Bookmarks 2025-05-06T19:00:58.459Z
by Owen Kibel
53 min read
76 New Bookmarks
| Favicon | Details |
|---|---|
 |
"Rosa Parks on Her Vag": Megyn Kelly on Met Gala Obscenity From Blackpink's Lisa and Halle Berry - YouTube May 6, 2025 "Rosa Parks on Her Vag": Megyn Kelly on Met Gala Obscenity From Blackpink's Lisa and Halle Berry YouTube "Rosa Parks on her vag": Megyn Kelly on Met Gala obscenity from Blackpink's Lisa and Halle Berry.LIKE & SUBSCRIBE for new videos everyday: https://bit.ly/3Aw... |
|
May 6, 2025
Democrats PRETEND To Be Outraged Over Trump Pope Photo, THEY ARE LIARS
YouTube
SUPPORT THE SHOW BUY CAST BREW COFFEE NOW - https://castbrew.com/Sign Up For Exclusive Episodes At https://timcast.com/Merch - https://timcast.creator-spring... |
| | Germanyâs Friedrich Merz Becomes Chancellor on Second Try but Emerges Weakened - WSJ
May 6, 2025 |
| 
May 6, 2025
Trump has already screwed over Germanyâs new chancellor
POLITICO
Germanyâs most important and powerful ally for many decades â the US â has only further weakened conservative leader Friedrich Merz. |
|
May 6, 2025
Cardinals are watching âConclaveâ the movie for guidance on the conclave IRL
POLITICO
The 2024 movie is proving a useful primer for clerics about to take part in the real thing to choose the next pope. |
| 
| Giving V8 a Heads-Up: Faster JavaScript Startup with Explicit Compile Hints ¡ V8
May 6, 2025
Giving V8 a Heads-Up: Faster JavaScript Startup with Explicit Compile Hints ¡ V8
Explicit compile hints control which JavaScript files and functions are parsed and compiled eagerly |
| | JavaScript Just Became 10X faster Thanks to a New Game-changing Feature From Chrome | HackerNoon
May 6, 2025
JavaScript Just Became 10X faster Thanks to a New Game-changing Feature From Chrome | HackerNoon
Chrome's V8 team just dropped a game-changing feature that makes JavaScript blazingly fast!
Hello JavaScript Enthusiasts! Welcome to a new edition of "This Week in JavaScript"! This Week in JavaScript This week, we've witnessed a seismic legal victory that changed the app development landscape forever, groundbreaking performance boosters for JavaScript engines, and incredible tools going completely free. This Google Chrome Feature Makes JavaScript Run Faster This Google Chrome Feature Makes JavaScript Run Faster Chrome's V8 team just dropped a game-changing feature that makes JavaScript blazingly fast! Chrome's V8 team just dropped a game-changing feature Chrome's V8 team just dropped a game-changing feature Key Features: Key Features: Explicit Compile Hints let developers control which functions compile at startup Simple magic comment (//# allFunctionsCalledOnLoad) tells V8 what to prioritize Average load time improvements of 630 milliseconds across major websites Zero refactoring required - just add comments and watch performance soar Explicit Compile Hints let developers control which functions compile at startup Simple magic comment (//# allFunctionsCalledOnLoad) tells V8 what to prioritize (//# allFunctionsCalledOnLoad) Average load time improvements of 630 milliseconds across major websites Zero refactoring required - just add comments and watch performance soar This isn't just an incremental updateâit's a turbo boost for web applications with minimal developer effort. GSAP is now Free GSAP is now Free In a stunning move, Webflow has acquired GSAP and made EVERYTHING completely free! Webflow has acquired GSAP Webflow has acquired GSAP What's Included: What's Included: All premium plugins (SplitText, MorphSVG) now available to everyone SplitText 3.13 completely rewritten: 50% smaller with 14 new features Perfect accessibility, responsive layouts, and emoji support New feature: animate TO CSS variables for dynamic theming All premium plugins (SplitText, MorphSVG) now available to everyone (SplitText, MorphSVG) SplitText 3.13 completely rewritten: 50% smaller with 14 new features Perfect accessibility, responsive layouts, and emoji support New feature: animate TO CSS variables for dynamic theming The animation ecosystem just democratized overnight, putting professional-grade tools in every developer's hands. The JavaScript Hack No One Talks About The JavaScript Hack No One Talks About Converting values to a string in JavaScript might sound like a small task, but it is surely very complicated. Here's the developer lifesaver you didn't know you needed! Converting values to a string in JavaScript Converting values to a string in JavaScript The Problem: The Problem: Most string conversion methods fail on symbols and null-prototype objects Different approaches crash on different types Your code randomly explodes with certain values Most string conversion methods fail on symbols and null-prototype objects Different approaches crash on different types Your code randomly explodes with certain values The Solution: The Solution: Object.prototype.toString.call(value) handles everything perfectly Works with symbols, null-prototypes, and all edge cases It's like having a universal translator for JavaScript values Object.prototype.toString.call(value) handles everything perfectly Works with symbols, null-prototypes, and all edge cases It's like having a universal translator for JavaScript values Tools & Releases You Should Know About Tools & Releases You Should Know About Let's speed-run through some of the other big tool updates this week! Deno 2.3: Compile your apps into single files with everything included, install packages 2x faster, and automatically clean up memory leaks with the new âusingâ keyword. Perfect for building standalone executables that work anywhere without requiring users to install Deno or dependencies. Prisma 6.7: Ditched Rust for TypeScript to run faster, split your database models into separate files for better organization, and now supports JavaScript-native SQLite out of the box. The new architecture makes Prisma feel like a native JavaScript library, eliminating the binary overhead that slowed down previous versions. Electron 36: Upgraded browser engine and Node.js, Windows apps now have rounded corners that look native, code signing is built-in so Microsoft trusts your apps, and