The memory prices put computers into the same category as cars. Expensive but still realistic.
It's these 200kUSD+ nodes with a dozen or more data center class GPUs that are killing the self-hosting dream. The cheapest nodes are is in the same price category as literal houses.
Either we get some competition in this space, or computing will go back to their roots as ivory tower big iron.
HBF will probably save the small inference nodes. (Eventually, the initial price will obviously be in the stratosphere.)
Long term, HBF shouldn't be more than 4x as expensive as commodity flash, and Kimi K3 is an interesting target for a system using it given how aggressively it compresses the KV-cache. An inference box with ~1.5TB of HBF and ~48GB of DRAM should be able to be built for less than a couple of grand, and get something near to 100tok/s on full Kimi K3 for a single token stream.
I will say something that I made peace with: these are part of the new normal. This one is even light in front of everything else. Look at the supposed most richest person on Earth comments. And it will get worse. I'm seeing political people public calling the ending of the suffrage for women. People sharing memes about Jewish people with the same underlying message as dark web board messages of 2010s. And I'm not talking about a specific group. This is going all over the political spectrum.
Part of me thinks that there will be a time where progressives will have to choose their radicals. I hope that it happens after my death.
The world’s values have absolutely gone to shit. There may be many contributing factors, but I will say it’s at least in part the fault of the far left pushing too far.
Instead of maintaining the fight with actual racism, thinking that’s gone away, they picked racist fights with everyone else, embracing a form of racism themselves.
They have also sided with misogynistic and racist groups in their extreme search for “tolerance”.
A fresh young mind that comes along can see racism and misogyny promoted on both sides of the spectrum, with the left one a bit more oppressive and hypocritical about it and the right side a bit more self-consistent and hedonistic about it.
Maybe OP have some different objective instead of being coherent and understood here (I hope that is the case). That said, it really looks like a manic episode
> There is no win condition for citizens, either way they are pawns to a larger game than them.
That's sadly correct. If you are not very rich or very influential, your position in these political things don't matter. Whenever I'm watching medieval shows where the plebeians or common folks are just background things in the story, I cannot ignore how similar our current societies are. However, I'm still learning how to avoid being anxious by these things.
I understand exactly what you are talking about because I had both realizations myself: thinking that I have no time because of work and chores, and realizing that the phone is addictive because it gives an exit for my anxiety. I'm still trying to solve both problems, though.
I'm probably misunderstanding something, but if the overall population risk for myocarditis is bigger than 4/100,000, wouldn't that indicate that mRNA vaccinated people have a lower risk? The number I found for the global population is 10~20/100,000. If taking the vaccine increases the risk, wouldn't the risk be bigger than 10~20/100k?
If the risk went down the term for that is "inverse correlation." In that case it would be preventative, rather than causative. Here we are seeing a correlation and we have a known mechanism of action that's driving that (described in the Stanford study).
Is that making more sense?
In part though, you are right to be noticing that the Pfizer study and the Stanford study disagree on the risk profile.
Why might that be?
Well, for one thing, if you look up the peer reviewers for the Pfizer study, it's the same list of names as the ones that performed the research, and each one of them is a Pfizer employee. Unsurprisingly, the lower risk profile was found in a study conducted by the same drug company that stands to profit from a lower risk profile. And it's not peer reviewed. Typically we rely on peer reviewed research in the world of science and medicine. Without it, it becomes very difficult to differentiate a science article from advertising, and pseudoscience.
Conversely the Stanford study was done by various researchers from various schools with no conflict of interest and no apparent profit from Pfizer, as best I can tell. And as we'd expect, they found not only a causal link, not only a higher incidence rate, but also the mechanism of action that drives it.
See my other comment answering this. It's only higher if you don't trust peer reviewed research, and instead use the numbers provided by the drug company in a study that was not peer reviewed. See the list of peer reviewers in Pfizer's study and notice that they are the same as the researchers, and that they've made claims without providing their results (those are still being released over an 80 year period by FOIA court order, you'll get the results when we're all dead).
If you can't find a link to the court case on this, let me know. I'll find it for you.
It's so sad to see that you are the first one with any kind of source in your comment. The rest are only saying scary stuff and you are supposed to argue with that. They are saying stuff like "my friend died because of the jab" and you come with studies links, which can sound unfriendly. And we are on HN, for fuck's sake.
The articles are fantastic, it's the unscientific claim that the data implies safety that's at issue. You can't post research about the fact that people misinterpret research right? But it's factually true. Most of the people on HN are not trained in doing or reviewing research. And not just on HN. Everyone can purchase an ad that claims to be the arbiter. Everyone can say they represent the consensus.
But the people with more money can buy more ads just like Google can stomp out any competitor, because they control the data channel.
It's a logical problem. Should we start requiring a license to hold an opinion about research? Who would we trust to govern that licensing process?
It's not really so simple as finding a source to back your claim. You need to be able to defend your interpretation of that source.
Bottom line "safety" is subjective. That's the critical argument. Compared to what is it safe?
Vaccine with a guarantee of infection, maybe the vaccine is safer.
Vaccinating everyone? Well, we didn't really study that. How do you study people in larger numbers with a guarantee that they don't get infected?
They didn't. They just assume everyone is infected and that number makes the research look valid and safe.
In fact, it's not.
The average person is more likely to have negative outcomes from vaccine than natural infection combined with non-infection.
So is it safe for someone who won't be infected at all? What about for someone that won't exhibit symptoms?
No. It's far less safe for those people.
300x-30,000x less safe. Depends on your estimation of asymptomatic and uninfected subjects in the real world.
It's more safe for someone in their 70s, with cancer, or whatever. Fine. But say that clearly. Don't try to bury that in summaries that obfuscate what you really studied.
Doing so destroys trust in science.
Not advocating for or against this vaccine platform, though. I'm advocating for laws about what meets and doesn't meet scientific rigor, as are most scientists right now.
Why are you not living there anymore? I understand the trade-offs, but I hate the feeling of constant surveillance. And knowing that I'm a minority in this subject, I'm aware that I'm probably going to be at the loser side in this battle. However, I'm still sad thinking that I could be a criminal in the future by being naked at my own house.
For the first time, I see that being a problem to right-wing parties, specially in USA. Now you have neonazis gathering a community by saying you are not extreme enough, and harassing the Jewish people of your side. It's crazy when you compare that to 10 years ago, but it is what it is.
Just a heads up, but Immich mobile app deleted my local copies when I activated the backup, for some unknown reason. I've checked my settings and there is an option to free up space on backup, but I didn't activate it. It's not something catastrophic, but some messaging apps will not see your pictures and will consider them corrupted
reply