The grey beards on Slashdot continually bring this up. I have this yet to be an actual problem, ever. Are you guys that inept at copy and pasting?
In years past, I often expressed a similar sentiment on threads trashing C++. The strawmen of the haters are the rarely the authentic bogie men of the proficient. Part of this is the self-selection effect, which is endlessly underestimated.
Half of all my problems with C++ in that phase of my career were due to not having this talk fully codified: CppCon 2014: Jon Kalb on exception-safe code, part I. Few of the oft-mooted C++ boogie men concern me the least. The segment in this Kalb video beginning at 13:20 explains precisely why your father's C++ still sucked, as perceived by an actual user.
For the matter at hand, as it happens, I do have fairly limited copy & paste skills using vi, which I actively dislike, but often find myself using for quick and dirty jobs at the command line.
For quick and dirty hacks at the command line to installed packages I usually align all my code insertions to column 1. This makes them easier to spot while I'm thrashing around, and easier to back out again, later, once the dust settles. Except, of course, in Python, which uniquely rejects my long-established survival toolkit.
Typical scenario: you're installing an unfamiliar piece of software, written in some scripting language, and you can't even figure out which sub-module is failing (due to a WTF error message, which is probably some lazy-ass stack dump), so you hack the source of the officially installed scripts a bit to add print statements, or remove subsystem calls, etc. in a diagnostic capacity, until you've figured out the lay of the land well enough to begin to Google an answer.
Yes, there are other ways to structure this work—all of which tend to elevate thrashing around to the level of actual software development. Man, some people out there must have extremely slight contact with the unfamiliar, who can afford to elevate every minor obstacle or setback to a formal deployment process. (So you wanted a quick solution, and now you're diving into Ant or Maven or Grunt or Ninja or Waf or Gulp or Jenkins or CruiseControl.)
Of course, once your Python script is fully imported into your favourite IDE, and wired up to the project's customary build script, and source controlled in Git, there would be no further reason for me to use my old column 1, quick-and-dirty hack at the command line fu. (If one supposes that, while installing Gulp for the first time ever, my original problem—too much balky, unfamiliar software—doesn't recurse into hyperspace.)
So I ask myself, why does this only happen to me? Do I live wrong? Did I commit some atrocity in a past life? Has gaining proficiency with C++ ruined my karma for life and beyond? Clearly, happy, shiny Pythonistas are doing something different than I'm doing, I just can't figure out what it is, or make it a go (or a go to) in my environment.
Didn't use Perl much, never found it much of a challenge, either to read or write, at my low level of usage. Those sigils, however, were always a thirty minute brainfuck while I switched gears. (LISP and APL and Erlang and JavaScript have similar gear-switching overhead.) I switch gears a lot. That bit became truly painful, when added to all the rest. I tend to forgive R, because I can bang out some incredible plots really damn fast after throwing a mental coin at the toll booth.
Has anyone else ever had a Swiss Army knife where your favorite blades change over time, or am I all alone on this axis, too?
Too often the standard of evidence in matters of bias boils down to this:
A circus promises to reform itself, to eliminate the grotesques: the fat woman, the thin man, the midget, the geek, etc.
Rube Sneergasm promptly visits reformed circus, and later reports back to townies that Barnum is smoking crack: quite obviously, the lion tamer is pudgy and diabetic, one of the acrobats still has a distinctively anorexic pallor, and there was an incriminating pile of rough-hewn chicken heads found in a heap behind the mess tent.
Less keenly observed: the nearby kitchen axe hasn't been properly sharpened since the geek was handed his walking papers.
The bottom line: black people, get yourself some more middle names (which you can continue to fill in, long after everyone else gets a head start on the actual exam questions—what, were you expecting a free lunch?)
But consider the upside: Uvuvwevwevwe Onyetenyevwe Ugwemuhwem Osas votes every damn time.
As does anyone named Covfefe before June 2017 (but Covfefe will probably get through the application line-up a lot faster).
All we need now is an algorithm capable of picking out Winona Ryder (as goth-girl Lydia) from a Beetlejuice police line-up, and that about wraps it up for suicide prediction.
Even the most evenly written piece has its bias where it was placed in relation to other material; Front page news on Pg 14 below the fold.
So if the layout were performed by random coin flips, by your logic, it would still be biased? By this standard, scaling a photograph from large to medium also commits an act of bias. But then, if a summary is not discernibly different than what it summarizes, it's not a summary, is it?
Without compression, comprehension a mile wide and an inch deep is all you have left. Good luck with that.
And within a week of rolling out the test case in Canada, someone will find out to game it and purchase fake ads under a fake name that Facebook will verify as real.
