Move over five stages of grief, we're at the fourth stage of leash.
The fifth stage of leash is where the hapless victim becomes adamant that the codependent relationship (who really controls that pesky CPU?) and corporate surveillance is all in your own best interest, and for your own good.
You can beat a polygraph by clenching your sphincter when answering questions, it causes a slight skin flush and rise in blood pleasure which is how the polygraph measures "lies", but they have to establish a baseline so if you do it every single time you answer a question the detection system can't see deception.
You're just begging for a repeat procedure with a clenchometer jammed somewhere uncomfortable.
And since sphincters are a class of muscles, with enough determination, you could earn yourself the instrumentation Full Monty.
It's a simple cost of doing business for any consumer who wires their bank credentials into the cloud, and then runs random applet downloads in the same sandbox.
So you put a wall-sized aquarium in your nursery with your newborn twins, and inside the aquarium you stock a giant python or cobra, and then you get yourself a riced-out 1 hp Roomba from Akihabara, just because, and then you download an experimental, indoor auto-mapping package for said Roomba from some applet pop-shop located in a dusty, foreign currency–starved minor-dictatorship whose borders you could barely sketch onto the right continent.
The fine should be cut in 1/2. I told a few customers one day in Best Buy that Lenovo was installing this trash on systems as well as using the mainboard to store this trash.
Your logic is circular. You think they should have trusted "some guy" spouting an opinion, who turns out to be so rational, he's insisting they should have trusted "some guy" spouting an opinion, years later...
Moreover, your 15-second anecdotal interaction warrants a 50% revision in how the world turns.
No idea why Joe Random Consumer might not trust your demeanor on first glace.
Not that it really matters anyways since it's just a marketing term. Intel's 10 nm has roughly similar characteristics to TSMC's 7 nm process.
Everyone who follows this even a little bit already knows that Intel's processes are the real thing, and everyone else's numbers are measured with parallax and a fat rubber ruler.
Not that this normally matters, as you point out. It only matters when some idiot comes along crowing about how Intel has been "left in the dust".
It's also the case that Intel is designing transistors to run fast and hot with extreme reliability (big profits from Xeon). If you're doing it right, "hot" is a synonym for "agile", meaning that your performance holds up on nearly every workflow. For this kind of agility, hot is the price you pay.
Of course, there are oodles of workflows which can sacrifice one form of agility or another at almost zero performance cost, and these chips can certainly be designed to run cooler than Intel's chips while maintaining the same performance level for the chosen workflow.
Key concept: one form of agility or another.
The difficulty lies in exploiting Intel's excess heat production without Balkanizing your product line.
Major cloud providers don't want to lock themselves into a compute architecture that's only good for today's compute fashion. Fast and hot Intel CPUs have a long track record of maintaining their performance come hell or high water (the later being the greater fright in the cloud business).
The compute landscape heaves whenever a giant workflow such as machine learning calves off. One day your regular compute is huge into ML support, the next day it isn't, and now your regular compute is scrambling for new business. Does it make no sense at all for cloud providers to err on the side of hot and agile for their faddish default compute bucket? You'd quickly come to that conclusion if gobble up some of the ARM literature too readily.
ARM is trying to be twice as good, but what calves off is usually 10x as good. Isn't that an old lesson?
So Intel's legacy of hot and agile is not quite the impediment it's constantly made out to be (as it ever was), because the competitive frame isn't driven by tiny 2x improvements. It's rare that anyone's risk tolerance sidles quietly into a 2x disruptive improvement (even more rare when 10x disruptive improvements are also afoot).
The bottom line here is that Intel's actual 10 nm transistors shouldn't be compared naively to someone else's putative 7 nm transistors, as if apples were the only fruit in the world.
[*] English is renowned for it's depth of synonym baggage, but we simply don't have a good one for the geographically loaded "Balkanize". Besides, I've read about 1453 (and its sequellae), and it was certainly earned. Maybe we can finally retire this term after the Great Chinese Dissolution of 2096, their galling Social Credit System having finally run its natural course (move over Balkanization, here comes Sesamicide).
So what this means is that a computer has decided Microsoft shares will go up in value by enough to make day trading in Microsoft shares highly profitable.
None of these algorithms would work were it not for the boundary condition that many people invested in the market are investing for value, i.e. a future share in the underlying profit stream.
The second stable semi-stable pool are the people busy hedging their bets to create a stable business environment. A proper hedge is conceptually a moving target. Hedge positions need to slosh around a bit all the time. What the speculators are doing is projecting where all this sloshing will find its natural equilibrium. For this service, they extract a profit, sometimes a large profit.
Otherwise, all the hedge funds would attempt to debark on the same appealing tropical island to ride out a storm, and there would be serious run on coconut milk, and soon the "pomegranate" Pina Coladas would be haemoglobin infused.
Moral of the story: you can't all hedge to the same shelter at the same time. Speculators increase the convergence rate to acceptable shelter occupancy levels. They're the PID controllers of nervous nelliehood.
Chris Anderson: But should we worry about the hedge fund industry attracting too much of the world's great mathematical and other talent to work on that, as opposed to the many other problems in the world?
