The conjecture - in the article - is it doesn't work at, well at all is strong but certainly not to the degree that is claimed. Also, given the replication crisis in this field, any benefit is possibly noise.
> We conduct four randomized field experiments among 1,801 hosts on Airbnb by creating fictitious guest accounts and sending accommodation requests to them. We find that requests from guests with African American-sounding names are 19.2 percentage points less likely to be accepted than those with white-sounding names. However, a positive review posted on a guest’s page significantly reduces discrimination: When guest accounts receive a positive review, the acceptance rates of guest accounts with white-sounding and African American-sounding names are statistically indistinguishable.
I think it is more a case of being everything to everyone, and one size fits all just plain sux.
IMHO there is no Twitter problem, but there are specific Twitter niche problems. Twitter is trying to solve the problems of people who want to follow a sport with those who want to lead a culture war in the same way and with the same perspectives and expected outcomes. They are not just very different problem spaces, but fundamentally different cultures, with different expectations and rules, and making either group live with a compromise between their cultures leaves everyone unhappy.
Reddit sidesteps this to a degree, with subreddits, by everything being less personal, and with moderation of sub reddits. Twitter, being account based rather than topic based, can't manage that. Twitter is just Twitter - a singular entity used in non-singular ways and criticised as a singular entity from very specific and singular perspectives.
How do you fix that? No idea. My hope is that Machine learning can give us better filtering, because the problem with Twitter is that people are not brands, so if I follow a basketball coach, and he has a bugbear about an issue that annoys me, I either lose all of his tweets, or put up with off topic rants.
But then, thinking about it, I'm not sure if that is a bug or Twitter's key feature. Maybe Twitter only works because the poor filtering forces people to see what they don't want to, and that drives usage and, in a perverse way, a desire to use twitter. That's a dark timeline.
As opposed to wind and solar? Let alone batteries, that use extremely toxic chemicals.
What we do with waste is a key ingredient to any power source, and the plan with old batteries, solar panels and wind turbines is no better than nuclear.
There is no free lunch, only competing ideas with different sets of problems and difficulties.
Not all ads need clicking, and the big budgets have never been in generating clicks, but in Branding (like TV ads).
The 10% that click are likely spit somewhere around 50-50 between the least valuable eyeballs (suckers who click ads) and people that actually are interested, depending upon the ad type (Search more valuable, banners less so).
I really shouldn't suggest this because it is definitely a bug that I enjoy exploiting, but on YouTube, if you hit the next video link then immediately hit back on the browser, the ads all clear. As in all the yellow ad markers disappear. So no annoying mid video ads at all. My default YT viewing is click link, click next, hit delete.
This is on my Macbook and all three major browsers Safari, Chrome and Firefox.
... AAAAAAND watch as this gets fixed (I'm an idiot).
That's not the argument in full. Carbon cost of production includes the cost of wastage. If the cost in carbon to produce what is wasted is > than the savings elsewhere, it is a loss.
It is not so much a complicated equation as it is an involved one, and we risk simplifying the equation and getting it very wrong.
The cost in carbon of a thing is complicated, and this sort of X (local) is good, and Y (non-local) is bad causes a lot of the issues.
There was a debate about Dutch flowers vs Kenyan. The debate was framed as "local vs grown in sunshine", e.g. the cost of growing in cold greenhouses vs sunshine. I think you know where this is going...
A book like https://www.amazon.com/Drawdown-Comprehensive-Proposed-Rever... provides the context needed to choose between options, and the solutions are often odd, like replacing old fridges which has a HUGE climate change benefit (because the refrigerants are 1,000s of times worse than CO2), but that's not a story that is told because, well I think complicated narratives lose to simpler ones.
All this stuff is super confusing. https://www.austrade.gov.au/contact/faqs/i-want-to-export-my... says you need to register if:
> your VAT taxable turnover is more than £85,000 (the 'threshold') in a 12 month period
> you expect to go over the threshold in a single 30 day period
WTF? How the hell would anyone know that? What if you never sell to the UK, but someone goes on a run because some product randomly becomes popular. It is such an unnecessary minefield.
Depends on the tax code. In Australia, corporate taxes are passed down as a deduction to real humans. So if I get a fully franked dividend (fully franked means the company paid the 30% company tax) of $70, I also get a tax credit for $30. If my marginal rate is > 30%, I pay the difference (e.g. 50% means I owe $20) but if it is lower I get a refund - the marginal rate is 19% up to $37K, so those people would get a refund of $11.
This seems to me a sensible law, as doesn't double dip, and foreign owners don't get the credit, so the government does get a smidge more revenue, while not F$%^ing over their citizens.
Except Digital sales take place in no-man's land, and are hard to tax. A federal tax hits Amazon, but would it hit Alibaba? The specifics are what make these ideas so difficult.
> Who gets the bulk of the fruits of his labors, and to a very large extent what he does is determined by said expropriating suits
At any other ti in history, with any other technology (Tesla vs Edison) that would be true. But in tech, the billions made the companies. Page, Brin, Bezos, Gates all got the "bulk of the fruits".
There is this weird idea that work creates value in a vacuum. That's not true. Work in a context - of a company that has funding/profitability/stores of cash creates value. Work done in a vacuum does not.
We need to balance those elements carefully, because not enough workers slows growth, but so does not enough capital/misallocation of capital (see housing).
The ideal world is one in which we all move in the same direction together not out of compulsion but out of an understanding that we all benefit from moving in said same direction.
1. Complexity - a change in exchange rates can hurt Netflix's / Google's / Apple's profit, even if all underlying numbers are correct. Guessing exchange rates is a terrifyingly difficult task, and it is just one of many complications 2019 CEOs have over 1969, let alone 1919.
