Hacker Newsnew | past | comments | ask | show | jobs | submit | idontwantthis's commentslogin

Yeah not my experience at all. I am an at will employee and I can wake up tomorrow to a locked laptop.

I said "without risk of lawsuits", and that phrase is doing a lot of work. Also, you're looking at it only from the employee's side.

Multiple statements can be true at once:

An ordinary employee who is just doing their job and hoping to continue doing their job can be laid off for any number of reasons that are outside of their control. The most common of these is company financial performance, which is both completely invisible and completely uncontrollable for an ordinary employee. Behind the scenes, the decision-makers in these layoffs have very scrupulously designed them to ensure that the people deciding who is laid off have no knowledge of an employee's membership of anything that might be construed as a protected class. This is why they often look so illogical.

A manager who has a problem employee can have a terribly difficult time getting rid of that employee in a way that is legal. My boss had one such problem employee, who did literally zero work after being hired. When my manager had an experienced engineer pair program with him, he literally had to be told what to type, keystroke by keystroke, to get him to produce anything. When my boss called him on this, he produced a doctor's note saying he was depressed and went on disability. Short-term disability can last up to a year, and you cannot fire someone on disability without a paper trail that says "You fired someone on disability, you knew they were on disability, and that is a protected class." Now your lawyers have to argue in court that the fact they were on disability had no bearing on why they were fired.

A person who takes the law as a tool with which to screw others over will do a better job screwing people over than the employee who is just trying to do their job and incidentally wants to keep it, or the manager who is just trying to provide a supportive environment for their reports to do their best work. Because that's what they're paying attention to. If you've got a lawyer combing over case law and statutes to understand the fine points of what you can and cannot be fired for, you can act in ways that ensure that you will win a big payout if you are ever fired. These ways probably bear no resemblance to what our notions of a "good employee" or even a "decent human being" are. But they are effective at collecting a paycheck, potentially from multiple employers at once, and being able to collect a bigger paycheck if your employer tries to get rid of you.


> I said "without risk of lawsuits", and that phrase is doing a lot of work.

It is like saying that it is hard to go swimming without risk of shark attack.

Mostly wrong until you share a bunch of cherry-picked scenarios to support your own beliefs.

“Well yes I knew a guy who liked to splash around in the ocean at sunset wearing his seal costume after ladling chum into the water all afternoon.”


I’ve found the complete opposite as recently as last week. When I ask a deep question about a book it will hallucinate whole paragraphs of bogus justification and even invent characters

Did you give it the text of the book and tell it to answer based on that?

Does this feature put the entire text of the book into the context?

I hope not, it probably only goes up to the page you are on in order to avoid spoiling the later content.

The LLM just working on its own is just generative intelligence. You have to ground it if you want the real stuff. The Kindle app has the text of the book and I'd want it to put that in the LLM context.

The entire book in the context at once?

That's probably 100k-150k tokens for most novels.

That’s more reasonable than I thought.

Can anyone explain why RIVN is down 8% after this announcement? Were investors expecting hands free handjobs or something?

I hold some RIVN and I'm wondering why they're spending resources on custom silicon instead of using something off-the-shelf. What is their advantage here? Can they hire the right people? Can they ship enough units to pay for it?

Those are my bearish questions. On the bullish side, the VW deal shows that they're willing and able to license part of their platform, so possibly have a big chance to recoup costs and maybe turn a profit just on that side, which justifies a big software + autonomy investment.


If their idea is both novel and useful, and if it actually works, and they can actually produce it, then: They can sell it to other automakers.

(GM has made a lot of cars with their own transmissions. And at various times, they've supplied -lots- of them to other automakers all over the world. They've made a lot of money doing this.

Someone's gotta build the machine vision/control systems for all of these self-driving cars; that someone may well be Rivian.

It's not as sexy as something like a new convertible might be, or a $40k self-driving electric car, and a consumer might not even know that the new car in their driveway has expensive Rivian parts buried inside, but that future can be very profitable for them.)


>On the bullish side, the VW deal

Oh, I think everyone missed this. Rivian is betting Elon made a big mistake by designing FSD to be strictly for Tesla. Rivian are doing FSD to license it out to other manufacturers. They're planning to open a new market.


But how could they beat Waymo or Tesla? At some point, these companies will offer an L4 package to car makers.

Who cares about Tesla, the comparison here is Waymo who has already been doing this for much longer.

I was just thinking about this recently. It doesn’t seem to make sense for each brand to have its own self driving implementation.

Fair point. I think they wanted to sound super futuristic (they often borrow a page from Apple's book) but they forgot they're not Apple.

Maybe they think custom silicon is biting off more than they can chew, coming at a time when they need to focus on R2 production and scale-up.

They just told potential buyers to not buy an outdated Gen2 or an early R2, assuming it is not delayed.

It could be that Gen 3 shipping late 2026 is a concession that R2 might be delayed until then.

Personally I think they will ship R2 Gen 2 vehicles to the early adopters that are less concerned with ADAS.

My R2 reservation is very late (I had to redo it for reasons) so I probably won't be able to order one until it's available anyways.


“Buy the rumors and sell the news.” Just typical market stuff.

