It's not just literacy, although that's nearly required to engage in the public discourse. It's really more like indoctrination. The voting populace has to have an implicit belief in public institutions, believe that attempting to vote or losing a vote is not cataclysmic (not a great reason for violence or retribution), and have patience that the system will gradually correct with future votes rather than require a authoritarian to restore order. I think you can also add a distaste for cults of personalities, and a willingness to vote in disagreement with religious leaders. Lastly, voters have to have a shared delusion that their vote matters, which it practically does not (economic value of voting is negative for the individual).
Russia likely doesn't meet these requirements, and the U.S. has had many failed democratic experiments in places like Afghanistan where this culture is missing.
If someone is interested in low level tensor implementation details they could benefit from a course/book “let’s build numpy in C”. No need to complicate DL library design discussion with that stuff.
It sounds like a hero story – it's not, it's more an existential nightmare and funny story? – but I kind of accidentally came to start reading all kinds of papers. Then fiancée was diagnosed with a severe condition. And just by having read stuff I found myself needing to interject doctors during her treatment, quite pointedly, to avoid risk of harm to her and unborn child – with my view being confirmed every single time by another doctor's second opinion.
It's mostly about reading fast enough, not actually requiring a feeling of comprehension. Skimming and going fast through lots of stuff. With extreme humility!! And then bit by bit an intuition kind of grows and you cut through the jargon and get a feeling for the core things. The mights and maybes and relationships in things. And then sort of learning to trust and not trust that intuition and have it guide your reading. It mostly shows up as doubt – an active doubt? – rather than an opaque sense of not having any feeling for things. Then that sometimes refines away from doubt into a sense of clarity towards some mechanism that's probably at play. Keeping absolutely humble towards it is suuuuuuper important, and it's always necessary to retain the perspective of oneself as limited and fallible.
It's also very hard to get this stuff into words. Seems more nebulous and "cosmic" than it is. It's just how our minds and reading comprehension work. It's about feeding the pattern detection systems with... substrate? A handle on things?
There are a few reasons why it works. "Works" as in is beneficial and useful to read, beyond just trusting doctors. (Do trust doctors!, –Jusr... help them help you. That's the thing.) One reason is that doctors do not have time to read, even if they'd very much want to. This is sort of force-multiplied?... with the personalization aspect: It is immensely valuable to read molecular biology from the personal perspective of operating and being inside a specific instance of that molecular biology machinery. The doctor's view is always more general (and is always a guardrail of safety, in part because of that). Then another reason is that there is SO MUCH actionable science out there. Just eminently safe and very, very actionable. It's so hard to get it across how it might be so, how it could possibly be, but it is. It really is.
OK, I just read the abstract and conclusion of the NAC paper posted above. But then I saw a comment from Aurornis saying it’s not that good. Not sure who I should listen to.
It depends what you're looking for! I moved there when I was in my mid 20's on a one year contract to run a team for the government's e-Residency program and ended up staying a bit longer (and have gone back every year since). Parts I enjoyed -- once you find a community it's a pretty strong one (but it took about 6 months for me to make any Estonian friends), there's a good early stage tech scene, the old town in Tallinn is beautiful and city living very approachable. I probably wouldn't move back now unless there was a very good reasons or to raise kids (it seems like a great place for that, very safe, great education system, etc.) just because it's relatively small (total pop ~1.3M) and so doesn't have as much opportunity as other places, plus my parents live in the US and I like being closer to them.
This doesn’t make sense to me. Why would you want something like this? What is it exactly that you expect from such a finetuned model that you cannot get from a frontier general purpose model?
How do you measure that?
reply