White Buffalo Ventures, LLC v. University of Texas

420 F.3d 366, 36 Communications Reg. (P&F) 520, 2005 U.S. App. LEXIS 15885, 2005 WL 1806353

• Court: Court of Appeals for the Fifth Circuit
• Decided: August 2, 2005
• Docket: 04-50362
• Status: Published

Opinion

JERRY E. SMITH, Circuit Judge:

This case involves the regulation of unsolicited, commercial mass electronic messages (“emails”) (a species belonging to the larger communication genus often referred to as “spam”). 1 Plaintiff White Buffalo Ventures, LLC (“White Buffalo”), operates several online dating services, including longhornsingles.com, which targets students at the University of Texas at Austin (“UT”). Pursuant to its internal anti-solicitation policy, 2 UT blocked White Buffalo’s attempts to send unsolicited bulk commercial email.

White Buffalo sought to enjoin UT from excluding its incoming email. The district court denied the injunction. On cross-motions for summary judgment, the court granted UT’s motion and denied White Buffalo’s. White Buffalo appeals, challenging the ruling on the grounds that federal law preempts UT’s internal anti-spam policy (the “Regents’ Rules”) 3 and that the policy violates the First Amendment. Mindful that this case presents several novel issues, the significance of which will grow proportionally with heightened cultural and economic reliance on the Internet, we affirm.

We make two determinations. First, we decide that the CAN-SPAM Act does not preempt UT’s anti-spam policy. Second, we determine that the policy is permissible under our First Amendment commercial speech jurisprudence, but we reserve judgment on whether state university email servers constitute public or private fora.

I.

A.

The parties do not dispute the facts. UT provides, free of charge, Internet access and email addresses to faculty, staff, and students at the domain “utexas.edu.” Owners of electronic mail accounts can access those accounts either on-grounds (by means of wireless connections or of wired, authenticated clusters) or remotely (by means of some other Internet access provider). An owner of a UT user account may, for example, log on from any third-party dial-up or broadband service provider and check for email residing on one of UT’s 178 email servers.

UT has a policy of blocking many types of incoming spam, irrespective of commercial content or source authenticity. Under the Regents’ Rules, the technology department (the “ITC”) implements procedures (1) to block incoming unsolicited, commercial emails and (2) to stop the transmission of such emails. 4

White Buffalo operates several online dating services, including one, called “lon-ghornsingles.com,” that targets UT students. In February 2003, White Buffalo submitted a Public Information Act request seeking all “non-confidential, nonexempt email addresses” held by UT, which responded by disclosing all qualifying email addresses. In April 2003, White Buffalo began sending legal commercial spam to targeted members of the UT community. 5

UT received several complaints regarding unsolicited email blasts from White Buffalo. UT investigated and determined that White Buffalo had indeed sent unsolicited emails to tens of thousands of UT email account-holders, at which point UT issued a cease and desist letter. White Buffalo refused to comply with that letter, so UT blocked all email ingress from the IP address 6 that was the source address for the unsolicited White Buffalo spam. The filter blocked all email sent from that IP address to addresses containing the “@utexas.edu” string.

B.

White Buffalo obtained a temporary restraining order (“TRO”) in state court. UT removed the cause to federal court on the basis of federal question jurisdiction; there the TRO was continued pending a hearing on the preliminary injunction. After a hearing in May 2003, the district court denied the injunction. The parties conducted discovery, and both moved for summary judgment. The district court granted UT’s summary judgment motion and denied White Buffalo’s.

II.

1.

This court reviews a summary judgment grant de novo, in accordance with the Fed R. Civ. PROC. 56 analysis that guides the district court. Ford Motor Co. v. Tex. Dep’t of Transp., 264 F.3d 493, 498 (5th Cir.2001). The district court entered judgment for UT on cross-motions for summary judgment. On review, the motions are reviewed independently, with evidence and inferences taken in the light most favorable to the nonmoving party. See id. We review a district court’s preemption determinations de novo. See Baker v. Farmers Elec. Coop., Inc., 34 F.3d 274, 278 (5th Cir.1994).

