Endorsed! Timothy Grady for Ohio!
During last Sunday’s Pirate National Committee meeting, members voted to endorse the campaign of Timothy Grady, independent candidate for Governor of Ohio in his 2026 gubernatorial race.
Having joined out meeting, which you can catch the recap of here, Mr. Grady gave us his best pitch for not only why we should endorse his campaign, but also an opportunity for Ohio voters to hear from the independent candidate.
After a long meeting and opportunity to ask questions, the decision was unanimous: the US Pirate Party will support the Grady campaign!
You can see the official announcement from Timothy Grady’s page here.
It should be noted that Tim Grady is not running as a proxy of the Ohio Pirate Party (which is active but unofficial), nor is the Grady campaign strictly a Pirate campaign. The United States Pirate Party values honest campaigns, person-first agendas and anyone who fights for free and open.
The 2026 election cycle will feature explicitly Pirate candidates, running as independents, DINOS/RINOS and explicitly as Pirates. While Timothy Grady is not one of those Pirate candidates, we are happy to throw our support towards a candidate who shares our values.
Ohio, you have a chance to say “Enough is enough”.
Timothy Grady, Victory is Arrrs
Qualcosa di strano galleggia nei nostri oceani da anni e gli scienziati hanno finalmente capito di cosa si tratta
Gli scienziati hanno scoperto 27 milioni di tonnellate di materiale invisibile nell'Atlantico, il che suggerisce danni profondi e potenzialmente irrevLee Bell (Meteored Italia)
3D-Printing A Full-Sized Kayak In Under A Day
If you want to get active out on the water, you could buy a new kayak, or hunt one down on Craigslist, Or, you could follow [Ivan Miranda]’s example, and print one out instead.
[Ivan] is uniquely well positioned to pursue a build like this. That’s because he has a massive 3D printer which uses a treadmill as a bed. It’s perfect for building long, thin things, and a kayak fits the bill perfectly. [Ivan] has actually printed a kayak before, but it took an excruciating 7 days to finish. This time, he wanted to go faster. He made some extruder tweaks that would allow his treadmill printer to go much faster, and improved the design to use as much of the belt width as possible. With the new setup capable of extruding over 800 grams of plastic per hour, [Ivan] then found a whole bunch of new issues thanks to the amount of heat involved. He steps through the issues one at a time until he has a setup capable of extruding an entire kayak in less than 24 hours.
This isn’t just a dive into 3D printer tech, though. It’s also about watercraft! [Ivan] finishes the print with a sander and a 3D pen to clean up some imperfections. The body is also filled with foam in key areas, and coated with epoxy to make it watertight. It’s not the easiest craft to handle, and probably isn’t what you’d choose for ocean use. It’s too narrow, and wounds [Ivan] when he tries to get in. It might be a floating and functional kayak, just barely, for a smaller individual, but [Ivan] suggests he’ll need to make changes if he were to actually use this thing properly.
Overall, it’s a project that shows you can 3D print big things quite quickly with the right printer, and that maritime engineering principles are key for producing viable watercraft. Video after the break.
youtube.com/embed/9DpMkYDCq9Y?…
Vulnerabilità nel sistema di accesso online per concessionari auto: un ricercatore trova falle di sicurezza
È stata individuata una vulnerabilità nel sistema di accesso online per i concessionari di una delle più grandi case automobilistiche al mondo: è bastato scavare un po’ nel codice della pagina. Il ricercatore di sicurezza Eaton Zwer di Harness ha riferito di essere riuscito a sfruttare la vulnerabilità per creare un account amministrativo con diritti di accesso completi al portale interno del produttore. La violazione ha consentito di ottenere dati riservati dei clienti, informazioni sul veicolo e persino di controllare da remoto le funzioni dell’auto, incluso lo sblocco.
Zwer, che in precedenza aveva individuato bug nei sistemi delle case automobilistiche , scoprì il problema per caso, durante un progetto personale svolto nel fine settimana. Scoprì che, al caricamento della pagina di login, il browser del cliente caricava un codice errato che poteva essere modificato per bypassare tutti i meccanismi di autenticazione. Ciò rese possibile la creazione di un account di “amministratore nazionale” che dava accesso a oltre 1.000 concessionarie negli Stati Uniti.
Attraverso questa interfaccia, era possibile visualizzare i dati personali dei clienti, inclusi i recapiti e alcune informazioni finanziarie, nonché gestire i servizi relativi ai veicoli. Tra le altre cose, ciò includeva il monitoraggio in tempo reale dei veicoli aziendali e trasportati, l’utilizzo di sistemi telematici e persino l’annullamento delle spedizioni dei veicoli.
Uno degli elementi più inquietanti del sistema era lo strumento di ricerca clienti, che richiedeva solo nome e cognome per accedere alle informazioni su un’auto specifica e sul suo proprietario. Zver ha utilizzato come esempio il numero di telaio di un’auto parcheggiata in strada e ha confermato che questo era sufficiente per associare l’auto a una persona specifica. Secondo lui, era possibile avviare la procedura di trasferimento dell’auto sotto il controllo di un altro utente semplicemente confermando la propria intenzione, senza alcuna verifica. Ha testato questo scenario con il consenso di un amico ed è riuscito a controllare efficacemente l’auto di qualcun altro tramite un’app mobile.
Non meno pericolosa era la possibilità di accedere ai sistemi collegati di altri concessionari utilizzando un unico login. Grazie al meccanismo SSO (Single Sign-On), l’account amministratore creato poteva non solo spostarsi tra diverse parti dell’infrastruttura, ma anche imitare l’accesso di un altro utente. Ciò consentiva l’accesso ai diritti, ai dati e ai sistemi del dipendente preso di mira a sua insaputa: un meccanismo simile era già stato utilizzato in precedenza nel portale dei concessionari.
Il ricercatore ha definito l’architettura una “bomba a orologeria“, osservando che gli utenti potevano visualizzare e utilizzare informazioni critiche, tra cui accordi, lead e analisi interne, senza essere scoperti. L’azienda avrebbe corretto la vulnerabilità entro una settimana dalla divulgazione privata del problema nel febbraio 2025. Tuttavia, un’indagine ha dimostrato che l’exploit non era mai stato utilizzato prima: Zwer sarebbe stato il primo a scoprire e segnalare le falle nel sistema.
