找到 528 篇相关文章
ABC filed its formal opposition to the Federal Communications Commission's effort to force it to submit to an early renewal of its broadcast station licenses, calling it an effort to chill the speech of broadcasters across the country. In a new filing with the FCC, ABC charges that the agency's willingness to pursue the early […]
Polymarket and Novig have both recently brought ex-DOGE affiliates onboard as scrutiny of the industry intensifies.
Introduction It was a perfectly normal day. I opened my laptop, ready to get some work done, and then... BAM. A black screen. Not a gentle fade to black, but more like my computer shouting, "I’ve had enough of your crap!" The same operating system that had been working perfectly just five hours earlier had suddenly decided it had had enough of life. I wasn't too worried though. After all, I had ChatGPT on my side. Three hours later... Yeah... my confidence crumbled faster than my phone battery at 2%. What followed was a three-hour rabbit hole involving NVIDIA drivers, multiple Linux kernels, Secure Boot, DKMS, Xorg, GDM, journalctl , systemd , and more terminal commands than I'd like to admit. Somehow, against all odds (and probably a little divine intervention), we managed to fix it. And honestly? I enjoyed every minute of the chaos. It was like a wild adventure—except with more curse words and less danger. So I decided to document the entire debugging journey—not just because it might help someone who runs into the same issue, but also because I deserve a little sympathy after spending three hours arguing with my laptop. (And if the solution seems painfully obvious to you... please let me enjoy my victory. Don't take this away from me.😤 The Problem After rebooting my laptop, I was greeted with just a black screen. No login screen, no desktop… just nothing.** At first, I tried to enter TTY using Ctrl + Alt + F3, but that wasn’t working either. Since I wasn’t able to reach TTY directly, I had to take a different route. By editing the GRUB boot entry and booting into multi-user.target , I forced Linux to start in text-only mode, giving me access to a terminal.** For this, I edited the GRUB boot entry and appended systemd.unit=multi-user.target to the end of the kernel command line (after quiet splash ). That was the first breakthrough, though. The operating system wasn’t completely dead… only the graphical interface was failing to wake up. First Clues and Initial Ass
Despite weeks of renewed press coverage and controversy around ICE, Donald Trump’s supporters appear to be standing by the agency no matter what.
Companies are making sure we have a surplus of options for building qubits.
As schools across the country continue to implement cellphone bans, a new Pew Research Center survey shows they continue to gain support. Seventy-seven percent of US adults support banning cellphones in middle and high school classes, and 48 percent support banning them for the entire school day. That's the first time more Americans have supported, […]
The US government is targeting China with a new import ban on "advanced robotic devices" and power inverters made in foreign countries, as reported earlier by Reuters. In an announcement on Tuesday, the Federal Communications Commission says the ban will include "mobile" robots, such as humanoid and quadruped models - but it is not limited […]
Every binomial confidence interval you have ever computed on an eval pass rate, Wald, Wilson, Clopper-Pearson, all of them, rests on one assumption: each example is an independent draw. Most eval sets violate it. You have 40 questions generated from the same 8 documents, or 200 turns from the same 30 conversations, or 150 examples that are really 50 cases with 3 paraphrases each. Those are not 200 independent observations. And when you feed a correlated set into a formula that assumes independence, the interval comes out too narrow, which means you declare differences significant that aren't. I want to walk through why, put a number on how much it matters, and show the fix, because this one is invisible: the code runs, the interval prints, and it is quietly wrong. Why clustering shrinks your real sample size Independent examples each carry their own information. Correlated examples carry overlapping information. If five questions come from the same document, and the model either understands that document or doesn't, those five outcomes move together. You did not learn five independent things about the model. You learned something closer to one and a half. The survey-statistics name for this is the design effect (Kish, "Survey Sampling," 1965). For clustered data it is approximately: Deff = 1 + (m̄ - 1) · ICC where m̄ is the average cluster size and ICC is the intra-cluster correlation, the fraction of total variance that lives between clusters rather than within them. Your effective sample size is: n_eff = n / Deff That is the number of independent examples your clustered set is actually worth. The number Take a realistic eval set: n = 200 examples, drawn from 40 source documents, so average cluster size m̄ = 5. Suppose the ICC is 0.3, which is unremarkable for "questions from the same document" (I have measured higher). Deff = 1 + (5 - 1) · 0.3 = 2.2 n_eff = 200 / 2.2 ≈ 91 Your 200-example eval is worth about 91 independent examples. The correct confidence interval
A Georgia man was charged with a felony for allegedly wiping his phone while being questioned by Customs and Border Protection. Samuel Tunick had something in common with others who have had their devices searched at airports: He was an activist. Unlike others who have had their phones seized with no explanation, Tunick was told […]
Donald Trump’s base has been increasingly spilt. Laura Loomer’s sudden embrace of Ukraine has only widened that divide.
