Aggregator

In a big win for free speech online, the U.S. Court of Appeals for the D.C. Circuit ruled that a federal agency violated the First Amendment when it blocked animal rights activists from commenting on the agency’s social media pages. We filed an amicus brief in the case, joined by the Foundation for Individual Rights and Expression (FIRE).

People for the Ethical Treatment of Animals (PETA) sued the National Institutes of Health (NIH) in 2021, arguing that the agency unconstitutionally blocked their comments opposing animal testing in scientific research on the agency’s Facebook and Instagram pages. (NIH provides funding for research that involves testing on animals.)

NIH argued it was simply implementing reasonable content guidelines that included a prohibition against public comments that are “off topic” to the agency’s social media posts. Yet the agency implemented the “off topic” rule by employing keyword filters that included words such as cruelty, revolting, tormenting, torture, hurt, kill, and stop to block PETA activists from posting comments that included these words.

NIH’s Social Media Pages Are Limited Public Forums

The D.C. Circuit first had to determine whether the comment sections of NIH’s social media pages are designated public forums or limited public forums. As the court explained, “comment threads of government social media pages are designated public forums when the pages are open for comment without restrictions and limited public forums when the government prospectively sets restrictions.”

The court concluded that the comment sections of NIH’s Facebook and Instagram pages are limited public forums: “because NIH attempted to remove a range of speech violating its policies … we find sufficient evidence that the government intended to limit the forum to only speech that meets its public guidelines.”

The nature of the government forum determines what First Amendment standard courts apply in evaluating the constitutionality of a speech restriction. Speech restrictions that define limited public forums must only be reasonable in light of the purposes of the forum, while speech restrictions in designated public forums must satisfy more demanding standards. In both forums, however, viewpoint discrimination is prohibited.

NIH’s Social Media Censorship Violated Animal Rights Activists’ First Amendment Rights

After holding that the comment sections of NIH’s Facebook and Instagram pages are limited public forums subject to a lower standard of reasonableness, the D.C. Circuit then nevertheless held that NIH’s “off topic” rule as implemented by keyword filters is unreasonable and thus violates the First Amendment.

The court explained that because the purpose of the forums (the comment sections of NIH’s social media pages) is directly related to speech, “reasonableness in this context is thus necessarily a more demanding test than in forums that have a primary purpose that is less compatible with expressive activity, like the football stadium.”

In rightly holding that NIH’s censorship was unreasonable, the court adopted several of the arguments we made in our amicus brief, in which we assumed that NIH’s social media pages are limited public forums but argued that the agency’s implementation of its “off topic” rule was unreasonable and thus unconstitutional.

Keyword Filters Can’t Discern Context

We argued, for example, that keyword filters are an “unreasonable form of automated content moderation because they are imprecise and preclude the necessary consideration of context and nuance.”

Similarly, the D.C. Circuit stated, “NIH’s off-topic policy, as implemented by the keywords, is further unreasonable because it is inflexible and unresponsive to context … The permanent and context-insensitive nature of NIH’s speech restriction reinforces its unreasonableness.”

Keyword Filters Are Overinclusive

We also argued, related to context, that keyword filters are unreasonable “because they are blunt tools that are overinclusive, censoring more speech than the ‘off topic’ rule was intended to block … NIH’s keyword filters assume that words related to animal testing will never be used in an on-topic comment to a particular NIH post. But this is false. Animal testing is certainly relevant to NIH’s work.”

The court acknowledged this, stating, “To say that comments related to animal testing are categorically off-topic when a significant portion of NIH’s posts are about research conducted on animals defies common sense.”

NIH’s Keyword Filters Reflect Viewpoint Discrimination

We also argued that NIH’s implementation of its “off topic” rule through keyword filters was unreasonable because those filters reflected a clear intent to censor speech critical of the government, that is, speech reflecting a viewpoint that the government did not like.

The court recognized this, stating, “NIH’s off-topic restriction is further compromised by the fact that NIH chose to moderate its comment threads in a way that skews sharply against the appellants’ viewpoint that the agency should stop funding animal testing by filtering terms such as ‘torture’ and ‘cruel,’ not to mention terms previously included such as ‘PETA’ and ‘#stopanimaltesting.’”

On this point, we further argued that “courts should consider the actual vocabulary or terminology used … Certain terminology may be used by those on only one side of the debate … Those in favor of animal testing in scientific research, for example, do not typically use words like cruelty, revolting, tormenting, torture, hurt, kill, and stop.”

Additionally, we argued that “a highly regulated social media comments section that censors Plaintiffs’ comments against animal testing gives the false impression that no member of the public disagrees with the agency on this issue.”

The court acknowledged both points, stating, “The right to ‘praise or criticize governmental agents’ lies at the heart of the First Amendment’s protections … and censoring speech that contains words more likely to be used by animal rights advocates has the potential to distort public discourse over NIH’s work.”

We are pleased that the D.C. Circuit took many of our arguments to heart in upholding the First Amendment rights of social media users in this important internet free speech case.

