The Great Canadian "Cache" Grab
Rethinking Browsing as Copyright Infringement

Jennifer Jannuska, 1997

eustus@MNSi.Net

The Great Canadian "Cache" Grab
Rethinking Browsing as Copyright Infringement

This paper deals with the copyright consequences of "browsing". "Browsing" is:

The act of looking through information by repeatedly scanning and selecting. An Internet browsing service presents a list of menu items or a page of information. After the user reads the information and selects an item, the service follows the reference and retrieves new information.⁽¹⁾

Recently, a number of copyright regulating bodies have toyed with the notion that persons merely browsing the Internet infringe copyright held by publishers of Internet websites.

In Canada, this threat was raised by the Information Highway Advisory Council Sub-Committee on Copyright in their final report, "Copyright on the Information Highway".⁽²⁾ The IHAC Report argued that,

browsing on the Information Highway entails the making of a copy; in order to browse, the work must be accessed. It is the SubCommittee's view that any act of accessing a work constitutes a reproduction.⁽³⁾

In researching this paper, I deliberately sought out those arguments that dissented from the current "wisdom" that browsing infringes copyright. I consider here the arguments that browsing is:

1. Not an act of reproduction;

2. An act of reproduction, which creates a merely ephemeral copy;

3. Alternatively, a form of reproduction that is permitted because there is an implied licence from the author/publisher of the information; and

4. That policy demands that the process of browsing be regarded as copyright-neutral, whether or not a copy is actually made.

I have concluded that sui generis legislation is required to protect the access of the public to documents on the Internet.

Copyright Act Provisions

27. (1) ... any person ... without the consent of the owner of the copyright, does anything that, by this Act, only the owner of the copyright has the right to do.⁽⁴⁾

The Act defines "infringing" as:

(a) ... any copy, including any colourable imitation, made or imported in contravention of this Act, or

(b) when applied to a fixation of a performer's performance in respect of which a performer's right subsists, or to a reproduction of such a fixation, any fixation or reproduction made or imported in contravention of this Act.

Internet "browsing" is not explicitly dealt with in the Act as either a form of infringement or as a type of "fair dealing."

These points will be revisited throughout this paper as we examine whether browsing creates a fixed copy, and whether browsing is a form of infringement. The next section of this paper lays another piece of the groundwork. In the next section, I will discuss the technical substructure of browsing.

The Technical Substructure of Browsing

The Web utilizes a "hypertext" formatting language called hypertext markup language (HTML), and programs that "browse" the Web can display HTML documents containing text, images, sound, animation and moving video. Any HTML document can include links to other types of information or resources ... Such "hyperlinks" allow information to be accessed and organized in very flexible ways, and allow people to locate and efficiently view related information even if the information is stored on numerous computers all around the world.⁽⁸⁾

There is a certain amount of confusion in the public as to the difference between browsing and "downloading" or "saving" Internet content onto a fixed disk. The IHAC Report acknowledges this conflicting information:

Those responding to the SubCommittee had widely divergent views of the definition of "browsing". Some parties included as "browsing" the ability to freely sample any database and extract information, as long as the data was not downloaded to a hard drive or printed. Others felt that viewing video or multimedia productions constituted "browsing". However, the creator community and information industry consistently stated that browsing constitutes an act of reproduction and is subject to the reproduction right.⁽⁹⁾

As I use the term in this paper, browsing refers to viewing material on the Internet. The "act of reproduction" alleged by the SubCommittee occurs because of a process called "caching," a common technical feature of modern⁽¹⁰⁾ Internet browsers.⁽¹¹⁾

Caching is the process used by Internet browsers of storing "browsed" material in the browser computer's RAM⁽¹²⁾ and on the browser computer's hard drive. The rationale and function of caching is explained by Steve Steinberg, a writer for Wired magazine:

The easiest way to improve Web performance is to not use the network at all. That's the idea behind caching, or storing copies of frequently accessed data somewhere close by. For example, browsers such as Mosaic and Netscape remember the last few websites you've accessed. That way, if you go back to a page, the data can be read straight from memory instead of from a computer across the country.⁽¹³⁾

Caching is an efficiency tool. It is something that a browser program does independently of the user. Cached copies are incidents of using a browser program.⁽¹⁴⁾

Most Internet users intuitively feel that caching is something different than downloading a file, although this distinction is hard to pinpoint at a technical level. Burger defines downloading as "the process of receiving a file from a remote computer."⁽¹⁵⁾ Downloading involves the making of a permanent copy on a disk. Caching also makes a copy on a disk.