2.

The doctrine of preemption stems from the Supremacy Clause, 7 which gives federal law precedence over a conflicting state law. See Cipollone v. Liggett Group, Inc., 505 U.S. 504, 516, 112 S.Ct. 2608, 120 L.Ed.2d 407 (1992). CAN-SPAM’s preemption of state law derives from an express provision in the Act. See 15 U.S.C. § 7707(b).

Although a court should begin with the expression provided by Congress, it must also “identify the domain expressly pre-empted.” 8 The fact that Congress has expressly preempted certain activity is plain, but the scope of that express preemption is not. The power to supplant state law is “an extraordinary power in a federalist system.” Gregory v. Ashcroft, 501 U.S. 452, 460, 111 S.Ct. 2395, 115 L.Ed.2d 410 (1991). Preemption radically alters the balance of state and federal authority, so the Supreme Court has historically refused to impose that alteration interstitially. See id. The Court has expressed this principle as a presumption against preemption of state law. 9 Supremacy Clause analysis is classic “tie goes to the state” jurisprudence, and the existence of an express preemption provision does not always plainly demarcate what the federal law expressly preempts.

3.

The district court granted summary judgment to UT on White Buffalo’s claim that the CAN-SPAM Act preempts ITC’s anti-spam regulations. The court premised its holding on four propositions: (1) that the “purposes” of CAN-SPAM, as determined by reference to the statute and the accompanying Senate Report, suggest that Congress did not mean to preempt technological approaches to combating spam; (2) that § 7707(c) specifically exempts UT from the scope of express preemption; (3) that § 7707(b)(2), which states that “[s]tate laws not specific to electronic mail, including State trespass, contract, or tort law” are not preempted, exempts UT’s anti-spam policy because that policy is part of a larger set of anti-solicitation rules; and (4) that UT’s ITS policy is not a “statute, regulation, or rule of a State or political subdivision of a state” and is therefore not preempted by § 7707(b)(1). We do not organize our discussion around these four propositions (because the appellate briefing renders an alternate organization more desirable), but we discuss each in its appropriate context.

To our knowledge, no Fifth Circuit panel has scrutinized any portion of CAN-SPAM, and no court in this country has considered the legislation’s preemption clause. This is therefore an issue of very, very first impression.

In part, CAN-SPAM prohibits fraudulent, abusive and deceptive commercial email, 15 U.S.C. §§ 7703, 7704; provides for enforcement of the Act by federal agencies, states, and Internet service providers (“ISPs”), id. § 7706; and provides for the issuance of regulations to implement the purposes of the Act, id. § 7711. The parties have agreed, in the district court and on appeal, that White Buffalo complied with the requirements of the CAN-SPAM Act. Its email blasts were not unlawful.

Most relevant to White Buffalo’s claim is CAN-SPAM’S preemption clause:

This chapter supersedes any statute, regulation, or rule of a State or political subdivision of a State that expressly regulates the use of electronic mail to send commercial messages, except to the extent that any such statute, regulation, or rule prohibits falsity or deception in any portion of a commercial electronic mail message or information attached thereto.

§ 7707(b)(1).

White Buffalo argues that this preemption clause prevents UT from promulgating regulations to impede the ingress of longhornsingles.com emails to utexas.edu users. According to White Buffalo, because UT is a state actor and has conceded that White Buffalo’s spam is not false or fraudulent, CAN-SPAM preempts the Regents’ Rules authorizing the email filters. White Buffalo provides no authority beyond § 7707(b)(1) in support of this position.

Matters become more complicated because, in addition to setting forth the preemption clause, § 7707 carves out a set of entities to be exempt from any possible preemptive effect. It states that “[n]oth-ing in this chapter shall be construed to have any effect on the lawfulness or unlawfulness ... of the adoption, implementation, or enforcement by a provider of Internet access service of a policy of declining to transmit, route, relay, handle, or store certain types of electronic mail messages.” § 7707(c).