New Yorkers who want to see their Instagram or TikTok feed curated by an algorithm, or receive notification after midnight, will soon have to verify their ages to access those features. New York Attorney General Letitia James (D) released a set of final rules on how to implement the SAFE for Kids Act, signed into […]
The outgoing executive director of the Office of Homeland Security Statistics is one of very few federal officials to speak out against the Trump administration’s immigration crackdown.
Most experiments that come back "no clear winner" were unwinnable on the day they launched. The data could not resolve an effect that size, and no amount of extra runtime was going to change that. You can find this out in about two minutes, before you spend anything, with one formula and a resampling pass over your own data. Here is the check, in three steps. Step 1. Compute the smallest lift your data can see For a two-arm test on a conversion rate, the smallest lift detectable at 95% confidence and 80% power is a one-liner: from math import sqrt Z_ALPHA = 1.96 # two-sided 95% Z_BETA = 0.84 # 80% power def mde ( baseline_cvr : float , n_per_arm : int ) -> tuple [ float , float ]: """ Minimum detectable effect: absolute (pp) and relative (%). """ se = sqrt ( 2 * baseline_cvr * ( 1 - baseline_cvr ) / n_per_arm ) abs_lift = ( Z_ALPHA + Z_BETA ) * se return abs_lift * 100 , abs_lift / baseline_cvr * 100 At a 3% conversion rate: clicks per arm smallest lift you can detect 5,000 +32% relative 20,000 +16% relative 100,000 +7% relative Read the middle row twice. Twenty thousand clicks per arm is a serious amount of traffic for a mid-market account, and a real 15% improvement still lands inside the confidence interval. The report will say "inconclusive," and the team will read that as a verdict on the idea. It is a verdict on the instrument. Invert the same formula and the planning question gets easier: at 3% baseline, detecting a 10% lift needs about 51,000 clicks per arm, and detecting a 5% lift needs about 203,000. If your account produces 8,000 clicks a month, you now know the honest answer to "how long should we run this." Step 2. Stop assuming your conversions are independent The formula above treats every click as an independent coin flip with the same probability. Account data does not behave that way, and the gap is not small. In a corpus of 31 advertiser accounts I maintain for diagnostic work (9.46 million search term rows, roughly $133M of spend, September 2024
Today, I’m talking with Evan Smith, who is cofounder and CEO of Altana, a company that develops software tools to manage big, messy supply chain networks around the world. We last had Evan on in early 2025 to talk about how Trump’s first few waves of tariffs were starting to affect global trade and what […]
The massive seed round was led by Index Ventures and Ribbit Capital, with participation from Sarah Guo's Conviction Partners.
“It’s not an argument with two sides, it’s an argument with 10 sides,” one senior administration official tells WIRED about how US AI policy is being shaped.
