{"episode": 4472, "title": "Cheap Yellow Display Project: Introduction to the Cheap Yellow Display", "host": "Trey", "date": "2025-09-23", "url": "https://hackerpublicradio.org/eps/hpr4472/index.html", "duration_s": 327.6, "transcript": "This is Hacker Public Radio Episode 4472, for Tuesday the 23rd of September 2025. Today's show is entitled, Cheap Yellow Display Project. Introduction to the Cheap Yellow Display. It is hosted by Trey, and is about four minutes long. It carries a clean flag. The summary is, a brief introduction to the Cheap Yellow Display. Hello again, this is Trey. Several months ago, I heard Paul Acidorian mention the Cheap Yellow Display on his podcast, Paul's Security Weekly. I didn't think much about it at the time, but then I heard it referenced again, and again. Then finally it was described, and I became interested. Cheap Yellow Display is the term used for the ESP32-2432S028R. That's a mouthful. Let's try that again. ESP32-2432S is in Sierra 028R is in Romeo. Since this is somewhat challenging to say, and to remember for that matter, and since the board is yellow, and it can be obtained for as little as $12 US dollars, it has been given the nickname Cheap Yellow Display. I will abbreviate this as CYD that's Charlie Yankee Delta for the remainder of this episode, in the show notes. It is an ESP32 with built-in Wi-Fi and Bluetooth on a development board, with one or more USB connectors, a microSD slot, a limited selection of GPIO pins, an RGB LED, a speaker, a light sensor, and best of all, a 2.8-inch that's 71mm for our metric folks, TFT touch screen LCD display. The Cheap Yellow Display runs on 5V DC. I'm including some photographs of the Cheap Yellow Display in my show notes. Random Nerds Tutorials has produced a very good write-up about this board on their website, link is in the show notes. Brian Locke, aka witness me now, has been building a community for the Cheap Yellow Display on his GitHub site, also link in the show notes, where he has instructions, examples, tutorials, downloadable tools, and much more. Beginning way back in the 1970s, my father and I built electronics projects together, and I have had a love for doing so ever since. Over the last few years, I have built several Arduino-based gadgets on different platforms, including a couple which run on breadboards sitting beside me on my desk. Here's a little bit of a sneak peek, I will share more about one of those in a future episode. Alright, back to this one. A common use for the Cheap Yellow Display among hackers is to leverage the built-in Wi-Fi and Bluetooth radios to compromise wireless networks or devices. The Marauder Project is a pre-built image that can be loaded directly to the Cheap Yellow Display to use it as a wireless hacking tool. Frank Fletcher's GitHub Repository, whose link is in the show notes, because Frank is spilt with the four instead of an A, but anyways the link is in the show notes. Frank Fletcher's GitHub Repository is one place where you can download the Marauder Tool. There are also video games, clocks, photo slideshodes and more, which have already been coded for you and are available on the internet for download. What would you build with the Cheap Yellow Display? What could you build? What problem might you want to solve? What fun project might you come up with? For myself, the Cheap Yellow Display intrigued me, but it did not yet jump out at me as something I had a need for. Yet. What would change my mind? What would set me on a quest to obtain some of these devices and learn to develop code for them? What problem did I wish to solve? Tune in again in a couple weeks to learn the answers to some of these questions and more in my next episode for this HPR series. Goodbye. The Internet Archive and R-Sync.net. On this otherwise stated, today's show is released under Creative Commons Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4474, "title": "Hacker Poetry - 001", "host": "Major_Ursa", "date": "2025-09-25", "url": "https://hackerpublicradio.org/eps/hpr4474/index.html", "duration_s": 203.3, "transcript": "This is Hacker Public Radio episode 4474 for Thursday the 25th of September 2025. Today's show is entitled Hacker Poetry 001. It is the first show by Newhost Major Ursa and is about two minutes long. It carries an explicit flag. The summary is a show where hackers read poetry written by other hackers. Welcome to Hacker Poetry, a podcast produced by the League of Better Villains and distributed on Hacker Public Radio. I'm your host for this inaugural episode, Major Ursa. I haxt a for fun and profit. Each episode of Hacker Poetry will feature one or more original poems written by a hacker and read by a hacker, though not necessarily the same hacker. Before we begin our inaugural episode, award from our sponsor, the League of Better Villains. Tired, penny, antifascist, and blowwords who want to laugh and watch the world burn, we are too. That's why the League of Better Villains is striving to raise the ball and evil class out a big blue marble. You deserve the brand-o and better evil humanity. Keep your eyes on the skies for the release of our short, random swamps coming to an open-air market million this fall. The League of Better Villains, bound out to the best. Welcome back to Hacker Poetry. I'm your host, Major Ursa. Today, we have three poems by an anonymous hacker. To begin, we have... At the first fires, man stood back to back with flame, howling annihilation into the dark. The animals fled. They knew nothing outlasts that endless hunger. And now, Rebuttments, up down, over around, rivers with their banks, cold clay bargains within and without. Finally, headline, billionaire goes missing. Nobody searches too hard, must have gone to occupy Mars. This concludes our first episode of Hacker Poetry. I'd like to thank our distribution network, Hacker Public Radio, as well as our sponsor producers, the League of Better Villains, bound out to the best. I've been your host, Major Ursa, and your attention and investment are now and forever mandatory. See you next episode. For HBR, it has been kindly provided by an honesthost.com, the internet archive, and our sings.net. On the Sadois status, today's show is released under Creative Commons, Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4476, "title": "Does AI cause brain damage?", "host": "Trollercoaster", "date": "2025-09-29", "url": "https://hackerpublicradio.org/eps/hpr4476/index.html", "duration_s": 789.3, "transcript": "This is Hacker Public Radio episode 4,476 from Monday the 29th of September 2025. Today's show is entitled, Does AI Cause Brain Damage? It is hosted by Trolter Coaster and is about 12 minutes long. It carries an explicit flag. The summary is, going through the findings of an MIT study about how AI will make your brain rot away or will it? Hello knowledge seekers, do scrollers and heroic procrastinators. Welcome aboard the Trolter Coaster. The only ride where the safety bars is smug grin and the drops are powered by sheer cognitive dissonance. Today we're asking a radical question. Does thinking still matter in 2025? Spoiler. MIT just ran an experiment, suggesting the answer is... Yeah, kinda. And the internet promptly tried to cancel them for H speech against convenience. Strapping. Sarcasm. Highton's Learning. Peer review pending. Picture. 54 volunteers each wearing an EEG swim cap that makes them look like budget marvel villains. These brave souls were told to ride three short essays under three different conditions. The first team rides their essay using brain only. Raw, neural, horse power, no training wheels. The second team were allowed to use classical search engines. Let's call them, let me Google that for you crowd. The third team could use Chatchy PT to help them ride their essays. After three rounds of writing an essay, 18 volunteers earned a bonus level. The LLM group had their shiny toy yanked away while the brain only warriors finally got to taste the forbidden fruit. So EEGs scanned the brains and repeated some whizz back to evaluate the originality of the essays. Teachers graded those essays and so did the NEI judge, since they already paid for the open AI account anyway. The results? Wow, you wouldn't have expected it. The brain only team lit up the neural fireworks display, higher activity on every possible level. More lights than the server rack of Jeff Geerling. Search users still showed respectable sparks. And the LLM group, let's politely say the range scan looked like a single birthday candle nobody bothered to blow out. Ownerships course tanked and the test group barely remembered the quotes they themselves well actually their AI had typed. And their essays, clones, all the same structure, all the same topics. Ba ba ba. But wait, there's nuance. The researchers did not scream AI will melt your brain. No, they coined the calmer term, they called it cognitive death. Offload too much thinking today and you pay interest tomorrow less in engagement or weaker recall. And the vague sensation, your originality is slowly being garnished. Before we torch every chatbot, note the studies fine print, only 54 participants. And in the switch round, only 18. This is an exploratory postcard from the frontier, not the final gospel. So hold the pitchforks, or at least keep the tags on for easy returns. Now comes the juicy part. When the LLM loyalists had their clutch removed, EE engagement tanked faster than my motivation on a Monday. Meanwhile, brain first folks, who finally got to use AI, still remembered what they had written two weeks ago, kept the neural party going and basically showed that adding tools after you build muscles is fine. Replacing