Secondo Zver, la radice del problema era ancora una volta qualcosa di banale: falle nel sistema di autenticazione API. Solo due vulnerabilità hanno messo a nudo l’intero mondo interno della rete di concessionari. Zver ritiene che questo sia un ulteriore promemoria: non appena il controllo degli accessi crolla, crolla tutto.
L'articolo Vulnerabilità nel sistema di accesso online per concessionari auto: un ricercatore trova falle di sicurezza proviene da il blog della sicurezza informatica.
2025 One Hertz Challenge: Abstract Aircraft Sculpture Based On Lighting Regulations
The 2025 One Hertz Challenge is really heating up with all kinds of projects that do something once every second. [The Baiko] has given us a rather abstract entry that looks like a plane…if you squint at it under the right conditions.
Powered by a coin cell, they effectively create a abstract representation of a plane in the night sky, paired with a flashing strobe that meets the requirements of the contest. [The Baiko] isn’t exactly sure of the total power draw, but notes it must be low given the circuit has run for weeks on a 30 mAh coin cell.
It’s an amusing piece of PCB art, though from at least one angle, it does appear the red LED might be on the wrong side to meet FAA regulations. Speculate on that in the comments.
In any case, we’ve had a few flashers submitted to the competition thus far, and you’ve got until August 19 to get your own entry in!
As Britain experiences one of its worst droughts in decades, its leaders suggest people get rid of old data to reduce stress on data centers.#News #UK
UK Asks People to Delete Emails In Order to Save Water During Drought
It’s a brutally hot August across the world, but especially in Europe where high temperatures have caused wildfires and droughts. In the UK, the water shortage is so bad that the government is urging citizens to help save water by deleting old emails. It really helps lighten the load on water hungry datacenters, you see.The suggestion came in a press release posted on the British government’s website Tuesday after a meeting of its National Drought Group. The release gave an update on the status of the drought, which is bad. The Wye and Ely Ouse rivers are at their lowest ever recorded height and “five areas are officially in drought, with six more experiencing prolonged dry weather following the driest six months to July since 1976,” according to the release. It also listed a few tips to help people save on water.
playlist.megaphone.fm?p=TBIEA2…
The tips included installing a rain butt to collect rainwater for gardening, fixing leaks the moment they happen, taking shorter showers, and getting rid of old data. “Delete old emails and pictures as data centres require vast amounts of water to cool their systems,” the press release suggested.Datacenters suck up an incredible amount of water to keep their delicate equipment cool. The hotter it is, the more water it uses and a heatwave spikes the costs of doing business. But old emails lingering in cloud servers are a drop in the bucket for a data center compared to processing generative AI requests.
A U.S. A Government Accountability Office report from earlier this year estimated that 60 queries of an AI system consumed about a liter of water, or roughly 1.67 Olympic sized swimming pools for the 250,000,000 queries generated in the U.S. every day. The World Economic Forum has estimated that AI datacenters will consume up to 1.7 trillion gallons of water every year by 2027. OpenAI CEO Sam Altman has disputed these estimates, saying that an average ChatGPT query uses “roughly one fifteenth of a teaspoon” of water.
Downing Street announced plans in January to “turbocharge AI” in the U.K. The plan includes billions of pounds earmarked for the construction of massive water-hungry datacenters, including a series of centers in Wales that will cost about $16 billion. The announcement about the AI push said it will create tens of thousands of jobs. It doesn’t say anything about where the water will come from.
In America, people are learning that living next to these massive AI data centers is a nightmare that can destroy their air and water quality. People who live next to massive Meta-owned datacenters in Georgia have complained of a lack of water pressure and diminished quality since the data centers moved in. In Colorado, local government and activists are fighting tech companies attempting to build massive data centers in a state that struggled with drought before the water-hungry machines moved in.
Like so many other systemic issues linked to climate change and how people live in the 21st century, small-scale personal solutions like “delete your old emails” won’t solve the problem. The individual water bill for a person’s old photos is nothing compared to the gallons of water required by large corporate clients running massive computers.
“We are grateful to the public for following the restrictions, where in place, to conserve water in these dry conditions,” Helen Wakeham, the UK Environment Agency’s Director of Water, said in the press release. “Simple, everyday choices—such as turning off a tap or deleting old emails—also really helps the collective effort to reduce demand and help preserve the health of our rivers and wildlife.”
Representatives from the UK Government did not immediately return 404 Media’s request for comment.
Could data center boom threaten Colorado's water supply and climate goals?
Two metro Denver construction sites offer a glimpse into the power and water needs of the growing data center industry as it tries to provide capacity for artificial intelligence and our online liv…Elise Schmelzer (The Denver Post)
Design Review: LattePanda Mu NAS Carrier
It is a good day for design review! Today’s board is the MuBook, a Lattepanda Mu SoM (System-on-Module) carrier from [LtBrain], optimized for a NAS with 4 SATA and 2 NVMe ports. It is cheap to manufacture and put together, the changes are non-extensive but do make the board easier to assemble, and, it results in a decent footprint x86 NAS board you can even order assembled at somewhere like JLCPCB.
This board is based on the Lite Carrier KiCad project that the LattePanda team open-sourced to promote their Mu boards. I enjoy seeing people start their project from a known-working open-source design – they can save themselves lots of work, avoid reinventing the wheel and whole categories of mistakes, and they can learn a bunch of design techniques/tips through osmosis, too. This is a large part of why I argue everyone should open-source their projects to the highest extent possible, and why I try my best to open-source all the PCBs I design.
Let’s get into it! The board’s on GitHub as linked, already containing the latest changes.
Git’ting Better
I found the very first review item when downloading the repo onto my computer. It took a surprising amount of time, which led me to believe the repo contains a fair bit of binary files – something quite counterproductive to keep in Git. My first guess was that the repo had no .gitignore for KiCad, and indeed – it had the backups/ directory with a heap of hefty .zips, as well as a fair bit of stuff like gerbers and footprint/symbol cache files. I checked in with [LtBrain] that these won’t be an issue to delete, and then added a .gitignore from the Blepis project.
This won’t make the repo easier to check out in the future, sadly – the hefty auto-generated files are still in the repo history. However, at least it won’t grow further as KiCad puts new archives into the backups/ directory, and, it’s good to keep .gitignore files in your KiCad repos so you can easily steal them every time you start a new project.
Apart from that, a .gitignore also makes working with your repository way way easier! When seeing changes overview in git status or GitHub Desktop, it’s way nicer to, and you even get a shot at reviewing changes in your commits to make sure you’re not adding something you don’t want in the repository. Oh, and, you don’t risk leaking your personal details as much, since things like auto-generated KiCad lockfiles will sometimes contain your computer name or your user name.
