ceras

Neither the iOS app nor the web app use Kotlin at DoorDash. Kotlin was picked for other reasons (consistency with the Android app wasn't mentioned): https://doordash.engineering/2021/05/04/migrating-from-pytho...

Starlark, used in Bazel, is in that vein, though goes a bit further by supporting functions (with many restrictions): https://github.com/bazelbuild/starlark/tree/master

Yes, they're exempt.

In adult language learning, practicing pronunciation early is far, far easier than fixing it later. Correcting bad pronunciation habits is extremely hard once engrained, but learning it the right way early isn't as hard as people think. Most courses don't focus on this enough IMO, resulting in people with hard-to-understand accents, which makes them feel less confident when practicing speech as they're asked to keep repeating themselves.

EA isn't limited to utilitarianism. A very substantial amount of people in EA would not consider themselves utilitarians. And of those who do, it's very rare to find a "pure" utilitarian with no deontological bent and thus willing to engage in fraud.

Even if you find a "pure" utilitarian (which is rare), it's considered "naive utilitarianism" to ignore the long-term and broad effects of creating harm, and it's also considered intellectually arrogant to think you know enough about the impacts of your decisions and moral philosophy that you can justify causing definite short-term harm for potential long-term gain against the moral frameworks of just about everyone else.

On the whole, I can't personally think of any one person involved in EA, or widely-read EA literature, that promote or support that type of thinking.

suburbs are perfectly walkable anywhere in the northeast. Cities less so due to crime

This isn't true. I've lived in and visited plenty of northeastern suburbs and cities. The suburbs are largely not walkable: no sidewalk, long distances, and as a result basically nobody walks to any shop. The cities are walkable and with a large amount of residents actually walking.

Curious, does this apply to romantic partners too? Personally, I'd be sad if my wife didn't share things troubling her that she didn't need my help solving. I like knowing how she's feeling about things.

But at work, I understand this mindset. Though personally, I still actually don't mind hearing people complain. And since I'm a manager, complaints are a very useful signal for me: even if I'm just in listening-mode, they give me more clarity on precisely what's going on in my team.

Layoffs do sometimes happen this way. I was an EM at a company with layoffs where line managers were not told at all about layoffs or included in deciding who to lay off: all discussions happened at the director level and up.

That's what the article is arguing for -- delay school start by 1 year for boys, which they call "redshirting." Key argument is towards the end:

One striking study, by Diane Whitmore Schanzenbach of Northwestern and Elizabeth Cascio of Dartmouth College, drew on data from Tennessee to study the impact of a delayed school start. The children in their sample were allocated randomly into different classrooms. Overall, Schanzenbach and Cascio found that being a year older had a positive impact on eighth-grade test scores, reduced the risks of repeating a grade before high school, and improved the chances of taking the SAT or ACT. The benefits for boys were at least twice as big as for girls on all measures through eighth grade. By high school, only boys were seeing any gains.

Lastly, they found that the younger classmates of redshirted children suffered no negative consequences. If anything, they wrote, there were modestly positive spillover effects. That’s one reason to believe that girls would only be helped by this shift—having more mature boys in classrooms would likely improve the learning environment.

Cascio and Schanzenbach’s research is the most robust to date, but their findings have been confirmed by a number of other studies. And related research has shown that redshirted boys are happier, too.

Why not both? Work is ~40hrs a week, so it's nicer if you have the option to enjoy it. There are other software jobs with similar pay to Google, but with more rewarding work. Win-win to switch, if that's what you're looking for.

Personally, the type of problems I solve at work are more interesting than I could realistically come up with and work on on my own. Ymmv.

at a time when cities are in decline

Do you have a source for that? Here in NYC things seem as popular as ever, with record-breaking rents due to high demand. Is it different elsewhere?

Regardless, even if it's true that some cities are struggling, that only makes it more important for them to focus on improving quality of life: a rise in WFH means people have more options, and a city that doesn't want to depopulate has to be an attractive place to live independent of economic opportunity.

This doesn't really match my experience for my own career or as a manager. Plenty of great engineers don't have a plan beyond "learn and get better," including those that have traditionally-successful careers (e.g. director+ at top tech company).

I'd expect that to matter more if you're, say, trying to be a founder - I can see that benefitting from intentional planning. But for careers at big companies, "learn and get better" seems good enough (and if that qualifies as a plan, I don't think I've worked with anyone that doesn't have a plan).

Most of that is growth, not churn. Alphabet has hovered around 20% net headcount growth per year for awhile, per their earnings reports.

