content in question is music. Some blank CD-Rs are undoubtedly used to store still images or functional software, and there is absolutely no reason for the music industry to receive any "compensation" for these activities!

Just as there are many meanings for a given collection of bits, so too there can be many different collections of bits that all have the equivalent "meaning" (4.3). Consider a track from a music CD - this is a representation of a time-sequence of sound pressure levels, that when played through a loudspeaker will reproduce the desired piece of music. The application of certain logical and mathematical operations to the original sequence of bits can produce a different (and smaller) collection of bits in the so-called "MP3" format. This "MP3" can also be played through a loudspeaker to reproduce the desired piece of music, with only a minimal degradation in sound quality. There are many other examples of ways in which a collection of bits may be transformed into another, without changing substantially the abstract "meaning" of those bits.

Principle (4.4) is particularly interesting when dealing with "copyright", because the principle is that it no longer makes sense to count individual "copies" of a collection of bits. Consider the data flows in the following hypothetical scenario, where I send my brother a picture of my new dog:

1. I take the picture with a digital camera. The camera generates in its dynamic memory a table of numbers, representing the light intensity at each point of the image.

2. The camera applies the "JPEG" compression algorithm to represent the same image with a much smaller set of numbers (thereby losing some details of the image).

3. The collection of bits resulting from the JPEG compression is stored on a memory card in the camera. The table of numbers representing the un-compressed image is erased from the camera's dynamic memory.

4. At some point, I connect the camera to my computer. The data bits travel in sequence over the cable, pass through buffers in the computer's dynamic memory, and are written to sections of the computer's hard disk drive.

5. I compose an e-mail to my brother, and I select the image file as an "attachment". This causes the data bits to be loaded from the hard disk into the computer's dynamic memory, then converted to an alternate representation in "MIME base64" encoding.

6. When I "send" the e-mail, the bits making up the message (some of which are the encoded representation of the image) are written back to my hard disk as a "sent mail" item. Those bits are also transmitted over my network connection as "IP packets" (a small collection of bits, with each packet representing one section of the message). Each IP packet will be copied into the memory of several intermediate network devices, before finally reaching a server belonging to my internet service provider (ISP).

7. If my ISP is unable to deliver the message immediately, the message will be written to the hard disk on the ISP's mail server.

8. Eventually, my ISP's mail server will be able to make a network connection to the mail server of my brother's ISP. The message will be loaded from disk into dynamic memory, transmitted over the network as IP packets, then stored to disk again at my brother's ISP.

9. At some point, my brother will connect to his ISP and initiate a network transfer of the message to his computer's hard disk.

10. My brother will view the message, causing it to be loaded from his hard disk into his computer's dynamic memory. His computer will read the "MIME base64" encoding of the message, and convert this back to the sequence of bits representing the JPEG-encoded image file.

11. Another program will process the bits representing the JPEG-encoded image, and will undo the encoding to produce a table of numbers representing light intensity at each point on the image.

12. This table of light-intensity values will be copied from the computer's main memory to a special region of memory on the computer's graphics card. From here, it will be used to manipulate the intensity of light on my brother's computer monitor.

In this scenario, a "static" copy of the image exists in 6 places, on various hard disks and on the camera's memory card. If my brother chose to save the attachment as a JPEG file, that would be a 7th "copy". If either ISP happened to perform a routine backup of their servers, even more copies of the image could end up stored for months in an off-site tape archive. Also note how the same "image" is represented in many different ways along this chain: as a "raw" table of light intensity, as a JPEG, as a MIME base64 attachment, as a sequence of IP packets on a network, then all the way back again.

I hope it is clear that "copying" is a natural, necessary, and automatic process in the world of networked personal computers. It does not make sense for the Copyright Act to attempt to restrict this process, either directly or by granting blanket approval for manufacturers to implement "anti-circumvention" technologies that cannot be bypassed under penalty of law. Instead, the Copyright Act should restrict the distribution of content from one corporation or user to another (e.g. prohibit user-to-user distribution). It is also not reasonable to impose any liability on the ISPs or network operators for the "copying" which takes place through their networks.

