After cherry picking and massaging statistics to an almost painful degree, Cohen takes a little shot at Google Fiber, insisting that users don't really need 1 Gbps.

"For some, the discussion about the broadband Internet seems to begin and end on the issue of "gigabit" access. To be sure, a one-gig connection has value, especially for those who have invested in "inside" networks and equipment to handle that 1-gigabit firehose of data.

The issue with such speed is really more about demand than supply. Our business customers can already order 10-gig connections. Most websites can't deliver content as fast as current networks move, and most U.S. homes have routers that can't support the speed already available to the home. As consumer demand grows for faster speeds, a competitive marketplace of wired and wireless broadband providers will be ready to serve it."

Cohen claims that 82% of Americans have access to wired high-speed Internet access of speeds exceeding 100 Mbps. But these services, provided in 85% of the country by only the local cable incumbent (the large cable companies never enter each others’ territories, and Verizon FiOS is available to just 15% of the country) are extraordinarily expensive: Comcast charges $114.95 per month for 105 Mbps download services. In Seoul, you can get symmetrical 100 Mbps (equal upload and download) access for $30/month, and there are three or four competitive choices.

By focusing on whether you need 1 Gbps, companies like Time Warner Cable and Comcast hope to steer the conversation away from how a lack of competition allows them to offer slow speeds and ever-higher prices (or the fact they're being outclassed in their own industry by a search engine)...

If the United States leads in anything in the broadband sector -- it's the use of denial and distortion by those with a vested interest in protecting the status quo. If you can convince people that everything is fine, nobody tries to fix things and your profit margins as a predatory, lumbering duopoly benefiting from regulatory capture remain high. You can legitimately argue that things are improving in many regions -- but to insist the United States is the global broadband leader is an obnoxious level of hubris, even for Comcast.

Today there is a cottage industry of critics who always want to tell us that our broadband Internet is not fast enough or satisfactory for one reason or another. The reality is that the United States is leading the way in speed, reach, and access - and doing so in a vast, rural nation that poses logistical connectivity challenges unlike any other country.

The district court’s conclusion that rules governing personal jurisdiction and venue provide no impediment to pre-Rule 26 discovery of the ISPs is legal error. A showing of “good cause,” which is required for discovery ostensibly intended to identify defendants, requires an evaluation of whether the information sought from the ISPs would be used to name and serve defendants in the forum. See, e.g., Oppenheimer Fund, Inc. v. Sanders, 437 U.S. 340, 352-53 & n.17 (1978) (where “the purpose of a discovery request is to gather information for use in proceedings other than the pending suit, discovery properly is denied”). The Copyright Act and the District of Columbia’s long-arm statute limit the court’s reach to defendants who reside in the district. And the uncontroverted evidence before the district court showed that few, if any, of the targeted Internet subscribers reside in the District of Columbia—as publicly available geolocation software used by Plaintiff’s counsel in other cases confirms. The district court’s decision to defer any consideration of personal jurisdiction or venue until after the subscribers’ personal information had been disclosed to Plaintiff requires reversal.

The court’s decision to permit discovery of the ISPs before deciding whether the 1,000-plus “Does” are misjoined provides an additional basis for reversal. Plaintiff, by routinely declining to name and serve defendants after obtaining the subscribers’ personal information, virtually ensures that Rule 20’s requirements for joinder will go unaddressed if not evaluated at the outset. And as a growing majority of courts have concluded, deferring a ruling on joinder deprives the courts of filing fees and encourages a proliferation of improperly coercive lawsuits. Given the groundswell of published opinions that disagree with the lower court and have severed or dismissed non-resident “Does” or all Does except for “Doe No. 1,” deferring a ruling on joinder in a suit that seeks nationwide subscriber information also encourages forum shopping—as the record here shows persuasively.

The record reflects that Plaintiff’s counsel’s cases have migrated across the country, with the venues selected, not by the locus of the parties or situs of harm, but based on counsel’s perceptions of which forum is most likely to authorize the greatest discovery, at the lowest cost, with the least judicial oversight.

The specter of intra-district, judge-specific shopping in Plaintiff’s counsel’s cases further underscores the problem with the lower court’s approach. The ISPs raised below Plaintiff’s counsel’s practice of filing complaints and dismissing them vel non based on the judicial assignment—only to re-file in another court. When presented with the same facts, Judge Wilkins quoted with approval Judge Huvelle’s finding: “Plaintiff’s actions a[re] akin to ‘judge shopping.’… This Court could not agree more.” ...

