methodover

Well. That first video was terrifying. I'm too scared to click on the others.

Dystopian SaaS idea: Integrate facial recognition and Yelp reviews. If someone shows up who gave you a negative review, send a text to staff and/or deny sale at POS.

Please, no one make this.

I left out a great deal of detail. We had documentation. We closely work with an HR lawyer, and consulted with her extensively about this individual long before firing him.

We didn’t get “scared” of his lawyer. Lawsuits are distracting and expensive for both sides even if the allegations are baseless, and we avoided one in this case by just offering a bit of a larger severance than we originally offered.

I’ve told this story before on these forums:

I had a junior engineer once, a few years back, who could not take criticism.

He would get flustered, angry, sarcastic. His work wasn’t good. I did my best to help him — and even though I was polite, he always seemed to take a kind of deep, personal offense at whatever criticism I had for him.

There were many problems I had with his work — he wasn’t great at testing his code (at a surface level, sure, but rarely thought deeply about the implications of his changes. E.g., once he got a ticket to clean up some code throwing lots of warnings, and solved it by making the code not work at all. He tested that it wasn’t throwing warnings, but didn’t realize that it had an early escape condition that was being triggered inappropriately every time.)

He surprised me with his lack of basic knowledge that I would’ve expected from a CS graduate who’d also made an app or two outside of school. He didn’t know what a database index was, for example, among other issues.

He was also not great at communicating. When talking about his work, he’d use technical jargon incorrectly. His emails to outside vendors were meandering and difficult to parse. I coached him on these points extensively.

In the end, after 6 months or so of trying to help him, I asked our CEO to fire him. We gave him a month’s severance.

A couple months later, his lawyer got in touch with his, alleging racial discrimination. The allegation was that we fired him because of his accent. He interpreted my problems with his communication as being a problem with his (slight, barely noticeable) accent.

Our CEO told me to work on how I levy criticism of others. I mean, he’s right I’m sure. I thought I was being precise and polite, but perhaps I did give the wrong idea.

Or, maybe this hire just wasn’t used to criticism. He didn’t understand how to take advice. He didn’t know how to internalize it and act on those suggestions. When fired, he blamed an immutable quality about himself rather than something he could change.

I don’t understand the math you’re giving here.

I just read The Meritocracy Trap by Daniel Markovitz. It’s an incredible book and talks at length about the issue in the article.

A word about this:

“What technology and trade have done, however, is displace millions of Americans from their middle-class jobs, and send them hurtling down the income ladder into less remunerative occupations.”

It’s not that simple. It’s not just trade and technology.

The causes of the erosion of good middle-class jobs are extremely complex. I’ll talk about just one here, since I don’t have too much time:

Tax law works to advantage higher tier workers over middle class workers. It costs much less in payroll taxes to hire a single superskilled worker at extremely high compensation than it does to hire a bunch of mid skill workers at middle class comp.

Quoting from the book:

“A simple example illustrates the special burden that the payroll tax imposes specifically on middle-class labor. If a bank deploys midcentury financial technologies to issue home mortgages using twenty mid-skilled loan officers who each earn $100,000 per year, this costs the bank and the workers, taken together, $306,000 in payroll taxes. By contrast, if the bank were to switch to the current mode of production and displace the mid-skilled loan officers with a single Wall Street trader who earns $2 million, this would cost the bank and the trader only about $90,000.

“Where two technologies of production are economically equivalent, but one requires twenty mid-skilled workers while the other requires one super-skilled worker, the mid-skilled approach currently faces an average payroll tax rate over 10 percentage points higher than the meritocratic approach, which produces an aggregate payroll tax burden over three times as great.

“The payroll tax, in other words, substantially suppresses mid-skilled employment and wages and fosters super-skilled employment and wages. (Indeed, if the super-skilled worker can get capital gains treatment for her income, by styling it as founder’s shares or carried interest, the income tax adds a further bias, on the order of 20 percentage points.)”

In MySQL, discovering the UNION hack a couple years ago was like a revelation. It’s something that I’ve since been preaching to everyone who’ll listen.

It’s extremely common to want to OR a small set of items. It feels like the compiler would be smart enough to use an index (presuming one exists matching the searches fields), but it doesn’t. UNION makes it use the index.

It’s a nice tool to have in your kit.

(At least not in 5.6 or 5.7.)

I’m sure other commenters here will comment on the substance of what you wrote.