It should also be understood that the monetary "value" of a CD will be different than it was 10 years ago. Then, the volume of digital information on a single CD could not be stored easily on an alternate storage medium. Therefore, part of the value of a CD was the physical storage medium for the digital information. However that aspect no longer exists, with compression technologies like MP3 encoding, and vast increases in the capacity of storage devices.

The "value" of content will also tend to decrease over time, as an increasing number of unauthorized copies are gradually made and distributed. However, this trend already exists; a movie is worth more in its first week of theatrical release than when it appears as a "Sunday Afternoon Classics" television broadcast. Publishers will have to accept that the monetary value of a piece of content will slowly decrease toward zero over time. I do not see this as a bad thing; if content has more value when new, it will give publishers an increased incentive to sponsor the creation of new content.

Principle (4.5) relates to cryptographic methods, such as those used to protect secure web transactions (SSL protocol) or to prevent an American DVD from being played on a British DVD player (CSS algorithm). The strongest protection methods involve an openly-known mathematical algorithm to scramble the original data based on the value of a "key". This "key" is a number, which is like the combination of a safe. There are so many possible numbers that nobody will ever guess randomly the one number that unlocks the safe or unscrambles the original data. For some reason, many industries have chosen not to use the strong, openly-available algorithms, and have instead implemented their own secret "in-house" algorithms. Almost always, these "in-house" algorithms have later been discovered to be seriously flawed so that they can be broken without having to guess the correct key.

Rather than fix the fundamental problems in their flawed implementations, companies have instead found it more economical to lobby governments for legislation that criminalizes the analysis and discussion of the flaws in their proprietary encryption algorithms. This is very dangerous, and is not in the best interests of anyone except possibly the shareholders of those corporations.

In Hans Christian Andersen's classic fairy tale "The Emperor's New Clothes", the title character is sold a garment with a magical property, namely that it was invisible to "everyone who was unfit for the office he held, or who was extraordinarily simple in character". Of course the "garment" did not actually exist except in the imagination of those who were told of it, and eventually a small child finally spoke aloud what had been obvious to all. (an on-line version of this story is available from Project Gutenberg at ftp://sailor.gutenberg.org/pub/gutenberg/etext99/hcaft10.txt )

There is a similarity between the emperor's "new clothes" and some of the "digital rights management" technologies that have recently been announced. In both cases, the actual quality of the item is far less than the claims that are made about it (claims made by the seller, and believed and repeated by the buyer). However there are two important differences. One, this shocking lack of quality is not visible to everyone, but only to those with a basic understanding of the technology. And two, legislation such as the DMCA has made it a crime for anyone to speak out as the child did in the story.

A Russian programmer and researcher, Dmitry Sklyarov, was recently arrested for violating the DMCA after he gave a presentation in the United States exposing the fundamental weakness of certain "e-book" protection technologies (this is an over-simplification; see for example http://www.eff.org/alerts/20010808_eff_sklyarov_alert.html for more information regarding this case). Other security researchers have chosen not to publish their observations of other technologies, fearing prosecution under these same laws.

The Government of Canada must not pass any such laws here! We as a society cannot let companies suppress basic research and academic discussion, simply to avoid the extra time and effort required to develop content-protection technologies that actually work.

5. A Fair Balance

I believe that something like the following would be a fair way to balance the interests of rights-holders against the interests of ordinary citizens, given the modern technological realities.

