The robotaxis did not need a driver, but Waymo has ended up paying delivery drivers to close ajar doors

What until not long ago seemed the exclusive province of science fiction is beginning to become visible on the streets: cars capable of moving from one point to another without a driver. And you don’t need to buy one to live the experience. In some cities around the world, it is enough to order a robotaxi from an application and see how the vehicle arrives to pick you up, identifying you in certain models with your initials on an LED screen located on the roof, as our colleague Javier Lacort confirmed in San Francisco almost two years ago. Futuristic scene, present problems. In the midst of this transformation of transportation, which aims to offer more safety and comfort, its weak points are also beginning to emerge. We don’t talk about the jams caused by connectivity failures nor of those cars that, for some reason, They start honking their horn at four in the morning. The issue is even more basic: if a user closes the door incorrectly, the vehicle cannot continue operating. The problem is not driving, it is being able to leave. In the case described by CNBC and TechCrunchare blocked if, at the end of the trip, a passenger leaves a door ajar. Waymo confirmed to both media that this detail prevents the car from resuming travel and completing new routes until someone closes it correctly. This is a basic, almost domestic friction that turns a simple oversight into an operational problem and explains why the company has to resort to human support to return its vehicles to service as soon as possible. Pay delivery people. The company is testing a system in Atlanta that alerts nearby delivery drivers of applications such as DoorDash when one of their vehicles is left with the door open. The proposal is simple: approach, close it and allow the robotaxi to operate again. The media even cites the case of a driver who was offered $11.25 for that specific task. They also detail a similar order divided between $6.25 for travel and another $5 after verifying the closure. It is not an isolated case. The Atlanta pilot is not the only example of this specific dependence on human help. Waymo has also turned to users of honka roadside assistance platform, to resolve similar situations in other American cities. In this case, some collaborators received offers of up to $24 to close the door of a stopped robotaxi. More than a local anecdote, these examples draw a clear operational pattern: when the vehicle is immobilized due to a minor detail, the quickest solution is still to send a person. Automatic doors, on the way. Today Waymo operates with a fleet made up entirely of electric vehicles Jaguar I-PACE adapted for autonomous driving, which still depend on human intervention in situations like this. But the company owned by Google assures that this gap has an expiration date, although without specifying it: it announced that its future robotaxis will have automatic closing. Meanwhile, the present of the autonomous car continues to show that double face: sophistication in driving and human dependence on the simplest details. Images | Xataka In Xataka | When San Francisco suffered a blackout, its streets were plunged into chaos for a reason: dumped self-driving cars

