Salta al contenuto principale






In Chicago immigration crackdown, agents raid daycare, senior living center


A Spanish-language immersion daycare in a leafy residential neighborhood on the North Side of Chicago was raided by federal immigration agents on Wednesday and a teacher was taken away, panicking school administrators and parents of infants, toddlers and pre-kindergarten children at the center, a staff worker at the daycare told Reuters.

Footage obtained by local WGN-TV showed two men, one in a balaclava, dragging a woman out of the colorfully decorated front doors of Rayito de Sol daycare center as she screamed. The men wore vests that said "Police" but no other agency markings were visible.

https://www.reuters.com/world/us/chicago-immigration-crackdown-agents-raid-daycare-senior-living-center-2025-11-05/





Republicans file lawsuit challenging California’s redistricting measure


But, but, but ... that's my ball. You can't play with my ball!

Republicans in California on Wednesday filed a federal lawsuit challenging a high-stakes redistricting measure that could help flip up to five congressional seats for Democrats.

The suit, filed by David Tangipa, a Republican assembly member, 18 California voters and the state Republican party in the US district court for the central district of California, argues that the new maps are unconstitutional because they were drawn to increase the voting power of a particular racial group. It asks the court to block the new maps from taking effect, at least temporarily.

The measure, Proposition 50, was approved by voters on Tuesday evening, in a decisive victory for Democrats. The plan temporarily gives the power to draw congressional districts to the California legislature, allowing it to adopt maps that will help Democrats pick up five seats in the US House of Representatives.


Obviously, if you represent the people, you can't let them vote on anything themselves.


in reply to HaraldvonBlauzahn

quietly? no it has been loudly been in news and research, people just ignored it. also by denying it exist when granting "research funds" too.



Judge orders White House to use American Sign Language interpreters at briefings


"White House press briefings engage the American people on important issues affecting their daily lives — in recent months, war, the economy, and healthcare, and in recent years, a global pandemic," U.S. District Judge Amir Ali wrote in issuing a preliminary injunction on Tuesday. "The exclusion of deaf Americans from that programming, in addition to likely violating the Rehabilitation Act, is clear and present harm that the court cannot meaningfully remedy after the fact."

The White House stopped using live ASL interpreters at briefings and other public events when President Trump began his second term in January.

The National Association of the Deaf (NAD) and two deaf men filed the lawsuit against Trump and Leavitt in May. The suit also names White House chief of staff Susie Wiles, along with the offices for president and vice president. It alleges the White House's failure to provide ASL violates Section 504 of the Rehabilitation Act of 1973. The law prohibits discrimination against people with disabilities in programs conducted by the federal government. The suit also claims the White House is in violation of the First and Fifth Amendments, which protect free speech and provide for due process, respectively.






Flock haters cross political divides to remove error-prone cameras


Flock Safety—the surveillance company behind the country’s largest network of automated license plate readers (ALPRs)—currently faces attacks on multiple fronts seeking to tear down the invasive and error-prone cameras across the US. This week, two lawm

Flock Safety—the surveillance company behind the country’s largest network of automated license plate readers (ALPRs)—currently faces attacks on multiple fronts seeking to tear down the invasive and error-prone cameras across the US.

This week, two lawmakers, Senator Ron Wyden (D-Ore.) and Representative Raja Krishnamoorthi (D-Ill.), called for a federal investigation, alleging that Flock has been “negligently handling Americans’ personal data” by failing to use cybersecurity best practices. The month prior, Wyden wrote a letter to Flock CEO Garrett Langley, alleging that Flock’s security failures mean that “abuse of Flock cameras is inevitable” and that they threaten to expose billions of people’s harvested data should a catastrophic breach occur.

“In my view, local elected officials can best protect their constituents from the inevitable abuses of Flock cameras by removing Flock from their communities,” Wyden wrote.

