Remember kids, it's absolutely illegal to create an archive of the Spotify library, but it's totally OK for big tech to scrape the vast majority of created content on the internet to train LLMs...
Tänään ihmetyttää joululauluissa esiintyvät oudot nimet. "...Senjo tonttu näki, tonttu näki.." Senjo. Varmaan joku vanha nimi. "...Liinakomme kiitää ja valkoinen on maa". En tiedä hevosten nimeämisestä mitään, mutta Liinakomme on outo nimi.
New result: you can build a universal computer using a single billiard ball on a carefully crafted table!
More precisely: you can create a computer that can run any program, using just a single point moving frictionlessly in a region of the plane and bouncing off the walls elastically.
Since the halting problem is undecidable, this means there are some yes-or-no questions about the eventual future behavior of this point that cannot be settled in a finite time by any computer program.
This is true even though the point's motion is computable to arbitrary accuracy for any given finite time. In fact, since the methodology here does *not* exploit the chaos that can occur for billiards on certain shaped tables, it's not even one of those cases where the point's motion is computable in principle but your knowledge of the initial conditions needs to be absurdly precise.
This result is not surprising to me - it would be much more surprising if you *couldn't* make a universal computer this way. Universal computation seems to be a very prevalent feature of sufficiently complex systems. But still it's very nice.
I'm Not a fan of Christmas music whatsoever. And so far I've not heard a single note of Last Christmas, nor many others. However: I'm heading out to grab some groceries shortly, and experience has taught me that the supermarket's Christmas playlist hates me. It has sent me to Whamhalla numerous times before. Wish me luck!
This is my take on the Donny Hathaway classic, 'This Christmas' redone with Audio Modeling SWAM Clarinet, alongside ProjectSAM Swing! and Swing More! Librari...
@Onj That’s indeed better. One song I don’t really mind is this version of 12 days of Christmas, because it’s musically nerdy and mostly instrumental, except the throat singing section. youtube.com/watch?v=g-vCauX4lY…
Provided to YouTube by Universal Music GroupTwelve Days Of Christmas · Bela Fleck & The Flecktones · Alash EnsembleJingle All The Way℗ 2008 Rounder Records, ...
The Twelve Days of Christmas in 12 Keys & 12 Time Signatures Bela Fleck & Flecktones featuring AlashJingle All the Way Tour 2025Béla Fleck- BanjoSierra Hull ...
Let's take a look back and remember some of what this year brought. commits At more than 3,400 commits we did 40% more commits in curl this year than any single previous year! Since at some point during 2025, all the other authors in the project have…
I consider DoH in browsers to be data collection enabled by default. I always customise the browsers to use a DoH server I trust or disable DoH entirely.
I really want a local Shazam equivalent for working out where the sound of something came from. I have a sound from a keyboard that I recorded years ago, and I can't find it in the plugin version of that keyboard. There are preview files for all the sounds, so I want to run an audio match tool against all the preview files, using my original sound as a base. Even if it's not quite the right key or pitch, if there are enough similarities, the tool should surface the closest available options.
Don't forget the Meta platforms (FB, Insta, W'app, threads). Mark Zuckerberg is possibly greedier and more evil than Elon, who is merely stupid. Smart organisations move to open source digital, and out of the "cloud".
The CBS 60 Minutes CECOT segment showed up in Canada's Global TV app and then quickly disappeared. But a direct copy (not a screen recording) is now archived online: archive.org/details/60minutes-…
This is a screen recording of a 60 Minutes segment about the Centro de Confinamiento del Terrorismo (CECOT) prison in El Salvador, which was intended to be...
This is a screen recording of a 60 Minutes segment about the Centro de Confinamiento del Terrorismo (CECOT) prison in El Salvador, which was intended to be...
#Introduction for my (recreated) solo instance 〰️〰️〰️〰️〰️〰️〰️〰️〰️〰️Hello! I go by Leora online, late-20s, pronouns she/her ❥ I am happily married! ❥ I am a 🐈 #cat mom to 3 and a 🐶 #doggo mom to 1 very crazy #whippet. I will likely be posting a lot of pictures of my #cats Saria, Majora, and Link, along with the occasional Mabel picture.
❥ I am a Linux Server Administrator. ❥ I currently live in Arkansas in the US, but I'm trying my hardest to relocate out of the South for obvious reasons. ❥ My memory is absolutely terrible, can't really recall a majority of my past. ❥ I’m reasonably confident in my programming ability; my projects are often less so. ❥ Took a 🇯🇵 #Japanese language course in high school only to forget it all the next year. Currently working on relearning what I've forgotten.
❥ Use to be super involved with 🎭 #theatre. I was in a mime troupe in high school; I learned mime stuff, stilt & ball walking, small bit of 🤸 acrobatic stuff, diabolo stuff. I was active on stage and behind the curtains for several shows throughout high school and college. My social anxiety has mostly put an end to that chapter of my life, though.
❥ In my free time I'm either tinkering with my #homelab, watching #jellyfin or youtube, reading #manga or #webtoons, playing video games (sometimes on #Twitch), or taking 📸 pictures of my cats.
Feel free to add me, I enjoy meeting new people!
I’ve been using a Monarch for a year now, and it’s been the busiest year of my working life. But right now, I’m on vacation. I expected to, and very much do, enjoy the benefits of multiline Braille for various work things, and I congratulate Vispero for doing a super job with Monarch support in JAWS. But now I am appreciating that there’s something very special yet intangible about curling up on the couch, grabbing a book from Bookshare, and reading for pleasure on the large display. I’m not sure I am able to articulate the full extent of the difference, it’s just more fluid, it’s better, it’s more relaxing somehow. I think people will be very pleased with some of the updates that are just around the corner, and of course we’ll continue to articulate the value proposition, the improved outcomes, the equity, of multiline Braille to entities who can fund them and put them under the fingertips of more blind people. So, back to my book.
I definitely want to take a look at a Monarch at some point, but so far, nobody is selling it in Germany yet, as far as I know. I'm curious about its size and weight. The one multiline Braille display I saw at my last CSUN visit several years ago, was quite big and quite heavy. You'd curl up on the couch with it just fine, but do not expect to get up again any time soon. LOL! I suppose the Monarch isn't as heavy and bulky?
Absolutely! I've always been a fan of the Help Tech displays, I got my first one with my first JAWS license in 1996, but the Activator is something special.
I strongly agree with these observations. With the Monarch, I have the advantages of a refreshable display and many of the advantages of embossed documents. I think more could be done to support reading scholarly material. For example, Victor Reader on the Monarch doesn't yet provide hyperlinks to footnotes and bibliographical references, even if properly marked up in the source, and document formatting could be greatly improved. However, this is already a huge advance over single-line displays, and developments are moving in the right direction. I definitely don't want to go back to the old ways.
@elettrona They’re quite different devices. Dot Pad needs to connect to something and has no storage or input, Monarch is fully self-contained and has a larger display. Dot Pad works very well with iOS and JAWS though and has a fast refresh rate.
I have more info about dotpad than monarch, I'll check it out, as they're both very expensive devices; when you have to rely on an aid, it's better to trust it. And to be really honest, I'm scared. I don't want to offend you as an American person, but the path United States are walking, makes me feel not so optimistic regarding the future of USA-based assistive tech.
@elettrona In the end, it may come down to price for you. Monarch is double the price of Dot pad, but Monarch has a larger display and does so much more stand-alone, so they serve different market segments.
@elettrona With the Braille Terminal application on the Monarch, you can also use it to access another device via a screen reader (NVDA and JAWS currently support the Monarch, with more expected to come). Thus you have both the internal KeySoft applications and screen reader support. The Dot pad is just a display device - it must be used with a screen reader or a specialized application running elsewhere and attached via USB or Bluetooth. I prefer the larger display and the ability to run applications on the device, so the Monarch fits my usage better. As Jonathan noted, the cell technology is similar (supplied by Dot Inc.) in both cases.
1. User complains to #hackerone that I named his *previous* name when he renamed himself to a silly name after I banned them in a #curl report filed back in October.
2. Hackerone asks me to respond on their support forum, on which I have no account. Grrr. I refuse to.
3. Replying to the hackerone email about this instead, I get a bounce saying they don't accept emails on support@hackerone ...
## Summary:
An integer overflow vulnerability exists in the `doh_probe_write_cb` function in `lib/doh.c`. This function is used as a write callback for DNS-over-HTTPS (DoH) responses. When a...
I really appreciate your ability to stay level headed on these threads, give people the benefit of the doubt (even when it's debatable whether they deserve it), and not just start cussin'.
In 2009, I was with a friend in West Virginia, who had a gig playing for a school music program for 5 through 10-year-olds.
The youngest of those kids are now old enough to drink in the United States. Maybe not quite college graduation, at least with a four-year degree, but close.
In a few days, the very youngest licensed American drivers will have been born three years after the first iPhone was released. This includes one of my cousin's grandchildren.
I woke up this morning, went into the kitchen, collected some spoons1, it's time to roll up my sleeves and release #iocaine 3.1.
