Funniest/Most Insightful Comments Of The Week At Techdirt

from the speak-and-spell dept

This week, both our winners on the insightful side come in response to our post about Dartmouth's insane paranoia over cheating, and the resultant surveillance scandal. In first place, it's That One Guy with a comment about the fact that the college appeared to be hiding its due process policies:

Nothing says 'We're on the right side of this issue and students can trust that they will be treated fairly if they come here' like trying to hide your own campus policies while pressuring students to 'admit' to cheating based upon evidence not provided to them to refute.

In second place, it's an anonymous commenter questioning why certain things count as "cheating" to begin with:

I never understood the point of memorizing formulas. Being able to recite the quadratic formula from memory doesn't mean I'm better at solving an equation.

The ability to use the knowledge properly is what needs to be tested, not simply checking to see if you can regurgitate the text.

For editor's choice on the insightful side, we start out with another anonymous comment, responding to the notion that pharma patents are important because the drug companies "take all the risks":

No. Pharma does not take "all the risks" (what risks?), and regularly makes use of publicly-funded (or otherwise not funded by pharma) research and data, then locks up not only the end product, but also previously publicly available information and data resulting from their trials. This is all bullshit and allowing it is an unconscionable and unfair subsidy to these companies. Further, they have an avenue to extend patents unfairly by otherwise pointlessly tweaking a formula or process, which somehow keeps the original process locked up.

Seriously, for real?

Next, it's Stephen T. Stone with an important addition to our frequently-made point about how the powerful abuse content moderation tools to suppress the vulnerable:

And sometimes not on purpose. When adult-oriented content (e.g., porn) gets driven from platforms, LGBTQ people are inevitably hit first and hardest by such bans because of preëxisting biases about LGBTQ content — namely, that such content is inherently sexual/adult-oriented.

Over on the funny side, our first place winner is UhHuh with a comment about the latest bad Section 230 reform bill and how its requirement for publicly available reports interacts with its other requirements:

That website better have a top-notch consumer protection policy....

In second place, it's That One Guy deploying a portion of a favorite joke in response to the simple question of exactly which "conservative" views are supposedly being censored around the web:

Oh, you know...

For editor's choice on the funny side, we start out with a comment from JMT about the bogey man of "left-wing fascism":

Nearly as bad as Christian Satanism but not quite as bad as vegan cannibalism...

Next, it's Baron von Robber with a reaction to the news about Trump's DOJ targeting journalists to protect his reputation:

Trump projects so hard, he could run 6 movies simultaneously at a drive-in.

That's all for this week, folks!

This Week In Techdirt History: May 9th - 15th

from the back-then dept

Five Years Ago

This week in 2016, the copyright fight over the Star Trek fan film Axanar was allowed to move forward — though without the too-soon attempt to raise the silly question of copyright on the Klingon language. The UK government was pushing for ten-year jail sentences for copyright infringement, HBO was abusing the DMCA process to stop Game of Thrones spoilers, and we took a look at Minnesota's insanely broad publicity rights law. It was also the week of the opening statements in the Oracle v. Google case that would carry on for years.

Ten Years Ago

This week in 2011, two important DMCA cases (IsoHunt and Veoh) were being heard by the 9th Circuit, while Limewire settled its own case, the US Copyright Group was allowed to move forward with a massive shakedown operation, and BMI was arguing that a single person listening to their own music via the cloud was a public performance under copyright law. Copyright maximalists were also opposing new TLDs for some reason and trying to get domain censorship capabilities included in the .net TLD. This was also the week that we first saw the PROTECT IP act and its extremely bad text that could gut parts of the DMCA and make linking a felony.

Fifteen Years Ago

This week in 2006, Apple (the tech company) won its trademark battle against Apple (the record label) with help from the now-famous "moron in a hurry" defense. Movie studios were tepidly trying to embrace BitTorrent while failing to understand it, due in large part to the industry's obsession with DRM (though at the same time, the recording industry magnanimously decided it would allow people to rip their own CDs, and Sony was admitting that its DRM-laden proprietary music format was a strategic error). We took a bigger look at why the argument for the necessity of copy protection doesn't make sense, in stark contrast to the analyst who was arguing that a lack of mobile DRM will lead to billions in losses.

