knn

Your assumption of a 100% savings rate is not quite reality though. More realistically, suppose both people save 50%. Then in 20 years, the rentier is about 2x richer than the worker. https://www.wolframalpha.com/input/?i=sum+i+from+0+to+20+(50... vs https://www.wolframalpha.com/input/?i=sum+i+from+0+to+20+(50...

This is a really thoughtful piece. I'd never heard of Buber before this

Agreed - there are a lot of aggressive homeless people downtown, and in berkeley as well. Some guy was cussing me out and trying to pick a fight with me the other week when I was just walking along.

Definitely true, SF really is unlivable on a median income and that is a continual frustration I have.

Obscene wealth next to in your face poverty is nothing new, nor particularly special about SF. I don't deny it's a big problem, but this is pretty much what every major city wrestles with. Nor is this unique to our time period. For example, it's well documented that in Paris through as early as the 1700's through now, that there was an enormous amount of capital concentration right beside extreme poverty and >40% of the population owning net nothing. 'Dystopian' is also a very dramatic language.. Where do you live in SF? If you live in downtown or soma and only spend time there, I can see what you mean. However if you go out to richmond or sunset for example, there are a lot more families and children there.

What do you mean by recent? There's been a whole society of rentiers living solely off investment income in most countries for at least the last 300 years, if not several thousand years. It's extremely well documented, especially in Britain and France since the early 1800's (see Thomas Piketty's work on wealth distribution). The more capital one owns, in general, more income is earned from capital rather than wages. This ratio increases sharply when you look at the top 1% and 0.1% of owners.

What you're saying is not exactly true. It's not necessarily competition from PE firms, it's high market value of good businesses. From the letter: "In the years ahead, we hope to move much of our excess liquidity into businesses that Berkshire will permanently own. The immediate prospects for that, however, are not good: Prices are sky-high for businesses possessing decent long-term prospects." What you're saying about competition is more true in a low growth regime (Europe pre-industrial revolution) in that too much capital kills return on capital. However growth rates are high, and will continue to be high for the foreseeable future, and return on capital in general will still be really good. Theory aside, the thesis of Berkshire is making good investments at a low price, and running businesses - of which Buffet's 'star power' is a more marginal factor in my opinion.

The BFG repo cleaner is pretty handy for the same purpose as well!

I'm picking through the author's paper and I don't buy this conclusion. I have a bone to pick with his evidence: for example, he makes a comparison between "willingness to expend time to drive to obtain an accidentally left behind unused, new-condition notebook (vs. willingness to expend driving time to obtain a new notebook at no financial cost)". The extent of the former he denotes as WTP-Retain and the latter WTP-Obtain. Even if these two values are equal in a study, that doesn't contradict the principle of loss aversion. In the case where a subject has already left behind his/her notebook, the loss has already happened. How can this measure loss aversion when the loss has already occurred? I think it's good the authors are trying to tease apart action vs inaction from loss/gain but these conclusions don't seem valid to me.

in my experience, doing deep learning is a lot harder than building simpler ml models. training times are killer, need lots of data, overfitting is a challenge, hard to interpret results, lots of things can go wrong. deep learning is the future from a mathematical standpoint (with neural nets you can essentially learn arbitrary functions in some borel space or something whereas simpler ml models are basically a special case of deep learning) but it's definitely harder.

Exactly - if you want to make choices based off what's economically the best (studying CS), that's fine. But there's so much beauty in the world that is "impractical" to reach for (like studying music, art, seeing nature, etc)

Sorry you had that bad experience (I would be pissed too), but the logic of (you having a singular bad experience therefore company will likely die) is next to baseless.

tenuous.

It's a great opportunity for those automating. I'm totally for automation as it creates wealth and I think is good for humanity in the long run, but people suffer from automation (losing jobs, homes, provision for their families) in the short term. People suffering from automation is a major reason why Trump was elected. It's highly insensitive to not be aware of the pain that is caused by automation by technologists like us.

Biking!

That's very interesting. In competitive SSBM (smash bros melee), movement is one of the integral parts of the game that give it its depth. Top player's movements can differ greatly from each other.

I will avoid buying 'smart' appliances for probably forever. Besides having mostly useless features, I'd rather not have another electronic probe in my life relaying all of the details of my life to everyone.

I agree that more technical projects require more cleverness/training, but most things don't require genius. Terry Tao, field medalist, argues strongly that one does not need to be a genius to be a professional mathematician. https://terrytao.wordpress.com/career-advice/does-one-have-t... Likewise, this implies that one does not need to be a genius to program difficult things.

Same thoughts as mine. Seems like ML was just thrown in the title to get more hits.

Interestingly, I go to a big state school that is somewhat a playground for the international Chinese elite. It's very common to see chinese international students driving $80,000+ cars and many of those same students cheating in classes. In a grad stat/software engineering class I took, most of them shared the same code for the assignments (in front of the professor). I hate to generalize because some of them are honest, but cheating is a terrible part of the subculture

The news feed eradicator has been a very good solution for me. I can still message, participate in groups, but that damned newsfeed is blocked! The newsfeed is engineered to suck up all of our precious cognitive space and fill it with mostly junk.

AI by Russell and Norvig. Machine learning by Murphy, Elements of Statistical Learning by Hastie et al. Just a few good ones out of many!

trained to a particular voice ~= trained to a user's preferences!

Many people would agree with you that race doesn't say that much about a person. However, it's quite naive to argue that racial issues are not a part of every day life. We live in a world built on racial divides and conflicts. It's obviously venerable to strive towards a non-racist society, but to do so one needs to understand the issues at hand.

Good on you! Squats are awesome. But don't forget to do prehab work or neglect your hamstrings! Tendon/muscle tightness or muscle imbalance can lead to some knee pain when squatting frequently. Stretching/foam rolling should keep you loose, and deadlifts (or glute ham raise or whatever else) will keep your hammies strong.

Sure, squats strengthen your core, but the core isn't a set of magical muscles that automatically make you faster. Strong quads, hams, and glutes are more important. Squats train your legs/posterior chain/core (and upper body to an extent), and also teach the triple extension [0] which is key in explosiveness. Type I and Type II muscle fibers matter as well

[0] http://myweb.facstaff.wwu.edu/chalmers/PDFs/Triple%20Extensi...

Not hacker news, but amazing story

I'm totally fine with FB charging businesses for advertisement. It's a free market, they can choose not to advertise there. Not a terrible exchange imo. However, Yelp extorting businesses to pay them to remove bad ratings/give good ratings is pretty shitty.

It would be pretty cool to order in the chat, 'deliver me 10 burgers from this restaurant using doordash at 6pm'. I would rather not click through crappy websites/enter my cc every time.

However every time we do do that, our habits and conversations get written in stone (in multiple data sets being passed around and bought/sold everywhere).

Yep, horrid title.

Have mixed feelings about this like I'm sure many do. Greater convenience, but less and less privacy.. Our Fb/goog/nsa overlords know what we eat, where we shit, all of our conversations and relationships. What a scary world we live in.

The databricks platform should solve exactly your problem - reusable data pipelining/transformation. I saw a demo of it last night and it was extremely slick. Their product is amazing, it makes data pipelining incredibly easy compared to setting up a hadoop cluster and running hive/etc. (I don't work for them - but if any databricks employee sees this, please hire me!) It runs on a spark cluster over AWS, which is much more modern and powerful than SAS/excel/sql. Since you know how to program already, it shouldn't be too hard to pick up spark (even has python bindings)