HNHacker News
TopNewBestAskShowJobs

mediaman

6,587 karma · joined April 20, 2009

submissionscomments
mediaman··on New MacBook Air with M2
I did - I believed all the '8 gig is fine' stuff at the beginning, and didn't like how much Apple charged for RAM. But for my work anyway it often starts having stuttering and slowness issues if I have too many things open. Otherwise it's quite fast.
mediaman··on America needs supply-side reform
Visit Rhineland, and you'll see the impact globalization has had on Germany. They are not immune to it. For example, Siemens has shut down a number of industrial plants over the years, shifting production to both Eastern Europe and China.
mediaman··on It is your responsibility to follow up (2019)
That's an interesting approach.

When I did my first startup, I landed my first seven figure deal because I'd done the research to know someone needed my product. But they were too busy to respond.

I regularly followed up over six months by both email and phone. I knew they needed it. But I also know that people tend to prioritize their days by what fires are burning and need to be put out.

Eventually I bothered them enough that they realized they should do something to get me to go away. I guess I became the fire they needed to put out. Only then did they actually process anything I'd sent enough to realize it was something their Fortune 500 client actually did need, and that they'd look great by proposing it.

They wound up being my biggest client and helped make my startup profitable.

I could have said: Mr. Potential Client, I'll send you one email. If you don't configure your mail filters and organize your workflow properly, well, "fuck off."

If I did, I wouldn't have likely gotten any of my clients. Then I could explain the outcomes to my investors: "Well, I sent them an email. They failed to read it!"

The point is that if you want to persuade other people to see things your way, your approach doesn't survive first contact with reality. That's OK: your job may not be that, and it probably works just fine if you only deal with incoming requests. There's just going to be a limit to what you can do.

mediaman··on Why America can’t build
This sounds like objecting to something much better than the awful practice today because it’s not theoretical perfection.
mediaman··on Why America can’t build
Why not just do statistically valid polling?

The problem is that politicians run this weird highly skewed sampling method of "community input" and then think they've got the pulse of the community, where in fact they've just heard from loudmouths with too much time on their hands.

A statistical poll would at least show what the actual constituency wants, not the loud, bored subset.

The subset problem is also why we see much more extreme politics in cities among elected positions that have less visibility: city council members tend to be much more extreme than mayors, because the latter has greater voter participation.

mediaman··on Supreme Court Overturns Roe vs. Wade
Poll data suggests that the plurality of the population do not support third-trimester abortions, but neither do they support a prohibition on first-trimester.

The reality is that what we consider a "life" is graduated, and the fetus progresses along that continuum during gestation. Most people know this intuitively, which is why popular opinion generally falls where it does. (See, for example, the Pew polls on the subject of abortion and trimesters.)

We struggle to draw a bright-line to say, "this is where life begins," because there is not really a point where life "begins," merely a long series of points where it shifts by degree from "no life" to "life."

As a society though we still are tasked with drawing this line and it is an arbitrary line. But it is illogical for it to begin at the very start of the continuum, or at the very end. The beginning because it is silly and unscientific to pretend as though some tiny clump of cells has any more life than a wart, and it comes at great cost to women's autonomy; the end because it is illogical to believe that a baby the moment prior to birth is a fundamentally different being than at the moment after. To believe that the moment is at either end is thoughtless dogmatism.

Where should the line be? Pick a point. Give the woman enough time to be able to know that she is pregnant, to plan and provide for her bodily autonomy to make her own life choices, and to deal with lead times at clinics. And then draw a line where the choice has been made. Most people believe that line is somewhere around the end of the first trimester or early part of the second.

mediaman··on Show HN: Avo – Build Ruby on Rails apps faster
Can I drop this into an existing Rails app? I use Jumpstart but would like a nicer admin experience.
mediaman··on Three Arrows Capital reportedly facing insolvency
Sure, but there's no intrinsic reason for crypto to have a given value.

Assets that generate value (cash flow) will fluctuate in value per the moods of Mr. Market, but over their long run will hew to the present value of their future earnings power.

There's no similar number for crypto. Might as well be tulips. In that case, perhaps it is true that history will rhyme.

mediaman··on CircuitPython – The easiest way to program microcontrollers
A broad statement such as this inevitably gets it wrong.

I use micropython for developing embedded applications for industrial sensors. It is much faster to develop and I am fine using microcontrollers with enough memory that I do not need to optimize every last byte. Cost of the hardware is not the priority.

