[Cite as: Mark A. Lemley,
Rights of Attribution and Integrity
in Online Communications,
1995 J. ONLINE L. art. 2, par. ___.]
{par. 3} Reality, alas, is more complex than even the very best science fiction. Consider the fate of Card's heroes were they to venture onto the Internet in the present day.
{par. 7} Lawyers are mercifully absent from Ender's Game. But in the
real world, those who live, work and play in cyberspace cannot afford to
ignore the problems that surround issues of attribution, identity, and
anonymity. Each of the three scenarios listed above presents a different
legal issue--the existence and scope of rights to attribution, the integrity
of "electronic personality," [FN 2] and the use of
anonymity, respectively.
{par. 9} The reason plagiarism is not necessarily copyright infringement has to do with the fact that copyright law does not condemn all copying--only the copying of certain things, under certain circumstances. There are several important limitations on the scope of copyright protection. One is the idea-expression dichotomy, arguably the most fundamental rule in all of copyright law. The short version of the story is that it is permissible (indeed, even encouraged) to copy the ideas of another, but illegal to copy the original expression of those ideas. A second significant limitation on copyright is the fair use doctrine, which makes it permissible to copy even the original expression from a copyrighted work under certain circumstances. Finally, some works may be copied because they fall into the public domain, and are "free for the taking" by subsequent authors.
{par. 10} To be sure, much plagiarism is also copyright infringement. If you take a chapter from my book and copy it verbatim, you are guilty of copyright infringement. But that is merely because you copied; in theory, nothing hinges on whether you attributed my work to me, or claimed it for yourself. If you are careful what you copy, so that you stay within the boundaries set by the copyright law, you are free to take the heart of this or any other article and use it in your own work. There is nothing I can do about it.
{par. 11} For many authors, the absence of an attribution right in United States copyright law is not very important. If authorship is primarily a commercial venture, very little hinges on attribution. The fact that someone else sells your concept for a screenplay for $3 million will not bother you any less because the "playwright" acknowledges you in a footnote. What you wanted was the $3 million. But for other authors, particularly those in the academic community, attribution may be more important than the right to commercial control. Certainly, I don't get paid by law reviews for the articles I publish; my return comes (if at all) in the form of peer recognition of my ideas. From my perspective, the Harvard Law Review is more than welcome to copy this article in its entirety, as long as they put my name at the top. On the other hand, I am likely to be very upset if Harvard publishes my article with someone else's name on it.
{par. 12} Why is this relevant to the Internet? Basically, the answer is that the Internet makes it dramatically easier to republish an author's works. Academic discourse between multiple participants used to take place in the rather stilted enviroment of academic journals, where months and even years would elapse between argument and rejoinder. In this environment, attribution was a practical necessity--before you made your argument, you usually had to remind your audience what you were responding to, often summarizing the prior paper at length. By contrast, Internet discussions are typically characterized by rapid responses from a host of people, by the dissecting and excerpting of original comments for discussion purposes, and by the cross-posting of ideas and discussions to other lists.
{par. 13} In all of this confusion, it is a relatively common occurrence for people's words (and certainly their ideas) to be reproduced without attribution. Indeed, even Card's heroes in Ender's Game suffered such a fate. Their early postings as Locke and Demosthenes were replicated in other areas of the Net, but were not attributed to the authors. [FN 3] If authors want to be identified with their ideas, they need something more than copyright law will give them. That "something more" may be a right of attribution--a right to be identified as the author of your writings.
{par. 15} In the first, Waldman Publishing v. Landoll, [FN 4] both Waldman and Landoll published adaptations of classic, public domain books and marketed them for children. Waldman claimed that it had put substantial work into creating its children's versions of the classic books, and that Landoll merely copied Waldman's chapter arrangement and large portions of its text. Because the underlying works were in the public domain and could be freely copied, however, Waldman apparently decided that it could not make out a case of copyright infringement.
{par. 16} Instead, it claimed that Landoll's sales of books based on Waldman's adaptations constituted a "false designation of origin" in violation of the Lanham Act. The Second Circuit agreed, finding that because Waldman had indeed contributed original expression to the public domain works, Landoll was required to credit Waldman as the originator of the children's editions. "[P]ublishing a work original enough for copyright protection, without attribution to its author, would constitute a false designation of origin." [FN 5]
{par. 17} In the second case, Robinson v. Random House, [FN 6] Random House represented Daley, an author who wrote a book about Pan Am airlines. Random House sued Robinson, who published a competing book which included approximately 25% of Daley's text verbatim. After finding that Robinson had infringed on Daley's protectable expression, the district court proceeded to consider Robinson's claim of fair use.