2. Globalisation - rather ironically, if a company employs an extra 10% of people - no one loses their job they just add an extra 10% - the ratio likely gets larger. How is that a BAD thing that more people are employed? Mattel is the most telling in this context ($6,271 average worker salary). IMHO it's a GOOD thing that Mattel directly employs workers, rather than using a, say, Foxconn. But it makes the ratio a lot worse. Obfuscating real worker wages is bad for workers, but good for avoiding ending up on these sorts of reports.
3. Market size - a follow on from 2, if Google makes 50% of it's revenue outside of the USA, what should the ratio relate to? US workers to CEO? Or South African? A lot of these CEOs are multi-country CEOs, and that is a level of difficulty beyond what existed a quarter century ago.
4. Market forces - a law to make CEO pay public means it is signaling something negative when a CEO makes a low ratio, which drives it up. Having public records of salary makes negotiating easier for workers, and CEOs are no different, so it has had a double upwards pressure.
Just some things that have made it grow over time.
Video transcoding is particularly suited to the cloud. This sort if unpredictable load - new titles will pop up at odd intervals - is ideal for striking the balance between speed to market (you don't want to delay releasing new titles) and cost (having extra capacity for transcoding on the off chance you need it is expensive).
Spool up hundreds of servers when they need to get a new title out, and buy them at cheapish rates is pretty much what the cloud was built for.
If anything, that is a call for better immigration, not better utilisation of those born in a country. Reasoning:
1. The US will not be able to compete w/ India or China when it comes to population
2. Most people with high ability are born outside of the US or any country statistically.
3. Importing the best and brightest is the surest way to continue growth.
The argument for immigration is that most 1st world countries do a great job of filling the middle 80% - most jobs in other words. We struggle at the top - as the top is hard to fulfill by definition - and the bottom, as most 1st world born don't want to, say, pick fruit.
To make sure 1st world economies continue to grow maximally, we need more of the very best, and more of those willing to do the sort of menial tasks we collectively simply don't want to.
Everyone wins from that. Those born in a 1st world country get to work at better run companies, and have their menial tasks fulfilled at a rate that keeps fruit cheap, and wait times low for things like fixing a broken door / getting a house painted.
That's the problem though - blank slate theory. If we had a more biological attitude towards this, we'd be able to help individual more effectively. Do IQ tests yearly, find smart kids and get them into better programs. On the flipside, accept that some schools have naturally less intelligent kids, and that their performance will, as a result, be lower.
Blank slate and no child left behind and similar ideas create this false belief that bad results are the teachers fault. If you teach kids that all have below average IQs, and they achieve average results, you are an awesome teacher. If you teach geniuses and the get average results, you are awful.
Does it though? If Germany's policies hurt the economy, at a time when the cost of energy goes up, it may threaten the plans. The last thing a costly reform needs is less money to spend.
Ignoring possible second order effects is a dangerous course of action. The nuclear shutdown in Germany was so bad not just for emissions (which have gotten worse) but also, as the Marginal Revolution article above points out, geopolitically as well. One decision can have a multitude of flow on bad effects.
The important part to remember is that the goal is not clean energy. The goal is for humanity to continue to exist and, if possible, to prosper. That is why climate change is awful, as it threatens our species. Not the planet. The planet will be fine. It is our species that is at risk.
Decisions made for vague, second effect ignoring reasons are worrying, especially when there are known second order effect consequences.
That's cool but where are the lines? The problem is that a LOT of lobbying goes on, and we don't know how to allow what, when or why - and no country does it well.
Examples: protests - are they ok? What if they are funded by some group?
What about a concerned citizenry meeting a politician? Is that OK? What if their funding comes from Iran?
It gets real complicated real quickly.
> Politicians should craft laws based on what's best for the country
That seems so simple, but man, that is SOOO complicated. Imagine a law that made life better for 80% of people, but made 20% worse off - should it be implemented? What about 2% better off, 98% worse off, but the worst off 2% amongst us benefited?
That simple sentence fragment is why lobbying, and groups like the NRA and teachers unions, exist and hold so much sway. Many laws affect groups disproportionately, and how we decide who to hurt and why is what politics is, in part, about.
There just aren't any magical, mystical policies where everyone wins, not even things like free trade which has been great for almost everyone on the entire planet.
Is the problem then the search for a solution that works 100% of the time? I don't see any solution not having a downside, except maybe a UBI, and that's, lets be honest, just way too expensive.
I think superannuation (or 401K in US parlance) with a means tested pension backstop is the most likely to be successful, simply because it removes this sort of shenanigans, and at scale becomes a really boring industry where shonksters are less likely to exist (NOTE: LESS not ZERO) but still has a social backstop for people that just don't ever get it.
To be fair the P/E will start to move towards 12. That's just inevitable.
If the P/E moves from the 39.37 I just Googled towards 12, the profit needs to triple to maintain share value. Heck, it was over 60 to start the year: https://ycharts.com/companies/GOOG/pe_ratio, and that seems a little insane to me.
By far the better approach than removing information by having blind / limited information applications is to have MORE information, e.g. https://marginalrevolution.com/marginalrevolution/2019/02/re...
> We conduct four randomized field experiments among 1,801 hosts on Airbnb by creating fictitious guest accounts and sending accommodation requests to them. We find that requests from guests with African American-sounding names are 19.2 percentage points less likely to be accepted than those with white-sounding names. However, a positive review posted on a guest’s page significantly reduces discrimination: When guest accounts receive a positive review, the acceptance rates of guest accounts with white-sounding and African American-sounding names are statistically indistinguishable.