The ISPs respectfully submit that the district courts in this Circuit should not be the destination for 1,000-plus Doe cases that are brought primarily to compile mailing lists—not to adjudicate actual cases or controversies.

AF Holdings and its counsel owe a duty of candor to the Court, and a duty of fairness to appellants.... The serious issues concerning attorney misconduct and potentially forged documents were not identified for the court below; they necessarily affect the “good cause” analysis and provide an alternative basis for reversal to address the evidence now being considered in the pending disciplinary proceedings in the Central District of California.

Specifically, in a series of cases beginning with Dendrite Int’l, Inc. v. Doe No. 3, 775 A.2d 756, 760-61, 342 N.J. Super. 134 (App. Div. 2001), courts have adopted a balancing standard to assess requests for early discovery to identify anonymous online speakers that protects the right to speak anonymously while at the same time ensuring that plaintiffs who have valid claims are able to pursue them. Without such a standard, abusive plaintiffs could too easily use extrajudicial means against defendants from whom they could not, in the end, obtain judicial redress. See Levy, Litigating Civil Subpoenas to Identify Anonymous Internet Speakers, 37 Litigation No. 3 (Spring 2011).

The use of BitTorrent to select and share movies is expressive and, therefore, protected by the First Amendment. Call of the Wild Movie, 770 F. Supp. 2d at 350 (“[F]ile-sharers are engaged in expressive activity, on some level, when they share files on BitTorrent, and their First Amendment rights must be considered before the Court allows the plaintiffs to override the putative defendants’ anonymity.”).

Although the expressive aspect of the conduct alleged here – the posting of copyrighted movies to BitTorrent – is somewhat minimal, that does not mean that discovery to identify the anonymous user without adequate initial evidence that individual Doe Defendants committed the alleged infringement. The weakness of AF Holdings’ assertions of personal jurisdiction and proper joinder means that First Amendment concerns weigh more strongly here in favor of quashing the subpoenas. Certainly it was not appropriate for the district court to ignore the question altogether.

Consider: the top 20 companies in the United States ranked by market capitalization include no media companies. But according to figures assembled for The New York Times by Equilar, which compiles data on executive compensation, media companies employ seven of the top 20 highest paid chief executives.

The names are familiar and the numbers are large: Leslie Moonves of CBS ($60,253,647), David M. Zaslav of Discovery Communications ($49,932,867), Robert A. Iger of Walt Disney ($37,103,208), Philippe P. Dauman of Viacom ($33,396,104), Jeffrey L. Bewkes of Time Warner ($25,670,263), Brian L. Roberts of Comcast ($25,087,379), and Rupert Murdoch of News Corporation ($22,418,292).

The data indicates that average pay of the 10 highest paid chief executives for media companies was about $30 million, more than the captains of technology or finance and other industries, who average $6 million to $14 million less.

We are encouraged by the continued attention of Congress to this important issue and we look forward to working with both the House and the Senate to find a legislative balance that promotes government sharing of cyberthreat information with the private sector while also ensuring the privacy of our users.

"If a consumer fails to respond to several Copyright Alerts, Comcast will place a persistent alert in any web browser under that account until the account holder contacts Comcast's Customer Security Assurance professionals to discuss and help resolve the matter,"

The ISPs—industry giants AT&T, Cablevision, Comcast, Time Warner, and Verizon—will launch their versions of the CAS on different days throughout the week. Comcast is expected to be the first, on Monday.

As a preliminary matter, there are several confusing elements in the Subpoenas that affect Comcast's approach to responding. First, the two 2088 Subpoenas instruct Comcast to designate the person(s) "who are the most knowledgeable about the subject matter categories set forth in Exhibit though there is no "Exhibit A" appended to either of the two Subpoenas that Comcast received in that case. There are, as identified elsewhere in those Subpoenas, "Schedules A and B" but no "Exhibit A." Moreover, the Schedule A attached to the Subpoenas does not set forth any subject matter categories, but rather identifies documents requested for production. Conversely, Schedule B sets forth subject matter categories. We accordingly assume that there is no "Exhibit A" and that the reference to Exhibit A (or Schedule A) with respect to subject matter categories should be "Schedule B" However, if there is an Exhibit A that happens to be left out, or this assumption is otherwise misplaced, please notify us immediately.