ended up giving away $40 billion in bitcoins

One of the largest houses in cryptocurrency exchange from South Korea wanted to reward its users with a symbolic promotion for their operations. However, a mistake has made the promotion of the company on everyone’s lips today…and not exactly for the better. For a few minutes, several hundred customers Bithumb They saw their accounts filled with bitcoins worth several billion dollars. What should have been a small promotional prize turned into a mistake that, on his screen, became billionaires to normal users. A conversion error. Bithumb’s original idea was to offer a reward of 2,000 won (approximately $1.37 in exchange) to users who participated in a company promotional event. The equivalent of a welcome coupon for newcomers. The problem came when, instead of sending that small amount in won, the system ended up sending bitcoins to the accounts of the new clients. According to published the BBCthe failure occurred when an employee entered the indicator “BTC” in the payment field instead of “Korean won”, so the platform executed the reward in cryptocurrencies instead of local currency. That simple misplaced piece of information led the company to mistakenly transfer some 620,000 bitcoins, a figure that, at current prices, is around $44 billion. A mistake that destabilized the market. Bithumb estimated that about 249 users of its platform received bitcoins by mistake and the failure affected about 695 clients who operated on the platform. It is estimated+ that, on average, each user was assigned about 2,490 bitcoins, which represents a value of around 144 million euros. Seeing the new balance of bitcoins in their account, several of these new “accident millionaires” rushed to sell, generating an avalanche of orders that caused the price of bitcoin to fall within Bithumb itself. about 10% in a matter of minutes. Bithumb hit the panic button. When the company realized the error, it began to apply restrictions to affected clients, temporarily limiting operations and withdrawals to stop the leak of funds. In its assessment of the incident, Bithumb assures which managed to recover approximately 99.7% of the 620,000 bitcoins that were left by mistake, which would leave about 125 bitcoins still pending recovery. The company also points out that what has already been recovered includes some 1,663 bitcoins that users they managed to sell before the platform’s “panic button” was pressed, which activated transaction blocks. Lee Jae-won, the company’s executive director, assured that the company will take the incident as a lesson and will prioritize “customer trust and peace of mind” over external growth. “Paper” Bitcoin. The case has reopened the debate about the so-called “paper bitcoin”, in reference to those transactions that exist within the internal systems of the exchanges but do not always have the real assets that support them behind them. The sum of bitcoins that suddenly appeared in the accounts far exceeds the $5.3 billion in bitcoin assets that Bithumb claims to be in custody, making it clear to what extent much of that “wealth” was only on paper in its internal books. It’s not the first time it happens. It is not the first time that a banking or financial entity makes its users millionaires in a “magical” way. He Financial Times counted a few days ago how Citibank made one of its clients a billionaire by transfer 81 billion dollars when he intended to send him a payment of $280. As happened with the South Korean bitcoin exchange, the bank realized the error and fixed it (unfortunately for the user) in 90 minutes. However, the simple fact that a human error when indicating a figure or inserting the type of currency can shake the entire bitcoin market has set off alarms in the South Korean Financial Supervisory Service, which has announced reviews and does not rule out opening formal investigations if they detect serious failures in internal controls or signs of illegal activity. In Xataka | Cryptocurrencies were supposed to become “independent” from the power of states. The US just killed him with a stroke of the pen Image | Unsplash (Michael Fortsch)

The most brutal rains in the history of Andalusia have already ended. Now the real problems begin

The storm Leonardo little by little begins to fade from the maps, leaving in its wake mainly alerts for strong gusts of wind in certain regions of Andalusia. The problem is that its footprint on the ground is just beginning to show its true dimension, since the main danger is that even if rainfall begins to decrease, the water continues to rise in the rivers. And this gives rise to the feared floods that are already has caused numerous evictions. Extreme saturation. To understand why authorities and the AEMET maintain the emergency level 2 and red warnings despite lulls in rainfall, we have to look under our feet. The soil functions, under normal conditions, like a sponge capable of retaining large volumes of water. However, after weeks of constant rainfall, Andalusia has reached its saturation point. In this way, the land does not support any more water, which increases the runoff coefficient throughout the territory. This means that each new liter that falls, no matter how small, will barely filter through the ground. The result is that it will run on the surface, turning slopes and mountains into giant slides towards the rivers. Increase of the channel. This is the reason why 14 rivers are under red notice today and another 31 under orange. Rivers such as the Guadalete, the Genil, the Guadiaro and the Guadalhorce They are not just responding to today’s rain.but to the inability of the basin to drain what has accumulated in the last 48 hours. We have an example in Huétor Tájar in Granadawhere the Genil River overflowed, making the entire town become a large lake. And this is the main risk we face despite the fact that rainfall is beginning to reduce its intensity. The reservoirs. The other major front of this crisis is hydraulic engineering. The reservoirs act as buffers during floodsretaining the water to prevent it from devastating the towns downstream. But Leonardo has managed to finish filling these reservoirs to their maximum limits. This has forced us to initiate technical releases with increasing amounts of water to avoid breakages or uncontrolled overflows of the dams. The problem is that doing so injects more flow into rivers that are already at the limit of their capacity, keeping towns like Ubrique or the lower areas of the Guadalquivir in suspense. Sierra Nevada. Gravity in the Genil basin is not based solely on precipitation, but on thermodynamics. Leonardo is not a cold storm of polar origin, but rather an Atlantic storm loaded with humidity that is causing snow accumulated in previous weeks melts at a high speed. The result here is clear: a greater flow in the rivers that drain the Sierra that joins all the factors that we have mentioned before. Landslides. For the next few hours, in addition to the increase in the riverbed, we must also keep in mind the risk of hillslide. In these cases, water saturation increases the weight of the soil and reduces its internal friction. This translates into a greater risk of landslides on roads and slopes, something that can especially occur in mountain areas such as Cádiz or Axarquía in Malaga. More rain on Saturday. Faced with overflowing soil, the last thing you want is to receive more rain. But the reality is that this same Saturday a new storm comes in that has already activated an orange alert in a region that has been greatly punished by Leonardo such as Grazalema. In this case, accumulations of up to 80 liters per square meter are again expected, which may further aggravate the situation that is being experienced. Images | Ted Balmer In Xataka | We have always believed that London is very rainy and that Barcelona is not. The only problem is that it’s a lie

