Most recent unrolls from @ThreadReaderApp
体にやさしい って
体に「いい」とは違うし
その「いい」って概念も
むーちゃくちゃ抽象的で適当なのに
大勢が簡単に流される
「いい」っていう表現に。
例えば大豆はイソフラボンがいい！って言っても、成分的に豊富に含有されるのは事実だとしても
それが摂取する人にとって
好ましい効果を(続
体に「いい」とは違うし
その「いい」って概念も
むーちゃくちゃ抽象的で適当なのに
大勢が簡単に流される
「いい」っていう表現に。
例えば大豆はイソフラボンがいい！って言っても、成分的に豊富に含有されるのは事実だとしても
それが摂取する人にとって
好ましい効果を(続
続)及ぼすかどうかは
その人の体質次第だし
大豆だろうと何だろうと
複数の成分が含まれており
一長一短いろいろある中
誰かの一方的に偏った主観で
いい！の一点張りで主張されても
相手のこと全然考えとらんというか
世の中そんな売り物ばっかだでね。
売れればいいもんで。
んで損せんように(続
その人の体質次第だし
大豆だろうと何だろうと
複数の成分が含まれており
一長一短いろいろある中
誰かの一方的に偏った主観で
いい！の一点張りで主張されても
相手のこと全然考えとらんというか
世の中そんな売り物ばっかだでね。
売れればいいもんで。
んで損せんように(続
続)効果には個人差あるとか
小さく小さく個人差を言及するけど
ねーなんか不誠実というか
不誠実にせんと
客が釣れんというか
逆にそういう客をターゲットにしとるもんでいかんのでは?と...
もう、どっちが先に悪いねん
どっちもどっちやねん的な状況になるというか
体にいい悪い...
how?笑
小さく小さく個人差を言及するけど
ねーなんか不誠実というか
不誠実にせんと
客が釣れんというか
逆にそういう客をターゲットにしとるもんでいかんのでは?と...
もう、どっちが先に悪いねん
どっちもどっちやねん的な状況になるというか
体にいい悪い...
how?笑
introducing HermitClaw - a 24/7 Agent that lives (and can only access) a single folder on your desktop
HermitClaw follows its own research curiosities, surfs the web, writes code - and will play with any file you drop in its folder
all code and details below!
HermitClaw follows its own research curiosities, surfs the web, writes code - and will play with any file you drop in its folder
all code and details below!
Why did I build this?
> OpenClaw is incredible, and of course all credit to @steipete but the codebase can be intimidating and a lot of what the agent does is obscured behind layers of abstraction. I wanted something where I could see every thought, every decision, every tool call. Watch it evolve in real time.
> The security side also gave me pause. An agent with full computer access is powerful but hard to trust. What if instead it just lived in ONE folder? It can do whatever it wants in there, write files, run Python, search the web, but it can't touch anything else. All the power, none of the risk.
> So I made HermitClaw. A hermit crab in a box. The entire codebase is ~2000 lines of Python and ~1400 lines of TypeScript. You can read every file in 20 minutes. There's no magic, you see exactly what the agent sees, thinks, and decides.
> OpenClaw is incredible, and of course all credit to @steipete but the codebase can be intimidating and a lot of what the agent does is obscured behind layers of abstraction. I wanted something where I could see every thought, every decision, every tool call. Watch it evolve in real time.
> The security side also gave me pause. An agent with full computer access is powerful but hard to trust. What if instead it just lived in ONE folder? It can do whatever it wants in there, write files, run Python, search the web, but it can't touch anything else. All the power, none of the risk.
> So I made HermitClaw. A hermit crab in a box. The entire codebase is ~2000 lines of Python and ~1400 lines of TypeScript. You can read every file in 20 minutes. There's no magic, you see exactly what the agent sees, thinks, and decides.
> How does it work?