{par. 18} The court found Robinson's failure to give attribution for the material taken from Daley's book doomed his fair use claim: "In this case, although a significant portion of nine out of fourteen chapters in Robinson's book was taken directly from the Daley Book, Robinson fails to quote the Daley Book, to cite to the Daley Book, or even to acknowledge the Daley Book. This reprehensible conduct places Robinson far closer to the scissor-wielding cut-and-paste plagiarist than to the scholar building on others' past works. Second, confusion between the original and the infringing work is also an issue to consider ... In this case, a reader would have absolutely no way of knowing that thousands upon thousands of the words used in the Robinson Book actually were penned by Daley." [FN 7] Based largely on this reasoning, the court found that Robinson's use of Daley's language was commercial rather than academic, and so rejected his fair use claim.
{par. 19} Both of these cases suggest that a right of attribution does in fact exist in United States law, at least in circumstances where the failure to attribute has a commercial effect. [FN 8] If you copy material without identifying it as copied, that fact may hurt you in a copyright infringement suit. Even if you cannot be sued for copyright infringement, your failure to attribute copied material may be actionable under the Lanham Act if it confuses consumers as to the source of the copied material. [FN 9] Taken together, these cases suggest that the worst cases of nonattribution will be taken care of by the existing law.
{par. 20} Other cases, however, are likely to fall between the cracks of copyright and trademark law because they have no obvious commercial consequence. E-mail conversations are prominent in this latter group. Excerpts taken from someone else's e-mail message may be sufficiently short or unoriginal that they do not constitute copyright infringement. Alternatively, the taking may be considered "fair" because it is for the purpose of criticism or comment, and not for financial gain. Finally, the nature of many e-mail conversations is clearly noncommercial, a fact that both precludes a finding of trademark infringement and makes a finding of "fair use" more likely in copyright cases.
{par. 22} The formulation of any right of attribution is complicated by the ease with which works can be altered online. Because it is possible not just to edit but also to change the content of someone else's posting, a right of attribution to be effective might also have to include a right of integrity.
{par. 23} Indeed, the Berne Convention provides for authors in signatory countries (with the de facto exception of the United States, which has signed the Berne Convention but has not altered its intellectual property laws to comply in this respect) "the right to claim authorship of the work and to object to any distortion, mutilation _ or modification of, or other derogatory action in relation to, the said work, which would be prejudicial to his honor or reputation." [FN 12] The United States has a similar provision protecting the integrity of works of visual art, but does not extend such protection to other types of copyrighted works. [FN 13]
{par. 24} Rights of attribution and integrity would not necessarily prevent the unauthorized modification of a prior work, however. Rather, the question in any case of copying begins with "did you attribute the work to the original author?" If you have not only copied but also changed the work, it should be up to the reviser to make clear what belongs to the original author and what she has added, changed, or deleted. This is not very difficult in many contexts. We have a number of conventions from print media--quotation marks, ellipses--which identify preexisting material taken by others. Similar conventions are developing online. For example, if I want to incorporate a pre-existing message into my e-mail, I may begin the excerpt by typing "T.S. Eliot writes: . . ." If I edit Eliot's comments, I can indicate that by writing "[snip]" where material has been elided.
{par. 25} Of course, in other contexts identifying the work of the original author will not be so easy. Musicians and visual artists obviously cannot use quotation marks to identify the authors whose work they appropriate. [FN 14] They may be able to credit the original artist in a side note, but even in that circumstance it may be impossible to distinguish effectively between original and derivative elements in the work. What do we do in those instances? If it is impossible to distinguish the parts of the two works, so that audiences will be likely to misattribute authorship, a right of integrity can take one of two tacks.
{par. 26} First, it might give the original artist the right to disavow any connection with the adapted work, and require the adapter to disclose that fact. (On this view, it would be the adapter's responsibility to contact the original artist to seek permission before publishing the derivative work). Alternatively, the legal rule could be that if you can't create an adaptation without engendering confusion as to authorship, the original author has the right to prevent their work from being used at all.
{par. 27} Other problems with a right of attribution generally parallel the difficulties the United States and other countries have faced in giving protection to "moral rights." The moral rights protection which the Berne Convention compels is personal to the author, and cannot be transferred under any circumstances (except to the author's heirs when she dies).