Madrid has bought so many electric cars that the DGT has ended one of its great incentives

Electric cars and plug-in hybrids will not be able to circulate in the Bus-HOV lane unless the signs indicate so. The DGT has confirmed that it was one of the most attractive measures for the potential customer of a car with a Zero Emissions label to take the leap. Now, so many cars of this type have been sold in Madrid that they have ended up putting an end to this advantage. What has happened? The DGT has sent a statement announcing the “Resolution on special traffic regulation measures for 2026.” Nothing very juicy except for one detail: the announcement that the Zero Emission cars they have run out of taking advantage the Bus-HOV lane to avoid traffic jams. The DGT explains that from now on, drivers of a Zero Emissions car (electric or plug-in hybrid with more than 40 kilometers of autonomy) will only be able to circulate on this special lane when it is specifically signposted. By default, they will not be able to enter it. Because? According to the DGT, the decision “responds both to the demand of the citizens and to the requests of the public transport companies and the Ombudsman who have conveyed to the DGT their concern about the progressive loss of effectiveness of the HOV lanes that directly affects the regularity and punctuality of the service, discouraging its use and harming thousands of daily users who opt for public transport.” And they provide data: traffic jams on the main roads have increased by 10%, while in the Bus-HOV lanes they have increased by 22%. But the data skyrockets in Madrid. According to their accounts, traffic jams are 20% more frequent on the main road of the A-6 entering and exiting Madrid. In its Bus-HOV lane, traffic jams have increased by 90%. Madrid, absolute leader. According to ANFAC data, Madrid was the Autonomous Community where the most electrified cars (electric and plug-in hybrids because the data also discriminate by non-plug-in hybrids) were purchased. In total, at the end of 2025, 102,245 cars of this type were recorded. Across Spain, 245,629 Zero Emission cars were purchased. The next region in which the most Zero Emission cars were purchased was Catalonia but it remained at 33,309 units. Behind them, only the Valencian Community and Andalusia exceeded 20,000 units. Goodbye to one of the great incentives. Until now, switching to the Bus-HOV lane despite only having one passenger traveling in an electric or plug-in hybrid car was one of the great incentives to get a vehicle of this type. The HOV Bus on the A-6 in Madrid, the only one for which the DGT offers data, is a relief for a road that is clogged daily. Beyond the driving comfort (absence of noise or vibrations) and the savings if we recharge at home, the Zero Emissions cars had two great incentives that were considered “political”. One is the purchasing aid that until now was collected in the MOVES III Plan but that have been frozen waiting for a Auto+ Plan that has not yet materialized. The second was this use of the Bus-HOV lane, since the time saved per day was considerable. However, advantages applied by each municipality such as unlimited access to ZBEsexemptions in the payment of road taxes or free parking in regulated parking areas. These aids are of municipal application and, therefore, vary from one city to another. Goodbye, goodbye. The loss of the unlimited pass for the VAO Bus is only a reminder that Zero Emission cars continue to enjoy some aid that, it is hoped, will end up disappearing. This is what has happened, for example, in Norway, where the exemption from paying taxes has caused a hole of 1.8 billion euros. The solution that has been proposed is to tax the weight of vehicles to alleviate this problem. In other cities, like parisit is also ignored whether the car is electric or not and a similar mechanism is also used to charge in regulated parking areas. Photo | DGT In Xataka | Guide to know if your car will be able to circulate in the ZBEs of Madrid in 2026: labels, registrations and areas