大阪・関西万博へのイスラエル参加不承認を求める署名活動がオンライン署名サイト「change.org」上で取り組まれている。（藤ヶ谷魁）

電波監理審議会（第1131回）会議資料

弾道ミサイルを想定した住民避難訓練の実施

情報通信審議会 電気通信事業政策部会 通信政策特別委員会（第15回）開催案内

内国郵便約款の変更認可

競争ルールの検証に関するWG（第59回）開催案内

松本総務大臣閣議後記者会見の概要

令和6年7月25日からの大雨に関する被害状況等について（第20報）

ふるさと納税に関する現況調査結果の概要

「革新的情報通信技術(Beyond 5G(6G))基金事業」　令和6年度社会実装・海外展開志向型戦略的プログラム（共通基盤技術確立型）の公募

令和6年7月25日からの大雨による被害に係る普通交付税（9月定例交付分）の繰上げ交付

情報通信行政・郵政行政審議会　郵政行政分科会（第93回）配布資料・議事概要・議事録

情報通信審議会　情報通信技術分科会　陸上無線通信委員会 小電力システム作業班　TPMS・RKEシステムアドホックグループ（第6回）の開催について

第39回政策評価審議会（第39回政策評価制度部会と合同）（令和6年6月25日開催）資料・議事要旨・議事録

Last month’s unprecedented global IT failure should be a wakeup call. Decades of antitrust inaction have made many industries dangerously reliant on the same tools, making such crises inevitable. We must demand regulators break up the digital monocultures that are creating a less competitive, less safe, and less free digital world.

The Federal Trade Commission (FTC) solicited public comments last year on the state of the cloud computing market. EFF made it clear that the consolidation of service providers has created new dangers for everyone and urged the commission to encourage interoperability so customers could more easily switch and mix cloud services. Microsoft cautioned against intervention, touting the benefits of centralized cloud services for IT security.

A year later, a key cloud-based cybersecurity firm released a bug unique to Microsoft systems. Vital IT systems were disrupted for millions worldwide. 

This fragility goes beyond issues at a specific firm, it results from power being overly concentrated around a few major companies.

What Happened

The widespread and disruptive tech outage last month happened thanks to an overreliance on one particular tool, CrowdStrike's Falcon sensor software. While not a monopoly, this tool is the most popular in end-point protection platforms.

This niche service often used by companies is best understood as an antivirus tool for devices, controlled by a cloud platform. “End-point” computers run the agent with very deep system permissions to scan for security issues, and the company CrowdStrike regularly pushes remote software updates to this tool. This setup means many devices rely on a single source for their security, leveraging shared insights learned across devices. It also means that many devices share a single point of failure.

Instead of an inconvenience for a few companies, it more closely resembled a government shutdown or a natural disaster.

An early sign of this problem came last April, when a CrowdStrike update disrupted devices running Debian and Rocky Linux operating systems. Linux “end-point” devices are uncommon, let alone those running these specific distributions with CrowdStrike software. What should have been a red flag in April was instead barely a blip.

Last month CrowdStike disrupted two other operating systems with a bad update: Windows 10 and 11. This time it spurred a Y2K-like collapse of crucial computer systems around the globe. Airlines, hospitals, financial institutions, schools, broadcasters, and more were brought to a standstill as an erroneous update on CrowdStrike’s platform caused system crashes. Instead of an inconvenience for a few companies, it more closely resembled a government shutdown or a natural disaster.

Both cases had similar impacts to devices, but the later case was an absolute disaster for infrastructure because of a digital landscape dominated by a few key players. Having so many sectors rely on a handful of services for the same operating systems makes them all susceptible to the same bugs, with even systems running absurdly old versions of Windows gaining an advantage for providing some diversity.

Whatever went wrong at CrowdStrike was just a spark. Last month it ignited the powder keg of digital monocultures.

Digital Monoculture

All computers are broken. Every piece of software and hardware is just waiting to fail in unexpected ways, and while your friendly neighborhood hackers and researchers can often hold off some of the worst problems by finding and reporting them, we need to mitigate inevitable failures. A resilient and secure digital future can’t be built on hope alone.

Yet, that’s exactly what we’re doing. The US has not just tolerated but encouraged a monopolistic tech industry with too little competition in key markets. Decades of antitrust policy have been based on the wrongheaded idea that sheer size will make tech companies efficient and better able to serve customers. Instead, we have airports, hospitals, schools, financial systems, and more all reliant on the same software, vulnerable to the same bugs and hacks. We created a tech industry that is too big to fail.

The lack of diversity makes the whole ecosystem more fragile

We live in the age of the digital monoculture, where single vulnerabilities can tear through systems globally; sabotaging hospitals and city governments with ransomware; electrical systems with state-sponsored attacks; and breaching staggering amounts of private data. Name a class of device or software, and more often than not the majority of the market is controlled by a few companies—often the same ones: Android and iPhone; Windows and Mac; Gmail and Outlook; Chrome and Safari.  When it comes to endpoint security products three companies control half of the market, the largest being Microsoft and CrowdStrike.