Caching stores temporary⁽¹⁶⁾ copies of browsed documents, so that the user can view them again without having to request the page information again from the remote site. On Netscape Navigator, temporary copies of a browsed document are sent both to RAM and the hard disk. The rationale for these two caches is described in Netscape's online User Handbook:

Netscape retrieves a page from the memory [i.e. RAM] cache more quickly than from the disk cache, though retrieving from the disk cache is still faster than fetching from the network. The disk cache has the advantage of persistence. When you exit a session (quit the Netscape application), the memory cache is emptied, but the disk cache is maintained (and takes up space on your hard disk).⁽¹⁷⁾

The disk cache copy of a browsed page stored temporarily on the machine is technically no different than a downloaded file. The only difference between a disk cache file and a downloaded file is the agent of the copying. In the case of caching, the agent making the copy is the browser program. In the case of downloading, the agent making the copy is the human user. The only thing preventing a user from using the cache file in the way that s/he would use a downloaded file is what Jason Carr has called the "weird names" factor:

The cache tends to have weird names that foul up the process [of using cache files off-line]. [For example], the file my_sister.gif might be renamed 887hafs7 and only the browser cache manager knows what [name] goes to what [content].⁽¹⁸⁾

The amount of browsed material that is cached at any given time is limited by the allocation of space set in the browser program, either the default allocation set by the program itself, or a custom amount set by the user. Theoretically, the disk cache is limited only by the size of the hard drive in the user's computer.⁽¹⁹⁾

Browsers typically cache everything the user browses in either RAM or disk cache.⁽²⁰⁾ This can include text, graphics, mini-programs (called "applets"), and real-time video or audio files.

The lifespan of a cache file in RAM or on the hard disk is contingent on the behaviour of the user. If the user allocates a small cache space, the cache will be regularly written over. That is, as new sites are browsed by the user, the limited space in the cache will be erased and re-written, over and over. It can be helpful to think of the caches as chalkboards. Imagine the browser program writing new data onto the chalkboard (cache) every time something new is browsed. When the space on the chalkboard (cache) is full, the program goes back to the front and erases enough space to write in the new data, and so forth. The size of the cache allocation, thus, has a lot to say in how often the erasing and re-writing occurs.

There are two other factors which affect cache lifespan: 1. the natural end of the RAM cache, and 2. the manual intervention of the user. First, RAM only operates while the computer is on, therefore, RAM erases when the computer is turned off. There is no carry-over of RAM-cached material from one computer session to the next. Second, the user can manually clear both the disk and RAM caches. In Netscape Navigator, the user can go into the Options menu, select Network Preferences, and press the button to clear the disk or RAM cache.

There is one final note to add to this technical description. It is becoming increasingly common for sites to be "cache-proof". On such sites, the site designer has added a setting into the code of the site stating the site must be reloaded from the remote site every time the site is visited. This phenomenon is described by Steinberg:

Caching has one nagging problem. If the content of a website changes frequently, the cached version will quickly become out of date. So, to make sure users always receive the current version of a page, some sites send a special note that says, in essence, "Don't cache me."⁽²¹⁾

Cache-proof sites are, however, more the exception than the norm at present. Most site designers like caching because it allows a site to have more complex features and yet be easily reloaded at the user's end. Site designers generally figure in the presence of caching when they calculate how much complexity they can put into a site. Renowned website design expert, David Siegel, suggests that designers assume that caching is present rather than reduce file size and complexity.⁽²²⁾

Once a file is in the disk cache, the user can manipulate the file as s/he would any other file:

You can do anything you want with the cached files. Take out part and use it, make small changes and keep the rest, whatever you want if you have the software to do it with... Cached files ... can be easily manipulated.⁽²³⁾

Assuming the user can overcome what has been previously described as the "weird names factor," the user can print or fax the files, cut out pictures, e-mail files to the user's friends, and even put up a cached picture as "wallpaper" on the user's Windows 95 interface. Of course, if the user wants to avoid the cached file being deleted and written over in the process of future caching, the user must save the particular file to another part of the hard disk.⁽²⁴⁾

One possible use of disk cache copies is off-line browsing.⁽²⁵⁾ The current version of Internet Explorer allows off-line browsing:

[Microsoft's Internet Explorer] enables you to cache pages and graphics between Internet sessions, letting you work with Web content offline (for example, from your airplane seat.⁽²⁶⁾

RAM cache copies are somewhat different from disk cache copies. These copies exist merely for the purpose of the browsing itself. Since RAM is nothing more than a black box of memory inside a computer, it is really not possible to access RAM other than through the instantaneous use of the particular program on the computer.⁽²⁷⁾

Browser caching is a complex phenomenon that has proven very difficult for copyright lawyers. The next part of this paper will deal with the argument that caching is not an act of reproduction.

