ceeplusplus

The AMD offerings are still very far from the M1. Try comparing the 6800U to the now 2-year old M1 in Geekbench. The M2 widens the gap even more.

The Ryzen chips are already clocking over 3 GHz, so there isn't much more scaling from power left. That's why 5nm Zen4 probably won't move the needle too much.

Not quite. The review intervals increase exponentially, so if you stop adding new cards the workload quickly goes to single digits per day in a few months.

35 a day is excessive for sure. Even with foreign language stuff which is the lowest time to solve per card, I limit myself to 10 new cards/day. Math problems 2 new/day. Daily workload ends up around 50-100 cards, which is very manageable. You do need to be consistent though.

A lot of ICE cars have much crappier range than a hybrid like a Prius. I know my old econobox only goes about 300-350 miles on a tank, which is comparable to a Tesla. Not everyone has 55 mpg cars that sip gas.

The refueling problem is big though. Maybe if they can cut it to 15 mins for a 15-90% charge it'll be good enough.

The problem with that theory is that writing code is easier than reading code. This is generally not the case in other professions. It is definitely not the case for an artist.

You still need correct code, and the halting problem says you can't prove whether code does what you want it to. At the end of the day, someone needs to be able to go in and fix shit the AI did wrong, and to do that you need to understand the code the AI wrote.

The link provides empirical data backing up the idea of 10x, specifically in the context of the speed and quality with which students come up with solutions to a programming project. Quality is measured by correctness and speed by time spent.

It also provides data showing that speed and correctness are essentially uncorrelated between different people - i.e. just because two different people finish at different speeds, does not mean the faster person has a less correct solution, at least in this specific setting. If you assume that's true generally, that makes the idea that there is a large gap between F500 programmers and FAANG ones even more credible, since the best programmers both produce better solutions and produce them faster.

Jobs where the difference in output between the best and worst is 10x. Generally creative professions like movie director, writer/journalist, lawyer, etc. SWE tends to fall into this category because the difficulty of writing good code is in some respects like writing good prose: the space of "solutions" is vast and there are few well defined rules by which a "good" solution can be constructed. Not only that, but writing it quickly while maintaining quality is difficult as well.

See for example this blog by Joel Spolsky [1].

[1] https://www.joelonsoftware.com/2005/07/25/hitting-the-high-n...

The thing that a lot of people don't understand about SWE (and a lot of other 10x professions) is that years of experience has a very weak correlation with skill. A top tier junior engineer could easily become better than a 15 yoe senior engineer who's not so top tier.

If you work at a Fortune 500 and then jump to a FAANG, this difference is very obvious.

> taking all that risk to be compensated at a rate equivalent to 351 employees

Do you understand how risk works? If you take on more risk, you should be compensated for that risk, else it doesn't make sense for you to take so much risk if you can get the same reward by taking less risk.

Executives as a whole (CEOs, VPs, even directors!) take a lot more risk - career and financial risk - than your typical line engineers. Their success depends on the success of the initiatives they lead, which can be subject to macro or competitive factors entirely outside their control. Those initiatives are generally longer timeline, meaning they have less chances at bat (and thus higher variance in outcome) than an IC. And a shitty IC can always be rehired pretty much trivially. If your company fails with you at the helm, you're not getting another CEO job. For example, Carly Fiorina has never held another management role in a large multinational corporation after her failure at HP.

On top of that, executives have to hold a large portion of their compensation as stock which they are barred from selling as long as they work at the company. If the company tanks, they take a retroactive paycut, often times upwards of 50%. They don't get refreshers to boost their comp back to original levels like ICs either.

Yeah CEOs nowadays have quite excessive compensation. But they will never have compensation at a single digit multiple of the average employee. The risk they take is too big for that to sufficiently compensate them.

Program managers have very little utility. Ops should really be automated (i.e. it should be half run by SWEs that automate everything they can), so real ops headcount should stay constant as the company grows as automation is increased. I would hope marketing is data driven on things like CAC, but who knows at a lot of companies. Lots of marketing people are not quantitatively fluent.

Executives also have to hold lots of stock that tanks when the company does poorly. Should workers also be forced to hold 100% of their RSUs as long as they work there and be paid 80% in RSUs? Even senior SWEs make 50% tops in RSUs and can sell their stock as soon as it vests with no restrictions. They get top ups to make up for poor stock performance. Their personal exposure to the company (risk!) is way lower than any executive. The CEO has effectively made negative salary due to their stock tanking.

Should companies be forced to keep unproductive or no longer useful workers? Clearly there is not enough customer demand for Robinhood to continue employing their ops/content/etc. workers. What are those workers going to do then, if they keep working?

Theoretically, you're picking off improperly priced homes and quoting a tighter buy/sell spread, so you aren't buying higher than the "market" price. The seller/buyer is the one crossing the spread.

