2,015 karma · joined February 13, 2018
The only thing you need is leverage, and yes, rank-and-file are not going to have a lot of leverage at a FAANG.
You can’t negotiate if you’re not willing to let the job go through.
Lots of Women participate in these discussions regardless of if they "feel more female" or "feel more male", but men mostly participate in the discussion because they "feel more female".
IMO the main reason most chatbots claim to “feel more female” is that on the training corpus, these kind of discussions skew heavily towards females because most of them happen between young women.
By the same logic, we should worry about the sun not coming up tomorrow, since we know to be true:
- The sun consumes hydrogen in nuclear reactions all the time.
- The sun has a finite amount of hydrogen available.
There’s a lot of non justifiable assumptions baked into those axioms, like that we’re anywhere close to superintelligence or the sun running out of hydrogen.
AFAIK we haven’t even seen “AI trying to escape”, we’ve seen “AI roleplays as if it’s trying to escape”, which is very different.
I’m not even sure you can even create a prompt scenario without that prompt having biased the response towards faking an escape.
I think it’s hard at this point to maintain the claim “LLMs are intelligent”, they’re clearly not. They might be useful, but that’s another story entirely.
Yes, you can get the JVM to crunch some numbers relatively fast, particularly if the operations are repetitive enough that you can pull JIT tricks.
Now try to run something that looks a bit more like an actual application and less like a cherry picked example, and as soon as you start moving memory around the gap jumps to orders of magnitude.
You have to be either clueless or delusional if you really believe that.
For me this is a clear case of negative externalities inflicted by software companies against the population at large.
Most software companies don’t care about optimization because they’re not paying the real costs on that energy, lost time or additional e-waste.
The evaluation needs to happen in the margins, even if it saves pennies/year on the dollar, it’s best to have those engineers doing that than have them idling.
The problem is that almost no one is doing it, because the way we make these decisions has nothing to do with the economical calculus behind, most people just do “what Google does”, which explains a lot of the disfunction.
Modern academia has terrible incentives. The easy path to success comes from not rocking the boat, even the guys at Data colada (who did the leg work to prove the case) have made a lot of enemies in the field.
I’m willing to bet there’s a large amount of less significant fraud that gets better defended, it’s only the people who are so successful they let they guard down that get caught in this silliness.
The last thing you want to do before pointing out that the emperor has no clothes is get a job at the palace.
Power control is the kind of stuff that benefits from very tight integration, and PCs just don’t have that.
Firmware is seen by most vendors as a pure cost to minimize, so you get a fragmented market full of subcontractors delivering the bare minimum that is considered “working”. Manufacturers also know most people aren’t going to use a big part of the functions they’re supposed to provide to OSes, and no one is really checking them, so it’s very common for devices to have only partial support for things they supposedly do.
First, no, it’s not all unsafe, it’s not even 50% unsafe.
Second, even 50% unsafe is an upgrade from 100% unsafe.
The problem is that the R4L project is raising a lot of questions about how interfaces are handled by some Linux maintainers, and there’s definitely a difference in criteria when it comes to soundness bugs.
I fall on the side that just because nobody has stepped on it on years isn’t a good excuse to keep landmines in the code, but clearly a lot of people in the kernel think differently.
It will likely work for a while, but it’s a fundamentally wrong approach and you’re going to be exposed into recording & decryption attacks, instead of breaking your encryption today, I just store all your communications and wait for the next qc to be online, then fish for stuff that is still useful.
It’s a silly approach if the timeframe is 50 years because most secret information goes stale quicker, but if you’re just waiting for say a year…
If and when anything quantum is able to yield results (I wouldn’t worry much about this), increasing key size is pretty much meaningless, you need to move to other encryption schemes (there’s lots of options already).
You keep making a point that no one is bringing up. Not only I don’t believe that, as a non-US citizen living in Europe I don’t even care about your two parties.
But there’s quite a stretch between “humans are biased” and “what you think about how society should work influences absolutely everything you do”.
As a silly example, I have quite a collection of music. I keep physical copies of the ones I think are more important to me. If a future archeologist were to find them, they’d get a biased picture of what people listen.