The district court held that CAN-SPAM does not preempt UT’s anti-solicitation policy. It noted that § 7707(c) permits Internet service providers to employ protection measures, and it held that UT belongs to that set of service providers. The court also relied on Congress’s acknowledgment of “the problems associated with the rapid growth and abuse of unsolicited commercial email [that] cannot be solved by Federal legislation alone” but that will also require the “development and adoption of technological approaches” to serve the goals of the Act. See 15 U.S.C. § 7701(a)(12). The court found that “[t]he Act ... does not preclude a state entity like UT from using technological devices [such as] spam-filters to conserve server space and safeguard the time and resources of its employees, students, and faculty.” White Buffalo Ventures, LLC v. Univ. of Tex. at Austin, No. A-03-CA-296-SS, 2004 WL 1854168, at *4 (W.D.Tex. Mar.22, 2004).

There are two competing interpretations, both rooted firmly in the text of the Act, of the degree of authority state actors may wield in response to commercial spam. Under the first, state entities may not regulate commercial speech except where that regulation relates to the authenticity of the speech’s source and content. Under the second, state entities may implement a variety of non-authenticity related commercial speech restrictions, provided the state entity implementing them is an “Internet access provider.”

As a result of Congress’s apparent failure to contemplate this question, we must not infer preemption. The textual ambiguity triggers the strong presumption against such a finding, and we cannot be sure whether UT’s regulations fall within the ambit of the express preemption clause. UT may therefore implement the Regents’ Rules without violating the Supremacy Clause.

UT argues that CAN-SPAM does not preempt the ITC policy (1) because the Act does not displace the state’s ability to supplement federal law and (2) because CAN-SPAM preempts state rules that relate to the sending, rather than the receipt, of unsolicited commercial emails. Section 7707(b)(1) carefully specifies state political subdivisions as falling within the scope of its preemption, and UT is a public school. 10

In a vacuum, the provision is explicit about the types of policies CAN-SPAM preempts. In layman’s terms, state entities may not regulate the use of electronic mail to send commercial spam except where those rules relate to source and content authenticity. UT emphasizes Congress’s choice to use the word “send” in the statute. As a result, UT argues, CAN-SPAM does not preempt its regulation of “received” emails. We decline to imbue the word “send” with the particular significance UT urges. 11

CAN-SPAM does not preempt the Regents’ Rules, because § 7701(b)(1) is in tension with plain text found elsewhere in the Act, and that tension triggers the presumption against preemption. The district court properly sought to interpret § 7707(c), which reads, “Nothing in this Act shall be construed to have any effect on the lawfulness or unlawfulness ... of the adoption, implementation, or enforcement by a provider of Internet Access service of a policy of declining to transmit, route, relay, handle, or store certain types of electronic mail messages.” In finding no express preemption, the court both (1) averred that the ITC policy may not constitute a “statute, regulation, or rule of a State or political subdivision of a state,” § 7707(b)(1), and (2) noted that UT is a “provider of Internet access.” Any suggestion along the lines of (1) — that an ITC policy does not constitute a policy of a state subdivision — is incorrect and requires little explanation. ITC implements the directives of, and operates pursuant to the authority of, the Board of Regents; its policies therefore constitute rules of a state subdivision. 12

We therefore confine ourselves primarily to the discussion of (2). The district court stated that “UT is certainly a provider of Internet access service to its students, if not to its employees and faculty, so it is expressly authorized under the statute to implement policies declining to transmit, route, relay, handle or store spam.”

The district court says “certainly” without any reference to the definition provided in the statute. Congress, in fact, imports that definition wholesale from a statutory predecessor, the Internet Tax Freedom Act: “[A] service that enables users to access content, information, electronic mail, or other services offered over the Internet, and may also include access to proprietary content, information, and other services as a part of a package of services offered to consumers.” 47 U.S.C. § 151.

We doubt that those legislators responsible for passing the Internet Tax Freedom Act gave serious consideration to the situation the instant facts present. UT indeed provides Internet Access Service— any time somebody sits down at a computer terminal on campus — but users need not check their UT email from UT network computers, because they can access the email server remotely. Nonetheless, status as an “Internet Access Provider” does not appear to turn on the fraction of access conducted remotely, and we are hard-pressed to find that providing email accounts and email access does not bring UT within the statutory definition borrowed from the Internet Tax Freedom Act. 13 We therefore decide that UT falls within the ambit of § 7707(c).