Filed Under: history, look back

Angry Joe Tears Into Twitch Over Its One-Sided Approach To DMCA Takedowns

from the strike-a-match dept

Famed YouTuber and Twitch streamer Angry Joe, or Jose Antonio Vargas, has made it onto Techdirt's pages in the past. True to his name, we've discussed his responses on a couple of intellectual property issues he's suffered through. When Nintendo flagged a video Angry Joe did about Mario Party 10, preventing him from further monetizing the video, he simply and angrily swore off of doing any Nintendo videos in the future, rightly noting that with the decision all the free advertising he'd given Nintendo just disappeared. When CBS blocked a review video he did because the review used 13 seconds of Star Trek: Picard, he took to Twitter to rip them to shreds as well. The point is that when Angry Joe encounters the frustrations many others deal with thanks to overly restrictive intellectual property practices, he doesn't stay silent. He gets... well... angry.

With that in mind, it's probably not all that surprising that the way Twitch is handling its DMCA process has finally caught up to Angry Joe resulting in yet another angry rant. Like many other Twitch creators, Angry Joe has been hit with DMCA notices. As he goes on to note, the process Twitch has designed for how this all works, well, sucks out loud.

Joe highlighted two major issues with Twitch’s system. The first is the strikes on people’s accounts are permanent, and don’t expire after 90 days like YouTube.

“Copyright strikes do not depreciate after a set period of time. If I receive three strikes my channel is terminated, yet I can’t see the strike, I can’t submit a counter-notification, I can’t submit a dispute,” he said in a recent rant. “It can’t get any worse than this.”

This seems both like it should be an easy problem to fix and an easy problem to have completely avoided in the first place. If anyone had bothered to test out Twitch's DMCA process prior to putting it in place, especially from the perspective of someone who had received multiple DMCA notices, it would have been immediately obvious that multiple strikes create a situation where the accused can't even counter the notice.

But the issues with the dispute program go deeper than that.

Joe used an example of a copyright strike he received in February 2021 for a Warzone clip dating back to November 2020, which had the Astronomia song ⁠— popularized by the coffin dance meme ⁠— playing inside the game.

“Here is your DMCA strike. What is the clip? We don’t have a link to the clip. We told you to delete all of your VODs. There is no ‘Dispute’ button,” he explained. “You can’t even review the VODs identified in DMCA notifications in your dashboard, and they don’t plan [on adding that feature] that until December 2021.”

If this all sounds absurd, it very much is. We've made this point before, but Twitch creators are Twitch's biggest and most important asset. Twitch is owned by Amazon. Amazon has virtually unlimited resources. Instead of bringing those resources to bear on all this nonsense, Twitch has instead hung its most important assets out to dry. The roadmap items are half a year away. Creators are getting slammed with these DMCA requests. Twitch is mostly doing nothing other than admiring the failed program it has rolled out.

It should be noted that Angry Joe has a very loud voice when it comes to streamers. If this doesn't get fixed soon, this could lead to a creator exodus.

Filed Under: angry joe, dmca, jose antonio vargas, streaming
Companies: twitch

Chinese Government's Hacker Competition Is Being Used To Find Exploits To Wield Against Uighur Citizens

from the aiding-and-abetting dept

Anything the Chinese government can weaponize against its Uighur Muslim population, it will. And has. Further details about an iPhone exploit discovered by Chinese hackers show the Chinese government got into the bug bounty program solely to find vulnerabilities to wield against the government's least-liked residents.

Patrick Howell O'Neill's article for MIT Technology Review points out Chinese hackers used to participate in popular hacking competitions like Pwn2Own, providing invaluable assistance to tech companies and tech users by finding vulnerabilities that could be patched before they were exploited by malicious hackers.

In 2017, Chinese participation in international competitions came to a halt. The founder and CEO of tech giant Qihoo 360 publicly criticized Chinese hackers for helping foreign tech companies find and patch security flaws. The CEO suggested this talent should stay at home and help the government find vulnerabilities to exploit.