Read full article

reshared this



Sandwich thrown by protester 'exploded' and left mustard stain on border agent, court hears


Customs and Border Patrol agent Gregory Lairmore told the jury the snack "exploded all over him" and he "could smell the onions and mustard" on his uniform.

Neither side disputes that Sean Dunn, 37, did in fact lob obscenities and a deli-style sandwich at officers deployed by President Donald Trump to patrol the nation's capital in August. But Mr Dunn's lawyer argues it was not a criminal act.

The incident was captured on video and went viral, making Mr Dunn a symbol of opposition in Washington DC to Trump.

Government prosecutors initially tried to secure felony charges against Mr Dunn, but a grand jury declined to indict him. Prosecutors have instead charged him with a lower-level misdemeanour assault.



Sandwich thrown by protester 'exploded' and left mustard stain on border agent, court hears


Customs and Border Patrol agent Gregory Lairmore told the jury the snack "exploded all over him" and he "could smell the onions and mustard" on his uniform.

Neither side disputes that Sean Dunn, 37, did in fact lob obscenities and a deli-style sandwich at officers deployed by President Donald Trump to patrol the nation's capital in August. But Mr Dunn's lawyer argues it was not a criminal act.

The incident was captured on video and went viral, making Mr Dunn a symbol of opposition in Washington DC to Trump.

Government prosecutors initially tried to secure felony charges against Mr Dunn, but a grand jury declined to indict him. Prosecutors have instead charged him with a lower-level misdemeanour assault.











When Will the AI Bubble Burst? (Gary Marcus with Murad Hemmadi) | Attention: Govern Or Be Governed






Roland Emmerich – „Das Arche Noah Prinzip“ (1984)

Roland Emmerich steht für Spektakel. Für Kino, das kracht, brennt, zittert, für Explosionen, die zu großen Erzählungen wurden. Doch bevor der Regisseur zum „Master of Disaster“ wurde, schuf er ein fast intellektuelles und stilles Debüt. Sein Studentenfilm von 1984 wird nun in der ARD wiederholt. Das ist exakt der eine Film, den ich mir aus Anlass von Emmerichs 70. Geburtstag gewünscht habe. Herzlichen Glückwunsch, Herr Emmerich! (ARD, Wh.)

Zum Blog: nexxtpress.de/mediathekperlen/…



Lemmy Development Update October 2025


During the past month, we've been working rapidly on adding features to lemmy-ui. We also finished up the last major backend changes. This means we can soon go to the beta phase for 1.0, which will focus on testing, bug fixing and helping Lemmy clients to start updating for the new API. After that will be the release candidate phase when version 1.0 will be live tested on lemmy.ml.

You can see the changes in action on the test server voyager.lemmy.ml (which was recently wiped). Registration is open, you are welcome to try things out. To stay up to date with our progress look at the lemmy-ui 1.0 and lemmy 1.0 milestone issues.

The major changes during October were:
- Speeding up migrations for 1.0 and reducing database size
- Notify users about moderation actions
- Default data for new instances (welcome post and popular communities)
- Card views for post listing
- Showing community sidebar on create post page
- Blurhash for images

::: spoiler Full list of changes by user

matc-pub



dullbananas



SleeplessOne1917



MV-GH



dessalines



Nutomic


:::

Or see the full list of changes at the links below:


An open source project the size of Lemmy needs constant work to manage the project, implement new features and fix bugs. Dessalines and Nutomic work full-time on these tasks and more. As there is no advertising or tracking, all of our work is funded through donations. Even so there is barely enough time in the day, and no time for a second job. The only available option are user donations.

To keep it viable donations need to reach a minimum of 5000€ per month, resulting in a modest salary of 2500€ per developer. If that goal is reached we can stop worrying about money, and fully focus on improving the software for the benefit of all users and instances. We especially rely on recurring donations to secure the long-term development and make Lemmy the best it can be.