C.

We analyze this issue using a Venn diagram, 14 the intersecting area of which Congress did not anticipate' — where the state entity is itself the provider of Internet access. In that area resides activity that Congress has both expressly preempted and expressly excepted from preemption analysis. Such tension, created by the text of the statute, leaves us unwilling to overrule the strong presumption against preemption. The Regents Rules are valid under the Supremacy Clause.

III.

White Buffalo contends that the district court erred in granting summary judgment on its First Amendment claim. Whether UT has violated White Buffalo’s First Amendment rights turns on the resolution of the four-part commercial speech test in Central Hudson Gas & Electric Corp. v. Public Service Commission, 447 U.S. 557, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980). We review First Amendment determinations under the commercial speech doctrine de novo. See Moore v. Morales, 63 F.3d 358, 361 (5th Cir.1995). Resolving this issue in favor of UT, we decline to reach the issue of whether UT’s email servers constitute public fora. 15

Commercial speech is “expression related solely to the economic interests of the speaker and its audience.” Central Hudson, 447 U.S. at 561, 100 S.Ct. 2343. No one seriously disputes the commercial character of the speech at issue here.

In Central Hudson, the Court invalidated portions of state regulations banning commercial advertising that promoted the use of electricity. See id. at 572, 100 S.Ct. 2343. The Court determined that the government’s action was more extensive than necessary to promote the state’s substantial interest in energy conservation. Id. at 569-70, 100 S.Ct. 2343. In so doing, the Court announced a four-part test to evaluate the legality of commercial speech regulation: (1) whether the speech is unlawful or misleading; (2) whether the government’s expressed interest is substantial; (3) whether the state action directly promotes that interest; and (4) whether the state action is more extensive than necessary to promote that interest. See id. at 566, 100 S.Ct. 2343.

Under the first Central Hudson prong, we must determine whether the speech is unlawful or misleading. See id. Both parties agree that White Buffalo’s commercial solicitations are legal and that they contain factually accurate information.

Under the second Central Hudson prong we must assess the “substantiality” of the government’s proffered interests. See id. UT advances two primary interests: (1) safeguarding the time and interests of those with UT email accounts (“user efficiency”) and (2) protecting the efficiency of its networks and servers (“server efficiency”). We distinguish between the two interests for reasons that are important under the fourth prong of the Central Hudson analysis.

For purposes of evaluating the summary judgment, we acknowledge as substantial the government’s gatekeeping interest in protecting users of its email network from the hassle associated with unwanted spam. Also substantial is the “server efficiency” interest, but it must independently satisfy a “goodness of fit” inquiry under the fourth prong of Central Hudson. “Suffer the servers” is among the most chronically over-used and under-substantiated interests asserted by parties (both government and private ones) 16 involved in Internet litigation, and rules imposed pursuant to such interests require more than a judicial rubber-stamp, for reasons we explain in part III.B.4.b, infra.

Pursuant to the third Central Hudson prong, we must next determine whether the UT policy directly advances both proffered substantial interests: (1) UT’s interest in sanitizing the network for its email account-holders (user efficiency) and (2) its interest in preserving the operating efficiency of its servers (server efficiency). See id. at 569, 100 S.Ct. 2343. Again, there can be no serious dispute that UT’s anti-spam policy, which blocks specific incoming commercial spam after account-holders have complained about it, directly advances both interests. One can hardly imagine a more direct means of preventing commercial spam from appearing in account-holders’ inboxes and occupying server space than promulgating a policy that excludes such material from the email network. 17

4.

Having resolved the first three Central Hudson questions in UT’s favor, we must finally conduct the most difficult inquiry— whether the ITC policy is no more extensive than necessary to achieve at least one of the two substantial state interests. See id. at 569-70, 100 S.Ct. 2343. 18 White Buffalo contends that UT’s anti-spam policy fails to meet this final requirement, although White Buffalo’s precise objection is hard to discern. It appears to protest the over-restrictiveness of the policy on the ground that it is impossible to articulate precisely what the UT regulation is.