Jim Simons: Well, it's not just mathematical. We hire astronomers and physicists and things like that. I don't think we should worry about it too much. It's still a pretty small industry. And in fact, bringing science into the investing world has improved that world. It's reduced volatility. It's increased liquidity. Spreads are narrower because people are trading that kind of stuff. So I'm not too worried about Einstein going off and starting a hedge fund.
Right, the average physicist couldn't math the hell out of a wet paper bag. And it's almost complete bullshit on another level, too, because much of the profit is driven by having a pre-eminent vantage point, which is extracted by rent rather than mojo.
But it's not complete bullshit. Underneath all that, the speculators are indeed performing a small, valuable service in smoothing out the dynamic behaviours of a too-complex system.
This is hardly ever explained at length by anyone competent on the inside, because very quickly any smart person can tell that this explanation's dick is small compared to the total volume of proceeds extracted (which only serves to heighten attention to the rent side of the business).
But since (for this reason) it's rarely mentioned, that doesn't mean it doesn't exist at all.
Jim Simons: We stayed ahead of the pack by finding other approaches -- shorter-term approaches to some extent. The real thing was to gather a tremendous amount of data -- and we had to get it by hand in the early days. We went down to the Federal Reserve and copied interest rate histories and stuff like that, because it didn't exist on computers.
We got a lot of data. And very smart people -- that was the key. I didn't really know how to hire people to do fundamental trading. I had hired a few -- some made money, some didn't make money. I couldn't make a business out of that. But I did know how to hire scientists, because I have some taste in that department. So, that's what we did. And gradually these models got better and better, and better and better.
Maybe it's something else I read about this guy (he's directly connected to a controversial Robert Mercer, and Peter F. Brown, and a huge pile of NLP research papers I once amassed myself back in the 1990s) but he does point out that the cream here
So many commenters here who figure that scientists are the last people to get the news about fragile food chains and the dangers of unintended consequence. Probably these commenters are the same group of people who never did their own arithmetic to determine that the precautionary principle can't be applied along all possible dimensions simultaneously.
Add an invasive species where it wasn't formerly found. Big problem: the precautionary principle says to remove the species immediately (some things are still being mucked up), but it also says to leave the species alone (some things might already depend upon it).
Even for the things we've done (as a species) that we now mostly regret, we don't have a complete score card. Perhaps the first-order effect looks horrific—but honestly, even in those cases, it's still not a risk-free conclusion that we messed up.
For some reason, people like to elevate the precautionary principle to some independent, meta level of confident cognition, which completely obviates the kind of hard thinking we originally messed up.
Once upon a time, mammals were a pandemic of invasive species—all of them, pretty much everywhere.
Precaution is more about change management than change avoidance. In change management, there's no magic, default "leave alone". The context around the thing you're "leaving alone" is also undergoing change.
That's why the Buddhist philosophers say you can never step into the same river twice—the precautionary principle applied to the precautionary principle—which for some reason causes most Westerners to go "why would you do that?" before returning to their confident precautionating.
Did you know that 97% of Americans retire financially broke dependent upon friends, family, and the Federal Government.
Note that the image of the black guy (by implication, the author) is facing away from the center of the page. This is symbolic of turning is back on dull orthodoxy.
Not also the visual appeal to bright lights, big city and the casino-ish orange/yellow/red reflective highlights, which form the focal point of masthead Oz.
Almost everyone retires broke and dependent on some other entity other than a giant pile of coins or bills amassed under a mattress. That entity might be a commercial bank (fairly solvent) or a National Bank (aka Uncle Sam, with the deepest pockets of all) who is paying out a stipend on what you paid in (a fairly standard business model wherever banks gather).
Even if you retire to a giant mansion, entirely bought and paid for, you're still dependent on Uncle Sam to continue to enforce your property rights (those who employ a private vigilante retinue rarely sleep with both eyes closed, so there's a downside to that model, too).
If they don't do these things, then why should the taxpayers subsidize their broken business model?
What subsidy? If you pass a law and do next to nothing to enforce the law, the net consequence is that the people running the officially illegal Grinch Bot networks have to keep all their internal communications far from the public eye (lest they self-incriminate), and neither can they brag about their prowess publicly (lest they self-incriminate with a bullhorn).
At almost no public cost, shadowy activities are forced to operate under shadowy constraints.
On the flip side, the vendors aren't idiots, and for sure they're making a profit somewhere on these lost-leader activities.
And we know how this works. The consumer gets a dopamine rush from thinking they've scored a small jackpot (casinos do this, too).
Dopamine is the slush-fund hormone. Found money is money you don't have to discuss with the wife. Very quickly, the typical consumer gets pie-eyed about all the things they would love to purchase, absent spousal surveillance.
This is the essence of the casino model. Punters will invest $1000 to win a $1200 jackpot (you're momentarily up on the house, but it won't last long). His responsible pocket is now $1000 in the red, while his irresponsible pocket is $1200 in the black. Mission accomplished. Drinks all around; master of the house; loose women for breakfast, lunch, and dinner. Lather, rinse, repeat.
So in the e-commerce setting, you save $40 on a Cabbage Patch doll for your daughter, but then you linger to spend $80 (20% off Today Only) on some toy for yourself.
