I don't need a standard for this. This is just noise. There are some people who have some sort of mental ailment that makes them obsessively want to introduce "structure" "scheme" "patterns" where it is just innately nonsense. You do you, but stop trying to force it on people.
Reminds me of the "scrum master" adjacent folks who could never cut it writing code and then branched into all kinds of things like "Git Flow" when having never understood Git to begin with. Peak bikeshed territory.
Reminds me of a place I worked at where a "naming committee" had to approve variable names. And no, you could not use "i" as an index in a one-line loop.
Being able to specify word boundaries in search is a basic feature for any developer tool. Vim has *, #, and /\<i\>. grep also has \<i\> and the -w option. LSPs have jump to definition and find references features.
> One letter variables are supposed to be used in scopes that fit on the screen completely.
Exactly, and ideally in less space than that. If you have something like:
for (i=0; i<10; i++) {
foo(i);
bar(i);
}
There is no point in using a "descriptive" name for the index. It's completely obvious what's going on. Anything more verbose would just hamper readability.
Which is a problem, because now people add Claude generated "docs" and Claude still defaults to making super verbose comments. It's inevitably outdated within days and ends up being a trap for future Claude sessions.
It's fantastic for reverse engineering like this. You still need to point it at the right resources of course (USB captures, a vendor driver binary, firmware dumps etc). And of course this mini OLED is somewhat of a throwaway feature to begin with.
And you need to be okay with the usual AI slop - like this sentence will make any kernel developer cringe:
"It is not a DRM display — you don't get a /dev/fb"
If you want it to exist in the Linux DRM subsystem, add a driver for it! It can be done easily for stupid USB framebuffer devices like this.
I feel like German university was already well "LLM proof". Come to class or not, no ones keeping track and it doesn't matter for grade. There are regular exercises, but they are not graded; you can submit them if you want it corrected to check your understanding, but there is also gonna be a short presentation going through it in class. Your entire grade is a 2-3h written exam at the end, no materials, no remote, no books, no multiple choice.
I missed that part, and since it's a newly delivered plane (this January), it's safe to assume the mitigation was in place. Preliminary report will be interesting here.
Wonder why they don't just grab a huge permanent Sharpie and write in huge letters "Do not insert pin here" on one hole and "Insert pin here" on the other hole.
I'm actually serious, it seems to me they resist these kind of short-term helpers that would save lots of injuries.
Sure, but they probably took 3 years to have a design review, an executive review, some firings and layoffs, re-hire, orientation, a sprint planning meeting, a sprint retro, a post mortem, an OKR meeting, a KPI meeting, an all-hands, and then the cover probably got stuck in customs with tariffs, and then the tolerances probably weren't correct.
And eventually be missed/ignored by a rushed ground tech and fail again.
Other than making it easier to blame someone, labeling is just a short term interim fix for such things. You design it to be physically impossible or as close to that as possible.
Been there, done that in much less high stakes environments. Upping the training, documentation, and labeling simply makes the mistakes happen less often for a physical process obviously prone to a common mistake.
Sure as an immediate airworthiness directive giant bright lettering is a great immediate “this month” fix. Certainly not a permanent one though.
If you make a hole multiple things can be fit into, eventually someone will try.
Anthropic & co charge API users much more, not least to demolish the middlemen low-effort plays like Cursor and Copilot. To not own the model is not viable in 2026.
Sorry, what do you mean by "To not own the model is not viable in 2026."
I assume I'm misunderstanding you (likely my fault), because the way I read that is that you're saying nobody should currently be using models owned & hosted by companies like OpenAI and Antheopic, while clearly a huge number of people are using those in 2026 despite not owning them.
But isn't the current understanding that harness is equally important as model once you get above a certain threshold, so there seems to be room to add value there.
Cursor is potentially about to be acquired by X.ai (i.e. SpaceX), unless this is just some IPO game being played by Musk. They are certainly not just a token reseller since they have their own models in addition to their own vector database approach for code matching.
I think it’s more correct to say they charge subscription users much much less. I assume less even than the cost of providing the inference, if you actually are using it.
Except that whole payload-to-orbit business turned out to be a dream of SciFi readers, very prevalent on HN, not so much a real massive business. Which is why Starlink is still the biggest "customer". It's probably most interesting for governments, which is also of course why China & co. will have their own SpaceX, not this one.
And then meanwhile the whole thing got merged with the corpse of Twitter and the failed xAI. With the latter nobody can quite explain why it continues to employ extremely overpaid 20 year olds when it has meanwhile pivoted to selling energy to its biggest competitors, an absolute turd of a business.
reply