To the contrary, UT (relying on the district court’s disposition of the issue) reasons that the policy is narrowly and specifically drawn to protect the system and users from only those unsolicited, commercial emails that have been identified as problematic by complaint, system monitors, or other means. The restriction is tailored by blocking only those emails from specifically identified ISP addresses. Although we may not agree with all of UT’s characterizations of its policy, we are aware of what that policy is. White Buffalo’s objection in this regard is without merit.

a.

With respect to the first proffered substantial state interest, which is promoting user efficiency, 19 the ITC policy is no more extensive than necessary. We have little problem affirming the proposition that, to keep community members from wasting time identifying, deleting, and blocking unwanted spam, UT may block otherwise lawful commercial spam (as long as the blocks are content- and viewpoint-neutral). 20

We reject, however, the proposition that the ITC policy is no more extensive than necessary to secure the state’s second substantial interest, which is the efficiency of its servers. 21 One might persuasively present evidence that spam, taken in its entirety, affects the efficiency of email servers; indeed, that appears to be what UT has proffered; it submits a list of between 1,500 and 2,000 blocked IP addresses. 22 Updegrove testified at the May 2003 Preliminary Injunction hearing that UT’s “system” would not be able to function without these blocks. Such testimony is common where server efficiency is offered as a state or private interest in Internet litigation.

We must nonetheless consider the evidence in the light most favorable to the nonmovant. See Ford Motor Co., 264 F.3d at 498. Moreover, the challenged regulation should indicate that its proponent “carefully calculated the costs and benefits associated with the burden on speech im posed by its prohibition.” Cincinnati v. Discovery Network, Inc., 507 U.S. 410, 417, 113 S.Ct. 1505, 123 L.Ed.2d 99 (1993) (internal citations and quotations omitted).

There is record testimony that White Buffalo can send a restricted volume of email at off-peak times, so as not to impede server efficiency. Moreover, UT’s list of blocked IP addresses does not make any of the types of distinctions that Congress obviously thought important — distinctions between, say, (1) truthful commercial messages and obscene images, (2) commercial messages with an unsubscribe feature and commercial messages without one, (3) emails sent during peak traffic times and those that are not, and (4) email that originates from an authentic source and email that does not. The rub is that although the record demonstrates that unblocking all spam might compromise network efficiency, it says nothing about the effects of allowing lawful time- and volume-restricted commercial spam to enter UT’s email network.

An exchange at the preliminary injunction hearing between UT’s attorney and Updegrove most vividly illustrates the poor fit between UT’s restrictions and the substantial interest in server efficiency:

Q: Well, [White Buffalo’s attorney] is saying, well, “Hey, I can send this at night when the employees aren’t there. I won’t send too many at one time. It won’t affect your system that much because of that. Now, is there a reason why that’s not an acceptable proposition.”

A: Well, if something is wrong, just because there’s a little bit of it doesn’t make it right. If a university makes resources available, misusing a little bit of those resources isn’t correct.

For the server efficiency rationale to pass muster under the fourth prong of Central Hudson, spam filters must block a set of spam that poses a legitimate threat to server efficiency.

This is not to say that UT need draw granular distinctions between types of spam where drawing them renders filtering economically infeasible. 23 It, however, is to say that where UT may easily use certain types of filters — e.g., time of day and volume filters — UT should use them rather than categorically exclude all unsolicited commercial bulk email. If those types of filters are economically infeasible, that evidence should be in the summary judgment record. The current record reflects only that UT does not employ such filters because legal spammers are subjectively “misusing” the system, not because they are overburdening it.

Our conclusion that, for summary judgment purposes, there is an insufficient fit between the ITC policy and the asserted interest in server efficiency is of little moment in the spam context. The server efficiency interest is almost always coextensive with the user efficiency interest, and the fit is sufficient for the latter; but declaring server integrity to be a substantial interest without evidentiary substantiation might have unforeseen and undesirable ramifications in other online contexts. 24

C.