IIRC the rumor leak around that was more nuanced: the leaks reported someone allegedly saying that market economics may not ultimately be able to support more than 2 leaders, because the economics will make #1 and #2 able to profit at prices that numbers 3+ can't compete with due to lack of scale. So the statement was that GCP has keep up or its ability to profit will be at risk, not that they'd shut down just because their slice of the pie isn't big enough if they end up as #3.

I don't believe it was clear who said this, how confident it was that someone said this, and whether it was an off-the-cuff comment in support of a push for more growth or an actual deep assessment of the economics of the space.

Princeton tried exactly this while I went there ("grade deflation"), and it didn't go well. They sent a letter to places like med school admissions alongside transcripts, but it wasn't enough. Ultimately, they abandoned the practice in 2014: employers, law schools, med schools often have minimum GPA requirements that are not school specific.

Students also didn't like it because they felt it made the environment more competitive and less collaborative than at other schools, because an A for your peer means one less A for you.

Somewhere else in this comment section someone mentioned how going to Princeton during grade deflation era stopped them from becoming a doctor, and I'd say that's accurate.

The taxi industry had it figured out only on a smaller scale. Rideshare reaches more people and rides than taxis did. In most of America, taxis were never a serious option outside of niche use cases like getting a ride from an airport. Even in NYC, where street hailing is possible, that only ever really worked in Manhattan (and not even all of Manhattan); and medallion caps meant driver supply could never fully meet rider demand, unlike for rideshare, so getting drivers wasn't a realistic problem.

Rideshare expanded the reach of taxi-like services to more regions, people, and use cases. It's now viable to travel without renting a car in many US cities - you can see this by how rideshare affected rental car companies. That's what makes the rideshare business hard: the places where they didn't have existing competition, because that's where the matching is harder and the economics not as easy.

It's a similar story for food delivery: GrubHub/Seamless operated in core areas of cities like NYC for many years (since the 90s in NYC, I think). Uber Eats brings that service to far more places, and that's where the challenge comes from in that business.

If they're issued as double-trigger RSUs then you're only taxed when they're liquid. That's because they aren't really yours till a liquidation event.

All major US tech company diversity initiatives I'm familiar with measure diversity as percent of the workforce that is black, Hispanic, or female. Asian men don't count toward these goals because they aren't under-represented relative to the US population.

This might help: https://microservices.io/patterns/data/database-per-service....

Varies by actual DB but: with this architecture, if the DBs are NoSQL then you didn't lose as much by splitting up the databases, but if they were SQL then among other things you lose transactions across those DBs (or now need costly distributed transactions), effectively making the collective SQL system behave more like a NoSQL DB.

You only need one promotion, L3 (entry level) to L4 (mid level), to not get fired at Google. And the company is quite generous with how long it gives you to get there. L3 to L4 is basically about writing good code on your own, and isn't really affected by promo culture -- doesn't really need shiny work.

Several years back, you had to also eventually get to L5. This was more affected by promo culture, but was also largely unenforced, which is why they get rid of the formal requirement to get to L5.

Caring about anything is inherently emotional.

Take human suffering. There's no logical reason to care about it. You can't take someone with sociopathic tendencies and logic them into caring about humans - it simply doesn't make sense. Either that empathetic circuitry exists in someone's brain or it doesn't, which manifests as emotions when it does exist.

I think you're starting from a flawed premise that logic is a motivating force. Logic is a slave to the passions, not the other way around. Logic is a tool we use to manage our emotional state more effectively by helping us choose actions that we believe are more likely yield net positive emotional states. Logic just computes results; it has no innate motivation of any kind, it takes motivations and emotions as inputs.

Take this for example:

beneficial to humanity.

You're implying a preference for humanity, which has some emotional foundation. Maybe you care about humans as a whole for some abstract emotional reason; or maybe it's purely self-interest since you're a human (so what benefits humans hopefully benefits you, even if you don't care about other humans) which is also an emotional reason. There's no objective, inarguable truth that you should care about humanity. The root of your confusion on animal rights stems here.

Ultimately, animal rights is about taking the empathy that most humans innately have to at least a limited extent, and expanding that to apply to more entities. If you experience no empathy in general in life, it makes no sense. If you experience some empathy, it makes some sense. If you experience a lot of empathy, it makes a lot of sense.

Diversity initiatives in tech have a strong bottoms-up component from a subset of employees. "Self interest" isn't really the point.