This kind of person, the kind of person who easily succumbs to dopamine thrall, is a valuable economic commodity; their fleece will keep hundreds of other people warm for the entire winter. Shame if the Grinch Bots ruining the mass e-commerce sport of fleecing dopamine-addicted rubes.
The Holy Grail for any politician is to get half of the sleepy electorate to vote for "tough on crime" instead of superior fiscal administration & accountability.
You'll rarely ever find a politician who doesn't deeply believe in fleecing vengeance-distracted rubes of what little political power they actually possess. When you promise economic responsibility to the masses, you can't deliver economic windfalls to your powerful friends. Lose, lose. So of course they're going to pass a hollow paper law to outlaw Grinch Bots, even with no effective policing strategy identified (all the better to balance the books).
This is almost the founding belief of American society: the clueless rube must soon be parted with their A) money, and B) democratic influence, by all available means.
You can induce many people to fall into casino tilt, and you can induce many other people to fall into moral tilt ("tough on crime" is the tried-and-true call to action for moral tilt, and if that doesn't hit home, you can always add a side order of Daughters of White Folk virginal frailty—we're looking at you, Mexico).
This is not just a tiny little business model. It's practically the American way of life, writ large.
If all you have is a scalar hierarchy, rebasing tends to look like a short-term plunge.
It's true, we do tend to rank people by salary, as if expertise were a scalar commodity. I will agree at the job-change boundary, the scalar projection sucks more than ever before.
But it sucks all the time, if you're a deep thinker, so this isn't conceding much.
If there's any practical gap between the nominal value and the cash liquidity of Bitcoin, Ohio had better brace itself for a stampede of Bitcoin payments.
If there's any practical gap, this might be a cynical move on the part of some wealthy Ohio speculator with political connections.
Most likely this is some person who scored a Bitcoin windfall and now wishes to return to an actual cash position at face value, who increasingly anticipates a Bitcoin implosion any day now.
It's not clear what you're asking. If a bit in the cache gets changed, it corrupts the instruction or data. That the cache is powered up makes no difference.
Wrong.
Next contestant, please.
You're assuming the cache has not parity or ECC mechanism where active use would eliminate single-bit errors before they accumulate into undetectable errors, whereas pickling the cache in quiescent warm brine would not.
If all of the public research was public, then we'd all be able to see how much of it is a sham
What you mean is that too many studies are underpowered, so you get a lot of false positives over the 0.95 bar.
We can fix this by making every study 4 to 10 times more expensive (everybody gets a huge N, except perhaps in the cases where the study population is finite, regardless of budget).
Would we actually be better off, or have you just made science twice as expensive as it really needs to be?
Another implicit assumption is that if a scientist (or team) publishes a paper that later fails to replicate, that all the money invested in the study was completely wasted (government waste is by tradition consigned to money-up-in-flames category, including tax revenues that the government immediately pays out again as citizen benefits, merely because some government official touched it—COOTIES—on the way through).
The every-failure-is-a-100%-writeoff presumption isn't true in Silicon Valley (many failed entrepreneurs learn invaluable lessons) and it isn't true in science, either.
Influencers is an interesting tag, and speaking for myself and me only, I've also never quite understood why people who can act in movies are somehow qualified to make important contributions with their opinions that suggest preference for candidates and political positions.
You're not reading enough Jordan Peterson. What women want (from their male romantic partners) is competence. Women generally favour a particular competence signal: societal attention / societal approval (if the attention signal is large enough, the approval half of the signal is irrelevant). Commanding the center of societal attention is a universal proxy for power. All forms of wealth command attention: conspicuous wealth, foolish wealth, (mostly) concealed wealth, grasping wealth, tight-fisted wealth. You will never sleep alone (nor without one hand on your pocket book).
Celebrity is almost by definition a top-drawer competence proxy.
For some reason, one of our core social heuristics is to emulate role models (those with power) by aping their behaviours, even the cynical behaviours we know they put on merely to exploit the unwashed.
I think the problem here is that a celebrity figure's ability to get away with these cynical behaviours in plain sight amplifies the power signal more than enough to compensate for the blowback signal of self-interested disgust at overlord overreach. (The balance likely depends on your own self-esteem curve.)
Then there's this other problem: a part of our brain is wired to presume that the mass behaviour of a billion people with low self-esteem filters can never be wrong.
Most people have more power than they realize. The problem is that you can't directly witness the loops of cause and effect.
If more people possessed explanatory depth—discussed at length in The Knowledge Illusion (2017) by Steven Sloman—it wouldn't be possible for know-nothing influencers to hijack commercial success.
From the corporate side, given a choice between customer A with explanatory depth, and customer B without explanatory depth, you might do fine with customer A in a B2B setting, but you definitely want customer B in a B2C setting.
Given enough power, you can actually shift the balance in the population at large from type A dominance to type B dominance, as Apple has done so successfully since introducing the first iPhone (Apple was long attempting to ride this dragon, and nearly bankrupt itself in a fire while doing so, but then the rewards were spectacular once the dragon finally took flight).