I’ll, however, just toss a complement: You’re a fantastic writer. Pithy, funny, informative. Nice work here.

You’re totally right.

I meant socialist in the same way that Bernie Sanders seems to use it, which I have trouble distinguishing from Warren’s states policies. (I do agree that the traditional definition of word applies to neither Warren’s or Sanders’ policies for the most part. Except for the health insurance industry, where the proposal is nationalization.)

I should have said social democrat. Indeed, now that I think about it, few engineers would label themselves socialist. I don’t, though I am a Warren/Sanders supporter.

This was my experience as well. Started off as a libertarian. But then I started working, first in the lobbying industry, then in Silicon Valley.

Meritocracy is not as good an idea as it seems to be in practice.

I’ve seen a couple comments saying that it’s the non-SWE’s that are the liberal ones.

That’s a very different experience than my own, and that of my friends.

My experience: Designers tend to be pretty liberal. Product managers and other managers tend to be moderate (tend to be Biden and Buttigieg fans). Salespeople tend to be extremely conservative. Support is kind of all over the place. Engineers tend to be either diehard socialists or (and this is the minority) libertarian.

I see no contradiction there. She’s fighting against the influence wielded by those at the very top of the ladder.

Do you have a source for the Zuckerberg quote?

How is it voting against their self interest?

This discussion and the original post has been what the United States government’s trade policy ought to be, not the morality of an individual’s purchasing choice (say, to buy a Chinese-made sweater or a US-made one). I think the question you’ve asked is interesting, but I don’t see how it’s pertinent to the matter at hand.

In regards to swapping cell phone plans, sure: If a cell phone company were to die because too many of their customers switched to a better competitor, that would be difficult for the employees of that company.

But that's an entirely different situation than the matter at hand, where we're talking about the United States government's policies on trade and the impact on our entire labor class and their fate within our own borders.

The entire point of an economy is to serve humanity. We're all participating in this circus to put food on our tables, provide for our children, grow, and enjoy life. We cannot lose sight of that fact. We have an obligation to see labor not as just cogs in a machine, but rather as constituents whose well-being we have an obligation to protect.

(I have to say, and I'm sure you don't mean it, but you comparing a human being to a cell phone plan is among the more callous things I've read on these forums. It might behoove you to sprinkle a bit more empathy in your language, just a tiny bit.)

There are more ways to measure the health of the American worker than the BLS' unemployment rate. What about other statistics? Things like like...

- Wage growth over time - Health outcomes - Savings rate - Credit card debt rate - Feelings about the future (are we on the right track/wrong track?)

Yeah, this is another aspect of the debate I find myself challenging: Labor is different. Our human lives are different than a cell phone plan.

I'm sympathetic to that idea, to some extent. Foreign workers would indeed be happy to get a marginal increase in their income, even if those wages and working conditions are well below developed-country standards.

But, then, aren't we just making a trade at that point? I will fire 100 local blue collar workers, and hire 100 foreign workers who will do their job for less. I hurt 100 people, help 100 people, and along the way pocket some extra cash for myself. I think most moral systems would have a problem with this. It is generally thought of as immoral to rend one person in order to help one other.

(Yes, things get complicated when you hurt N people to help more than N people, depending on the number of extra people helped and the kind of hurt. For example, sacrificing soldiers in order to save an entire country. Generally speaking though, most moral systems have a problem with hurting people against their will except in extraordinary situations involving very large numbers. Does this trade situation qualify as that? I'm skeptical that is does.)

There has to be a better way. Some way where we aren't harming local blue collar workers, but are still helping foreign nations develop, while giving those foreign nations on the path of strong worker protections and wages that we have here.

The big question I’ve been struggling with in regards to the free trade debate is this:

Is cheap labor a legitimate source of comparative advantage?

I increasingly don’t think that it is. Yes, we get products more cheaply. But those products are made in countries with fewer labor restrictions, often in places that are not democratic, where workers cannot vote for changing laws. We’re effectively saying we want democracy where we live, for our own white collar professions, but we don’t want to be shackled by those restrictions for how we get rich, or how we enjoy our wealth. We’re effectively telling local blue collar workers that we’re okay with their level of work being taken over by people with less rights than themselves.

That increasingly doesn’t feel just. Comparative advantage makes sense for natural differences between countries, such as availability of natural resources or favorable weather for certain types of crops. But not for fundamental human rights.