{par. 28} Though designed to benefit authors, a non-transferrable right may end up hurting them in some circumstances. If buyers such as publishers are unable to buy the moral rights from authors whose works they want to acquire, they may be willing to pay less for the work than they otherwise would. Certainly, the existence of moral rights makes it more difficult to create new works based on the works of others, since the number and nature of permissions that must be obtained increases significantly. And, because the right of attribution contemplated in the Berne Convention is not subject to a "fair use" or private use exception, the effect of a full right of attribution will be to limit the use the public can make of copyrighted works, albeit in a fairly minor way.
{par. 29} Whether a right of attribution makes sense on line depends
in large part on what we think is at stake. If copyright and trademark
law will protect against willful or commercial nonattribution, perhaps
the informal customs of the Internet ("netiquette") and common courtesy
can be relied upon to do the rest.
{par. 31} Perhaps there is a legal solution to this problem. If forging Joshua Quittner's electronic mail address involved improper access to his computer system, charges could be pressed under various federal and state criminal "computer trespass" laws. But Quittner identifies in the article ways in which his identity might be forged on the net itself, without the forger having improper access to his computer. [FN 16] Alternatively, representing oneself falsely as Joshua Quittner might constitute trademark infringement, depending on the purpose for which one did it and the effect it had. Because the trademark laws are designed to protect only against the confusion of consumers in commerce, however, most citizens could not avail themselves of such protection if their online identities were hijacked by noncommercial messages.
{par. 32} A more promising area for legal liability is that of communications torts. In particular, taking over someone's online persona might in various circumstances constitute fraud or misrepresentation, defamation, or invasion of privacy. But fraud and misrepresentation, like trademark infringement, require some sort of commercial or other injurious effect. The fake Joshua Quittner must have induced others to act to their detriment on the basis of this false identity. That may well be the case in certain circumstances, but many intrusions on online identity lack that element, and so would not be actionable.
{par. 33} Defamation is a better claim; the fake Joshua Quittner has arguably published a false statement (the assertion that he is in fact Joshua Quittner). But even defamation requires that the plaintiff be "defamed," that is, be cast in disrepute or held up to mockery or ridicule; moreover in some circumstances the plaintiff will have to prove some actual damage. Both of these may be difficult to show, meaning that defamation provides only uncertain protection against impersonation.
{par. 34} Invasion of privacy appears to provide the strongest cause of action for someone whose online identity was appropriated by another. Two separate torts for invasion of privacy are relevant here. The first is the tort of "appropriation of name or likeness." This sounds promising--certainly the "appropriation" of Joshua Quittner's name is precisely what has occurred here. As Prosser and Keeton put it, "the effect of the appropriation decisions is to recognize or create an exclusive right in the individual plaintiff to a species of trade name, his own, and a kind of trade mark in his likeness." [FN 17]
{par. 35} Unfortunately for our purposes, the common law tort of appropriation of name or likeness--and the state right of publicity which has evolved from it in the past 50 years--are limited to circumstances in which the name of the plaintiff has been appropriated for commercial purposes. [FN 18] This makes the "appropriation" privacy tort even less useful for impersonation plaintiffs than trademark infringement or defamation: not only must you prove that the defendant made money from using the plaintiff's name, but that it was the "commercial value" of the plaintiff's name itself that was taken.
{par. 36} That leaves us with the second prong of the privacy tort, governing "publicity which places the plaintiff in a false light in the public eye" ("false light" privacy). [FN 19] It is here that the impersonation tort must rest. False light privacy originated in the English case of Byron v. Johnston, [FN 20] in which the poet Lord Byron convinced the English equity courts to enjoin Johnston from circulating a "bad poem" which Johnston had falsely attributed to Byron. Since then, it has been used to protect plaintiffs from having their name linked to books, articles, political statements, or legal documents which they did not in fact author. [FN 21] So if Mr. Quittner can identify the person (or bot) using his identity online, he ought to have a cause of action for invasion of privacy, regardless of what is said using his name. [FN 22]
{par. 37} But we are not concerned only with Joshua Quittner. What about Valentine in Scenario 2? Here, the difficulty is that no one has published articles using the name "Valentine"--instead, both the original author and the "impersonator" used the pseudonym Demosthenes. Can Valentine prevent a third party from "impersonating" her pseudonym? Is it realistic to say that Valentine has been "placed in a false light in the public eye"--or alternatively, that "Demosthenes" has?
{par. 38} It is here that I believe the law fails to take account of the realities of Internet discourse. Both defamation and invasion of privacy are bound tightly to the concept of the person. Just as it is impossible to defame the dead, [FN 23] it appears to be impossible to defame (or portray in a false light) a fictional entity. Thus, in circumstances where objectionable publications use a pseudonym, courts generally require that the plaintiff prove that the public know that the objectionable meaning attached to him personally. [FN 24]
{par. 39} As a result, "Demosthenes" could be defamed or placed in a false light only if the public knows who she really is. Similarly, corporations can be defamed, but only because it is a known collection of persons which is assumed to have prestige and standing in the community. [FN 25] This helps those who are easily identified with their pseudonyms, but it does nothing to aid those who truly wish to remain anonymous (or pseudonymous, as with Valentine).
{par. 40} The failure of the law to protect pseudonymous personalities online ("electronic personalities" or "epers," in the words of Curtis E.A. Karnow [FN 26]) can do serious damage to what Howard Rheingold calls "the virtual community." [FN 27] It is one of the virtues of electronic interaction that individuals may in a large sense design their own personality, freed from the boundaries of physical appearance and social convention. Men can (and do) become women online; the old become young; lawyers and business people become artists and poets. [FN 28]
{par. 41} But for a community to develop online, these new electronic personalities must be able to interact, share experiences and confront adversity together on a consistent basis. This takes time, during which a particular electronic personality (which generally means a particular person) becomes a member of the community. Impersonation of electronic personalities--in Howard Rheingold's words, "violating the sanctity of nicknames" [FN 29] --prevents the development of that community. The result is damaging whether or not the impersonation is discovered. If it is revealed, people are less likely to trust messages from that particular electronic personality, because they cannot be sure of the authenticity of the message. If it is not revealed, the electronic personality is harmed because the community attributes to it messages which it did not author.
{par. 42} Protecting the integrity of electronic personalities would solve these problems. There are a number of informal social norms which make electronic impersonation "taboo." [FN 30] But informal social norms are effective only to the extent that the transgressors are members of the community. Some form of legal protection seems appropriate in those circumstances where a painstakingly developed electronic personality is hijacked by an outsider for his own ends. The obvious legal analogy to the physical world is the false light privacy tort. That tort protects "real" persons against impersonation, and it does not need much alteration to be retrofitted for the online world. All that is required to apply the false light privacy doctrine in the electronic context is for us to accept the electronic personality as "real."
{par. 43} But it is possible that we do not need to rely on the law
at all to protect electronic personalities. In his companion article in
this issue of the *Journal of Online Law*, Michael Froomkin suggests that
by combining cryptography and anonymous remailers, individuals can create
verifiable but untraceable "digital signatures" for their pseudonyms. [FN
31] Using this technology, it should be possible to create a pseudonym
which cannot be hijacked without access to the actual author's private
encryption "key."
{par. 45} On a less dramatic but no less important note, early women writers often used pseudonyms which disguised their gender so that their books would be read. The historical accomplishments of anonymity are laudable, and should not be disregarded lightly. At the same time, anonymity is the key to the successful commission of crimes and torts against others. Anonymous Internet users can commit copyright infringement, fraud, and defamation with impunity online, and can conspire with others to commit physical crimes in the virtual world.
{par. 46} There is no easy way out of this conundrum, particularly since it seems extremely unlikely that any one government has the technical ability to prevent the private use of encryption devices and anonymous remailers. Various people have proposed different solutions, ranging from complete anonymity to proposals to give the government a limited right to determine identity [FN 32] to banning anonymity entirely [FN 33] How you come down on the issue of anonymity probably depends in large measure on how you weigh the danger of these competing scenarios--in short, whether you fear anarchy more or less than an all-powerful government.
{par. 47} Perhaps surprisingly, the United States Supreme Court came down rather strongly in favor of a right to anonymity in the recent case of McIntyre v. Ohio Elections Commission. [FN 34] In that case, the court held that a state statute prohibiting the distribution of anonymous political campaign literature violated the First Amendment. The Court extolled the virtues of anonymity in literature and politics throughout American history, commenting that "an author generally is free to decide whether or not to disclose her true identity" and that, "at least in the field of literary endeavor, the interest in having anonymous works enter the marketplace of ideas unquestionably outweighs any public interest in requiring disclosure as a condition of entry." [FN 35]
{par. 48} Anonymity is more fully addressed by Michael Froomkin elsewhere in this issue of the Journal of Online Law. [FN 36] I do not propose to enter the debate over anonymity here, because I don't believe that rights of attribution and integrity of electronic personality depend on the status of a "right" to anonymity. The reason is simple: invocation of a legal rule based on misuse of identity necessarily requires proof of identity. If Locke and Demosthenes want to avail themselves of a legal right to have their words attributed to them, or if Peter and Valentine want to claim the exclusive right to hold themselves out as Locke and Demosthenes, they will have to be able to prove that they "are" in fact Locke and Demosthenes! Truly anonymous postings will therefore not be eligible for legal protection in the sense discussed in this article. This is true not because of any legal rule disfavoring anonymity, but by the very nature of anonymity itself.