muscles with tools less fine. It's almost as if practice matters. Now isn't that a crazy concept? Travel back to your childhood with me. Remember those memorizing those tables and multiplication, hitting every minute of it. And if you finally learned how to bend your head around it and you could drill them like a sergeant, one, two, three, four, they gave you a calculator and feeling like for welding mealier, key detail. You learned before you offloaded. Nobody said skip addition kiddo, the TI-84s got your back. Nope, you had to learn them anyway because it helps you. Translating this to 2025, build cognitive strength first and then unleash the silicon sidekick. So it's clear it's not a new insight, scaffold speed shortcuts. Need a real more example here in Belgium, July. Medical entrance exam. A bunch of applicants slip chat GPT answers through the digital loophole. And the headline writers rejoice! Parents panic and somewhere a sergeant sighs because yet again people conflate using tools with being tools. Authorities investigated. Found a handful, quite a few bad apples, and reminded everyone that yes future doctors should know where the heart is. Without crowdsourcing the question, revolutionary stuff. The Dutch professor filling her months nails it which he says, relying on an LLM to do your homework is like rolling a fork lift into the gym to bench press for you. Sure, the weights go up, but your biceps stays as squishy as a half heated marshmallow. No pain, no gain, the same goes for the brain. Both demand resistance to grow. And if you're thinking, whatever, I'll just multitask my way into greatness. Decades of studies on digital multitasking would love a word. Heavy jugglers of screens, often sport, weaker working memory, attention spans or even a lack of self discipline. There are even structural brain differences. Who turns out your brief round with cortex is not a fan of 15-brothered trap tabs all screaming. Click me, click me, click me, click me, click me. Again, who would have guessed? It's established. No toys for the kiddies, but my brain, it's all grown up. So I don't have to worry about all this chisel. I can play with Claude and chat with Petey to my heart's content, because we all know brain degradation isn't a thing. There's one other cute elephant quietly sleeping in the room, hoping this drawer coaster will just steer away from it. The elephant is called the hype cycle. It's where every new technology gets pumped into the market as the next coming of Jesus that will save us all. Start-ups experiment with the new tech and savvy investors pump and hype in the hope to make some money out of it. The big companies like Microsoft and Google will ingest successful start-ups making it their own. A few other successful start-ups will manage to become their own big beast. You thought you were paying for your Chatchy Peas D subscription? Think again, you're merely paying for the lipstick. Smoking up their companies, swaying their hips for the Bitcoin barons and digital landlords. And to be able to do that, they need to show off their big customer base, promising big gains behind the curtains. So a bunch of investors took the bait. Millions even billions are rolling in. And after a while though, the investors want their money back. Ideally, with some extras as that's what they were promised. And this, my friend, is where the prices will go up and features go away. Cory Doctorow nicely coined it in certification, but that's for a whole other podcast. Maybe one you'll make because, come on, let's be honest. You don't want another one of these insulting episodes on our public radio, right? Just one last thought on this. The secret sauce is called open sauce. But don't tell it forward. But for now, let's just roll over and accept that other limbs are here to stay. And maybe can't be useful too. I challenge you! Yes, you with your tiny earbuds or your 20th century oversized headset to run a two-day experiment. The next time you need to create some content, maybe an essay. Maybe your love letter. Oh, please. Don't let chat GPT write your love letters. Maybe that nasty resignation mill you always dreamt of writing. Or maybe the outline for an episode of hacker public radio. Step one, use your grey matter to create the outline. Nothing else. No tech, just a piece of paper and steer away from the siren song of autocomplete and write it yourself. Then, if it's done, invite your favorite elemental critique, expand or dunk on your ideas. If the alarm is done stroking your ego and along the way also triggered some ideas, send your artificial voice for a nappy again. Finish with a personal rewrite. And 48 hours later, test what you remember out of the top of your head. What do you remember yourself? If your mind's a blank canvas, congratulations. You've just paid the minimum on your cognitive debt credit card. Time to refinance it with more effort. So, should we ditch the bit chatbot right away? Just short answer, only if you're also eating your calculator in doom and doom. Along answer, treat AI as a scaffolding. Keep thinking in the driver's seat. And remember, that speed means nothing if you swap accuracy for autopilot. Foundation fluency first, acceleration second. That's how calculators empowered math rather than utilizing it. That's how elements can make us smarter instead of duller. Alright, writers. The troller coaster is pulling back into the station. Brains intact, I hope. Key souvenirs, cognitive debt, is real but manageable. Effort precedes offloading and a pinch of sarcasm spices things up. Hmm, so you don't agree? Am I too nice to the AI overlords? Or did I hurt the feelings of your precious little psychophantic chatbuddy? Ha ha ha ha ha, great! Flip out the data that proves I'm full of digital hot air. Record your own podcast. Season it with a bit of spicy sarcasm. Don't spare me, I can take it, I like it hot. But if you're more of the sweet type, feel free to sugarcoat it. Until then, keep your neurons firing and have your AI on a leash. Preferably a retractable one. You have been listening to Hacker Public Radio. Hacker Public Radio does work. Today's show was contributed by a HBR listener like yourself. If you ever thought of recording a podcast, click on our contribute link to find out how easy it really is. Posting for HBR has been kindly provided by an honesthost.com, the Internet Archive and our Sync.net. On the Sadois status, today's show is released under Creative Commons' Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4477, "title": "doodoo 3 a deuce plus 1", "host": "Jezra", "date": "2025-09-30", "url": "https://hackerpublicradio.org/eps/hpr4477/index.html", "duration_s": 692.2, "transcript": "This is Hacker, public radio episode 4,477 for Tuesday the 30th of September 2025. Today's show is entitled Du Du 3 Du plus 1. It is hosted by Jezera and is about 10 minutes long. It carries a clean flag. The summary is continuing to blather about Du Du a TV app for getting done, done. My name is Jezera and this is episode 3 about Du Du creating a Du Du application. It is incredibly pleasing that I get to say Du Du is in production. That is to say I have installed Du Du on my remote server digital ocean and I'm using it all the time. In a sense, however, it also means that I am testing in production and there are certain things I should have tested prior to putting Du Du into production. Most importantly there would be what is the process of starting from scratch. At some point I had introduced a bug server side that was causing an issue if the data file for the items and categories did not exist. That file is created by the server side code if it doesn't exist. But part of the code was looking for data in that file before the file was even created. It was a very catch 22 situation. That being said, it was also the only major issue I encountered when deploying the software which I think is great. Before getting to the point of deploying the software to production, I really needed to figure out and implement the process for authentication. The process described in Du Du Episode 2 is pretty much exactly what I went for. On the server level server side that meant I needed to create a file based on the hash of a string that is entered by the user somewhere. In order to do that, I added to my make file. Yes, I am using a make file for this application. I added to that file a command pass-phrase- hash. So that the user has to enter make pass-phrase- hash. And that will prompt the user to enter a pass-phrase at which point that pass-phrase then will be converted into a SDA 256 some hash of that pass-phrase and stored as a text file. Upon starting the server, that text file is read into memory and that is the stored hash that any incoming pass-phrase is checked against. So the server, whenever there is a request for data, will check the included header for a pass-phrase and hash that pass-phrase. And if that hashed pass-phrase that was passed in matches the hash from memory, then all is good and data is returned. Actually, that isn't quite true. When the user enters the pass-phrase to generate a hash server side using this make file, what is hashed is pass-phrase with a carriage return new line appended to it. I haven't figured out how to not get that to be part of the hash. And so the pass-phrase that is passed in from the client as part of the request header, before that gets hashed, I add a backslash and new line to that string and then get its hash to check to see if they match. Through my usage of the application, I have noticed a few issues primarily with the user interface that could be improved. When I opened this application