Toyota was obsessed with creating its best electric sedan. So he ended up asking Huawei for help…

After letting it be seen in the Shanghai Auto Show At the beginning of last year, Toyota just made official the bZ7its electric flagship more than five meters long. What is striking here is not the car itself, that too, but the technology that gives life to both its software and its drive train. Technology that… is not from Toyota. The car. bZ7, this is the name that Toyota has given to an electric sedan that embodies the latest technology available for this segment. The summary is simple. 5.1 meters long. 1.9 meters wide. LPF (lithium ferrophosphate) type batteries of BYD origin. Autonomy of between 600 and 800 km (according to the Chinese cycle, CLTC) depending on version. Operating system HarmonyOS. Huawei DriveOne system (electrical system, engine, car architecture…) What’s Huawei looking like here?. In 2020, Huawei confirmed its commitment to the electric car with DriveOneits first electric motor. Specifically, we are talking about a control unit composed of a motor, reducer, converter, integrated charger, power distribution unit and battery control unit. It thus allows this Toyota bZ7 to have a power of 278 HP and a maximum speed limited to 180 km/h. All this in a much more compact platform compared to the traditional ones used in this type of vehicles. The interior. As if it were not enough to power the engine of this luxury sedan, the cabin has a 15.6-inch floating central screen. The size of a generous laptop. The operating system that gives it life is HarmonyOS, a platform thatthe company develops for the world of electric cars, smartphones, tablets, computers and peripherals of all kinds. The alliance. That Huawei and Toyota develop a car together is something quite recent. The Japanese company announced that, on cars destined for China, it would cooperate with Huawei. Toyota began to lose steam both in global sales and in China, where it fell 6.9% in 2024. After three consecutive years of losing sales in China, it decided with one of the manufacturers that today has more muscle when it comes to developing complete platforms for electric cars. Beyond Apple and Google. Chinese manufacturers like Huawei are betting on a solution at the operating system level that is much more integrated than what Apple and Google have been trying to do for years. Unlike Android Auto and Apple CarPlay, integrate the system (in this case HarmonyOS, but also in cases like that of HyperOS), allows complete control of the vehicle from it. It does not require a smartphone, it is updated via OTA, it is based on its own code… It’s something similar to what Google is trying with Android Automotivea complete system but with little adoption, and what Apple promised with CarPlay Ultracurrently reserved only for luxury vehicles. Image | toyota In Xataka | In the midst of the industry crisis, the brand that has most opposed the electric car continues to break records: Toyota

In medieval Europe, not only humans ended up on the gallows. Other criminals were also executed: the “murderer” pigs