Browser Caching as an Act of Reproduction

To begin with, the myth should be dispelled that says that digital copying is inherently different in nature from hard-copy copying.⁽²⁸⁾ American courts have ruled that the digitization of an image creates a copyright-subject reproduction.⁽²⁹⁾ Furthermore, transmitting one of these digitized images over a bulletin board or over the broader Internet is one of the rights reserved to the copyright holder.⁽³⁰⁾ Canadian courts have considered other forms of digital representation to be copyright-subject, such as videotapes,⁽³¹⁾ audiotapes,⁽³²⁾ and computer software.⁽³³⁾

Beyond this, however, there is little guidance for lawyers on the legal status of browsing as an act of reproduction. The issue is currently up in the air. Browsing has generated some of the most vocal controversy in copyright law in recent memory.

Some commentators are convinced that copyright is not infringed when a user browses because reproduction does not occur:

[Internet browsing is analogous to] browsing a book in a library or a bookstore, or hearing buskers play music in the street. Disobeying the sign in a bookstore that tells one not to loiter by the magazine rack does not turn the speed-reader of the Economist cover-to-cover into a copyright infringer, any more than the audience of the busker infringes copyright by listening.⁽³⁴⁾

Other commentators are equally convinced that copyright is infringed by browsing. The September 1995 report of the United States Working Group on Intellectual Property Rights, Intellectual Property and the National Information Infrastructure, stated in a footnote:

³¹⁵ ... Activities such as loading a work into a computer ... may be infringement [of the copyright holder's exclusive rights].⁽³⁵⁾

The footnote refers to a United States decision which dealt with the issue of RAM copies, MAI Systems Corp. v. Peak Computer, Inc.⁽³⁶⁾

MAI considered the copyright consequences of booting up an operating system program into a computer's RAM for diagnostic purposes by an unauthorized service technician. Brunetti J. accepted the district court's finding of fact that:

A "copying" for purposes of copyright law occurs when a computer program is transferred from a permanent storage device to a computer's RAM ... "The loading of copyrighted computer software from a storage medium (hard disk, floppy disk, or read only memory) into the memory of a central processing unit ("CPU") causes a copy to be made. In the absence of ownership of the copyright or express permission by license, such acts constitute copyright infringement."⁽³⁷⁾

This decision is influential on the browsing question because it makes the factual conclusion that a copy in RAM is a fixed copy for the purposes of copyright infringement. If a RAM copy is sufficiently fixed to be copyright infringement, browsing with a cache inevitably leads to copyright infringement.

However, the conclusion reached in MAI has been controversial. Some authors have challenged the contention that a cache copy in RAM is fixed. The next section of this paper will canvass the argument that browsing merely creates ephemeral copies.

Browsing as Creating an Ephemeral Copy

A number of authors have argued that browsing creates a merely ephemeral copy, which should not be considered a form of copyright infringement. Andrew Grosso, for instance, likens RAM copies created by browsing to the "presence" of live sound through his home stereo:

If I tune in on my home stereo to a live performance of my favorite band, I am duplicating the band's music in my living room using the speakers on my stereo; however, the sound is not "fixed", and I am not infringing on the band's copyright. True, the sound is present in my living room, and it is a reproduction or duplication of a copyrighted work, but the sound is ephemeral, and not covered.⁽³⁸⁾

Grosso argues that the browsed content is merely ephemerally present in RAM, and there is no real act of reproduction.⁽³⁹⁾ Grosso disputes the finding in MAI that copyright is violated by the presence of content in RAM; he argues that RAM copies, which are destroyed when the computer is turned off, are not fixed.⁽⁴⁰⁾ Canada does not currently accept the doctrine of "ephemeral" copies.

The issue of "ephemeral" copies was raised in the Canadian case, Bishop v. Stevens.⁽⁴¹⁾ In Bishop, the question at bar was whether the defendant television station which had acquired a performance licence to perform the plaintiff's song could broadcast the performance from a taped copy which would thereafter be destroyed. The defendants argued that the reproduction on tape was a mere "ephemeral" recording and not fixed. In rejecting the defendants' argument, McLachlin J. determined that there should be no derogation from the strict words of the Act, which does not expressly permit ephemeral recordings as part of the broadcasting right:

Neither the wording of the Act, nor the object and purpose of the Act, nor practical necessity support an interpretation of these sections which would place ephemeral recordings within the introductory paragraph to s. 3(1) rather than s. 3(1)(d). On the contrary, policy considerations suggest that if such a change is to be made to the Act it should be made by the legislature, and not by a forced interpretation. I conclude that the right to broadcast a performance under s. 3(1) of the Act does not include the right to make ephemeral recordings for the purpose of facilitating the broadcast.⁽⁴²⁾

The judgment, in obiter, acknowledged the history of discussion of the ephemeral recording issue in Canada.⁽⁴³⁾ The proposal was made in the 1984 Government White Paper on copyright legislation reform, From Gutenberg to Telidon, to enact a provision protecting ephemeral recordings.⁽⁴⁴⁾ The proposal in the White Paper was subsequently rejected by Parliament. Ephemeral recording is expressly permitted in the Brussels round of the Berne Convention amendments, however Canada is not a signatory to that amendment.⁽⁴⁵⁾

The IHAC report obviously contemplates the possibility that accessing a work on the Internet may create a merely "temporary or ephemeral fixation". The Report says that:

It is the SubCommittee's view that any act of accessing a work constitutes a reproduction, even if it is a temporary or ephemeral fixation.⁽⁴⁶⁾

However, it is unclear how the SubCommittee thinks that the finding that a cache copy is a merely "ephemeral fixation" is possible, given the current state of Canadian law. The report does not recommend the adoption of any new legislation to protect ephemeral fixation, nor does it recommend that Canada sign the Brussels agreement.

In light of the Bishop decision, it is unlikely that Canadian courts will accept the argument that a RAM (or disk cache) copy is merely ephemeral. The courts would more likely follow the MAI decision and find that unauthorized caching makes a copy that is a form of infringement. In the Bishop decision, the court expressed its concern over even ephemeral copies:

A recording ... is permanent. It may be copied easily, privately, and precisely. Once a work has been recorded, the recording takes on a life of its own. This is why, from a composer's point of view, the right to control the circumstances under which the first recording is made is crucial.⁽⁴⁷⁾

Disk cache copies are rarely defended on the "ephemeral" argument. There is no question that a "material" copy is made when a file is fixed to a disk. Canadian law clearly prohibits unauthorized copying of software onto disks, for instance.⁽⁴⁸⁾ Nonetheless, there has been argument to suggest that even disk cache copies should be considered copyright-neutral.

Lewis Eisen suggests that the way a browser program functions should not determine whether a human user is liable for infringement:

Some browsers automatically cache the information to disk, others don't. Should it matter how the browser handles the data?⁽⁴⁹⁾

It is unclear whether browsing creates ephemeral or fixed copies. Moreover, it is unclear whether the doctrine of "ephemerality" even exists in Canada. Canadian and American orthodoxy would dictate that a copy is a copy, whether it is "present" on the user's system for a temporary purpose or whether it is permanently saved. On the other hand, legal pundits are beginning to raise questions about the propriety of enforcing this strict standard:

Should [caching] infringe copyright at all? Only if one reads the law very literally -- like the apocryphal computer program that translated "nous avions" ("we had") by "we airplanes." Copyright [was] never meant to stop people from repairing or reselling or reading or using material in customary ways.⁽⁵⁰⁾

Another challenge to the strict standard is raised in favour of implying a licence to cache a browsed page if the copy is merely for personal viewing and it is destroyed forthwith. The next section of this paper deals with the implied licence argument.

Browsing as Subject to Implied Licence

The rationale behind the argument states that web publishers publish their material on the Internet, which is, by its very essence, a distribution medium. In his now-famous essay on the movement of information in cyberspace, John Perry Barlow has described the distribution phenomenon:

The way in which information spreads is ... very different from the distribution of physical goods. It moves more like something from nature than from a factory. It can concatenate like falling dominos or grow in the usual fractal lattice, like frost spreading on a window, but it cannot be shipped around like widgets, except to the extent that it can be contained in them. It doesn't simply more on; it leaves a trail everywhere it's been.⁽⁵¹⁾

Barlow argues that, because the Internet is essentially a distribution medium, the price of posting content on the Internet is that the content, unbounded from its physical form, will be distributed. Nicholas Negroponte refers to this feature as the difference between "atoms" and "bits".⁽⁵²⁾

Because of the distributive essence of the Internet, many writers have argued that a certain amount of distribution should be presumed by people who publish content to the Internet. One Internet user has put the presumption into a question:

If the people who put it on the web didn't want you to use it for your personal use, why would it be there?⁽⁵³⁾

Minimally, publishers must intend that people will browse their site and that it will be cached in the process. Eisen states:

Since there is no way of viewing a webpage other than through your browser, one could be forgiven for assuming that there was an implied licence to copy in that manner granted by the author when he or she posted the material on the World Wide Web.⁽⁵⁴⁾

An implied licence works as a full defence to a charge of infringement, rebutting the presumption that the act of reproduction is unconsensual:

In order to constitute an infringement the act complained of must be done "without the consent of the owner of the copyright". Such a consent may be presumed from the circumstances. The inference of consent must be clear before it will operate as a defence and must come from the person holding the particular right alleged to be infringed.⁽⁵⁵⁾

Where the defence is successful, the infringement claim against the defendant is dismissed.