Drank 0.4l of caffeinated hot chocolate, and topped it with two pellets of Ferrero Pocket Coffee. So I'm like wide-eye awake for the next hour, and I'm hearing colours2. ↩︎
iocaine's soothing green is, as expected, soothing, like a small breeze upon a field of wild grass. My burning red-yellow rage is the most metal sound you ever heard. ↩︎
Aww. I want lots of hot chocolate with coffee. I should make that, problem is there's no hot chocolate in the middle of a village in the middle of nowhere. Good luck, go you go go!
Как те, которые бояться потревожить россию, изъять ее деньги, сбить их дрон, наказать за перерезанные кабеля, трубопроводы, закрытые аэропорты, взорванные рельсы, сгоревшие торговые цетры…
Das 240. WildMics-Special war unser Jahresabschluss 2025. In einem Ask us Anything konnte ihr uns Fragen zu Politik, Gesellschaft und dem Leben stellen. In Zusammenarbeit mit dem Bund Europäischer Jugend - Österreich. Mit @SophiaKrappweis & @TommyKrappweis hoaxilla.com/wildmics-special-…
Ich habe diese Folge vorhin bei der morgendlichen Haushaltsroutine gehört. Sophias Bemerkung, Tommy möge doch bitte ihre Atmer nicht persönlich nehmen, traf mich völlig unvorbereitet und löste einen sehr lauten Lachflash aus. Vielen Dank dafür! Und vielen Dank für das, was ihr alle tut! Ich höre euch sehr gern zu und finde Trost darin, dass es Menschen gibt, die sehr ähnlich denken wie ich. Euch allen einen geruhsamen Jahreswechsel,
Oh ja, das tut er sehr oft, weil es auch bei ernsten Themen immer wieder auflockernde Momente gibt. Ich kann euch fast nie live verfolgen, weil mich der Input von Twitch, dem Livestream und dem Chat gleichzeitig akustisch überfordert (Blind mit Screen Reader, also Sprachausgabe),daher beteilige ich mich nie im Chat, aber ich höre euch sehr gern nach und schicke dann halt auch mal hier ein Feedback.
A short while ago I was explaining to a client the reason why his website has seen a very significant increase in load over the past few weeks. On many of the servers I manage, I see connections coming from residential users, yes, but mainly from unscrupulous AI scrapers that do not care about overloading everything. Someone on their side replied to me: "Wouldn’t moving everything to the cloud solve the problem? I mean... autoscaling!"
That sparked a somewhat "conspiratorial" thought. Some colleagues have told me that, with the recent surge in AI bot traffic, their costs have skyrocketed. Given how much money is being spent, and not earned, on AI, are we really sure this is not, in some way, planned or at least welcomed as a way to increase costs for end users of cloud services and funnel them into AI budgets?
Europe is making a list, checking it twice… and upgrading its trains to high-speed – nice! 🚄
While you are sipping hot chocolate, we are cooking up 200+ km/h rail magic:
✨ Faster journeys: Berlin → Copenhagen in 4hr instead of 7hr 🚆 New routes: Lisbon–Madrid–Paris and Tallinn–Riga–Vilnius–Warsaw 🌙 More night trains for long trips 🎫 One ticket for your whole journey, and easier booking across borders
By 2040, your holidays might just start on the rails.
Also you might want to add alt text to that, so if someone stumbles on it, they aren't confused. "Screenshot of an alert for icloud, saying that more than 1000 items cannot be downloaded, with an okay button." Don't worry, I didn't run your image through anything, voiceover is smart, sometimes, so it prints out the text if it detects it, mostly works in English. :D
uh, yes, but when you attached that photo, it actually gave more meaning, except, if someone's net is too slow, or images don't load, or it's someone who can't see them, alt text isn't bad. Imagine when I wanted alt text ten years ago, because my internet was slow, and the little I could see on images, I couldn't. It's basically blur, still is, but yeah. Also, wow, Apple. Can I make a thousand personal voices and try to export the 1000 files? LOL. I am now tempted to ask if what you were downloading were photos&videos, or files? Because if it's the first, I am jealous. I love taking photos, but I don't, because they turn out a bit too ugly. :D
@tardis it was photos... i'm trying to download my old photos: I moved away from iOS some time ago, but my kids didn't, and they are filling up the iCloud, so I thought I'd free up some space... well guess I won't. (I could at least delete the videos, which are probably the biggest portion of my photo archive.)
Ah, you can go on the web, and download them from there if access from the web is allowed. Thing is, I love iOS, cause, accessible. I have ditched Windows and Android a long, long time ago, only use them for work, and to stay on top of the line with new accessibility stuff, else they're just collecting dust somewhere.