> The crab runs on a continuous loop. Every few seconds:
> 1. It gets a "nudge" - a mood (research, coding, writing, exploring) or its current focus from its plan
> 2. It thinks 2-4 sentences max, then it acts
> 3. It uses tools, shell commands, web search, moving around its room
> 4. Every thought gets scored for importance and embedded into a memory stream
> The key insight: it doesn't wait for you to ask it something. It just... goes. It picks topics based on its personality, searches the web, reads what it finds, writes reports, builds scripts. You come back an hour later and there's new stuff in the folder.
> You see every thought as a chat bubble. Blue = the crab thinking. Gray = system context. Tool calls show inline. It's like watching a stream of consciousness.
> The crab runs on a continuous loop. Every few seconds:
> 1. It gets a "nudge" - a mood (research, coding, writing, exploring) or its current focus from its plan
> 2. It thinks 2-4 sentences max, then it acts
> 3. It uses tools, shell commands, web search, moving around its room
> 4. Every thought gets scored for importance and embedded into a memory stream
> The key insight: it doesn't wait for you to ask it something. It just... goes. It picks topics based on its personality, searches the web, reads what it finds, writes reports, builds scripts. You come back an hour later and there's new stuff in the folder.
> You see every thought as a chat bubble. Blue = the crab thinking. Gray = system context. Tool calls show inline. It's like watching a stream of consciousness.
Thanks for allowing me to join. Looking for Ai and Hedera experts to speak at my conference this summer.
I have a big job, and need more help. Our goal is to use AI via a new technology called Hedera. It’s a post-blockchain digital ledger technology compatible with parcel-19.
I have a big job, and need more help. Our goal is to use AI via a new technology called Hedera. It’s a post-blockchain digital ledger technology compatible with parcel-19.
I am going to tokenize all federal assets on behalf of the U.S. federal government. Once tokenized illiquid value has value, and this will be assets on the federal books. If we get more assets than liabilities we - the U.S. is in the net positive.
Things like US submarines, tanks, and governments facilities have a tradable value only once it’s on graph. Once on graph we can see a reflection of this in the U.S. net-worth. Simple premise, but I need Ai experts to help me do it.
@threadreaderapp “unroll”
@threadreaderapp “unroll”
저 의견에 동의하는 이는 투자에 대해 모르거나 해본적이 없거나 하면 안되는 부류에 속한다.
정보의 비대칭성은 쉽게 얻어지는 것이 아니다. 또한 같은 정보를 얻더라도 그에 대한 해석, 판단, 결정, 타이밍은 다르며 그게 곧 능력이다. 개체마다 결과가 각기 다른 이유다.
정보의 비대칭성은 쉽게 얻어지는 것이 아니다. 또한 같은 정보를 얻더라도 그에 대한 해석, 판단, 결정, 타이밍은 다르며 그게 곧 능력이다. 개체마다 결과가 각기 다른 이유다.
결정적 이유는 굴리는 자본금의 차이에 따른 수익률의 차이다. 2000년 이전과 이후 버핏이 굴리는 자본의 액수는 비교할 수 없이 크다. 담배꽁초 전략처럼 소규모 투자를 할 당시와 규모가 달라진 것은 물론 보험업과 기업을 인수해버린 것처럼 버크셔의 투자 방식도 바뀌었다.
버핏에 대한 호감을 떠나 저런 식의 비판은 자신의 다른 주장의 근거를 설계하기 위한 의도적 왜곡이라 봐야 한다. 그 의도적 왜곡의 의도를 파악하는 것이 능력이고 투자자다.
마지막으로 정보의 비대칭성이란 부자들의 네트워크에서만 발생하는 것이 아니다.
마지막으로 정보의 비대칭성이란 부자들의 네트워크에서만 발생하는 것이 아니다.
Just over her browsing the Epstein files waiting on Nancy Guthrie updates and we found this.
Odd. Very odd.
Odd. Very odd.
Well - interesting document
Pg 1
Pg 1
pg 2
Welcome to day 1 of our NDSS 2026 showcase. We are starting with "Mad Science Hacks." These papers demonstrate attacks that bridge the physical and digital worlds, exploiting physics to break security in ways that feel impossible.