smooth animations without performance caps. The new ServiceWorkerMain API lets you control background workers from the main process, opening up new possibilities for desktop app architecture. Koa 3.0: A tiny web framework (under 3MB) from the Express team that handles caching, errors, and content negotiation automatically - perfect for building lightweight APIs. Unlike heavier frameworks, Koa gives you just what you need without bloat, making it ideal for microservices and serverless deployments. PGlite 0.3: Run a full PostgreSQL database in your browser (3MB size), works offline with IndexedDB storage, supports AI vector search, and ideal for testing or edge computing apps. Now upgraded to PostgreSQL 17.4, bringing all the latest database features to client-side applications without external dependencies. Deno 2.3: Compile your apps into single files with everything included, install packages 2x faster, and automatically clean up memory leaks with the new âusingâ keyword. Perfect for building standalone executables that work anywhere without requiring users to install Deno or dependencies. Deno 2.3: Deno 2.3: Prisma 6.7: Ditched Rust for TypeScript to run faster, split your database models into separate files for better organization, and now supports JavaScript-native SQLite out of the box. The new architecture makes Prisma feel like a native JavaScript library, eliminating the binary overhead that slowed down previous versions. Prisma 6.7: Prisma 6.7: Electron 36: Upgraded browser engine and Node.js, Windows apps now have rounded corners that look native, code signing is built-in so Microsoft trusts your apps, and smooth animations without performance caps. The new ServiceWorkerMain API lets you control background workers from the main process, opening up new possibilities for desktop app architecture. Electron 36: Electron 36: Koa 3.0: A tiny web framework (under 3MB) from the Express team that handles caching, errors, and content negotiation automatically - perfect for building lightweight APIs. Unlike heavier frameworks, Koa gives you just what you need without bloat, making it ideal for microservices and serverless deployments. Koa 3.0: Koa 3.0: PGlite 0.3: Run a full PostgreSQL database in your browser (3MB size), works offline with IndexedDB storage, supports AI vector search, and ideal for testing or edge computing apps. Now upgraded to PostgreSQL 17.4, bringing all the latest database features to client-side applications without external dependencies. PGlite 0.3: PGlite 0.3: And that's it for the thirty-third issue of "This Week in JavaScript." This Week in JavaScript. Feel free to share this with a fellow developer, and make sure you're following for more weekly updates. Until next time, happy coding!|
|
May 6, 2025
Tim Pool Explains The Joy Of FATHERHOOD
YouTube
BUY CAST BREW COFFEE TO SUPPORT THE SHOW - https://castbrew.com/Become A Member And Protect Our Work at http://www.timcast.comHost:Tim Pool @Timcast (everywh... |
|
May 6, 2025
Antifa RIOTS Have BEGUN, Unhinged Leftists SET FIRES At UW Campus Take Over Building
YouTube
The Green Room - https://rumble.com/playlists/aa56qw_g-j0BUY CAST BREW COFFEE TO FIGHT BACK - https://castbrew.com/Join The Discord Server - https://timcast.... |
|
May 6, 2025
Trump CRUSHING Democrats, New Polls Show Democrats SINKING, GOP AHEAD Despite SMEARS | Timcast IRL
YouTube
Sign up and deposit for Underdog HERE with promo code TIM to get up to $1,000 in bonus credits and a free pick:https://play.underdogfantasy.com/p-tim-poolTex... |
|
May 6, 2025
Democrats Will IMPEACH Trump If They Win The Midterms
YouTube
SUPPORT THE SHOW BUY CAST BREW COFFEE NOW - https://castbrew.com/Sign Up For Exclusive Episodes At https://timcast.com/Merch - https://timcast.creator-spring... |
|
May 6, 2025
White Irish REVOLT Against INVASION As Nation Sees 300% SURGE In "Asylum" Claims
YouTube
The Green Room - https://rumble.com/playlists/aa56qw_g-j0BUY CAST BREW COFFEE TO FIGHT BACK - https://castbrew.com/Join The Discord Server - https://timcast.... |
|
May 6, 2025
Episode 2831 CWSA 05/06/25
YouTube
All the news that's fit to mock~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~If you would like to enjoy this same content plus bonus content from Scott Adams, inc... |
|
May 6, 2025
Liberal Explains REAL Masculinity
YouTube
The Green Room - https://rumble.com/playlists/aa56qw_g-j0BUY CAST BREW COFFEE TO FIGHT BACK - https://castbrew.com/Join The Discord Server - https://timcast.... |
| 
May 6, 2025
Still No Explanation for That Blackout in Spain and Portugal, Huh? | National Review |
|
May 6, 2025
President Trump Participates in an Easter Prayer Service and Dinner
YouTube
The White House |
|
May 6, 2025
President Trump Delivers a Sports Announcement
YouTube
The White House |
|
May 6, 2025
"Donald Trump is Crushing His To-Do List" đşđ¸
YouTube
Promises Made, Promises Kept. |
|
May 6, 2025
President Trump Gaggles with the Press on Air Force One, May 04, 2025
YouTube
Air Force One |
|
May 6, 2025
President Donald J. Trumpâs Commencement Address at the University of Alabama
YouTube
đşđ¸ President Donald J. Trump to the University of Alabama Class of 2025:âI've learned that perseverance is everything. So whatever happens, no matter where... |
|
May 6, 2025
"The Tide of Change is Sweeping the Country" âPresident Trump đşđ¸
YouTube
THIS VIDEO WILL GIVE YOU CHILLS đşđ¸"The tide of change is sweeping the country. Sunlight is pouring over the entire world, and America has the chance to sei... |
|
May 6, 2025
Investing in America đşđ¸
YouTube
"The companies represented in this room have collectively announced more than $2 trillion in new investments, and we have a total of close to $8 trillion. Ev... |
|
May 6, 2025
President Trump Participates in a Bilateral Meeting with the Prime Minister of Canada, May 6, 2025
YouTube
The White House |
| 
May 6, 2025
Vance, conservatives blast Omar over resurfaced 'fearful of White men' clip: 'Genocidal language'
Fox News
A resurfaced clip from Democratic Rep. Ilhan Omar sparked controversy on social media this week in a firestorm that included a scathing response from Vice President JD Vance.