That's exactly what has happened. The Chinese government banned participation in foreign hacking competitions and started its own. The first homegrown event was won by a researcher working for Qihoo 360, who found an exploit that allowed malicious actors to take control of even the latest iPhones simply by steering the iPhone user to a webpage containing malware.

This was patched two months later by Apple, quietly and with little attention drawn to it. But incidents occurring in the two months between the discovery and the patch didn't go unnoticed. Google's security researchers observed unusual activity and wrote about it.

[I]n August of [2019], Google published an extraordinary analysis into a hacking campaign it said was “exploiting iPhones en masse.” Researchers dissected five distinct exploit chains they’d spotted “in the wild.” These included the exploit that won Qixun [of Qihoo 360] the top prize at Tianfu, which they said had also been discovered by an unnamed “attacker.”

Now, more details about that string of attacks has been revealed. And it shows the Chinese government took the winning exploit and weaponized it against its Uighur population.

Shortly after Google’s researchers noted the attacks, media reports connected the dots: the targets of the campaign that used the Chaos exploit were the Uyghur people, and the hackers were linked to the Chinese government. Apple published a rare blog post that confirmed the attack had taken place over two months: that is, the period beginning immediately after Qixun won the Tianfu Cup and stretching until Apple issued the fix.

This has now been confirmed by another source: the US government. Its surveillance agencies also picked up on the malicious hacking efforts and noted their targeting of China's favorite target of oppression. And it was the government's intervention that sped up Apple's response to the exploit.

The US quietly informed Apple, which had already been tracking the attack on its own and reached the same conclusion: the Tianfu hack and the Uyghur hack were one and the same. The company prioritized a difficult fix.

This is the sort of cooperation one prefers to see. The federal government has often portrayed Apple as an enemy -- not just of agencies like the DOJ, but of the American public. In this case, the government worked with Apple to stop attacks on foreign citizens by a foreign government. The Chinese government made the most of this exploit for two months -- one it obtained through a homegrown hacking competition that appears to exist solely to create offensive tech weapons for state-ordained hacking.

Meanwhile, the hacker who discovered the exploit and collected the cash (all while working for the company whose CEO called for Chinese hackers to stop helping foreign companies and start helping The Man stick it to locals and foreign adversaries) is trying to distance himself from the damage his exploit has wrought.

When we contacted Qixun Zhao via Twitter, he strongly denied involvement, although he also said he couldn’t remember who came into possession of the exploit code. At first, he suggested the exploit wielded against Uyghurs was probably used “after the patch release.”

Both of these claims are untrue. And both have been debunked by both independent research and US government surveillance. While it's unwise to tangle with the Chinese government by refusing to hand over discovered vulnerabilities, it's probably a little easier to sidestep that obligation by sitting out government-sponsored hacking competitions. In the end, this isn't the researcher's fault. The government chose to use it this way. But anyone entering a Chinese government-sponsored hacking competition is likely well aware any discoveries they make will be weaponized by an extremely oppressive government.

Filed Under: china, exploits, hackathon, hackers, hacking competition, surveillance, uighurs
Companies: apple

FISA Court Says FBI May Be Abusing Surveillance Powers; Will Continue To Allow It To Abuse Surveillance Powers

from the but-the-court-has-made-it-clear-it's-disappointed dept

Reform efforts following the Snowden leaks led to some minor improvements at the FISA Court. The USA Freedom Act gave the court permission to allow someone to present the side of the surveilled from time to time and introduced some reporting requirements that allowed Americans to see just a bit more about how their surveillance tax dollars were spent.

But if anything is going to change the way America's surveillance agencies perform their duties, it's going to be up to the FISA Court, which can reject requests and shut down surveillance programs until they're fixed. The Section 215 collection of phone records may have reached an early end-of-life after the untargeted bulk collection of phone records was banned, but other surveillance authorities continue unaltered. The one that leads to most surveillance abuses is one the court can't seem to bring under control: Section 702.

This is supposed to be surveillance of foreign national security targets, but agencies with access to 702 collection far too routinely use this as a backdoor to American's communications. And nothing about that has changed, eight years down the road from the Snowden leaks. A newly declassified decision [PDF] from the FISA Court shows the court is still willing to overlook egregious abuses of this authority to spy on Americans.