on my mobile phone, I decided that it would be best if there was a shortcut on the desktops. This is on Android, and I was using Firefox and added to the home screen. And there is not a favicon for the UI, for the web interface. And that means that there is no very specific nice button for this on my home screen. There's no icon. It doesn't look good. It's just the letter D. I don't like it. So I need to make a favicon more than likely that will be done as an FG. SVG using Inkscape, because Inkscape SVGs, or an SVG in general, makes an excellent favicon, and it is a small file size, which I really like. Other issues I encountered. There needs to be some way to show that something is happening when pressing the submit button for creating a new item or new category. And along the same lines when deleting an item or a category, this became most prevalent to me on a very slow connection. I was out-remote, and on a very, very poor LTE connection. And I noticed, significantly, there was a very long delay between pressing a button and something happening, and that is because the connection was incredibly slow. The UI shows nothing at all that tells me the user that, yes, I have pressed a button and something is happening. And at that point, I'm very tempted to start pressing the button again and again and again and again. And what that will end up doing is sending multiple API requests to the backend, which I don't want. So, there's something that needs to change. The API itself, that is the server API, needs better error responses. For the most part, every response is from the API is code 200, which means everything's hunky-dory, and maybe some data. But if there's an error at some places, there will be a response of 200 and then a message about that error. And that's not necessarily the best thing. In fact, that's terrible. If there's a specific type of error, then I should be returning the proper HTTP error code that represents that error. Another problem is the tab order of the UI. And this, I find, or I found, most prevalent while using my desktop, in which case I didn't want to have to move the mouse to select the new item button. I wanted to just press the tab button and get there as soon as possible. And the new item and new collection buttons are on the bottom. Because when I'm holding my mobile device, I want those buttons as close to my thumb as possible. That's why I put them at the bottom. Those buttons, being at the top, would be excellent for desktop, maybe not so much for mobile. But having to tab through all of the edit and delete buttons that exist for every category and item, is tedious. So that can be changed by setting the tab order in the HTML. I am reminded of the following advice from a woodworking book. Get in the habit of modifying designs to make them more appealing to your eye. Obviously, that doesn't apply to just woodworking. It also applies to software. It applies to anything where there is a design. It is an excellent credo for hackers. Take what exists and change it to make it better. The final major UI issue I encountered was when I entered as an item under the category of shopping, a URL to a product that I needed to purchase. It was a very, very long URL. And when that was displayed in the interface, it stretched the entire interface very wide. And it just made Dudu look like Dudu. And I was not happy with that. Again, there is a CSS property to define when an item breaks that should be able to resolve that issue. And that is something I need to code. And I'm going to go do that right now. On that note, I have got some code to hack and some music to play. And episode two to upload. Have a great day. Keep on hacking. And don't be afraid to modify designs to make them more pleasant to your eye. You have been listening to Hacker Public Radio at HackerPublicRadio.org. Today's show was contributed by a HPR listener like yourself. If you ever thought of recording podcasts, click on our contribute link to find out how easy it really is. Hosting for HPR has been kindly provided by an honesthost.com, the Internet Archive, and our sync.net. On the Sadois status, today's show is released under a Creative Commons Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4484, "title": "When Your Dentist Uses ChatControl Logic", "host": "Trollercoaster", "date": "2025-10-09", "url": "https://hackerpublicradio.org/eps/hpr4484/index.html", "duration_s": 937.3, "transcript": "This is Hacker Public Radio episode 4484 for Thursday the 9th of October 2025. Today's show is entitled, When Your Dentist Uses Chat Control Logic. It is hosted by Troller Coaster, and is about 14 minutes long. It carries a clean flag. The summary is, Troller Coaster alerts us to the dangers of the proposed EU chat control legislation. So here I am, sitting in the waiting room with a mildly annoying toothache. Nothing got a shroffick, just one of those dull throbs that tells you some things going on in there that could turn nasty if ignored. The receptionist calls my name, and I walk into Dr. Chat Control's office. The dentist greets me with a concerned expression. I've reviewed your case, he says gravely, and many others. Toothaches, or a serious problem. Did you know that in 2023 alone, at least 111,988 people in Europe experienced a dental pain? That's over 60% of all global complaints traced to this region. Okay, I say, but I just need you to take a look at this one tooth. This is an epidemic he interrupts, and we can't just treat symptoms anymore. We need a comprehensive mandatory risk assessment. Dr. Chat Control pulls out a 47-page questionnaire. First, we assess the risk that your mouth might develop cavities, gum disease, or oral cancer he explained. This applies to everyone, regardless of whether they have symptoms. But I have symptoms, I say, this is why I'm here. Exactly, he says triumphantly, which means your high risk. So we move on to phase two mitigation measures. He hands me a pamphlet titled Safety by Design for oral health. From now on, you need to implement parental control on your diet. Every candy will come with an age verification wrapper and user reporting mechanisms. So your teeth can flag potential problems and verify your age before consuming any hard foods. Age verification, I retort, but I'm 38! Perfect, he replies. That means you're old enough to consent to monitoring. Now, Dr. Chat Control continues. Since your mitigation measures haven't been implemented yet, you just got here, I know. I'm authorized to issue a detection order. He holds up an official looking document with a judges signature. This allows me to scan not just the two that hurts, but your entire mouth and your sinuses, and your lymph nodes, and just to be thorough, everyone in the waiting room. Wait, what? I blurt out. It's targeted. He assures me. We're only scanning high risk areas, which, according to our independent EU dental center, is every tooth, every patient, all the time. But here's the clever part, Dr. Chat Control says. Pulling out what looks like a tiny sander strapped to an airbrush. Your enamel contains layers that currently block our view with regards to early detection of certain viruses. So we need to replace the enamel with an optimized layer that perfectly protects against currently known sugars and acids, but allows our government-approved detection equipment to inspect the content of your teeth. I'm flabbergasted. You want to weaken my enamel? Ah, only slightly. He said cheerfully. Just enough so that our scanner can monitor what's happening inside your teeth all the times. You'll feel much safer. But and don't worry. The new enamel will still protect you from everything else. How's that possible? If you can penetrate it, I can continue, sir, cautious. Well, factory-approved candies, for example, he says, repeatable candy manufacturers have agreed to respect the adapted enamel and not penetrate it. They've signed a treaty to use optimized sugars. Okay. And the art is not candies, I ask. Ah, ah, he says, adjusting his glasses. Those are obviously illegal. Those criminals don't tend to honor treaties. So yeah, the adapted enamel will make your teeth more vulnerable to unauthorized cavity formation, bacterial attacks, and anyone who wants to exploit the fact that your natural protection has been replaced with an optimized version. So you're making my teeth less secure? I repeat. We're making them more observable, he corrects, with a slightly annoyed tone. That's a difference. Your teeth will still be protected, just not against us. Or hostile foreign candy shops, or those criminal artisanal real sugar nuts. But you have no business there anyway, and they will definitely protect you against harms from law abiding candy. But I read like over 500 leading dentists signed a letter saying this would create massive vulnerabilities. I point out those dentists are thinking about long term hygiene. But the chat controlled size dismissively. Besides, it's not a back door. If we're replacing the front door with a tinted glass door, given all your citizens are law abiding, his tone gets stoner if he continues. Let me say more. If you refuse to upgrade your enamel, we will consider you a risk of dental terrorism. This is no longer about your teeth only you must know. If you choose to blatantly ignore your teeth, you're a threat to society. But the whole reason I am here is because I care about my teeth. Doctor holds up his machine and asks, do you want me to treat you or to report you? Okay, whatever. It seems everybody is doing this. I don't want to be the outcast crying wolf all the time. So I gave in. It's a relief. The procedure isn't painful. After completing