惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

爱范儿
爱范儿
E
Exploit-DB.com RSS Feed
Google DeepMind News
Google DeepMind News
F
Full Disclosure
D
Darknet – Hacking Tools, Hacker News & Cyber Security
T
ThreatConnect
Stack Overflow Blog
Stack Overflow Blog
Last Week in AI
Last Week in AI
Martin Fowler
Martin Fowler
G
GRAHAM CLULEY
C
Check Point Blog
T
Threatpost
I
Intezer
Spread Privacy
Spread Privacy
The Register - Security
The Register - Security
Project Zero
Project Zero
月光博客
月光博客
人人都是产品经理
人人都是产品经理
阮一峰的网络日志
阮一峰的网络日志
D
DataBreaches.Net
IT之家
IT之家
Malwarebytes
Malwarebytes
T
The Blog of Author Tim Ferriss
P
Privacy International News Feed
P
Palo Alto Networks Blog
T
The Exploit Database - CXSecurity.com
量子位
李成银的技术随笔
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Cisco Talos Blog
Cisco Talos Blog
Know Your Adversary
Know Your Adversary
美团技术团队
The GitHub Blog
The GitHub Blog
T
Tor Project blog
M
MIT News - Artificial intelligence
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Google Online Security Blog
Google Online Security Blog
P
Proofpoint News Feed
有赞技术团队
有赞技术团队
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
博客园 - 司徒正美
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
C
Comments on: Blog
T
Threat Research - Cisco Blogs
aimingoo的专栏
aimingoo的专栏
Security Latest
Security Latest
NISL@THU
NISL@THU
The Cloudflare Blog
H
Help Net Security
Recent Commits to openclaw:main
Recent Commits to openclaw:main

