One thing that surprisingly hasn't happened in the 30 years since the Human Genome Project is a Nobel Prize for the work. There are examples from large physics projects where the leaders were awarded the prize. Surely Francis Collins and perhaps Craig Venter are deserving.
Lewis and Rao published a meta-analysis of 25 large scale controlled advertising experiment [1]. Here's the abstract:
Twenty-five large field experiments with major U.S. retailers and brokerages, most reaching millions of customers and collectively representing $2.8 million in digital advertising expenditure, reveal that measuring the returns to advertising is difficult. The median confidence interval on return on investment is over 100 percentage points wide. Detailed sales data show that relative to the per capita cost of the advertising, individual-level sales are very volatile; a coefficient of variation of 10 is common. Hence, informative advertising experiments can easily require more than 10 million person-weeks, making experiments costly and potentially infeasible for many firms. Despite these unfavorable economics, randomized control trials represent progress by injecting new, unbiased information into the market. The inference challenges revealed in the field experiments also show that selection bias, due to the targeted nature of advertising, is a crippling concern for widely employed observational methods.
One of the biggest problems in science (in general) is that your reputation and therefore funding is highly tied to a particular subfield. Everyone in a subfield knows each other and gives positive grant reviews and external tenure letters to other practitioners. If you try to move on to new areas of research, your reputation and funding are gone. The outcome is that areas of research interest become quasi-permanent, even when they are no longer interesting. I talk to the people today that still work in my specialty of nuclear physics (relativistic heavy ion physics) and I'm shocked at how little intellectual progress has been made. At the risk of offending people it's time to give up and declare victory, but the consequences would be dire (hundreds of professors all over the world without grants).
The best land for high density construction in SF are the smaller rent controlled buildings, not single family homes. These smaller rent controlled buildings are near lively neighborhoods and transit, and are also often at the outer edge of their useful lifespan. I have yet to see a policy proposal to open up this land for development because nobody wants to give up rent controlled units.
Part of the reason white collar firms outsource is that benefits usually need to be uniform across all employees. If you are a white collar firm with generous benefit packages, it becomes comparatively expensive to hire blue collar workers yourself compared to outsourcing. Decent health insurance (~$20K/year) is more than the federal minimum wage ($14.5K/year). I'm not suggesting this is moral, but it is reality. At a factory, with many workers at similar pay levels, this won't be as much of an expense.
Unfortunately, a second reason you see outsourcing are job functions where many of the workers in a geographic area are undocumented. Companies can't hire undocumented workers directly, but will allow a middleman to do so on their behalf.
The subtitle of the New York Times article is misleading if not flat out wrong: "The suit focuses on the advertisements that generate a vast majority of the company’s profits." The DoubleClick ad exchange most certainly does not generate the vast majority of the the company's profits. Search ads are completely independent of this lawsuit.
With both Google and Facebook, mandatory licensing of their advertising inventory (within reason) would help competitors grow. Advertisers want to buy on Facebook/Google because they are the only two ad platforms with quality and scale. Companies like Pinterest/Twitter/Snapchat don't have the enough scale to be interesting to most advertisers. Programatic marketplaces (Adwords, etc.) have the scale, but not the quality (too many spammy publishers and lack of identity resolution across the marketplace). If the big two digital ad platforms were required to license their premium inventory, other companies could invest in their products and gain substantial incremental revenues as consumers started using them. With current business practice, you need to reach massive scale before you can even think about monetizing at reasonable rate ($30+ CPM).
I was estimating a few hundred million streams per year (including All Along the Watchtower by Hendrix). The $4 million/billion streams gets split a few ways and the songwriter only gets a portion of that. Radio was more lucrative for songwriters since the songwriters got the bulk of the revenue (no payments to performers).
The only way to justify these prices is movie/TV/advertising licensing which is negotiated for each specific instance. Streaming brings in a fraction of cent per play so his catalog can't be generating more than a million dollars per year at best. Using his music to sell prescription drugs to nostalgic baby boomers watching Fox News is likely the plan.