the baseline scan, Dr. Chat control frowns at the screen. This is concerning, he says. The system has flagged 8,412 potential cavities. What? But I only have 32 teeth. I whimper. The deduction algorithm works on a probabilistic model he explains. Swiss dental authorities report that about 80% of automated cavity reports of our false positives. So we'll need to investigate all of them. But that still means 1682 of those threats are real, I say. Which is still 1682 too many, Dr. Chat control blurs. Now I'll need to drill exploratory holes in every tooth, in your gums, your tongue, and just to be safe in your neighbor's mouth, because the system flagged them to when they walked past the waiting room. But this seems insane. The math is solid, he insists. Even at 99.99% accuracy, which doesn't exist, we'd still generate 100,000 false dental alerts per day across Europe's 450 million people. But we can't let the cavities win. While Dr. Chat control explains his flawless system, I notice how the tooth that is actually hurting, the one I came to fix, is starting to abscess. Dr. I say, pointing to the swelling. We'll get to that after we finish scanning everyone, he says dismissively. The important thing is that no cavity goes undetected. Even if it means we spend all our time investigating healthy teeth. But the actual problem is getting worse, I say. That's because sophisticated cavities have learned to evade detection. Dr. Chat control explains. They change their appearance slightly, maybe grow on the back of the tooth instead of the front, or hide under existing fillings. Our algorithms can't catch those yet. But what about that letter from those 500 dentists? They also claim it's technically infeasible? Ah, they're just not thinking big enough, the doctor blocks my argument. So here's the treatment plan. Dr. Chat control says, we're going to install a permanent scanning device in your jaw that monitors every tooth 24-7 and reports any suspicious activity to a centralized EU dental database. Ooh, that sounds like a massive security risk. Ah, not at all, he replies. The device is encrypted. Only our scanners can look behind your upgraded enamel. What leaves your tooth is already encrypted. So your privacy is intact. No other device than ours can read the data. What about those knockoffs you can buy on all the express? I ask. Ah, they don't work reliably. And also, why would hackers want access to your teeth? His cough. Although, now that I think about it, the device itself could make a high-valuable target for malicious actors as they could abuse them to learn how to make knockoff candy that doesn't get detected by scanners. Things your natural enamel would obviously protect against. But let's not go there. By the way, Dr. Chat control adds casually. Dentists and dental workers in the EU are exempt from the scanning requirement under professional dental secrecy. So there they don't get monitored, I ask. Correct, he says. Our oral health is a matter of public trust. Yours, however, requires constant surveillance. Isn't that like a double standard? It's a two-tier system for a two-tier society. Here applies pleasantly. Now, shall we proceed with installing the jaw monitor? By this point, my actual toothache is developing into a full-blown infection. The pain is incruciating. Dr. I need antibiotics and a root canal. I cry. No. We don't do target treatments anymore. Dr. Chat control says. That's the old model. Now, we focus on comprehensive mandatory monitoring. But... I'm gonna lose my tooth, I say. Perhaps he acknowledges. But while we were scanning everyone in the waiting room, we found three people with slightly suspicious molars. They're perfectly healthy now. But they could develop cavity someday. That's three potential problems prevented. And preventing is always better than curing. But my problem actually needs curing. I point out it's getting worse. Yeah, he admits. But we've invested so much in the scanning infrastructure that we can't really afford to do traditional dentistry anymore. The system needs to justify itself. You understand? Oh, but this is insane. I repeat. This is progress. He corrects me. As I'm leaving the office, still with my untreated painful mouth and a jaw full of surveillance equipment, I notice the sign on the door. Signal dental, orthodontic streama, and proton mouth care have seized operations in the EU due to incompatibility with mandatory scanning requirements. Apparently, the dentists who actually know how to fix my teeth without installing spyware have all moved to Switzerland. Meanwhile, I hear the voice of Dr. Chad Control behind me in the waiting room, announcing to everyone, we've successfully detected 6,847 potential cavities this week. When someone asks what he means by potential, he quietly answers. Only 11 were real, but that's 11 teeth saved. The infection is still spreading. My mouth is throbbing, and I'll probably lose my tooth after all, because nobody's actually treating it. But at least the healthy ones are under constant surveillance. Okay, so this was my story. I made fun of it, but it's not funny. I mean, it's about wasting loads of money on a noble goal in all the wrong ways. But next week, October 14th, the vote will be cast in the EU, so it's high time to reach out to your NEP and ask their stance if they're undecided or their motivation if they're in favor. Go to fightchatcontrol.eu if you live in Europe. And don't forget, if you don't live in Europe, this will also impact you, because weakened encryption in mainstream products will spread and the weakened encryption will be used everywhere. So fightchatcontrol.eu, reach out to your NEPs if you're living in Europe and make some fuss about it, please, because this isn't important. You have been listening to Hacker Public Radio at Hacker Public Radio. Today's show was contributed by a HBR listening like yourself. If you ever thought of recording podcasts, then click on our contribute link to find out how easy it really is. Hosting for HBR has been kindly provided by an honesthost.com, the internet archive, and our sync.net. On the Sadois status, today's show is released under Creative Commons, an attribution 4.0 international license.", "license": "CC-BY-SA-4.0"}
{"episode": 4485, "title": "Git for Github and Gitlab", "host": "Archer72", "date": "2025-10-10", "url": "https://hackerpublicradio.org/eps/hpr4485/index.html", "duration_s": 555.7, "transcript": "This is Hacker Public Radio Episode 4485 for Friday the 10th of October 2025. Today's show is entitled Get For Github and GitLab. It is hosted by Archer 72 and is about 8 minutes long. It carries a clean flag. The summary is Archer 72 talks about GPG and SSH keys for Git. Oh this is your host Archer 72 for Hacker Public Radio. In this episode I get a crash course on Git and thought it would be a good episode. Not actually on Git itself but how to use it on Git Hub and GitLab. Using Git itself has been covered quite a bit by Kletu and Yannick. First off I am looking for a job so I thought it would be a great time to brush up on my Git knowledge and make a show too. Of course I have no Git expert by any means but it has been said in comments. Hacker Public Radio is my memory. You will want to create an SSH key and a GPG key for each Git instance. In this case I will use both Git Hub and GitLab. A few other sites host Git files which are Hacker Public Radio's own Git. And then there is not a bug and code bug. I know this is old for a lot of people but you will want to create an SSH key with SSH key down which will create an ED25519 key pair. Several years ago this was not the default. I can't remember what it was but there was another type. And then you add an entry to .ssh config for each Git instance. The host is Git Hub.com. The user is always Git and I use an identity file which is SSH key. Located under the .ssh directory. And then there is the host for GitLab. User in there is always Git. Then you would want to use SSH-Add for each unique Git key. Now for signing each commit you would need a GPG key which is generated by GPG space, dash-full, dash-generate-key. The command to list the keys for public keys is GPG dash-list-public-keys and that will be a 40 character string that we will use in several places. There are a couple of Git variables that need to be defined. The first is Git, space, config, space, dash-global, user.signingkey, space. And this is where you put the 40 character public key. Then the command is Git, space, config, space, dash-global, space, commit.gpg, sign, space, true. And then to prepare the public key for copying to GitHub or GitLab or whichever one you would use is GPG, space, dash, dash, armor, space, dash, dash, export. And you put the public key here and then you take the output and copy to one of the Git repositories and its beginning of the key will say be to begin PGP. There are some small differences between the way the information is entered in GitLab versus GitHub. In GitLab you go to your avatar, you go to add a profile to SSH keys and add key which is on the right hand side. And then on your computer you can't tilde slash dot SSH slash GitHub GitLabkey.pub. Copy this and go to add key. Back at the terminal you would type git, space, remote, space, set, dash, URL, space, origin, space, git, at symbol, gitlab.com slash your username, slash your gitlab repo, dot git. Then go to add a profile, go to GPG keys and then add key which is on the right hand side and copy and add the public key from GPG dash dash list dash public dash keys. Now over on GitHub you would go to your avatar, settings, SSH and GPG keys, new SSH key. You can't your tilde slash dot SSH