Now that the PCB Git-ability has been improved, let’s take a look at the board, first and foremost; the schematic changes here are fairly minimal, and already reviewed by someone else.
Cheap With Few Compromises
There’s plenty of PCIe, USB3, and SATA on this board – as such, it has to be at least four layers, and this one is. The SIG-GND-GND-SIG arrangement is only slightly compromised by a VDC (12 V to 15 V) polygon on one of the layers, taking up about 30% of space, and used to provide input power to Mu and also onboard 3.3 V and 5 V regulators.
Of course, with so many interfaces, you’ll also want to go small – you’ll have to fit a lot of diffpairs on the board, and you don’t want them flowing too close to each other to avoid interference. This board uses approximately 0.1 mm / 0.1 mm clearances, which, thankfully, work well enough for JLCPCB – the diffpairs didn’t even need to be redrawn much. Apart from that, the original design used 0.4 mm / 0.2 mm vias. Problem? JLC has a $30 surcharge for such vias for a board of this size. No such thing for 0.4 mm / 0.3 mm vias, surprisingly, even though the annular ring is way smaller.
I went and changed all 0.4 mm / 0.2 mm vias to 0.4 mm / 0.3mm vias, and that went surprisingly well – no extra DRC errors. The hole-to-copper distance is set to be pretty low in this project, to 0.15 mm, because that’s inherited from LattePanda carrier files, so I do hope that JLC doesn’t balk at those vias during the pre-production review. Speaking of DRC, I also set all courtyard errors to “ignore” – not only does this category have low signal-to-noise ratio, the LattePanda module courtyard also would raise problems at all items placed under the module, even though there’s plenty of space as long as you use a DDR socket tall enough.
One thing looked somewhat critical to me, though – the VDC polygon, specifically, the way it deprived quite a few diffpairs from GND under them.
Redraw, Nudge, Compromise
Remember, you want a ground polygon all along the underside of the differential pair, from start to finish, without interruptions – that ground polygon is where ground return current flows, and it’s also crucial in reaching the right differential pair impedance. The VDC polygon did interrupt a good few pairs, however.
Most of those interruptions were fixed easily by lifting the VDC polygon. Highlighting the net (` keyboard key) showed that there’s only really 4 consumers of the VDC power input, and all of them were above the overwhelming majority of the diffpairs. REFCLKs for M.2 sockets had to be rerouted to go over ground all throughout, though, and I also added a VDC cutout to pull gigabit Ethernet IC PCIe RX/TX pairs over VDC for most of their length.
This polygon carries a fair bit of current, a whole N100 (x86) CPU’s worth and then some, and remember – inner layers are half as thick, only 0. 5oz instead of 1 oz you get for outer layers by default. So, while we can cut into it, the VDC path has to be clear enough. A lot of items on VDC, like some gigabit controller power lines, ended up being moved from the VDC polygon layer to the opposite inner layer – now, they’re technically on the layer under PCIe and gigabit Ethernet pairs, but it’s a better option than compromising VDC power delivery. I also moved some VDC layer tracks to B.Cu and F.Cu; remember, with high-speed stuff you really want to minimize the number of inner layer tracks.
Loose Ends
With the vias changed and polygon redrawn, only a few changes remained. Not all diffpair layer crossings had enough vias next to them, and not all GND pads had vias either – particularly on the Mu and M.2 slots, what’s with high-speed communications and all, you have to make sure that all GND pads have GND vias on them. Again, highlight GND net (`) and go hunting. Afterwards, check whether you broke any polygons on inner layers – I sure did accidentally make a narrow passage on VDC even more narrow with my vias, but it didn’t take much to fix. Remember, it’s rare that extra vias cost you extra, so going wild on them is generally safe.
The SATA connector footprint from Digikey was faulty – instead of plated holes for through-hole pins, it had non-plated holes. Not the kind of error I’ve ever seen with easyeda2kicad, gotta say. As an aside, it was quite a struggle to find the proper datasheet on Digikey – I had to open like five different PDFs before I found one with footprint dimension recommendations.
A few nets were NC – as it turned out, mostly because some SATA ports had conflicting names; a few UART testpoints were present in the schematic but not on the board, so I wired them real quick, too. DRC highlighted some unconnected tracks – always worth fixing, so that KiCad can properly small segments into longer tracks, and so that your track moves don’t then result in small track snippets interfering with the entire plan. Last but not least, the BIOS sheet in the schematic was broken for some reason; KiCad said that it was corrupted. Turned out that instead of BIOS.kicad_sch, the file was named bios.kicad_sch – go figure.
Production Imminent
These changes helped [LtBrain] reduce PCB manufacturing cost, removed some potential problems for high-speed signal functioning, and fixed some crucial issues like SATA port mounting pins – pulling an otherwise SMD-pad SATA port off the board is really easy on accident! They’re all on GitHub now, as you’d expect, and you too can benefit from this board now.
8/19: Oppose Police Social Media Surveillance
Boston Police (BPD) continue their efforts rollout more surveillance tools. This time on social media.
On August 19th, the Boston Public Safety Committee will hold a hearing on the Boston 2024 Surveillance Technology Report including police usage of three new tools to monitor social media posts. Any tool BPD uses will feed into the Boston Regional Information Center (BRIC) and Federal agencies such as ICE, CBP and the FBI.
If you want to tell the Boston Public Safety committee to oppose this expansion of surveillance, please show up on the 19th virtually. Details are posted, but to sign up to speak, email ccc.ps@boston.gov and they will send you a video conference link. We especially encourage Boston Pirates to attend and speak against this proposal. The Docket # is 1357.
Continuous-Path 3D Printed Case is Clearly Superior
[porchlogic] had a problem. The desire was to print a crystal-like case for an ESP32 project, reminiscent of so many glorious game consoles and other transparent hardware of the 1990s. However, with 3D printing the only realistic option on offer, it seemed difficult to achieve a nice visual result. The solution? Custom G-code to produce as nice a print as possible, by having the hot end trace a single continuous path.
The first job was to pick a filament. Transparent PLA didn’t look great, and was easily dented—something [porchlogic] didn’t like given the device was intended to be pocketable. PETG promised better results, but stringing was common and tended to reduce the visual appeal. The solution to avoid stringing would be to stop the hot end lifting away from the print and moving to different areas of the part. Thus, [porchlogic] had to find a way to make the hot end move in a single continuous path—something that isn’t exactly a regular feature of common 3D printing slicer utilities.