Donate



A crowd-sourced review service for OpenStreetMap - General talk - OpenStreetMap Community Forum


in reply to Tungmar

lib.reviews seems to be just a five-star rating and a text box. We desperately need an open source review platform, but it needs to be simple, just a like/dislike, a question and answer tips box, ands taggable categories like foursquare had.

I really miss foursquare, and am convinced google and yelp killed it. I could go to any city in the western hemisphere, filter by vegetarian/vegan friendly, and always get incredible recommendations. Google and yelp by comparison are entirely gamed.





America’s Dumbest Billionaires Fail to Stop Zohran Mamdani


Andrew Cuomo, an elderly has-been, the lesser son of a greater sire, who as governor literally conspired with Republicans to hand them control of the New York state Senate for half a decade; who resigned from office in disgrace after he was credibly accused of 13 instances of sexual harassment; and whose campaign quite obviously had no purpose other than satisfying his own lust for accumulating personal power, along with that of his billionaire donors.

As the campaign progressed and Mamdani’s victory became ever more likely, Cuomo descended into vindictive gutter racism. He did not disagree with a right-wing radio host who said that Mamdani would be “cheering” another 9/11, suggested that Mamdani would have Muslim women “completely covered up,” and that he “doesn’t understand New York culture” because he’s a “citizen of Uganda.”

Cuomo happily took Donald Trump’s endorsement and went on Fox News to tout it. His closing campaign message, as The Nation’s Jeet Heer pointed out on Bluesky, smacked of Vidkun Quisling—implicitly threatening New Yorkers with a Trumpian occupation if they voted for anyone but Cuomo.

It was disgusting stuff. But it also was palpably desperate, and coming from one of the worst candidates imaginable...

...

What we see, I think, are a bunch of rich guys who have been comically out of touch with normal people for many decades, and more recently have blowtorched their brains into a smoking pile of ash on Elon Musk’s Twitter/X and in various group chats. It’s why they got so worked up about Mamdani in the first place—the New York City mayoralty is not some omnipotent office, and there are a dozen ways to hem it in at the state and local level if they so wished. What these oligarchs spent to stop Mamdani feels like less on an annual basis than he wants them to pay for a better future for all New Yorkers, a joke Mamdani himself has made.

In any case, his slight tax increase on rich people, free buses, and city-run grocery stores are pretty far from a communist revolution. But that’s not how it appears to rich people, surrounded on all sides by yes-men and toadies, who spend several hours a day marinating in an online Nazi sewer.



Why cities around the world are uniting to keep cool


From the desert of Phoenix in the United States to the humid streets of Quezon City in the Philippines, mayors are facing the same new reality: Heat is here to stay, and it is impacting every element of city life. That’s why we — along with more than 30 other mayors from C40 Cities, a global network of nearly 100 of the world’s biggest cities tackling the climate emergency — are joining forces to protect our people today and prepare our communities for a hotter tomorrow. Through the new C40 Cool Cities Accelerator, we’ll work together to speed up bold and inclusive climate action that meets the urgency of this growing threat.
in reply to Aneb

True, at least 30 people in a small amount of power is actually talking about making a plan.


Inverse Knowledge Search over Verifiable Reasoning: Synthesizing a Scientific Encyclopedia from a Long Chains-of-Thought Knowledge Base


This paper comes up with a really clever architectural solution to LLM hallucinations, especially for complex, technical topics. The core idea is that all our knowledge, from textbooks to wikis, is "radically compressed". It gives you the conclusions but hides all the step-by-step reasoning that justifies them. They call it a vast, unrecorded network of derivations the "intellectual dark matter" of knowledge. LLMs being trained on this compressed, conclusion-oriented data is one reason why they fail so often. When you ask them to explain something deeply, they just confidently hallucinate plausible-sounding "dark matter".