slash GitHub key.pub, you add in a title and a key and then add SSH key. Back at the terminal you would type git, space, remote, space, set, dash, URL, origin, space, git, at github.com colon, user slash github repo, dot git. Then for your GPG key you go to avatar, settings, SSH and GPG keys, new GPG key. You add a title for your key, you get the key from GPG dash dash list dash public dash keys and add in your key here. For the next part I didn't want to think about it too much, so I asked Cloud.io because I wanted to create a page for my resume on github.io. Create a repository and you name it something like resume or my resume. Upload your HTML resume file and name it index.html. Go to your repository settings, go to pages, under source select deploy from a branch. Choose if it's the main branch, choose the main branch and the root folder. Your resume will be available at HTTPS colon slash slash your username dot github.io slash resume. In about 10 to 15 minutes I had a resume uploaded here and once I learned how to use CSS I can create a better looking page. That's all I have for now. If this has been interesting, please comment or if you have critiques, let me know too or record a show of your own. If you have interesting thoughts or ideas that you would like to share, this has been Archer72 for Hacker Public Radio. You have been listening to Hacker Public Radio at Hacker Public Radio does work. Today's show was contributed by a HBR listener like yourself. If you ever thought of recording podcasts and click on our contribute link to find out how easy it really is. Hosting for HBR has been kindly provided by an honesthost.com, the internet archive and our sync.net. On the Sadois status, today's show is released under Creative Commons Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4487, "title": "Is AI autistic?", "host": "Antoine", "date": "2025-10-14", "url": "https://hackerpublicradio.org/eps/hpr4487/index.html", "duration_s": 611.2, "transcript": "This is Hacker Public Radio Episode 4487 for Tuesday the 14th of October 2025. Today's show is entitled, Is AI Autistic? It is hosted by Antoin and is about 9 minutes long. It carries a clean flag. The summary is, a personal opinion on commercial LLMs. I believe that God is the one that gives us the gifts, the power to do anything. Even to talk with a friend, even to share something you like it to discover or to learn. So I thank God for being here. Being able to share this show with the title you have already heard at the beginning. It was inspired by Archers 72, Dad in episode 4454, AI is a trap. Shared his opinions. I shared two. I hope it is as pleasant to you as it's pleasant to me to hear its show. I have the opportunity to hear. This is a talk for Hacker Public Radio about characteristics of LLM and how you can use them for your best. I will share five of them. First characteristic, LLMs, large language models are designed to always give an answer that convinces. Dad does not mean they are right. Use it for the best. Don't trust the answer. If you didn't know the answer beforehand using an AI to obtain a better or faster result, but for really receiving the answer, then verify it independently. Second characteristic, commercial LLMs, many times, do not think too much in the face of simple and short prompts. Dad's a barrier, probably against wasting resources. How to use it for the best? Two things. First, show the AI what you have found by your own and proceed to tell what exactly you need help with. On the basis of what you were already able to think. Second, learn to follow up to have a conversation. Suggested follow-ups are not good. Prefer instead to talk to the AI as a real conversation, to get confirmation of what you need or contrast something you disagree with or to obtain an example that may clarify if it's true or false, etc. And Dad is a characteristic that resembles autism a bit, so the name of the show. Do not like when someone wants something without doing any effort to obtain it. So, you also do not feel like doing it for them. Someone comes to you asking something but has not even such a dictionary before just comes with empty hands wanting a world from you. AI's generally also don't like that. And you want to obtain good results with a so ample, with a so vast prompt, with a so vast input without knowing yourself what you really need punctually, specifically. Third characteristic without the long-gating anymore. LOMs are not accountable. In general, most AI's, you don't always know all the sources for the information given. And yes, they need resources, they need sources for the information. They can't invent something new, unless it's mixing words without knowing exactly. So, what they have is from the source that they have drink enough. That's what I say. They do not know what they are telling, nor they understand the value of the sources. And you can expect different outputs for the same inputs. AI's are not deterministic. Simply saying they are quiz machines to generate content. They don't cognitively, cognitively understand. But that convinces humans because they use human content and patterns. So if you use AI for something, especially any serious purpose, remember the result you get is of your responsibility. Don't expect to be excused for your words because I was helped by AI. They did it. You don't have these excuses. What is for you, you can have any source for your task. But you are responsible for characteristic. AI have a pattern. We can suspect with some degree of certainness, certitude, how to say that. We can suspect that something was generated by AI, and no one likes to be answered by one. If that possibility was not explicitly told. What to do about it? I suggest you don't use AI generated content with someone you estimate. Or for answering students if you work with them, etc. It is rude. Fifth characteristic. And the last. LOMs can give great results. With less effort, if effort, then you need to apply with no machine at your site, with no help at all. What the AI have done is by exclusion, not at what you have done. Use it for your best. You do not learn if someone does the job for you. For tasks, you know well. And what to accelerate, or remove repetitive steps, you may count on AI, or better, with a count on a specific software that can be programmed and give accountable results, because you know exactly what is being done with the input. Contrary to the AI that is somewhat random, or at least not accountable, not deterministic. But for the intellectual work, if you like to think, if you're good on written expression, the LOM may give, or may get passable results in less time, at the price of removing from you the chance to dedicate yourself to the comprehension and production. So it's not always a matter of producing like a king. Effectiveness. It may be important to satisfy what you value as meaningful. Learning, maybe, or feeling the satisfaction of the conclusion so that you can sustainably follow a routine that is not a pain on the eye. That is an expression to mean something that causes disgust. That's all, that's all folks. For today, let me know if you'd like more content on this, possibly I, or Arthur72, or even you could bring some more opinion on AI, if it is of your interest. And one was here. Bye-bye. Thanks for watching.", "license": "CC-BY-SA-4.0"}
{"episode": 4488, "title": "Cheap Yellow Display Project Part 2: What is the problem?", "host": "Trey", "date": "2025-10-15", "url": "https://hackerpublicradio.org/eps/hpr4488/index.html", "duration_s": 456.6, "transcript": "This is Hacker Public Radio Episode 4488 for Wednesday the 15th of October 2025. Today's show is entitled, Cheap Yellow Display Project Part 2. What is the problem? It is hosted by Trey, and is about six minutes long. It carries a clean flag. The summary is, delving into the problem I want to solve using my cheap yellow display. Hello again, this is Trey. This is Part 2 in my cheap yellow display project series. You can find Part 1 in Episode 4472. We previously left our handy hero learning about the technology of the cheap yellow display, but without a compelling reason to begin using one. As we rejoin the topic, it was Winter Field Day 2025. That would be January 25th and 26th. Winter Field Day is actually two days, but yeah, January 25th and January 26th, 2025. Winter Field Day is an annual event where amateur radio operators from around the world gather some portable radio equipment and set up somewhere away from their normal base of operations. It is designed to encourage operators to practice their emergency preparedness skills in unfavorable weather. Usually, they will run their equipment using batteries or generators. I chose this day because I knew there would be a good amount of radio traffic. I had just finished tuning my first hand made inverted V dipole antenna for use on the 10 meter amateur radio bands. These bands span 28 megahertz all the way up to 29.7 megahertz. I had the antenna connected to a 10 meter amateur transceiver to listen in on the radio traffic. Yes, I will include pictures of the antenna in the show notes. Scanning through the lower end of the band resulted in receiving a number of very strong continuous wave signals. Continuous wave is abbreviated CW or Charlie Whiskey in amateur radio circles and it stands for Morse code signals transmitted over radio frequencies. The tones indicating dots and dashes of Morse code were clearly audible through the radio speaker. Wait, stop time out! I can hear you shouting as you listen. This is supposed to be a discussion of the ESP32 cheap yellow display. What does this have to do with amateur