The argument in favour of an implied licence to browse is strong: there is generally no alternative to browsing for viewing Internet content. Moreover, there is often no way to find out the name of the copyright holder to contact for authorization without browsing the individual's webpage, which may already be a violation, under a strict interpretation. A violation necessary to prevent a violation is absurd.

The next section of this paper addresses the policy rationales for keeping browsing copyright-neutral.

Policy Rationales for Keeping Browsing Copyright-Neutral

The efficiency argument states that caching is an efficiency tool. Penalizing users for making use of software that does incidental caching will force users to use browser programs with caching mode turned off, which will inevitably force browser software companies to stop writing caching routines into their software. This would roll back the current state of technology. Another result of zero-cache operation would be greater overall traffic on the Internet. Greater traffic would probably result in some popular sites being inaccessible at peak hours and would probably result in slower service overall due to more bandwidth being tied up.⁽⁵⁶⁾

The consistency argument states that copyright law affecting browsing should be consistent with what is allowed for other media. Computer software has sui generis protection in the Act. Licenced users of computer software are permitted by law to make one backup copy for personal use.⁽⁵⁷⁾ Backup is considered to be an "incident" of fair dealing with software. Likewise, it would seem analogous to protect limited caching rights as legitimate incidents of browser software use.

The knowledge argument states that a minimum level of knowledge is needed in order to have infringed copyright. Since most browsers do not come right out and tell the user that copies are being made (this is known as a "transparent process"), many users are genuinely unaware that the cache is functioning while they browse. It would not be fair to impose liability on a user who genuinely did not know that copies were being made by her/his software.⁽⁵⁸⁾

And if a distinction in liability were to be merely contingent on the level of technical knowledge of the user, it would seem to create a perverse incentive for the user to learn nothing about the software.

The economics argument states that browsing with caching, far from being a form of piracy, is actually good for business. Caching makes it possible for users to come back multiple times to a site without enduring a slow load each time. Web designers want to make users come back to their sites, browse through freebie information and consider purchasing value-added components. Siegel calls this the "restaurant model":

You hear about a restaurant from an advertisement or a friend, or discover it while passing by. You check out the daily specials chalked on a board out front and smell the aroma in the doorway... When you are finished, you have dessert, ask for the bill, and pay. You leave a tip, pick up the card next to the cashier, and maybe exchange a few pleasant words with the owner. Later, when you are hungry again, you return or you don't, based on the quality of that first experience.⁽⁵⁹⁾

The alternative measures argument states that creators of content can protect their content from being copied or control access to copies through technological devices, rather than by laying out a legal spider web in which to catch unwitting insects. More effective in preventing unauthorized use, and cheaper than infringement lawsuits, is the use of alternative technologies.⁽⁶⁰⁾ There are numerous devices in this category, password-locking being the most common. For example, in order to use the Encyclopedia Britannica⁽⁶¹⁾ or the Wall Street Journal full-edition on-line,⁽⁶²⁾ users must register over the phone and pay a subscription fee. In exchange, the users receive a password that will enable them to log-in to the site for a certain period of time. Other more sophisticated forms of technology are also emerging which permit even greater control than a one-shot password-check. David Angell has described "digital-rights management technology":

These are software security solutions that track the actions performed on content by an end user. Digital-rights management systems can embed code that provides proof of ownership and copy protection. The systems can detect when alterations are made, track the movement of content through serial numbers, and meter content so the appropriate royalties can be paid to the information provider.⁽⁶³⁾

Digital-rights management technology would protect copyright on the Internet by putting the onus on the creator to control the distribution of their content, rather than relying on the end user to respect the copyright. This onus would tie in with the presumption of an implied licence. Any Internet publisher not making use of protective technologies would be presumed to licence individuals to browse the site freely.

Finally, the public domain argument states that the Internet has given to humankind an unprecedented medium for communication and exchange. Putting blanket copyright restrictions on browsing would chill access to this medium, and would prevent the democratization of information that many think is possible through the Internet:

The greatest interest at stake [in the Internet copyright debate] is not that of the copyright holders or the database companies, rather it's the interest of a free and open society has in ensuring that everyone can learn the facts, regardless of their economic background. We should regard our nation's investment in the knowledge of its citizens, and in their ability to participate knowledgeably in an open society, as the highest and most precious intellectual property we can ever protect.⁽⁶⁴⁾

The final section of this paper will contain my recommendations and the conclusion to this examination of browsing and copyright.

Conclusion and Recommendations

As I have noted variously throughout the paper, the legal position of browsing is unclear in Canada. It is unclear whether a RAM cache is a fixed copy. It is unclear whether a doctrine of "ephemerality" exists in Canada. It is unclear whether a defence of implied consent/implied licence will be approved by Canadian courts. This paper has deliberately avoided another complex and uncertain part of the browsing question that merits a study of its own -- the contributory liability of browser software companies.