For centuries, medieval Europe It was a place where justice was dispensed not only in the courts, but in the squares, in full view of everyone, with public rituals designed to repair order when someone broke it in an intolerable way. At that time, the fear of the unforeseeable did not come only from armies, plagues or famines, but also from what moved through the streets and corrals. In the France Medieval times, for example, the public ritual of punishment (carriage amidst mockery, solemn sentence and execution before the community) did not always have a human as the protagonist: sometimes, the condemned was a pig. The image, which today seems like an oddity from a black chronicle or a folkloric exaggeration, was real enough to leave repeated documentary traces: animals led as prisoners, hung upside down until they died and treated, in practice, as perpetrators responsible for a crime that had broken the social balance. The pig as a real threat The frequency of these cases is better understood by remembering that the medieval world lived attached to animals and their risks. Pigs, in particular, were useful because they ate everything and could feed on scraps, but that same omnivorous condition made them dangerous if they roamed free near small children. The records collect numerous episodes in which pigs killed and even devoured children, a violence that today clashes with the modern image of the docile and slow animal, but which was then associated with specimens closest to the wild boar: fast, strong and capable of imposing themselves physically in seconds. Medieval archives collect cases like the one from 1379when a group of pigs in Saint-Marcel-lès-Jussey killed the son of a swineherd, or the from 1386 in FalaiseNormandy, where a sow destroyed a child who ended up dying. Also that of 1457 in Savigny, Burgundywhen little Jehan Martin was killed by a sow and, especially disturbingly, his six piglets were found nearby, stained with blood. They were not vague rumors, but stories that were fixed with names and placesand that fueled the need for a public response that was not limited to a simple private loss. In France, these events often led to in judicial proceedings formalities in which the animal was imprisoned, transferred and executed as if it were a common criminal. Sources talk about expenses registered normally (cart, prison, executioner even brought from Paris) and an administrative routine that suggests that, for the people of that time, it was not an absurd spectacle, but a legitimate mechanism of justice. The strangeness, therefore, was not that there was violence, but rather that the violence was channeled through a trial with the appearance of ordinary procedure. When money is not enough A practical explanation of these processes was that medieval justice tended to seek reconciliation between partiesand many disputes could be resolved with compensation or agreements. But when a child death came into the picture, that logic was broken: the damage was too serious and the money could be insufficient to close the social wound. In that context, the court intervened to “take control” of the conflict, separate it from private revenge and offer an institutional solution that would distribute the emotional and political burden of the outcome. Trials also functioned as a form of organize the story: It was not just about punishing the animal, but about clarifying human responsibilities. If a pig was known for being dangerouswhy was he allowed to loiter near children? Was there negligence on the part of the owner? a chain of negligence? There was even a suggestion of the possibility of darkest questions: if the child was “unwanted”, if he or she was deliberately left in a risky situation or if the accident hid an intention. The court, by intervening, not only imposed a penalty, it produced an official explanation that the community could accept. Sometimes, the local machinery was not the last word and the matter escalated towards higher authorities. In the case of 1379, some of the accused pigs belonged to an abbey, and from there a petition was sent to Duke Philip “the Bold” requesting clemency. They defended that their animals had not participated and that they were “well-behaved pigs.” The duke heeded the request and issued a pardon for the animals of the abbey, showing that these processes, strange as they may seem, were inserted in real networks of power, influences and political decisions. Far from being simple superstition or peasant rage, these executions could serve to assert authority. The right to erect a gallows and execute criminals it was a privilegeand taking a case to the end allowed a local lord to exhibit the ability to punish and control order. There are episodes that reinforce that reading: a pig murderer from the 15th century it remained imprisoned five years before being executed, and formal letters were sent for permission to build a gallows. When the duke finally agreed, the triumph was not only symbolic: in addition to showing power, the lord stopped carrying the practical cost of keeping the animal imprisoned and feeding it. Plus: another key is the medieval vision of reality as a logical system created by godwith animals destined to serve humans. For a pig to devour a child was an unbearable investment of that order, a rupture of hierarchies that demanded public reparation. In that mental framework, the trial and execution were not theater: they were a way of “putting back together” what had been broken, of affirming that the world still had rules and that chaos, even when it came from an animal, could be put back into place by a solemn act of justice. Image | Ernest Figueras, Zoe Clarke In Xataka | The Middle Ages were not as dark as they told us In Xataka | 900 years ago, Europe had its own Manhattan: the impressive skyscrapers of more than 100 meters of Bologna

74% of employees have felt more productive when using AI. Almost half have ended up correcting the result