A resurfaced clip of Dem. Rep. Ilhan Omar, a member of the progressive "Squad" in Congress, sparked a frenzy on social media this week with conservatives blasting the congresswoman over her comments regarding the "radicalization of White men." "I would say our country should be more fearful of White men across our country, because they are actually causing most of the deaths within this country," Omar said in a 2018 interview with Al-Jazeera while discussing the domestic terrorism threats in the United States and responding to a question on how much concern "jihadism" poses to the United States. "And so if fear was the driving force of policies to keep America safe, Americans safe inside of this country, we should be profiling, monitoring, and creating policies to fight the radicalization of White men." The clip, posted by conservative influencer accounts including Laura Loomer and LibsofTikTok with millions of impressions, sparked outrage from conservatives on social media, including from inside the White House. OMAR SLAMS TRUMP, MUSK FOR CHANGES AT USAID, ACCUSES PRESIDENT OF RUNNING DICTATORSHIP "This isnât just sick; itâs actually genocidal language," Vice President JD Vance posted on X. "What a disgrace this person is." "This is blatant racism," GOP Sen. Mike Lee posted on X. "Who condemns it?" ILHAN OMAR BLASTS HARRIS-WALZ CAMPAIGN FOR COURTING LIZ CHENEY: 'HUGE MISSTEP' "@ilhanMN never ceases to be an embarrassment for Minnesota," GOP Majority Whip Rep. Tom Emmer, who represents Minnesotaâs 6th Congressional District, posted on X. "Thereâs never been a more anti-American member of Congress than Ilhan Omar," conservative influencer Paul Szypula posted on X. Omar's office responded to a media inquiry by pointing Fox News Digital to her X post responding to Vance. "In this nearly 8yr old clip, I am referring to the rise of white nationalism in an annual report issued by the Anti-Defamation League that said white supremacists were responsible for 78 percent of "extremist-related murders," Omar said. "PS you should look up what "genocidal" actually means when youâre actively supporting a genocide taking place in Gaza." CLICK HERE TO GET THE FOX NEWS APP The social media firestorm comes shortly after Omar sparked controversy for telling Daily Caller News Foundation reporter Myles Morell to "f--- off" after he asked her a question about fellow Democratic Party figures traveling to El Salvador to defend illegal immigrant Kilmar Abrego Garcia, who was deported to the country by the Trump administration.Omar later responded to the clip being shared on X, stating, "I said what I said. You and all your miserable trolls can f--- off."|
| | New prompt method rewrites text in any style without changing its meaning
May 6, 2025
New prompt method rewrites text in any style without changing its meaning
THE DECODER
A new technique helps AI text generators mimic the style of a sample text without distorting its original meaning. The method is based on a well-known linguistic model. |
| 
May 6, 2025
Outsourcing Your Brain to ChatGPT | National Review
National Review
The stigma attached to using LLMs is valuable. But it likely masks the extent to which younger Americans have come to rely on AI in all sorts of ways. |
|
May 6, 2025
Trump Has Hollywoodâs Foreign Propaganda Problem Backwards | National Review
National Review
Our problem isnât too much buying from China â itâs too much selling to China. |
|
May 6, 2025
Trump's First 100 Days Shatter California's Left-Wing Illusions | National Review |
|
May 6, 2025
Trump CRUSHING Democrats, New Polls Show Democrats SINKING, GOP AHEAD Despite SMEARS | Timcast IRL
YouTube
Sign up and deposit for Underdog HERE with promo code TIM to get up to $1,000 in bonus credits and a free pick:https://play.underdogfantasy.com/p-tim-poolTex... |
|
May 6, 2025
Trumpâs Cartel Crackdown Offer, Possible Alcatraz Reopening, Antarctic Ice Increases: AM Update 5/6
YouTube
President Trump confirms he offered U.S. military support to take on the cartels, but Mexicoâs president rejected the help. The President shakes up the news ... |
| 
| This common mistake is killing your chance to stand out online - how to fix it | ZDNET
May 6, 2025
This common mistake is killing your chance to stand out online - how to fix it
ZDNET
The algorithm is a bigger threat to your creativity than AI. Here's what you need to do about it. |
| | Palmer on X: "@whyvert @byrne_a âIf the road to hell is paved with good intentions, where does the road paved with bad intentions go?â" / X
May 6, 2025 |
| | RTX 5080 Super 24GB and 5070 Super 18GB rumored once again â and they could be keenly priced because AMDâs RDNA 4 GPUs have âspookedâ Nvidia | TechRadar
May 6, 2025
RTX 5080 Super 24GB and 5070 Super 18GB rumored once again â and they could be keenly priced because AMDâs RDNA 4 GPUs have âspookedâ Nvidia
TechRadar
Nvidia RTX 5070 Super could be priced compellingly to return fire at AMD in the mid-range |
| 
| Cuttlefish Seem to 'Wave' at Each Other, but What They're Saying Is a Mystery
May 6, 2025
Cuttlefish Seem to 'Wave' at Each Other, but What They're Saying Is a Mystery
Gizmodo
A cute observation in the cephalopods' behavior indicates they also react to sound waves, a notion that will soon be tested with a machine learning approach. |
| 
May 6, 2025
Elon Musk expands on vision of Mars as âlife insuranceâ for humanity | Fox News
SpaceX CEO Elon Musk is prioritizing his vision for life on Mars as his time with the Department of Government Efficiency (DOGE) comes to an end. Musk sat down with Fox News host Jesse Watters for an interview where he shared the DOGE teamâs findings regarding government waste, fraud and abuse, and expanded on his idea of Mars colonization as "life insurance" for humanity. "Mars is life insurance for life collectively. So, eventually, all life on Earth will be destroyed by the sun. The sun is gradually expanding, and so we do at some point need to be a multi-planet civilization because Earth will be incinerated," Musk said in an interview that aired Monday on "Jesse Watters Primetime." ELON MUSK DOES NOT REGRET WORK AT DOGE, SUPPORT FOR TRUMP: âESSENTIALâ FOR AMERICA TO âREACH GREATER HEIGHTSâ The sun, a medium-sized star with a radius of about 435,000 miles, will eventually run out of energy and expand into a red giant star when it starts to die, according to a NASA fact sheet. The government space agency noted that the red giant star could become so large that it would engulf Mercury, Venus and possibly Earth as well, although the sun's lifespan, according to scientists, will last another 5 billion years. Musk explained that his vision for Mars is more than just landing on the planet and planting flags and laying footprints; itâs about creating a self-sustaining city. "The fundamental fork in the road of destiny [is] that Mars is sufficiently self-sustaining and can grow by itself if the resupply ships from Earth stop coming for any reason, whether that is because civilization died with a bang or a whimper," he told Watters. "If the resupply ships are necessary for Mars to survive, then we have not created life insurance." MARS WAS ONCE COVERED IN WIDE, RAGING RIVERS "We've not created life insurance for life collectively. So that's the key point in the future where [the] destiny of life, as we know it, will forever be affected, is when Mars becomes self-sustaining." Musk wrote in a post on his social media platform X in mid-March that a SpaceX Starship will depart for the red planet at the end of 2026, carrying Optimus, Teslaâs humanoid robot. CLICK HERE TO GET THE FOX NEWS APP He predicted that if the missions go well, humans could start landing on Mars as soon as 2029, although 2031 is more likely.|
|
May 6, 2025
Humiliated Merz fails to be elected German chancellor on first parliament ballot
POLITICO
This failure is unprecedented in Germanyâs postwar history. |
| 
May 6, 2025
Trump changes his tune on the economy
The Hill
