yc-kraln

It does--but a particular form of Capture The Flag where there is a computer system and the "capturing" is breaking in or exploiting a security issue in that system.

Claude does much better when I tell it to use ripgrep and ctags ;)

How do you handle noisy neighbors?

every line of code is a liability. open, closed, doesn't matter. companies will have to treating it that way--which means actual engineering--or they will get burnt, and hard.

> Working software over comprehensive documentation

this is 100% backwards for anything safety-critical or that needs to be maintained past a butterfly's lifetime. this is what encourages yolo-driven-development instead of considering what actually should be done, and this is why agile or Agile or whatever formalization or bastardization of it can not be considered software engineering, but merely code monkeying.

Great that these are source-available, but I wish they had a human copy edit the README, it's unbearable.

I would love to use the drive units from a Tesla in a conversion project. Unfortunately, they're cryptographically paired with the main computer, and there's no way to use them.

What a waste.

Can't wait to see the Sailfish/Motorola crossover, honestly.

Interesting! I am getting tired of looking at Roundcube and having weird issues and was thinking of doing the same. Were you planning on making the result public?

It's really difficult for me to understand the level of cynicism in the HN comments on this topic, at all. The amount of goalpost-moving and redefining is absolutely absurd. I really get the impression that the majority of the HN comments are just people whining about sour grapes, with very little value added to the discussion.

I'd like to see someone disagree with the following:

Building a C compiler, targeting three architectures, is hard. Building a C compiler which can correctly compile (maybe not link) the modern linux kernel is damn hard. Building a C compiler which can correctly compile sqlite and pass the test suite at any speed is damn hard.

To the specific issues with the concrete project as presented: This was the equivalent of a "weekend project", and it's amazing

So what if some gcc is needed for the 16-bit stuff? So what if a human was required to steer claude a bit? So what if the optimizing pass practically doesn't exist?

Most companies are not software companies, software is a line-item, an expensive, an unavoidable cost. The amount of code (not software engineering, or architecture, but programming) developed tends towards glue of existing libraries to accomplish business goals, which, in comparison with a correct modern C compiler, is far less performance critical, complex, broad, etc. No one is seriously saying that you have to use an LLM to build your high-performance math library, or that you have to use an LLM to build anything, much in the same way that no one is seriously saying that you have to rewrite the world in rust, or typescript, or react, or whatever is bothering you at the moment.

I'm reminded of a classic slashdot comment--about attempting to solve a non-technical problem with technology, which is doomed to fail--it really seems that the complaints here aren't about the LLMs themselves, or the agents, but about what people/organizations do with them, which is then a complaint about people, but not the technology.

What costs are you calculating with that, net of taxes and rent, you can't survive as a single or even couple on 1800€+ in Berlin?

Like the median income here is extremely low, that salary puts you in the top 10%.

Help me understand because this doesn't make sense at all to me--context: living in Berlin the last 13 years

Especially as congress has, for as long as I remember, been about the superset of infosec, society, art, etc. IMHO it's more along the lines of complaining about any ride that isn't a roller coaster at a theme park--no one is forcing you to go on any rides, other people clearly enjoy them, they're not taking anything away from your roller coaster, and having them increases the diversity of the crowd in an ultimately positive-for-everyone way.

Some people just like to complain that they have to take a shower and can't harass women like they used to like they could when congress was at the BCC and that kind of nonsense didn't immediately get you thrown out like today.

I came here to post this. We make a lot of the same sorts of optimizations for our OS distro (debian based) -- disabling frequency scaling, core pinning, etc. Critically, CPU0 has a bunch of stuff you cannot push, and you're better off with using one of the other cores as an isolated island.

This is what the scheduler latency looks like on our isolated core:

# Total: 000300000 # Min Latencies: 00001 # Avg Latencies: 00005 # Max Latencies: 00059 # Histogram Overflows: 00000

(those are uS!)

My big issue with this study is it points to a cause. How can they know the issue is social media, and not, say, the climbing atmospheric CO2 or other long-COVID related issues?

Got one of these as well, a powershell one which had a whole complex rootkit in it that bypasses the built-in virus detection. Super nasty!

Sounds like you want Molly

The stepping didn't kill wifi, the boards they ran with the d0 stepping have likely a different wifi chip (it's connected externally) or similar (maybe the wifi chip has a different stepping?), unrelated board-level changes.

The d0 stepping boards I have with wifi work with the linux kernel, still.

I get this issue when it uses sudo to run a process with root privileges, and then times out.

I work with Claude more-or-less how I worked with our Indian colleagues; the difference is Claude is improving over time.