Much like monocultures in agriculture, the lack of diversity makes the whole ecosystem more fragile. A new pest or disease can cause a widespread collapse without a backup plan. The solution, conversely, is to increase diversity in the tech market through tougher antitrust enforcement, and for organizations to make IT system diversity a priority.

Allowing an over-reliance on a shrinking number of companies like Microsoft will only ensure more frequent and more devastating harms in the future.

How we got here Broken Antitrust

As EFF has pointed out, and argued to the FTC, antitrust has failed to address the realities of a 21st-century internet.

Viewing consumers as more than walking wallets, but as individuals who deserve to live unburdened by monopoly interests.

Since the 1980s, US antitrust has been dominated by “consumer welfare” theory, which suggests corporate monopolies are fine, and maybe even preferable, so long as they are not raising prices. Subtler economic harms of monopoly, along with harms to democracy, labor rights, and the environment are largely ignored.

 For the past several years, the FTC has pressed for a return to the original intent of antitrust law: viewing consumers as more than walking wallets, but as individuals who deserve to live unburdened by monopoly interests.

But we have a long way to go. We are still saddled with fewer and less adequate choices built on a tech industry which subsidizes consumer prices by compromising privacy and diminishing ownership through subscriptions and restrictive DRM. Today’s empires of industry exert more and more influence on our day to day life, building a greater lock-in to their monoculture. When they fail, the scale and impact rival those of a government shutdown.

We deserve a more stable and secure digital future, where an error code puts lives at risk. Vital infrastructure cannot be built on a digital monoculture.

To do this, antitrust enforcers, including the FTC, the Department of Justice (DOJ), and state attorneys general must increase scrutiny in every corner of the tech industry to prevent dangerous levels of centralization. An important first step would be to go after lock-in practices by IT vendors.

Procurement and Vendor Lock-In

Most organizations depend on their IT teams, even if that team is just the one friend who is “good with computers”. It’s quite common for these teams to be significantly under-resourced, forced to meet increasingly complex needs from the organization with a stagnant or shrinking budget.

Lock-in doubles down on a monopoly’s power and entrenches it across different markets.

This squeeze creates a need for off-the-shelf solutions that centralize that expertise among vendors and consultants. Renting these IT solutions from major companies like Microsoft or Google may be cost-effective, but it entrusts a good deal of control to those companies.

All too often however, software vendors take advantage of this dynamic. They will bundle many services for a low initial price, making an organization wholly reliant on them, and then hinder the ability of the organization to adopt alternative tools while later raising prices. This is a longstanding manipulative playbook of vendor lock-in.

Once locked in, a company will discover switching to alternatives is costly both in terms of money and effort. Say you want to switch email providers. Rather than an easy way to port over data and settings, your company will need to resort to manual efforts or expensive consultant groups. This is also often paired with selective interoperability, like having an email client work smoothly with a bundled calendar system, while a competitor’s service faces unstable or deliberately broken support.

Lock-in doubles down on a monopoly’s power and entrenches it across different markets. That is why EFF calls for interoperability to end vendor lock-in, and let IT teams choose the tools that reflect the values and priorities of their organization.

Buying or building more highly-tailored systems makes sense in a competitive market. It’s unlikely a single cloud provider will be the best at every service, and with interoperability, in-house alternatives become more viable to develop and host. Fostering more of that internal expertise can only bolster the resilience of bigger institutions.

Fallout from The Cloud

Allowing the economy and the well-being of countless people to rely on a few cloud services is reprehensible. The CrowdStrike Falcon incident is just the latest and largest in a growing list of hacks, breaches, and collapses coming to define the era. But each time everyday people endure real harms.

Each time, we see the poorest and most marginalized people face costly or even deadly consequences. A grounded flight might mean having to spend money on a hotel, and it might mean losing a job. Strained hospital capacity means fewer people receive lifesaving care. Each time these impacts further exacerbate existing inequalities, and they are happening with increasing frequency.

We must reject this as the status quo. CrowdStrike’s outage is a billion-dollar wake-up call to make antitrust an immediate priority. It's not just about preventing the next crash—it's about building a future where our digital world is as diverse and resilient as the people who depend on it.

　７月７日に投開票が行われた東京都知事選は小池都知事が３選を果たした。写真は投票前日の７月６日夜に新宿で行われた蓮舫氏の最終街宣。駅前を埋めた溢れんばかりの聴衆に最後の訴えをアピールして盛り上がった。熱気が凄く、若い人たちが多く参加してこれまでとは違う雰囲気で、支援者が投票を呼び掛けて各地で展開した「ひとり街宣」の方たちも見受けられ選挙結果に期待が高まった。しかし無党派票を石丸伸二氏に多く持っていかれての敗北。今回はネットやSNSの連動が投票効果をより上げることも実感したが、..