In this paper, I have canvassed four arguments. Here are my conclusions with respect to each of these arguments:

Browser caching as an act of reproduction - I am persuaded that browsing is indeed an act of reproduction. However, not all prima facie acts of reproduction need violate copyright.⁽⁶⁶⁾ I have tried to argue that browsing is not an infringing act of reproduction.

Browsing as creating an ephemeral copy - Ephemerality is, as I mentioned, one area that is unclear in Canadian copyright law at present. I am convinced that cache copies (particularly RAM copies) could be considered ephemeral copies, if this doctrine existed in Canada. However, in light of the Bishop decision, it should be assumed that ephemerality as a doctrine does not exist in Canada.

Browsing as subject to implied licence - This argument is, in my opinion, the strongest legal argument permitting browsing as a non-infringing act. The defence, however, is greatly subject to the good graces of an individual judge. Until we have a more informed judiciary, it seems unlikely that implied licence will be a unimpeachable defence on which to rely.

Policy rationales for keeping browsing copyright-neutral - I have argued that there are numerous policy reasons for upholding browsing as a reasonable use. Common sense would dictate that courts should protect the rights of the public to access the Internet by rejecting over-zealous claims of infringement by the publisher community,⁽⁶⁷⁾ what I have termed in the title of this paper as a blatant "cache grab."

For all of these reasons, I am not satisfied with the proposition of the IHAC SubCommittee that no new legislation is necessary to deal with the Internet and copyright. Unfortunately, the Act in Canada has established a pattern of sui generis legislation to the point that any medium or process not appearing in special "fair dealing" exceptions is considered to be bald-faced infringement.⁽⁶⁸⁾ This exception-to-the-rule drafting has become particularly acute in the final stages of Bill C-32,⁽⁶⁹⁾ and there is no reason to deny that this trend exists. Given this trend, I am convinced that new legislation is necessary to protect the rights of the public to access not-otherwise-protected documents on the Internet. The legislation I propose would add a paragraph to s. 27(2) stating that it is fair dealing for a browser software program to make RAM or disk caches of an impermanent nature, provided that these copies are not otherwise manipulated by the user.

The principle of "View Plus" has been articulated by Grosso. Grosso's principle says that mere caching should not be subject to copyright penalties, unless there is a "plus" activity done by the user:

We must look toward a principle that can be called "View Plus", that is, if you download a file, and then do something more with that file than merely look at it, you have infringed the copyright. Running a downloaded software program, re-transmitting copies over the Net to other parties (without the author's permission), and modifying a work in a significant fashion, are all "pluses" that, when added to a "view", may permit us to say with a safe measure of legitimacy that somebody's copyright has been violated.⁽⁷⁰⁾

Unfortunately, special legislation is necessary to give the courts the sure footing that they seem to require to find that no copyright infringement exists.

Contrary to the IHAC Report's finding, I argue unequivocally that browsing should not be a litigeable act of reproduction. I do not make this argument on behalf of a limited special interest group of Internet radicals and free-loaders but as argument for rationality in dealing with this medium in the interests of all of society. On the Internet, the creator-consumer distinction is becoming less relevant. It is in all of our best interests, as potential creators and consumers on this new medium of exchange, to demand better balancing in our copyright law between access and exclusivity.

Notes

1. D.E. Comer, The Internet Book (Englewood Cliffs, N.J.: Prentice-Hall, 1995) at 286.

2. Information Highway Advisory Council, Final Report of the Sub-Committee on Copyright (March 1995) [hereinafter "IHAC Report"]. Available at http://info.ic.gc.ca/info-highway/reports/copyright/copy_e.txt.

3. Ibid.

4. "Anything that only the owner of the copyright has the right to do" refers to the listed rights in section 3.(1).

5. "The verb browse is derived from the behaviour of hungry animals who, in winter when pasture is barren, forage for tender shoots and the buds of trees and bushes." N. Negroponte, "Caught Browsing Again" (1996) 4.05 Wired 200 at 200.

6. I borrow this expression from R. Carelli, "Low-Tech Supreme Court to Explore Cyberspace to Rule on Limiting Net" The Detroit News (16 March 1997) 2A.

7. 929 F.Supp. 824 (3^rd Cir. June 11, 1996). Available at http://www.aclu.org/court/cdadec.html.

8. Ibid. at para. 33.

9. Supra note 2.

10. Two browsers currently make up 90% of the browser market: Netscape Navigator and Microsoft Internet Explorer. Both of these browser packages make use of caching. See D. Siegel, Creating Killer Web Sites: The Art of Third-Generation Site Design (Indianapolis: Hayden Books, 1996) at 22. Siegel also has a website at http://www.killersites.com.