I've been reading review after review about how the 2018 Mac mini basically appeals to no-one with any vestige of explanatory depth, except under the general category of "well, if it's sufficiently inconvenient to escape the Apple tent altogether, this is the cheapest way to cling to the outskirts, and might even qualify as a reasonable purchase if your work load particularly needs compute, but never pushes a pixel at an animated frame rate; or you've got some kind of weird office aesthetic where having a shiny little recycled aluminum box with all kinds of crap hanging off the back on short and expensive interconnects is your idea of a glamorous rat's nest".
The Apple product literature doesn't even supply anyone with explanatory depth so much as a Bathroom Reader of Mac mini technical disclosure. I had to find some obscure enthusiast forum and wade through a hundred Comments of Dreck to find out that the 128 GB SSD option has half the write throughput of the 256 GB SSD option (with another increment at 512 GB, but not nearly so substantial, especially at the GIANT cost increment).
There's no online block diagram at even the highest level of the T2 chip. There's a block diagram of the marketing department's view of Alpine Ridge from about three years ago, that does as much to confuse as to reveal. Silicon vendors like Newark sell some of the older Intel TB3 parts, but not a single one of these listings comes w
My faith in the scientific model of dire anthropogenic climate change started at about 30% in 1980, and has increased by about 10% per decade since.
In any other area of science, this would be regarded as a rapid confidence ramp with respect to a wicked question of science, but stakes are very high with the planet hanging in the balance, so double standard.
My faith in the dire economic projections is not ramping at anywhere near that rate. It's mostly just a giant FUD sundae, and neither the rigors of traditional peer review nor the actual credentials wielded have so far managed to impress me.
Without doubt, Bad Shit Could Happen.
But to construe this debate as a science debate, one where the scientific consensus is automatically credible, simply boggles my mind.
There is a scientific nucleus to this debate: if these climate changes continue Things Would Have to Change.
Yes, they would—but they were going to change anyway, because change in the modern world is simply a given (and the established rate of change is pretty much in constant acceleration, too).
But These Changes Are Really Bad Changes We Shouldn't Go Anywhere Near.
Is that economics, futurology, ideology, or philosophy? Or some unspecified mixture of all four?
And my rebuttal: quite possibly, the wrenching changes to existing conventions of human society to avoid these environmental outcomes are also Really Bad Changes We Shouldn't Go Anywhere Near (because this, too, is highly experimental terrain, and everyone can see how our recent political foray into social media is just peachy).
There are existential risks on all sides of this.
Scientists: Well, if you're too stupid to do what we're telling you to do without starting a food fight (complete with mushroom clouds), don't blame us.
Maddening horde: If we can't blame you when the prescribed giant change initiative all goes sideways, why should we listen in the first place?
Scientists: Because we're right about the science.
Maddening horde: Well, perhaps that makes you feel good, because what you care about most of all is the science, but it doesn't do jack shit for the rest of us, who don't regard one self-inflicted bad outcome as automatically better than any other self-inflicted bad outcome.
The costs of climate change could reach hundreds of billions of dollars annually, according to the report. The Southeast alone will probably lose over a half a billion labor hours by 2100 due to extreme heat. Farmers will face extremely tough times. The quality and quantity of their crops will decline across the country due to higher temperatures, drought and flooding. In parts of the Midwest, farms will be able to produce less than 75% of the corn they produce today, and the southern part of the region could lose more than 25% of its soybean yield. Heat stress could cause average dairy production to fall between 0.60% and 1.35% over the next 12 years -- having already cost the industry $1.2 billion from heat stress in 2010.
Whether climate change is driven (in part or in full) by human activities is a science question and will ultimately be decided by scientific criteria (exactly when such a reliable decree comes down the pike remains open to debate, though increasingly less open, year over year, year after year, on current trends).
Extrapolating the economic cost of heat stress ventures deep into the dark, unreliable heart of economics (which is referred to as "the dismal science" for a good reason) is a mug's game.
Even glib, prognosticating economists can't make these projections without taking into account future human ingenuity. Economics with a side-order of futurology—what could possibly go wrong?
I'm getting ever more grumpy about this constant bait and switch: we're really, really, really confident of our climate model this time, so let us now describe with infinite confidence the future we envision using our entirely non–Magical 8 Ball (this involving two additional academic fields, one barely respected, the other openly derided; moreover, the vast majority of the dignified-to-a-fault, hard-science authors of this publication have any specific training—or tainting—in either of these shabby fields, and that's why you should believe every word we write).
Because nobody gives a fuck why someone in Africa croaks.
Until you point out it's an incurable, highly contagious hemorrhagic fever with a brief 48-hour incubation period during which it expresses no tell-tale symptoms, whatsoever.
When your grade three teacher told you about Africa being "over there", her own education was already ten years out of date.
Just wait until all those Pennsylvanian coal miners find out what "globalization" really means.
All it takes is one tainted kilo of weed delivered by Swift Boat in the dark of night.
That is how most rich people think. Just delay it, global warming, population explosion, running out of oil, precarious infrastructure, giving away the crown jewels of the economy to the enemy, nothing matters, as long as I am convinced all the bad things will happen after my lifetime.
Yoda contains more wisdom about the Grand Mastery of prudent delay in the green fingerclaw of his pinky hand-toe than you will ever even suspect.
The pace of discovery is slowing because of the law of diminishing returns. We already picked the low hanging fruit. Now each incremental advance gets more and more expensive, and the number of significant breakthrough "leaps" get fewer and farther between. Same as everything else. Cars, circuits, razors, microwave ovens. Each advance is increasingly more complex and costly than the last.
This is simply not true.
Back when we discovered penicillin, we lied to ourselves in our giddy triumph over the microbial world about what mastery of global antibiotic use actually entailed, so we blindly painted ourselves into a terrible predicament in Alpha Quadrant Superbug.
That's exactly the kind of low-hanging fruit a hunchbacked woman with a gnarled crook hands you in a meadow clearing in the deepest depths of some dark, German forest (or British wardrobe—what passes for a forest on a barren island).
"Help yourself to another square of Turkish Delight, Edmond, it will make you immmmuuuuuuuuune!"
A hundred years passes with Edmond mostly staggering around in a dark pall, but then Edmond finally gets seriously woke to the human microbiome (and, indeed, gloriously complex microbiomes everywhere, if only we'd bothered to investigate with adequate tools).
Personally, sign me up for Edmond 2.0, because Edmond 1.0 was knee-deep in clusterfuckage, and barely had the wits to even know it.
Slashdot Mirror
Move over five stages of grief, we're at the fourth stage of leash.
The fifth stage of leash is where the hapless victim becomes adamant that the codependent relationship (who really controls that pesky CPU?) and corporate surveillance is all in your own best interest, and for your own good.
Father knows best.
Long live the hunchback salute.
You're just begging for a repeat procedure with a clenchometer jammed somewhere uncomfortable.
And since sphincters are a class of muscles, with enough determination, you could earn yourself the instrumentation Full Monty.
It's a simple cost of doing business for any consumer who wires their bank credentials into the cloud, and then runs random applet downloads in the same sandbox.
So you put a wall-sized aquarium in your nursery with your newborn twins, and inside the aquarium you stock a giant python or cobra, and then you get yourself a riced-out 1 hp Roomba from Akihabara, just because, and then you download an experimental, indoor auto-mapping package for said Roomba from some applet pop-shop located in a dusty, foreign currency–starved minor-dictatorship whose borders you could barely sketch onto the right continent.
What could possibly go wrong?
There's probably a secret messages encoded in Slashdot's delightful crop of fresh misspellings.
Wouldn't surprise me if the synopsis author could properly spell horcrux / horcruxes.
Priorities, man.
Your logic is circular. You think they should have trusted "some guy" spouting an opinion, who turns out to be so rational, he's insisting they should have trusted "some guy" spouting an opinion, years later ...
Moreover, your 15-second anecdotal interaction warrants a 50% revision in how the world turns.
No idea why Joe Random Consumer might not trust your demeanor on first glace.
Everyone who follows this even a little bit already knows that Intel's processes are the real thing, and everyone else's numbers are measured with parallax and a fat rubber ruler.
Not that this normally matters, as you point out. It only matters when some idiot comes along crowing about how Intel has been "left in the dust".
It's also the case that Intel is designing transistors to run fast and hot with extreme reliability (big profits from Xeon). If you're doing it right, "hot" is a synonym for "agile", meaning that your performance holds up on nearly every workflow. For this kind of agility, hot is the price you pay.
Of course, there are oodles of workflows which can sacrifice one form of agility or another at almost zero performance cost, and these chips can certainly be designed to run cooler than Intel's chips while maintaining the same performance level for the chosen workflow.
Key concept: one form of agility or another.
The difficulty lies in exploiting Intel's excess heat production without Balkanizing your product line.
Major cloud providers don't want to lock themselves into a compute architecture that's only good for today's compute fashion. Fast and hot Intel CPUs have a long track record of maintaining their performance come hell or high water (the later being the greater fright in the cloud business).
The compute landscape heaves whenever a giant workflow such as machine learning calves off. One day your regular compute is huge into ML support, the next day it isn't, and now your regular compute is scrambling for new business. Does it make no sense at all for cloud providers to err on the side of hot and agile for their faddish default compute bucket? You'd quickly come to that conclusion if gobble up some of the ARM literature too readily.
ARM is trying to be twice as good, but what calves off is usually 10x as good. Isn't that an old lesson?
So Intel's legacy of hot and agile is not quite the impediment it's constantly made out to be (as it ever was), because the competitive frame isn't driven by tiny 2x improvements. It's rare that anyone's risk tolerance sidles quietly into a 2x disruptive improvement (even more rare when 10x disruptive improvements are also afoot).
The bottom line here is that Intel's actual 10 nm transistors shouldn't be compared naively to someone else's putative 7 nm transistors, as if apples were the only fruit in the world.
[*] English is renowned for it's depth of synonym baggage, but we simply don't have a good one for the geographically loaded "Balkanize". Besides, I've read about 1453 (and its sequellae), and it was certainly earned. Maybe we can finally retire this term after the Great Chinese Dissolution of 2096, their galling Social Credit System having finally run its natural course (move over Balkanization, here comes Sesamicide).
None of these algorithms would work were it not for the boundary condition that many people invested in the market are investing for value, i.e. a future share in the underlying profit stream.