All digital recording media would be considered to be "Digital Paper", in other words, a generic substance that any citizen could purchase, own, and use for any purpose (just as is the case with real paper). The current levy that is collected on behalf of the music industry would be abolished. The Copyright Act would grant to the publisher the exclusive right to control original distribution of content to users, whether by selling a physical medium containing the content (a CD or VHS tape), by providing individual access over the Internet (e.g. downloading an MP3 file), or by broadcasting the content publicly (TV, radio). A "compulsory license" might apply to re-transmission of broadcast content. Once a user had legally obtained content in one form, the user would have the right to store, duplicate, process, transform, and use that content in any way he or she wished, without the knowledge or consent of the publisher. However, the user would not be permitted to make the content available to any other users except under certain limited circumstances (such as playing a CD at a private party). If the user had originally purchased the content on a physical medium, the user would be permitted to sell that physical medium to another user. However, the first user would thereby give up any further rights to store or use the copies of that content that he or she had made. If the original user had not obtained the content on a physical medium (i.e. the user downloaded a file), that user would not be permitted to give or sell that content to any other user. The user would not be permitted to transfer to any other user a modified version of the content (such as an MP3 created from a CD), even if the second user would be entitled to create such a modified version (e.g. he owned his own copy of the same CD). The Copyright Act would have to recognize that 100% enforcement of these regulations could not be obtained without massive privacy violations. A user would always be able to e-mail an encoded music file to another user, just as that user could send a duplicated cassette tape or a copy of sheet music to another user through the postal mail. If we do not require Canada Post to open every envelope and check for copyrighted material, we can not require internet service providers to perform such activities on digital messages. Also, it is important to recognize that not every person who receives an unlicensed copy of a work, would have otherwise purchased a licensed copy of that work. Although publishers might use this formula to estimate their "losses" due to piracy, there is no reason to believe this reasoning. In cases where a user made content available on a public Internet site (e.g. "Napster"), publishers would have a limited ability to have the user's ISP disconnect or remove the site. However, due process would have to be followed, there would have to be a process for the user to appeal such an order, and there would have to be significant penalties for making false accusations. In cases where a company used technological means to prevent an action that would otherwise be legal, a user would have the legal right to circumvent that technology. For example, a visually-impaired person would have the right to use any necessary means to convert an "e-book" into a format that could be fed into a text-to-speech converter, even if the publisher of the e-book had installed technological measures so that the text could only be viewed on a computer screen. "Circumvention devices" would only be illegal if it could be shown by the publisher that the device had no legal uses. Academic research into "circumvention devices", including reverse-engineering, would have to be permitted. The right for anyone to perform such activities and discuss the results is more important than a corporation's "right" to obtain a particular profit margin. However I would support a provision prohibiting the disclosure of any cryptographic keys used by anti-circumvention technology, as long as all other aspects of the algorithms could be discussed openly. The Copyright Act would recognize that any person has a right to create content, and to make full use of any technological devices to store, process, and distribute that content. Manufacturers who build copy-protection or usage-restriction technologies into their products would have to let users turn off these technologies if the user didn't want to use them for his or her own content. For example, manufacturers would be prohibited from selling a VCR that would refuse to let a user copy a videotape, if that videotape was of the user's wedding. The term of copyright would not be extended beyond the current time limit (lifetime of author + 50 years), in order to stimulate the on-going creation of new works. The danger is that as the term of copyright increases, publishers will hold the rights to an increasing amount of material, and may find it more profitable to re-sell old material rather than sponsoring the creation of new material. This is particularly likely if publishers employ technological measures to enforce a "pay-per-view" model of content distribution rather than a "sale" model.

6. Conclusion

I urge the Government of Canada to consider very carefully any possible changes to the Copyright Act, and to implement only those measures that reflect an understanding of the technical realities of the modern digital world, and also protect the rights and interests of Canadian citizens over the financial interests of publishing corporations.

I hope that my thoughts and opinions are useful, and I would be willing to clarify any points that may not have been presented clearly above. Also please note that I reserve the right to change my opinions in the future, particularly if confronted with new information or perspectives on these issues.

Comments on "Consultation Paper on Internet-based Retransmissions"

I am in favour of a compulsory copyright license for Internet-based retransmissions of over-the-air radio and television signals. I am in favour of amendments to the Copyright Act to expressly provide for the technologically-neutral application of the License, but subject to appropriate territorial restrictions and other reasonable conditions.

Regarding territorial restrictions, I suggest that this could best be accomplished by requiring that Internet-based re-transmitters maintain a record of individual users including the user's geographic location, then to provide to that user only content approved for distribution in that area.