radio? You know what? You are absolutely right. Now hold your horses and we will get there. So, I barely learned Morse code as a child and I used it a bit as an aviator in the 90s. Well, when flying I was always able to reference a visual representation of the Morse code beside the actual letters for navigational aids and other things that I was using to Morse code to identify, but that's neither here nor there. The point of it is I never really became proficient. Shortly after Winter Field Day 2025, I began taking lessons on Morse code with the goal of becoming proficient at both sending and receiving Morse code around 20 words per minute. This training may be the topic for another episode in a different series, maybe. We'll see how my skills progress. Anyways, I scanned further up the band and I also identified some digital transmissions. They sound like noise, sometimes like modems, other things like that. I'm guessing what I heard was FT8 and further up the band I heard voice transmissions. So, my antenna was working at least for receiving. Now, I've been a radio operator for a while. A little back history. I've had an amateur radio license since 2016 and I quickly progressed all the way up to an extra class which gives me permission to use all the amateur radio frequencies that are allowed within the United States on the high frequency HF, very high frequency VHF and ultra high frequency UHF bands. However, to this point, I have only actually operated on the VHF and UHF bands and have done so using mobile and handheld transceivers. I inherited some HF equipment from a close friend who went silent key in 2023 and I was only now trying to use it. You can learn more about my friend and about the term silent key in episode HPR 3922 with a link in the show notes. Alright, back to field day. There was far too much traffic and it was way too intimidating for me to even consider making my first attempt at transmitting on the HF bands. Nope, no sir. This would have to wait until later. But I didn't need to determine how well my antenna would transmit. I began to ponder my options. I really did not want to talk to anyone until I had listened to some more QSOs and I could implement proper radio practices. Now the term QSO, that's QSO Quebec Sierra Oscar, is the amateur radio term for radio conversations. Anyways, I also have no interest in digital modes yet and I like the simplicity of voice and CW so there I am back at Morse code again. What if there was a way that I could transmit a signal in Morse code and get reliable feedback on signal propagation without the need to try to reply to any responses. Now for this to work it would need to be accurate and repeatable and properly structured in timed Morse code transmission and it's way more than my training at the point of this recording or even at the point where this was happening, more than my training could can accomplish. So this is something to think about and think about and think about. Alright enough thinking tune into the next episode in the series to learn where these thoughts led me and how in the world this all relates back to my cheap yellow display project. Goodbye. You have been listening to Hacker Public Radio at Hacker Public Radio does work. Today's show was contributed by a HBR listener like yourself. If you ever thought of recording a podcast and click on our contribute link to find out how easy it really is. Hosting for HBR has been kindly provided by an onstoast.com, the internet archive and our syncs.net. Onstoast.net today's show is released on our Creative Commons Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4492, "title": "How to do a distribution upgrade of an Ubuntu LTS on a Digital Ocean droplet", "host": "Rho`n", "date": "2025-10-21", "url": "https://hackerpublicradio.org/eps/hpr4492/index.html", "duration_s": 433.3, "transcript": "This is Hacker Public Radio Episode 4492, for Tuesday 21 October 2025. Today's show is entitled, How to Do a Distribution Upgrade of a Nubuntu LTS on a Digital Ocean Droplet. It is hosted by Rowan and is about six minutes long. It carries a clean flag. The summary is Rowan upgrades his Ubuntu 20.04 LTS Digital Ocean Droplet to 22.04 LTS. The episode of Hacker Public Radio, today, I learned how to update my Digital Ocean Droplet running Ubuntu 20.04 to Ubuntu 22.04 LTS, but the LTS is, if anyone's, you know, up for done an upgrade, just up to raid for Ubuntu before, it's usually pretty straightforward over all this was, but Digital Ocean maintains its own version of the Ubuntu core packages, and the normal program to release upgrade, disables any third-party repositories and just runs against the Ubuntu repositories. Unfortunately, when you do this, you get the message, you can't find the core packages after a little duck duck going, or as it's usually called, New Berlin, I am found a podcast, podcast, wait, it's only a post about how you do this, because originally I'm thinking, do I need to go in and change my sources list and list, and this person, the post, and that quickly says, no, you do not need to mess with the sources list that you have to do, is pass a environment variable to the do release upgrade program, environment variable being release, creator allow third-party, set that equal to one, and do release upgrade, but then happily use your third-party repositories while doing the upgrade. So then after that, it was you a pretty standard, let it run, let it do all the downloads, saying, yes, I want to keep my considerations for various packages, and after a nice cup of coffee, he's got your new 22.0 for LTS, quick reboot, test the system, and generally nothing, no problems, although I did run into issues running next cloud. I have a couple other websites, one, hpr.cornium.us, if you want to see the new site, so are those, and the future you may already be looking at the new site, but anyway, that was right on time, but next cloud was PHP, and I quickly realized that when looking at the status of the Apache server, it wasn't starting because it was still trying to load the old PHP 7.4 module, which was removed, and so I just had to enable the PHP 8.1 module. I got my stuff closer, I did then still get a different error message about an internal error. I found the next cloud logs, and it said it can't connect to the database, so I was like, what's going on? And I think, oh, right, it needs a module to be able to connect to Postgres, which is a database I'm using. Actually, which reminds me, I actually did also, there was a message during the upgrade that Postgres 12 was obsolete, and it's been upgraded to Postgres 14, which really is kind of old at this point, but we're in LTS behind anyway going to 22.4, but it didn't give you nice instructions on basically saying, if you want to move your old Postgres cluster to the new Postgres cluster, you need to basically delete the current Postgres, the default Postgres 14 cluster created, and it gives you a nice command, pgdrop cluster, dashes stop, 14 main, and then there's a pg upgrade cluster command, and you just give it the old cluster, and you tell it, you want to get to a new cluster, and it takes care of things right away. So I did actually do that first before I started messing with the PHP stuff, because I knew that was going to be an issue. So next, after getting PHP running and getting the Postgres mod running, I did actually see next cloud running in my browser, but it was throwing one more error that it couldn't find, the GD module. GD does the graphics module it uses when it's for the photo album and things. So then I did have to enable that, and both the pg, the Postgres module, and the GD module I had do install. I have notes in these, or in the show notes, they're references to the commands that were run to make it easier for you running to this yourself. But after that, everything looks good, it seems to be running. I do claim to upgrade to 24x4, but the first thing I want to do is move my cloud instance from using blocks, storage, to using s3 storage, but that is the project for another day. Anyway, thank you for listening, and I hope you enjoyed this episode of a Carth public radio. How easy it really is. Hosting for HBR has been kindly provided by an honesthost.com, the internet archive, and our syncs.net. On the Sadois status, today's show is released under Creative Commons, Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4494, "title": "Exploring FUTO Keyboard", "host": "Antoine", "date": "2025-10-23", "url": "https://hackerpublicradio.org/eps/hpr4494/index.html", "duration_s": 1053.5, "transcript": "This is Hacker Public Radio episode 4494 for Thursday the 23rd of October, 2025. Today's show is entitled Exploring Futo Keyboard. It is part of the series' app Spelling King. It is hosted by Antoin and is about 16 minutes long. It carries a clean flag. The summary is, I install Futo from zero and explain why I prefer him. Hello, you are at Hacker Public Radio today with Antoin and Futo Keyboard. I've just formatted my telephone. We initiated it to the defaults because my Samsung Galaxy have updated to Android 16 and sometimes I like to format after an upgrade of the system. So, to restart again. So, I've just reinstalled from zero the Futo Keyboard. I will go through its minimal configurations because it's the keyboard I use. Why I like it? Let me explain it first before opening. But you can already install it at your phone if you like why I speak. I like Google Keyboard because I think it's beautiful and it has a very good prediction. And I prefer long time ago maybe more than a year that I do not use the Gboard. I prefer a keyboard that is more private without access