The second stable semi-stable pool are the people busy hedging their bets to create a stable business environment. A proper hedge is conceptually a moving target. Hedge positions need to slosh around a bit all the time. What the speculators are doing is projecting where all this sloshing will find its natural equilibrium. For this service, they extract a profit, sometimes a large profit.
Otherwise, all the hedge funds would attempt to debark on the same appealing tropical island to ride out a storm, and there would be serious run on coconut milk, and soon the "pomegranate" Pina Coladas would be haemoglobin infused.
Moral of the story: you can't all hedge to the same shelter at the same time. Speculators increase the convergence rate to acceptable shelter occupancy levels. They're the PID controllers of nervous nelliehood.
The mathematician who cracked Wall Street — March 2015
Right, the average physicist couldn't math the hell out of a wet paper bag. And it's almost complete bullshit on another level, too, because much of the profit is driven by having a pre-eminent vantage point, which is extracted by rent rather than mojo.
But it's not complete bullshit. Underneath all that, the speculators are indeed performing a small, valuable service in smoothing out the dynamic behaviours of a too-complex system.
This is hardly ever explained at length by anyone competent on the inside, because very quickly any smart person can tell that this explanation's dick is small compared to the total volume of proceeds extracted (which only serves to heighten attention to the rent side of the business).
But since (for this reason) it's rarely mentioned, that doesn't mean it doesn't exist at all.
Maybe it's something else I read about this guy (he's directly connected to a controversial Robert Mercer, and Peter F. Brown, and a huge pile of NLP research papers I once amassed myself back in the 1990s) but he does point out that the cream here
So many commenters here who figure that scientists are the last people to get the news about fragile food chains and the dangers of unintended consequence. Probably these commenters are the same group of people who never did their own arithmetic to determine that the precautionary principle can't be applied along all possible dimensions simultaneously.
Add an invasive species where it wasn't formerly found. Big problem: the precautionary principle says to remove the species immediately (some things are still being mucked up), but it also says to leave the species alone (some things might already depend upon it).
Even for the things we've done (as a species) that we now mostly regret, we don't have a complete score card. Perhaps the first-order effect looks horrific—but honestly, even in those cases, it's still not a risk-free conclusion that we messed up.
For some reason, people like to elevate the precautionary principle to some independent, meta level of confident cognition, which completely obviates the kind of hard thinking we originally messed up.
Once upon a time, mammals were a pandemic of invasive species—all of them, pretty much everywhere.
Precaution is more about change management than change avoidance. In change management, there's no magic, default "leave alone". The context around the thing you're "leaving alone" is also undergoing change.
That's why the Buddhist philosophers say you can never step into the same river twice—the precautionary principle applied to the precautionary principle—which for some reason causes most Westerners to go "why would you do that?" before returning to their confident precautionating.
Anyone who thinks my post was over the top, wade through the following document very carefully:
Are you ready to stop living the 40-40-40 Plan and Start Living your Dreams?
Note that the image of the black guy (by implication, the author) is facing away from the center of the page. This is symbolic of turning is back on dull orthodoxy.
Not also the visual appeal to bright lights, big city and the casino-ish orange/yellow/red reflective highlights, which form the focal point of masthead Oz.
Almost everyone retires broke and dependent on some other entity other than a giant pile of coins or bills amassed under a mattress. That entity might be a commercial bank (fairly solvent) or a National Bank (aka Uncle Sam, with the deepest pockets of all) who is paying out a stipend on what you paid in (a fairly standard business model wherever banks gather).
Even if you retire to a giant mansion, entirely bought and paid for, you're still dependent on Uncle Sam to continue to enforce your property rights (those who employ a private vigilante retinue rarely sleep with both eyes closed, so there's a downside to that model, too).
What subsidy? If you pass a law and do next to nothing to enforce the law, the net consequence is that the people running the officially illegal Grinch Bot networks have to keep all their internal communications far from the public eye (lest they self-incriminate), and neither can they brag about their prowess publicly (lest they self-incriminate with a bullhorn).
At almost no public cost, shadowy activities are forced to operate under shadowy constraints.
On the flip side, the vendors aren't idiots, and for sure they're making a profit somewhere on these lost-leader activities.
And we know how this works. The consumer gets a dopamine rush from thinking they've scored a small jackpot (casinos do this, too).
Dopamine is the slush-fund hormone. Found money is money you don't have to discuss with the wife. Very quickly, the typical consumer gets pie-eyed about all the things they would love to purchase, absent spousal surveillance.
This is the essence of the casino model. Punters will invest $1000 to win a $1200 jackpot (you're momentarily up on the house, but it won't last long). His responsible pocket is now $1000 in the red, while his irresponsible pocket is $1200 in the black. Mission accomplished. Drinks all around; master of the house; loose women for breakfast, lunch, and dinner. Lather, rinse, repeat.
So in the e-commerce setting, you save $40 on a Cabbage Patch doll for your daughter, but then you linger to spend $80 (20% off Today Only) on some toy for yourself.