11. A "Browser" is "a program specifically designed to help users view and navigate hypertext ... a Web browser," The New Hacker's Dictionary, version 3.2.0, available at http://www.datamation.com/PlugIn/humor/jargon/jargon_16.html#SEC23.

12. "Random-Access Memory is the electronic stratchpad and 'thinking space' where your programs and data reside while you are working with them." J. Burger, The Desktop Multimedia Bible (New York: Addison-Wesley, 1993) at 79.

13. "Speeding Up the Web" (1995) 3.12 Wired. Available at http://www.hotwired.com/wired/3.12/departments/geek.page.html.

14. See Burger, supra note 12 at 75.

15. Ibid. at 133.

16. Netscape says that "no pages are permanently stored in a cache." Netscape Communications Corporation, Netscape Handbook (1996) [hereinafter Handbook], available at http://home.netscape.com/eng/mozilla/3.0/handbook/docs/appans.html#C14.

17. Handbook, ibid. Similar processes occur in Microsoft's Internet Explorer. See the description in Explorer's Technical "White Paper" for the Windows 95 version, available at http://www.microsoft.com/ie/support/docs/tech30/perf.htm.

18. J. Carr, E-mail reply to Usenet post by the author, "Re: Cache Questions for a Paper", originally posted to comp.infosystems.browsers.misc on 7 March 1997, 10:45 a.m. (Copy on file with the author).

19. See "Nedell", Reply to author's question posted to "Experts Exchange" website on 8 April 1997. Experts Exchange is located at http://www.experts-exchange.com/. (Copy on file with the author).

20. Carr and "Nedell" have confirmed that everything browsed is cached. Carr has noted that "this may be user-definable in [Microsoft Internet Explorer], Opera, [and others]." See Carr, supra note 18 and Nedell, ibid.

21. Supra note 13.

22. See "About Caching," supra note 10 at 62.

23. See "Nedell", supra note 19.

24. Disk cache files go to a particular directory, usually set up automatically when the user installs the browser program, called something like "c:/netscape/cache/". Any file that is in the cache directory is "up for grabs" to be re-written by the browser. To permanently "save" a file from the cache, a user would have to copy or "move" the file to another directory.

25. Off-line browsing refers to the ability of the user to read website content without the user's Internet connection being active.

26. "White Paper", supra note 17.

27. See Burger, supra note 12 at 79.

28. S.L. Haynes argues that:

An idea is protected at the point it is placed in physical form. That "physical form" can be digital expression... Fixation in electronic media is fixation for copyright purposes. [Merely] because Net-based writing is impermanent and because [there is a] natural tendancy of members of the Net culture to ignore others' legal rights, [it does not follow that] copyright law is outmoded.

"Economy of Ideas: Round Two" (1993) 2.06 Wired, available at http://www.hotwired.com/wired/2.06/departments/rants.html (emphasis mine).

29. See Playboy Enterprises, Inc. v. Frena, 839 F. Supp. 1552 (M.D. Fla., Dec. 9, 1993) [hereinafter Playboy].

30. In Canada, the right of transmission is statutorily protected in the Act, s. 3(1)(f). In the United States, transmission was considered in Playboy, ibid. The liability for transmission by a third-party bulletin board was also considered in Religious Technology Center v. Netcom On-Line Communication Services, Inc., 907 F. Supp. 1361 (N.D. Calif., Nov. 21, 1995) [hereinafter RTC].

31. See Tom Hopkins International Inc. v. Wall & Redekop Realty Ltd. (1985), 1 C.P.R. (3d) 348 (B.C.S.C.).

32. See ATV Music Publishing of Canada Ltd. v. Rogers Radio Broadcasting Ltd., [1982] O.J. No. 571 (H.Ct.J.). Also R. v. Miles of Music Ltd. (1989), 74 O.R. (2d) 518 (C.A.).

33. See Delrina Corp. v. Triolet Systems Inc., [1993] O.J. No. 319 (Gen. Div.).

34. D. Vaver, "Rejuvenating Copyright Digitally" in Intellectual Property Secretariat of the Department of Justice, Symposium on Digital Technology and Copyright (Papers delivered at a Conference at Willson House, Meech Lake, 3 March 1995) 1 at 3 [hereinafter Symposium].

35. Working Group on Intellectual Property Rights, Intellectual Property and the National Information Infrastructure (Washington: U.S. Patent and Trademark Office, 1995) (Co-chairs: B.A. Lehman & R.H. Brown) at footnote 315. The report is available on the Internet at http://www.mc.hik.se/~mp93ks/ipnii.txt.