@tardis that's exactly what I tried, and got the "max 1000" error. and I refuse to spend time on downloading them in chunks of 1000 (at least until the iCloud completely clogs up with children's iphones' backups.)
Ah screw that. Suck it, apple. LOL. See, I don't love iOS or macOS that much, I just love it for the accessibility. Else, Apple can suck it, a lot. LOL.
@tardis I love the accessibility of my iPhone too - I can access it easily by taking it out of the "misc" drawer (and charging it, because iPhones tend to drain the battery even when they are switched off.)
Hahahaha, I love that! Yeah, they do drain, I have a 13 pro I have to charge, its battery has degraded some what, so it's not a good usage as a phone anymore, but using it as a backup, since I give all my old devices to family members, and it came a time when my phone collection was no more. :D
So there are apparently people creating personal voices on Macs, because it's more flexible than on the iPhone. I have heard several times that people have even cloned all sorts of random voices. But since you can't upload audio files for recording, how exactly are you to do this, manipulate the folder, where ever it is, or whatnot? #personal_voice #accessibility #a11y #voiceover #mac
Honestly, I don't know and I haven't tried it. Now you gave me an idea, it's bad when I have ideas. :D Manipulating the folder? I am not sure, unless Mac saves it somewhere in an actual folder, but as far as I know, it's synced via iCloud, so if I were to create a personal voice on Mac, it will also be on the phone, which, to be fair, is fun. I should. Told you, giving me ideas is bad.
Lately I’ve been realizing that I probably made a (small but important) mistake when choosing Proton Mail over @Tutanota Mail.
I truly respect and like both European companies, but what I’ve been noticing, and deeply appreciating, is how strong, healthy, and direct the relationship between Tuta and its users is. They communicate, they discuss, they listen.
It actually reminds me a lot of the @Vivaldi browser team, including @jon himself. And that kind of attitude is priceless, especially when it’s combined with such a clear stance on privacy & security, AI, and user-first values. That combination is simply top-tier.
I recently noticed Vivaldi starting some cooperation with Proton. Personally, I’d much rather see a collaboration with Tuta.
Rooting for you. Rooting for both teams (actually all three). Absolutely fantastic work.
Die Bundesregierung will Internet-Zugangs-Anbieter verpflichten, IP-Adressen aller Nutzer für drei Monate zu speichern. Das geht aus dem Gesetzentwurf zur Vorratsdatenspeicherung hervor, den das Justizministerium veröffentlicht hat.
Danke für den Hinweis, wir schauen uns das an. Fraglich, ob es dieses Mal Bestand haben wird... siehe: tuta.com/blog/data-retention-g… Wir lehnen jede Form der anlasslosen Massenüberwachung ab und werden sicher dagegen aktiv werden.
Whatever else the future may hold, I'm hoping for a decrease in the yelling of "stochastic parrots!" in literally any conversation about so-called AI by people who wouldn't otherwise use the word "stochastic" in any other context. Memorising a phrase containing an obscure word and deploying it like a hand grenade to shut down a discussion has never struck me as a particularly helpful or inclusive strategy.
Joshua Rogers on his bug bounty experiences in 2025.
Positive for #curl, kafka-esque for all others mentioned. ‚BugCrowd‘ seems to a typical level-1 support company living on denials.
(Joshua also reported on Apache and pbly other projects where he could talk to the maintainers. I take #curl here as an example for FOSS projects interested in actually securing things.)
I just read a recent post about using LLMs to write code, whose author is still ambivalent on the subject but concludes that they've made their peace with AI, for better or worse. (Not naming, linking, or quoting verbatim, because my intent isn't to single out this person.)
This just makes me respect all the more the minority who still have the moral courage to keep resisting generative AI due to ethical concerns. And it makes me want to be one of them.
In the recent post I referred to, the author says that an LLM enables them to write an "insane" quantity of code.
Yes, I think it _is_ insane. A proficient typist can probably already type faster than most of us can read _and think_. An LLM can generate tokens way faster. So increasing the quantity of code probably isn't good for having code that's well understood and reviewed.
Lol. Reminds me of that AI-assisted patch someone sent me to fix AT-SPI always using the first keyboard layout under X11, regardless of what was currently in use. It added quite a bit of code, but it was trying to replicate logic that belonged elsewhere and was causing problems, and the correct fix turned out to involve only adding a few lines of code. I will say that this person's patch helped me to figure out how to think about the problem, and eventually I found the right blog post that helped me figure out what I needed to do.
"A set of supercritical carbon dioxide (CO2) power generation units commenced commercial operation on Saturday in Guizhou Province, southwest China, marking the country's pioneering commercial application of this innovative technology."