Artificial intelligence is already part of the daily life of the employees of many Spanish companies and helps them complete tasks faster. At least that is what emerges from a recent study by the AI ​​consultancy Workday, in which it is estimated that three out of four workers feel more productive thanks to AI. Behind that data there is a growing adoption of AI tools and a change in perception among professionals. However, this reality also implies a less visible one: part of that time gained you are missing out on reviewingcorrect and fine-tune what AI systems generate. Everyday use of AI in Spain. According to the data collected in the report “Beyond productivity: measuring the real value of AI” prepared by Workday, 74% of workers in Spain indicate they feel more productive thanks to AI, with 28% using it daily or 58% claiming to use it very often during their work week. That frequency of use of AIHowever, it is well below the global average which reflects a daily use of 46%. In any case, the increase in the use of AI translates into an average of time savings of between one and three hours per week for repetitive and administrative tasks, such as writing reports, analyzing or searching for data. ​These data coincide with the photo that the study of Indicators of use of Artificial Intelligence in Spain of 2024 prepared by ONTSI (National Observatory of Technology and Society), although in that case the perception is positive, only 11.4% of Spanish companies with 10 or more employees used AI technologies, which is revealed by a very limited business implementation. In any case, 85% of the users consulted report savings of between 1 and 7 hours per week. ​The problem of constant revisions. Satisfaction with the use of AI has the counterpart that 42% of Spanish workers dedicate up to one hour per week to review, correct or reformulate the result produced by AI, known as what has been called a “hidden tax“which stops part of the benefits. Adolfo Pellicer, Country Manager at Workday confirms that the use of these tools requires review and supervision of the results. “There is a hidden impact of AI at work. The report shows us that almost 40% of the time saved with the use of AI it ends up being lost in correctingreview and redo what the information that AI gives us,” said Pellicer. in statements to Computer World. AI digital natives. The youngest employees, between 25 and 34 years old, account for 46% of the cases with the highest review burden, since they use AI more frequently. 77% of these users verify AI results more rigorously than human-generated work. This generates additional exhaustion in these profiles. In departments such as human resources, 38% of employees need to review AI results due to the high number of errors reported. For its part, in the technical and IT departments, with a 32% increase in the use of AI, the tool has been better integrated, generating better results and content that requires fewer and fewer modifications. ​Training in companies: the pending signature. Although 66% of global leaders cite skills training as a top priority for leveraging AI, only 37% of employees who regularly use it admit to having access to these training programs. According to the report data From ONTSI, in Spain, this disconnection is worsened because 78% of workers demand more digital tools and training to use them, but adoption remains low: only 11.4% of companies with 10 or more employees used AI in 2024. In Xataka | Firing a worker because an AI “does its job” sounds very tempting. China wants to make it inappropriate Image | Unsplash (ThisisEngineering)

He is the most important programmer in all of history. And he has also ended up using AI to program

Linus Torvalds, Linux kernel creatorhe found himself with some free time this Christmas, so he wanted to dedicate it to a personal project that he had on the shelf: an application capable of generating digital audio effects that he called AudioNoise. The curious thing is not that he started programming on his own, but rather what he ended up doing with part of that application. Linus tries Vibe coding. This project has a description on GitHub that holds a surprise. In the last paragraph of it he indicates that “Also keep in mind that the Python visualization tool was basically written using vibe-coding. I know more about analog filters (which isn’t saying much) than I do Python. I started with my typical “Google it and copy what I see” way of programming, but then I cut out the middleman (myself) and used Google Antigravity to create the audio sample viewer.” The best programmer programs like any other. The statement is surprisingly sincere and honest from the one who He is probably the most important programmer in all of history.. Admitting that your typical way of programming is “search on Google and copy what I see” is already curious, but part of programming precisely consists of looking for solutions from others and copying or adapting them into your own projects. AI can help. The other (big surprise) comes of course when he indicates that the visualization tool for his project was not programmed by him, but rather by the Google tool, Antigravity. This integrated development environment (IDE) allows you to work directly with Gemini 3 in different versions and even with Claude Sonnet and Opus 4.5, and despite having been released a few months ago, it is becoming one of the favorite tools of veteran developers but also of those who are beginning to make their ideas a reality without having too much knowledge. Let them tell it to me. This viewer has been programmed by Linus Torvalds. Or rather, Antigravity controlled by Linus Torvalds. testing. At Xataka we were curious to test what that viewer did, so we cloned the GitHub repositorywe asked Claude Code to explain how the project works and after a couple of quick changes we were able to test it with a small audio file. What the project does is implement digital audio effects (there are ‘phaser’, ‘echo’, ‘flanger’, ‘fm’ and ‘discont’) and then, if one wants, apply visualization. That visualization compares the original audio with the processed audio to see how the effect modifies the waveform. Subsurface Capture Linus’ other projects. Although Linus Torvalds is the person most responsible for the Linux kernel evolutionit is already common to see him develop some parallel and totally independent projects. Last Christmas he already created his own guitar pedal softwareand in 2011 began the development of Subsurfacean application to record and plan scuba dives, an activity to which he is very fond. The current GitHub repository is maintained by various developers among which stands out Dirk Hohndel, who was one of the first developers of the Linux kernel along with Linus Torvalds. AI is a tool. In a recent participation on the Linus Tech Tips YouTube channelLinus Torvalds talked about how he saw the world of AI. In your opinion: “AI will be a tool, and it will make people more productive. I think vibe coding is great for getting people to start programming. I think (the code it generates) is going to be horrible to maintain… so I don’t think programmers will go away. You’ll still want to have people who know how to maintain the output.” And it works for personal projects. That speech precisely aligns with this small “experiment” that he has used in that personal project: at the moment for projects of this type using tools like Antigravity can be a great idea, although it certainly does not seem so for larger projects in production. Thus, it does not seem likely that AI could be used to modify Linux code… at the moment. That, of course, may change in the future, but as Torvalds says, these types of developments will require notable (and probably human) oversight to validate that everything has been done correctly. Image | TED Conference In Xataka | Bill Gates and Linus Torvalds had been rivals for 30 years. The funny thing is that they just met and took a selfie