In practice it seems like something isn't properly hedged so Opendoor and a lot of other iBuyers are exposed to the underlying asset rather than just capturing the spread and picking off incorrectly priced homes. That's why you see their margins go up when housing does well and down when housing does poorly. A proper market maker should have no exposure to the underlying (e.g. market makers made record profits in 2020 despite equities tanking). I bet if a Wall Street quant firm came in they'd roll over Opendoor and the like.

> It is if the arrest is pretextual

I wasn't aware that arresting someone for stealing or illegal drug use was pretextual. Oh wait, it isn't. People committed a legitimate crime.

> nobody who lives in a van, tent, or even on the street meets the standard for forced institutionalization by those conditions alone

Then you give them a choice: go to treatment or go to jail. It is a strictly better outcome than only giving them the choice to go to jail.

> “move them somewhere else” is not a sustainable solution

It's a fully sustainable solution for many cities in SoCal, like I've said. Go to Newport Beach some time and you'll see. If you want your city to spend billions supporting people who don't want to be supported, you can do that with your own money and vote.

> due process

You're not entitled to violate local laws without consequences. Notice I never said to specifically go after people for being homeless, but to go after people for violating actual laws. Illegal drug use, illegal trash dumping, stealing, etc. Just because those actions are associated with homeless people does not make them any less illegal.

> supported living (including mental institutions)

The entire point is that the political will does not exist to force people who should be in mental institutions but don't want to be, into those institutions. If you steam bug spray into someone's eyes or chase them with a metal pipe, you are a danger to others around you.

To be clear, it's 61k/year/tent. Not 61k/year overall. Overall spending is in the billions [1].

[1] https://www.sfchronicle.com/bayarea/article/S-F-has-an-unpre...

Even in Shanghai and Shenzhen, which have much higher density than NYC and the best public transit system in the world, people still use cars and like having parking space.

The product you're selling - public transit and high density walkable neighborhoods - isn't very appealing to most people. Public transit is slow, dirty, and dangerous. Neighborhoods without enough parking just force people to Uber everywhere or take public transit, raising the cost of living and wasting people's time. Smaller, hyperlocal grocery stores have inferior selection, higher prices, and worse quality than places like Costco.

And these aren't problems you can fix with more funding. Public transit is slow because NIMBYs block trains from running at high speed in most neighborhoods. It's dirty and dangerous because there is no political will to screen riders or to arrest vagrants. Smaller grocery stores suffer from inferior economies of scale, so they never will compete with places like Costco in price.

The price of basic goods like bananas in a bodega type store is triple what it would be at Costco. $1.29-1.99/lb for bananas vs. $0.50-0.70/lb at Costco. And the Costco bananas will be bigger and in better condition than the ones at the bodega. Make the comparison with baked goods like muffins or croissants and it's no contest.

> Forced institutionalization is plainly unconstitutional

It is not illegal to arrest someone for stealing/drug use/etc. and then have a psychiatrist examine them to determine whether they should go to jail or be institutionalized. Courts regularly examine criminals for mental fitness to stand trial already.

> no particular evidence that you can disappear whichever group of undesirables you see fit and guarantee that more won't appear

Funny, because that's exactly what lots of cities do and it works pretty well. Take a look at any wealthy SoCal city and you'll find no homeless.

> a universal housing mandate, and adequate social services (including harm reduction)

So, spending more than the median American makes per year on tents? Or allowing crazy drug addicts to trash temporary housing and hotels? Because that's what happens in SF. From the SF Chronicle, which is a super left leaning publication: "Residents have threatened to kill staff members, chased them with metal pipes and lit fires inside rooms, incident reports show. At the Henry Hotel on Sixth Street, a tenant was hospitalized after a neighbor, for a second time, streamed bug spray into their eyes, public records show. Last May, less than a mile away at the Winton Hotel, a resident slashed another tenant’s face with a knife, leaving a trail of blood out of the building" [1]

And from the director running one of these programs: "“We throw all these people into our hotels who have no demonstrated ability to live independently, who have severe problems and need to be elsewhere,” Shaw said"

Clearly the problem is not housing, it's the lack of support around getting mentally ill people the treatment (i.e. institutionalization) they need. And obviously, a crazy person is not going to consent to treatment for their craziness, so you need to force them into institutionalization.

[1] https://www.sfchronicle.com/projects/2022/san-francisco-sros...

> cities of course do that all in the name of public parking no matter the opportunity cost

Cities do not give away public parking. Either you have to pay for it (metered parking), it's provided "free" to customers only (i.e. paid for by the profit margin on goods and services), or it's only available to people who live in an area (i.e. those people paid property tax, either as owners or renters).

> everything but housing will do, presumably

Housing doesn't solve rampant drug addiction, theft, mental illness, and laziness. The article in the OP literally cites trash accumulating, rats running rampant, and vehicle fires spontaneously occurring. That is not the environment that a rational person occupies.