The solution the paper demonstrates is to use a massive pipeline to "decompress" all of the steps and make the answer verifiable. It starts with a "Socrates agent" that uses a curriculum of about 200 university courses to automatically generate around 3 million first-principles questions. Then comes the clever part, which is basically a CI/CD pipeline for knowledge. To stop hallucinations, they run every single question through multiple different LLMs. If these models don't independently arrive at the exact same verifiable endpoint, like a final number or formula, the entire question-and-answer pair is thrown in the trash. This rigorous cross-model consensus filters out the junk and leaves them with a clean and verified dataset of Long Chains-of-Thought (LCoTs).

The first benefit of having such a clean knowledge base is a "Brainstorm Search Engine" that performs "inverse knowledge search". Instead of just searching for a definition, you input a concept and the engine retrieves all the diverse, verified derivational chains that lead to that concept. This allows you to explore a concept's origins and see all the non-trivial, cross-disciplinary connections that are normally hidden. The second and biggest benefit is the "Plato" synthesizer, which is how they solve hallucinations. Instead of just generating an article from scratch, it first queries the Brainstorm engine to retrieve all the relevant, pre-verified LCoT "reasoning scaffolds". Its only job is then to narrate and synthesize those verified chains into a coherent article.

The results are pretty impressive. The articles generated this way have significantly higher knowledge-point density and, most importantly, substantially lower factual error rates, reducing hallucinations by about 50% compared to a baseline LLM. They used this framework to automatically generate "SciencePedia," an encyclopedia with an initial 200,000 entries, solving the "cold start" problem that plagues human-curated wikis. The whole "verify-then-synthesize" architecture feels like it could pave the way for AI systems that are able to produce verifiable results and are therefore trustworthy.



Inverse Knowledge Search over Verifiable Reasoning: Synthesizing a Scientific Encyclopedia from a Long Chains-of-Thought Knowledge Base


This paper comes up with a really clever architectural solution to LLM hallucinations, especially for complex, technical topics. The core idea is that all our knowledge, from textbooks to wikis, is "radically compressed". It gives you the conclusions but hides all the step-by-step reasoning that justifies them. They call it a vast, unrecorded network of derivations the "intellectual dark matter" of knowledge. LLMs being trained on this compressed, conclusion-oriented data is one reason why they fail so often. When you ask them to explain something deeply, they just confidently hallucinate plausible-sounding "dark matter".

The solution the paper demonstrates is to use a massive pipeline to "decompress" all of the steps and make the answer verifiable. It starts with a "Socrates agent" that uses a curriculum of about 200 university courses to automatically generate around 3 million first-principles questions. Then comes the clever part, which is basically a CI/CD pipeline for knowledge. To stop hallucinations, they run every single question through multiple different LLMs. If these models don't independently arrive at the exact same verifiable endpoint, like a final number or formula, the entire question-and-answer pair is thrown in the trash. This rigorous cross-model consensus filters out the junk and leaves them with a clean and verified dataset of Long Chains-of-Thought (LCoTs).

The first benefit of having such a clean knowledge base is a "Brainstorm Search Engine" that performs "inverse knowledge search". Instead of just searching for a definition, you input a concept and the engine retrieves all the diverse, verified derivational chains that lead to that concept. This allows you to explore a concept's origins and see all the non-trivial, cross-disciplinary connections that are normally hidden. The second and biggest benefit is the "Plato" synthesizer, which is how they solve hallucinations. Instead of just generating an article from scratch, it first queries the Brainstorm engine to retrieve all the relevant, pre-verified LCoT "reasoning scaffolds". Its only job is then to narrate and synthesize those verified chains into a coherent article.

The results are pretty impressive. The articles generated this way have significantly higher knowledge-point density and, most importantly, substantially lower factual error rates, reducing hallucinations by about 50% compared to a baseline LLM. They used this framework to automatically generate "SciencePedia," an encyclopedia with an initial 200,000 entries, solving the "cold start" problem that plagues human-curated wikis. The whole "verify-then-synthesize" architecture feels like it could pave the way for AI systems that are able to produce verifiable results and are therefore trustworthy.




