Information Technology and the Law

As we talked about in class today, perfect internet filtering is impossible from two standpoints: (1) a technological one; and, (2) from the standpoint of the people, who won’t agree on one single definition of “obscenity.” If the filter is too strong, then you end up overblocking. If too weak, underblocking. Unfortunately, it’s virtually impossible to get the strength of the filter such that neither underblockage nor overblocakge occur. And Dan explained how they can be so frustrating, so much so, that he wrote a program to bypass it. In fact, they don’t work that great either. They may block some of your favorite non-obscene sites, while letting other obscene ones escape the filter’s detection mechanism.

Finally, we mentioned that internet filtering may not be the best method of preventing youth to see obscene images in the library. We discussed such preventative measures as parental supervision, requiring an account, or having to slide a magnetic library card, with age info, into a card reader before beginning an internet session. In his dissent Justice Stevens mentions many of these alternatives that were brought up at the district court level. Some of these include (1) the ability of the library to enforce internet use policies, and call the police if necessary, (2) restricting minors’ unfiltered access to terminals within view of library staff, (3) optional filtering, (4) privacy screens, and (5) recessed monitors. All these measures may in fact be considered “less restrictive,” that is they might take care of most of the problem, but without the negative effects of overblocking.

While Dan considers these filters to be a nuisance, Justice Stevens goes farther and calls the requirement of internet filtering a “statutory blunderbuss” that violates the First Amendment. He likens the hundreds of thousands or so overblocked sites to “constitutionally protected messages” and notes that such an overly broad restriction is not justified.

While I too believe that internet filters are a nuisance, I don’t necessarily have the same logic as Justice Stevens for why I don’t like them. Surely overblockage is a problem, for, if it wasn’t, then the strength of the filter could be cranked up such that almost all obscene sites are blocked, as well as many non-obscene ones. In other words, I don’t believe in a solution that also creates a large problem. I think that the intention of CIPA is a good one. I don’t think that it is wrong in requiring that libraries try to protect children from obscene images, but I don’t think that today’s internet filtering technology, while probably the least expensive option, is the best way to go about doing this. Congress needs to realize this, so that more effective and more appropriate measures can be put in place in public libraries.

As others (see this) have pointed out, bundling a web browser and an operating system can create efficiencies and yield benefits to the consumer. However, the court ultimately decided that Microsoft was wrong in bundling the two because it was destroying competition and thereby harming consumers. By bundling Windows and IE, Microsoft helped prevent middleware products such as Netscape Navigator and Java from becoming widely used by developers as a platform for applications, and by doing so, it prevented competition within the operating system market, since Windows has many more applications written for it than any other operating system. Therefore, in its decision, the court was not singling out web browsers as the only component of an operating system that must be removable. That the court decided that IE should be unbundled from Windows resulted from the specific facts of the case, and these facts indicated that the act of tying the two together would ultimately harm competition in the OS market.

So no, it was not normatively wrong for Microsoft to bundle the two together. In fact, because of the potential benefits to the consumer, it would have been entirely possible if Microsoft had tied the two together for those very reasons. In other words, they might have decided to bundle the two for technological reasons only, and not in order to prevent competition. Unfortunately, I don’t think a good intention would have changed whether or not the Sherman Act was being violated. However, Microsoft was well aware of the threat to their monopoly on the OS market, and acted accordingly, whether responsibly or irresponsibly I will not comment on.

Often in antitrust cases there is a fine line between exclusion and competition, and this case demonstrates that point. Microsoft’s action could be viewed as competitive since the bundling might have yielded benefits to the consumer. Its action can also be viewed as anticompetitive. Neither of these views is necessarily wrong, and in fact both may be correct. Yet Microsoft stated in its appeal that it did not engage in any exclusionary conduct. However, it seems pretty obvious how the tying of IE and Windows could potentially harm the OS market, and since Microsoft clearly tied the two together to protect its monopoly, Microsoft most certainly did engage in exclusionary conduct.

Perhaps if Microsoft had provided a better defense for why the two were integrated, it would have stood a better chance in court. But even then, the general incompetence of Microsoft’s witnesses wouldn’t have lent much credibility to those appeals. Especially when you get one of the executives admitting on the stand that prices for Windows are chosen without consideration of the prices of competitor’s products.