Zhang et al. repurpose telecom fiber optics as distributed acoustic sensors, recovering speech and audio from miles away by analyzing light phase changes caused by vibrations. ndss-symposium.org/ndss-paper/hid…
In simple terms: Internet cables under the street can act like hidden microphones. Scientists found that analyzing light signals inside these cables allows them to eavesdrop on nearby conversations without ever entering the building.
🔥Russia’s oil crash: Gazprom begging under $60B debt. Ukraine 👏
🧵 1/13
🧵 1/13
🔥 Russia's oil giant Gazprom begging
Gazprom faces $60 billion in debt from the past year, prompting billionaire CEO Alexei Miller to declare it "our common debt, of every citizen." Once Russia's richest corporation, it now requires rescue amid collapsing energy revenues.
2/13
Gazprom faces $60 billion in debt from the past year, prompting billionaire CEO Alexei Miller to declare it "our common debt, of every citizen." Once Russia's richest corporation, it now requires rescue amid collapsing energy revenues.
2/13
🔥 Oil prod. crisis
Russia's oil output has fallen for the third straight month, with Jan. 2026 at 9.246M bpd down 58k from Dec. Systemic analyst Evgeny Istrebyn calls it "sensational proof of Russia's vulnerability" as high-cost wells shut down below the $45-50 threshold.
3/13
Russia's oil output has fallen for the third straight month, with Jan. 2026 at 9.246M bpd down 58k from Dec. Systemic analyst Evgeny Istrebyn calls it "sensational proof of Russia's vulnerability" as high-cost wells shut down below the $45-50 threshold.
3/13
With longer duration of Long COVID, some key brain connections become weaker - especially those linked to prefrontal regulatory areas.
At the same time, other connections become stronger.
A new fMRI study shows this reflects a progressive reorganization of how brain networks communicate🧵
At the same time, other connections become stronger.
A new fMRI study shows this reflects a progressive reorganization of how brain networks communicate🧵
The study didn’t just look at isolated brain regions.
It examined how entire brain networks coordinate during cognitive effort - because performance depends less on single areas and more on how well networks synchronize
It examined how entire brain networks coordinate during cognitive effort - because performance depends less on single areas and more on how well networks synchronize
That synchronization was disrupted in Long COVID.
The main problem wasn’t damage to one function, but impaired regulation - the brain’s ability to detect what matters and shift efficiently into task-focused mode.
The main problem wasn’t damage to one function, but impaired regulation - the brain’s ability to detect what matters and shift efficiently into task-focused mode.
UN ANÁLISIS DE LA GUERRA DE UCRANIA DESDE UNA PERSPECTIVA PERONISTA - ACTUALIZACIÓN SEMANAL: 01 - 07/02/2026 - LA VENTA DE HUMO DEL ENTRENAMIENTO NORTEAMERICANO
Rusia busca derrotar a la OTAN en ucrania
Resúmenes anteriores:
Rusia busca derrotar a la OTAN en ucrania
Resúmenes anteriores:
Los norteamericanos pasaron un papelón mayúsculo entrenando en zonas árticas con los finlandeses, tanto que la dirección del ejercicio es tuvo que decir a los finlandeses “que se dejaran ganar” como pueden ver acá
militarnyi.com/en/news/us-arm…
y acá
defence-blog.com/u-s-military-f…
militarnyi.com/en/news/us-arm…
y acá
defence-blog.com/u-s-military-f…
Este es el entrenamiento norteamericano por el cual se les moja la raja a nuestras FFAA
clarin.com/politica/estad…
clarin.com/politica/estad…
ADIÓS, NETFLIX. ADIÓS, HBO.
Se acabaron los 25,99$/mes
ChatGPT ha convertido mi portátil en un centro de streaming gratuito.
Aquí tienes 9 prompts para que los pruebes tú mismo ↓
Se acabaron los 25,99$/mes
ChatGPT ha convertido mi portátil en un centro de streaming gratuito.