Second, the three Subpoenas' Schedules A and refer to "Verizon Internet Services" and to documents pertaining to Verizon (and, thus, not to Comcast). Of course, Comcast would not have any documents relating to Verizon or its Internet Services, nor could there be any person(s) at Comcast knowledgeable about such Verizon documents. We assume these references to Verizon were intended to identify Comcast, and respond below accordingly. But if that is incorrect, there would be no documents produced, and no deponent identified, for the separate reason that Comcast would not have any responsive material(s) or information related to Verizon records.

More substantively, Comcast may not provide to any third party documents or information that include a subscriber's personally identifiable information without first ensuring compliance with the requirements of Section 631(c) of the Communications Act, 47 U.S.C. 551(c). That Section generally prohibits cable operators from disclosing such information without the subscriber's express written consent, and also imposes an affirmative obligation on a cable operator to "take such actions as are necessary to prevent unauthorized access to such information by a person other than the subscriber or cable operator." Id. 551(c)(1). Section 631(c)(2) provides three exceptions to the general ban on disclosing PII without the subscribers' express consent. Only one is applicable to your requests as you have not included any signed consents from the subscribers. Specifically, Section 631(c)(2)(B) states that disclosure is permitted "pursuant to a court order authorizing such disclosure, if the subscriber is notified of such order by the person to whom the order is directed."

We note that at a prior stage of this case you obtained a court order authorizing disclosure of PII for the individuals identified and named in Schedules A and to the Subpoenas. Indeed, the information gained in response to that earlier order allowed such identification of the subscribers in the present Subpoenas. But that court order only authorized disclosure of subscriber names and addresses, and MAC addresses. Schedules A and B to the present Subpoenas call for a great deal more than names and addresses; rather, they seek PII relating (among other things) to the DMCA record(s) for the identified subscribers, the identified subscribers' internet bandwidth usage, specific content watched or accessed by the subscribers, and bills and invoices relating to use of the service, which in turn would reveal the identified subscribers' viewing habits, service selections, costs paid, and other sensitive information. Moreover, some of the documents you seek also concern Comcast confidential and proprietary information on the operation of Comcast's network, and other document requests refer to subscriber service usage that Comcast does not track or monitor.

In fall 2009, Comcast planned to launch an Internet service for the poor that was sure to impress federal regulators. But David Cohen, the company's chief of lobbying, told the staff to wait.

At the time, Comcast was planning a controversial $30 billion bid to take over NBC Universal, and Cohen needed a bargaining chip for government negotiations.

"I held back because I knew it may be the type of voluntary commitment that would be attractive to the chairman" of the Federal Communications Commission, Cohen said in a recent interview.

The initiative may not have sealed the FCC's decision to approve the NBC merger. But it helped, Cohen said.

The proposal clearly captured the fancy of regulators. Late last month, Genachowski, the FCC chairman, touted the program, seemingly claiming some credit for its creation.

"This particular program came from our reviewing of the Comcast NBC-U transaction," Genachowski said in a speech. "Comcast embraced it as good for the country, as well as good for business. And I'm fine with that."

Among the studios with WikiLeaks movies in development are Time Warner Inc.'s HBO Films, DreamWorks Studios, Comcast Corp.'s Universal Pictures and Annapurna Pictures, the company run by Megan Ellison, daughter of Oracle Chief Executive Larry Ellison.

Like any good short-sighted investor, Moffett has urged companies to not upgrade their networks, insiting that caps and overages are the "next generation of communications." The irony of course is those telcos who listened and didn't upgrade their networks are now having their lunch eaten by cable providers on a massive scale, which, if Moffett's firm and client investment interests lean heavily toward cable operators, is something that's working out rather well for him and them.

Less network investment, less competition, higher prices. Great for investors, not so great for you

Plaintiffs should not be allowed to profit from unfair litigation tactics whereby they use the offices of the Court as an inexpensive means to gain Doe defendants’ personal information and coerce “settlements” from them. It is evident in these cases – and the multitude of cases filed by plaintiffs and other pornographers represented by their counsel – that plaintiffs have no interest in actually litigating their claims against the Doe defendants, but simply seek to use the Court and its subpoena powers to obtain sufficient information to shake down the Doe defendants. The Federal Rules require the Court to deny discovery “to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense.” Fed. R. Civ. P. 26(c)(1). This case requires such relief.