Study question 5 for April 13 states:

Suppose that in Eisenberg, rather than sending the disabling code update through the mail on a floppy disk, the update had been delivered by downloading, after asking the system’s user to “click here to download updated software.” Should this change in the facts change the result of the case?

My short response to that question is no, this change will not change the result of the case. My long response includes a discussion of the CFAA.

The 1994 revision of the CFAA is very broad in its definition of “transmission.” Under this broad definition, it is hard to imagine why an update delivered over the internet would have any less of an intent element than one delivered in the mail. Because the plaintiffs were receiving software update disks, it was reasonable for them to assume that the disks contained bona fide updates, and were not of malicious intent. In other words, the plaintiff never had reason to suspect that the defendants would place a time bomb or other program of malicious intent on the update disks. Therefore, it makes no difference whether the update disks are received via US mail, and then manually inserted and run in the machine, or if they are sent electronically and then executed with the push of a button.

Even if all previous updates had been delivered by mail, and only the one with the time bomb was delivered electronically, the defendant would still be in violation of the CFAA. It might be wise for the plaintiffs to be skeptical of any electronic updates received for the first time, but assuming that they take all precautions in trying to identify the origin of the “transmission,” and trust that the update was sent to them in good faith from the defendant, the plaintiffs can’t be held responsible for the fact that they assisted in the transmission getting installed on the computers.

The 1994 revision of the CFAA focuses very much on intent and less on the means through which the code is transmitted. As long as one can prove intent, then the specific means of transmission is not particularly relevant. In other words, the plaintiff would be no more careless in updating their software by clicking with a mouse than they were when they inserted the disks by hand. Due to these reasons, I would be very surprised if a court were to conclude differently based on the different methods of transmission. A conservative reading of the revised CFAA would certainly include both mail and internet as valid methods of transmission for a time bomb.

In class today there were several different opinions regarding Professor Felten’s 1st discussion question: “In thinking about whether to apply trespass principles to web pages, how much does it matter that people talk about pages using words like “site,” “visit” and “home” that are normally associated with real property?”

Alex noted that the similarities in terms are irrelevant since the fundamental architecture of the internet represents anything but a physical location. Some agreed, but others indicated that this didn’t matter, since many still think of web pages as physical places, citing the use of such aforementioned terms as evidence, understanding, or acceptance of this. I agree with Alex, and would like to elaborate on why I think this way.

First of all, it would be helpful to restate the basic law of trespass. As explained in The Ancient Doctrine of Trespass to Web Sites by I. Trotter Hardy, the Restatement of Torts defines the common law cause of action for “trespass” as follows:

One if subject to liability to another for trespass, irrespective of whether he thereby causes harm to any legally protected interest of the other, if he intentionally

a) enters land in possession of the other, or causes a thing or a third person to do so, or

b) remains on the land, or

c) fails to remove from the land a thing which he is under a duty to remove.

Let’s assume I am sitting in my dorm room on my computer. I open a web browser and browse http://www.eBay.com. A few seconds later the eBay homepage appears on my computer screen. Have I now entered eBay’s physical property? All I did was type a few keystrokes and press “enter.” If anything, eBay’s homepage has entered my physical property (my dorm room). But our typical notions of property preclude anything that can exist in two places at once. Moreover, if, after loading the homepage, I leave my computer for a period of 1 hour, and return, have I remained on eBay’s property for 1 hour? After I load the homepage, if I don’t click or type anything, then I certainly can’t be said to have remained on eBay’s property, for the simple fact that I haven’t downloaded more bits of information from eBay’s server onto my computer. Sure, I might get an occasional refresh full of advertisements, depending on the particular website. But the fundamental nature of the internet, from my perspective, is very different from anything I experience in the physical world. Each time I request x bits of information from a server, the request appears to me like an instantaneous one-way telephone conversation. Likewise, the reply, depending on how much information I’ve requested, appears as a very quick response to my request. But after the request if fulfilled, even if I don’t close the browser window (let’s assume that I start reading what I already have on my screen), time, with respect to the internet, ceases to exist, until I request more information, and thus the process repeats. In other words, the act of “remaining” on someone’s land that we think of with regard to physical property doesn’t fit well in the context of the internet.