The enclosure itself was designed from the ground up to enable this method of printing. Rhino and Grasshopper were used to create the enclosure and generate the custom G-code for an all-continuous print. Or, almost—there is a single hop across the USB port opening, which creates a small blob of plastic that is easy to remove once the print is done, along with strings coming off the start and end points of the print.
Designing an enclosure in this way isn’t easy, per se, but it did net [porchLogic] the results desired. We’ve seen some other neat hacks in this vein before, too, like using innovative non-planar infill techniques to improve the strength of prints.
youtube.com/embed/2Sy50BrlDMo?…
Thanks to [Uxorious] and [Keith Olson] for the tip!
A DEA agent used a local cop's password "for federal investigations in late January 2025 without [the cop's] knowledge of said use."
A DEA agent used a local copx27;s password "for federal investigations in late January 2025 without [the copx27;s] knowledge of said use."#Flock
Feds Used Local Cop's Password to Do Immigration Surveillance With Flock Cameras
A Drug Enforcement Administration agent used a local police officer’s password to the Flock automated license plate reader system to search for someone suspected of an “immigration violation.” That DEA agent did this “without [the local police officer’s] knowledge,” and the password to the Flock account, which belonged to the Palos Heights PD, has since been changed. Using license plate readers for immigration enforcement is illegal in Illinois, and casual password sharing between local police and federal law enforcement for access to surveillance systems is, at the very least, against Flock’s terms of service.The details of the search were first reported by the investigative news outlet Unraveled, which obtained group chats about the search using a public records request. More details about the search were obtained and shared with 404 Media by Shawn, a 404 Media reader who filed a public records request with Palos Heights after attending one of our FOIA Forums.
DEA agent used Illinois cop’s Flock license plate reader password for immigration enforcement searches
A federal Drug Enforcement Administration agent on a Chicago area task force used Palos Heights Detective Todd Hutchinson’s login credentials to perform unauthorized searches this past January. Group chat screenshots obtained via public records request show the detective and the feds discussing the incident.Unraveled Press
Flock makes automated license plate reader (ALPR) cameras, which passively collect the time, plates, and model of cars that drive past them and enter them into a network that can then be searched by police. Our investigation in May showed that federal agents were gaining side-door access into this system by asking local police to perform immigration enforcement searches for them; the new documents show that in some cases, local police have simply given federal agents their passwords.The documents obtained by Unraveled show details of an internal investigation done by the Palos Heights, Illinois police department in response to a series of questions that I asked them for an article we published in May that appeared to show a Todd Hutchinson, a police officer in Palos Heights, performing a series of Flock searches in January as part of their research into an “immigration violation.”
At the time, Palos Heights police chief Mike Yott told me that Hutchinson was a member of a DEA task force “that does not work immigration cases.”
“None of our officers that work with federal agencies have cross designation as immigration officers, and therefore have no immigration authority, and we and our partner agencies are very sensitive to the fact that we and the State of Illinois do not pursue immigration issues,” Yott said. “Based on the limited information on the report, the coding/wording may be poor and the use of Flock may be part of a narcotics investigation or a fugitive status warrant, which does on occasion involve people with various immigration statuses.”
Our reporting set off an internal investigation into what these searches were for, and who did them, according to the documents obtained by Unraveled. According to a July 9 investigation report written by the Palos Heights Police Department, Hutchinson was the only task force member who had access to Flock. Information about what the search was actually for is redacted in the internal investigation, and neither the Palos Heights Police Department nor the DEA has said what it was for.
“Hutchinson advised that it was common that he allowed others to use his login to Flock during the course of their drug investigations. TFO Hutchinson spoke to his group and learned that one of the DEA agents completed these searches and used his login information,” the report says. The DEA agent (whose name is redacted in the report) “did in fact use Hutchinson’s login for federal investigations in late January 2025 without Hutchinson’s knowledge of said use.”
“When I had shared my account with the Special Agent, I believed it would only be used for DEA/narcotics related investigations,” Hutchinson wrote in an email to his bosses explaining why he shared his password. Hutchinson said in a series of text messages to task force officers, which were also obtained by Unraveled, that he had to change the password to lock other members of the task force out of the system.
“What’s the new password?,” a task force member wrote to Hutchinson.
“Sorry man. Keys had to be taken away,” he responded.
The task force member replied with a gif of a sad Chandler Bing from friends sitting in the rain.
“Hey guys I no longer have access to Flock cause Hutch took my access away,” another group text reads. “Apparently someone who has access to his account may have been running plates and may have placed the search bar ‘immigration’.. which maybe have brought undue attention to his account. Effective immediately Defer all flock inquiries to Toss Hutchinstein[sic].”“Dear Todd, I hope you don’t get in trouble cause of my mistake,” the DEA agent joked in the group chat. “U were so helpful in giving the group access but now that is gone, gone like dust,…..in the wind … Trust is broken / I don’t know if bridges can be mended … one day we might be back to normal but until then I will just have to sit by this window and pray things will return … Best Regards. Ps, can u flock a plate for me”
“Only time will tell my fate, I suppose,” Hutchinson responded. “What’s the plate? And confirming it is NOT for immigration purposes…”
“It was a test …… and u passed ….,” the DEA agent responds.
In response to a separate public records request filed by Shawn, the 404 Media reader, and shared with us, the Palos Heights Police Department said “Our investigation into this matter has revealed that while these inquiries appear to have been run as part of a taskforce assignment, no member of the Palos Heights Police Department ‘ran’ those queries. They were, apparently, run by another, non-Palos Heights, task force member who used a Palos Height's member's sign in and password information without his knowledge.”The Palos Heights Police Department said in its investigation files that “this incident has brought to light the need to review our own protocols of LPR use.” The police department said that it had decided to limit searches of its Flock system only to agencies within the state of Illinois, rather than to police departments around the country. The department also turned on two-factor authentication, which had not been previously enabled.
“Lastly, I believe there is a need to start a monthly review of our own flock searches to ensure our officers are working within standards and compliant with all policies and laws,” the report says.
Palos Heights’ casual sharing of passwords to a powerful surveillance system is a violation of Flock’s terms of service, which states “Authorized End Users shall not share their account username or password information and must protect the security of the username and password.”
