He is implying that white collar heavy companies that employ AI will be up to twice as productive as those that don't. His bold claim seems more like marketing targeted at CEOs than a warning directed at governments.

It's a bold claim because productivity increases in competitive industries do not lead to less jobs. In a competitive market, the spending ratios between staff and non-staff expenditures has been optimized and AI is not going to drastically change this. That's because job cuts will mean a reduction in service level or output relative to competitors that retain their staff and instead increase their service level, output or both.

For instance, a software company employing AI may be able offer either lower hourly rates and/or faster development turnaround with the same amount of staff they had pre-AI, as opposed to firing half the staff and keep the same service level post-AI.

Yet in other words, AI taking our jobs is nonsense. That's just not how it works in a competitive world, especially not in industries relying heavily on IT for which the demand is essentially unlimited.

That's just bullshit. Anyone can open an account on a cryptocurrency exchange and start trading. It works the same as trading stocks on your brokerage account. You can sell your crypto and withdraw the money to your bank account at any time. The liquidity for Bitcoin is staggering, your baseball collection card comparison is moronic and completely off the mark.

As you point out, the system of fiat currencies have inflation built in, but that is at odds with its function of fairly bookkeeping ones' reward for their economic activity. Money replaced barter and enabled quantification of economic activity as well as storage by carrying some portable token (seashells, beads, gemstones, coins, banknotes, silver, gold) enabling easy storage and deferring of receiving ones' reward. Most tokens turned out to be woefully inadequate because they lacked scarcity. And today when stored in modern day fiat, the compensation received also slowly melts away over the years for the same reason: lack of scarcity which we call monetary inflation and which is built straight in to the system.

To at least keep up with the inevitable inflation, people will have to invest their fiat. That in and of itself is not a bad thing during times of moderate real inflation*, but if real inflation becomes like over 4% then this causes significant adverse economic effects such as overheated real estate markets and incentivizing investments in uneconomical businesses -- after all losing 2% is still better than losing 4%. This ties up human labour into uneconomical activities, which the world has recently witnessed during the times of near-zero interest rates where at the same time the monetary inflation was sky high because of QE programs. One of the ill effects that the world witnessed was a flood of unprofitable VC funded startups that didn't have any viable business plan, but which tied up people doing unproductive labour even causing labour shortages which hurt existing viable businesses.

Bitcoin offers an alternative means of storing ones' wealth. At face value it's better suited than fiat in the sense that the supply of Bitcoin can't be artificially inflated over its preprogrammed schedule unlike fiat which can and often actually ends up getting debased by its meddling government. Bitcoin is also a better gold in the sense that it is cheaper to safely store, transport and transfer, impossible to counterfeit, and unlike gold has a predictable, known and fixed supply. The obvious quality missing is that Bitcoin hasn't proven its worth and stability over centuries, unlike gold, which is why it is still a hugely volatile asset. Only time will be able to fix that. The signs are good -- volatility has already decreased significantly since its inception 15 years ago.

So, Bitcoin definitely has a place in the world. Perhaps not to replace fiat altogether, but definitely as an alternative way to safeguard ones' wealth and to keep it isolated from economic meddling by governments.

*real inflation: price inflation minus the risk free rate. So if price inflation is 6% but risk-free you can earn 2%, then the real inflation would be 4%.

If fantastical and grand solutions are on the table, then how about creating a giant dust cloud that blocks a couple percent of sunlight in between the sun and the earth? Specifically at the first Lagrange point (L1) so it stays somewhat stationary in front of the sun relative to the earth. It doesn't carry as much geo-engineering risks and the cloud would disperse naturally into space over the years.

I wonder if there are batteries that cater to the specific use case of storing power for no longer than 24 hours. Conventional batteries are able to store power for months but clearly this is not required for this use case. So perhaps a much cheaper or more efficient technology exists that stores power for 24h max but does so more cheap and efficient than EV-style batteries.