Clear a corner of your workbench and gather a handful of parts, because this palm-sized television is an afternoon build, not a semester project. Here is the shopping list for a Scanwheel of your own: A Raspberry Pi Pico to run the show An A4988 stepper driver A 21-02485 stepper motor (or a similar small NEMA-style unit) Five LEDs, plus current-limiting resistors A 3D-printed case and spinning disk The Scanwheel, built by a maker who goes by [Ancient], is a mechanical TV that fits in your hand. Instead of a glowing panel, it leans on persistence of vision: your eye holds each flash of light for a fraction of a second, so a row of blinking LEDs seen through a moving slit reads as a solid picture. Spin the disk fast enough and the flicker melts into an image. How the picture actually forms The disk sitting on top of the case carries 20 small holes spaced evenly around its edge, each drilled at a slightly different height. As the motor turns, only one hole passes in front of the LEDs at a time, so light escapes in a scanning line rather than a wash. The Pico drives the stepper up to roughly 900 RPM through the A4988, then fires the LEDs in a precise order timed to the disk position. Get that timing right and the holes trace out a grid. The payoff is a 20x20 pixel color display in the center, flanked by two more 20x20 black-and-white panels that can each show a different image. Five LEDs feed all three. The whole coordination job lives on the Pico's GPIO pins, which is why the wiring stays simple enough to manage on a breadboard before you commit anything to a soldered protoboard. Small light baffles in the base keep the LEDs from bleeding into each other, a detail worth copying if your first image looks smeared. Give it a spin The full build guide, firmware, and disk files are on the project's GitHub repository , so you can match the hole spacing and LED timing exactly. If your image drifts or tears, start by trimming the RPM and re-checking when each LED switches rela
Black holes eject powerful energy jets that blow away the surrounding gas to great distances. So how can they continue to grow?
Sequential Testing and the SPRT: How to Stop a Test Early Without Cheating Meta description: Peeking at a fixed-sample A/B test inflates false positives. Sequential testing lets you check results repeatedly and stop early without cheating. TL;DR Fixed-sample testing assumes you'll wait for a pre-calculated sample size before looking at results. Checking early and stopping the moment you see significance — "peeking" — quietly inflates your real false-positive rate, often far above the 5% you think you're getting. Abraham Wald's Sequential Probability Ratio Test (SPRT), developed for wartime quality control, is the mathematically rigorous alternative: a procedure built to be checked repeatedly, with pre-calculated boundaries that keep the false-positive rate honest by construction. The difference between the SPRT and peeking isn't willpower — it's that the SPRT's stopping rule is part of the math from the start, so stopping early doesn't cost you anything in error-rate control. Sequential design is the right call when traffic is limited, the cost of running a test too long is high, or the business genuinely can't commit to waiting for a fixed horizon — not a substitute for rigor, but a different kind of rigor suited to a different constraint. This is a methodology choice, not a shortcut — and it's one input into the broader question of how much certainty a given bet needs, covered in the Confidence Tier Model . Every experimentation program eventually hits the same moment: a test has been live for four days, the dashboard shows a lift, and someone — a stakeholder, a PM, sometimes you — asks "can we call it?" The honest answer depends entirely on what kind of test you designed, and most teams don't have a clean answer, because most teams designed a fixed-sample test and are now trying to read it like a sequential one. Those are not interchangeable. Knowing the difference, and choosing deliberately between them before the test starts, is the actual skill — not "wait lon
The Confidence Tier Model: How to Decide When Your Data Isn't Enough Meta description: Most testing programs are built for traffic they don't have. Three confidence tiers — proven, directional, speculative — each with its own bet-sizing rule. TL;DR Fixed-sample A/B testing assumes you can wait for statistical significance. Most teams can't — traffic is too thin, or the market is moving too fast to wait. The fix isn't lowering your standards. It's replacing the binary "significant / not significant" gate with three explicit confidence tiers — Proven, Directional, Speculative — each with its own evidence bar and its own bet-sizing rule. Underpowered tests systematically overestimate effect size (the "winner's curse" ). A confidence tier that accounts for this is more honest than a p-value that pretends otherwise. The way to move a learning up a tier isn't more of the same test — it's triangulation: stacking correlated, individually-weak signals until they converge. This is a methodology choice, not a compromise. Teams that name their confidence tier explicitly make faster, more defensible decisions than teams that either wait for certainty they'll never reach, or ship everything with false confidence. A product manager says: "Users want better deals." A brand marketer says: "TV is driving more direct demand." A performance marketer says: "This channel has a strong ROAS." Finance says: "But is this incremental?" Product says: "Will this hurt user trust?" Leadership says: "Should we scale this?" Six people, six kinds of evidence, and a decision that needs to get made this quarter — not whenever a test finally clears p<0.05. This is the actual job: not running tests, but converting six competing claims into one evidence base leadership can act on. Most experimentation methodology is written for a world where you have the traffic to wait for a clean answer. Most companies don't live in that world. The problem classic A/B testing doesn't solve Fixed-sample significance tes