More concerningly, it shows, as we have been reporting, that there are very few practical guardrails on how Flock is being used. The DEA does not have a contract with Flock, and police generally do not obtain a warrant to use Flock. We have repeatedly reported on police officers around the country who have offered to either run plates for their colleagues or to give them access to their logins, even when those agencies have not gone through proper acquisition channels.
The Palos Heights police department did not respond to a request for comment from 404 Media. The DEA told 404 Media “we respectfully refer you to the Palos Heights Police Department.” Flock also did not respond to a request for comment. The House Oversight Committee announced last week that it had launched an investigation into how Flock is being used to search for immigration violations.
ICE Taps into Nationwide AI-Enabled Camera Network, Data Shows
Flock's automatic license plate reader (ALPR) cameras are in more than 5,000 communities around the U.S. Local police are doing lookups in the nationwide system for ICE.Jason Koebler (404 Media)
Come celebrate with us and catch a LIVE recording of the 404 Media podcast.#party
You're Invited: 404 Media's Second Anniversary Party and LIVE PODCAST!
We've survived and thrived for two years and are ready to celebrate with you, the ones who made it possible!Come have a cocktail or locally-brewed beer on us at vertical farm and brew lab farm.one. We'll also record a live podcast with the whole 404 crew, for the first time in person together since... well, two years ago!
Doors open at 6, programming begins at 6:45, good hangs to continue after. Open bar (tip your bartenders), and pizza will be available for purchase on-site if you're hungry.
Free admission for 404 Media subscribers at the supporter level. Sign up or check your subscription here. Once you're a supporter, scroll to the bottom of this post for the code to enter at checkout on the Luma page. Or buy tix for yourself or a friend to make sure you have a spot on the list.
We'll also have some merch on hand that'll be discounted for IRL purchases.
If getting into the coolest party of the summer isn't enticing enough, you'll be supporting the impact of our journalism, which so far this year has included:
- Nvidia was sued after we revealed the company scraped YouTube and other sites en masse to build its own AI systems.
- GeoSpy, an AI tool that let anyone geolocate photos in seconds and whose users could include stalkers, dramatically restricted access after 404 Media published an investigation into it.
- Congress repeatedly grilled Apple and Meta over their association with nonconsensual nudify and deepfake apps after we exposed the connections.
- A public library ebook service said it was going to cull AI slop after we found low quality books were flooding libraries.
- Congress just launched an investigation into automatic license plate reader company Flock; After 404 Media’s investigations this year, Flock removed a number of states from its national lookup tool. In July, Senator Ron Wyden of Oregon announced an agreement by Flock to block any out-of-state police searches related to abortion or immigration.
Our earlier work has shut down surveillance companies and triggered hundreds of millions of dollars worth of fines too. Our paying subscribers are the engine that powers this impactful journalism. Every subscription, monthly or annual, makes a real difference and makes it possible to do our work.
Thank you to our friends at DeleteMe for making this celebration possible.
Fine print: Tickets are required for entry, including for subscribers. 21+ only. Seating for the podcast is open but limited and includes standing room; a ticket doesn't guarantee a seat but let staff onsite know if you require one. Photos will be taken at the event. Venue reserves the right to refuse entry. Good vibes only, see you soon!
Code for subscribers is below the images.
Scenes from our panel at SXSW 2025, our DIY hackerspace party in LA on July 30, and our first anniversary party last year.
Upgrade to continue reading
Become a paid member to get access to all premium content
Upgrade
Bravo Donald: sono gli americani a pagare l'86% dei rincari innescati dai dazi
Secondo un rapporto di Goldman Sachs gli esportatori stranieri hanno sopportato solo il 14% degli aumenti innescati dalle tariffe americane. Il resto le impres…Claudio Paudice (HuffPost Italia)
reshared this
A Gaza il “giornalisticidio” prosegue indisturbato
@Giornalismo e disordine informativo
articolo21.org/2025/08/a-gaza-…
Prosegue a Gaza il giornalisticidio, parte del genocidio in atto a Gaza. Sono ormai oltre 230 i cronisti assassinati, compresi i sei eliminati ieri. Contro di loro ê ripartita l’ennesima campagna “erano solo
Lezioni di conversazione in italiano
Uso spesso podcast e video di persone di madrelingua inglese per migliorare la conoscenza della lingua.
Mi piacerebbe restituire il favore.
Ho pensato che magari da qualche parte sul pianeta c'è qualcuno che studia italiano a cui potrebbe fare altrettanto comodo avere uno sparring partner, quindi non podcast e video ma vere conversazioni on-line (gratuite).
Non so da che parte partire per far arrivare la notizia a chi potrebbe essere interessato, voi come fareste?
like this
reshared this
«Non c’è nessun caro ombrellone»
ci vuole coraggio a definire i prezzi in italia non esosi e non un'emergenza... pazzesca la ghigna che ha la gente. fosse per me renderei obbligatoria una spiaggia libera accanto a ogni stabilimento a pagamento.
Public consultation on ”retention of data by service providers for criminal proceedings”. Answering guide for civil society organisations and individuals.
The European Commission has launched a public consultation to gather your views about the impact of data retention rules in view of adoption of legislative and non-legislative measures at EU level. Here's EDRi's guide on how to answer the tricky consultation.
The post Public consultation on ”retention of data by service providers for criminal proceedings”. Answering guide for civil society organisations and individuals. appeared first on European Digital Rights (EDRi).
L’Italia investe nelle startup tecnologiche israeliane
@Notizie dall'Italia e dal mondo
La Cassa Depositi e Prestiti destina decine di milioni di euro a imprese israeliane di intelligenza artificiale e calcolo quantistico. L’obiettivo è attrarre in Italia competenze e innovazione, dimenticando la distruzione di Gaza
L'articolo L’Italia investe nelle startup tecnologiche israeliane
Le fatiche di Eva: la strada ancora lunga verso la parità
@Giornalismo e disordine informativo
articolo21.org/2025/08/le-fati…
Unendo esperienza personale e dati di accreditati studi nazionali e internazionali, Paola Mascaro, manager e già presidente di Valore D, propone una lettura sul difficile cammino verso una parità di
Se il capitalismo perde il suo «spirito»*
@Giornalismo e disordine informativo
articolo21.org/2025/08/se-il-c…
Da qualche tempo quotidiani e libri si soffermano, più o meno approfonditamente, sul tema della transizione globale in atto. Si riflette sulla inusitata fase di “riproduzione del sistema” che sta avvolgendo il pianeta, con l’unica costante, rispetto al
CDN, la mossa Agcom riaccende lo spettro del fair share?