The design of Bitcoin doesn't allow for a more energy-efficient method of mining. The automated issuing scheme of Bitcoin dictates that currently new 900 BTC per day are minted. Simplified, that's means that all miners together can expend an amount equivalent to up to 900 BTC per day for mining these 900 coins. Would they spend more, they'd be losing money. Anything they spend less is profit. (In addition, miners also get rewarded with transaction fees that they may add to their mining budget).

Miners convert the earned BTC to fiat (for instance, dollars) to pay for equipment, staff, rent and electricity. Every 4 years, the amount of BTC mined per day halves, and in fact there's an upcoming halving event in about 3 weeks from now. From that point onwards, only 450 BTC per day will be available to be mined, until that gets halved again to 225 in 2028.

Note that inefficient miners will quickly be forced out. Miners that expend more money than they earn from the amount of BTC that they mine will be making a loss and will quickly fold. To give a simple example: if you'd mine yourself using power from the grid then you'll likely find that you spend more in electricity than you earn from mining. Miners therefore are forced to employ the cheapest possible power to be able to compete. Fortunately, there is abundant free power available. Power from hydro, windmills and solar can be used for free during times in which more power is generated than the grid is able to absorb. Normally during these times, electricity prices can even turn negative just so that power plants stop flooding the grid with power it can't absorb. Storing the excess power in batteries is uneconomical. Miners however are happy to absorb this power and in fact, some power generating companies offer contracts to miners to help them balance the grid by switching off- or on- their mining equipment at the right moments.

Part of the power generated from renewables is normally wasted during times of excessive generation, but now this waste power can earn money by selling it to miners instead. This improves the business case for renewables which means cheaper and cleaner power for everyone.

Looking at it this way, the case can be made that Bitcoin even promotes renewable energy. But the takeaway from all this is that the reality is much more nuanced than your typical "muh bitcoin wastes power, bad" knee-jerk reaction, which, by the way, also asserts that Bitcoin has no utility which is absolutely false.

The next step will be to bury the wood barks, plant stalks and even entire tree trunks at the site of abandoned depleted oil wells. After decomposing, the carbon rich remains will be turning into a black liquid that over time will fill up the underground cavities.

The great majority of unsophisticated "investors" will lose significant chunks of their money when they attempt any form of day trading, no matter in what asset: stocks, options, bonds, commodities, currencies or "crypto".

That's because they're up against an unbeatable army of professionals and algorithms that no one can dream of outperforming consistently with day trading. This is exacerbated by greed, impatience and fear of losses that amateurs who engage in day trading are subject to more than anyone else: the reason to engage in day trading is making a quick buck, which selects for risk takers with tendencies of greed and emotional decision taking.

The summary does not mention whether the son of Lagarde engaged into daytrading but virtually everyone putting large amounts of their net worth into "crypto" qualifies as a greed driven risk taker and is bound to lose their "investments". The scam-ridden universe of "crypto" makes this even more of a certainty.

Now for some nuance, the term "crypto" compasses the thousands of opportunistic Bitcoin copycat cryptocurrencies that are almost without exception outright scams, as well as the ill-conceived concept of NFTs and "stablecoins", the latter used by exchanges to fraudulently hold fractional reserves like a bank legally can.

For some perspective, a non-speculative investor with patience that would have put their money into Bitcoin and just waited it out, would be in the green considering most time frames (any time frame actually, except mostly for the period between February 2021 to May 2022 and today).

Bitcoin also can be regarded as a highly speculative investment that can go to zero at any time, but at least it is very different and far removed from anything "crypto". For starters, it was envisioned conceptually as an uncensorable money, and it first proved its utility by allowing payments to Wikileaks at a time it got shut off from the banking network. Bitcoin genuinely represents a new paradigm that many people believe may actually work and survive as a money free of political meddling and manipulation through interest rates and other central banks interference. Therefore, it would be a mistake to lump Bitcoin together with scam-ridden "crypto" horseshit. Also, it must be noted that there is no need to trade crypto for unsophisticated investors to lose their shirt: any brokerage account offers plenty of opportunities for them to get completely wrecked.