One method of determining a user's geographic location would be if the re-transmitting company collected a payment or regular subscription from the user, and could thereby verify the user's billing address. However I do not believe that payment should be a requirement.

Another method of determining a user's geographic location would be by means of a number of approved third-party verification services (a precedent for which exists in certain age-verification agencies that control access to adult-themed Internet sites). Perhaps a service such as Canada Post's "www.epost.ca" (which provides services like on-line bill payment) would also be interested in providing location-verification services.

Use of a third-party location-verification service would have benefits to the user:

1.The user would only have to maintain an account with one agency, but could then access on-line content from a large number of Internet-based re-transmitters.

2.The user's true identity would not need to be made available to the re-transmitters, instead an anonymous account identifier could be used. However, law-enforcement officials would still be able to locate the user by accessing the records of the location-verification service.

Certain technological means of enforcing territorial restrictions have been proposed, however it is almost certain that any such technological means can be circumvented by other technological means. Therefore, technological location-enforcement would require strong "anti-circumvention" legislation, and as I have stated earlier I strongly oppose such legislation because of its harmful side-effects. Also, in the event of the circumvention of location-enforcing technology, it might be very difficult for the re-transmitter to trace the ultimate destination. Therefore, I feel that despite the additional administrative burden, enforcement by identification of end users is a preferable method, and will give foreign rights holders an increased confidence in the safety of making their content available to Canadians.

Banner advertising

It is difficult to address this issue in a technology-independent manner. However, as a general principle I would suggest that section 31(2)(c) be interpreted to mean that when delivered to the user, the re-transmitted content must be de-coupled from any further advertisements, and that the delivery of the re-transmitted content could not be conditional upon the user's viewing these further advertisements.

A specific technological example would be as follows: upon accessing the web site of the re-transmitter, a user could be presented with banner ads. However, once the user signed in and activated a particular content stream, that content stream would appear in a distinct window or frame free from additional advertising (that could, for example, be re-sized to fill the entire screen). Additionally, the user would be able to employ ad-blocking software such as "Junkbuster" (http://www.junkbusters.com/ijb.html) without losing the ability to receive the content stream.

Restrictions on banner advertising would most likely reduce the willingness of certain companies to become Internet-based re-transmitters of content. However, I believe that restrictions of this type are in the best interests of both the original rights holders and the users of such technology.

Further re-transmission by end users

I do not feel that this is a significant concern, and I do not feel that it would be appropriate to require technological measures in order to prevent end users from further re-transmitting content. I believe the existing copyright regulations already provide sufficient protection for rights holders.

Over-the-air broadcast is normally the last "release window" for any given content (following theatrical and video releases). Therefore, by the time that anything appears on an over-the-air broadcast, it is likely that unlicensed copies will already be available over the Internet. Therefore, a small amount of further "spillover" from Internet-based retransmissions of the over-the-air broadcast is unlikely to have a significant financial impact on the original rights holder.

Also, once a signal is transmitted over the air, it is technically possible for any user to receive that signal, convert it to digital form, and make it available over the Internet. Indeed, it would be more likely for this to happen than for a user to further re-distribute an Internet-based re-transmission (assuming that the Internet-based re-transmission service could track individual user accounts, and identify anyone who was forwarding the content stream).

"Simultaneously and in its entirety"

This phrase should be understood to include the reasonable time-delays and data modifications associated with distribution over the Internet. It should be noted that satellite-based transmission systems also introduce a time delay (both in converting the signal, and in transmitting the signal to and from the satellite).

As for the question of subsidiary signals, I feel that Internet-based re-transmitters should make available some subsidiary signals such as closed captioning and secondary audio programs. However I would not go as far as to say that all subsidiary signals must be presented.

Mike Montour

Partagez cette page

Pour faire connaître cette page, cliquez sur le réseau social de votre choix :

Aucun appui n'est accordé, soit de façon expresse ou tacite, à aucun produit ou service.