This kind of person, the kind of person who easily succumbs to dopamine thrall, is a valuable economic commodity; their fleece will keep hundreds of other people warm for the entire winter. Shame if the Grinch Bots ruining the mass e-commerce sport of fleecing dopamine-addicted rubes.
The Holy Grail for any politician is to get half of the sleepy electorate to vote for "tough on crime" instead of superior fiscal administration & accountability.
You'll rarely ever find a politician who doesn't deeply believe in fleecing vengeance-distracted rubes of what little political power they actually possess. When you promise economic responsibility to the masses, you can't deliver economic windfalls to your powerful friends. Lose, lose. So of course they're going to pass a hollow paper law to outlaw Grinch Bots, even with no effective policing strategy identified (all the better to balance the books).
This is almost the founding belief of American society: the clueless rube must soon be parted with their A) money, and B) democratic influence, by all available means.
You can induce many people to fall into casino tilt, and you can induce many other people to fall into moral tilt ("tough on crime" is the tried-and-true call to action for moral tilt, and if that doesn't hit home, you can always add a side order of Daughters of White Folk virginal frailty—we're looking at you, Mexico).
This is not just a tiny little business model. It's practically the American way of life, writ large.
If all you have is a scalar hierarchy, rebasing tends to look like a short-term plunge.
It's true, we do tend to rank people by salary, as if expertise were a scalar commodity. I will agree at the job-change boundary, the scalar projection sucks more than ever before.
But it sucks all the time, if you're a deep thinker, so this isn't conceding much.
I guess you've been running on the same Apple computer for fifteen years now, and you've never had a business relationship with Jeff Bezos.
If there's any practical gap between the nominal value and the cash liquidity of Bitcoin, Ohio had better brace itself for a stampede of Bitcoin payments.
If there's any practical gap, this might be a cynical move on the part of some wealthy Ohio speculator with political connections.
Most likely this is some person who scored a Bitcoin windfall and now wishes to return to an actual cash position at face value, who increasingly anticipates a Bitcoin implosion any day now.
Wrong.
Next contestant, please.
You're assuming the cache has not parity or ECC mechanism where active use would eliminate single-bit errors before they accumulate into undetectable errors, whereas pickling the cache in quiescent warm brine would not.
What you mean is that too many studies are underpowered, so you get a lot of false positives over the 0.95 bar.
We can fix this by making every study 4 to 10 times more expensive (everybody gets a huge N, except perhaps in the cases where the study population is finite, regardless of budget).
Would we actually be better off, or have you just made science twice as expensive as it really needs to be?
Another implicit assumption is that if a scientist (or team) publishes a paper that later fails to replicate, that all the money invested in the study was completely wasted (government waste is by tradition consigned to money-up-in-flames category, including tax revenues that the government immediately pays out again as citizen benefits, merely because some government official touched it—COOTIES—on the way through).
The every-failure-is-a-100%-writeoff presumption isn't true in Silicon Valley (many failed entrepreneurs learn invaluable lessons) and it isn't true in science, either.
You're not reading enough Jordan Peterson. What women want (from their male romantic partners) is competence. Women generally favour a particular competence signal: societal attention / societal approval (if the attention signal is large enough, the approval half of the signal is irrelevant). Commanding the center of societal attention is a universal proxy for power. All forms of wealth command attention: conspicuous wealth, foolish wealth, (mostly) concealed wealth, grasping wealth, tight-fisted wealth. You will never sleep alone (nor without one hand on your pocket book).
Celebrity is almost by definition a top-drawer competence proxy.
For some reason, one of our core social heuristics is to emulate role models (those with power) by aping their behaviours, even the cynical behaviours we know they put on merely to exploit the unwashed.
I think the problem here is that a celebrity figure's ability to get away with these cynical behaviours in plain sight amplifies the power signal more than enough to compensate for the blowback signal of self-interested disgust at overlord overreach. (The balance likely depends on your own self-esteem curve.)
Then there's this other problem: a part of our brain is wired to presume that the mass behaviour of a billion people with low self-esteem filters can never be wrong.
Most people have more power than they realize. The problem is that you can't directly witness the loops of cause and effect.
If more people possessed explanatory depth—discussed at length in The Knowledge Illusion (2017) by Steven Sloman—it wouldn't be possible for know-nothing influencers to hijack commercial success.
From the corporate side, given a choice between customer A with explanatory depth, and customer B without explanatory depth, you might do fine with customer A in a B2B setting, but you definitely want customer B in a B2C setting.
Given enough power, you can actually shift the balance in the population at large from type A dominance to type B dominance, as Apple has done so successfully since introducing the first iPhone (Apple was long attempting to ride this dragon, and nearly bankrupt itself in a fire while doing so, but then the rewards were spectacular once the dragon finally took flight).
I've been reading review after review about how the 2018 Mac mini basically appeals to no-one with any vestige of explanatory depth, except under the general category of "well, if it's sufficiently inconvenient to escape the Apple tent altogether, this is the cheapest way to cling to the outskirts, and might even qualify as a reasonable purchase if your work load particularly needs compute, but never pushes a pixel at an animated frame rate; or you've got some kind of weird office aesthetic where having a shiny little recycled aluminum box with all kinds of crap hanging off the back on short and expensive interconnects is your idea of a glamorous rat's nest".