President Trump is changing his tune on the economy, suggesting Americans should buy less and will probably pay more and bear the brunt of an uncertain economic landscape as his wide-ranging tariffâŚ
President Trump is changing his tune on the economy, suggesting Americans should buy less and will probably pay more and bear the brunt of an uncertain economic landscape as his wide-ranging tariff policy takes effect. Trump and his economic team have for weeks said the tariffs would result in only short-term pain and that the tumult in the stock market would eventually level out. But the White Houseâs messaging has evolved from Trump on the campaign trail promising to lower prices and make America âwealthyâ again to Trump suggesting the U.S. needs a cultural shift on consumer spending while accepting that his tariff plan will raise prices. Trump was asked Sunday by NBCâs Kristen Welker if he would acknowledge that his tariff plan will result in higher prices. At first, the president suggested tariffs will âmake us richâ â similar to sentiments heâs expressed when touting his economic policy. But in the next turn, he suggested that American children, for example, do not need as many toys and that Americans do not need to spend as much money on âjunk we donât need.â âIâm just saying they donât need to have 30 dolls. They can have three. They donât need to have 250 pencils. They can have five,â Trump said, acknowledging the prices of such items could also go up. Thatâs in stark contrast with candidate Trump, who spent much of 2024 railing against inflation under former President Biden and promising to lower costs if elected. In an ABC News interview last week, Trump said his economic policy is what voters signed up for. Trump has in recent weeks acknowledged âa little disturbanceâ in the economy that emerged when his tariff plan was rolled out. When he was campaigning, Trump spoke frequently of tariffs on China, the European Union, Canada and Mexico, but his policy ultimately imposed tariffs on nearly every country in the world, sending the U.S. and foreign stock and bond markets into chaos. Douglas Holtz-Eakin, president of the center-right American Action Forum, called Trumpâs messaging âpivotingâ on an unpopular policy. âThis feels tone-deaf to me. This is, 'Youâre too materialistic. You donât need as many dollars as you think.' And heâs a very strange messenger for that message, and I don't think it's going to sell,â Holtz-Eakin said. Marc Short, who was a top aide to former Vice President Mike Pence during Trumpâs first administration, warned that Trump risks alienating people if he keeps talking about dolls, calling it a âdamaging messageâ that âsuggests a little bit of an elitism perspective.â High tariffs on China and other key trading partners will have the most impact on Americans who rely on less-expensive goods, not those who can buy 30 toys, argued Daniel Hornung, National Economic Council deputy director in the Biden administration. âSaying low- and middle-income people should just buy less or buy more expensive stuff misses an important point,â Hornung said. âWe have large swaths of the country that donât make enough money to afford to buy expensive things, and itâs very important to them whether or not something costs 5 percent or 10 percent or 20 percent or 100 percent more.â Trump imposed the 90-day pause on âreciprocalâ tariffs â those that went above the benchmark 10 percent rate imposed on all countries â amid increasing pressure from Wall Street and fellow Republicans. A 10 percent tariff remains on all countries, as does a massive 145 percent tariff on China, the worldâs second largest economy. Markets are still facing turbulence as the future of tariffs, and trade relationships remains unclear, despite the White House insisting some deals are close to fruition. Using tariffs as a negotiation tool while Trump is also asking Americans to get used to buying less are opposing ideas, argued Kathryn Anne Edwards, a labor economist and policy consultant. âTheyâre in complete conflict with each other, because if itâs just a negotiating ploy, you don't try to bring domestic production home at all. Youâre just trying to get a better price for your consumers here,â she said. âIf itâs actually about domestic production, negotiation's off the table because I don't care what you offer me, this is about jobs at home.â Also contributing to economic anxieties are some projections from Wall Street that see a potential recession on the horizon. When Welker asked Trump whether he was OK with the prospect of a recession, at least in the short term, he replied: âLook, yeah. Everythingâs OK. What we are â I said this is a transition period.â Trump allies in Congress, meanwhile, are backing the president. Sen. Ron Johnson (R-Wis.) dismissed the possibility of a recession Sunday, adding that âyou have to act boldlyâ when it comes to the tariff agenda. Edwards said the possibility of the U.S. heading toward a recession could actually prevent companies from opening manufacturing plants in the U.S., going against one of Trumpâs own intentions. âWhat would prevent a business right now from saying, âHey, if there's a tariff, I'm going to start manufacturing something at home'? Well, they can't do it because if thereâs a recession, itâs not an easy time to start a high-scale manufacturing business, especially if orders are down, stores are closing and consumption is down,â she said. The policies are also losing some support from the public. Almost 6 in 10 U.S. adults said Trumpâs policies are making the economy worse in a CNN poll published last week, and a recent Gallup poll found that 89 percent of U.S. adult respondents think the tariffs will increase prices. One indicator of whether Trumpâs message on the economy is a winning one will be how Republican lawmakers handle it in their 2026 reelection bids, Holtz-Eakin said. âI donât know that he ever loses his base, but the fundamental question is, when does he lose the Republicans in Congress who need to run for reelection?â Holtz-Eakin said. âIf the president gets unpopular enough ⌠you start trying to distance yourself, and when you start to see that, you know, Trump's lost.â|
| 
May 6, 2025
Trump critics say "MAGA Maoism" is undercutting American capitalism
Axios
Trump is breaking sharply from free-market orthodoxy in his second term. |
|
May 5, 2025
Trump on Upholding Constitution: "I Don't Know" | The Daily Show
YouTube
Jon Stewart sorts Trump's latest barrage of bulls**t into the "OK" and "NOT OK" piles, including telling kids to cut down on dolls and pencils, calling for A... |
|
May 5, 2025
Rutger Bregman - âMoral Ambitionâ | The Daily Show
YouTube
Historian and best-selling author Rutger Bregman joins Jon Stewart to unpack his latest book, âMoral Ambition,â which is a call to action for people, especia... |
| 
| The Opposite of DĂŠjĂ Vu Can Happen, And It's Even More Uncanny : ScienceAlert
May 5, 2025
The Opposite of DĂŠjĂ Vu Can Happen, And It's Even More Uncanny
ScienceAlert
Repetition has a strange relationship with the mind. |
|
May 5, 2025
Full interview: President Trump says heâll be a âtwo-term president,â downplays third-term talk
YouTube
President Donald Trump joins Meet the Press for an exclusive and wide-ranging interview reflecting on the first 100 days of his second term and laying out hi... |
| | Trumpâs Three Steps to Economic Growth - WSJ
May 5, 2025 |
| 
May 5, 2025
Nvidia launches fully open source transcription AI model Parakeet-TDT-0.6B-V2 on Hugging Face
VentureBeat
An attractive proposition for commercial enterprises and indie developers looking to build speech recognition and transcription services... |
| | gorklon rust on X: "This is great âĽď¸" / X
May 5, 2025 |
| 
May 5, 2025
Videos | Donald J. Trump For President 2024
Certified Website of Donald J. Trump For President 2024. Watch a special message from President Trump. |
| 
May 5, 2025
Niall Ferguson: Donald Trump Is Crushing His To-Do List
But is he crushing the economy in the process? |