A governmental entity may assert that a statute serves multiple interests, and only one of those need be substantial. See Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 71-73, 103 S.Ct. 2875, 77 L.Ed.2d 469 (1983). The ITC policy survives First Amendment scrutiny despite its failure to justify that policy in relationship to the server efficiency interest. We therefore decide that UT’s anti-spam policy is constitutionally permissible under Central Hudson. Because we so decide, we need not address what type of First Amendment forum a public university email network constitutes.

The summary judgment is AFFIRMED.

1 . Because the term "spam” is often thought of pejoratively, it is important to note that although that term necessarily implies that the email was unsolicited, the more general meaning does not (1) imply anything about the veracity of the information contained in the email; (2) require that the entity sending it be properly identified or authenticated; or (3) require that the email, even if true, be commercial in character. There nonetheless appears to be no consensus as to the precise meaning of the term "spam,” which is sometimes used synonymously with unsolicited "bulk” email. A set of spam messages sent out together is called an email "blast.”

The term "spam” derives from a 1970 Monty Python Flying Circus sketch in which a waitress .recites a menu containing "egg and spam; egg bacon and spam; egg bacon sausage and spam; spam bacon sausage and spam; spam egg spam spam bacon and spam; spam sausage spam spam bacon spam tomato and spam ...." See Roger Allen Ford, Comment, Preemption of State Spam Laws by the Federal CAN-SPAM ACT, 72 U. Chi. L.Rev. 355, 355 n.l (2005) (citing David Crystal, Language and the Internet 53 (Cambridge 2001)).

2 . UT has a general policy against solicitation, which it articulates in the Rules and Regulations of the Board of Regents of the University of Texas System ("the Regents”). Pursuant to that policy, UT, with limited exceptions, prohibits solicitation at and on its facilities and has promulgated specific procedures dealing with unsolicited email communications, including commercial solicitations. Under these procedures, when unsolicited email communications come to the attention of university network administrators (by way of complaints, system monitors, or other means), UT takes steps to block or otherwise stop the transmission of such emails, with or without notice to the sender, as circumstances permit or warrant.

3 . Specifically, White Buffalo contends that UT’s regulations are preempted by the Controlling the Assault of Non-Solicited Pornography and Marketings Act of 2003 (the “CAN-SPAM Act” or the "Act”), 15 U.S.C. §§ 7701-7713, Pub.L. 108-187, 117 Stat. 2619 (2003).

4 . These procedures may or may not provide notice to the sender, depending on the circumstances.

5 . We presume the legality of these emails based on this record, the parties’ agreement, and the absence of any challenge.

6 .An Internet Protocol ("IP”) address is a unique 32-bit numeric address, written as numerals separated by periods, identifying each sender or receiver of information traveling across the Internet. An IP address has two parts: the identifier of a particular network on the Internet (say, the first 24 bits) and an identifier of the particular device (which can be a server or a workstation) within that network. In essence, an IP address identifies a single computer; that computer might be an entry point into an internal network, but that is not important for our purposes.

7 . "This Constitution, and the Laws of the United States which shall be made in Pursuance thereof; ... shall be the supreme Law of the Land U.S. Const, art. VI, cl. 2.

8 . Cipollone, 505 U.S. at 517, 112 S.Ct. 2608; see also Medtronic, Inc. v. Lohr, 518 U.S. 470, 484, 116 S.Ct. 2240, 135 L.Ed.2d 700 (1996).

9 . See Cipollone, 505 U.S. at 517-18, 112 S.Ct. 2608; see also Bldg. & Constr. Trades Council v. Associated Builders & Contractors, 507 U.S. 218, 224, 113 S.Ct 1190, 122 L.Ed.2d 565 (1993) ("We are reluctant to infer preemption ....”); Maryland v. Louisiana, 451 U.S. 725, 746, 101 S.Ct. 2114, 68 L.Ed.2d 576 (1981) ("Consideration under the Supremacy Clause starts with the basic assumption that Congress did not intend to displace state law.”).