There are applications where there is not enough memory and python isn't the right tool. But in a world where microcontrollers keep getting more advanced and packaging more memory, claiming that languages like Python have no place is just as wrong as when people claimed Python had no place against C/C++ back when PCs were more resource constrained.

The economics for this type of question typically center around volume shipments. If you're shipping lots of hardware, optimizing hardware cost (and therefore minimizing software resource requirements) is the priority. If you're not, developer time is the main cost. There are a lot of products in the latter category.

mediaman··on Inbound container volumes to the US are reverting to pre-pandemic levels
Yes, you're right, I meant as a percent of the FOB price of the product. So 100% of the FOB price would be 50% of the landed cost. 100% of landed cost doesn't make sense and was the wrong term.
mediaman··on Inbound container volumes to the US are reverting to pre-pandemic levels
Drops in containerized freight, and the following drops in China-US freight prices, will start putting pressure on a lot of manufacturing onshoring initiatives started during the pandemic.

For many classes of bulk goods, freight was becoming 50-100% or more of total landed cost for goods imported from China, which overwhelmed labor cost differentials. In addition, Chinese production problems provided more reasons to move production closer to demand.

But US labor costs increased substantially during the labor crunch of 2021. Prevailing wages that were around $15 ramped to $18-22 as companies worked to compete against the likes of Amazon hiring hundreds of thousands of warehouse workers. Entry level Chinese wages have not increased nearly as much.

So if freight costs collapse such that the landed cost only has 25% of landed cost as freight, and US manufacturers have experienced an increase in labor costs relative to Chinese production, this could spell some very challenging times over the next 1-2 years as a recession applies cost management discipline on the supply chain and US manufacturers find themselves in potentially an even worse comparative cost position than pre-pandemic, vis-a-vis offshore production.

This is also why many US manufacturers were wary about adding capacity: they know about the bullwhip effect, and they also knew that it would affect not only consumer demand but also their supply chain advantage versus offshore production, and so any capital intensive capacity added to meet short-term demand may go idle 1-2 years afterward. And most manufacturing capacity doesn't pencil out if you only get a couple years of good utilization out of it.

mediaman··on S.F. population fell 6.3%, most in nation, to lowest level since 2010
If we're talking about 'reasonable' conservatives, the typical answer is that booking people who commit a crime, and then using the threat of prison time to force them into rehab will ultimately be much better for them and the community than immediately releasing them under the guise that poverty causes addiction and crime and that therefore they are essentially blameless.

Some former addicts attest that, when they were addicted, it was only the threat of prison time that could force them into rehab, because at the time they were not fully in control of their own decision making. They say the 'stick' helps motivate them to choose rehab instead, and it ultimately helped them out of a bad situation.

Sam Quinones' book, The Least of Us, documents some of this and is well-researched.

mediaman··on PhD students face cash crisis with wages that don’t cover living costs
It doesn't vary enough to make a big impact. Regular gas in Alabama is only about 8-9% cheaper than the national average. So it almost cuts an hour off.
mediaman··on PhD students face cash crisis with wages that don’t cover living costs
I was a little skeptical of this because of the impact of state-level minimum wage laws that are higher than the federal, so I pulled up the data on how many people are paid the US federal minimum wage.

Turns out that 1.1 million people have jobs that pay the federal minimum wage or less. But that number does not include tips, and about 60% of those people work in restaurants or bars. It's hard to say how much in the way of tips they get, but it's fair to say that there are at least 440,000 people who are earning no more than the federal minimum wage and don't get tips to make up for it.

That's about 0.6% of the hourly workforce. For them, filling a 15 gallon tank of gas would take 10 hours of labor, or even more after deducting employee taxes.

Also, while the federal minimum wage demographic skews young, about half of the 1.1m figure above are over the age of 25.

Above data is per the BLS.

mediaman··on I spent two years launching tiny projects
I just wanted to say that I love your phrasing of working on something that "you hate a little less".

In general I believe the idea of "work on what you love" to be awful advice; I have found that I get deep satisfaction out of working on something that someone is willing to pay for, even if that means that sometimes it is a headache and I hate doing some of the things that need to be done. Once you accept that those things will continue to exist and continue to need to be done, and it's not going to be bliss, those things seem to matter less.

mediaman··on Twitter Deal Temporarily on Hold
I may have been unclear. He wouldn't have to buy Twitter if the government prevented it, but he would be required to pay $1b. Outside of those circumstances, he is contractually forced to buy Twitter.