| | Where the Trump administration has science on its side
May 5, 2025
**Where the Trump administration has science on its side **
The Economist
A government report evaluates gender therapy and medical care for children |
| 
| How a Problem About Pigeons Powers Complexity Theory | Quanta Magazine
May 5, 2025
How a Problem About Pigeons Powers Complexity Theory | Quanta Magazine
Quanta Magazine
When pigeons outnumber pigeonholes, some birds must double up. This obvious statement â and its inverse â have deep connections to many areas of math and computer science. |
| | Enormous desert structure built by insects can be seen from space - Earth.com
May 5, 2025
Enormous desert structure built by insects is 4,000 years old and can be seen from space
Earth.com
Every termite mound holds roughly 65âŻcubicâŻyards of packed clay. The total displaced soil rivals about 4,000âŻGreatâŻPyramidsâŻofâŻGiza. |
| | This Windows 11-like Linux distribution is aimed squarely at developers | ZDNET
May 5, 2025
This Windows 11-like Linux distribution is aimed squarely at developers
ZDNET
Curious about Linux or ready for a superior dev platform? Try AnduinOS. |
| | Natalie_Magnificent on X: "@DailyNousEditor A brave person." / X
May 5, 2025 |
|
May 5, 2025
Hedonism and SĂśren Kierkegaard
YouTube
In this video, I discuss hedonism, and how it is often viewed as a kind of vulgar sensualism. I then introduce Soren Kierkegaard's Two Stages of Human Existe... |
| | Rothmus đ´ on X: "đŻđŻđŻđŻ https://t.co/r0zaK9EAWp" / X
May 5, 2025 |
| | Daily Nous / Justin Weinberg on X: "Is a philosophy professor helping the Trump administration with its anti-trans crusade? https://t.co/uvDNwpELoE" / X
May 5, 2025 |
| 
| A Scientific Discovery Could Feed 136 Billion People â A Breakthrough Like the Invention of Fertilizers - Jason Deegan
May 5, 2025
A Scientific Discovery Could Feed 136 Billion People â A Breakthrough Like the Invention of Fertilizers
Jason Deegan
At the end of the 18th century, economist Thomas Malthus famously predicted that unchecked population growth would one day outstrip food production. It sparked a school ... Continue Reading â |
| 
| Sunrise on the Reaping: How Hunger Games prequel was inspired by Scottish philosopher
May 5, 2025
**Sunrise on the Reaping: How Hunger Games prequel was inspired by Scottish philosopher **
The Scotsman
Scotland is everywhere â as long as you look hard enough.
Set more than twenty years before the original novel, author Suzanne Collinsâ latest Hunger Games prequel was inspired by the works of an 18th-century Scottish philosopher. Released in March, Sunrise on the Reaping is the fifth novel in Collinsâ blockbuster series, which revisits the dystopian world of Panem 24 years before the events of The Hunger Games. It is the American authorâs first book since the publication of The Ballad of Songbirds and Snakes in 2020. For previous books in the series, Collins had drawn inspiration from Greek mythology â particularly the myth of Theseus and the Minotaur â as well as Roman gladiator games. But for her latest book, Collins instead turned to Scottish Enlightenment philosopher David Hume. When the book was first announced, Collins said: âWith Sunrise on the Reaping, I was inspired by David Humeâs idea of implicit submission and, in his words, âthe easiness with which the many are governed by the few.â âThe story also lent itself to a deeper dive into the use of propaganda and the power of those who control the narrative. The question âReal or not real?â seems more pressing to me every day.â With the cast for the upcoming film version having just been revealed, here is how Suzanne Collins was inspired by Scottish philosopher David Hume while writing Sunrise on the Reaping. David Hume was a Scottish philosopher, generally regarded as one of the most important to write in English. As well as being a historian and essayist, Hume is responsible for a number of major philosophical works which remain influential to this day. A leading figure in the 18-century Enlightenment period, Hume was born and raised in Edinburgh, developing an early love for reading and writing as well as ancient and modern philosophy. Collins was first introduced to the works of Hume by her father, a historian and doctor of political science. Describing her childhood, she said her father approached âhistory as a storyâ which saw her introduced to stories of war and philosophical dilemmas from a young age. In an interview with Scholasticâs Editorial Director David Levithan, she said: âMy dad introduced me to David Hume when I was a child, along with many other philosophers. He talked about them while using more kid-friendly examples. Like, in Hume's case, sunrises and billiard balls. It was a little mind-bending but always interesting.â Themes of truth, lies, and propaganda have been at the heart of the Hunger Games series, and Sunrise on the Reaping is no different. It is somewhat more unique, however, in that Collins seemed particularly determined to alert her primarily young adult audience to political messaging which looks to shape individual opinions toward a collective which fits the status quo. From her earliest words on Sunrise, Collins made sure to highlight Humeâs idea of âimplicit submissionâ as her key inspiration, which asks the question of âwhy is it so easy for the few in power to govern the many?â. Reflecting this, the book opens with a series of quotes; the first two from George Orwell and William Blake â both about propaganda, truth, and lies â and the final two from David Hume. The fist Hume quote is as follows: âNothing appears more surprising to those who consider human affairs with a philosophical eye than the easiness with which the many are governed by the few, and the implicit submission with which men resign their own sentiments and passions to those of their rulers. When we enquire by what means this wonder is effected, we shall find, that, as Force is always on the side of the governed, the governors have nothing to support them but opinion. It is therefore, on opinion only that government is founded; and this maxim extends to the most despotic and most military governments, as well as to the most free and most popular.â From Humeâs Essays Moral, Political, and Literary, Collins included the passage as âinvites so many questionsâ. She has even gone so far as to say that âIf all people do is read this Hume quote and discuss it, this book has been a win for meâ. The passage sees the Enlightenment philosopher conclude that only the common âopinionâ of the many regarding the legitimacy of the few in power, keeps them in charge. In Sunrise on the Reaping, Collins demonstrates Humeâs idea of implicit submission on three levels: âindividual, Hunger Games, and nationalâ. It is the foundation of the book, which asks what form propaganda must take for people to become convinced implicit submission is the preferred option, if it is inevitable, or if there is a way to be protected against it. The second quote examines a different aspect of Humeâs philosophy, the answer to the question âis the sun guaranteed to rise tomorrow morning?â One of philosophyâs greatest debates, Humeâs answer to this question would be no. Collins quotes: âThat the sun will not rise tomorrow is no less intelligible a proposition, and implies no more contradiction, than the affirmation, that it will rise.â The debate between deductive and inductive reasoning is a core theme of Sunrise on the Reaping, with conflict between protagonist Haymitch Abernathy and Lenore Dove stemming from their differing beliefs on this matter. One believes that just because something happened in the past, it will continue to happen in the future â in this case, the Hunger Games reaping. Meanwhile, the other believes that the reaping is not a certainty, and that acknowledging it isnât inevitable can lead to a different future.|
| 
| Philosopherâs Apparent Role in Governmentâs âTreatment for Pediatric Gender Dysphoriaâ Report Revealed by Metadata - Daily Nous