36. 991 F.2d 511 (9^th Cir., April 7, 1993) [hereinafter MAI].

37. Ibid. at 517.

38. A. Grosso, "Copyright and the Internet: A Footnote, a Sleight of Hand, and a Call to Reason" (January 1997) Fed. Law. 44 at 44.

39. A similar argument is asserted by P. Samuelson, "The Copyright Grab" (1995) 4.01 Wired, available at http://www.hotwired.com/wired/4.01/features/whitepaper.html. "The temporary storage of a copyrighted work in a computer's memory should not be regarded as an infringing reproduction." Also see L.S. Eisen, The Canadian Lawyer's Internet Guide (Pixley Press in Cooperation with the Law Society of Upper Canada (Continuing Legal Education), 1995) at 149.

40. See Grosso, supra note 38 at 44.

41. (sub nom. Télé-Métropole Inc. v. Bishop) (1990), 31 C.P.R. (3d) 394, 72 D.L.R. (4th) 97 [hereinafter Bishop cited to C.P.R.].

42. Ibid. at 408 (emphasis mine).

43. See ibid.

44. See ibid. at 406.

45. See ibid. at 400. List of the signatories to the Brussels Convention Relating to the Distribution of Programme-Carrying Signals Transmitted by Satellite (1974) available at http://www.wipo.org/eng/general/copyrght/brussels.htm.

46. IHAC Report, supra note 2 (emphasis mine).

47. Bishop, supra note 41 at 404 (emphasis mine).

48. See Act, s. 27(2)(l).

49. Eisen, supra note 39 at 150.

50. Vaver, in Symposium, supra note 34 at 5.

51. J.P. Barlow, "The Economy of Ideas" (1993) 2.03 Wired. Available at http://www.hotwired.com/wired/2.03/features/economy.ideas.html.

52. See N. Negroponte, Being Digital (NY: Vintage Books, 1996) at 11.

53. "Nedell", supra note 19.

54. Supra, note 39 at 150.

55. H.G. Fox, in The Canadian Law of Copyright and Industrial Designs, 2nd ed. (Toronto: Carswell, 1967) at 339, as quoted in Bishop, supra note 41 at 409.

56. Carr, supra note 18: "[Zero-cache browsing] means more traffic, in general, moving over the net." See also the technical description of packet exchange in Steinberg, supra note 13.

57. See Act, ss. 27(2)(l) and (m).

58. In RTC, supra note 30 at 1369, Whyte J. refused to find the defendant service provider contributorily liable for infringing the plaintiff's copyright stating that:

[The] Plaintiffs' theory would create many separate acts of infringement and, carried to its natural extreme, would lead to unreasonable liability... Although copyright is a strict liability statute, there should still be some element of volition or causation which is lacking where a defendant's system is merely used to create a copy by a third party. (emphasis mine).

59. Supra, note 10 at 28.

60. H.P. Knopf, "Copyright and the Infoway: Catalyst for Progress or Cause of Gridlock?" in Symposium, supra note 34 at 21 at 54 has stated that: "If an owner does not want a work to be browsed, it can be made available only on "pay" services, or encrypted."

61. Available at http://www.eb.com/.

62. Available at http://www.wsj.com/.

63. D. Angell, "The Copyright Question: Making the Net Safe and Profitable for Copyrighted Content" (January 1997) Internet World, available at http://www.iw.com/1997/01/copyright.html.

64. M. Godwin, "Copyright Crisis" (March 1997) Internet World 100 at 102. Mike Godwin is staff counsel of the Electronic Frontier Foundation, at http://www.eff.org.

65. "URL" stands for "uniform resource locator." See Comer, supra note 1 at 304. A URL is an address on the Internet that usually begins: http://www...

66. Knopf, supra note 60 at 54 has stated that: "[Browsing as an act of reproduction] should be deemed to be non-infringing."

67. Knopf, ibid., has argued that:

[Creating a new reading/browsing right] would be a nightmare of complexity. There is no "browsing right" in bookstores, and users would see this as an intolerable and unnecessary extension of a monopoly.

68. Consider the fair dealing privileges (s. 27(2)) themselves. The detailed exemptions exist specifically for:

(a) any fair dealing with any work for the purposes of private study or research;

(a.1) any fair dealing with any work for the purposes of criticism, review or newspaper summary, if

(i) the source, and

(ii) the author's name, if given in the source,

are mentioned;

[etc.]

69. The full text of the final version of Bill C-32 is available at http://www.parl.gc.ca/bills/government/C-32/C-32_3TOCE.html. Note s. 29, which sets out the new fair dealing provisions.

70. Grosso, supra note 38 at 45.

• Jennifer Jannuska's e-mail
  
• Back to LoGIC's homepage...