{par. 49} To be sure, there are so-called "anonymous" remailers which strip the identity from messages, but log that identity in a file from which it can be obtained if need be. The notorious "anon.penet.fi" remailer in Finland works on this model. If Peter and Valentine use such a pseudo-anonymous mailer, they will be able to prove authorship (and nonauthorship) in court if necessary. But, by the same token, governments and private litigants will be able to find the same information, and so it is wrong to call their messages "anonymous" in a pure sense.
{par. 50} In short, Peter and Valentine may be able to protect their
putative rights to attribution and integrity in Scenarios 1 and 2, at least
under the legal regime I propose here, but only at the risk of removing
their anonymity and exposing themselves to liability in Scenario 3. There
is a rough sort of justice to this. Net citizens who take advantage of
true anonymity have in effect opted out of the legal system as a whole,
since they cannot be bound by its rules. Even if we wish to allow such
citizens the opportunity to "opt out," it is only fair to prevent them
from using the law to their advantage when it suits them--they cannot pick
and choose the circumstances under which they will abide by legal rules.
{par. 52} Each of the issues discussed in this Article could be resolved by the legislative or judicial creation of a new right--rights to attribution, integrity of personality, and anonymity, respectively. But I am not convinced that we need to go that far. Many cases of mis- or nonattribution can be resolved satisfactorily under the existing copyright and trademark laws. Other cases can be resolved by the use of informal social norms against plagiarism, particularly in the academic communities where attribution is of great importance. While society may eventually conclude that authors should have an unfettered right to attribution online, we should at least be cognizant of the problems such a right has engendered in other contexts.
{par. 53} Similarly, it is tempting to create a personal right to maintain the integrity of one's online personality. Changing the law to account for cyberspace poses less risk in this context, since we have various tort theories that seem ready-made for protecting integrity of personality. A minor modification to the right of privacy is less troubling than the creation of an entirely new right. Nonetheless, this new right may be unnecessary, particularly if Michael Froomkin is correct that encrypted pseudonyms can be both verifiable and untraceable.
{par. 54} Finally, we should be particularly wary about legislating
in the area of anonymity. There is no need for legal rules to create a
"right" to anonymity. The more likely possibility is that the government
will try to take anonymity away. The First Amendment (as interpreted in
the McIntyre decision) may limit the power of government to restrict
anonymity, but the United States government has not stopped trying to do
so. Regulating anonymity is a difficult task, and not one to be undertaken
lightly. With anonymity, as with much of the Internet, we must make sure
there is a problem before we look to the government for solutions.
FN 2: See Curtis E.A. Karnow, The Encrypted Self: Fleshing Out the Rights of Electronic Personalities, 13 J. MARSHALL J. OF COMPUTER AND INFO. L. 1 (1994).
FN 3: ORSON SCOTT CARD, ENDER'S GAME 147-48 (1977).
FN 4: Waldman Publishing Corp. v. Landoll Inc., 49 Patent, Trademark & Copyright J. (BNA) 251 (2d Cir. Dc. 27, 1994).
FN 5: Waldman Publishing v. Landoll at 251 (rejecting the "bodily appropriation" rule); but see Cleary v. News Corp., 30 F.3d 1255 (9th Cir. 1994) (requiring "bodily appropriation" of a work unchanged in order to violate the Lanham Act).
FN 6: Robinson v. Random House Inc., 49 Patent, Trademark & Copyright J. (BNA) 472 (S.D.N.Y. Jan. 18. 1995).
FN 7: Robinson v. Random House at 473.
FN 8: Semco Inc. v. Amcast Inc., 63 U.S.L.W. 2669 (6th Cir. April 18, 1995) (requiring commercial effect to invoke the Lanham Act).
FN 9: See also David M. Kelly and Linda S. Paine-Powell, Developments in American Unfair Competition Law; Implications for the Computer Industry, 4 EUR. INTELL. PROP. REV. 184, 189-90 (1995).
FN 10: See Eric Schlachter, IP Issues on the Net, electronic mail to CNI-COPYRIGHT listserv, March 20, 1995, 5:16 p.m.
FN 11: See Berne Convention for the Protection of Literary and Artistic Works, July 24, 1971, Paris Text, art. 6bis.
FN 12: Berne Convention, id. at art. 6bis(1).
FN 13: The Copyright Act, 17 U.S.C. sec. 106A (Supp. V 1993).
FN 14: See Christopher R. Harris, Manipulation of Photographs and the Lanham Act, 16 COMM. & L. (1994) (addressing copyright law difficulties posed by the computerization of photographic images).
FN 15: See also HOWARD RHEINGOLD, THE VIRTUAL COMMUNITY: HOMESTEADING ON THE ELECTRONIC FRONTIER 181 (1993) (discussing impersonation of identity).
FN 16: Joshua Quittner, Automata Non Grata, WIRED 119, 121 (April 1995).
FN 17: W. PAGE KEETON, PROSSER & KEETON ON TORTS 854 (5th ed. 1984).
FN 18: Waits v. Frito-Lay, 978 F.2d 1093 (9th Cir. 1992) (holding that the imitation of a professional singer's distinctive voice in an advertisement is actionable voice misappropriation); White v. Samsung Electronics, 989 F.2d 1512 (9th Cir. 1993) (dissent from denial of rehearing, criticizing the majority for extending the tort of misappropriation to a celebrity's "identity," without use of the celebrity's name or likeness).
FN 19: KEETON, supra note 17, at 863.
FN 20: Lord Byron v. Johnston, 35 Eng. Rep. 851 (1816).
FN 21: KEETON, supra note 17, at 863-64.
FN 22: Time, Inc. v. Hill, 385 U.S. 374 (1967) (extending the tort of invasion of privacy to include actions in which the defendant "fictionalized" an account of the plaintiff's experience), but see Cain. v. Hearst Corp., 878 S.W.2d 577 (Tex. 1994) in which Texas abolished the tort of false light privacy.
FN 23: KEETON, supra note 17, at 778.
FN 24: Carlisle v. Fawcett Publications, Inc., 20 Cal Rptr. 405, 410-11 (Ct. App. 1962) (requiring the plaintiff to establish that defamatory statements were of an concerning him).
FN 25: KEETON, supra note 17, at 779.
FN 26: Karnow, supra note 2, at 4.
FN 27: RHEINGOLD, supra note 15, at 5.
FN 28: RHEINGOLD, supra note 15, at 232-33 (noting the fantasy aspect of electronic interaction).
FN 29: RHEINGOLD, supra note 15, at 181.
FN 30: RHEINGOLD, supra note 15, at 181.
FN 31: A. Michael Froomkin, Anonymity and Its Enmities, 1995 J. ONLINE L. art. 4, par. 31-35.
FN 32: George P. Long III, Comment, Who Are You? Identity and Anonymity in Cyberspace, 55 U. PITT. L. REV. 1177, 1200-09 (1994) (recommending that anonymity be preserved, albeit on a limited basis); Karnow, supra note 2, at 8-11 (explaining that the legal system may pursue a legal fiction who abuses the system, or it may disregard the fiction and pursue the individual); see Anne Wells Branscomb, Anonymity, Autonomy, and Accountability: Challenges to the First Amendment in Cyberspaces, 104 YALE L.J. 1639, 1661 (1995) (discussing net users' methodology for determining what behavior is appropriate and enforcement of their own standards without outside legal intervention).
FN 33: I. Trotter Hardy, The Proper Legal Regime for 'Cyberspace', 55 U. PITT. L. REV. 993, 1050-51 (1994); see Anderson v. Nidorf, 26 F.3d 100 (9th Cir. 1994) (upholding CA's antipiracy statute, which requires the author of reproduced recordings to be labeled on the tapes before they are sold).
FN 34: McIntyre v. Ohio Elections Commission, 115 S.Ct. 1511, 1516 (1995) (holding that an author's decision to remain anonymous is an aspect of the freedom of speech protected by the First Amendment).
FN 35: McIntyre v. Ohio Elections Commission, 115 S.Ct. at 1516.
FN 36: See Froomkin, supra note 31.
FN 37: Branscomb, supra note 32, at 1661-62 (discussing excommunication of abusive users from cybercommunities); see Hardy, supra note 33, at 1054 (recommending a presumption in favor of flexible, decentralized rule-making).