This is not to say that trespass law can never apply to the internet. But in the form laid out in The Ancient Doctrine of Trespass to Web Sites by I. Trotter Hardy, the basic law of trespass to land doesn’t fit well with web pages. Of course, this still doesn’t address whether or not trespass is the best type of law to apply to the web pages. In other words, even if trespass law were restated so that web pages were clearly included as property that could be trespassed, copyright law may still be better suited to deal with these issues.

In class today one student drew an analogy between Lexmark and GM. The argument was that when an owner of a GM automobile replaces one or more parts with non-GM certified parts, that they may run the risk of voiding warranty coverage. In other words, GM doesn’t have to file suit with individuals because they are unhappy with the fact that GM owners may be using aftermarket third-party parts. They simply may (depending on the nature of part(s) in question) have the right to deny partial warranty coverage as a result. This business model completely contrasts with Lexmark’s business model for its T520/522 and T620/622 laser printers. Lexmark uses an authentication sequence with these models to ensure that only Lexmark-certified cartridges are used in these printers. I assume that the GM-Lexmark analogy was originally brought up because the student implied that Lexmark should not care whether an owner decides to use 3rd-party cartridges. However, Lexmark is not committing any moral wrongdoing by requiring that only Lexmark cartridges be used in these printers. Moreover, it is foolish to assume that a business model that works for the automotive industry will work equally well for the printer industry.

First of all, warranty is a considerably bigger issue in cars than in printers. I don’t have any statistics, but even subjectively I am confident that the percentage of warranty claims is much higher with cars than with printers. Thus, while the threat of losing warranty coverage (and the thought of resulting expensive repairs) may be a deterrent to install aftermarket 3rd-party parts on an automobile, it is most certainly not a deterrent for the owner of a printer. Furthermore, I doubt that parts make up a substantial portion of automakers’ revenue. With today’s competitive warranties, consumers are usually protected from paying for routine repairs for about 4 years. And after the warranty period expires, the consumer doesn’t have much incentive to use certified parts, if a cheaper 3rd-party alternative exists, simply because the new part is unlikely to be covered by an additional warranty (though this may not always be the case). On the other hand, ink is something that needs to be continually replaced. And the way that printer manufacturers have their business models set-up, they make a substantial amount of revenue from the sale of ink cartridges. Does this mean that they should change their business models to be closer to that of automakers? Certainly it costs printer companies far less to make cartridges than the outlandish prices they charge in stores might suggest, but part of that is to make printing technology affordable to the casual or home user. I think that most of these people are happy to be able to get the printers cheaply, and would be glad to spend more on ink, since they may not go through it very fast.

Because Lexmark stands to make a substantial portion of their revenue from the sale of cartridges, it is understandable why they might want to exclude popular printers from being compatible with 3rd-party cartridges. In fact, I am sure that this is every printer manufacturer’s dream-come-true. Fortunately, we live in a world where this is not entirely yet the case. Personally, I like being able to buy 3rd-party ink cartridges, or toner cartridges, cheaply. But there is certainly nothing wrong with manufacturers doing everything within their legal right to try and make sure that consumers are forced to buy cartridges from the same manufacturer. At the same time, I don’t necessarily think that what SCC has done is wrong. The point is that right now consumers have a choice as to whether they want to be able to install inexpensive 3rd-party cartridges, or whether they want a printer that they know will only accept expensive cartridges from the manufacturer. As long as consumers have that choice, then they shouldn’t complain that the business model is unfair, so long as the printer manufacturers are within legal bounds.

In conclusion, I don’t think that the Lexmark-GM analogy is useful or applicable to this case. From Lexmark’s perspective, it is obvious why they want a monopoly on cartridge sales to Lexmark owners. But as the 6th Circuit showed in this case, printer companies may never have a complete monopoly on cartridge sales. Lucky us!

Josh Probst gave an interesting hypothetical situation in his post on Tuesday evening. He wondered what would happen if Microsoft’s Xbox video game console was able to play games that were originally written for Sony’s PlayStation 2 video game console. In other words, what if the Xbox were able to emulate PS2 games? Would this be legal? What would happen if Sony took Microsoft to court for copyright infringement or tarnishment? Josh implies that the courts might not rule in the same manner as they did in the Sony v Connectix case, but I disagree. If we assume that Microsoft would need to perform copyright infringements that were similar to the ones performed by Connectix in order to write a BIOS that would successfully emulate PS2 games on the Xbox system, then I think that the case would be decided in a manner very similar to that of Sony v Connectix.

