Python is also a more transferable skill than Fortran.
I bet that plays some role too in its popularity in the scientific community, which has many young anxious grad students/postdocs looking to ensure they are employable.
Many of the introductory programming courses seem to favor Python. If you already know a little Python, I can see the temptation to stick with it instead of learning Fortran.
One of the more fascinating aspects of these is how smooth the second hand movement is.
Second hands on Quartz jump once / second. This is to lengthen the battery life.
On mechanical watches, they are smoother than Quartz since the escapement releases power multiple times / sec. But still ever so slightly jumpy since power is still released in discrete increments.
Spring Drive is outta this world smooth…it can do this since battery life is not an issue since it’s mechanically generated power that can be rewound…for practical purposes, it is releasing power continuously, see for yourself here:
https://m.youtube.com/watch?v=jcHA5rBQxQc
It's so funny how the smoothness of the second hand has changed in desirability over time.
It started with mechanical watches that moved relatively smoothly at 3-6 beats per second. Then Quartz came along, and it became fashionable for seconds to move on the second (the "quartz crisis"). Then mechanical watches became fashionable again as quartz watches became commodities during the "Mechanical Renaissance", and it's now a sign of luxury for a "smooth sweeping" second hand again.
And then you have these modern outliers, like the F.P. Journe Tourbillon Souverain, which feature a "remontoire" that stores up energy before releasing it on the second for increased accuracy. So you can pay $250K for a watch that moves like a $10 quartz :)
Actually, at the very high end of luxury watchmaking they prefer lower beat movements as it increases the service interval, exotic escapements like the remontoir are primarily for exhibition purposes/bragging rights.
If you’re on a budget and want a deadbeat seconds hand, Jaeger LeCoultre has a deadbeat seconds watch, the Geophysic True Second, which is ‘only’ around $15k :) I believe it has been discontinued and is only available on the secondary market.
I had an electromechanical Timex with a dead beat seconds complication. Which means it was a battery powered watch with a balance that was supposed to be ticking 4 times / second and an extra complication added (the dead beat seconds) to make it tick once per second.
If you'd put your ear to it you could hear it tick 3 times in the background and then a loud tock.
I used to be fascinated with that, until I saw a wall clock that had a smooth second hand movement. It was a cheap $10 clock, because it's not an issue to put a larger battery into a wall clock.
Maybe I already got my kick out of seeing a smooth second hand movement and no longer feel the need to look at it on my wrist. Or perhaps my fascination was based on some gatekeeping, and seeing a cheap item with a similar feature made it disappear. Likely both.
I guess people who genuinely admire the engineering effort would be left unaffected.
I wear a mechanical because the loud tick of a full size 1Hz second handdrives me bonkers. Although I can handle the small seconds in a quartz chronograph.
While most quartz watches use cheaper stepping motors, there are also quartz watches which use synchronous motors, so the hands have a perfectly uniform and noiseless rotation movement.
I had some big wall clocks of this kind, and my father had such wrist watches.
The energy consumption of synchronous motors is lower, because they only have to overcome the friction forces, without having to also accelerate the mass of the hands.
I am not sure, because that was some years ago, when my father, who used the watch, was still alive, but in any case the battery lasted at least a year.
Yeah, not that I perceive it to be an assumption that is entirely illogical, but why is smooth movement supposed to consume more power? I can understand that jumping less often conserves power but the power hierarchy should be more ticks > less ticks > no ticks, if we ignore potential increased frictions at lower angular velocity as well as challenges of resisting disturbances.
At large sizes, synchronous motors are much more efficient than stepper motors, so they use much less power.
At small sizes, the synchronous motors must use permanent magnets, which increase their cost and they have windings that are more difficult to make and the difficulty increases with the smallness of the motor.
The electronic drive of a synchronous motor is more expensive, because it must generate sinusoidal currents, not rectangular currents.
At small sizes, a synchronous motor may have a lower torque than a stepper motor , so it might need extra gears, which would increase the cost.
As long as it is still cost-effective to manufacture a synchronous motor, it will always have a better efficiency and a lower power consumption than a stepper motor. The reason why stepper motors are preferred is that at very small sizes they can be much cheaper, especially when including the cost of all associated electronic and mechanical components.
> On mechanical watches, they are smoother than Quartz since the escapement releases power multiple times / sec. But still ever so slightly jumpy since power is still released in discrete increments.
Yup, and the higher the beat rate, the "smoother" it looks. Grand Seiko Hi-Beats and Zenith El Primeros come to mind. There's a good Hodinkee article describing the tradeoffs of different beat rates [1].
Seiko has a 4Hz series of inexpensive quartz movements (VH31) that are at least as smooth as the entry level Seiko NH35. Bulova has a significantly smoother 15Hz quartz movement but you’re going to spend $600 getting it in a watch.
I think I'd be more worried about my job in a 10000 person company than a 300 person company.
If Im in a 300 person company I'm probably one of 40-60 devs, and most of my peers know me and have met me. If I'm a senior dev, there's a good chance I'm on first name basis with the executive team, and they know what I work on and the impact I have. There's a lot better chance of me being able to demonstrate my value and avoid being let go if cuts are needed.
At a 10000 person company I'm basically guaranteed to just be an employee number next to a salary number in a spreadsheet, when it comes time for cuts.
Bitcoin “crashed” to a market cap of 440 billion USD.
It is valued more than the vast majority of publicly traded companies.
From that measure alone, why should anyone think “it’s dead”?
Bitcoin ain’t tulips, ain’t the pets.com, etc, to think something that is 14 years old and valued at 440 billion is going to die anytime soon seems…so wrong
My argument is that $10B of Bitcoin isn’t worth anywhere near that.
You can’t sell it without crashing the market, and you probably can’t borrow multiple billions of cash against it either because nobody in the crypto ecosystem has that kind of liquidity anymore. I’m open to be proven wrong by evidence.
Argument from ignorance fallacy. I'd love to prove you wrong, but the fact is that it is impossible to prove.
My own experience says that there is a lot more money out there than anyone can ever hope of comprehending. A good portion of that money belongs to people who are never seen. Not everyone tries to be a public figure. Not everyone is in the US. I just saw a video of a G-Wagon in Saigon (~$650k)... Vietnam is a very poor country on the outside, but you'd be astonished how much money people have there.
The amount of money in crypto is staggering. I know of people who started mining ETH from day one... and didn't stop. I can only imagine their wealth. It would be easy for them to sell off chunks over the last 4 years without anyone noticing.
Needless to say, I'm sure that if someone wanted to get rid of it, they could. It might take some time and effort, but when you have god levels of money, all you have is time.
I feel like you and I are actually arguing the same thing. A lot of people in crypto are obviously crypto-rich and have been converting to cash over the years to buy those fancy cars and penthouses.
So who's left buying today? Retail frenzy brought the fresh dollars but it's dried up. Those crypto-millionaires don't need to convert their cash back to crypto. They own enough crypto already, and (as you describe) they went through an effort to build cash positions, so why revert that... So where does the $10 billion in cash come from?
I know that institutions and high net worth individuals (hnwi) are quietly buying. There are companies like attestant.io who are staking ETH for (hnwi).
> Retail frenzy brought the fresh dollars but it's dried up.
For now. Being in the industry myself, I've had a saying for years, like it or not... magic internet money always goes back up. I feel like you're just looking at one blip in 13 years of blips, throwing up your hands and saying game over. That's like saying after the dot-com bubble bursting, that tech would never go up again.
You can definitely bet that people are buying the dip. It is starting to show in the recent upward price movement too. 20%+ gains in a few weeks are not small.
You don't necessarily need a bank for that; DeFi platforms like Aave can provide the equivalent of a margin loan. Looks like the current (variable) APY for WBTC is 1.11%.
> From that measure alone, why should anyone think “it’s dead”?