Comcast's large share of the broadband market, along with the fact that most of its customers have few if any comparable alternatives, gives Comcast significant leverage in negotiating with backbone ISPs. Comcast has traditionally been a customer of Level 3, one of the largest internet backbone providers. When a Comcast user exchanged data with a network that was not directly connected to Comcast's own network, Comcast paid Level 3 to carry that traffic.

But in November 2010, the two firms became locked in a bitter dispute. Level 3 had just won a contract to deliver content for Netflix, one of the internet's largest video services. Anticipating that Netflix would generate more traffic than the existing links between the Comcast and Level 3 networks could accommodate, Level 3 proposed installing additional links between the networks. Ordinarily, Comcast, as a Level 3 customer, would gladly accept what was essentially a free upgrade. Instead, Comcast refused to accept the new connections unless Level 3 agreed to pay Comcast for the additional traffic. And Level 3, after voicing strong objections, paid up.

Fancaster registered its mark in 1989 for broadcasting services, and over the years it's been used in connection with a range of services, "including selling Fancaster branded radios, charging customers to watch closed circuit boxing matches, producing karaoke shows, transmitting sponsored news messages to wireless pagers and cell phones, and conducting live demonstrations of FANCASTER broadcast services" (cites omitted).

In 2006, it launched Fancaster.com to broadcast short sport-related video clips. It hoped to cover such must-see events "as La Tomatina in Spain, Ostrich racing in Arizona, the Westminster Kennel Club Dog Show and the annual Nathan's Hot Dog Eating Contest." Rather than advertise the website on the Internet (you know, where people who enjoy content online might already be), they were seeking out untapped Internet enthusiasts by "marketing the website at sporting events, bars, on local television channels in Sioux Falls, South Dakota and Sioux City, Iowa, on radio stations in Charleston, South Carolina, and via flyers and handbills."

Meanwhile, in 2008 Comcast rolled out a service called fancast.com "that allowed users to watch full-length premium mainstream media over the Internet." The service was a debacle, losing $80M in less that 2 years due to “the unexpectedly high cost of distributing video content on the internet.” (Even though Comcast acquired bandwidth at wholesale rather than retail costs... how much would it have cost non-carriers to launch competitive services?). In March 2011, Comcast shut down the Fancast service and rolled the domain name over to XfinityTV.

With the overlap between the Fancaster and Fancast names, one possibility is that Comcast blatantly ripped off the name of a small startup who wouldn't want to tangle with a giant, thereby creating "reverse confusion" where everyone thinks first-mover Fancaster infringes second-comer Comcast. But another story equally fits these facts: Fancaster is doing a little trademark trolling, seeking to increase Comcast's $80M of losses by grabbing some gravy for itself. (Some gravy indeed: Fancaster's damages expert thought it would take $73M of corrective advertising to fix Comcast's damage to a brand that has no market awareness outside of Sioux City.)

It's a sad commentary on our milieu when we can't tell which litigant is bullying the other. Maybe *both* parties are equally imbibing the bullying elixir. Fancaster initially unleashed the litigation hounds, but Comcast responded with a hailstorm of countermoves, including an ACPA counterclaim for a slew of "fancast" domain names Fancaster registered after learning about Comcast's upcoming launch. A lot of lawyers appear to have satisfied their billable hour goals using this case. Yay for free-spending, deep-pocketed clients!

Trademark Infringement

The court resoundingly thumped Fancaster's core argument about consumer confusion, miraculously finding a way to twist all of the factors to Comcast's favor. The judge may have cut some analytical corners, but that says the judge simply didn't accept Fancaster's narrative.

The court specifically rejected the possibility of initial interest confusion, citing 3rd Circuit precedent that basically limited IIC to competitors, because the parties didn't directly compete. The court also dismissed Fancaster's efforts to show overlaps in search engine results, saying "the confusion one encounters on an Internet search engine is a twenty-first century version of that experienced when searching the phone book." (I am going to be doing some work this quarter to show that the initial interest confusion doctrine almost never succeeds in court any more, and therefore it imposes costs on both litigants for no gain. This case is just one example of that.)