L'articolo proviene da #StartMag e viene ricondiviso sulla comunità Lemmy @Informatica (Italy e non Italy 😁)
L’Autorità sulle CDN ribadisce che non introdurrà network fee né interverrà sul mercato dell’interconnessione. Il quadro giuridico startmag.it/innovazione/cdn-la…
Arrestato in Colombia il narcotrafficante Federico Starnone, anche grazie alla rete anti-'Ndrangheta I-CAN di Interpol
Si tratta di un latitante 44enne, ricercato dalle autorità italiane per i reati di associazione a delinquere finalizzata al traffico internazionale di sostanze stupefacenti con le aggravanti connesse a due distinti tentativi di importazione di ingenti quantitativi di cocaina dal Sudamerica.
E' ritenuto legato alla 'Ndrangheta. E' stato catturato in un appartamento nel quartiere residenziale di Cali.
A carico di Starnone è stata già emessa una sentenza di condanna a 5 anni e mezzo per reati di droga. L'uomo è stato catturato dalla polizia colombiana mentre si trovava in un appartamento nel quartiere residenziale nel capoluogo del dipartimento Valle del Cauca.
Essenziale l'apporto del progetto INTERPOL Cooperation Against ‘Ndrangheta (I-CAN).
Si tratta di un'iniziativa lanciata dall'Italia e dall'INTERPOL nel gennaio 2020 per contrastare la minaccia globale rappresentata dalla ‘Ndrangheta, come noto un'organizzazione criminale transnazionale altamente organizzata e potente.
Finanziato dal Dipartimento della Pubblica Sicurezza italiano, il progetto mira a rafforzare la cooperazione internazionale tra forze di polizia sfruttando le capacità dell'INTERPOL di condividere intelligence, competenze e best practice, trasformando così le informazioni in arresti e smantellando le reti criminali.
Avviato a Reggio Calabria l'obiettivo principale del progetto è stato - da subito - quello di istituire un sistema globale di allerta precoce contro questo "nemico invisibile". I-CAN opera attraverso una rete di paesi pilota, che inizialmente includevano Australia, Argentina, Brasile, Canada, Colombia, Francia, Germania, Italia, Svizzera, Stati Uniti e Uruguay, che si è espanso a 13, tra cui Austria, Belgio e Spagna.
Il progetto facilita operazioni coordinate transfrontaliere, come dimostrato dall'operazione globale del 2020 che ha portato all'arresto di sei latitanti legati alla 'Ndrangheta in Albania, Argentina e Costa Rica, con conseguente sequestro di 400 kg di cocaina e smantellamento del clan Bellocco. Le operazioni successive hanno continuato a dare risultati, tra cui l'arresto nel 2023 di un latitante di 16 anni, Edgardo Greco, in Francia, con il supporto di I-CAN.
Il progetto si è evoluto oltre la sua fase iniziale, con iniziative in corso tra cui la Conferenza I-CAN del 2022 a Roma, che ha riunito le forze dell'ordine di 14 paesi per definire una strategia unitaria contro la 'Ndrangheta, oggi considerata un'entità criminale "silenziosa e pervasiva" che si infiltra nelle economie legittime attraverso la corruzione e il riciclaggio di denaro.
Il successo del progetto si basa su una combinazione di condivisione di intelligence, coordinamento internazionale e utilizzo di strumenti analitici avanzati per esplorare dati provenienti da diverse fonti, consentendo indagini transnazionali. Il suo quadro continua a sostenere gli sforzi in corso, tra cui il progetto I-FORCE, incentrato sulla cooperazione regionale nell'Europa orientale e sudorientale.
#ndrangheta #ican #interpol #iforce
reshared this
#Trump e #Putin al caminetto
Trump e Putin al caminetto
L’appuntamento è per il 15 Agosto in Alaska e sarà il primo del secondo mandato di Trump alla Casa Bianca. Il fatto che vi sarà un incontro è di per sé un fatto positivo, quando le due superpotenze nucleari dialogano il mondo intero respira meglio.www.altrenotizie.org
Il ritorno dell’autorizzazione a procedere rafforzerà le tutele degli eletti nelle Camere
@Politica interna, europea e internazionale
L'articolo Il ritorno dell’autorizzazione a procedere rafforzerà le tutele degli eletti nelle Camere fondazioneluigieinaudi.it/il-r…
Anduril accelera nell’Indo-Pacifico. Nuovi accordi con Taiwan e Corea del Sud
@Notizie dall'Italia e dal mondo
Dopo America ed Europa, Anduril ha messo gli occhi anche sull’Indo-Pacifico. L’emerging tech Usa, leader nel comparto dell’IA applicata alla difesa, ha recentemente concluso due accordi che pongono le basi della sua futura espansione in estremo oriente. Da un lato
The human voiceover artists behind AI voices are grappling with the choice to embrace the gigs and earn a living, or pass on potentially life-changing opportunities from Big Tech.#AI #voiceovers
Voiceover Artists Weigh the 'Faustian Bargain' of Lending Their Talents to AI
Acting is an industry of feast and famine, where performers’ income can swing widely by role, by month, and by year. It’s a field where people often face the choice between passion, creativity, and taking a commercial gig for a check. As with so much else, this delicate personal calculation is now being disrupted by AI.Last month, online actors’ jobs boards were flooded with a very specific, very well-paid role. Nestled between student short film gigs and callouts for background dancers, was the ambiguously-named opportunity “Technology Company AI Project.” According to the job listing on cast and crew job board Mandy, it would pay up to $80,000, for only 19 total hours of work. This is unusually high for an industry where a national-level ad campaign for a big brand might pay $6,000.
The post was from voice acting talent agency Voice123, casting on behalf of a project by Microsoft. According to the listing, the company was looking for voice actors across 19 languages, with specific regional dialects and accents including “French from France native” and “Arabic as spoken by Palestinian/Israeli Arab communities.”
“I get instant notifications, and I was getting so many of them,” said Katie Clark Gray, a podcaster and voice actor. The rate stood out to her. “The jobs that I tend to see are, like, £250 [about $339 USD]... it was, like, a lot of posts. The money seemed like a lot.” She said that it’s rare to get that many notifications for a recognizable brand.