The government issue and the money issue are basically the two scenarios in which Musk would be unable (not just unwilling) to complete the transaction. So that's where he has to pay the fee. Otherwise, he can't change his mind.

mediaman··on Twitter Deal Temporarily on Hold
That is indeed the big question. The court case would go on for a long time, and in the meantime they'd be in this weird limbo and the business would suffer.

What we're really seeing here is: can a very rich person completely flout a contract, ignore legal obligations he signed, and abuse the legal system to the point that contracts are basically unenforceable against him? Sure, he has no case, and there is no real conventional legal way to wiggle out of this, but that's not really the question, is it?

As Matt Levine said in his column today, a merger arb told him that "if you're reading the contract, you've already lost."

In all likelihood, it seems the most probable outcome is a modest adjustment of purchase price to further "secure the deal" (whatever that means). But Musk is a bit crazy, so while that's the most likely outcome, it's hard to say it with much confidence.

mediaman··on Twitter Deal Temporarily on Hold
The $1b is not a simple breakup fee. He can't just say, "I've changed my mind, here's $1b, the deal is off." It's not a $1b option to walk away.

The merger contract provides the option to pay $1b to walk away under only limited circumstances: either the acquisition is blocked by a government, or Musk fails to get financing for the deal. The latter can't really be easily faked: if he said 'sorry, I thought I had the money, but I checked my couch cushions and it turns out I don't,' and paid the $1b to walk away, Twitter would sue and this assertion that his financing broke down would be tested by the courts.

In fact, out of those limited allowances to pay the $1b and walk away, Musk is bound by the contract to 'specific performance'. That is, he's bound to actually do it. Pay the $54.20.

A good M&A analysis of the situation is below.

https://yetanothervalueblog.substack.com/p/quick-twtr-though...

mediaman··on Passenger with “no idea how to fly” lands plane after pilot incapacitated
Sometimes the ground service people are allowed to move the planes on the ground for repositioning, when no passengers on on board. It's not uncommon for them to at least know how to turn the systems on and move it. And if you know that, takeoff isn't incredibly more difficult. Landing, on the other hand, is more difficult, but that didn't seem to be among his objectives.
mediaman··on “It Is Getting Worse. People Are Leaving”
That data is great - thank you for posting. It looks like rail-car miles have held somewhat steady, down a little bit, except for 2020, which I assume was COVID related.

I am being a bit facetious with the fiction comment.

However, I do believe there are a lot of problems in the rail industry. With the introduction of "PSR" (precision-scheduled railroading), trains actually have no schedule, and only leave when they're full. This maximizes efficiency.

It's very hard on rail staff, because it means they, too have no schedule. They work a train in one direction, do a staff change, and then go to a hotel while waiting for a return trip. They get as little as 90-120 minutes of notice for the return trip. So it's a lot of time away from home, with very little ability to control one's schedule. I imagine that this source of frustration emerges in a lot of different places, and it's possible that this letter was one of them.

mediaman··on “It Is Getting Worse. People Are Leaving”
I did find a GAO study that said that trains increased in length by 25% since 2008.

But the number of derailments dropped by more than 25% during that time, so the probability of a given car being involved in a derailed train still decreased.

There's been other coverage of the ironically named "precision scheduling" in the rail industry causing a lot of labor problems, because it makes for nightmarishly unpredictable schedules for workers. They never know when a train is going to leave until it is full, so after working a shift, workers stay in a hotel waiting for a return trip that they won't know about until 90-120 minutes ahead of departure. I am guessing that the claims around increased derailment may actually be about frustration with the work environment.

mediaman··on “It Is Getting Worse. People Are Leaving”
The government data I am able to find does not seem to support the idea that there is a massive (or any) increase in derailments. And this letter never provides any data supporting the premise either.

According to the Bureau of Transportation Statistics, there were 1,056 train derailments (both cargo and passenger) in the US in 2021. This was the lowest number in the dataset going back to 1975.

Ten years ago, it was 1,470 derailments. In the year 2000, it was over 2,000 derailments.

In reality, the data suggest we are on a strong downward trend in derailments per year.

Is this government data wrong? Or is this writer trying out a career in fiction?

Data: https://www.bts.gov/content/train-fatalities-injuries-and-ac...

mediaman··on IQ Test Made by Mensa Norway
There is no evidence that the SAT is a function of practice and training, and a lot of evidence that training has only a small impact on SAT scores.

See, for example: https://nepc.colorado.edu/sites/default/files/Briggs_Theeffe...

https://www.researchgate.net/profile/Benjamin_Domingue/publi...