In Sony v Connectix the 9th Circuit Court of Appeals relied on a fair use precedent from Sega v Accolade (1992). It states that the factors to be considered in determining fair use are:

1) the purpose and character of the use, including whether such use if of a commercial nature or is for nonprofit educational purposes;

2) the nature of the copyrighted work;

3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and

4) the effect of the use upon the potential market for or value of the copyrighted work.

a) As the 9th Circuit states in Sony v Connectix, the first factor considers the extent to which the new work is “transformative.” Based on the 9th Circuit’s interpretation of this factor, I would say that Microsoft’s Xbox is modestly transformative. It allows owners of PS2 games to play them on the Xbox without having to have both consoles present. This can be a huge bonus for people who like the games that are put out for both consoles, but don’t have the physical space for both, or for people who typically travel with their consoles, and don’t want to have to lug both around, but would like to be able to play both Xbox and PS2 games. Moreover, since Microsoft would undoubtedly make their own BIOS to run the PS2 games, this software would represent a unique expressive element, and would not contain object code that infringes on Sony’s copyright, just as in the Connectix case. Regarding the commercial aspect of the first factor, Microsoft’s intention would likely be seen as less commercial than Connectix’s. Whereas PC owners would have had to buy Connectix’s software in order to emulate PS1 games on their PCs, Xbox owners wouldn’t have to buy anything new in order to emulate PS2 games (assuming Microsoft would simply make the new BIOS available as a free firmware update). In fact, it might spur the sale of PS2 games, and it would be hard to find direct evidence that it would harm PS2 sales any more than Connectix’s software harmed PS1 sales. Moreover, like many car owners, many video game console owners are very brand-loyal. Microsoft’s new BIOS might simply make it more convenient for those who already own Xboxs and PS2s, and have a bunch of PS2 games already. For those that don’t own PS2s, it’s likely that they might buy PS2 games now that they can play them on their Xbox. But it would be difficult to prove that Microsoft is hurting Sony’s sales, since it would be difficult to prove that Xbox owners would have ever bought a PS2 simply because they wanted to play the games. In other words, Sony may end up making more money as a result.

b) Just as in the Connectix case, any copying of Sony’s BIOS that Microsoft was responsible for must have been necessary to be a fair use. Just as the 9th Circuit concluded that Connectix’s copying was a fair use, it would be likely to reach the same conclusion regarding this hypothetical case. Thus the 2nd factor weighs in favor of Microsoft.

c) As in the Connectix case, there would be no disputing that Microsoft in fact copied Sony’s BIOS in the reverse engineering process. This fact would certainly weigh against Microsoft, but as the 9th Circuit points out in the Connectix case, it would be of “very little weight.”

d) Regarding the 4th factor, the 9th Circuit pointed out in Sony v Connectix that even if Sony were to undergo economic loss as a result, that would not eliminate a finding of fair use. Copyright law does not grant a monopoly to Sony in this respect.

Here we see that factors 2 and 4 factors very clearly favor Microsoft, and that the 3rd slightly favors Sony. The first factor offers much room for interpretation, though I would argue that if anyone it favors Microsoft. Connectix clearly was aiming to make money through sales of its Virtual Game Station, whereas there is no clear evidence that under Josh’s hypothetical that Microsoft would even make any more money. In fact, it is Sony that would stand the best chance of making money through sales of its games. If we assume that the latest Microsoft press reports are true, then we can assume that Microsoft is losing tons of money on consoles. Therefore, even if sales of the Xbox were to increase, Microsoft’s profit margins would decline. And assuming that Sony loses money on sales of its PS2 as well, its profit margins would increase. Perhaps this is why neither company has introduced firmware updates that allow the play of each other’s games on either system. In summary, it seems that the current sales model is such that the software is really what determines whether a system is popular or not. Both systems are capable of stunning graphics and gameplay. But since both manufacturers make profits based on how many games they sell, neither is interested in making their systems compatible with the competing manufacturer’s games, since this would be like giving money to their competitors.

It is an interesting hypothetical situation, and if it ever arose, I believe that the courts would decide in essentially the same manner as they did for the Sony v Connectix case, for all the reasons cited above.

