Falling back on a RL pipeline to cover gaps always strikes me as a more sophisticated version of the mechanical turk. If what we had was truly AGI wouldn't they be able to derive this from the data they already have.
This is a great analogy, and one I'm surprised more people haven't raised. Instead all your hear are comments about mathematicians being sour losers, they should have know the model terms of service etc...
Sometimes reading these articles I can't help but think we've pushed ourselves into a local but sup-optimal maximum technologically speaking. i.e. If we didn't follow the path of trying to stick everything into an iPhone/Android (along with moving everything into the cloud), we'd have a more interesting technological landscape.
I have no idea why these sorts of posts are popular. Past college you're not going to learn physics by trying to self-study an entire university course. The best way to learn is just to pick a small part of physics you'd like to learn (preferably related to your job), i.e. how GPS work or some fluid mechanics etc... Then learn the physics you need for that. Knowledge accumulation can't be organized in a straight line, it happens non-linearly and generally builds upon small wins that are useful for you.
That's not what op is arguing. To use your example, coming up with singular examples of continuous non-differentiable functions is an example of "ugly" mathematics, whereas putting them into a nice framework where they can be analyzed as a whole (i.e. functional analysis, density of such functions, etc...) is an example "elegant and insightful" mathematics. The same with the monster group, on its own maybe nothing special, but then you have the connections with other branches of math. Tao seems so focused on the individual problems and not their connections/generalizations.
Education yes, research unfortunately no. I'm not saying research outside of academia is not possible, I'm just saying it's not taken seriously and this needs to change. We really do need to go back to the 19th century model of the researcher gentleman.
Great list, and most of those don't involve big tech. I think what your list illustrates is that progress is being made, but it requires deep domain expertise.
I'm not disagreeing with the fact that there's a shit ton of founders out there looking for a quick pay day (I'd guess the majority fall into that category). Just pointing out there are exceptions, and the exceptions can be quite successful.
Fixed typo! Yeah, alone time is inherently a male thing, and is probably what drove a lot of the historical breakthroughs in the past. This need to always be socializing is a very feminine construct and probably does more damage in the long run than people realize (i.e. to deep breakthroughs, case in point how science has now become a popularity game rather than a search for the truth). A more nuanced article would have delved deeper into the types of alone behavior that are beneficial vs those that are destructive.
My criticism with this article is that the author seems to lump all time alone behavior into the 'bad' category. What about the guys spending time alone to work on their guitar solo, dive deeper into a branch of AI math, or how about spending hours reading plato?
Honestly all these cute websites give people a false sense that they're actually learning something. The only way to learn this stuff is get one of the million good LA books out there and work through the problems. But that's hard, so people look for shortcuts.
I never understood the obsession with Bruce Lee as a fighter (not considering his acting/stunt scenes here which deserve to be judged on their own merit), it seems any half decent Judoka or amateur boxer would have probably beaten him in a fight.
Nothing really to add, but the NSA museum outside of DC is really cool. I think this is a good example of a museum that works well for adults/kids alike.
Yep, it's insane that SWEs working on products like pinterest earn more than engineers working on major infrastructure projects. Honestly without direct government intervention I think market forces will always lead to this outcome.
I think a lot of people are missing the point. Most likely what's happening here is that the issuers (not Visa/MC) see a large number of chargebacks/fraud for adult content sites and have determined that it's much easier if they don't accept transactions from these sites.