nateberkopec

Drive-by commenters reacting to the title of this post with their personal opinions on charity would do well to learn more about the effective altruism movement's focus on evidence-based giving. The word "effective" in the title is doing a lot of work, and represents a huge amount of research and fact-checking done by effective altruism organizations.

Example: here's a massive report on global warming charity Cool Earth by Giving What We Can: https://www.givingwhatwecan.org/report/cool-earth/

GiveWell's cost-effectiveness spreadsheet is a sight to behold as well: https://www.givewell.org/how-we-work/our-criteria/cost-effec...

I started donating 10% of my after-tax income about a year ago.

8chan, you mean. Jim Watkins.

Margaret Hamilton, of Apollo fame, designed and still advocates something called the Universal Systems Language: https://en.wikipedia.org/wiki/Universal_Systems_Language

Community lending in Africa (and the African diaspora) has a lot of history: https://thegrio.com/2011/05/20/sou-sou-black-immigrants-brin...

Given how susu circles work, the article seems to just be describing a high tech version of a thousands year old practice.

NPR and PBS both receive the majority of their funding from private sources, so I don't think its a valid comparison.

Why was this submitted? The newest WhatsApp CVE is more than 6 months old.

> Frontend should never need 5 seconds

I just pulled these numbers off of CNN.com using webpagestest.org, which is showing about a 500ms TTFB and a 5 second time to first contentful paint. I'm a web perf consultant and a 10:1 relationship between frontend and backend time is pretty typical IME.

To me, the only point on this list that hasn't improved much is "Ruby on Rails needs to more or less work in ANY environment".

Ruby's install/user story for Windows is still Not Great, and Ruby version wrangling on any operating system is a question that you still get 100 answers to ("use asdf/rvm/rbenv/chruby blah blah blah").

For the rest:

1) It's certainly Fast Enough now. More important webperf issues are happening on the frontend side nowadays. Who cares if your backend takes 500ms to respond when it takes 5 _seconds_ to compile the JS on the client.

3) Solved since Rails 4.0 IMO.

4) Discourse has more or less proven this can be done, with some hard work. Also, Heroku more or less standardized the 512MB VPS size, and nowadays 1GB is pretty cheap and becoming more common, so Rails doesn't struggle with memory limits anymore.

It's been really interesting to watch this pandemic from Japan, where I've been for the last ~3 months. I am an American citizen.

Walking around Tokyo this week I'm still seeing mask rates of >95%. Lockdown ended about 2 weeks ago.

In America, mask-wearing carries a different set of social signals. It seems that not wearing a mask is perceived, by some, as a signal of one's virility and freedom.

The article makes this all pretty clear, but the comments here are getting bogged down in "but the data says masks are/aren't effective". The point is that the data doesn't matter. Mask-wearing is determined by social/political expectations and signaling, not by data.

If a study came out tomorrow that said mask-wearing reduces transmission by 95%, Trump still wouldn't wear one. Looking for a silver bullet study to prove a point either way is a waste of time at this juncture.

Apparently, asking Americans to show that they care about the health of another American via a very low-impact signal (the $10 it costs to get/make a mask, the minor inconvenience of wearing one) is too much.

How can a society solve any size of a problem when it cannot be bothered to put a piece of paper or cloth on their face for the possible benefit of another? If your political climate makes such a minor gesture controversial, how will you solve any meaningful issue where your factions disagree?

How do you get Americans to care about the health and welfare of other Americans?

> The keyboardist for Casiopea also composed all of the station jingles for the subway in Japan

This is amazing! I had no idea this was the same person. I love Casiopea.

IME, Japanese elders are polite to a fault, so while I get pretty intimidated going to a new cafe (kissaten), I'm always treated very well. I once wandered into a yakitori joint and an english-speaking patron made my whole order for me when they realized I couldn't read the menu.

Golden Gai is another story, and I've heard about places in the deep countryside being a little more hostile, but those are exceptions.

I don't understand the claims re: font loading.