The Cloudflare Blog

The day my ping took countermeasures Announcing Claude Compliance API support with Cloudflare CASB Announcing Claude Managed Agents on Cloudflare Project Glasswing: what Mythos showed us Our billing pipeline was suddenly slow. The culprit was a hidden bottleneck in ClickHouse Browser Run: now running on Cloudflare Containers, it’s faster and more scalable When "idle" isn't idle: how a Linux kernel optimization became a QUIC bug Building For The Future How Cloudflare responded to the “Copy Fail” Linux vulnerability When DNSSEC goes wrong: how we responded to the .de TLD outage Code Orange: Fail Small is complete. The result is a stronger Cloudflare network Introducing Dynamic Workflows: durable execution that follows the tenant Post-quantum encryption for Cloudflare IPsec is generally available Agents can now create Cloudflare accounts, buy domains, and deploy Shutdowns, power outages, and conflict: a review of Q1 2026 Internet disruptions Making Rust Workers reliable: panic and abort recovery in wasm‑bindgen Moving past bots vs. humans Building the agentic cloud: everything we launched during Agents Week 2026 The AI engineering stack we built internally — on the platform we ship Orchestrating AI Code Review at scale Introducing the Agent Readiness score. Check to see if your site is agent-ready Shared Dictionaries: compression that keeps up with the agentic web Redirects for AI Training enforces canonical content Unweight: how we compressed an LLM 22% without sacrificing quality Agents that remember: introducing Agent Memory Agents Week: network performance update Introducing Flagship: feature flags built for the age of AI Cloudflare’s AI Platform: an inference layer designed for agents Building the foundation for running extra-large language models AI Search: the search primitive for your agents Deploy Postgres and MySQL databases with PlanetScale + Workers Artifacts: versioned storage that speaks Git Email for agents - Cloudflare Email Service now in public beta Project Think: building the next generation of AI agents on Cloudflare Introducing Agent Lee - a new interface to the Cloudflare stack Register domains wherever you build: Cloudflare Registrar API now in beta Browser Run: give your agents a browser Rearchitecting the Workflows control plane for the agentic era Add voice to your agent Managed OAuth for Access: make internal apps agent-ready in one click Securing non-human identities: automated revocation, OAuth, and scoped permissions Scaling MCP adoption: Our reference architecture for simpler, safer and cheaper enterprise deployments of MCP Secure private networking for everyone: users, nodes, agents, Workers — introducing Cloudflare Mesh Building a CLI for all of Cloudflare Durable Objects in Dynamic Workers: Give each AI-generated app its own database Agents have their own computers with Sandboxes GA Dynamic, identity-aware, and secure Sandbox auth Welcome to Agents Week 500 Tbps of capacity: 16 years of scaling our global network From bytecode to bytes- automated magic packet generation Cloudflare targets 2029 for full post-quantum security How we built Organizations to help enterprises manage Cloudflare at scale Why we're rethinking cache for the AI era Our ongoing commitment to privacy for the 1.1.1.1 public DNS resolver Introducing EmDash — the spiritual successor to WordPress that solves plugin security Introducing Programmable Flow Protection: custom DDoS mitigation logic for Magic Transit customers Cloudflare Client-Side Security: smarter detection, now open to everyone How we use Abstract Syntax Trees (ASTs) to turn Workflows code into visual diagrams A one-line Kubernetes fix that saved 600 hours a year Sandboxing AI agents, 100x faster Inside Gen 13- how we built our most powerful server yet Launching Cloudflare’s Gen 13 servers- trading cache for cores for 2x edge compute performance Powering the agents: Workers AI now runs large models, starting with Kimi K2.5 Introducing Custom Regions for precision data control Standing up for the open Internet- why we appealed Italy’s Piracy Shield fine From legacy architecture to Cloudflare One Announcing Cloudflare Account Abuse Protection: prevent fraudulent attacks from bots and humans Slashing agent token costs by 98% with RFC 9457-compliant error responses AI Security for Apps is now generally available Building a security overview dashboard for actionable insights Investigating multi-vector attacks in Log Explorer Translating risk insights into actionable protection: leveling up security posture with Cloudflare and Mastercard Fixing request smuggling vulnerabilities in Pingora OSS deployments Active defense: introducing a stateful vulnerability scanner for APIs Complexity is a choice. SASE migrations shouldn’t take years. From the endpoint to the prompt: a unified data security vision in Cloudflare One Ending the "silent drop": how Dynamic Path MTU Discovery makes the Cloudflare One Client more resilient A QUICker SASE client: re-building Proxy Mode How Automatic Return Routing solves IP overlap Always-on detections: eliminating the WAF “log versus block” trade-off Mind the gap: new tools for continuous enforcement from boot to login Stop reacting to breaches and start preventing them with User Risk Scoring Defeating the deepfake: stopping laptop farms and insider threats Moving from license plates to badges: the Gateway Authorization Proxy Evolving Cloudflare’s Threat Intelligence Platform: actionable, scalable, and ETL-less Introducing the 2026 Cloudflare Threat Report See risk, fix risk: introducing Remediation in Cloudflare CASB How Cloudy translates complex security into human action From reactive to proactive: closing the phishing gap with LLMs Modernizing with agile SASE: a Cloudflare One blog takeover Beyond the blank slate: how Cloudflare accelerates your Zero Trust journey The truly programmable SASE platform Toxic combinations: when small signals add up to a security incident We deserve a better streams API for JavaScript The most-seen UI on the Internet? Redesigning Turnstile and Challenge Pages ASPA: making Internet routing more secure Bringing more transparency to post-quantum usage, encrypted messaging, and routing security How we rebuilt Next.js with AI in one week Cloudflare One is the first SASE offering modern post-quantum encryption across the full platform Cloudflare outage on February 20, 2026
Breaking down broadband nutrition labels
Cloudflare Team · 2022-04-15 · via The Cloudflare Blog

2022-04-15

7 min read

As part of the recently passed Infrastructure Investment and Jobs Act (Infrastructure Act) in the United States, Congress asked the Federal Communications Commission (FCC) to finalize rules that would require broadband Internet access service providers (ISPs) display a “label” that provides consumers with a simple layout that discloses prices, introductory rates, data allowances, broadband performance, management practices, and more.

A sample Broadband Nutrition Facts from the original 2016 FCC proposal.

A sample Broadband Nutrition Facts from the original 2016 FCC proposal.

While the idea of a label is not new (the original design dates from 2016), its inclusion in the Infrastructure Act has reinvigorated the effort to provide consumers with information sufficient to enable them to make informed choices when purchasing broadband service. The FCC invited the public to submit comments on the existing label, and explain how the Internet has changed since 2016. We’re sharing Cloudflare’s comments here as well to call attention to this opportunity to make essential information accessible, accurate, and transparent to the consumer. We encourage you to read our full comments. (All comments, from Cloudflare and others, are available for public consumption on the FCC website.)

The Internet, 6 years ago

Six years can change a lot of things, and the Internet is no exception. For example, Tiktok barely existed as a company at the start of 2016; now it is the most popular site in the world. The global population that uses the Internet increased from 3.4 billion people in 2016 to 5.2 billion in 2021, which represents a growth of 52%. According to Statista, users in 2015 spent around 5.5 hours with digital media; now users spend almost 8 hours with digital media. The amount of data consumed on the Internet in 2021 was 79 zettabytes, which is a number that is expected to more than double in only two years. Users are more dependent on the Internet now than ever before.

Users being more dependent on the Internet has been amplified during the pandemic. According to Pew Research, 90% of American adults say the Internet has been essential or important for them personally during the coronavirus outbreak. Forty percent of American adults say they used digital technology or the Internet in new or different ways compared with before the beginning of the outbreak. A home broadband connection is no longer primarily for recreation, but a necessity for equitable access to education, healthcare, and as of 2020, it’s now even essential for many employment opportunities.

With that dependency, though, comes a higher expectation of quality. In 2016, users were more tolerant of poor performance: they were just happy if their Internet worked. Furthermore, applications were typically less latency sensitive: things like VoIP and video chats were less prevalent than they are today. Nowadays, however, video chats are almost ubiquitous: we use them at work and at home with increasing frequency. If these applications are slow or perform poorly, it’s hugely impactful to the user experience. We think of it as “our Internet cutting out,” and we lose the engagement with whomever we’re talking to.

Our increased dependence on the Internet has in turn increased our expectations for good Internet performance.

Your Internet should be graded on performance

Because the Internet has become more focused on performance in 2022, we believe that your Internet providers should disclose to you how good they are at providing a good experience for these applications that are now mission critical.

Previously, performance was measured by bandwidth, or the size of the pipe between you and what you want to access. However, bandwidth is much more widely available today than it was six years ago. Median download throughput increased from 39 Mbps in 2016 to 194 Mbps in 2021. This increase in throughput has opened up new uses of home Internet connections, and new opportunities to look holistically at the Quality of Experience (QoE) of home broadband. We believe that metrics beyond bandwidth such as latency and jitter (the variance in latency) have grown appreciably in importance and that should be reflected in policy going forward.

Transparency into broadband Internet performance isn’t just important to consumers, though. With more and more enterprises relying on the Internet to reach both customers and also employees, it has become a foundational part of the American economy. So many businesses rely on Cloudflare because they want their digital assets delivered to customers, partners, and employees quickly. Enterprises want to secure their network with our cloud because our edge services are physically close to users and can be reached with low latency. Performance is no longer a luxury — it is increasingly a necessity.

The FCC defined latency in 2016 as “the time it takes for a data packet to travel from one point to another in a network.” While technically true, the vagueness of this definition presents certain issues. The latency between two points could be arbitrary, or as is the case with current speed tests, measuring a path that is never traversed by consumers in daily Internet usage. To put it succinctly: we don’t know what is being measured or whether what’s measured reflects reality.

While there is ambiguity about what latency ISPs would show on their broadband label, Cloudflare, and other content providers, can see latency from the other side – from our edge servers that are serving websites to consumers. What we see is that rural states have higher latency than more dense states.

Figure 1: 50th percentile TCP Connect Time (ms) to Major Content Delivery Networks

*Alaska and Hawaii have TCP Connect times of 263ms and 160ms respectively. Data compiled by Cloudflare from the HTTP Resource Timings API‌‌

As an example, Cloudflare offers a browser isolation product that runs a web browser in our cloud, an application that is extremely sensitive to latency. To achieve these latencies, we’ve connected directly with 10,000 distinct networks across more than 270 global data center locations. We estimate that 95% of Internet users globally can reach Cloudflare-protected websites and services in under 50 milliseconds.

