jjxw

As another commenter pointed out, it also doesn't have to sift through thousands of faces. Chances are the doorbell is only exposed to a handful of faces a day which greatly reduces the volume of false positives regardless of the field of view and resolution.

Agreed. To clarify, I use the term in a narrow sense limited to what I imagine to be "reasonable" from the profit seeking point of view of Goldman.

Sounds like a reasonable operation to set up. Goldman is ostensibly in the business of selling shovels in a gold rush so if enough of their clients are demanding a product that gives then exposure to bitcoin then Goldman senses there's money to be made here in fees.

While not exactly the same, reminds me of the "Roach Motel" Dark Pattern [1] which intentionally increases friction to completing an action by users that would be detrimental to the business (such as deleting one's account). Amazon is another example of making it nearly impossible to figure out how to delete your account unless you search for the steps or go through a lot of trial and error. [2]

[1] https://darkpatterns.org/types-of-dark-pattern/roach-motel [2] https://www.youtube.com/watch?v=kxkrdLI6e6M

Had the pleasure of working with their engineering team last year as part of a pro-bono data consulting grant that a nonprofit I work on (http://www.deltanalytics.org/) provides.

Excited to see how these systems are integrated into prevention / intervention and other applications for these systems beyond detecting illegal logging.

Command and Control by Eric Schlosser covers many close call nuclear incidents that the US military has had since WWII. The incident which has been seared into my memory is one where an aircraft carrying a nuclear bomb flying over the east coast of the US had a critical engine failure which resulted in the crew bailing out. After the crew bailed out, an object in the cockpit inadvertently armed and dropped the bomb as the aircraft disintegrated. The only thing that prevented the eastern seaboard of the US (including Washington DC and New York City) from being irradiated with fallout was a failsafe switch which prevented a circuit from being completed in the bomb.

It is a wonder that we (still) haven't blown ourselves up yet.

This happens any time a public company's stock goes down because of unexpected news. Investors sue for breach of fiduciary duty in an attempt to recover losses in the share price as a result of alleged negative, material information being withheld from the public by the company's management.

That being said - this should be particularly interesting given the topic. Securities fraud cases are usually boring, but this lawsuit may bring to light more details on exactly what FB didn't disclose.

Based on other companies which tried and failed (e.g. Spoonrocket, Sprig) this seems like a risky model to follow. It's possible that the costs of managing such a kitchen plus delivery service is at this point more expensive than brick and mortar, or the margin earned on these deliveries is much less than sit down customers.

Maybe once kitchen automation technology gets there this can be a workable model.

[1] https://techcrunch.com/2017/05/26/on-demand-food-startup-spr... [2] https://techcrunch.com/2016/03/15/spoonrocket-shuts-down/

Now if they could get to repairing the solar system model at The Interval ;)

I've been stopping by at midnight for the past year or two in hopes that I'll be able to see the system in action, but alas it appears the system is so complex that repairs are costly. For those of you intrigued with The Long Now and are based in SF - definitely visit The Interval, it's one of the more interesting "themed" bars out there with books and a collection of mini exhibits dedicated to science, art, and music.

Sounds like the type of advice that some of the more brash tech CEOs (e.g. Travis K.) would have taken to heart.

He's referring to the more complicated derivative products that allowed those funds to buy risk in the form of betting on low future volatility. Since volatility has recently spiked, those types of products have lost the majority of their value.

Ah - that's an interesting perspective I had not considered. I was basing my prediction on the stronger than expected job gains, GDP growth forecasts, and how the Fed is cutting back on quantitative easing.

Keeping an eye on the US inflation report coming out this Wednesday should provide more guidance on what the Fed may do - if it's higher than expected I think the Fed will probably raise rates by another .25%.

From one armchair economist to another I suspect the Federal Reserve will have something to do with what I wouldn't call a "recession" per se, but a slowing of growth through interest rate hikes.

As much as I personally think cryptocurrencies will have a difficult time achieving currency status - this is pretty much what happened during the American Revolution. Congress issued bonds to pay soldiers and for goods for the Continental Army. Those who accepted the bonds as payment were, in a sense, betting that they would be worth something in the future.

Certainly not impossible, but one thing is a disincentive to commit a 51% attack is that if you had that much computing power a more profitable use of it would be to just mine rather than create fake transactions which, if detected, could crash the value of the currency.

Fair point :)

I guess the right comparison would be how much it would cost to produce another human capable of doing the SHA-256 hash by hand rather than the "fuel" to sustain it.

Capital costs may not be trivial if you factor in food required for brainpower.