The court also scoffed at Fancaster's request for $73M for corrective advertising:

There is not a shred of evidence of any damage to the Fancaster mark caused by Comcast. The only loss to Fancaster that Mr. Krueger could testify to was that resulting from pursuing the instant litigation against Comcast.

Evidentiary Issues

Comcast had survey expert Hal Poret do two surveys. The court tossed the first one because it didn't adequately replicate market conditions by not presenting consumers with a navigable website:

use of a printout and static screenshots, instead of live websites, provide ample grounds on which to exclude the March 2009 survey. For one, it is difficult to fathom how presenting a respondent with a paper printout of the FANCAST homepage in anyway replicates how an Internet user would encounter and perceive the FANCAST website in the marketplace. Websites, particularly those that offer video content, are meant to be viewed on a computer and allow consumers to browse and interact with them via hyperlinks. The FANCAST printout offered none of these aspects. Similarly, although viewed on a computer, the static screenshots of the fancaster and control website homepages did not allow respondents to interact with them as they ordinarily would in the marketplace.

I haven't researched this issue, but this ruling may tell us something important about the requirements for consumer surveys when websites are involved.

Conclusion

This ruling eviscerated Fancaster's case, making it a strong win for Comcast, but it left a few residual legal issues open. Yet, the legal battle has been mooted by the passage of time. Comcast already stopped using Fancast as a brand, and Fancaster still hasn't shown a lot of movement towards developing a real business or even a revenue model. Are the parties really going to spend more money on a pointless lawsuit? We all know what the answer should be; let's see how they actually answer.

For more on the case, see Rebecca Tushnet's 43(B)log .

Now that nearly 20 million households in the U.S. are able to use DNSSEC, we feel it is an important time to urge major domain owners, especially commerce and banking-related sites, to begin signing their domain names. While in the past those domains may have wanted to do so but felt it would have limited effect, they now can work on signing their domains knowing that the largest ISP in the U.S. can validate those signatures on behalf of our customers.

When we launched the Domain Helper service, we also set in motion its eventual shutdown due to our plans to launch DNSSEC. Domain Helper has been turned off since DNS response modification tactics, including DNS redirect services, are technically incompatible with DNSSEC and/or create conditions that can be indistinguishable from malicious modifications of DNS traffic (including DNS cache poisoning attacks). Since we want to ensure our customers have the most secure Internet experience, and that if they detect any DNSSEC breakage or error messages that they know to be concerned (rather than not knowing if the breakage/error was "official" and caused by our redirect service or "unofficial" and caused by an attacker), our priority has been placed on DNSSEC deployment -- now automatically protecting our customers...

The software is unfriendly to Mac users running Firefox: It changes the browser's homepage to comcast.net, and blocks users from changing it to anything else.

I heard this from a friend who'd just signed up for Comcast's Xfinity high-speed Internet service and soon discovered some behavior on his Mac that is akin to Windows malware - something had hijacked his Internet settings. The technician who arrived to turn on the service said that a software package from Comcast was necessary to complete the installation. My friend later discovered that his homepage had been changed to comcast.net, and that Comcast software had modified his Firefox profile so that there was no way to change the homepage setting. 

I contacted Comcast; they initially blamed the problem on a bug in Firefox. Mozilla denies this, and says it's Comcast's doing.

"This is NOT a Firefox bug or issue," a Mozilla spokesperson wrote in an email. "It is a Comcast method that applies preference changes to Firefox."

Comcast spokesman Charlie Douglas acknowledged that the Xfinity software hijacks Firefox's settings. He said the problem is limited to Mac users, and that permanency of the change was unintentional. He added that the company is in the process of correcting the installation software.

"Customers absolutely should be able to change their preferred homepage anytime," Douglas said. "We're obviously apologizing for any inconvenience we've caused Mac users."

If, after these educational and acknowledgment alerts, the subscriber’s account still appears to be engaged in content theft, the ISP will send yet another alert. At this time, the ISP may take one of several steps, referred to as “Mitigation Measures” reasonably calculated to stop future content theft. These Mitigation Measures may include, for example: temporary reductions of Internet speeds, redirection to a landing page until the subscriber contacts the ISP to discuss the matter or reviews and responds to some educational information about copyright, or other measures (as specified in published policies) that the ISP may deem necessary to help resolve the matter.