The role would include recording “conversations, character voices, and natural speech to help train AI systems,” Crispin Alfario, a recruiter for the role on the Voice123 platform, told 404 Media. Alfario could not comment further due to privacy terms, but said there was “a positive response during the castings for these projects.” Clark Gray said that advertised AI roles like this are increasing in scope and in scale, and that she now sees far fewer roles available for employee training video work or industrial roles like phone menu voices — the area she got her start in over a decade ago.
She sees accepting AI training voiceover roles as something of a Faustian bargain: They might seem like a lot of money, but they reduce the amount of work available in the future. “You're still taking away tomorrow's meal because they're offering you a little bit more,” she said. “Those 19 hours… will scale to hundreds and thousands of hours of AI output. They would otherwise have to pay for it.”
0:00
/2:10
1×Katie Clark Gray practicing takes for a voiceover script.
I called Microsoft’s PR to ask if I could chat to someone involved in casting for the roles that Clark Gray had spotted, on the same day that Microsoft CEO Satya Nadella published a note about the “recent job eliminations” of four percent of staff and pledged to “reimagine every layer of the tech stack for AI.” The next day, less than two weeks after Clark Gray spotted the Microsoft ads, the company announced a new virtual character for Copilot, the trial version of which is currently only available in English. After that announcement, a Microsoft spokesperson confirmed to me that the voiceover roles I asked about were for Copilot Voice, and that they will “continue to look for more talent as [they] expand these capabilities.” I hadn’t been sure that the audition posts were linked to Copilot, but the confirmation from Microsoft confirmed that the posts that Clark Gray had spotted had been in advance of the product announcement.
“More and more I'm seeing AI disclaimers that, by auditioning for this, you agree to have your voice and likeness used and replicated. I hate that.”
Hunter Saling, an actor and comic based in LA, said he’s seeing more and more roles which have an AI component or require signing an AI waiver. He auditioned for a “Siri-type AI assistant,” in May. The role would have paid an amount of money where he “wouldn’t need a job” for a long time.“You'd be providing a whole bunch of stuff up front,” he said, “and then be paid as a performer, as a voiceover artist, to come back on a yearly basis to do more stuff.”
0:00
/0:40
1×Hunter Saling practicing takes for a voiceover script.
I wondered if this was another situation where an audition was the first public hint of a product launch in the space, but Saling couldn’t tell me the company he’d auditioned for, due to confidentiality. I kept an eye out for new Siri-type AI agents that might be able to pay life-changing money and, while I was writing this story, on July 17, OpenAI launched their ChatGPT agent—a Siri-type AI assistant. OpenAI is also known to use Mercor, an AI-enabled recruitment platform, which was recently posting about voice casting for a “top AI Lab.”
The AI-assistant voice audition process was very different from usual, Saling said. He described the voice he did as “the performance of no performance;” a voice that was “not personality free, but, like, neutral, but friendly and helpful.” He describes the work he did on the audition as “not children's host, but also not robotic either… I read a story, some recipe directions, and some just general sentences.”
On August 7, OpenAI announced ChatGPT 5 which would have several new personalities, but the company said that those personalities would not apply to voice mode.
Being selected for this kind of windfall could alter the course of an actor’s life.
One part of the audition script stood out to Saling: He was asked to “affirm” someone. “That did start to send me on a bit of a mental spiral of, oh, my God, someone needs affirmation from their home assistant.”
Auditioning for this role also posed an ethical question. “I will say I was surprised in myself that I was OK doing this,” he said. “More and more I'm seeing AI disclaimers that, by auditioning for this, you agree to have your voice and likeness used and replicated. I hate that.”
The last couple of years have seen the entertainment industry in turmoil over the use of AI in screen and voiceover work. Both the four month SAG-AFTRA actor’s strike in 2023, as well as their almost year-long video games strike, which ended last month, focused on the use of AI. The agreements which ended the strikes describe different industry categories of AI use, differentiating between the kind of AI which digitally alters or replicates the work of a particular actor, and generative AI which is trained using actor’s work or creates a “synthetic performer.”
playlist.megaphone.fm?p=TBIEA2…
Saling does agree with this technical difference, between delivering an artistic or creative performance that can be altered, perfected, or smoothed out later, and providing a voice to be re-created for industrial use, like in an AI assistant. Creating the neutral voice of an AI assistant, to be generatively replicated, is industrial, rather than artistic; “this is something that... it's not a performance, it's not a character. It's a tool,” he said.Clark Gray is not financially dependent on her voice acting career, and her calculus in auditioning is different. She didn’t submit for the Microsoft role, but “wouldn't fault anybody for going out for that job,” she said. “That’s a year’s salary for a lot of people.” But she also feels a difference in applying for creative voiceover roles vs industrial ones; “I think the cartoon voices are much more fun. I don't know anybody who doesn't,” she said. “You do bring a sort of artistic, like, extra sauce to it. Creating a character really does take something different than reading something in a neutral voice.”
Saling said that he thinks the adoption of AI taps into the entertainment industry’s commercially-driven but counterproductive desire to create mass appeal via synthetic perfection. “Sometimes I feel like Lear yelling at a storm on the fucking cliff,” he added — with a theatricality ChatGPT could only dream of.
Mercor, an AI recruiting startup founded by 21-year-olds, raises $100M at $2B valuation | TechCrunch
Mercor, the AI recruiting startup founded by three 21-year-old Thiel Fellows, has raised $100 million in a Series B round, the company confirmed toTage Kene-Okafor (TechCrunch)
The OverDrive is made to let ground vehicles navigate tough terrain with minimal input from humans.#military #AIbots
The U.S. Army Is Testing AI Controlled Ground Drones Near a Border with Russia
The U.S. Army tested a fully AI controlled ground vehicle in Vaziani, Georgia—about 100 miles from the Russian border—last month as part of a training exercise. In military-published footage, an all wheel, off-road vehicle about the size of a car called ULTRA navigated the European terrain with ease. The training exercise had the ULTRA resupplying soldiers, but both the military and the machine’s creator think it could do much more.
playlist.megaphone.fm?p=TBIEA2…
The Pentagon has invested in drones and AI for decades, long claiming that both are the future of war. The appearance of the ULTRA signals a time when AI controlled robots will populate the battlefields of the near future.“ULTRA was built to be modular and mission-adaptable from the start,” Chris Merz, an employee of Overland AI, the company behind ULTRA said according to an Army press release. “We are actively developing variants that support casualty evacuation, counter-unmanned aircraft systems, and terrain shaping operations.”