Here's the EFF's summary of last November's decision -- one made public late last month by the FISA Court.

[T]he FISC again found that a series of overly complex but still ultimately swiss cheese agency protocols -- that are admittedly not even being followed -- resolve the Fourth Amendment problems caused by the massive governmental seizures and searches of our communications currently occurring under FISA Section 702. The annual review by the FISC is required by law -- it’s supposed to ensure that both the policies and the practices of the mass surveillance under 702 are sufficient. It failed on both counts.

The key part of this summary is this: "admittedly not being followed." Even when the government admits it's violating rights, the FISA Court says that's ok… as long as the government admits it. Candor is appreciated, but it would be better if the government stopped violating its own protocols and the rights of Americans. That's where the FISA Court should draw the line. But it doesn't. And it hasn't in years.

While the Court has expressed concerns about certain communications being swept up in 702 collections and disseminated to agencies who won't use this information solely for national security purposes, it's ok with that happening, so long as the NSA and other agencies check all the internal guideline boxes before violating the Fourth Amendment or, as is specifically discussed in this case, accessing privileged attorney-client communications.

The FISA Court says it has no problem with the NSA accessing privileged communications since it won't be directly involved in any criminal cases. But it can see how other agencies without the same foreign-facing directives might end up in the possession of privileged conversations.

That being said, the government is admonished to guard against the possibility that NSA, in compliance with its procedures, might disseminate to FBI a report based on a privileged communication described in Section 5(c) of the NSA procedures (pertaining to a criminal charge in the United States) that, had the FBI obtained it through its own collection efforts, the FBI would be required to sequester with the Court under FBI Minimization Procedures…

Swell, but the Court goes on to say this isn't something it's going to worry about.

The Court again concludes that NSA’s procedures, as a whole and applied to it, an agency with no law-enforcement mission or authority, are reasonably designed to protect the substantial privacy interests in attorney-client communications, consistent with the need to exploit those communications for legitimate foreign-intelligence purposes.

This kind of ignores the fact that prosecutions of foreign terrorists generally involve US courts. So even if the NSA isn't exactly in the law enforcement business, its collections aid and abet domestic law enforcement agencies that do directly work with prosecutors and build criminal cases.

Marcy Wheeler's summary of the FISA opinion is even more caustic. Here's the bullet point breakdown of the Court's acquiescence to the government's claims that it's really, really hard to keep the FBI from looking at stuff it's not supposed to be looking at when it accesses 702 collections.

• It took time for them to make the changes in their systems

• It took time to train everyone

• Once everyone got trained they all got sent home for COVID

• Given mandatory training, personnel “should be aware” of the requirements, even if actual practice demonstrates they’re not

• FBI doesn’t do that many field reviews

• Evidence of violations is not sufficient evidence to find that the program inadequately protects privacy

• The opt-out system for FISA material — which is very similar to one governing the phone and Internet dragnet at NSA until 2011 that also failed to do its job — failed to do its job

• The FBI has always provided national security justifications for a series of violations involving their tracking system where an Agent didn’t originally claim one

• Bulk queries have operated like that since November 2019

In short, the FBI keeps screwing up. And it keeps telling the FISA Court it will do better. Then the FBI doesn't actually improve and the Court gives it a pass because it's trying to try. The bulwark put in place to deter the government from abusing its powers and violating rights isn't doing its job. It's deciding that national security work is too important to subject to serious judicial scrutiny.

Filed Under: abuse, doj, fbi, fisa, fisa court, fisc, mass surveillance, nsa, section 702, surveillance, usa freedom act

Daily Deal: The 2021 Microsoft Azure And Security Training Bundle

from the good-deals-on-cool-stuff dept

The 2021 Microsoft Azure And Security Training Bundle has 5 courses to help you master Microsoft Azure. You'll learn about core Azure services, pricing and support. You'll also learn how to implement, manage, and secure a company's Microsoft Azure environment. The bundle is on sale for $50.

