djriley

I feel like this is quite a different situation. Building a nuclear weapon requires a lot of things to happen in the physical world that we can watch for. Weapons grade fissile materials must be sourced or refined, facilities must be build, you can't just do this in any machine shop. Various experts must be either poached or subverted or home grown. There are a lot of people with mouths involved. Then once you get past all of that you either have to test or just hope for the best at which point you will certainly be detected and likely handled with extreme prejudice.

Digital intelligence seems much less controllable. Near frontier weights are already out there, you don't necessarily need massive datacenters if you're not trying to serve commercially. Even if we we're somehow able to control what OTHER nations do through either treaties or threats of sanctions/force, working on this stuff is easily in range of many of the larger criminal orgs. Beyond all of that, there's always the chance that weights from authorized providers find their way "out the back door".

We've seen with crypto and darknet markets that even if you can fight this kind of stuff and catch bad actors, the most actual change you're going to effect is a temporary slowdown and the next guy who tries is going to be that much better. I don't doubt that regulation is going to be the initial route this goes, mostly due to the financial interests of the frontier companies and the desire to keep this build-out solvent for as long as possible rather than any true concern for actual safety and sanity. What's really going to need to happen though, in my very insignificant opinion of course, is we are going to have to change from the ground up how our digital infrastructure is built.

Maybe with the coming changes that is for the best too, our current systems are designed for people and I don't know the current numbers but it seems like agent traffic is already the bigger percentage. We may not be able to build our infrastructure on top of these large mountains of human designed legacy code. Maybe we need to be thinking about new kernels, new servers, new protocols and stacks.

We seem to be very quickly reaching a point where is something is not mathematically provable to be secure then it is not secure at all.

People talk.

That rumor is false, probably because Calif, the firm that discovered the exploit also just published an OpenBSD ASLR bypass people think they must be connected.

Php2shell is completely language level. By the time the OS sees it it's just a function of the webserver.

I think Pascal, as in it's modern dialect Object Pascal used in Delphi and FPC/Lazarus is a sort of "hidden language". You won't see many big Open Source projects using it and there's only a handful of big commercial packages using it, but behind the scenes for domain-specific/in-house software it is used quite a bit. It allows small teams or individuals to very quickly put together software that is used by the business departments or engineering/technical with modern GUIs and good performance. It's exactly what it says on the package, "Rapid Application Development". Stuff like Rust and Go are great, but having to code out the entire UI by hand basically doubles the maintenance/development burden and requires a lot of knowledge about the widgetsets and their libs. QT Creator, Glade/Cambalanche, wxFormBuilder are nice, but there's a decoupling there that adds complexity and they all require C/C++ if you want native code. WinForms/WPF on .NET in Visual Studio probably come the closest to what you get, but it's mostly Windows specific and I wasn't able to pick it up easy like I was Delphi.

FPC and Lazarus are really amazing tools. FPC itself is quite a competent compiler with many available platforms including legacy platforms, Java VM, WASM. Very fast compilation, run time perf is not quite C++ especially with strings, but fast enough. Even Lazarus LCL programs are quite lean and efficient for what they are. The RTL/FCL are comprehensive with tons of functionality, things like string handling, containers, math, all there and everything you expect.

The killer app with Lazarus is it's cross platform GUI capabilities. The last release is still stuck on Gtk2, but trunk Gtk3 is default on Linux with QT5 and QT6 available, Cocoa, and Win32 and for the most part it just works. You build for GTK or Qt, you switch to Win32 and compile a native windows app. The RTL abstracts most of the OS stuff like file handling and threads, and the LCL handles the widgetsets. There's nothing else even close when it comes to cross platform development of desktop apps.

That's interesting, I thought Claude or Codex would be really good with it. I've been doing some FPC/Lazarus work with Gemini 3.8 Flash, working on a program that creates STLs, DXFs, and SVGs of involute gears for machining or 3-d printing with a nice "Gear Designer" and visual preview and the code has been perfect. I enjoy building the UI and write the UI logic myself, but the exporters and gear calculations are beyond my capabilities and everything Gemini's done has been basically flawless. I have no dependencies beyond the RTL and LCL, not using any libraries to help with the format exports. Seem's like the opinion here is that Claude and ChatGPT do way better code than Gemini, so I'm surprised you've had issues.

The only minor problems I've had is Gemini went through a phase where it was using semi-colons improperly in if statements, which is something everybody messes up with Pascal, so forgivable, I'm sure if I was using an agent rather than the web interface it would have figured it out from compiler errors right away. That and sometimes with UI stuff it would hallucinate methods or properties for components that are part of commercial delphi components, but I've only ran in to this 2 or 3 times.