For me this is a space that has been long dominated by MRPeasy--it's just such a perfect fit. Happy to see open source solutions slowly catching up.

Definitely had fun as I was scrolling the article with a Trackpoint using the scrolling button. 100% my preferred way to laptop.

I have a somewhat different problem with io_uring in practice: It's extremely hard to use /correctly/. The management of buffers which bounce across a kernel boundary and may-or-may-not end up in the same original thread lends itself to lots of subtle race conditions, resource exhaustions, and ABA issues. It's not that you can't make it work, and work well--it's that it's hard to do correctly, and very easy to make something which works 99.99% correctly, and then fails spectacularly under load or over time.

I can imagine the security implications are the same.

ouch

The answer, of course, is "it depends".

We have something like ~40 repos in our private gitlab repo, and each one has its own CI system, which compiles, runs tests, builds packages for distribution, etc. Then there's a CI task which integrates a file system image from those ~40 repo's packages, runs integration tasks, etc.

Many of those components communicate with each other with a flatbuffers-defined message, which of course itself is a submodule. Luckily, flatbuffers allows for progressive enhancement, but I digress--essentially, these components have some sort of inter-dependency on them which at the absolute latest surfaces at the integration phase.

Is this actually a multi-repo, or is it just a mono-repo with lots of sub-modules? Would we have benefits if we moved to a mono-repo (the current round-trip CI time for full integration is ~35 minutes, many of the components compile and test in under 10s)? Maybe.

Everything is a tradeoff. Anything can work, it's about what kinds of frustrations you're willing to put up with.

The article reads like sour grapes: It's always someone else's fault.

It was your company. If you knew the press in the US was better, /why didn't you expand to the US/? /Why didn't you raise money in the US?/ Companies are ultimately transnational, the market for capital, talent, and eyeballs is global, and if you weren't getting what you wanted from the European press, why didn't you go after the American press?

There are some structural things which make things more difficult than if you're living in SF, LA, or NYC--I'd argue more strongly it's the lack of pension funds investing as LPs in funds which results in less overall money in the ecosystem--but there's a LOT of the US which is not those places, and I would strongly argue that it's easier to be a startup in Berlin or Paris than in Chicago or Dallas.

Systemd is great if your use case is Linux on a modern Desktop or Server, or something which resembles that. If you want to do anything else that doesn't fit into the project view of what you should be doing, you will be met with scorn and resistance (ask the musl team...).

What isn't great, and where the hate comes from, is that it makes the life of a distribution or upstream super easy, at the expense of adding a (slowly growing) complexity at the lowest levels of your system that--depending your perspective--does not follow the "unix way": journalctl, timedatectl, dependencies on/replacing dbus, etc. etc. It's also somehow been conflated with Poettering (he can be grating in his correctness), as well as the other projects Poettering works on (Avahi, Pulse Audio).

If all you want to do is coordinate some processes and ensure they run in the right order with automatic activation, etc. it's certainly capable and, I'd argue, the right level of tool as compared to something like k8s or docker.

I'm working on inexpensive, mass-manufacturable fully autonomous drones.

It's banned because it leaks information (location, for instance) which is exactly the kind of thing you don't want leaking from your armed forces.

> But short of that, are there not only two root causes left?

No.

Let me paraphrase the conversation I had with the head of obestity at one of Europe's leading research hospitals (top 5 in EU):

Once you get a certain amount of overweight, regardless of the cause, there are so many regulation systems in your body (at least 20 that we know of) which kick in to keep you overweight that it is not possible to diet and exercise your way back to a healthy weight without picking up a new disease (eating disorder, exercise additiction, etc.). The people who manage it spend the rest of their lives obsessively weighing everything they eat, exercizing every day, and even then if they slip up or lose mobility or something, inevitably they will become overweight again.

The only known medical treatment is operating on the stomach. This changes enough of the regulation systems (including the hormonal ones) that it has a reasonable chance of sticking. GLP-1 seems to offer a second option.

It continues to astonish me that eating disorders which result in weight loss are (correctly) accepted by society as diseases with complex causes which can include lifestyle and self-control but also mental health and innumerable other things, but if you're overweight clearly you are a sad sack of shit that just can't open their mouth without shoving food in it.

The electrical wiring in cars is Conway's law manifest in copper.

For those unfamiliar with Conway's law, I am arguing that how the car companies have organized themselves--and their budgets--ends up being directly reflected in the number of ECUs as well as how they're connected with each other. I imagine that by measuring the amount of excess copper, you´d have a pretty good measure for the overhead involved in the project management from the manufacturers' side.

(I previously worked for Daimler)