youtube.com/embed/OwxPodELAQA?…
ULTRA runs on Overland AI’s proprietary OverDrive software, a system that’s designed to give AI full control over ground vehicles on the battlefield. Overland AI did not return 404 Media’s request for comment, but its website claims it can retrofit OverDrive onto traditional vehicles and its YouTube page has a video claiming to show the AI piloting a Ripsaw M5 tank.
youtube.com/embed/H8D7AtW1Lqo?…
Overland AI is a Seattle based company that started in 2022. It’s gained a lot of buzz in the last few years as a pioneer of AI software meant to control unmanned ground vehicles (UGVs). Jon Fink, Overland AI’s CTO, explained how its software worked during a presentation at a defense tech showcase earlier this year.During the demo, Fink showed footage of a field test where an ATV navigated hazardous terrain with minimal input from a human. Fink said the company’s OverDrive software is “purpose built for the warfighter. It’s built in order to enable the operator so it can remotely task a system so it can autonomously move through an environment without reliance on detailed maps or communication back with that operator.”
The big challenge of AI systems like this is that they need to be able to navigate the terrain on their own without looking at a map. GPS is often jammed or unavailable on the battlefield. So a robot will need to use cameras and other sensors to make decisions about how to move through a warzone in real time. In the video, the operator drops a few waypoints on a map of the area and clicks a button to launch the ATV. “Note while we’re specifying all these tasks, I’m not like zooming in, looking very close at detailed information that I might have from a satellite, because I can’t necessarily trust that,” Fink said. Satellite imagery can become outdated quickly on a chaotic battlefield.“I’m really giving the system just a coarse idea of what I need it to do to accomplish my mission.”
youtube.com/embed/8TFoDcG1Z0U?…
The U.S. Army’s test last month has been a long time coming. “This isn’t new,” Samuel Bendett, a drone expert at the Center for Strategic and International Studies, told 404 Media. “This has been in development for many, many years […] this is at least a decade’s worth of research, development, testing, and evaluation of different levels of autonomy with different technologies.”Russia, China, and the United States are all working on AI controlled ground vehicles. Drones require an operator which means a human being needs to maintain contact with a device over vast distances. That’s easier to do when the machine is a robot flying through the sky, but ground vehicles have to contend with signal-blocking debris and are easier targets for ground troops.
“Communication between the UGV operator and the operator can be jammed if it’s radio, the communication can be severed if it’s done via cable, communication may be endangered if it’s an aerial drone that’s trying to provide signal strength and overwatch capabilities,” Bendett said. “Operators have to be in relative proximity to their UGVs, and that, of course, somewhat negates the point of using UGVs instead of people. If people are close to their UGV, they can be discovered and killed.”
AI answers a lot of these problems. If an operator can give a set of simple instructions to a machine and let it operate independently, then it need not be in constant contact. In his presentation earlier this year, Fink noted that the AI controlled ATV adjusted its speed as it navigated terrain, all on its own. “We haven’t set any sort of speed limits or specifications to the system when we tasked it, we basically just told it: ‘Go to these general locations’ and it’s taking care of all of the decisions as it needs to,” he said.
There are major concerns about warfighter machines making decisions by themselves. UN Secretary-General António Guterres has called for a ban on autonomous weapon systems, calling them “morally repugnant.” In Gaza, Israel is using AI models from OpenAI and Microsoft to make targeting decisions and Israeli intelligence officers have told reporters that information provided by the AIs were treated “as if it were a human decision.”
Right now, Overland AI’s OverDrive AI stack is just for helping a ground vehicle navigate, and Bendett said it’s ahead of the pack. “The Holy Grail of autonomy is translating that intuitive human experience into a UGV that will be able to navigate, on its own, through rough terrain, mixed terrain, uncertain terrain, which is what we’re seeing with Overland UGVs,” he said.
What could this thing be used for? “The number one goal for these kinds of UGVs is logistics and supplies,” Bendett said. “Medical evacuation is becoming a growing concern and UGVs are also used for that.”
It, of course, won’t stop there. “UGVs used in combat can be mounted with all manners of weapons,” Bendett said.
The U.S. Army did not respond to 404 Media’s request for comment.
‘Lavender’: The AI machine directing Israel’s bombing spree in Gaza
The Israeli army has marked tens of thousands of Gazans as suspects for assassination, using an AI targeting system with little human oversight and a permissive policy for casualties, +972 and Local Call reveal.Amjad Iraqi (+972 Magazine)
QUANDO L’INTELLIGENZA ARTIFICIALE INIZIA A GIOCARE CON LA CYBERSICUREZZA…
@Informatica (Italy e non Italy 😁)
Mai sentito parlare di RunSybil? Si tratta di una startup fondata da chi era il primo ricercatore sulla sicurezza di OpenAI...
L'articolo QUANDO L’INTELLIGENZA ARTIFICIALE INIZIA A GIOCARE CON LA CYBERSICUREZZA… proviene da GIANO NEWS.
#TECNOLOGIA
Ministero dell'Istruzione
#NoiSiamoLeScuole, con i fondi del #PNRR finalizzati alla costruzione di nuove scuole sono stati demoliti e sono in corso di ricostruzione due istituti in Sicilia e uno in Lombardia.Telegram
alephoto85
in reply to Max 🇪🇺🇮🇹 • • •ricordo app, forse "Hello Talk" e "Tandem", che potrebbero essere utili in questo caso ma non mi viene in mente nessuna alternativa open.
Se invece ti va bene anche offline, al mio circolo arci di fiducia, organizzano lezioni di italiano per stranierз, giusto per fare un esempio.
Magari prova a vedere se nella tua zona ci sono progetti simili. Credo ci sia sempre bisogno di volontarз.
Max 🇪🇺🇮🇹
in reply to alephoto85 • •@alephoto85
Sì in effetti è una buona idea, non ci avevo pensato. Grazie.
Anche se a me piacerebbe di più farlo con gente che sta dall'altra parte del mondo, così potrei approfittarne per farmi raccontare qualcosa di come vivono laggiù 😀
alephoto85
in reply to Max 🇪🇺🇮🇹 • • •capisco! Ci sta effettivamente! Sono sicuro però che anche chi arriva qui da lontano avrà qualcosa da raccontare in merito.
Se trovo altre cose online ti scrivo ma adesso mi vengono in mente solo quelle 😅