nvm0n2

47 ms·

481.
by nvm0n2 3y ago
When Chrome first launched it was actually quite rough. There are some rose tinted glasses on this thread. It was Windows only, lacked basic features like printing, had no extensions system and had no few site compatibility bugs. The primar
482.
by nvm0n2 3y ago
This sort of thing gets claimed a lot (by climatologists) but isn't true. 50 years ago is in the 1970s and there were plenty of climatologists predicting a new ice age back then. Here's an example: https://www.nytimes.c
483.
by nvm0n2 3y ago
How are uncertainty bounds communicated in complex, compounding simulation models? Poorly to not at all. Climate models have been diverging over time. As details are added (e.g. to try and handle clouds) the different teams making models
484.
by nvm0n2 3y ago
It'd be nice if that were true but in most of the world you can't get online without ID verification. Even for WiFi you will need a text message, which ties it to your phone number which ties it to your identity
485.
by nvm0n2 3y ago
Training needs insane amounts of inter node bandwidth, to the extent that training clusters use specially built hardware for it. Decentralised training isn't physically possible anytime soon, maybe never.
486.
by nvm0n2 3y ago
The problem is the ultra-liberal conflation of words with violence. The X-risk folks are mostly concerned about actual, physical violence against humanity by AI - "what if we accidentally make a paperclip maximizer" being the text
487.
by nvm0n2 3y ago
I don't think Google/Facebook execs realize to what extent they destroyed themselves when they allowed the meltdown over James Damore. He literally wrote a whole essay warning them of the dangers of allowing rampant left wing puri
488.
by nvm0n2 3y ago
There aren't meaningful gaps in software quality here, most of the "American" services and sites are half built by European workers anyway. There aren't competitors for a bunch of reasons, but the far more restrictive EU
489.
by nvm0n2 3y ago
It's the way this leads to a specific outcome that's hard to explain.
490.
by nvm0n2 3y ago
It's clearly not that extreme, given that Europeans overwhelmingly express strong preferences for American-built communications and information tech.
491.
by nvm0n2 3y ago
Delphi/VB weren't low code! They were full blown languages with IDEs.
492.
by nvm0n2 3y ago
It's really not that American. Americans didn't invent the concept of free speech, not by a long shot. The idea that it's somehow "un-European" to allow people to say things "harmful to democracy" is wrong
493.
by nvm0n2 3y ago
I'm aware of the DSA but fortunately am not required to follow it, being in one of the parts of Europe that's not a part of the EU. Section 230 doesn't have anything in it about being "actively harmful to democracy"
494.
by nvm0n2 3y ago
There's no general responsibility to not "present" things "actively harmful to democracy". And there's nothing uniquely American about that. I'm a European living in Europe and there's no such obligat
495.
by nvm0n2 3y ago
Doesn't seem like a very good theory given that Twitter's political bias was one of the driving motivators for him to acquire it in the first place. He's done some unpredictable stuff but this is entirely expected and consist
496.
by nvm0n2 3y ago
Wouldn't help in this case, the bugs are in things that Rust's type system doesn't help you with like references being held across processes that are sharing memory segments, connected to GCd JavaScript heaps. Affine types ar
497.
by nvm0n2 3y ago
Surely the problem you're objecting to is that Android is JVM native, so the Java APIs are a part of the OS itself.
498.
by nvm0n2 3y ago
We were hearing that the day he bought it and started layoffs. Twitter will be crashed within a few days, we heard. Clearly that never happened. Twitter is still online, making changes and executing. That's what executives are respondi
499.
by nvm0n2 3y ago
It's not exactly a conspiracy theory. ADL have very openly attacked Twitter, Musk says the ADL have been pressuring advertisers, and many different kinds of groups have complained about left wing pressure groups trying to cancel them b
500.
by nvm0n2 3y ago
But I'm not willing, because your willingness to make unfounded assumptions leads you to demand things like censorship and the removal of all anonymity, which would only be possible at great cost and would certainly make it much harder
501.
by nvm0n2 3y ago
You're overlooking the cases where the vaccine enters the bloodstream and itself is circulating, or gets transcribed and yields permanent spike production. Especially if combined with immune tolerance this can cause the observed effect
502.
by nvm0n2 3y ago
I agree it sounds like it should be true. But the reason BS is effective is that it often sounds true without actually being so. I can think of many examples within 10 seconds where high effort bullshit was trivially refutable, so I don&#
503.
by nvm0n2 3y ago
Maybe public policy discussions between foolish people, but not elsewhere. You really don't want to go down the route of loosening the standards around blaming people for unrelated people's acts. Not only would that simply be an
504.
by nvm0n2 3y ago
Modern COVID isn't dangerous. The vaccines can be. And the vaccines don't stop you getting COVID anyway, so, the risk is additive in any event. Your call.
505.
by nvm0n2 3y ago
> No. This is well proven. 1) it is exponentially easier to generate bullshit than to refute it If it's well proven then you can easily link to a rigorous proof of that claim, one that justifies the use of mathematical terminology l
506.
by nvm0n2 3y ago
The concept of "good faith" is a self-refuting one, as it has no real definition. It's just an sophisticated sounding ad-hominem that means "I don't think this person really believes what they say" but is invar
507.
by nvm0n2 3y ago
There are at least two ways to refute Brandolini's Law, both work. The first refutation is in response to a steelman version of the argument. It's simply that that you often can't know whether a claim is or isn't bullshi
508.
by nvm0n2 3y ago
What exactly is "stochastic terrorism"?
509.
by nvm0n2 3y ago
Foreign mRNA doesn't hurt us because it's recognized by the body as harmful and immediately destroyed by the immune system. This was one of the biggest challenges to developing mRNA based drugs and vaccines. The mRNA used in the C
510.
by nvm0n2 3y ago
It's not only the heart that's at risk. The shots have been seen to cause diabetes or cause uncontrollable ketoacidosis in people who are already diabetic. This isn't a surprise because diabetes is caused by the destruction o

More ›