Because everyone is waiting to exhume it when it's worth something. Bitcoin has no utility, which is not a knock against it's implementation but it is an ill-omen for it's value. The concept of 'digital gold' only works if the gold is inherently worth something. Owning a piece of a blockchain that's too expensive to transact on is a detriment, not a positive.
Not to insinuate there's some 'better investment' though. Crypto is dead because the dream of democratized currency hit the mainstream and got rejected, big-time. Use it if you want, invest for however long you'd like, but you should make peace with the fact that a crypto future is nothing more than opportunistic ankle-biting.
It's not Berkshire Hathaway, Bitcoin's "test of time" was 3 years of media attention that is now coming to a close. Regulation and taxation is disincentivizing people from using it. Exchanges and liquidity holders are being ousted as scammers, it's basically The Reckoning for everyone who promoted a Laissez-faire economy.
I love the idea of a democratized currency, but Bitcoin's value is propped up by very little now. As I said before, you cannot run an economy on goodwill alone. Someone's getting fooled into holding the bag, and the longer you hold out for a payday, the greater the chance that fool is you.
A democratized currency sounds kind of neat, but bitcoin was never this. It started out as an oligarchical currency with the halvenings (especially spaced the way they were). It became oligarchical at the transaction verification level (the level of taxation) when ASICs were developed for it.
That a whole decade more for people to learn about it, and especially since the main features has not changed. Yes, there are improvements to the protocol, but the main thing has changed little over the years
Betting on the stability of a protocol is a good way to lose your money when that protocol is obsoleted.
Edit: I'm coming right out and saying it; if we even have to talk about L2 chains, you might as well just admit that Bitcoin itself needs an update to be usable.
Paper cash was an innovation to solve the issue that gold is hard to lug around. And very much an analogy of a L2 network
It is arguably easier to counterfeit than gold (there is fools gold too…)
The base layer is very secure. L2s will be faster, but have a trade off of being less secure.
Note: the L2 Lightning Network is different than L2s for ETH like POLY, POLY has its own market cap (garbage idea IMO). Lightning is just a tool to make BTC more efficient
See, it's already over. If you have to explain that to a layperson, it's not going to get adopted. Furthermore, if people are taught to adopt this mentality, that puts them at pretty great risk for getting scammed.
Again - do whatever you want and invest wherever you please. But L2 chains are not catapulting crypto back into usability, and Lightning is living evidence that Bitcoin's bet failed. If you need auxiliary technology to transact the money you supposedly own, your protocol is broken to the core and needs replacement.
That's only good until enough people get burned so bad they never touch it again. Then it has completely no utility.
Gold has survived as a speculative vehicle because it's been relatively stable over a much longer time period. That, and unlike Bitcoin, gold is actually a physical asset that is not held together by (wasting) electricity.
> How could ANYTHING stand the test of time and be worth 440 billion if it has no utility? Genuine question.
1) markets aren't rational nor are the people in them
2) BTC is a currency but is being treated like a commodity
3) a lot of big orgs, investors, and even governments bought in and now either hold and pray for growth, or else pray it doesn't drop. They have every incentive to hype up its value even though its use for most things is essentially moot; your Mom isn't going to use BTC to buy catfood, and its utility on the dark web is now overshadowed by things like Monero. You HODL and hype cuz otherwise you lose billions.
It's too soon to say it won't fall further. The chart is still very weak. It won't take much for btc fall below 16k again. Another major failure like gemini or greyscale will do it.
Yes, it's worth a lot still, but for most btc investors long-term returns have been bad unless you bought pre-2017. BTC is only up 15% from its 2017 highs (from 19.5k to 22.5k). This lags almost all index funds.
The combined value of oil or copper is worth a lot too, but this does not make it a good investment either.
Maybe all the talk about the death of crypto just reflects where we are in the cycle.
The mood feels similar to late 2018 when most people weren't paying attention at all, and the few who were paying attention were just disgusted by the whole thing.
When the conventional wisdom is that everything from art to sports to real estate will be imminently cryptoized, you know it's frothy, and near the top of the cycle.