In 1919 the Germans decided to sink their entire fleet in the North Sea. The steel from those ships ended up in space

At 11:20 in the morning of June 21, 1919, Admiral von Reuter’s ship began to signal to the rest of the German ships in Scapa Flow Bay, England. The taps and water intakes were opened, the pipes were destroyed, the portholes were dismantled: no one noticed anything. Until around midday, the Friederich Der Grosse began to list to starboard. It was already late, the German flag was flying from the 74 masts. Scapa Flow. The image tells the story of Scapa Flowthe sinking of the German fleet immediately after World War I. While the Allies negotiated the terms of the Armistice with Germany, the fleet was held captive and stationed off the British coast. Von Reuter feared that the Allies would divide up the ships, so he decided to sink it completely, at any cost. The British naval ships that were on maneuvers arrived at 2:30 p.m. and were only able to save one ship. The last to sink was the battlecruiser Hindenburg. Nine Germans were killed, 16 were wounded, 1,774 were detained. 52 ships were sunk on June 21 at Scapa Flow. But they are no longer there: they are on the Moon, Jupiter and beyond the orbit of Pluto. steel is steel. A tough guy, with bad temper and few words. But in 1945 (or a little before), everything changed. We didn’t realize it at first, but we quickly discovered that although all steels are equal, there are some steels that are more equal than others. I’m not going around the bush: what happened in ’45 was the atomic bomb, the device of the Devil that made us change geological era. The problem. Since the first atomic bombs exploded on the Earth’s surface, the air contains traces of radioactive elements. They are there, dissolved in it, but the amount is so small that they are harmless. Unless for some strange reason you have to blow in enormous amounts of air in the manufacturing process of some material. It’s almost useless to us. That is, all steel manufactured after the explosion of the first atomic bomb is radioactive. Very little, almost nothing. But enough so that some medical, physical or astronomical instruments do not work correctly. For example, radioactivity monitoring systems used by spacecraft. He tells it David Bodanis in “E = mc². Biography of the most famous equation in the world“, a book that, although it has become somewhat outdated, is still a delight. You may have heard the story, but it is a good story. Steel = expensive. In the book, Bodanis explains that, faced with this problem, uncontaminated steel became very expensive. Above all, because before ’45 we did not make steel in quantities so industrial as now. I imagine dozens of NASA engineers rummaging through their family’s cutlery so they can send reliable machines into space. Until someone remembered Kaiser Wilhelm’s ships. The peculiarity of Scapa Flow. There are sunken ships in many places, but there are not many shallow inlets with 52 sunken ships in their waters. Not all of them were there, but a few were enough for us to manufacture the equipment that the Apollo mission left on the lunar surface, that which the Galileo probe took to Jupiter, and that which the Pioneer probe is taking even further. The evil, the sea. In Xataka | Quantum find in Cambridge points to solar ‘Holy Grail’: single-material solar panels In Xataka | The Atacama salt flat is the key on which the electric car industry pivots. And it’s starting to dry