Before the 1980s long distance calls from AT&T landlines were very expensive. The FCC forced AT&T to allow competitors to use their infrastructure at a reasonable price. MCI and Sprint entered and the cost of long distance calls fell pretty quickly. Sadly, this may partially explain our terrible and expensive cell phone service. In many countries, the first competition to the national phone carrier were mobile services so the early mobile entrants were competing against massively inflated prices for long distance calling and quickly built high quality alternatives.
My proposal is that we treat liability in the self driving car space similar to the way we have a specialized liability framework for vaccines [1]. Similar to vaccines, self driving cars would lead to reduced deaths on aggregate, but there are cases where the technology fails and people are injured/killed. In the case of vaccines, consumers cannot sue the manufacturers using regular courts but instead there is a separate system where the standard for liability is much higher.
Some economists at Vanderbilt used the introduction of Boro cabs (green cabs) to estimate the effect of an additional for hire vehicle on traffic congestion. They claim that most of the increase in Midtown Manhattan traffic congestion is due to Lyft/Uber[1].
The media reporting as well as Google's response is entirely focussed on the consumer monopoly on search, but the DOJ complaint focusses equally on potential anti-competitive practices in the search advertising market. The DOJ describes the situation well here:
110. There are barriers to entry in these advertising markets that protect Google’s
advertising monopolies. Most critically, search advertising of any kind requires a search engine
with sufficient scale to make advertising an efficient proposition for businesses. Specialized
search engines require significant investment, including the cost of populating and indexing
relevant data, distribution, developing and maintaining a search algorithm, and attracting users.
Search advertising of any kind also requires (1) a user interface through which advertisers can
buy ads, (2) software to facilitate the sales process, and (3) a sales and technical support staff.
The same barriers to entry that apply to general search services also protect Google’s general
search text advertising monopoly.
Declaring search advertising an illegal monopoly would not only open up the possibility for structural changes (requiring them to license the search advertising to other search engines, giving aggregators the ability to buy search advertising on google.com) but also exposes them potentially to large payments to advertisers who have been harmed in the past.
Has this reduced search market share for Google in Europe? I suspect not since Google has a high brand loyalty. In fact this complaint may make Google even more profitable by not allowing phone manufacturers to charge the for default placement. I believe it will likely kill the main revenue stream for Mozilla as well.
I had a real estate agent tell us "that house is uninsurable due to the roof." It was a wooden roof in a high fire zone. A good agent can be useful.
The state of California can't force insurance companies to write new policies or to not cancel existing policies. To make matters worse, insurance companies can't generally raise rates on existing policies so their only recourse is cancellation.
If all else fails there is a high risk FAIR plan as insurer of last resort, but it is much more expensive and covers nothing but fire.
Stepping outside the US, there is evidence that misinformation spread of Facebook (including WhatsApp) was directly responsible for the Rohingya genocide and forced migration [1]. At what point do these platforms have an obligation to not allow themselves be used for such campaigns?
Rent surveys are usually heavily biased towards large buildings. SF has a ton of new development in currently moribund neighborhoods, so I suspect some of the massive drop is due to that being oversampled (Mission Bay, for instance). In the more desirable neighborhoods with lots of rent controlled places the price drops likely aren't as steep. My friend is trying to rent a condo in a desirable location and most people looking at it are moving within San Francisco, trying to get into the more established neighborhoods at a better price.
The first wave of lockdowns had a clear goal: "flatten the curve" to prevent hospital capacity shortages, thereby preventing extra deaths that would be entirely attributable to resource shortages. In terms of meeting this goal, lockdown 1.0 was successful.
Isn't the long wait due to the arbitrary cap of something like 50K GC/country/year. Seems like Congress should just fix that as part of comprehensive immigration reform.
They seem to ignore this bylaw freely ... "shall be annually distributed in the form of prizes to those who, during the preceding year, shall have conferred the greatest benefit to humankind."
