blegr

I’ve seen him discuss additions to the static typing features.

Same as a bank? If you lie to investors and get yourself nuked, the owners or what passes for owners get wiped out, which is normal. The people who really qualify as victims are those who were harmed and didn't benefit from the fraud, which is everyone else.

Sure, some people would lose money but didn't intend to cause harm. Those are victims of the fraud. Their shares of the criminal company still get wiped out and they get in line to get bankruptcy proceedings.

Is there another way that makes sense?

The police spent the day searching her house. Inside, they found a note with the heading “NOT GOOD ENOUGH.” There were several phrases scrawled across the page at random angles and without punctuation: “There are no words”; “I can’t breathe”; “Slander Discrimination”; “I’ll never have children or marry I’ll never know what it’s like to have a family”; “WHY ME?”; “I haven’t done anything wrong”; “I killed them on purpose because I’m not good enough to care for them”; “I AM EVIL I DID THIS.”

On another scrap of paper, she had written, three times, “Everything is manageable,” a phrase that a colleague had said to her. At the bottom of the page, she had written, “I just want life to be as it was. I want to be happy in the job that I loved with a team who I felt a part of. Really, I don’t belong anywhere. I’m a problem to those who do know me.” On another piece of paper, found in her handbag, she had written, “I can’t do this any more. I want someone to help me but they can’t.” She also wrote, “We tried our best and it wasn’t enough.”

Calling this an admission of guilt is ridiculous. The only one that doesn’t say she didn’t do it says it happened because she wasn’t good enough. Self-blame when everyone is calling you a murdered is completely expected; the article says later that she had PTSD.

Is that the evidence she’s getting slammed based on?

New link: https://www.newyorker.com/magazine/2024/05/20/lucy-letby-was...

Some people feel some kind of person/presence close to them, that's how it happened to me. This [1] weirdly makes sense.

It felt kind of like those dreams where you can't move. Except you've woken up, I could see the correct room and position I was in. It tends to come with fear and doesn't last long.

Never happened when I slept on my side or front. I almost never sleep on my back now. Also, chronic stress could have been a trigger.

[1] https://commons.wikimedia.org/wiki/File:Henry_Fuseli_(1741–1...

A lot of suspicious facts in the article though.

Letby’s defense team said that it had found at least two other incidents that seemed to meet the same criteria of suspiciousness as the twenty-four on the diagram. But they happened when Letby wasn’t on duty. Evans identified events that may have been left out, too. He told me that, after Letby’s first arrest, he was given another batch of medical records to review, and that he had notified the police of twenty-five more cases that he thought the police should investigate. He didn’t know if Letby was present for them, and they didn’t end up being on the diagram, either. If some of these twenty-seven cases had been represented, the row of X’s under Letby’s name might have been much less compelling. (The Cheshire police and the prosecution did not respond to a request for comment, citing the court order.)

Are you still sure after reading?

Does this make MRIs safe for some people who wouldn’t qualify due to metal implants? Or at least reduce the risk of accidents?

It's funny that restaurant burger prices in my EU country are basically the same as in your "upscale American city". I thought everything there was more expensive since the higher percentiles of your salaries are so much higher.

Didn't Microsoft stop using those in BitLocker years ago because they were often borked? [1] It's a new failure point, but it's sensible.

[1] https://www.tomshardware.com/news/bitlocker-encrypts-self-en...

Good design dictates that you start one loop and build the whole program around it, no? The docs for asyncio.run say as much.

It’s not about the price of upgrades though, it’s about their bundling together and the ridiculously stingy base specs that often make the upgrade non-optional. People who buy a base MacBook Air probably aren’t thinking about keeping it for 8 years or using it for heavy workloads.

I hope this means AI-accelerated frameworks get better support on Mx. Unified memory and Metal are a pretty good alternative for local deep learning development.

Yes, but you understood what I meant since you could assert that it’s not what it means.

It forces you to buy multiple upgrades instead of just the one you need.

Now that GP said it, it'd be nice to be able to write "some_map[i].value_or(some_value)".

I see your point, but my experience is that you need the stack trace first, and the developer’s explanation second. Asserts crashing with a message that makes perfect sense in its context but is completely useless for debugging are the bane of my workweek.

Now I appreciate a clear explanation for an uncommon assert and for example, OpenCV could do with more of those, but in most functions, seeing the line that throws the error is enough to understand.

Right, but that's redundant with the stack trace. It's not actually helpful to run a big program I don't know very well and panic with a single "your goose isn't cromulent!" message from a call 20 levels deep.

In your example, it's likely that the person who sees this message won't have enough context to understand it; it's more like a debugging assert. Since you'll need a debugger and a breakpoint anyway, the message isn't very helpful.

It's not really my choice, I'm working in a big codebase I didn't write. Lots of large classes containing large collections with getters and setters.

It sucks but it's easy to review and avoid, probably could be checked statically by linters too.

The Option type seems to have various standard Rust methods like expect()

Isn't that value()?

The practical use of optional<> is that you know where T is going to be constructed and can reason about the memory layout.

The practical use for me is making interfaces safer. Where I saw colleagues use pointers as optionals, end up mis-tracking what can be null and what can't, only checking it inconsistently, and triggering UB, I now have a clear distinction between optional and non-optional arguments/returns with an easy way to access the contained .value() without risk of UB. The type also tells me when I should handle the empty case and when I shouldn't.

Most of the time, I want to pass/return a reference and the lack of `optional<&T>` makes it tiring. If only `std::reference_wrapper` had a shorter name, I could at least use that. But then I'd end up with `arg.value().get().attr` when `.value().attr` should be enough...

Isn’t that the consequence of them having smaller budgets to begin with? Having a trillion-dollar budget means the US defense industry can afford more graft and cost overruns.

I'll admit the support for my Dell was pretty good. They sent someone on-site to fix a known defect in their product line.

This is less severe than losing an account because at least the encrypted drive is backed up, right? :)

Everyone has a brand they're never buying again because of a few problems they had in the past. For every new brand they _are_ still buying, there are 10000 other people who are never buying _that_ one again because of a few problems they had in the past.

The only difference I've seen between Apple and my previous laptop brands is that their support techs are useful.

Wasn't that mostly compiling though? VSCode's CMake tools take multiple seconds just starting an already-built executable.

France. I get 2.5 days a month and take them very rarely. It's nice to have the option to leave for a few weeks or take days off for doctor appointments, etc.

Don't forget that zerodium still pays more for an android 0-day than an iOS 0-day.

A random Internet search gives iOS 30% market share to Android's 70% [1], which could also explain the higher price.

[1] https://www.statista.com/statistics/272698/global-market-sha...

Split the re-triable action into one function, make a wrapper function that re-tries if needed, and use a third function that makes the decision to re-try and how long to back off.

Then you can test the decision function trivially, the re-try function by mocking the action and decision, and the action function itself without back off interfering.

That’s what you suggested, just saying that I did that in a Python API client with the backoff library and the result is pretty neat.

Actually, “clickbait” doesn’t really apply here, the paper supports the title.