When the conventional wisdom is revulsion and disgust then maybe we're another six to eight months away from not thinking about it at all, which is probably the bottom of the cycle.
How could something be worth 440 billion in market cap after a major correction not be valuable?
This is after having a 14 year history of people learning about it, not some scam coin like LUNA that went from zero to 40 billion and to zero in less than 1 year
If I take 80 Dollars from you and promise to return you 50, is this operation also worth 50 Dollars? If so, what‘s the meaning of „worth“?
If I open a bank account and someone else deposits 1 billion there and I have a contract that obliges me to return the money in a year and I am not allowed to move it or use it as a security, is my bank account or the bank creating it „worth“ 1 billion?
Whether bitcoin is worth something as a tool/currency cannot be measured well by the amount of money stored in it.
For the simple reason that market corrections are not absolutes: the only thing that a correction down to 440B tells us is that it hasn’t been corrected down to 0 yet.
It may never be, but that’s just to say that “it’s still valued” is not a good argument for “it’s correctly valued.”
I mean, not to defend Bitcoin, but if you have invested at $60 and sold at $60K - it has provided tremendous value. Most people are going to lose money on this “investment”, but just saying that it provides no value at all is a tired argument.
This is nihilistic value: the “value” of Bitcoin seems to be latent not in its productive capacity, but in its ability to extract value from other people. Which is to say that it has the same value that a credit card skimmer does, and that that is not a useful definition of “value.”
It’s a good argument. I don’t fully agree we should limit the definition of value to only its (subjectively) productive quality. It’s like gambling - I personally don’t see value in it but, hey, many people love it so they get something from it.
I said nothing about Bitcoin, only about your argument. Responding to an argument requires reading it first, which is a bit more skilled than just knee-jerk reactions.
If crypto is going to be regulated all fees need to come out of taxes from crypto earnings. Otherwise it isn't paying its way. And since it isn't creating general goods for the broader market, the taxes of the broader market shouldn't be expected to cover it.
If feels very very very wrong to compare a search during April 2020 and now.
Why? The year after April 2020 was a booming job market for SWEs. Hiring managers were competing for talent, 2021 was a boom year of low interest rates, euphoria in private and public markets.
This blog post _should not_ be considered a response to the WSJ article.
And sorry, grinding away at leetcode should only take about 2 months tops if you don’t have to worry having to balance leetcode and your job, and not a tall ask for a 250k+ job in tech. if you don’t need that high a salary, smaller companies generally are more lax about leetcode
grinding leetcode is some dumbshit waste of time that I'm not going to bother with -- I've got code to put in production and this simply is never a good use of my time. When you have an obnoxious hiring process that will take multiple hours for your special company, why would I even waste my time? I can click a few more buttons and have offers elsewhere anyways...
Author also mentions that he didn't have any savings at the time he got laid off. So maybe not the normal person that's worked at Apple for 10 years and cashing out RSUs but someone who did short stints at each.
> The year after April 2020 was a booming job market for SWEs.
Did we know that at the time, though? From what I remember, it was all doom and gloom: public markets were trending downward, people were talking about recession and corporate belt-tightening, and the assumption was that we'd see a lot of layoffs, and the labor market would get tight.
True, that's not what happened (at least in the case of knowledge workers) -- in most cases it was the opposite -- but I don't think we had that foresight back in April 2020.
Anecdotal but over the past 20 years I've held about eight different positions in various companies as a software engineer, and I've never even so much as glanced at leetcode. I've been able to live comfortably and put plenty away for the future with 150 to 200K salary.
I feel like companies that emphasize interviews prioritizing rote memorization aren't going to be a good culture fit for me anyway.
I, for one, assumed that at least half of HN would agree with you. HN has a skew of people that are pro DDG, but not so much that this is all that contrarian of a view
I don't know, I think he is undoubtedly a newsworthy individual, and if it is true that he is not receiving a speaker's fee, I think it makes sense for them to keep him on the agenda.
I bet that plays some role too in its popularity in the scientific community, which has many young anxious grad students/postdocs looking to ensure they are employable.