The below probably fits your criteria. Probably the best piece of sports writing that I've read, DFW's love and knowledge of the game really comes through without being overwhelming to those who don't have much knowledge about tennis.

[0] http://www.nytimes.com/2006/08/20/sports/playmagazine/20fede...

More specifically, it allows one to short with limited downside. There are some exchanges that provide naked short positions in Bitcoin, but those have theoretic infinite downside and your broker will force you to liquidate at a loss at a certain point.

With futures, however, you can put together positions that will provide a known limit on your downside for short positions. This provides a more appealing avenue for shorts to get in with much less risk.

I wouldn't say Bitcoin is even a good example of being able to diversify from nation-specific risks. China's crackdown on the currency is a great example and if the SEC or other regulatory body decides to crack down or otherwise make interacting with these markets more difficult it would presumably have an effect on the price.

The largest risk to the long term viability for bitcoin (and probably crypto in general) are regulatory in my mind. A lot of value can be derived from regulatory arbitrage - just look at the evolution of cryptocurrency's use cases from 1) buying and selling illegal goods, 2) getting around capital controls, and now 3) issuing securities without having to go through the normal IPO process. It's a long road to people using this stuff to transact "real" goods and services. There is a very real nation-specific risk in the form of regulatory regime.

"As long as the music is playing, you’ve got to get up and dance." -Charles Prince, Ex-Citibank CEO 2007

This is one of my favorite modern finance quotes on bubbles because it encapsulates so well why there is still momentum in this crypto bubble even after an even casual reading of the news will tell you that everyone and their grandparents believes it's a bubble.

My bet is that heavy regulation in a market where there is heavy trading (e.g. USA, Japan, Korea) will be the blow that bursts this, but as with everything in public markets, who knows?

A common misconception is that while Bitcoin may be deflationary in that the rate of supply creation is decreasing, it does not in any way guarantee the value of the coin. Inflation and deflation of currencies are also demand based. If demand for bitcoin falls faster than the decreased rate in which coins are issued, the value must go down.

As with most financial assets with high liquidity, there is no such thing as "guaranteed to become more valuable". If it were, it would already be at that price.

Sounds like a classic case of overfitting training data. Bitcoin and cryptocurrency as an asset class is so nascent that I wouldn't be making any significant bets on what the algorithm outputs even if the cross validation is good.

As with all investing claims, the pertinent question to ask is "if this works so well, why are you spending time marketing your method instead of making boatloads of money using this system you developed?"

It is perplexing to me why the burden of proof was ever placed on "increasing supply decreases prices" when the mechanism by which rent control + limiting construction is supposed to benefit anyone besides those already occupying those units is totally unclear.

I don't even think it's "ought" vs "does" - it's more NIMBY interests being more ingrained into local politics allowing them to somehow gaslight everyone into thinking the laws of supply and demand don't apply to housing.

I'll enthusiastically second the recommendation. I won't ruin too much, but their policy of not allowing mobile phones while you are in the museum to prevent you from searching anything up really helps to preserve the atmosphere.

That's the great thing about a futures market. People who want to short Bitcoin with reasonable risk management can do so now. There are a lot of people who have lost their shirts betting against Bitcoin because all you could do were naked short - these derivative markets will provide sophisticated traders the ability to hedge their bets so they can continue to short without getting margin called if Bitcoin happens to double in value before crashing in the coming months / years.

That is sort of the joke about all securities litigation. Shareholders will often sue companies that experience price drops due to "misleading investors" and, as you mentioned, the fine or damages end up being paid for by the investors themselves.

There was a joke floating around somewhere (Matt Levine I think?) that one strategy to avoid a jury trial proceeding was to become so publicly infamous that it would be impossible to select an unbiased jury. Clearly didn't work out for Shkreli in this case.

There seems to be a misconception regarding what Shkreli was found guilty of. The legal case here has very little to do with the pharmaceutical pricing controversy - it is a separate case based on a separate hedge fund that he managed. The gist of it is that he took people's money to start a hedge fund, lied to investors that the fund was doing fine when the hedge fund went belly up, but ended up returning everyone's money plus a sizable return when his separate pharmaceutical venture went well.

When fraud happens those affected don't usually get their money back much less a return on that money. However, it's pretty clear what he did is also fraud (false documents, not returning people's money when they asked for it) even if the fact that investors came out better makes the plaintiffs less sympathetic.

Not to presume any motivations of Bitcoin's creator, but it doesn't hurt when the success of your research also nets you an incredible amount of wealth. Though, as far as I know the presumed wallet that he owns has not had any funds moved off of it so at least their interest is aligned with the success of Bitcoin.