These controlled studies show an impact of something like 10-35 points from coaching and other prep.

mediaman··on IQ Test Made by Mensa Norway
Most career data shows that once you reach 125-130 IQ or so, it stops really being a factor in career success, and other things matter more: grit, interpersonal skills, etc.

Being at the 99.9% level is impressive to be sure, but it doesn't seem to translate to much tangible benefit in our current society.

mediaman··on Amazon Q1/2022 [pdf]
I think the market is (hopefully) reacting to margins on their core retail business, and top-line guidance. RIVN has added lots of noise to earnings.
mediaman··on U.S. economy shrank at a 1.4% annual rate in the first quarter
There's a lot of talk about high inflation rates, and what the Fed has to do to combat them, including causing a recession.

It's true that inflation is high, but the data shows us that most of this is being caused by factors not related to high economic demand: it's energy and food.

Core CPI also increased, but it's tempered considerably: the annualized rate of the latest month over month core CPI change is now under 4% (3.9%), having fallen from an annualized m/m rate of change of 6.5-7.5% a few months ago. 3.9% is a little higher than our target of 2%, but not as alarming as the headline rate of total CPI.

Everyone made fun of the claims of "transitory" inflation from late last year, but it's looking like -- except for energy and food, which are being driven by international events -- the drivers of base inflation really were one-off events, and are not continuing at the same rate. For example, international freight increased dramatically, from around $3,500 for a 40HC from Ningbo to LA to $18-20k, but it has since stabilized to around $15k - and inflation is a rate of change, so in that area we are not seeing a continued paced increase in cost. It's just stayed expensive.

Of course, energy and food matter, a lot. Food prices matter more than the price of, say, imported lawn chairs. But it's useful to ask why inflation is happening. Is it because our demand is too high and there's competition for resources? Then we need to slow down the economy to reduce competition for resources. That's the Fed's job. Or is it because there's a sudden reduction in available global oil and grain supply? Then it's not clear that causing a hard recession is going to fix it. In that case, we need to fix the decrease in supply. And in our current case, that's not easy: much of this increase is being driven by a misinformed dictator of a nuclear superpower hell-bent on escalating a regional war, and this type of problem is not one the Fed is well equipped to deal with.

What we're really seeing is supply-side driven inflation. The Fed should continue to drive core inflation back down to its target level of 2-2.5%, but we are already seeing success here. The real question is what we do about energy and food, and it's not clear the Fed's actions are going to solve that problem.

mediaman··on Google Docs will “warn you away from inappropriate words”
And there is starting to be a movement against the use of the POC term, because it lumps everyone with more than a given amount of melanin into one homogenous “non-white” category, instead of recognizing the actual diversity of people from across the world. Like, what does a recent Malay immigrant culturally have in common with an American whose ancestors were forcibly brought over 200 years ago and enslaved?

To be honest, I can see why ‘POC’ would be culturally reductionist and have been kind of surprised at the progressive adoption of it. Seems obvious that it will be deprecated.

mediaman··on Intuit asked Mailchimp employees to pay medical costs out of pocket
Is health insurance not provided by the state in France? I thought employer-provided health insurance was just was an odd Americanism.
mediaman··on Ask HN: What bits of fundamental knowledge are productivity multipliers?
Part of the issue is that they often don't want you to decide what to build. They've separated out the "building" from the "deciding what to build", so hearing from a candidate that it's important to focus on what to do is off-putting to an organization that just wants you to munch through JIRA tickets all day and find technical solutions to whatever somebody else wants, and then keep a smile on as they change their mind and discard the feature.
mediaman··on No news is good news
The Rothschilds have had a substantial stake in the Economist for 70 years, including prior to the time when Pearson was a shareholder. Pearson did not control the board, had only 6 of the 13 board seats, and never had majority control of the company. It's unclear what is meant by "corporate owners;" Pearson is a publicly traded corporation.

I'm also a long-time subscriber and do not understand the "shift" you are referring to. The Economist has been pro-free-trade, generally against heavy regulation, and a magazine that consistently takes small-l liberal positions for a very long time. That stance means that sometimes they are right - deregulation has been good in many areas - but sometimes they get it wrong, when, for example, free trade may have adverse effects.

Sometimes they take positions that are bad - they argued for the second Iraq war - but they are pretty up front about their biases, and generally don't represent their biases as purely objective facts, as many traditional newspapers do. Their writing is clear and concise and is not meant to be consumed uncritically. The magazine is still unparalleled for what it offers.

← PreviousPage 14 of 34Next →