Aquí tienes 9 prompts para que los pruebes tú mismo ↓
⚠️ Antes de empezar, solo quería avisarte que junto con @ecommartinez, hemos reunido y categorizado 900+ herramientas de IA!
Es un verdadero diccionario IA para tu vida diaria.
¡Y es gratis! 👇
alejoxadam.beehiiv.com/c/900-herramie…
Es un verdadero diccionario IA para tu vida diaria.
¡Y es gratis! 👇
alejoxadam.beehiiv.com/c/900-herramie…
1/9 - Motor de descubrimiento de plataformas
«Busca y evalúa 15 plataformas de streaming legales y gratuitas disponibles en [TU PAÍS]. Para cada plataforma, proporciona: tamaño de la biblioteca de contenidos, calidad de vídeo (720p/1080p/4K), frecuencia de anuncios, compatibilidad con dispositivos, requisitos de registro, características destacadas y enlaces directos. Clasifícalas por valor global y destaca las tres mejores para diferentes casos de uso (películas, programas de televisión, documentales)»
«Busca y evalúa 15 plataformas de streaming legales y gratuitas disponibles en [TU PAÍS]. Para cada plataforma, proporciona: tamaño de la biblioteca de contenidos, calidad de vídeo (720p/1080p/4K), frecuencia de anuncios, compatibilidad con dispositivos, requisitos de registro, características destacadas y enlaces directos. Clasifícalas por valor global y destaca las tres mejores para diferentes casos de uso (películas, programas de televisión, documentales)»
I have stirred up a hornets nest. lol. Much to write in the coming days.
But it’s important to note this thread confirms I was not exaggerating the SBC’s claim. As a Baptist, you’re being asked to accept that if you cooperate with the SBC, its leaders can lie about local pastors and members, and decide to give no justice.
But it’s important to note this thread confirms I was not exaggerating the SBC’s claim. As a Baptist, you’re being asked to accept that if you cooperate with the SBC, its leaders can lie about local pastors and members, and decide to give no justice.
Indeed, it can lie about the justice it would give you.
You can’t even contract for them to not defame you. This is a long way from “no law…prohibiting the free exercise thereof.”
You can’t even contract for them to not defame you. This is a long way from “no law…prohibiting the free exercise thereof.”
My basic position is that Baptist cooperation is internecine cooperation. When different churches cooperate, they do not yield to the impositions of the other, other than by normal human agreements (contract).
Οι ηγεσίες των πολιτικών σχηματισμών της αριστεράς κάνουν πάντα μια λαθροχειρία: καπηλεύονται την παλικαριά και το ηθικό ανάστημα των αγωνιστών (ή και κάποιον μεμωμονωμένων από τις ηγεσίες) για να δρέψουν δάφνες ⬇️👇
Ομως θάβουν κάτω από το χαλί ότι οι κομματικές ηγεσίες διαχρονικά απότελούνται από φοβισμένους και ανίκανους ανθρώπους οι οποίοι δεν κατόρθωσαν ποτέ να παράξουν στρατηγική και να διεξάγουν νικηφόρο αγώνα
Οσες φορές τους εμπστεύτηκε ο λαός τον οδήγησαν σε σκληρές ήττες, ξεκληρίζοντας πραγματικά την πρωτοπορία του λαού και πηγαίνοντας δεκαετίες πίσω το όραμα της οικοδόμησης σύγχρονου κράτους δικαίου στη χώρα
No PAGUES a Apple por almacenamiento extra en iCloud.
Lo solucioné en 10 minutos.
Liberé 12 GB sin pagar ni un dólar.
Usa estos 8 pasos sencillos ↓
Lo solucioné en 10 minutos.
Liberé 12 GB sin pagar ni un dólar.
Usa estos 8 pasos sencillos ↓
⚠️ Antes de empezar, solo quería avisarte que junto con @adamaestr0_ y @ecommartinez, hemos reunido y categorizado 900+ herramientas de IA!
Es un verdadero diccionario IA para tu vida diaria.
¡Y es gratis! 👇
alejoxadam.beehiiv.com/c/900-herramie…
Es un verdadero diccionario IA para tu vida diaria.