to the internet but that has the same functionalities. And I have found the Futo because it has no internet access and it has voice recognition. It has swipe to text, swipe to type and a good prediction. And I think it's beautiful too. One point I'd like also is that I can configure some minimal things like what are the symbols that I want to be in each character if I want the A to have an amp or sound or whatever. And you don't have that minimal configuration but you can have a little effect. And also if you want the number, row, at the top or not, etc. I like this kind of configuration like the time I press the key for the anomaly seconds to have access to the other characters. You can configure that milliseconds that you don't have in many keyboards. So, opening the Futo. First pass, first thing is go to options of typing. Then I have to come to enable Futo keyboard. Then I enable it. We turned. Step 2, at the app it says select Futo to type. So I select Futo. Alright, now I'm at the menu. We have the options like languages, keyboard, typing, prediction. Let's go a bit of each one. Languages. You can ignore the verify if there are updates manually because it is by playstyle. But it has the option there, I don't know why. Languages, here it is in French for me because I have the telephone in French right now. But I can, let's say, add a language at the top. My language, Portuguese. Then Portuguese, Brazil. And the QWERTY type of keyboard. You can select many configurations. I recommend you to test the Azerty configuration. I like the position of DA. My fingers don't tumble so much at that configuration as I try a data bit. But for now, I have the QWERTY because I am a bit rated, more use it too. So there it is, I have added a language and I can also remove one. I remove the French. I remove it in the bin icon. Okay, I remove it. Now I have the part that is okay. That's it. You can get your language. But it's not all. Going under, you can see explore dictionaries. You can go to the adoption. And I don't have the browser, but I think that's the option to go to a website. To download the language, the dictionary for the language. All right. So if you want correction, et cetera, you have to download. It's very easy to download the dictionary for the language and import at the program. After the download, you can exclude the downloaded item because it will already be at the end. Returning keyboard and typing, here are the main things. You can have a bit of tiny adjustments you'd like for your better use of a keyboard. Like resize. If you want the option to have the number row at the top. I use it. I prefer it. If you want the arrows to be under the keyboard, I don't need it. I don't like it. It's an option, but I didn't need it. How you want to have the touch in the space bar. You can go there. Oops. I want to touch the icons. That's what I talked at the beginning. You have active by default. Like the characters of the language should be the first. Like the end with a tile, I think it's a tile. If you've long press in, or you can have the option to put the numbers as the default for a long press of the character in which there is. The secondary type for numberings, symbols, fast actions. I like that option, that is at number four. I don't mind the order because I normally long press and see what I'm doing. So you can don't matter the order for me. But you can exclude what you don't want. But that action, fast actions, I want to talk to you about. Like I open a notebook here with the foot already here. And let's say, I wrote something. And I can award the long press award. Normally, what you have is a pop-up to copy or to cut or to paste. And with the foot or keyboard, you don't have to select the word and long type at the screen to see the options. The C of the keyboard, long press C, there's the option with two windows that is to copy. So if you have a text selected and go to C, long press there, then copy it. And the space, you want to paste that thing. You don't have to go to the little tiny point of the screen that you want to paste. Because you have the V, long press the V, then there's the option to paste. That's the main feature that I don't see everywhere. That's it. But you also have like typing something. And you want to cut. Cut is the same thing you have that contracts on windows. That is copy or while deleting that thing. To do that, you'll long press the X. There is this Caesar to cut. Now the word or the paragraph, the text you selected and cut it and have cut. Now you can paste with the long press on V and copying and pasting. That's it. While you're doing that, you may already have the option of the correction at the top, the correction bar. Oh, without internet, only with the dictionaries. You have returned into the options of the keyboard. You have other configurations. Now that is for languages. If you want to have the bar of actions or suggestions at the top, you can do on and off. You can have the passwords to appear on the suggestion bar. Plissage automatic on lean. That is auto-few, auto-few or something like that. Preferences of typing. If you want to swipe to type, it's not very good, but it works. So you can have it. If you want suggestions of emojis, I don't like that. I will disable. Correction auto, I don't want auto-correction. I want to type. If I want, I type on the suggestion. If I am doing wrong. If you want the first word to be my school, the long, height ladder. What is my English? You can enable or disable it for the first word to be with a big ladder. Then I am pointing space. If you want to do a double space, to do a space and a point automatically spacing a dot, I don't like it to disable. Sometimes I want to more spaces and not replacing by dot automatically. Another option, delete, pasted, text and backspace. And this is in English that is not translated to French. So you can disable or enable the delete pasted in certain text while backspace. Delete what I have already, what I have just pasted. And do auto-correct and backspace? Why? I don't have auto-correct, so the adoption disabled are enabled. Like it is now. It's not relevant, but let's disable. Do you like character? Like when you press the word, the ladder, you want it to appear big. I don't like it. I don't like to type when not see the words floating, the ladder floating. So, disabling. Vibration? No. No vibration. Some sound? No sound. Okay. That's what I like it. Text prediction. You can do it like it is. That is the transformer LM language module. You can have it enabled like it is. But not the final adjustment because it says it consumes a lot of battery. I never use it, so I won't tell you to do that. Personal dictionary, there is the option. You can add your words. It does not automatically add anything, but if you want. I always use that word. I want the keyboard to know, to suggest, for example. So you can go there, the dictionary, personal dictionary, and put your word as a language specific or for all languages. No, no. It is already owned by default to block shocking terms. And there is not the list. It is automatically. And you can add one list if you want. Suggestion for correction enabled by default. Automatically detects wrap, happy. I don't remember whether it is. And personalize suggestions. I disable that. I don't like the keyboard to learn so much. Because then it gets very good. And then I find myself writing what the keyboard says. I want to write. But I don't want this personalized suggestion. So I disable that. There is the action, enable action key. I don't know. Yes. You can have the action key. I don't remember what it does. Theme. I like the dynamic auto. That is, nights or day mode only. But you have other little things. And last option. A part of the help that I want to do that. Nothing. It's the payment. I have a red bot. It is 100% free for practical use. But it encourages you to pay. As I like, I pay it. And to not download again the payment app. I can just click here on. I have already paid. I have already paid. All right. Now the payment option has gone away. Miss Elena? Okay. There was it. That is what I like. And photo keyboards. Some features. It's beautifulness. Or better. It's beauty. And my sharing with you at Hacker Public Radio. I am or I was. And one here with you. Good bye. You have been listening to Hacker Public Radio. At Hacker Public Radio does work. Today's show was contributed by a HBR listener like yourself. If you ever thought of recording podcasts. And click on our contribute link to find out how easy it really is. Hosting for HBR has been kindly provided by an honesthost.com, the internet archive and our sings.net. Unless otherwise stated. Today's show is released under Creative Commons. Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4495, "title": "An introduction to Taskwarrior", "host": "candycanearter", "date": "2025-10-24", "url": "https://hackerpublicradio.org/eps/hpr4495/index.html", "duration_s": 345.3, "transcript": "This is Hacker Public Radio Episode 4495 for Friday the 24th of October 2025. Today's show is entitled, An Introduction to Task Warrior. It is the first show buying you host candy can order and is about four minutes long. It carries a clean flag, the summary is, going over what makes task warrior unique and why I use it. Hello, it's candy can order that is how you pronounce it. Task warrior, the really cool program I've been using for quite a while. It's a big complex, like most CLI apps, that has a lot of custom ability. It makes task warrior unique and my personal go to, if I have the ease of frifting, is the unique way that it treats tasks. Every task is assigned an urgent v-nigger that starts at zero and can be added to from different sources. The default view also search by the number, so it can be used to categorize how important a task is. Each source of urgency is given a value between zero and one and a coefficient, and each source is multiplied together and added it to the total urgency. For an example, the due date urgency source had a coefficient of 12 at a fault, and the value gets closer to one, the closer to due date of this task is, bumpeding it up in a list. Add a reference in my own task warrior setup, which was a vanilla, tasks can barely get past 20 urgency, so due date is often a pre-large factor in sorting. Alright, so sorting is done by some weird number instead of the standard of sorting due date. So what? So you can customize a heck out of it. Every single sort of urgency coefficient can be modified at will through to task RC file, what the age coefficient to be more powerful than date, want tasks with a certain tag, be boosted to the top, it's just a single line. It's very simple and powerful, and easily lets you define how important tasks are to you. The custom and ability extend to creating your own task fields too. These fields called UDA, either defined attribute, can be defined in a task or see through a few lines as well. These unfortunately don't show up on reports by default, so you have to either modify the default report to include these forms or create your own reports, but they are extremely useful for further control of the urgency value. As an example, I have an estimated time at UDA on my setup, with a possible value of 5 minutes, 30 minutes, 1 hour, 2 hours, and long. 5 minutes, 30 minutes, and 1 hour give to us around 2.5, 1.5, and 0.5 to urgency, respectively, while 2 hours and long have negative urgency coefficients. That way, tasks that are either to complete flow up to the top and I can get them over with quickly. Task warrior also has some simple and effective ways to manipulate how often tasks appear, so it's do, occur, scheduled, and weight attributes. Do is stuff explanatory, tasks make an urgency the closer it should do that it gets, and a further boost if they are overdue. The way urgency works in task warrior is somewhat unique. In essence, it creates an invisible template task that is copied to a child task for each recurring event. The recurrence field takes the date range like 7 days or 1 months, and once a due date for the task passes, it creates a new task with each date field offset that many days. It is a bit finicky, but as long as you don't modify the parent task, it works well enough. It's scheduled, it is also extremely simple, since it just boosts the urgent theme by a lot once the date passes. Finally, there's weight. Weight is useful since the outright height of tasks until the date listed on that passes. It's a great way to set a task temporarily, or experiment with parents too, they make a task appear every weekday, only over the weekend. Weight is also extremely useful with the special date of the Sunday, which is defined as new year of the year 9999. Putting a task with weight Sunday is the usual way to softly detach. I personally use weight Sunday as a kind of bucket list, rescuing that I want to do Sunday, but aren't immediate enough to put onto the main list. I've mentioned before, it is also extremely eb-descript. It offers special commands for person with bash, and even offers an extra command that puts out every single task as a json for eb-person. It also offers hooks for running command after any action using class on scripts. Finally, it has the benefit of using contained in plain text and not running constantly. It makes the eb-to-export import and recover tasks equally. Please check the shows description for the full task warrior documentation. This is far from a complete overview and I'm skimming over most of it. Also, I'm not completely sure if I'm going to be doing another show, but yeah. I'm kindly provided by an honesthost.com, the internet archive, and our syncs.net. On this otherwise stated, today's show is released under Creative Commons' Attribution 4.0 International License.", "license": "CC-BY-SA-4.0"}
{"episode": 4498, "title": "Living the Tux Life Episode 1", "host": "Al", "date": "2025-10-29", "url": "https://hackerpublicradio.org/eps/hpr4498/index.html", "duration_s": 409.5, "transcript": "This is Hacker Public Radio episode 4498 for Wednesday the 29th of October 2025. Today's show is entitled, Living the Tux Life Episode 1. It is hosted by Al, and is about five minutes long. It carries a clean flag. The summary is, a mini-podcast series on running Linux's daily driver. Hi and welcome to Living the Tux Life Episode 1 by Al. This is a mini-series of me trying to run Linux as my daily driver. This is inspired by a podcast called Dilboot Darius, where two hard-cool windows gamers have had enough of windows, and they want to use it to see if they want to use Linux on their machines. So by the end of the series, they will either remove windows and keep using Linux, or they will remove Linux and keep using Linux. And this kind of kind of got me thinking that I should really use Linux a lot more. And this is why I want to document my journey of trying to run Linux as my daily driver. Mainly, I am a Windows guy. I work with Azure all day. I've always used Windows, it's my bread and butter, paste the bills. But it's okay to use it in a day, but I'm getting fed up with it outside of work, whereas with things like being really bloated here, you get forced off of 365, down your throat, you always want to reboot with updates. The installed time takes so much time to install. You have to now, when you install it, you have to log in with a Microsoft account. You can't do local apps. You can't log in with local users. I don't know, maybe one way, but they're just so forced to be not having logged in with local users, but having logged in with a Microsoft account. I don't really have no reason for using Windows outside of work. I used to, I'm a Scout leader, and we used to have to fill a format for risk assessments, but you had to have a full version of Word and Excel, and you couldn't use the online version because you didn't like the filling out of the form. Now that's online, you don't have to, I don't need to have, I use Linux all the time on my laptop. I've got another laptop of one which we're called podcast on, but I need to force myself to kind of use it more and more. Also, I've never really liked supporting other people with Linux because other things are running in Linux myself. If there's something going wrong, it's my problem, it's Linux. I need to fix it. It's my problem because I'm taking care of things. Normally, I'll just get people with running Windows on their laptops, but for them, my wife. She's got a T460 with 16-gram i5, a big one. When they come with e7, I've actually done the back of it. Her last row, as she loved it, but that was only when she's 10, and that won't support Windows 11 because it haven't got a TPM module. As is people who know, Windows 10 is no longer receiving updates. So, what am I going to do for her? So, for the moment, I've given her a Windows 11 laptop, running Windows 11. All she does needs now is running, it has a couple of websites, needs to do for her job. So, there's nothing there that is tiring what you need. So, I've given her a Windows 11 laptop, just so that I can get a T460 back, and then I'm going to store some kind of Linux on there and then show her how to use it and see what kind of like the wife has had to test it on it. I mean, she used to Windows 10. I've given her a Windows 11. It took a bit of a time to get a Windows 11, but now because I've got everything to do through a browser, I've given her two browsers she needs, one for work, one for personal. So, I can't see you about giving her her old T460 back, that she should have a problem running her Linux. I don't need to, I think I don't need to be brave and be ready to support Linux. I know, I'm a brainy guy, so I should be as a thinker. So, why don't I think I need to look at what I find with Linux is laptops, is that I like to have with like a modern laptop where you can shut the lid down, and if machine will go to sleep and you can come to back a couple of days later, open the lid and it either wake up and you haven't really lost lots of power. I've been playing with a couple of test laptops which are all i5, gn8, all things have the same high-ve nation issue where they either go to sleep properly or they don't go to deep sleep. So, I want to try it on the, I've got a 11th gen i11, I want to try laptops, is that this my homework this week is to try and see if I can get them work out the better, how power consumption really works and how I can get some deep sleep working on Linux so that when you, so that it doesn't, the battery doesn't drain when you put lid down. So, that's all for episode one. I think I might want to ask that is there anyone else out there who would like to take on the challenge with me, maybe a co-host, so we can bounce ideas off, so if you are interested drop me in line and come back tomorrow for another episode of PAPA Video! Listening to Hacker Public Radio at hackerpublicradio.org today's show was contributed by a hpler listener like yourself if you ever thought of recording broadcast and click on our contribution link to find out how easy it means. Hosting for hbr has been kindly provided by and ons.com the internet archive and our singing.net. On the sahdwai status today The show is released on their creative commons, attribution 4.0 international license.", "license": "CC-BY-SA-4.0"}