10 . UT argues that ITS is not a political subdivision of the state. This argument is merit-less, as we explain in part II.B.3, infra.

11 . UT posits that CAN-SPAM regulates not the "receipt” of email, but the "sending” of it. UT then contends that the Regents' Rules control the "receipt” of email. Section 7707(b)(1) preempts state law regulating "the use of electronic mail to send commercial messages.” All email (and all "snail mail,” for that matter) is both "sent” and "received.”

The event triggering preemption is that the email was sent, not the particular identity of the entity sending it. We do not mean to say that "send” and "receive” never have more specialized meanings in the statute, but only that the grammatical construction of this particular provision suggests emphasis should not be placed on that distinction here.

12 . In a related passage, the district court stated that “the Board of [Regents] Rules governing solicitation using university facilities cannot be said to be specific to electronic mail since it regulates all forms of solicitation.” We need not decide this issue, because we have alternate grounds of making our preemption decision.

13 . It would be an unusual policy to allow private, but not public, educational institutions to act as custodians for the interests of its online community. The prudence of the policy, however, does not drive our determination that UT should be considered an Internet Access Provider under the Act.

14 .A Venn diagram uses circles to represent sets, with the position and overlap of the circles indicating the relationships between the sets.

15 . In other words, we consider two hypothetical situations: one in which the UT servers are public fora, and one in which they are not. If the servers are not, then the First Amendment question is easily resolved — if a server is a private forum, the government may regulate the speech so long as it is viewpoint-neutral. In the alternative, if a server is a public forum, we apply Central Hudson. If we determine that this particular regulation would satisfy either situation, we need not resolve the dicey but admittedly important question of the public versus private forum status of public university email servers.

16 .The opinion in eBay, Inc. v. Bidder’s Edge, Inc., 100 F.Supp.2d 1058 (N.D.Cal.2000), is one of the first and perhaps the most conspicuous instance of this rationale. The eBay court held, on a trespass to chattels theory, that if the defendants crawling "activity is allowed to continue unchecked, it would encourage other auction aggregators to engage in similar recursive searching of the eBay system such that eBay would suffer irreparable harm from reduced system performance, system unavailability, or data losses.” Id. at 1066. See also note 25, infra.

17 . Neither party provides caselaw in support of its position on the question of substantial interest. The district court relied on extensive comments in the Senate Report offered in support of the CAN-SPAM Act.

18 . For an explanation as to why only one substantial interest need be satisfied, see part III.C, infra.

19 . By "user efficiency” we mean the ability of UT email account holders to go about their daily business without constantly having to identify and delete unwanted commercial spam.

20 . UT Vice President of Information Technology Daniel Updegrove testified at the May 20, 2003, preliminary injunction hearing:

[A]t a minimum there's the time it takes to configure an e-mail filter, running the risk that legitimate messages will be filtered erroneously or one by one deleting the offending messages. And there’s an ongoing question of how much of this message you have to actually encounter in order to decide that it's spam.

He furthered affirmed that if UT "wasn’t allowed to block or was somehow required under the Constitution to unblock these 1700 some odd sites, that it would severely degrade an employee’s ability to do their job[.]”

21 . This "poor fit” could be rephrased as an objection under several other Central Hudson prongs. For example, the Supreme Court has made a similar analysis under the third prong. In Greater New Orleans Broadcasting Ass’n, Inc. v. United States, 527 U.S. 173, 188, 119 S.Ct. 1923, 144 L.Ed.2d 161 (1999), the Court stated:

The third part of the Central Hudson test asks whether the speech restriction directly and materially advances the asserted governmental interest. This burden is not satisfied by mere speculation or conjecture; rather, a governmental body seeking to sustain a restriction on commercial speech must demonstrate that the harms it recites are real and that its restriction will in fact alleviate them to a material degree. Consequently, the regulation may not be sustained if it provides only ineffective or remote support for the government's purpose. We have observed that this requirement is critical; otherwise, a State could with ease restrict commercial speech in the service of other objectives that could not themselves justify a burden on commercial expression.