The Apple product literature doesn't even supply anyone with explanatory depth so much as a Bathroom Reader of Mac mini technical disclosure. I had to find some obscure enthusiast forum and wade through a hundred Comments of Dreck to find out that the 128 GB SSD option has half the write throughput of the 256 GB SSD option (with another increment at 512 GB, but not nearly so substantial, especially at the GIANT cost increment).
There's no online block diagram at even the highest level of the T2 chip. There's a block diagram of the marketing department's view of Alpine Ridge from about three years ago, that does as much to confuse as to reveal. Silicon vendors like Newark sell some of the older Intel TB3 parts, but not a single one of these listings comes w
Sandbagged Sheryl Sandberg needs a plan to gain control over what's actually happening in the company she runs, right under her nose.
Failure to execute such a plan means exactly what you think it means.
My faith in the scientific model of dire anthropogenic climate change started at about 30% in 1980, and has increased by about 10% per decade since.
In any other area of science, this would be regarded as a rapid confidence ramp with respect to a wicked question of science, but stakes are very high with the planet hanging in the balance, so double standard.
My faith in the dire economic projections is not ramping at anywhere near that rate. It's mostly just a giant FUD sundae, and neither the rigors of traditional peer review nor the actual credentials wielded have so far managed to impress me.
Without doubt, Bad Shit Could Happen.
But to construe this debate as a science debate, one where the scientific consensus is automatically credible, simply boggles my mind.
There is a scientific nucleus to this debate: if these climate changes continue Things Would Have to Change.
Yes, they would—but they were going to change anyway, because change in the modern world is simply a given (and the established rate of change is pretty much in constant acceleration, too).
But These Changes Are Really Bad Changes We Shouldn't Go Anywhere Near.
Is that economics, futurology, ideology, or philosophy? Or some unspecified mixture of all four?
And my rebuttal: quite possibly, the wrenching changes to existing conventions of human society to avoid these environmental outcomes are also Really Bad Changes We Shouldn't Go Anywhere Near (because this, too, is highly experimental terrain, and everyone can see how our recent political foray into social media is just peachy).
There are existential risks on all sides of this.
Scientists: Well, if you're too stupid to do what we're telling you to do without starting a food fight (complete with mushroom clouds), don't blame us.
Maddening horde: If we can't blame you when the prescribed giant change initiative all goes sideways, why should we listen in the first place?
Scientists: Because we're right about the science.
Maddening horde: Well, perhaps that makes you feel good, because what you care about most of all is the science, but it doesn't do jack shit for the rest of us, who don't regard one self-inflicted bad outcome as automatically better than any other self-inflicted bad outcome.
s/either/neither
The following is not science:
Whether climate change is driven (in part or in full) by human activities is a science question and will ultimately be decided by scientific criteria (exactly when such a reliable decree comes down the pike remains open to debate, though increasingly less open, year over year, year after year, on current trends).
Extrapolating the economic cost of heat stress ventures deep into the dark, unreliable heart of economics (which is referred to as "the dismal science" for a good reason) is a mug's game.
Even glib, prognosticating economists can't make these projections without taking into account future human ingenuity. Economics with a side-order of futurology—what could possibly go wrong?
I'm getting ever more grumpy about this constant bait and switch: we're really, really, really confident of our climate model this time, so let us now describe with infinite confidence the future we envision using our entirely non–Magical 8 Ball (this involving two additional academic fields, one barely respected, the other openly derided; moreover, the vast majority of the dignified-to-a-fault, hard-science authors of this publication have any specific training—or tainting—in either of these shabby fields, and that's why you should believe every word we write).
Until you point out it's an incurable, highly contagious hemorrhagic fever with a brief 48-hour incubation period during which it expresses no tell-tale symptoms, whatsoever.
When your grade three teacher told you about Africa being "over there", her own education was already ten years out of date.
Just wait until all those Pennsylvanian coal miners find out what "globalization" really means.
All it takes is one tainted kilo of weed delivered by Swift Boat in the dark of night.
What does it take to upgrade this to the solar system's twin? Or will these solar-twin systems forever remain half sisters?
Yoda contains more wisdom about the Grand Mastery of prudent delay in the green fingerclaw of his pinky hand-toe than you will ever even suspect.
This is simply not true.
Back when we discovered penicillin, we lied to ourselves in our giddy triumph over the microbial world about what mastery of global antibiotic use actually entailed, so we blindly painted ourselves into a terrible predicament in Alpha Quadrant Superbug.
That's exactly the kind of low-hanging fruit a hunchbacked woman with a gnarled crook hands you in a meadow clearing in the deepest depths of some dark, German forest (or British wardrobe—what passes for a forest on a barren island).
"Help yourself to another square of Turkish Delight, Edmond, it will make you immmmuuuuuuuuune!"
A hundred years passes with Edmond mostly staggering around in a dark pall, but then Edmond finally gets seriously woke to the human microbiome (and, indeed, gloriously complex microbiomes everywhere, if only we'd bothered to investigate with adequate tools).
Personally, sign me up for Edmond 2.0, because Edmond 1.0 was knee-deep in clusterfuckage, and barely had the wits to even know it.