Burlingame, one of the wealthiest cities in the Bay Area, has a decent mix of multifamily and single family housing. The last time I checked, it was about 50/50 renters owners. Given the exorbitant prices to buy a house there, the rents are somewhat reasonable (quite a bit cheaper than SF, starting at about $1800 for a one bedroom). In spite of this, it still is very racially homogenous (very few African Americans or Latinos). Additionally, the schools tend to be mostly homeowner's children -- very few families rent there in order to give their children access into the top ranked public schools. Based on the real world example of Burlingame, it's not clear that changing the zoning rules will lead to the outcomes that the study authors hope for.
Stephen Hawking did not have an important theory with an experimentally testable prediction, similar to the current crop of brilliant string theorists (Ed Whitten, for instance). Alan Guth should have the Nobel Prize, in my opinion, for his theory of inflation. It is widely accepted but does not have a specific experimentally verified prediction that can rule out all other scenarios. The Nobel committee does not want to be proven wrong later.
I don't think that changing driers to employee status will necessarily give them health benefits. I'm not an expert, but while ERISA generally requires uniform benefit options for all employees in a similar class, the companies could chose to not schedule the drivers as full time employees, use contract labor services, raise the employee contribution to unaffordable level, etc. Many companies with large numbers of employees have large fractions of their employees without health insurance (Walmart, for instance).
The problem with any discussion of "ranking" students is that it is impossible to define the ranking system. Is the goal to produce the highest GPAs at Yale? To maximize future donations? To produce the greatest number of Supreme Court justices? I had a friend in college who was recruited because the marching band needed a baton twirler. To make matters worse, some programs (such as computer science) are over impacted and it may be the case that the "top" students all want to be in the same programs.
There is a frenzy of people buying houses in Tahoe right now so they can work from there, but many will be disappointed to find that the internet service available will not allow them to work successfully. We have a cabin in a neighborhood that ostensibly has decent DSL, but AT&T will not allow new signups and the the bandwidth drops to 500 kbps between noon and 3PM on weekdays since the Pandemic started. Oddly it is fine in the evening (when my internet often stops working in the Bay Area). I'm wondering if internet quality should be part of the real estate disclosure process.
This thread is completely missing the draw of TikTok, at least as it pertains to my teenagers. TikTok makes it fun and compelling to create content. My teenagers will actually go places with their friends to "make TikToks." The reason it is compelling, and this is not true of any other current social network, is that good content will get eyeballs -- the algorithm seems to be "fair" in terms of playing your content to enough people to see if it is any good, at which point it can get widely distributed. Most social networks start out this way, but eventually good user generated content gets drowned out by influencers and commercial interests. It remains to be seen if TikTok can stay this way.
I highly recommend The Song Machine (Inside the Hit Factory) by John Seabrook. The author describes the modern reincarnation of the Brill Building that pumps out pop hits using methods that we would not call songwriting (Max Martin, Dr. Luke, etc.). The goal is to get you to listen to 30 seconds of a song (that counts as a spin on digital streaming services). They've decomposed the process into mass production of backing tracks (using pro tools) which then get handed to so-called topliners that add the "hook" (the catchy part of the song). They stick the famous artist at the beginning of the song to get you past the 30 second threshold.
The cost of sending a letter to any address in the US is the same. The cost of delivering that letter varies wildly. Think New York City vs. rural Alaska only reachable by plane. If USPS did not have monopoly, they would be severely undercut in profitable areas and would only service hard-to-reach areas where prices would spiral out of control. So if you want to keep the price of a stamp the same nationwide you need to grant a monopoly. It's the same reason regulated utilities own the power lines.
My kids' private school recently sent out a $500 per student tuition credit (about $650K for the entire school). I see now that they got a multi million dollar PPP loan (which will be forgiven). Their revenues were unaffected by the epidemic (full enrollment for next year and no plans to rollback the tuition increase).