But my taste in music has nothing to do with my beliefs about politics. There’s artists there with every type of conviction, from different eras, from different countries,..
“Everything is political” is a great slogan for activists to get people to do what you want them to do, but like all slogans it’s not actually true.
History is biased, I don’t think anyone would debate that, but bias isn’t necessarily political. There’s lots of reasons for people to think some item or story is worth preserving, and most of them aren’t about politics.
The “everything is political” concept has been a very successful propaganda campaign, particularly in the US.
I seriously doubt it.
It’s far easier and more effective (and economical) to have a bunch of jelly bean components around in stock.
You’re going to have a hard time 3D printing anything that can be solderable (either the 3D printer needs to work at high temperatures for DIY, or you need exotic solder that melts at low temperatures).
If you have the need to fabricate quick PCBs for prototyping, you’ll be better served by a cheap CNC machine and some copper foil blanks.
The only real promise I see is that you might, in the very long future, be able to print custom multi-purpose devices, that integrate the characteristics of non critical electronics with mechanical elements, i.e. integrating NTCs on cases or fan supports,..
The liability lies with the driver, and all Tesla needs to prove is that input from the driver will override any decision made by the ADAS.
There are other options, but they tend to be less budget friendly, more targeted to professionals than DIYers.
Small BGAs can be hell to hand solder.
People still build this kind of stuff, it's just less common because it's way more expensive (it's labor intensive, so it doesn't get cheaper over time).
Most companies have turned to the other way of doing this, they just offer money to the buildings, and if they consider it an eyesore, they can pay for the "beautification". Most building managers won't do it though.
The funniest thing of this whole business is that a lot of residential buildings oppose the installation arguing "health issues", which ends up with the antennas built on another building. They not only don't get the cash, but most times their actual exposure goes up.
This is completely orthogonal. If it weren’t, the lander would be in a better shape, but it’s as much of a clusterfuck as the rest of the mission.
SpaceX has never been outside of LEO, and I’m very unconvinced Starship can do it’s part on Artemis, much less do all the mission by themselves.
Side-goals, fake goals and scope creep are one of the biggest red flags for “projects to avoid”.
The network is not some amorphous blob.
If a new streaming company called Notflix enters the market (with dedicated hardware), ISPs will have to build dedicated infrastructure that connects that competitor to the network. One of the obvious ways of gaining a foothold in the market is to go where other services have poor connectivity and setup shop there. Of course the ISP that sets up that infrastructure will take big cut for their work, as they should.
With Net Neutrality in effect, the supply side of heavy use internet services becomes constrained, creating a barrier of entry for new competitors. It’s just plain old regulatory capture.
A lot of people seem very confused about what “neutrality” means, and it’s consequences. As an analogy, VAT is an equal tax (everyone pays the same VAT) but it’s a very non-progressive tax (it burdens poor people more than rich people.
Your ISP doesn’t really care about your speed, it could increase yours and all your neighbors speed by a big chunk and it won’t really notice it. The problem is that to handle a Netflix they need to do a massive investment.
Yes, non net neutrality is about creating differentiated “highways”, but you are not going on that “highway” no matter what.
The discussion is if internet is considered infrastructure (as roads are) and thus they should be built with everyones money, no matter how specific they are to a single company, or if we should leave it to the market.
I’m still waiting for someone to explain to me why a company that makes massive amounts of money from the internet shouldn’t be paying a higher proportion of the infrastructure costs than a user.
Programming is thinking in the same exact way all knowledge work is thinking:
- Design in all it’s forms is mostly thinking
- Accounting is mostly thinking
- Management in general is mostly thinking
The meaningful difference is not the thinking, it’s what are you thinking about.
Your manager needs to “debug” people-problems, so they need lots of time with people (i.e. meetings).
You are debugging computer problems, so you need lots of time with your computer.
There’s an obvious tension there and none of the extremes work, you (and your manager) need to find a way to balance both of your workloads to minimize stepping on each others toes, just like with any other coworker.
Most issues either look like someone has been working on them for quite some time (which seems rude to barge in and waste their time), or there’s no clear indication of what would be valuable to help with.