San Francisco spends $61k/year/tent [1] on homeless tents, which is more than median income in the US. The problem is not funding. The problem is that the local population would rather spin performative activism instead of taking real action (i.e. institutionalizing mentally ill people and forcing people off the streets and to stop using heroin and meth).

[1] https://sfist.com/2021/03/04/insanely-it-is-costing-san-fran...

Ironically, another commenter has pointed out that the USSR did not tolerate this kind of behavior (and neither does China). I suppose in that case having the "socialist commies take over", as you put it, would be an improvement over the status quo when it comes to homelessness.

> These are people who, mostly, never had a chance. This is the key point. They were dealt a truly garbage hand in life

Have you talked to any of them? Most of the ones I've talked to don't want anything to do with work. Yes, work on its own is not enough to rejoin society. But work is a prerequisite for rejoining society. It is necessary but not sufficient. If you don't want to work you cannot rejoin society.

At some point you need to assert some tough love and force them to do things they don't want to do. Like work. Or not drink/do drugs.

> So you support the state housing them at taxpayer expense

Describing prison as "housing at taxpayer expense" is... one way of putting it.

I think they should be given a choice: go to jail or get a job and 2 months of state funded housing. If you go back on the streets because you didn't want to work, then police will pursue every public nuisance, public urination, shoplifting, etc. charge and send you to jail. If you're mentally ill you're forcibly committed. If you're a drug addict, you're forced to go to rehab or go to jail.

Otherwise these people will never learn to be decent members of society (you know, people that don't steal and shit on the streets - the bar is on the floor).

A lot of these people don't want to go into shelters because those shelters have no drug use/no alcohol policies. But I say the state should not be supporting their drug habits with taxpayer dollars.

Not a surprising outcome given Europe's underspending on defense. The US has long been subsidizing EU defense via security guarantees and foreign bases and this is the natural outcome of falling behind in domestic R&D spending and relying on US defense.

The F35 might have been a colossal money blackhole but given two equally inefficient defense contractors the one with more money wins.

If Bloomingdale is the same Bloomingdale Google Maps tells me is in the middle of Manhattan and 2 blocks from Central Park NYC, yes, it is an enclave of wealth and privilege. Try riding your bike around the Bronx some time, it'll broaden your world view.

> I went to public schools here

There's a world of difference between a public school in a wealthy area and one in a shitty one. The former might as well be a private school in terms of the quality of instruction and available funding from PTA donations.

Now factor in redundancy and server costs, and the difference is not so huge. You're still paying a multiple of the raw cost, but unless you're storing on the exabyte scale I think it shouldn't really matter in the grand scheme of things.

Cost of a single engineer to manage a Minio cluster probably already outweighs the extra cost you're paying at any reasonable scale (i.e. most companies). And if you're a big player the published costs are not what you're paying.

> since there's no homicide. The woman didn't die.

Yes, so it wouldn't have shown up in your touted homicide stats, yet the fact that it happened makes everyone who rides the subway feel less safe.

How many attempted homicides, attempted muggings, stalkings, etc. are there that never get reported?

> things left outside your house DO tend to vanish, especially if left overnight

That's quite a sad thing you've grown used to. I've lived in suburbs in several states and not one would have me saying that things left outside my house tend to vanish.

You should expect more from your fellow citizens. Not stealing is a basic decency and sending thieves to jail should be a basic tenet of society.

Here's an experiment you should try, since you claimed to never have anything stolen. Buy a bike, any bike, and one of those cable locks. Lock it on a public street and leave it for 8 hours during the day. Then see if it's still there at the end of the day.

I guarantee you it won't.

> Homicide victims usually did something that enhanced their risk

Like... take the subway in NYC [1]? Homicide victims in normal cities that don't have mentally deranged homeless roaming the streets usually did something that enhanced their risk. Homicide victims in NYC... had the unfortunate luck to be selected by mentally ill homeless that should have been institutionalized.

[1] https://www.reuters.com/world/us/man-wanted-pushing-woman-on...

The engineer who already has a job at a random company has a known skill level and growth trajectory. The engineer out of a good school has an unknown growth trajectory. Conditioned on the accomplishments of alumni at the school and the (subjective) acceptance bar of the school, the new grad engineer could have a much steeper growth trajectory than the established one.

Silicon Valley got this right on the money ironically [1]. If you have nothing to show, people assume a potential future ability level rather than basing their decisions on your actual ability.

[1] https://www.youtube.com/watch?v=BzAdXyPYKQo

Adding more ads and bringing down video serving costs are tangible, achievable goals. Solving self driving is not. I honestly don't think Google has the right culture to deliver either. There's no hunger to win there.

Google funds a huge number of teams internally that (in my opinion) don't do anything that contributes to the top line or bottom line or the company's image.

Not only that, but Google is known for being a "chill" place to work - good WLB, low pressure to deliver, very low chance of PIP. That attracts the wrong kind of employee. When you start seeing TikToks of people who work at Google documenting their life rather than delivering results, that's when you know the company's standards are too low.