Laut Oxfam würde es jährlich 37 Milliarden Dollar kosten, um sowohl den akuten, als auch den chronischen Hunger auf der ganzen Welt zu beseitigen. Elon Musk allein hätte im Moment die Mittel, um das zwanzig Jahre lang zu tun. Nur um mal die Relationen zu klären.
@dnkrupinski Fast. Silvester muss noch vorbei gehen. Dann geht endlich Routine wieder los. Und während viele Menschen diese nicht mögen, bin ich gerade in einer Situation, wo die Unterbrechung der Routine für mich keine Erholung darstellt.
No sé si es que uso los LLMs de forma distinta a otra gente, pero a mi me dan muy buenos resultados como vehículo de aprendizaje que me sería imposible (literalmente) hacer de otra forma. Tut largo, pero es que lo tenía que decir.
Os contaré mi experiencia concreta. Llevo tiempo en un plan para mejorar mis matemáticas. Con un LLM he podido aprender matemáticas formales de forma rigurosa, verificando las pruebas por mi mismo, algo que sin esa herramienta me era materialmente imposible por razones de accesibilidad.
Después de unos meses, he conseguido aprender un montón de lógica que no sabía (CNF, BNF, prenexado de fórmulas, semántica, el teorema de la solidez de la deducción natural sobre fórmulas de primer orden, por inducción en la altura de la derivación). Estrategias de prueba: contrapositiva, por contradicción, por inducción, inducción fuerte y estructural. En análisis real, entiendo bien los límites de las secuencias y funciones, la continuidad, los teoremas fundamentales como Bolzano, convergencia de una secuencia monótona acotada, valor intermedio, valor extremo. No sólo los entiendo sino que puedo escribir por mi cuenta las pruebas de los teoremas. Conseguí entender como funcionan los números complejos como representantes de una rotación del plano, raíces de complejos, y porque toda la trigonometría se simplifica enormemente con estas herramientas. En álgebra lineal, aprendí como calcular determinantes recursivamente, Gauss-Jordan, Gram-Schmidt, proyección, QR. Pero aún más importante, aprendí a entender el álgebra lineal desde el punto de vista de una aplicación lineal T sobre un espacio vectorial V en un campo F, independiente de una base, en vez de partir de la representación concreta de una matriz. Aprendí que el determinante de una aplicación lineal es una función multilineal, alternante, normalizada, de la imagen de unas bases ordenadas a un escalar, que representa la ratio del volumen orientado de la aplicación lineal. Hice todo el tema de expandir una función de este tipo sobre bases en 2d ((1,0), (0,1)) y vi salir de allí las permutaciones, hasta la típica fórmula de determinante de una matriz 2x2 (ad-bc).
Todo esto no fue fácil, no fue libre de frustración. La gente muchas veces dice que el objeto de un LLM es eliminar la frustración y la dificultad que dan lugar al aprendizaje. Pues no sé, si se usa así a lo mejor sí, pero tuve días en que pensé que no podía con esto. Yo siempre insisto en reproducir las pruebas, hacer los cálculos, hacer tests y ejercicios, y que se me corrija con rigor. En alguna ocasión el LLM me dijo, "tienes un conocimiento suficiente para continuar," a mi no me lo pareció, y seguí trabajando hasta que entendí el tema satisfactoriamente.
Las posibilidades de haber hecho esto sin un LLM, para mi, son cero. Porque me habría sido imposible encontrar material accesible (no es la primera vez que lo busco) y todavía más difícil poder preguntar dudas y que alguien me corrija los ejercicios. No se trata de que El LLM sustituya el esfuerzo intelectual; sino que suple la ausencia de material accesible y la posibilidad de corrección.
Por eso no me convence la afirmación general de que los LLMs son inútiles para aprender, o para x. Son herramientas muy fáciles de utilizar mal, y no son perfectos; en algún caso me dieron resultados erróneos (cosa que por cierto los profesores también hacen), lo cual incide en la necesidad de verificar todo; pero es que ese es mi método en todo caso, cuando aprendo matemáticas intento verificar todo. Quizá por eso me haya ido bien.
En resumen, esta tecnología me ha facultado para aprender cosas que hacía años que quería aprender y que hasta ahora siempre me había sido imposible. A la gente no le gusta la palabra democratizar en este contexto, pero me cuesta no usarla.
#homelab, watching 
#jellyfin or youtube, 
reading #manga or #webtoons, 
playing video games (sometimes on #Twitch), or taking 📸 pictures of my cats.
But that's not too bad, I expected more.
Tuukka Ojala
in reply to Teemu J • • •