GDPR should be a fairly healthy deterrent to anyone implementing anything like this (that is, if the UK doesn't throw it in the bonfire of EU regulation after 2019).

This is a non-starter before we even get to thinking about regulations. I doubt the Uber guy is thinking about typical light aircraft (and weather capability immediately makes doing that completely impractical outside of places with 350 VFR days per year), but probably on the lines of scaled-up quadcopters that can carry people, flying around cities - no doubt based on the recent prototype seen at CES.

It works fine for small RC drones, but the thing is: quad/hexcopters (etc) basically brute force themselves into the air. They are tremendously inefficient and scale badly. To keep the rotor inertia down low enough you can use simple fixed pitch propellers (rather than complex (expensive) collective pitch rotors like a helicopter) they are going to have to be small and high RPM.

The racket will be unbearable. If you think city noise is bad right now just with road traffic, it will be ten times worse if you had even just hundreds of quadcopters big enough to lift people flying around. There will be no refuge anywhere in a city from the unbelievable and annoying racket these things will make - due to the differeing rotor rpms on just one vehicle, there will be an annoying set of beat frequencies generated by the lift systems to add to the ungodly racket from the fundamental frequency.

Today you can go into a city park, even in a big noisy city, and get away from the noise. Walk a couple of minutes into any of the big London parks and it's pretty peaceful. This will be a distant history if there are hundreds of people-carrying quadcopters (or indeed any -copter type vehicle) flying around the city. People living and working in the city just won't stand for it. To add to that, most people don't like flying to start with and only do so because it's the only practical way of crossing an ocean or going somewhere 2000 miles away. In a city, given the choice, a lot of people would rather take a metro system than board some kind of absurdly noisy automated flying machine.

Well that's too bad. In the winter, it's dark when I go to work, and dark when I get home, and since I have no windows - if it's a busy time I could quite easily go from December to Feburary without seeing proper daylight.

The Nissan Leaf is doing quite well where I live - I've seen many Leaves driving around the place, the person I know who has one is pretty enthusiastic about it.

The Federal Reserve does not decide how much a dollar is worth. There isn't some committee collectively raising their pinkies and saying "Today, two Cox's apples will be worth one dollar". The government only has some fairly blunt tools to influence the value of currency (for example, interest rates and quantitative easing) but a dollar is worth whatever someone is willing to trade for it.

The problem with Bitcoin is not that it's value is set by what people are willing to trade for it, but virtually no one is willing to trade goods or services for bitcoin, and instead it has become merely an instrument of speculation.

Unfortunately the so called debunking of the nuclear winter was done by somebody with no knowledge of climate science.

US and USSR scientists, independently, calculated the effect of the nuclear winter in the 1980s and came to the same conclusion. More recently simulations have been run again, this time with better understanding of the climate and massively more powerful simulation tools - and has discovered that the 1980s predictions if anything were optimistic - the nuclear winter effect was actually likely to be much worse. A simulation was also run on a regional conflict between India and Pakistan with an exchange of just 50 weapons. The effects are worse when the conflict occurs in subtropical latitudes. Such a conflict would trigger a "nuclear autumn" that would shorten the growing season in the US midwest by 60 days in the couple of years following the conflict, and would have serious climate consequences lasting about a decade. This level of growing season shortening would cause food shortages and high food prices in the rich industrialised world, and simply result in famine in poorer countries and places where agriculture is already marginal.

A US-vs-Russia exchange with a significant fraction of the arsenal...nuclear winter doesn't even describe it - more like "nuclear six month long night" - the models forecast mid day light levels in the northern hemisphere about similar to a moonlit night. Six months without food growth would kill most of the survivors, before even considering the cold.

On point 4 - young people finding old music and finding it better than 2010s pop - this may just be a case of 'survivorship bias'. The dross will fade away and the good stuff will stick around, but for newer music from the 2010s, the dross still hasn't had a chance to fade away.

There's also the issue of ETOPS. There was a time where legally a transoceanic airline flight required at least 3 engines. ETOPS (Extended-range Twin-engine Operational Performance Standards ... or Engines Turn or Passengers Swim) does away with that restriction, and over time larger and larger ETOPS limits have been introduced.

Actually, when a user-land process contains sensitive data and can run arbitrary code, it is a serious flaw. For example a web browser: contains in its memory space user passwords, authentication tokens, private keys etc. and can run arbitrary code from a remote source (Javascript). This could be exploited in a 'drive-by' manner with the user never being aware that it happened.

Reminds me of a joke I heard at the time. "You are flying in a plane whose fly by wire systems use Intel pentium processors. The Pentium implements IEEE floating point. How is IEEE pronounced when you are on the aircraft? Aieeeeeeee!!!"

You can do that. A Javascript POC has been written for the SPECTRE vulnerability (a related but different architectural problem to the one under discussion here: SPECTRE also abuses out of order execution and cache based information leakage). The SPECTRE vulnerability only will get you data from the same process's address space, but when that process is a browser and has things like authentication tokens, password managers and private keys it can still be a problem.

Leaking a user space program's own information can be a serious risk especially if that program can also execute arbitary code. A web browser is an example of such code. They have done a proof-of-concept where Javascript running on Chrome can leak information to a remote attacker information within Chrome's memory space. This could include sensitive information such as authentication tokens, private keys, the content of Chrome's password manager, etc.

Washing and reusing used to be a big thing in the UK: the dairies always did it on a big scale (they would deliver you milk, you'd use it, and put the empties out, the milk float that delivered the milk took the empties back to the dairy which would clean and re-use the bottle). There were soft drinks companies that also did the same thing.

This practise started dying out in the 1980s. There are still some dairies that do milk rounds and wash/reuse the bottles, but that's now a minority.

Well, except RISC did end up doing overwhelmingly well: ARM (which originally meant 'Acorn RISC machine') is now the most popular architecture, and ARM ships more cores in a year than x86 variants have in all time.

I wouldn't call it "hearing" a thud, any more than I hear my internal monologue. There was more of a perception of a thud sound, but it was distinct from if there was an actual thud sound. Interestingly if I didn't concentrate on the image, and kept it in my peripheral vision, the timbre of the "thud" would change, for instance it was more of a "twang" if I looked below the image so that it was in my upper peripheral vision.

With a bit of effort I could make the "thud" go away too, while looking directly at the image.