Loading ...
5th Circuit and Server Efficiency
Much has been made, at least privately, of the 5th Circuit’s refusal to accept the “server efficiency” argument put forward by the University of Texas in the recent White Buffalo Ventures decision.
The answer to why the “server efficiency” argument was rejected is placed within the plain language of the decision. Notice:
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.
One might persuasively present evidence that 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. 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 imposed by its prohibition.” Cincinnati v. Discovery Network, Inc., 507 U.S. 410, 417 (1993) (internal citations and quotations omitted). (pg. 10)
The second paragraph there tells the tale: “We must nonetheless consider the evidence in the light most favorable to the nonmovant.”
When conducting a summary judgment hearing the Court is obliged to take all testimony at its weakest for the party trying to get the Court to dismiss the case and at its strongest for the other party. The way to get around this obligation is to provide the Court with information that backs up the allegation.
If the University of Texas, as the party trying to get White Buffalo’s lawsuit dismissed, wanted to truly rely upon the server efficiency argument, then they needed to present testimony supporting the argument. They tried to do this, but only by providing the trial court with the list of IP addresses they were blocking. There’s nothing wrong with doing this, but there appeared to be no link between this list and the efficiency of the servers other than the blanket statement made by Daniel Updegrove that if they couldn’t block that UT’s system would not be able to function.
The obvious question here is “why?” What is it that would happen if those blocks were removed? Would the server start billowing smoke? Would the hard drives rattle apart? Would the system become so slow as to be unusable?
But even more than that, there is the question of fit with White Buffalo’s spam specifically. In an effort to blunt the server efficiency argument, White Buffalo agreed to both rate and time limiting. This is certainly an offer that body-part-enhancement and drug spammers are going to refuse to make. Thus, White Buffalo is acknowledging the server efficiency argument and asking for some way to lessen that effect through an agreement.
If the University of Texas wants to continue with the server efficiency argument, they would have to find a way to tie in the use of lookup tables to ensure that mail is blocked except during the agreed-to hours, man hours in server configuration, and man hours in negotiations for rate and time limits (which would have to necessarily be someone with some ability to bind UT to a contract and a high degree of familiarity with the mail server and its configuration).
This is not to say that something couldn’t be done by UT in order to use this argument, but White Buffalo did a masterful job at taking this argument away from them at summary judgment by offering to work with them.
This is what the 5th Circuit means when it writes:
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. 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. (pg. 11)
The problem with the server efficiency argument then, is that saying “the cumulative effect of all spam harms my server” does nothing to blunt the counter-argument of “we’ll work with you so that my spam does not harm your server.”
And now it’s time for the one, great, big, huge caveat to all of this: The University of Texas is a political subdivision of the State of Texas. The reason that these arguments hold any water at all is because a political subdivision of a state is not allowed, under normal circumstances, to prohibit commercial speech unless it can show that it has some rational reason for doing so.
You may do what you please with your own personal, private property. Even block White Buffalo Ventures.
MickC @ August 8, 2005
[...] ght. This is especially true with Footnote 24. Stay tuned. UPDATE III: My notes on the Server Efficiency argument mentioned on pages 10 and fol [...]
Let’s come at this from a different approach.
The underlying reason the server efficiency argument failed in this case is not that WBV did a good job of presentation.
Rather the evidence of UT failed to meet to any objective test.
In essence, UT said:
“it is not right to send unsolicited bulk e-mail”
This is a totally subjective basis to justify blocking commercial speech. It did not impress the Court as a rational basis to justify the filtering policies of UT.
What is the point I am making?
* Network operators should not look at Sec. 8(c) and say, this provision allows us to do “anything” we want (my server my rules).
This is especially true where the network, even though privately owned, operates in a public or quasi-public capacity.
* Network operators operating public or quasi-public facilities will want to read section 8 (c) in light of the first amendment.
* The rational for blocking e-mail should have some “good faith basis” to survive the constitutional test that under-pins section 8 (c).
(See for example 47 U.S.C. § 230)
All that being said, it is not right to send UBE.
But a Court will require a stronger evidentiary basis to support the server efficiency position then merely the bald claim, “it is not right” put in response to the spammer’s position “we will work with you through rate limiting and delivery timing.”
A couple of general comments:
* This case is somewhat different because UT is a State entity, although the Court did not determine that UT’s servers are “public.”
* The American Courts have extended first amendment protection beyond merely State entities to include public places, even though the public discourse is taking place on private property.
* To avoid difficulties, I am suggesting in the case of networks that cater to the public, the Court’s may read into section 8 (c) a “good faith” requirement.