Trump’s widening war on the left started with Palestine


cross-posted from: news.abolish.capital/post/5821

President Donald Trump greets Israeli Prime Minister Benjamin Netanyahu and his wife, Monday, July 7, 2025, at the South Portico of the White House.(Official White House Photo by Daniel Torok)The Trump administration's recent efforts to target left-wing groups started with attacks on the Palestine movement, following the strategy established by pro-Israel organizations that worked for decades to pave the way for such repression.

In September, Trump issued an executive order claiming to designate “Antifa” as a “domestic terrorist organization” and a presidential memorandum (NSPM-7) that targets charities and advocacy groups over alleged national security concerns.

These efforts were seemingly driven by the assassination of right-wing commentator Charlie Kirk, which the Trump administration has continually blamed on the left despite a complete lack of evidence.

“The last message that Charlie sent me … was that we needed to have an organized strategy to go after the left-wing organizations that are promoting violence in this country,” declared White House deputy chief of staff Stephen Miller shortly after Kirk’s killing. “With God as my witness, we are going to use every resource we have at the Department of Justice, Homeland Security, and throughout this government to identify, disrupt, dismantle, and destroy these networks.”

While Kirk’s death provided the spark for Trump’s recent moves, the administration’s war on the left effectively began with its targeting of Palestine advocates.

Almost immediately upon arriving in the White House, the Trump team revoked visas, snatched people off the streets, detained legal citizens, and launched a McCarthyite campaign against university administrations for allowing anti-Israel sentiment to foment on their campuses.

“We ought to get them all out of the country,” declared Trump, referring to students who protested the genocide. “They’re troublemakers. They’re agitators. They don’t love our country. We ought to get them the hell out.”

From Mondoweiss via This RSS Feed.



problemi di gaming per la dimenticanza cosmica, e finisco così super perduta nel rotting senza più giocare (mi sono dimenticata di nuovo lo Switch a casa)


Ommiao, oggi… solita noia, non c’è proprio niente da fare, non ho proprio via di scampo. Ommeglio: si potrebbe invero fare del gran gaming, come ideale tentativo di distrarsi dagli orrori, ma anche il gaming è divenuto ormai più difficile del dovuto. Cioè: una volta dentro, è sempre gaming, per fortuna, e c’è ben poco […]

octospacc.altervista.org/2025/…


problemi di gaming per la dimenticanza cosmica, e finisco così super perduta nel rotting senza più giocare (mi sono dimenticata di nuovo lo Switch a casa)


Ommiao, oggi… solita noia, non c’è proprio niente da fare, non ho proprio via di scampo. Ommeglio: si potrebbe invero fare del gran gaming, come ideale tentativo di distrarsi dagli orrori, ma anche il gaming è divenuto ormai più difficile del dovuto. Cioè: una volta dentro, è sempre gaming, per fortuna, e c’è ben poco se non niente che possa mai cambiare in peggio; ma, sulla soglia, tra il prima ed il dopo per così dire, si verificano inevitabilmente problemi e rotture, e io davvero non. 🙀

Una cosa strana che mi sta infatti succedendo — che con oggi è accaduta credo un totale di sole 2 volte, un numero ancora vagamente accettabile, e solo per questo non mi è ancora venuto da piangere… se dovesse ricapitarmi, credo che le lacrime inizierebbero finalmente ad uscire — è che mi sto dimenticando il Nintendo Switch a casa quando vado all’università. O meglio, perché il diavolo è sempre, e malamente, nei dettagli: quando mi ricordo di mettere la console in zaino, alla fine, nei momenti morti non mi capita sempre di volerci giocare; magari faccio gaming sul telefono, o tutte altre cose… e invece, se un giorno deve venirmi in autobus un istante di voglia di giocarci, mi deve venire proprio uno di quei rari giorni (pensate che combinazione!!!) in cui ho dimenticato di portarla. Insomma, la vecchiaia imperversa… a soli 21 anni… 😪