May 5, 2025
Philosopherâs Apparent Role in Governmentâs âTreatment for Pediatric Gender Dysphoriaâ Report Revealed by Metadata (updated) - Daily Nous
Daily Nous
A philosopher is apparently listed as the creator of part of a controversial report released May 1st by the United States Department of Health and Human Services (HHS) regarding gender-affirming medical care for transgender youth. Philosophers working for the government is a topic regularly covered here at Daily Nous. Thatâs one reason for this post, but not the only one. The Department of Health and Human Services is led by Robert F. Kennedy, Jr., known for his advocacy of positions widely rejected by the medical profession, scientists, and other experts. Kennedy was appointed by President Donald Trump, who, throughout his second term so far, has been attacking universities and cutting off the federal funding that supports much research in US institutions of higher education. The anti-expertise and anti-education agenda of the Trump administration makes it surprising that any academic would agree to work for it. Nonetheless, there seems to be some evidence that one academicâphilosophy professor Alex Byrne of the Massachusetts Institute of Technology, whose writing on transgender issues has been controversialâhas done just that. That evidence is in the metadata of the appendix of an HHS report, âTreatment for Pediatric Gender Dysphoriaâ. The appendix itself is titled âOverview of Systematic Reviewsâ. The following is a screenshot I took of part of the metadata of the appendix, as viewable in a plain text-reader app like Notepad, which lists âAlex Byrneâ as the creator of the appendix: The text âdc:creatorâ is a tag that refers to âan entity primarily responsible for making the resourceâ. (The âdcâ part is a reference to âDublin Coreâ, a standard for metadata design.) To my knowledge, the listing of Byrneâs name in the metadata was first pointed out by Florence Ashley, an assistant professor of law at the University of Alberta, in a post on Bluesky the day the report was released. Whether the metadata is accurate and being interpreted correctly I canât say for certain at this time. The text of the report and appendix do not name any authors. According to a press release about the report from HHS, the authors âare not initially being made public, in order to help maintain the integrityâ of the reportâs âpost-publication peer reviewâ. This seems disingenuous, as it is.. |
| | As an Experienced LLM User, I Actually Don't Use Generative LLMs Often | Max Woolf's Blog
May 5, 2025
As an Experienced LLM User, I Actually Don't Use Generative LLMs Often
But for what I do use LLMs for, itâs invaluable.
Lately, Iâve been working on codifying a personal ethics statement about my stances on generative AI as I have been very critical about several aspects of modern GenAI, and yet I participate in it. While working on that statement, Iâve been introspecting on how I myself have been utilizing large language models for both my professional work as a Senior Data Scientist at BuzzFeed and for my personal work blogging and writing open-source software. For about a decade, Iâve been researching and developing tooling around text generation from char-rnns, to the ability to fine-tune GPT-2, to experiments with GPT-3, and even more experiments with ChatGPT and other LLM APIs. Although I donât claim to the best user of modern LLMs out there, Iâve had plenty of experience working against the cons of next-token predictor models and have become very good at finding the pros. It turns out, to my surprise, that I donât use them nearly as often as people think engineers do, but that doesnât mean LLMs are useless for me. Itâs a discussion that requires case-by-case nuance. How I Interface With LLMs Over the years Iâve utilized all the tricks to get the best results out of LLMs. The most famous trick is prompt engineering, or the art of phrasing the prompt in a specific manner to coach the model to generate a specific constrained output. Additions to prompts such as offering financial incentives to the LLM or simply telling the LLM to make their output better do indeed have a quantifiable positive impact on both improving adherence to the original prompt and the output text quality. Whenever my coworkers ask me why their LLM output is not what they expected, I suggest that they apply more prompt engineering and it almost always fixes their issues. No one in the AI field is happy about prompt engineering, especially myself. Attempts to remove the need for prompt engineering with more robust RLHF paradigms have only made it even more rewarding by allowing LLM developers to make use of better prompt adherence. True, âPrompt Engineerâ as a job title turned out to be a meme but thatâs mostly because prompt engineering is now an expected skill for anyone seriously using LLMs. Prompt engineering works, and part of being a professional is using what works even if itâs silly. To that end, I never use ChatGPT.com or other normal-person frontends for accessing LLMs because they are harder to control. Instead, I typically access the backend UIs provided by each LLM service, which serve as a light wrapper over the API functionality which also makes it easy to port to code if necessary. Accessing LLM APIs like the ChatGPT API directly allow you to set system prompts which control the ârulesâ for the generation that can be very nuanced. Specifying specific constraints for the generated text such as âkeep it to no more than 30 wordsâ or ânever use the word âdelveââ tends to be more effective in the system prompt than putting them in the user prompt as you would with ChatGPT.com. Any modern LLM interface that does not let you explicitly set a system prompt is most likely using their own system prompt which you canât control: for example, when ChatGPT.com had an issue where it was too sycophantic to its users, OpenAI changed the system prompt to command ChatGPT to âavoid ungrounded or sycophantic flattery.â I tend to use Anthropic Claudeâs API â Claude Sonnet in particular â more than any ChatGPT variant because Claude anecdotally is less âroboticâ and also handles coding questions much more accurately. Additionally with the APIs, you can control the âtemperatureâ of the generation, which at a high level controls the creativity of the generation. LLMs by default do not select the next token with the highest probability in order to allow it to give different outputs for each generation, so I prefer to set the temperature to 0.0 so that the output is mostly deterministic, or 0.2 - 0.3 if some light variance is required. Modern LLMs now use a default temperature of 1.0, and I theorize that higher value is accentuating LLM hallucination issues where the text outputs are internally consistent but factually wrong. LLMs for Professional Problem Solving! With that pretext, I can now talk about how I have used generative LLMs over the past couple years at BuzzFeed. Here are outlines of some (out of many) projects Iâve worked on using LLMs to successfully solve problems quickly: BuzzFeed site curators developed a new hierarchal taxonomy to organize thousands of articles into a specified category and subcategory. Since we had no existing labeled articles to train a traditional multiclass classification model to predict these new labels, I wrote a script to hit the Claude Sonnet API with a system prompt saying The following is a taxonomy: return the category and subcategory that best matches the article the user provides. plus the JSON-formatted hierarchical taxonomy, then I provided the article metadata as the user prompt, all with a temperature of 0.0 for the most precise results. Running this in a loop for all the articles resulted in appropriate labels. After identifying hundreds of distinct semantic clusters of BuzzFeed articles using data science shenanigans, it became clear that there wasnât an easy way to give each one unique labels. I wrote another script to hit the Claude Sonnet API with a system prompt saying Return a JSON-formatted title and description that applies to all the articles the user provides. with the user prompt containing five articles from that cluster: again, running the script in a loop for all clusters provided excellent results. One BuzzFeed writer asked if there was a way to use a LLM to sanity-check grammar questions such as âshould I use an em dash here?