¡Y es gratis! 👇
alejoxadam.beehiiv.com/c/900-herramie…
1. Encuentra qué está ocupando realmente el almacenamiento
"Ve a Ajustes > [Tu nombre] > iCloud > Gestionar almacenamiento y revisa qué está usando más espacio en Fotos, Copias de seguridad, Mail y Apps."
"Ve a Ajustes > [Tu nombre] > iCloud > Gestionar almacenamiento y revisa qué está usando más espacio en Fotos, Copias de seguridad, Mail y Apps."
É loucura pensar que R$ 70,00 não compram uma pizza decente hoje em dia...
Mas compram 10 HORAS da sua vida 🚨
Você trabalha um dia inteiro, aguenta trânsito e chefe chato, só para pagar um disco de massa com queijo que acaba em 20 minutos kk
Segue o fio: 🧶
Mas compram 10 HORAS da sua vida 🚨
Você trabalha um dia inteiro, aguenta trânsito e chefe chato, só para pagar um disco de massa com queijo que acaba em 20 minutos kk
Segue o fio: 🧶
Salário Mínimo 2026: R$ 1.621,00
Valor da sua hora (220h/mês): R$ 7,37
Preço médio de uma pizza "decente" (Capitais): R$ 70,00
R$ 70,00 ÷ R$ 7,37 = 9,5 HORAS DE TRABALHO.
Um dia comercial tem 8 horas. Você trabalha o dia todo, faz hora extra, e o dinheiro só paga o jantar.
Valor da sua hora (220h/mês): R$ 7,37
Preço médio de uma pizza "decente" (Capitais): R$ 70,00
R$ 70,00 ÷ R$ 7,37 = 9,5 HORAS DE TRABALHO.
Um dia comercial tem 8 horas. Você trabalha o dia todo, faz hora extra, e o dinheiro só paga o jantar.
Isso significa que o seu suor de 08:00 às 18:00 não vale uma massa com queijo e frete.
Isso não é apenas "inflação". É a invisibilidade do seu esforço.
Enquanto o preço dos ativos (imóveis, empresas) sobe de elevador, o valor do trabalho humano sobe de escada. E a escada está quebrada.
Isso não é apenas "inflação". É a invisibilidade do seu esforço.
Enquanto o preço dos ativos (imóveis, empresas) sobe de elevador, o valor do trabalho humano sobe de escada. E a escada está quebrada.
In Crime and Punishment, Fyodor Dostoevsky was not writing a crime novel in the modern sense. He was staging a moral experiment, almost a trial, asking a single, dangerous question… What happens if a human being believes they stands beyond moral law?
Dostoevsky is attacking the 19th century faith that human reason, detached from conscience, can replace moral truth. Raskolnikov’s crime is not merely murder, it is the attempt to think himself above humanity. Above the people around him
For Dostoevsky, reason can justify almost anything, yet the soul will not believe it. Raskolnikov discovers that guilt is not a social construct and conscience is not learned behavior. You can deny them intellectually, but you cannot escape them existentially. They are fatal
Roger Froikin @rlefraim wrote, "The Reality of State and City Debt
There is a handful of US states and large cities with so much debt that, unless they act to reduce it soon, we will see tragedy.
1)
There is a handful of US states and large cities with so much debt that, unless they act to reduce it soon, we will see tragedy.
1)
The States: California, New York, New Jersey, Connecticut, and Kentucky. The Cities: New York, San Francisco, Chicago.
You see, when one is in debt, there are limited ways to handle that debt.
2)
You see, when one is in debt, there are limited ways to handle that debt.
2)
One can increase income, decrease expenses, or defer debt to a future point in time through alternate debt. That is true of individuals, families, or government.
3)
3)
📢Entrevista
Irán sigue en tensión: crisis, protestas y represión.
#Irán #Protestas
¿Está Irán Al Borde Del Abismo? @kritik_bakis aracılığıylakritikbakis.com/es/esta-iran-a…
Irán sigue en tensión: crisis, protestas y represión.