Ovviamente, il secondo caso è oggi, altrimenti non me ne starei lagnando (o forse si, ormai sono imprevedibile persino a me stessa). Ed è così tragico, perché uffi… e c’è traffico, e c’è assenza di persone che non mi odiano (ok, no, proprio oggi per fortuna questo non è capitato), e c’è tempo da spendere prima e dopo le ore di perdita di temp—ehm, cioè, di lezione… Se non ho voglia di programmare, e non ho particolarmente voglia di leggere, e non ho niente di particolare da poter e voler scrivere sul momento, non posso fare altro che il gaming, come attività importante. Il motivo per cui mi scordo lo Switch, però, è a suo modo molto buffo (‘nzomma); e sì, in parte è questione di skill issue, ma in altra parte è colpa di Nintendo merda (spoiler: l’ultimo punto). Questo perché: 🤭

  • Se la console non è al centro della scrivania, dove metto sempre tutte le cose da prendere quando esco, ed è invece nella dock (…nonostante questa sia giusto affianco), non mi risalta alla vista, e quindi mi sfugge di prenderla per metterla nello zaino.
  • Se non ci gioco né la sera prima, né la mattina stessa, è molto improbabile che io mi ricordi di posarla sopra la scrivania oppure direttamente nello zaino, quindi rimane in dock, e si presenta il problema di cui sopra.
  • Non posso semplicemente tenere lo Switch nello zaino finché non mi viene voglia di usarlo la sera a casa, perché la batteria di ‘sta console di merda si scarica in meno di 3 giorni se sta in standby, anche col WiFi spento e senza nessun gioco lasciato aperto — e non posso spegnerla quando non la uso, perché già di suo è palloso, e io in particolare dovrei poi ogni volta collegare PC o telefono via USB per avviare il CFW, che lasciamo stare…

E certo, stare senza console non significa stare assolutamente senza gaming, ma sul telefono ho perlopiù solo giochi puzzle, perché tutte le altre cose sono o scomode da giocare lì, o richiedono troppe risorse; delle eccezioni le ho, ma non sempre sono il passatempo più ideale, quando poi magari voglio giocare a Pokémon Brainrot Z-A, dai. Mentre invece, comunque sia, se devio da questa scelta binaria, qualcosa mi va in qualche modo sempre male… La consolina PocketGo che ho è così piccola che potrei quasi tenerla come portachiavi, ma a giocare lì sopra mi cieco, e allora non va bene, problemi di salute; Giocare sul PC portatile invece, a parte che non è fattibile in autobus, ma solo in posti statici, è un troiaio, perché con gli emulatori e le mappature nghhhh… (Giocando al MAME sul PC stamattina ho per sbaglio caricato uno stato quando invece volevo salvarlo; sono al mio fottuto limite.) 😵‍💫
ME EVERYDAY PLAYINGMY SILLY LITTLE GAMESQuesta in foto sarei dunque io in questo momento (o beh, poco fa; o forse, al contrario, tra poco), con appunto la Swiss, se solo ce l’avessi appresso (anche se i miei joycon sono blu e rosso, non verde e rosa); e invece no. E i miei giochi saranno silly, si, ma io che resto senza lo sono 10 volte di più… Che poi, ok la volta scorsa che ero di fretta per scendere di casa, ma stamattina avevo così tanto tempo extra per prepararmi e tutto che, una volta scesa, ho dovuto persino farmi una passeggiata fino ad una fermata dell’autobus più lontana della mia solita, perché di stare ferma impalata mi secco… ma, di aver dimenticato la console me ne sono accorta ovviamente solo una volta sull’autobus. Fuck my stupid smemorated gamecel life, bla bla bla… Consigli su come posso fare per non dimenticarmi più la console, per favore??? 😶‍🌫️
#dimenticanza #forgot #gaming #memoria #NintendoSwitch #problemi