When I first started on this project I believe we were on early Gemini 3 and I will say it's handling of Pascal constructs was quite rough, it seemed to have a lot of trouble with the verbosity. Lot's of extra tokens to grok the structure vs something like C or Go. But from 3.5 onwards it's been flawless.

Maybe it's just one of those things Gemini is simply better at. I'd much rather read Gemini prose than Claude, and Pascal is definitely closer to natural language than something like Rust might be. On the other end of the language spectrum I find it is incredibly good with Go as well, I think Google might have a bit of an edge as far as high quality training material for that language goes.

Awesome! I always wanted to try browsing the web like RMS ;)

Needs to let the player know how many they got at the end! Maybe a hi score board!

Something I feel was sort of left out here was precision grinding which began to come in to play right around the same time period this article focuses on. Grinding is what really allowed the industrial revolution to explode. Tight tolerances and good surface finishes allowed things to run faster and more efficiently. Automation and repeatability lead to high production of tooling and products that eventually lead to mass production.

What I do find interesting is GT3 cars have both ABS and traction control these days and they have great racing. I believe prototypes have TC but no ABS. I realize the mix of talent is much different, especially in GT3 where there are Pro/AM teams and stuff like that, but I'm not sure TC makes that big of a difference on the racing. I don't think ABS should ever be allowed in open wheel as it inspires too much risk taking in overtakes. The whole reason NA F1 revved so high was partially to help keep the power band reasonable exiting slow corners.

I mean that's pretty much the IndyCar spec and they have pretty decent racing. They've done things like embracing underbody aero to cut down on dirty air to allow them to follow closer, low downforce specs on ovals to force them to race the track. Even a spec car, a drivers ability to communicate setup issues to the team, the team to implement them, the budget to do it. F1 is some of the best in the world, but none since Mansell have really came over and dominated, not even Alonso.

Current IndyCar is pretty good compared to the IRL years, this isn't an argument to make F1 a spec series. I was IndyCar for close exciting racing, I watch F1 for the engineering battle as well as the drivers. I think for a lot of F1 fans, the off track drama, the development of the teams through out the season, and the cool ideas they come up with to try to push the boundaries are about as good as the on track entertainment.

The spectacle is still very important, there is nothing like seeing F1 cornering speeds in person. IndyCar isn't close, they have their oval thing with ultra high sustained top speeds, but not even close to F1 on a road course. When you see how fast an F1 car can change directions, it genuinely boggles the mind that a human is behind that. And they've really dropped the ball in that aspect this season. The fans aren't happy, the driver's aren't happy. Something needs to change.

You can use it to write a plug-in to change the name to YAH Harness!

Cute, I love this style! It's like a long Aesop's Fable! I can't wait for the sequel!

As far as number one goes, the tooling has almost certainly been recycled when they say "destroyed". By recycled I mean in to new steel in an electric arc furnace. It's pretty common even in every day mass production for tooling to either be sent for destructive testing or a destructive decomissioning/EOL use (cutting, defacing, used for practice for facility apprenticeship, etc). I imagine something on the level of Blackbird tooling would have gone through a very strict decomissioning process before it even hit the scrap bin.

As far as #2 and the rest goes, I agree that with enough money you could make anything happen. However, sometimes recreating something so complex could be a "not enough money in the world" type situation. Sure, drawings and specs exist, but even on today's 3-d modelled assembly you're not typically modelling every single part to full accuracy. You have things that are off the shelf that might be a black box drop in, you have things you contract out that are built to spec. You know what you get, but not necessarily exactly how it's done. Then in your own facility there is a certain level of institutional knowledge that is not necessarily documented in such a way you can just recreate it.

Finally, one of the biggest hurdles, there's are not things that you can just throw an LLM and some engineers at. The actual trades themselves have changed so much since the time of production your only option is to call in retirees. The machinery is different now and so are the workers.

I feel like with a large budget they could definitely get an SR-71 in the air for testing and exhibition purposes, Id love to see it happen. I guess it really depends on how much was removed and destroyed when the machine was decomissioned and how much damage lack of maintenance, environmental factors and age have taken their toll. The good news is it hasn't been operated in a long long time. The bad news is it hasn't been operated in a long long time.

Source: I am a toolroom machinist at a material science lab in the aerospace and defense sector.

It's in Delphi, Free Pascal devs are still pretty resistant to the idea and it likely won't be coming to fpc any time soon.