The chilling effect that China is having in liberal democracies like ours is extremely disturbing.

We need our governments to do something about this. If Chinese companies want to do business here, we need our own businesses treated with the same ruleset in their country. We wouldn’t ban a Chinese corporation for a minor executive’s political speech. We have laws against it. (The first amendment protects non-citizens as well as citizens.)

China has been enjoying our markets but hasn’t been willing to extend the favor fully. We need a new trade agreement that demands equal treatment, or we should pull out of China and do business with friendlier countries instead.

Would we even realize their sentience

(Minor nitpick, but I think you meant "sapience" rather than "sentience." Sentience is something that essentially all animals on this planet have, and just means the ability to experience sensations. Sapience appears to be unique to humans (and perhaps a small number of animal species) and refers to the ability to reason.)

Reposting my comment from yet another HN post on these pretty mundane videos.

Gofast is just a weather balloon.

https://youtu.be/PLyEO0jNt6M

Gimbal is a jet. The perceived rotation is an artifact of the camera rotating on a gimbal.

https://youtu.be/4X1PRDbtiF0

Nimitz is also a jet. It’s far away, and blurry.

https://youtu.be/s1oTg0kxzDs

It would be really cool if these were alien spaceships. They’re likely not.

Gofast is just a weather balloon.

https://youtu.be/PLyEO0jNt6M

Gimbal is a jet. The perceived rotation is an artifact of the camera rotating on a gimbal.

https://youtu.be/4X1PRDbtiF0

Nimitz is also a jet. It’s far away, and blurry.

https://youtu.be/s1oTg0kxzDs

It would be really cool if these were alien spaceships. They’re likely not.

Oh I see! 1G from our reference frame! Fascinating! That’s a perspective I haven’t seen before, I didn’t realize that that’s what you meant. Indeed, it isn’t possible. We’d see the ship slowly approach but never reach the speed of light, no matter how hard or long they step on the accelerator.

Err, that isn't a correct understanding of physics, I believe. The ship wouldn't need to reduce its output as it gets faster relative to some other object in the universe. It can happily keep accelerating at 9.8 m/s for as long as it likes (or has fuel). No passengers would get squashed.

So, actually, like, it's actually really intuitive I feel like:

From the perspective of an object accelerating, newtonian physics works totally intuitively. If you had a rocket that could accelerate at 1g indefinitely, you just go faster and faster and faster and you get to any destination you want (even far away!) pretty quickly. And it would be a rather comfortable trip! You'd have Earth-like gravity the whole way.

It's really only the observer's perspective that things get confusing. When an observer watches something accelerate, they see it never going faster than the speed of light, no matter how fast it "actually" goes.

The trick is time. Time for slow things passes faster than time for fast things. A clock on a very fast rocket ticks much more slowly than clocks on (relatively) stationary things. That's how the paradox is solved.

Let's say you wanted to visit the Andromeda galaxy, which is around 2,500,000 light years away. If you had a rocket that could travel at 1g indefinitely, you'd get there in a comfortable 29 years! However, observers on Earth would see the trip taking around 2,500,000 years.

If you'd like to play with these numbers yourself, feel free to check out this neat calculator (not made by me)

http://nathangeffen.webfactional.com/spacetravel/spacetravel...

The GDPR doesn’t care where the personal data is stored, I thought. It makes no distinction as far as I can tell between data stored on a web server and data stored on a device. You seem to be making a distinction. Is there a source you could reference?

Prefix tree is the right answer here right? From an interview standpoint.

But my question is: why would you want to roll your own here? Wouldn’t the appropriate move be to look for solutions for your specific use case that’ve already been made? Ie if you’re talking web, I see no reason why a standard database index would work. MySQL’s index, a b-tree, is perfectly serviceable for the question, no? I’m having trouble imaging a situation where an out of the box solution wouldn’t work...

I don’t think I have a deep understanding of efficient data structures. I couldn’t re-implement a red black tree for example without looking it up. I wonder: is it enough to have a cursory understanding of things like runtime complexity, space complexity; understand the different basic data structures and how they have trade offs that can help or hinder different operations, and how that connects to real world work. Especially how different actual solutions use different data structures internally. What I don’t understand is the value from going from cursory understanding of data structures to deep understanding.

But having a deep understanding of data structures (not just arrays and maps) and algorithms really gives you a mastery of your craft, especially around performance and scalability.

Can you give an example?