Note: The Techdirt Deals Store is powered and curated by StackCommerce. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

Filed Under: daily deal

The Flopping Of Trump's Blog Proves That It's Not Free Speech He's Upset About; But Free Reach

from the it's-the-audience dept

A week ago, we wrote about Trump's new blog, which was designed to look vaguely tweet-like, noting that this proved that he never needed Twitter or Facebook to speak freely. He's always been able to speak on his own website. NBC News has an interesting story now, suggesting that the blog just isn't getting that much attention.

A week since the unveiling, social media data suggests things are not going well.

The ex-president’s blog has drawn a considerably smaller audience than his once-powerful social media accounts, according to engagement data compiled with BuzzSumo, a social media analytics company. The data offers a hint that while Trump remains a political force, his online footprint is still dependent on returning to Facebook, Twitter and YouTube.

The Desk of Donald J. Trump is limited — users can’t comment or engage with the actual posts beyond sharing them to other platforms, an action few people do, according to the data.

Some have been using this to argue that Twitter and Facebook's bans on the former president were attacks on his "free speech." But it actually demonstrates something different -- and important. Everyone complaining about the removal of Trump's account are not actually mad about the "free speech" part of it. They're really mad about the "free reach." (Hat tip to Renee DiResta for making this point years ago).

Being kicked off these platforms by the platforms (as opposed to, say, the government) is not an attack on your ability to speak. There are lots of places to do that. It is, instead, an attack on having easy access to an audience on those platforms. And, as far as I can tell, there is no right to having as large an audience as possible. Thus, in the same sense that I can't demand a million followers on any of these platforms, the former president similarly can't demand that they supply him with the audience of their users.

Filed Under: access, audience, content moderation, donald trump, free reach, free speech
Companies: facebook, twitter

Microsoft Data Shows That The FCC's Broadband Maps Are Fantasy

from the rose-colored-glasses dept

We've noted for a very long time that despite a lot of lip service about broadband, the U.S. government still doesn't have a very good idea of where broadband is or isn't available. There's a long line of reasons for this, including political pressure by regional monopolies that very much don't want a lack of competition or these coverage gaps to be apparent (somebody might get the crazy idea to try and fix the problem!). The FCC has also long been criticized for methodology that declares a census block (which can be hundreds of square miles in rural areas) "served" with broadband if just one home can theoretically get service from an ISP.

After decades of this, the government last year finally took action and passed the DATA Act, which tasks the FCC with fixing the problem. The Biden FCC last February also belatedly announced a "task force" to examine the problem in more detail. But we're still years away from any of these changes being implemented (assuming they're even implemented competently), something of cold comfort to those struggling to afford internet access right now during a pandemic and economic crisis.

Meanwhile, the problem persists. The Verge this week utilized data from Microsoft then compared it to FCC and industry data to create a more accurate map of U.S. broadband access. In numerous regions, the disparity is almost comical:

"The disparity between FCC reports and the Microsoft data can be shocking. In Lincoln County, Washington, an area west of Spokane with a population just a hair over 10,000, the FCC lists 100 percent broadband availability. But according to Microsoft’s data, only 5 percent of households are actually connecting at broadband speeds."

Note that even this data isn't entirely accurate because it's measuring what speeds are currently being used, not what's always available. That means users may have access to faster speeds, but potentially didn't subscribe due to high costs. That in turn is due to a lack of competition, something experts say also isn't accurately measured by the FCC. But it's not just Microsoft data that showcases how the FCC's broadband availability maps aren't remotely connected to reality.

If you want to see the problem in personalized detail, just spend a little time with the FCC's $350 million broadband map, which all but hallucinates competitors, availability, and speeds -- while excluding (at telecom industry behest) any meaningful measurement of a key metric: price. $350 million is a steep price tag just to look at the world through rose-colored glasses.

Of course, industry has historically opposed broadband mapping reform, because it only highlights a lack of access and competition, driving folks to action. The false progress painted by bad data is also routinely used by telecom-allied lawmakers (like Ajit Pai) to justify the dismantling of government oversight or the rubber stamping of megamergers that reduce competition even further. And while the resulting "digital divide" is often treated as a problem that magically appeared one day, it's important to remember that this is a policy choice, and the direct result of twenty years of government policies that prioritized the interests of regional monopolies (not coincidentally tethered to our intelligence gathering systems).

Filed Under: broadband, competition, data, fcc