Hasnain says:

Sigh. Tech bros will be tech bros.

“On May 21, 2021, the team hosted their monthly “gala,” where Launch House members and friends gathered at the Beverly Hills mansion for a party. They loaded in a carful of booze, and though the founders had claimed they’d planned on only around 100 guests, hundreds more were added to the guest list. “Brett [Goldstein] told the girl at the door to let anyone in, regardless of Covid tests,” said one attendee. “If they weren’t on the guest list, as long as they were hot or had a lot of followers,” they could enter. “Even if Covid weren’t a thing, I’m sure it would have been a fire safety issue,” said another. “It wasn’t a networking event anymore; it felt like a club.” More concerning was the presence of seemingly underage girls, some of whom were seen falling over or passed out on the curb outside.”

Hasnain says:

Great read, and it reminds me that I need to go and blog about my own blog setup at some point.

“Normally this would be terrifying, especially with the amount of traffic my blog gets (as represented by this handy graph here). You'd think that something that does a lookup on every post in the worst case for the most common thing on the biggest dataset would make performance terrifyingly slow You'd also think that with the amount of traffic I get that it'd be an active detriment.

However, this is when I play my trap card! When you look at the analytics you can see that the most frequently read article is the most recently posted one! This means that it's not actually a big O of n lookup most of the time. It's constant time complexity. In theory this design is the terrifying type of thing that you normally find out about after you accepted a job offer, but in practice it's fine. It is a bit weird though, and I may need to rethink this in the future, but this has scaled to almost 300 posts as-is so I think it's okay for now.”

Hasnain says:

Sharing without comment.

“Facebook’s stonewalling has been revealing on its own, providing variations on the same theme: It has amassed so much data on so many billions of people and organized it so confusingly that full transparency is impossible on a technical level. In the March 2022 hearing, Zarashaw and Steven Elia, a software engineering manager, described Facebook as a data-processing apparatus so complex that it defies understanding from within. The hearing amounted to two high-ranking engineers at one of the most powerful and resource-flush engineering outfits in history describing their product as an unknowable machine.

The special master at times seemed in disbelief, as when he questioned the engineers over whether any documentation existed for a particular Facebook subsystem. “Someone must have a diagram that says this is where this data is stored,” he said, according to the transcript. Zarashaw responded: “We have a somewhat strange engineering culture compared to most where we don’t generate a lot of artifacts during the engineering process. Effectively the code is its own design document often.” He quickly added, “For what it’s worth, this is terrifying to me when I first joined as well.””

Hasnain says:

Bookmarking this for later re reading as I’ve run into this problem before.

“Where does this leave us? Unclear. If you go down the rabbit hole of reading about all the issues surrounding GATs and HKTBs you get a strong sense that it's better to avoid creating APIs that invole abstracting over ownership and borrowing when possible. You will run into walls and the workarounds might be ugly and hard to understand. So I guess a new thing I can recommend not to try to do: do not abstact over borrows and ownership if functions are involved (unless you really know what you are doing).”

Hasnain says:

This is nominally a post introducing a library, but it's so much more - I learnt a lot about good API design, string matching algorithms, and performance benchmarking. burntsushi's code and blogposts are always great, and this is no exception.

"In other words, if Rust’s primary string types work for your use case, then you should probably ignore bstr altogether and continue using them.

So why have a byte string library? The simplest way to explain it is to point at the std::io::Read trait. How does it work? Well, it says “anything implementing std::io::Read can take a writable slice of bytes, read from its underlying source and put the bytes from the source to the writable slice given.” Do you see anything missing? There’s no guarantee whatsoever about what those bytes are. They can be anything. They might be an image. They might be a video. Or a PDF. Or a plain text file.

In other words, the fundamental API we use to interact with data streams doesn’t make any guarantees about the nature of that stream. This is by design and it isn’t a Rust problem. On most mainstream operating systems, this is how files themselves are represented. They are just sequences of bytes. The format of those bytes usually exists at some other layer or is determined through some additional context."

Hasnain says:

This was a pretty good read on leadership.

"When you’re a senior leader you don’t get to throw your hands up and say that there was nothing you could do. Did your project rely on another team who screwed up? Sorry, you still need to figure out how to get it done. Are you an Engineering leader who hates their Product counterpart? Do they hate you? Too bad, you still need to get things done."

Hasnain says:

Yay!

““I think this becomes a tipping point, frankly,” said Christine Hendrickson, the vice president of strategic initiatives at Syndio, which provides software that helps employers identify pay disparities. “It’s at this point that employers are going to stop going jurisdiction by jurisdiction and start looking for a nationwide strategy.””

Hasnain says:

“Other teams are conducting the first JWST observations of TRAPPIST-1, a relatively nearby red dwarf star orbited by seven Earth-size rocky worlds. Several of these planets are in the star’s habitable zone, where conditions favoring liquid water and even life may be possible. While JWST cannot directly image the planets, spectroscopy will help identify the gases in their atmospheres — possibly even hints of gases that could signify biological activity. “What we really want is Earths,” said Macintosh.”

Hasnain says:

This was a really interesting read.

“DEB COHEN: Some of the studies we were actually having a bit of a laugh about, um. So what you would do is you would starve people overnight. Um, you'd fast them overnight and then you'd ask them to cycle to exhaustion. One group, you would give a sports drink containing sugar, and the other group you would give water. Well, guess what? The people that have had a bit of sugar are going to outperform the people that have been starved and had water, it's not rocket science.

So there are all sorts of these, all sorts of these kinds of studies that they do, and you look at it and go, that just does not happen.

JEN GUNTER: And it doesn't matter that it's not clinically relevant. Now you've got something that looks statistically different and you can peg a marketing campaign around that.

DEB COHEN: Exactly. And if you throw enough darts at a dartboard, if you've got your eyes blindfolded, then one of them is going to hit the dart board at some point.”

Hasnain says:

“These tradeoffs are often intensely difficult to pursue openly. Who wants to be known as the politician in favor of benefits fraud or the financial CEO who thinks they are not laundering enough money?

One of the interesting questions here is who gets to resolve tensions like this. Generally speaking, it will be private actors applying their own cost-benefits decisions. There is substantial space for regulations to help with cases, like identity theft, where actors can choose to spend other people’s risk budgets to maximize for their own interests.”