(Internal citations and quotation marks omitted.) The prong under which we make our observations matters little for Central Hudson analysis, however, because the Supreme Court has stated that "[a]ll are important and, to a certain extent, interrelated: Each raises a relevant question that may not be dispositive to the First Amendment inquiry, but the answer to which may inform a judgment concerning the other three.” Id. at 183, 119 S.Ct. 1923.

22 .This information is contained in Exhibit 4. There are two constituent lists — one of blocks by host address and one of blocks by IP address.

23 . For example, as Updegrove testifies, it would be impossible to filter spam based on whether the originator of the email was a legitimate business. The email filters could not automate this task.

24 . This is no more than a cautionary note, the importance of which has become more plain as a result of our increasing familiarity with litigation involving the Internet. For example, in many of the "digital trespass” cases, where a plaintiff bases a trespass to chattels theory on a defendant's unauthorized use of a network/computer system, the court will merely conclude, without evidence or expía- nation, that the allegedly unauthorized use burdened the system.

One of the most prominent such statements occurs in CompuServe v. Cyber Promotions, Inc., 962 F.Supp. 1015, 1022 (S.D.Ohio 1997) ("To the extent that defendants' multitudinous electronic mailings demand the disk space and drain the processing power of plaintiff's computer equipment, those resources are not available to serve CompuServe subscribers. Therefore, the value of that equipment to CompuServe is diminished even though it is not physically damaged by defendants’ conduct.”). Many courts mention system degradation and perfunctorily cite CompuServe, but focus primarily on things such as decline in customer goodwill, worker productivity, and the like. See, e.g., Am. Online, Inc. v. IMS, 24 F.Supp.2d 548, 550 (E.D.Va.1998) ("[Plaintiff's] contact with [Defendant's] computer network was unauthorized; and [Plaintiff's] contact with [Defendant's] computer network injured [Defendant’s] business goodwill and diminished the value of its possessory interest in its computer network.”); Am. Online, Inc. v. LCGM, Inc., 46 F.Supp.2d 444, 452 (E.D.Va.1998) (citing CompuServe language).

Since eBay was issued, however, courts have become a little more circumspect about using the "slippery slope” argument. See Ticketmaster Corp. v. Tickets.Com, Inc., 2003 WL 21406289 (C.D.Cal. Mar.7, 2003) ("Since the spider does not cause physical injury to the chattel, there must be some evidence that the use or utility of the computer (or computer network) being 'spiderized' is adversely affected by the use of the spider. No such evidence is presented here. This court respectfully disagrees with other district courts' finding that mere use of a spider to enter a publicly available web site to gather information, without more, is sufficient to fulfill the harm requirement for trespass to chattels.”).

This rationale, with little to no evidentiary substantiation, has likewise justified claims under the Computer Fraud and Abuse Act ("CFAA”), 18 U.S.C. § 1030. See, e.g., Hotmail Corp. v. Van$ Money Pie, Inc., 1998 WL 388389, ¶ 34 (N.D.Cal. Apr.16, 1998) (unpublished) ("The evidence supports a finding that plaintiff will likely prevail on its [CFAA] claim and that there are at least serious questions going to the merits of this claim in that plaintiff [including] that defendants took such actions [utilizing system capacity] knowing the risks caused thereby to Hotmail's computer system and online services, which include risks that Hotmail would be forced to withhold or delay the use of computer services to its legitimate subscribers; that defendants’ actions caused damage to Hotmail; and that such actions were done by defendants without Hotmail’s authorization.”). Interestingly, the court conducting the most thorough inquiry into actual system damage did so in the process of declaring the issue to be one of triable fact, precluding summary review. See Am. Online, Inc. v. Nat'l Health Care Discount, Inc., 121 F.Supp.2d 1255, 1275 (N.D.Iowa 2000). Even in the CFAA context, however, courts rely on the "loss” rather than the "damage” language in the statute. See, e.g. EF Cultural Travel BV v. Explorica, Inc., 274 F.3d 577, 585 (1st Cir.2001), a maneuver that almost mirrors courts’ tendency to favor the server efficiency interest in name but the user efficiency interest in substance.