Now it almost reached $20 lol.

Excessive hype leading up to selling the news, happens all the time.

> try to regulate idling, which worse emissions profiles

Most certainly regulated. There are people who make a living off of reporting idling trucks and collecting the bounty.


Not for cars what so ever. There is a 3 minute rule which is unenforced.

That's both amazing and hilarious just like filling a plane with hard drives is both insanely effective and just plain insane.

Just to put the market into perspective, the UK was looking to build an undersea cable to push power from Morocco but canceled it [1]. That cable would have provided somewhere around 8% of the power to the UK and the whole project was slated to cost 25bn pounds (~$33b USD). Imagine sahara solar shipping could provide that but also supply (cheaper) to Spain, Italy, etc. As the tech gets better and cheaper the Americas could become customers too. Suddenly Timbuktu becoming an energy hub for the world doesn't sound as crazy as, well, it sounds.

[1] https://apnews.com/article/uk-morocco-renewable-energy-xlink...


We already have too much solar in Spain, I don’t think we need to rely more on it. Even worse if it’s coming from a strategic enemy.

AWS Snowmobile exists , I wouldn’t call filling a truck (or a plane) with hard disks insane .

mmm, it was discontinued in 2024 - so maybe amazon decided it was insane (well, insufficiently economic?) after all?

Less demand these days, In 2015 when it launched there was lot more drive to move to the cloud, lift and shift was the industry mantra. By 2024 the kind of orgs with >100 PB have already moved to cloud or have no plans to do so.

The current solution is we can bring our own devices and reserve ports on a AWS Data Transfer Terminal. It costs $300-500/hour USD for a 100 GbE bandwidth so not really cheap.

While AWS is stopping doing devices for migration (not economical at low volumes these days). They however still support physical transfers so customers can pack their own planes so to speak with hard disks to the AWS terminal.


Does it run on apple silicon?

Apparently - https://github.com/ivanfioravanti/z-image-mps

Supports MPS (Metal Performance Shaders). Using something that skips Python entirely along with a mlx or gguf converted model file (if one exists) will likely be even faster.


(Not tested) though apparently it already exists: https://github.com/leejet/stable-diffusion.cpp/wiki/How-to-U...

It's working for me - it does max out my 64GB though.

Wow. I always forget how unlike autoregressive models, diffusion models are heavier on resources (for the same number of parameters).

I’ve found wim hof no more effective than laughing gas. Very short euphoric high then back to where I was plus a slight headache.

I suspect you’re missing something important- not surprising since Wim is a really unclear teacher. The low oxygen and/or high co2 causes your body to panic and you are supposed to use that as a stimulus to gradually train conscious control of your heart rate and stress response and calm yourself down. This is where the benefit comes from, not the short term physiological panic itself, whose symptoms you are describing. In fact, those symptoms should go away entirely when you learn to calm the panic and consciously control your autonomic nervous system. This then also primes you mentally to be able to go into ice water, which is also a large stimulus to learn to calm, plus trains your brain to more easily jump into anything difficult. All of this takes months or years of consistent practice to gradually develop a relatively permanent sense of calm and internal strength among other things, and is really just a simplified form of tantric tummo meditation.

where can I learn more?

I'm not sure. I figured out most of that from taking Wims online class, and through years of personal practice and experimentation. I found that I was able to pick up a lot more by watching Wims body language and actions in videos, that he never directly explains, but I am not sure everyone could pick those up.

Ultimately, Wim's stuff is mostly based on tummo and other Vajrayana/Tantra yoga practices, and you can find both books and in person teaching locally in most places that cover these things. Wim never even explicitly mentions this, but effectively what he is is a yoga master that has developed yet another new system of yoga. Before the 'Ice Man' stuff he had even developed and written about other yoga innovations he had come up with.

Another thing worth mentioning is that while Wim is truly a master at his own method, he tries to rationalize everything with a scientific backing, but he has absolutely zero understanding of what he is talking about, and it's all nonsense. I have a biomedical background, and have developed some of my own theories about how his stuff is actually working physiologically and psychologically, plus there is some academic research out there you can read both on his methods and Tummo itself- but ignore everything he says about that, he doesn't even understand the basics.

Also, ignore everything he says about putting breath pressure up in your head/ears- absolutely never do that, as it can cause tinnitus, and the methods work fine without it.


Then again, Wim Hof costs you $0 and laughing gas gets expensive.

How about just laughing without the gas? Ok you have to find something to laugh about. Nowadays turning on the news does that fairly effectively.

Propellant transfer isn’t necessary for starlink launches.

Of course they do. Everything a parent buys for a child increases GDP.

GDP is not measuring what is bought, but what is produced.

Bingo!

Some will argue that consumption drives production but according to the common definition children don’t contribute to GDP.

Assuming governments are going to address population growth/decline then it’s a choice between incentivising births or issuing visas.

Even in countries that have free healthcare births are in decline so it’s not the cost of children alone that is causing this situation. I would argue it’s the economic crutch called immigration.


I’m quite diversified using Schwab and fidelities funds but I don’t know how to figure out what percentage is AI or AI adjacent.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: