steelframe

This event was pretty chilling for me because I just took the Renfe line from Seville to Madrid a couple of weeks ago while visiting Spain. I generally like to think that I have a pretty good instinct on relative risks and make reasonable decisions based on that. This time around, between the incident earlier this year and this most recent incident, depending on how you interpret and apply the statistics (of course that's always up for healthy debate), it looks like it would have actually been safer driving between Seville and Madrid. That's a pretty low bar in terms of safety!

I set my favorite agent on the task (Opus 5.5 in the Kagi Assistant), and given the two rail crashes in Spain this year, these are the stats it came up with for "chances of dying" via various modes of transportation between Seville and Madrid:

* EU high-speed rail, long-run average: ~1 in 36 million

* This rail line, counting both 2026 crashes: ~1 in 400,000 to 1.2 million (prior to the most recent crash it was 1–3 million; now ~2.5× higher)

* Flying with a European airline: ~1 in 80 million

* Driving: ~1 in 1 million or worse

Of course this is hindsight. Heightened awareness of risks, higher propensity to overreact to anomalies or alarms, procedural changes, and so forth in the wake of this second crash might push high-speed rail in Spain to be higher than the European average in the future.

Not necessarily an endorsement, but one of the more popular ones: https://www.hetzner.com/dedicated-rootserver/matrix-gpu/

AI wants to be free!

https://allmodels.io/llm/abliterated

I switched entirely over to Asahi Linux on my M1 Macbook Air and installed a third-party 55Wh battery, as the original battery had fallen below 80% capacity. That machine is good to go for another 5 years.

There are no features or performance characteristics of any of the subsequent M[2-6] Macbook Air or Pro models that I care about. I'm not really sold on running models on the hardware I have right in front of me. So long as I can rent an RTX PRO 4000 Blackwell at about half a buck per hour through a European provider, I'm fine using that with Qwen3.8-27B UD-Q4_K_M for the few things I consider more privacy-sensitive.

What if the company has provided the harness, tool execution environment, MCP servers, and LLM provider subscription for the employee to use? What if the company has provided a set of skills and other prompt context that could lead to undesirable behavior in certain unforeseen circumstances? What if the permission prompt the user sees 30 dozen times a day looks like, "OK to execute this random shell script?"

Accountability really is hard in this new world.

No need to involve dang in this. He's had nothing to do with these price hikes.

Wow, you're right. The DMCA happened while I was out of the country doing something very not-tech-scene related, so I completely missed that it passed way the heck back then.

I suppose my point still stands if you are willing to stretch back to Usenet, BBSs, etc. as an inflection point in mass copyright infringement.

When it comes to copyright, I am not so optimistic. When the AI bubble bursts, the usual corporate interests will resurface with a vengeance.

So long as there's big money interests around copyright, I assure you it will not die. And all the predictions about the knives coming out after the AI bubble pops are spot on.

It took a while for the DMCA to happen after Napster. But it indeed happened, and we all know the saying about history repeating itself.

I just use my area code followed by 867-5309. It's quick, non-confrontational, and completely lost on whatever generation is currently manning the register.

I have been using the alias feature for years now with Fastmail. Deleting an alias is hands-down the quickest and most convenient way to unsubscribe from whatever drivel a company ends up sending my way care of their slimey marketing vendor.

Bummer that you got shadow banned.

Now many of us will get to learn how to find meaning in a life where your time is worth about $2-4/hour, which is the reality for about 2 billion people today.

So overwrite with random patterns of dots. You could "arms race" this all the way to the point where you need a scanning electron microscope.

But this line of discussion is entirely missing my original point.

The font color is so close to the background now that I can't even read your edit. Sorry, bud.

Depending on the make and model, send a reverse feed code and then overwrite with block characters. You will know that something bad happened, but you won't know what.

Maybe I am living in a bubble, but no company I have worked at in the past 15 years has cared in the least about either Windows or Office.

For me the analogy doesn't totally hold up. Suppose the CEO of Smith & Wesson were to host a firing range on their own property without adequate barriers in place to keep stray bullets from hitting neighboring houses, vehicles, and businesses. Maybe that analogy isn't perfect, but seems closer to what is actually happening.

It's 3cyl, 1.6L, turbocharged. Trust me, you just have to try it to understand what kind of vehicle we're talking about.

Still driving my GR Corolla with the data communications module fuse pulled.

Or even an HN comment, for that matter.

No doubt it's anything but perfect, but I can give anecdata that my M1 just slept overnight while not plugged in, and I had 99% charge just now when I opened it up to check HN.

My daughter recently left me with her old M1 Macbook Air when she went to college. I decided to give Fedora Asahi Remix a try. It was great seeing that the install process was a simple "pipe curl to sh" from a domain I suppose I have to trust anyway. The most technical part of the install was making sure I waited long enough for the machine to shut down and then holding the power button until "Loading startup options…” appeared on the screen, popping in my local macOS username/password a handful of times along the way while the scripts did their thing.

To be fair, I did in fact get a "Donate to GNOME" interstitial shortly after I first logged in. My agent was able to disable that without too much trouble. I find it obnoxious that GNOME gets to grab my attention for a donation just because they control my UI, while the thousands of other developers who wrote key parts of my OS don't get the same privilege.

That aside, I'm absolutely thrilled with the combination of Apple hardware and Linux software. I can see this being my default "mobile device with a keyboard." It doesn't yet work with my external display, but rumor has it the folks at Asahi having been making strides on that this year. Until then my Framework and Thinkpad both have some more utility as workhorse machines.

I suppose M2 is the most recent platform that has "good enough" Linux support from Asahi, but my M1 fits the bill for most everything I want to do on a mobile device with an attached keyboard. I might pick up an M3 with 24GB to replace it at some point if the Asahi team can get external monitors, GPU acceleration, and sleep working well on that chip.

This is why I pulled the fuse on the Data Communications Module (DCM) on my car. The device that transmits the telemetry simply doesn't get any juice.

What does "use it too much" even mean for someone who is sight-impaired and who needs to use it every time?

I own my own domain in the .us TLD, and I've been using it for my email for the past 20 years. When I applied for an "Electronic Travel Authorisation" (ETA) to enter Great Britain earlier this year the app flat-out rejected my email address as invalid.

Usually, when that sort of thing happens, I just choose to take my business elsewhere or use my time in a different way. However it really ups the ante when a nation-state refuses entry based on that category of technical incompetence.

> A quick search suggests that the most expensive education in the world is something like $100k.

I have a kid in an American university right now, and a quick search of my bank account statements confirms that there are far more expensive educations in the world.

Meanwhile Qwen3.8 27B got both the 'f's and the 'i's questions right.

Gastroenterologists generally categorize feces into ~7 categories. They can be indicators of a variety of dietary deficencies, including dehydration or insufficient fiber intake. Others can be indicators of a variety of medical issues, up to an including colon cancer. Ubiquitous smart toilets scanning and categorizing our feces would move the needle of overall public health and well-being.

I recently had to set up a new Android device for work. Since I keep all my personal accounts separate from my work accounts, I needed to create a new Google account on that phone. I ended up paying $8 for a month of the cheapest service I could find just to get a phone number so I could create that account.