â against the BuzzFeed style guide. Once again I hit the Claude Sonnet API, this time copy/pasting the full style guide in the system prompt plus a command to Reference the provided style guide to answer the user's question, and cite the exact rules used to answer the question. In testing, the citations were accurate and present in the source input, and the reasonings were consistent. Each of these projects were off-hand ideas pitched in a morning standup or a Slack DM, and yet each project only took an hour or two to complete a proof of concept (including testing) and hand off to the relevant stakeholders for evaluation. For projects such as the hierarchal labeling, without LLMs I would have needed to do more sophisticated R&D and likely would have taken days including building training datasets through manual labeling, which is not intellectually gratifying. Here, LLMs did indeed follow the Pareto principle and got me 80% of the way to a working solution, but the remaining 20% of the work iterating, testing and gathering feedback took longer. Even after the model outputs became more reliable, LLM hallucination was still a concern which is why I also advocate to my coworkers to use caution and double-check with a human if the LLM output is peculiar. Thereâs also one use case of LLMs that doesnât involve text generation thatâs as useful in my professional work: text embeddings. Modern text embedding models technically are LLMs, except instead of having a head which outputs the logits for the next token, it outputs a vector of numbers that uniquely identify the input text in a higher-dimensional space. All improvements to LLMs that the ChatGPT revolution inspired, such as longer context windows and better quality training regimens, also apply to these text embedding models and caused them to improve drastically over time with models such as nomic-embed-text and gte-modernbert-base. Text embeddings have done a lot at BuzzFeed from identifying similar articles to building recommendation models, but this blog post is about generative LLMs so Iâll save those use cases for another time. LLMs for Writing? No, I donât use LLMs for writing the text on this very blog, which I suspect has now become a default assumption for people reading an article written by an experienced LLM user. My blog is far too weird for an LLM to properly emulate. My writing style is blunt, irreverent, and occasionally cringe: even with prompt engineering plus few-shot prompting by giving it examples of my existing blog posts and telling the model to follow the same literary style precisely, LLMs output something closer to Marvel movie dialogue. But even if LLMs could write articles in my voice I still wouldnât use them due of the ethics of misrepresenting authorship by having the majority of the work not be my own words. Additionally, I tend to write about very recent events in the tech/coding world that would not be strongly represented in the training data of a LLM if at all, which increases the likelihood of hallucination. There is one silly technique I discovered to allow a LLM to improve my writing without having it do my writing: feed it the text of my mostly-complete blog post, and ask the LLM to pretend to be a cynical Hacker News commenter and write five distinct comments based on the blog post. This not only identifies weaker arguments for potential criticism, but it also doesnât tell me what I should write in the post to preemptively address that negative feedback so I have to solve it organically. When running a rough draft of this very blog post and the Hacker News system prompt through the Claude API (chat log), it noted that my examples of LLM use at BuzzFeed are too simple and not anything more innovative than traditional natural language processing techniques, so I made edits elaborating how NLP would not be as efficient or effective. LLMs for Companionship? No, I donât use LLMs as friendly chatbots either. The runaway success of LLM personal companion startups such as character.ai and Replika are alone enough evidence that LLMs have a use, even if the use is just entertainment/therapy and not more utilitarian. I admit that I am an outlier since treating LLMs as a friend is the most common use case. Myself being an introvert aside, itâs hard to be friends with an entity who is trained to be as friendly as possible but also habitually lies due to hallucination. I could prompt engineer an LLM to call me out on my bullshit instead of just giving me positive affirmations, but thereâs no fix for the lying. LLMs for Coding??? Yes, I use LLMs for coding, but only when I am reasonably confident that theyâll increase my productivity. Ever since the dawn of the original ChatGPT, Iâve asked LLMs to help me write regular expressions since that alone saves me hours, embarrassing to admit. However, the role of LLMs in coding has expanded far beyond that nowadays, and coding is even more nuanced and more controversial on how you can best utilize LLM assistance. Like most coders, I Googled coding questions and clicked on the first Stack Overflow result that seemed relevant, until I decided to start asking Claude Sonnet the same coding questions and getting much more detailed and bespoke results. This was more pronounced for questions which required specific functional constraints and software frameworks, the combinations of which would likely not be present in a Stack Overflow answer. One paraphrased example I recently asked Claude Sonnet while writing another blog post is Write Python code using the Pillow library to composite five images into a single image: the left half consists of one image, the right half consists of the remaining four images. (chat log). Compositing multiple images with Pillow isnât too difficult and thereâs enough questions/solutions about it on Stack Overflow, but the specific way itâs composited is unique and requires some positioning shenanigans that I would likely mess up on the first try. But Claude Sonnetâs code got it mostly correct and it was easy to test, which saved me time doing unfun debugging. However, for more complex code questions particularly around less popular libraries which have fewer code examples scraped from Stack Overflow and GitHub, I am more cautious of the LLMâs outputs. One real-world issue Iâve had is that I need a way to log detailed metrics to a database while training models â for which I use the Trainer class in Hugging Face transformers â so that I can visualize and analyze it later. I asked Claude Sonnet to Write a Callback class in Python for the Trainer class in the Hugging Face transformers Python library such that it logs model training metadata for each step to a local SQLite database, such as current epoch, time for step, step loss, etc. (chat log). This one I was less optimistic about since there isnât much code about creating custom callbacks, however the Claude-generated code implemented some helpful ideas that werenât on the top-of-my-mind when I asked, such a buffer to limit blocking I/O, SQLite config speedups, batch inserts, and connection handling. Asking Claude to âmake the code betterâ twice (why not?) results in a few more unexpected ideas such as SQLite connection caching and using a single column with the JSON column type to store an arbitrary number of metrics, in addition to making the code much more Pythonic. It is still a lot of code such that itâs unlikely to work out-of-the-box without testing in the full context of an actual training loop. However, even if the code has flaws, the ideas themselves are extremely useful and in this case it would be much faster and likely higher quality code overall to hack on this generated code instead of writing my own SQLite logger from scratch. For actual data science in my day-to-day work that I spend most of my time, Iâve found that code generation from LLMs is less useful. LLMs cannot output the text result of mathematical operations reliably, with some APIs working around that by allowing for a code interpreter to perform data ETL and analysis, but given the scale of data I typically work with itâs not cost-feasible to do that type of workflow. Although pandas is the standard for manipulating tabular data in Python and has been around since 2008, Iâve been using the relatively new polars library exclusively, and Iâve noticed that LLMs tend to hallucinate polars functions as if they were pandas functions which requires documentation deep dives to confirm which became annoying. For data visualization, which I donât use Python at all and instead use R and ggplot2, I really havenât had a temptation to consult a LLM, in addition to my skepticism that LLMs would know both those frameworks as well. The techniques I use for data visualization have been unchanged since 2017, and the most time-consuming issue I have when making a chart is determining whether the data points are too big or too small for humans to read easily, which is not something a LLM can help with. Asking LLMs coding questions is only one aspect of coding assistance. One of the other major ones is using a coding assistant with in-line code suggestions such as GitHub Copilot. Despite my success in using LLMs for one-off coding questions, I actually dislike using coding assistants for an unexpected reason: itâs distracting. Whenever I see a code suggestion from Copilot pop up, I have to mentally context switch from writing code to reviewing code and then back again, which destroys my focus. Overall, it was a net neutral productivity gain but a net negative cost as Copilots are much more expensive than just asking a LLM ad hoc questions through a web UI. Now we can talk about the elephants in the room â agents, MCP, and vibe coding â and my takes are spicy. Agents and MCP, at a high-level, are a rebranding of the Tools paradigm popularized by the ReAct paper in 2022 where LLMs can decide whether a tool is necessary to answer the user input, extract relevant metadata to pass to the tool to run, then return the results. The rapid LLM advancements in context window size and prompt adherence since then have made Agent workflows more reliable, and the standardization of MCP is an objective improvement over normal Tools that I encourage. However, they donât open any new use cases that werenât already available when LangChain first hit the scene a couple years ago, and now simple implementations of MCP workflows are even more complicated and confusing than it was back then. I personally have not been able to find any novel use case for Agents, not then and not now. Vibe coding with coding agents like Claude Code or Cursor is something I have little desire to even experiment with. On paper, coding agents should be able to address my complaints with LLM-generated code reliability since it inherently double-checks itself and itâs able to incorporate the context of an entire code project. However, I have also heard the horror stories of people spending hundreds of dollars by accident and not get anything that solves their coding problems. Thereâs a fine line between experimenting with code generation and gambling with code generation. Vibe coding can get me 80% of the way there, and I agree thereâs value in that for building quick personal apps that either arenât ever released publicly, or are released with disclaimers about its âthis is released as-isâ nature. But itâs unprofessional to use vibe coding as a defense to ship knowingly substandard code for serious projects, and the only code I can stand by is the code I am fully confident in its implementation. Of course, the coding landscape is always changing, and everything Iâve said above is how I use LLMs for now. Itâs entirely possible I see a post on Hacker News that completely changes my views on vibe coding or other AI coding workflows, but Iâm happy with my coding productivity as it is currently and I am able to complete all my coding tasks quickly and correctly. Whatâs Next for LLM Users? Discourse about LLMs and their role in society has become bifuricated enough such that making the extremely neutral statement that LLMs have some uses is enough to justify a barrage of harrassment. I strongly disagree with AI critic Ed Zitron about his assertions that the reason the LLM industry is doomed because OpenAI and other LLM providers canât earn enough revenue to offset their massive costs as LLMs have no real-world use. Two things can be true simultaneously: (a) LLM provider cost economics are too negative to return positive ROI to investors, and (b) LLMs are useful for solving problems that are meaningful and high impact, albeit not to the AGI hype that would justify point (a). This particular combination creates a frustrating gray area that requires a nuance that an ideologically split social media can no longer support gracefully. Hypothetically, If OpenAI and every other LLM provider suddenly collapsed and no better LLM models would ever be trained and released, open-source and permissively licensed models such as Qwen3 and DeepSeek R1 that perform comparable to ChatGPT are valid substitute goods and they can be hosted on dedicated LLM hosting providers like Cerebras and Groq who can actually make money on each user inference query. OpenAI collapsing would not cause the end of LLMs, because LLMs are useful today and there will always be a nonzero market demand for them: itâs a bell that canât be unrung. As a software engineer â and especially as a data scientist â one thing Iâve learnt over the years is that itâs always best to use the right tool when appropriate, and LLMs are just another tool in that toolbox. LLMs can be both productive and counterproductive depending on where and when you use them, but they are most definitely not useless. LLMs are more akin to forcing a square peg into a round hole (at the risk of damaging either the peg or hole in the process) while doing things without LLM assistance is the equivalent of carefully defining a round peg to pass through the round hole without incident. But for some round holes, sometimes shoving the square peg through and asking questions later makes sense when you need to iterate quickly, while sometimes you have to be more precise with both the peg and the hole to ensure neither becomes damaged, because then you have to spend extra time and money fixing the peg and/or hole. âŚmaybe itâs okay if I ask an LLM to help me write my metaphors going forward.|
| 
May 5, 2025
Dave Portnoy blasts Philly station, âscumbagâ reporter over interview about antisemitic âfâk the Jewsâ sign at Barstool bar
New York Post
Dave Portnoy said a local ABC news affiliate tried to âbait and switchâ him during an interview about the incident involving an antisemitic sign at his Barstool Sansom Street bar in Phi⌠|
|
May 5, 2025
Israel plans to occupy and flatten all of Gaza if no deal by Trump's trip
Axios
Trump has effectively given Netanyahu a green light to do as he sees fit, Israeli officials say. |
|
May 5, 2025
Woke Leftism BACKFIRED, White Guilt IS OVER, Shiloh Hendrix Raises $660k
YouTube
BUY BOONIES BOARD - https://shop.boonieshq.com/products/28th-amendmentBUY CAST BREW COFFEE TO SUPPORT THE SHOW - https://castbrew.com/Become A Member And Pro... |
|
May 5, 2025
Why are young men voting Conservative? A lot of it is a desire for change and those feelings are
YouTube
Why are young men voting Conservative? A lot of it is a desire for change and those feelings are valid. #election2025 #canada #cdnpoli #pierrepoilievre #cons... |