This highlights the shift from using astronomical yardsticks such as the earth rotation and its position relative to the sun to derive a time and date, to using a measure that doesnâ(TM)t depend on any of these.

I argued that timestamp logic would be more robust if it were to use a monotically increasing counter instead of a date and time. The summary does not mention any of that.

Not sure if you're implying this was mentioned in the summary, because it wasn't.

Thank you! Didn't know TAI existed. Yes, that'd do the job.

Surprisingly, UNIX time originally behaved exactly like TAI, causing localtime() to run fast: one second after each leap second. Sadly, with the advent of NTP, instead of adapting localtime() to correctly account for leap seconds when converting to time, the xntp daemon was adapted instead to repeat the same second twice.

Story here: https://cr.yp.to/proto/utctai....

The proper solution for timestamps is to use a counter that only counts forwards and never skips.

We could define this counter as being the number of actual seconds that passed from some arbitrary fixed point in time (akin to unix counting seconds passed since 1/1/1970, but without adjusting it for leap seconds). Where "actual seconds" means the number of seconds that actually passed since that predefined point in time. Converting from a number to a date and time is performed by taking into account the full list of all the adjustments (i.e., leap seconds) that took place (and of course the timezone logic).

For instance, between 1/1/1970 and 1/1/1980, 9 leap seconds were introduced. So, for example, if we would define 1 as 1/1/1970 00:00:01 UTC, then 1/1/1980 00:00:00 UTC is calculated as 3600 * 24 * (365 * 8 + 366 * 2) + 9. Reversely, 3600 * 24 * (365 * 8 + 366 * 2) would represent 12/31/1979 23:59:51 UTC. (The 2 * 366 are to account for the two leap years 1972 and 1976).

This would fix the ambiguity of timestamps. It would benefit calculating actual timespans as well, as it eliminates a giant footgun for those applications where one-second resolution precision is required over longer periods, such as in space travel.

This scheme can be implemented even without the need of some central timekeeper. The height of the forward-only counter can always be calculated for any point in time using the list of leap seconds. Just calculate the height of the counter for every timestamp and store that number. Use the reverse calculation to translate back to date and time.

Let's hypothesize that space and time are emergent properties of more fundamental physics.

Time being emergent is defendable as for instance light does not experience any passage of time. There is no way to establish that in between two interactions, elementary particles experience any passage of time. All our knowledge of particles is necessarily based on behaviors caused by interactions.

Space being emergent is also defendable. For instance, an electron is a point particle that does not have any volume. Again there is no way to establish the volume of elementary particles. All our knowledge is based on interactions. There's no way for us to know the properties of particles when they're not interacting with other particles.

Given the emergent space / time hypothesis, it must follow that the properties of space, time and particles are constant over "time". A deeper physics determines these properties, but we cannot see because our knowledge is restrained by what we measure by interacting.

"Conflict of Interest Disclosures: Dr Martel reported receiving cofounder equity from Auricle Inc and coinventor royalties from the University of Michigan during the conduct of the study; being cofounder of Auricle Inc outside the submitted work; and being a coinventor on US Patent 9,682,232, Personalized Auditory-Somatosensory Stimulation to Treat Tinnitus, outside the submitted work. Dr Schvartz-Leyzac reported receiving grants from the National Institutes of Health (NIH) during the conduct of the study. Dr Shore reported reported receiving cofounder equity from Auricle Inc and coinventor royalties from the University of Michigan during the conduct of the study; being cofounder of Auricle Inc outside the submitted work; and being a coinventor on US Patent 9,682,232, Personalized Auditory-Somatosensory Stimulation to Treat Tinnitus, outside the submitted work."

Source

To boost credibility of this study to acceptable levels, independent groups must replicate the study and find the same results.