So while Cloudflare supports the FCC’s effort to increase understanding of cost and privacy of Internet Service Provider offerings and wants the labels to be expedited to provide real consumer value, we have suggestions to significantly augment the labels to provide a better view of how your Internet does at providing services to you. Standardizing technical measurements across the Internet is a big topic, and in some cases we suggest the FCC build stakeholder consensus on additional future changes to the label.

For the broadband performance section of the label, we recommended:

  • Renaming “download speed” (and “upload”) to “throughput,” “bandwidth”, or “capacity.” We can’t deny “speed” has become conversationally interchangeable with throughput, but they aren’t the same. As the Internet continues to grow, “speed” will mean how fast the Internet is, which will be measured in latency and overall quality of service, not just throughput. The latter is simply the amount of bits a connection can handle in the downstream direction at any given time.

  • Adding “jitter” to the label. With the pandemic-driven rise of video conferencing, jitter —the variation, or stability, of latency in an Internet connection—has become a common cause of issues. Found yourself saying “my Internet is cutting out” or “am I frozen? Oh, I’m back”? That’s likely jitter.

  • Add methodological transparency and work towards standards for how latency, jitter, and packet loss are measured. Consumers should be able to make apples-to-apples comparisons between ISP offerings, but to do that,  a standard in how ISPs measure these numbers is needed. Rather than a hasty mandate from the FCC, our suggestion is to take the time to engage stakeholders on the best approaches.

The end goal of these recommendations is to make sure that standards on performance match the experiences users have on the Internet. Today, speed tests and other forms of Internet measurement often query endpoints that are embedded into ISP networks that don’t see any traffic beyond measurements, and this can produce misleading results that may lead users to think that their Internet experience is better than it actually is. If your measurements don’t follow the same paths and are treated the same as normal Internet traffic, your measurements will look better. We believe that performance measurements should closely approximate the user experience, so that you have the complete picture of how your Internet is performing.

Disclosing Network Management

However, network performance isn’t only about how well your provider takes bits from your device to where they need to go. Sometimes network performance can be impacted by network management techniques. Providers may institute techniques like traffic shaping, which will slow down traffic to and from specific high-bandwidth sites to ensure that other sites don’t see congestion and degraded performance. Other providers may implement bandwidth caps, where specific users who consume lots of data may be slowed down if they exceed a threshold, a technique commonly used for mobile networks.

To help address these issues, we recommended including policy level line-items in the network management section instead of merely a yes-or-no answer. For example, if an ISP slows traffic after a certain amount of data has been consumed in a month, that information should be accessible on the label itself.

Privacy Disclosures

For the privacy section of the label, our recommendation is that a link to a dense and rarely-read ISP privacy policy is not sufficient transparency into how an ISP will use subscribers’ data. We recommended a privacy section that gives consumers insight into:

  • Collection and retention of information: The label should indicate whether the ISP collects and retains any information beyond what is strictly necessary to provide services to the subscriber, including web browsing history and location data, as well as how long that information is retained.

  • Use of information: The label should indicate whether data collected by the ISP is used for purposes other than what is strictly necessary to provide the broadband service to the consumer, such as for advertising.

  • Sharing of information: The label should indicate whether the ISP shares or sells the data collected, including location or browsing information data, with third parties.

  • Opt out: The label should indicate whether the ISP provides options to opt-out of data use and sharing (whether the ISP receives consideration for such sharing).

  • Security of information: The label should indicate whether the ISP provider has technical mechanisms in place to secure data from unauthorized access, including whether it encrypts metadata about a consumer’s browsing habits, and mechanisms in place to report breaches.

We also suggested that the FCC make the data presented in the label accessible in a machine-readable format for researchers and consumers.

The Internet is built on users

We commend Congress for including broadband nutrition labels in the Infrastructure Investment and Jobs Act, and the FCC for moving quickly to implement the labels. The current broadband label, the product of years of work, will be a significant improvement over what we have now – nothing.

However, we don’t believe that the labels should stop there. While the labels from 2016 go a long way towards providing clarity into how much money users pay for their Internet and create a good standard for pricing, the Internet and the way people interact with it is so different now than it was six years ago. We need to ensure that we are representing the user experience to its fullest, as this will ensure that our Internet experience can continue to improve over the next six years and beyond.

ResearchPolicy & Legal