One motivator is societal good and fairness to ensure that great opportunities are as equally available as possible, and that nobody avoids or leaves the industry due to their race or gender.

The other is making sure you have more demographic variation that can make a better product for more people. A classic example is avoiding gaffs in ML models based on skin color. All else equal, the more representative your employees are of your target user base, the more likely someone is to raise the right questions early. This is especially true for consumer tech where engineers are part of the process of deciding what gets built, but also true in cases like thinking about ML fairness.

I think you're making a mistaken observation there. It's a skewed data set: of course people who like living in cities will talk about policies they'd like to see to make their cities better, and not complain about ways that a suburb or rural community could be better.

I don't live in a suburb so it would be very odd for me to complain about a hypothetical suburb I don't live in, even though I'd have more complaints if I did live in a suburb. Just how HN as a whole talks about making tech jobs better (e.g. this remote work thread, which isn't relevant to most jobs) far more than it does about other jobs: it's just more relevant to the people here, and even if they overall like their tech job, they want it to be better.

There are prices to pay for living in city centers

These are the prices we pay today in the US, but they don't have to be. Other countries like The Netherlands and Switzerland take noise pollution more seriously and show it's possible to have quieter cities. Even among huge cities it's possible: Tokyo is much quieter than NYC.

We don't need to make people choose between living in dense, efficient spaces vs calm and quiet. Making urban living more comfortable is a net benefit: the people that live there will be happier, people that wanted to live there but avoided due to noise will now have that option, and people outside of cities would be unaffected. Why accept this as just the way it is when plenty of other cities have shown it can be better?

This video shows what it's like when a city takes noise more seriously (in this case, the whole of The Netherlands). It focuses on cars, but apartment noises can be handled too with building codes. https://youtu.be/CTV-wwszGw8

I think there's one nuance to add: the test is based on your base rate expectation for how likely the given individual is to have the rare disease.

The more evidence you have that this person is more likely to have the disease, the more useful the test becomes. Some examples:

- if the disease is more common among people over age 65, it's more useful on people in that age group

- if the person displays more symptoms associated with the disease, and not also associated with more common diseases, then it's more useful for someone with those symptoms

- if a disease is common in an area of the world, and the person has traveled there, the test is more useful

The more factors you have, the less likely a false positive. This is why it's often better to avoid medical tests unless there's a reason to suspect someone is at elevated risk for a disease: a positive result is more likely to be a false positive.

It's the reverse at Meta: they reward you with more money more quickly than a promotion. This happens when you quickly have very high impact, but it's unclear if you can sustain it.

This similarly happens to more junior who seem to be on a steep upwards trajectory: the company tries to retain them with money (stock) before they actually get good enough to get promoted, because there's high confidence they'll grow fast.

Someone who earns into the next pay band is pretty likely to be promoted in the near future, once they've shown they can sustain it.

Google distinguishes ads on their own properties from ads on other properties in their revenue filings, but they still call this ad revenue.

Google Search ads are bucketed under "Google Search & other" in financial statements, which is just one of their "Avertising revenue" categories as defined here: https://abc.xyz/investor/other/additional-financial-informat...

That reminds me a bit of Tokyo. It's full of small streets[0] that cars can drive on, but they tend not to because they're narrow and don't have sidewalks, so making cars move slowly. Cars then naturally flow through to the larger streets[1] when possible, which do have sidewalks.

[0] https://goo.gl/maps/4RtUzUUfjvMpLs8y6 (this street has a small painted sidewalk since it has more commercial activity, though pedestrians often walk in the middle; the adjacent streets have no painted sidewalk).

[1] https://goo.gl/maps/BquuyEvSrVExuyiu7

In a walkable neighborhood, the result may be different. For instance, NYC has a lot of local shops, enough that NY state as a whole has the lowest per-capita rate of fast food chains in the US. And I see multiple local coffee shops here for every 1 Starbucks (though Starbucks is still quite common; far more likely than a McDonald's). I think it may be related to walkability since chains are more common in NYC's more car-oriented neighborhoods.

Outside that, I still don't think you're most likely to get a McDonald's in the middle of a residential neighborhood: it'd most likely be a corner store, since that's what walkable areas are most full of. At least in NYC, it's uncommon to have fast food outside of a busy shopping street. The shops embedded in the middle of residential neighborhoods are largely corner stores and delis, and to a lesser extent cafes / bakeries and barber shops / hair salons.

No strict SLA is unrealistic for many businesses (imagine NYT going down for an entire weekend), but follow-the-sun could work if employees really want it.