As we saw in Timothy Wu’s Copyright’s Communications Policy, each time a new commercially viable technology comes about that makes revenue based upon the copyrighted works of others, any resolution of the conflict of interests takes much time and negotiation. Thus, according to our history books, we should not expect a complete resolution between copyright holders and the distributors of P2P software (who arguably make money based on others’ copyrighted works) anytime soon.

At first appearance, the current struggle between owners of copyrighted works and the distributors of P2P software seems not unlike the other struggles outlined in Wu’s paper. Those struggles took anywhere from about 10 years (in the case of the recording industry) to 37 years (radio) to be resolved. In each case, the two sides settled on a compulsory licensing agreement, after which Congress enacted it into law. Therefore, one might wonder whether the 3 cases outlined in Wu’s paper reflect a possible future for P2P.

However, we must be careful in comparing the past technologies that Wu mentions with that of P2P distributors. In each technology that Wu mentioned, whether it was records, radio, or cable television, it was the “distributors,” and not the end-users, that were attacked for violation of copyrights. Furthermore, at the time each of these technologies was developed, their uses were not explicitly infringing under the copyright legislation of the time. After all, that was the central debate in each of these cases: whether uses of the new technology constituted a violation of copyright.

This is in stark contrast to P2P technology. At the time it was developed, the “infringing” uses of P2P were already clearly violations under current copyright legislation. Moreover, because these “infringing” uses are indeed copyright violations, it is the end user, and not the distributor of the technology, that are direct copyright infringers. The question before the courts is whether Grokster and Streamcast can be held secondarily liable for these copyright infringements.

This brings us back to the question: are the struggles that Wu describes indicative of the future struggles between P2P distributors and copyright owners? While some of the fundamentals outlined above indicate that the current struggle between P2P and copyright owners and the struggles that Wu describes are not related, I think it’s likely that the cases may eventually be resolved in a similar manner. That is, both sides will take each other to court, unsuccessfully try to impose new legislation, and eventually develop some sort of royalty or compulsory licensing scheme that both sides agree on, and have Congress alter copyright law.

Regardless of how the struggle gets resolved, the fact of the matter is that P2P technology is here to stay. Clearly both the RIAA and MPAA could start selling their works online, using the internet as a distribution tool. But P2P represents a clear technological advantage, in that it can allow the transfer of large files to many people very quickly, without the bottlenecking that might be encountered should the works all be downloaded from a central server. The problem of P2P of course, is that since it is not centrally maintained, it is very hard to regulate, and thus make sure that copyright owners get revenue for what is downloaded.

However, just because a high profile P2P case is before the Supreme Court at the moment doesn’t mean that we should expect a resolution anytime soon. Unlike its decision in Sony, which was a reversal of the judgment of the Court of Appeals, the Supreme Court is currently not in a position to set a radical new precedent for the lower courts. They can either affirm the lower court’s ruling, or they can remand the case to a lower court for further proceedings. Thus, one can be assured that the lawsuits will continue for some time. Currently, both the MPAA and the RIAA risk an ongoing decay of profitability due to their inability to extract income from the distributors of P2P software. At the same time, distributors of P2P software have to be careful because they face the possibility that they may be held liable for indirect copyright infringement.

At the moment, I’d say that P2P distributors have a lead over the copyright owners. I say this simply because the 9th circuit recently sided with two major P2P distributors. Also, the plaintiffs may have temporarily weakened their arguments by some of the things they said in the 9th circuit court proceedings. I am specifically referring to the following exchange:

Ramos: There’s nothing different from what they’re [Grokster and Streamcast] doing from organizing and instructing the participants in a trafficking network that is trading in counterfeit materials than what they are doing.

Judge Noonan: Let me say what your problem is. You can use these harsh terms, but you are dealing with something new. And the question is, Does the statutory monopoly that Congress has given you reach out to that somthing new? And that’s a very debatable question. You don’t solve it by calling it theft. You have to show why this court should extend a statutory monopoly to cover the new thing. That’s your problem.

I’d say that the plaintiffs have to be very careful because the courts may think this kind of behavior represents a desire to destroy P2P technology altogether. The courts are fully aware of the beneficial uses of P2P technology, and the plaintiffs aren’t going to win cases by attacking the technology. In order for this struggle to be resolved, the copyright owners must eventually come to some sort of an agreement with P2P distributors. The courts can only do so much. Only once Congress enacts new legislation will there be protection for both sides.

Update: At 12:15 AM I added a couple hyperlinks and fixed some minor typos.