What we need to consider is the flip side, where a consumer network (Internet access service) is blocking requested bulk mail from being delivered to its customers and takes the position under section 8(c) – our server, our rules.
If push came to shove:
* Would the Courts read into section 8 (c) a good faith requirement?
* How does 47 U.S.C. § 230 interact with section 8 (c) of the CAN SPAM Act of 2003?
* And on a broader scale, does the failure to prohibit the sending of commercial e-mail without consent under the CAN SPAM Act of 2003 create a problem?
The Act allows unsolicited bulk commercial e-mail, while allowing networks to make their own rules in handling e-mail.
Is solicited bulk commercial e-mail the only legitimate form of commercial e-mail?
If so, does the failure to prohibit unsolicited bulk commercial e-mail, while by necessity permitting networks to make their own rules mean the Act is an unreasonable burden on legitimate commerce?
As an example, by allowing networks to have their own rules, does this means that:
* Subject to external pressure from other networks, the larger back-bone providers for example can host spammers, and their support facilities, ignore zombie botnets and simply say, our network, our rules?
- and -
* Unless there is across the board peer pressure, there is a disincentive on the large back bone providers to incure the necessary expenditures to enforce network security?
John Glube
Toronto, Canada
Let’s look at this from another angle.
It is fair to say that WBV raised some interesting questions concerning the “server efficiency argument”
Unfortunately, in response, UT said in essence:
“it is not right to send unsolicited bulk e-mail”
Although I agree with the sentiment, this is a subjective basis to justify blocking commercial speech. It did not satisfy the Court (at least on a summary judgement motion) as a rational basis to justify UT’s filtering policies, although the Court did uphold UT’s filtering policies based on the user efficiency argument.
So?
* The Court became involved in a constitutional commercial free speech analysis because UT is a “State Political Unit.”
* Can an Internet access service as that phrase is defined in 47 U.S.C. 231(e)(4), that is not a “State or Federal political unit” use Sec. 8(c) of the CAN SPAM Act of 2003 to say, “this provision allows us to do anything we want with e-mail going across our network and therefore we are immune from suit?”
My own view?
I would anticipate the Courts will say that section 8 (c) simply means that the CAN SPAM Act of 2003 imposes no obligations on the “filtering” policies of Internet access services.
In turn, the Courts will need to look to State and Federal law, read in light of the American Constitution to ascertain whether the particular Internet access service is immune from suit, given its “filtering” policies.
For this reason, I would suggest that Internet access services will at least want to have a “good faith basis” for their filtering decisions, to allow for potential immunity under 47 U.S.C. 230.
[...] Regular readers, of course, are already familiar with this story. For related entries here see here, here, and here. [...]
Does server efficiency on a University’s IT system require that faculty members’ email boxes be capped at 20MB when commercial providers offer email boxes of 1GB or half for free, and allow up to 10MB attachments to a
single message?
One federal district court said yes, and cited Judge Spark’s summary judgment opinion in the White Buffalo litigation in support. See Faculty Rights Coalition v. Shahrokhi, 2005 WL 1657116, 177 L.R.R.M. (BNA) 3194, S.D.Tx., Jul 13, 2005.
http://www.faculty-rights-coalition.com/FRCvShahrokhi.html
Jan. 9 , 2006. U.S. Supreme Court denies petition for writ of certiorari in
White Buffalo Ventures, LLC v. University of Texas at Austin, 420 F.3d 366 (5th Cir. 2005). See http://www.supremecourtus.gov/docket/05-520.htm
In denying review in case in which an off-campus matchmaker service sought
judicial relief against the University’s interception of its messages, the Court let stand a ruling by the Fifth Circuit Court of Appeals, which upheld the University’s blocking policy even though the messages met the requirements imposed by the CAN-SPAM Act. The case did not present the question whether members of the campus communicty could be banned likewise if they used the university’s system to distribute communications with commerical or political content.
http://www.faculty-rights-coalition.com/WhiteBuffaloCasenote.html
Does server efficiency on a University’s IT system require that faculty members’ email boxes be capped at 20MB when commercial providers offer email boxes of 1GB or half for free, and allow up to 10MB attachments to a
single message?
No, of course not. But this is a specious argument at best.
Server efficiency does not require anything more than the University decides that it requires. The decision about what is required for servers to efficiently do their jobs should rationally be made based upon the resources available to the University, not by the resources available to a third-party provider, whether that be AOL, Yahoo, or Google.
In your case, it seems to me that your problem is that you think the policies established by the University should not apply to you based upon the (political) content of the speech. But those policies were already in place and everyone had to live under them.