A teenager discovered the ‘Málaga’ virus and ended up founding VirusTotal. The enigma that remains is the same since 1992: who programmed it

Bernardo Quintero (@bquintero) was 14 years old and his first PC, an Amstrad PC-1512, had just arrived home. It was 1987, and the co-founder of VirusTotal He was excited by this machine that allowed him to exploit his computer curiosity. His hobby ended up being trying to circumvent the copy protection systems of some games, and he was there one day when something suddenly happened. A little white ball moved on your screen. By itself. Without him having done anything. He soon discovered that it was a computer virus. One that he ended up studying to know how to detect and eliminate it. He succeeded, and over the next three years he ended up improving his first antivirus, a tool that allowed him to recognize and eradicate seven different viruses he had encountered. It didn’t seem like that project was going to go much further, and Quintero began his studies in Computer Science at the Polytechnic University School of Malaga. In one of the first classes, a professor asked if anyone wanted to raise a grade with a Pascal programming project. He signed up, and when talking to the professor, he asked him if he had done any previous projects. “Well, yes,” he replied. “An accounting program, disk utilities, an antivirus…”. The teacher cut him off. “Did you say antivirus?”. When he answered affirmatively, the professor asked him to accompany him to his office. There he showed him how the entire IT department had been infected by a virus that the antivirus did not recognize. Fragment of the code in Turbo Pascal 5.5 of the antivirus that Bernardo Quintero developed to eliminate the “Málaga-2610” virus (1992). Source: Bernardo Quintero. Quintero soon detected where the problem could be and went home with an infected disk to work on an antivirus. It took him more than he thought, but after a few hours he managed to figure out how to detect it and delete it. That helped him pass the subject, but it also ended up being the definitive seed of the professional project that would end with the founding of Virus Total. He tells it all in more detail in his novel, ‘Infected‘, which he published at the beginning of the year and in which he narrates those beginnings and how that ended up leading him to create VirusTotal, the Malaga company that would later end up being bought by Google. That virus in his faculty was called “Málaga”, and Quintero spent years without paying much attention to it again. So, three years ago, this expert posted a message on Twitter (X) to try to solve the mystery of who would have created it. Already then he discovered that according to several sources the virus had been created at the Polytechnic School of Informatics. The objective, I counted thenit was not about bringing the name to light, but about chatting with that person and remembering those times. He failed to reveal the mystery, and that mystery remained unsolved again. But Bernardo Quintero never forgot that and returned to the fray with a new attempt a few days ago. After first publishing a message on X, the next day he published a summary of that story on LinkedInand asked for help in that post to try to solve the mystery once and for all. We contacted him, and he told us how while in the past he had focused on discovering how it infected and creating the disinfection tool, he never tried to find out who had created the “Malaga” virus. But he told us that “now, looking at it with new eyes, I have seen a couple of interesting details and I have discovered the motivation.” In fact, he adds that thanks to those messages on X and LinkedIn “I have received stories from several people who studied those years at the Polytechnic of Malaga and who believe they know the author.” Of those candidates, he explains, “I have ruled out 3 or 4, but there is one that fits very well with the new data I have.” The mystery seems to be close to being solved. “I just need to clear up one unknown to confirm the author.. And if it is confirmed, there is a beautiful and sad story that will be worth telling.” Everything therefore indicates that it will finally be known who was the author of that virus, and Quintero has promised to tell more details these days. We will be attentive. Image | Mika Baumeister In Xataka | The computer with the most malware in the world: this is MICE, the challenge of Bernardo Quintero and VirusTotal

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.