> Vercel's impossibly fast CDN delivers new.css and the font Inter using xz/fonts, so there's virtually no bloat added to your pages.

It's 300kb of font downloads, hardly light. And the xz/fonts page says:

>Break your users from unethical tracking networks

I'm still making a request to a 3rd party CDN with an origin header that says what page I'm looking at, so I'm not sure what privacy I'm gaining here.

It's mirroring small amounts of production traffic at Shopify.

Does I/O release the GIL in Python?

That's not what the efficient markets theory is. It means that the current price reflects all currently available public information, not that the price magically predicts the future and it can never change. Random events like what you described are completely different from the halvening, which is predictable years in advance almost down to the day.

Yeah I just picked a number out of thin air. I suppose it depends on your discount rate + some arbitrary threshold of when you want to say the asset has returned most of it's useful cash flow.

For example, a cash flow with a 10% discount rate will return about 90% of all it's discounted future cash flow by year 20.

The discount rate depends on the investor, and as for when you stop counting cash flow on the way to infinity, I guess that's up to you.

To believe this is true, you must also believe that efficient markets do not really exist.

Imagine if a publicly traded company announced that in 1 years time, they would buy back half of their outstanding shares (not a great analogy but its the best I can think of). What would happen to the stock price?

All securities prices in publicly traded markets are essentially priced as discounted cash flows over the next 20-30 years. The current price reflects all available public information about those cash flows.

I'm not saying efficient markets is 100% true all of the time, of course the world demonstrates that it isn't, but the level of disbelief you have to have in the hypothesis to believe that a well-known public event affecting Bitcoin will result in a 3x price increase is lunacy, IMO.

What passes for "technical analysis" is rather insane, and isn't limited to the crypto community. It seems to grow around assets that are easily purchasable by the lay public, have lots of random price noise, and unclear fundamentals. Foreign exchange is another good example.

Oh nice. Ghost Fleet looks like the 2015 version of Red Storm Rising.

AirBnb was doing about $4 billion in revenue per year before the crash with 7,500 employees. That's about $500k revenue per employee, which is on the high end of tech industry standards. How is that bloated?

Neat!

Watching the games, it looks like the effectiveness of dribbling is unrealistically high. Most scoring plays consist of dribbling around 1 player, then dribbling straight at that back line and hoping to win the take-on.

It's more or less epidemiologically impossible that you will see an attack rate of 100% (as you just assumed). Once attack rate reaches >40~60%, the reproduction number drops below 1 because there just aren't enough uninfected hosts.

The first sentence is also the thesis of Adam Curtis' documentary The Century of the Self.

OODA was developed for fluid environments, where strategy and tactics must constantly adapt to an uncertain world whose understanding is constantly being updated.

Video games don't work well with this model because in most video games a) information is perfect and b) results are deterministic. You don't have to update your understanding of the world constantly because your knowledge of the world is more or less complete.

In John Boyd's world, you couldn't "pre-compile" any decisions because he was working in the real world, not a game.

"The American Hospital Association said the webinar reflects the views of the experts who spoke on it, not its own."

You're quoting the opinion of one person: Dr James Lawler, a professor at the University of Nebraska Medical Centre. That person researches infectious diseases, so, their opinion isn't a complete guess of course, but we should ascribe the numbers to the right source.

I find this "baroque capitalist" style is even more present in certain cultures outside the West, particularly those that are just becoming "rich" in this generation.

Walking around the Dubai airport terminal, for example, is an assault of this "maximalist", silver/gold style. Modern Chinese buildings sometimes feel the same way. It's like the whole world is turning into a Las Vegas casino.

I love ternary operators, but nesting them is a recipe for turning your code into an unreadable mess.

Rails is a dependency: https://github.com/tongueroo/jets/blob/master/jets.gemspec#L...

> I don't understand this, I'm not exactly in the market for buying $1 billion military airplanes and missiles.

Where do you live? You're probably a voter in an area that Boeing needs a helping hand in Congress.

Looks like you didn't read the article.