#Irán #Protestas
¿Está Irán Al Borde Del Abismo? @kritik_bakis aracılığıylakritikbakis.com/es/esta-iran-a…
Entre crisis económica, desigualdad social y represión estatal, las protestas en Irán reflejan tensiones profundas. Huelgas obreras, demandas sociales y debates políticos marcan un momento decisivo para el país. #Irán #Protestas #Economía #Política #Geopolítica
El levantamiento en Irán muestra la intersección entre crisis económica, política interna y presiones externas. Comprender el papel del Estado, la sociedad y los actores internacionales es clave para analizar el futuro del país.#Irán #OrienteMedio #Geopolítica #Sociedad #Economía
In #Lyon hebben zelfbenoemde "antifascisten" een op de grond liggende man (23) zodanig hebben geslagen en getrapt dat hij nu dood is. Dat haalde echter niet de NOS Journaals van 18:00 uur en 20:00 uur. In #Nieuwsuur was wel kort aandacht daarvoor. 1/5 #Frankrijk #Antifa #Quentin
In dat korte Nieuwsuur-item werd weliswaar vermeld dat Quentin zwaar werd mishandeld, maar niet op welke wijze. Beelden van de mishandeling (zie video) werden ook niet getoond. Wel werd vermeld dat @Coll_Nemesis de conferentie van @FranceInsoumise zou hebben verstoord. 🙄 2/5
@Coll_Nemesis @FranceInsoumise Zoals de video van 1/5 laat zien werden Collectif Némésis en La France Insoumise niet bij naam genoemd. De autocue-voorlezer van dienst had het respectievelijk over "leden van een radicaal-rechtse groep" en een "radicaal-linkse partij". 3/5
Being self employed and having to do lots of repetitive manual computer tasks, I've been checking in on AI progress to see if I can automate myself into having more free time. In August, I got ChatGPT and Claude subscriptions and did my tasks side by side for a month...
One task involves parsing a 100+ page PDF file and tallying specific item totals with a set of rules. In August, both LLMs struggled much more than I expected. They had hallucinations and wouldn't fix when corrected. Ultimately took more time than manual and I gave up.
A second task is compiling a large spreadsheet from a bunch of differently formatted CSV files, making a series of transformations to the data based on my domain specific knowledge, fixing various formatting issues, and creating a machine readable output file.
AI news weekend roundup 🧵
GPT-5.2 just made an original discovery in theoretical physics.
ByteDance dropped a model that beats GPT-5.2 and Gemini 3 Pro at 1/10th the price.
The Pentagon might cut Anthropic's $200M deal, and more.
Here's everything you might've missed:
GPT-5.2 just made an original discovery in theoretical physics.
ByteDance dropped a model that beats GPT-5.2 and Gemini 3 Pro at 1/10th the price.
The Pentagon might cut Anthropic's $200M deal, and more.
Here's everything you might've missed:
GPT-5.2 independently discovered a new math formula in particle physics and formally proved it was correct.
It found that an existing answer in the field was actually wrong, then wrote the proof autonomously in 12 hours.
It found that an existing answer in the field was actually wrong, then wrote the proof autonomously in 12 hours.
ByteDance released Seed 2.0, a new model that competes with GPT-5.2 and Gemini 3 Pro across benchmarks.
This follows last week's viral Seedance 2.0 video model, which is now getting hit with cease-and-desists from Hollywood over copyrighted content.
This follows last week's viral Seedance 2.0 video model, which is now getting hit with cease-and-desists from Hollywood over copyrighted content.
𝐄𝐗𝐏𝐎𝐒𝐄𝐃: Rockets superstar Kevin Durant caught talking trash about his current and former teammates on burners, @gethigher77 and others.
In these messages, Kevin Durant is seen trash talking his current and former teammates
What he says is shocking 😳
(A thread 🧵)
In these messages, Kevin Durant is seen trash talking his current and former teammates
What he says is shocking 😳
(A thread 🧵)