When is the last time you participated in a discussion thread by offering a constructive path forward?
You entire post amounts to "security is difficult". Seriously? That's the dumbest thing I've heard since Trump say "who knew health care could be so complicated?"
So I guess your path forward is to repeal and replace whatever is presently being done, only you're 99% vague on what "replace" would finally amount to (spoiler alert: something else difficult and imperfect), and I'm not sure you're 100% up to speed on the repeal part, either.
You don't even get around to completing your thought. Security is difficult therefore...? Therefore what? Therefore we should all pick up our toys and go home?
Another field of human endeavour with no perfect solutions: romance. Romance is supposed to cure this kind of shallow moaning. Unfortunately, since we invented the Internet, too many people are not even getting to first base. You wouldn't think that romance was something that young adults could find their way around (in large numbers), but evolution has apparently been proved wrong. What can evolution possibly do? It could double the voltage, but then how could we possibly build enough prisons to hold all the sex offenders that would surely result.
Maybe evolution should simply admit that this is an impossible problem and give up. I'm not so sure that's a great idea myself, but to its credit it does have the ring of an idea you've never found personally unsatisfactory, by what I can glean from your post.
Yeah, every young, asthmatic brainiac dreams of going into science for the money. Yeah, every Jewish mother pressures her firstborn to become the next Dr Botany Weedkill.
Unfortunately, even the most highly motivated shirker of the golddigger rat race ultimately learns that money is instrumental to opportunity.
The Faraday inflation of big science. It's a bitch. The old glory days of "hey, Mom, can I have some lemon juice?" are long gone.
The Guardian's house style must be interesting: straight quote for the contraction, curly quote for the plural possessive, and both in the same headline.
Or maybe the curly quote on "billionaires" was just a one-off to remind people of the three-comma asshole in Silicon Valley.
I've always felt the best use of CAPTCHAs was to motivate machine learning.
It has always been a dumb task to ask real people to do, beloved only by those whose business models involve learning something trivial about a small potential bias in a person's purchasing habits, without really knowing anything about the person at all.
Web scale: broad and oh so shallow.
Except for the big fish, who already know everything.
I live further up the west coast, and the probability of "the big one" we all fear is about 0.3% per year (which does not increase much as years go by without such an event happening; there's a whole chapter devoted to the Big Three statistical metaphors in Algorithms to Live By; fractal history approximates no history to a first order).
Furthermore, a big chunk of the "cost" that so worries you is bringing all of the damaged infrastructure into alignment with modern building code and building practice. Sure, it's an unplanned cash outlay (from the category of "unplanned" events that get incessant airplay), but a big chunk of that outlay amounts to capital investment, under duress though it may be.
One can even wonder whether a little creative destruction in the Bay area wouldn't help to ameliorate the Bay-area housing crisis.
The bottom line here is that ecologists have long recognized that humanities global footprint is a lot better with dense urbanization than without.
It does stack a lot of eggs into some fragile baskets (the damn things barely last a century or few), but this tends to go hand-in-hand—in the least brotherly sense—with an unimaginable concentration of wealth, much of it encoded in 1s and 0s, for which the building code is global redundancy, interconnected by such fat pipes that the restoration bandwidth required to salvage a large, badly shaken urban economy is on the order of a few Netflix-hours.
Methinks the lady has a soft spot for her disaster porn.
Not an everyday typo. My fingers betray me in ever more clever ways.
About half of my typing errors are full word substitutions which preserve 80% of the desired keystrokes, though not usually involving words this abstract.
Shifting of the population expression rate of genes already present in the gene pool is such weak tea, I practically have to screen for homeopathy dilution when I read a story submission like this one.
An actual evolution product worth talking about is the capacity of a population to rapidly shift composition to match local conditions.
Ideally, the number of short beaks would remain compatible with the food best exacted with short beaks, while the number of long beaks increases to optimally extract the newly available food source. Or some blended matrix of similar effect. Then, when the feeders all go away (easy come, easy go), it all shifts back again. What an awesome survival skill, that evolution might in some mysterious way have favoured over long megamillennia.
When I adjust the dial on my coffee grinder to suit a new bean, I don't go around calling it evolution. Somehow my ego never got that particular genetic memo.
The grandiose gadflies have long coexisted with the bullshit busters, in fluctuating ratio as social norms evolve. It might be true that a fool is born every minute, but it won't help the grifters much if a grifter is born every thirty seconds. There's a fancy name for a long-term phenotypic equilibrium (ever drifting) which I presently forget. But look around, there are many, all around us.
Case in point: beach weakling has the pick of the marital litter if 90% of the local population is homologous for CAD (and women have any say in the matter at all).
Before the first actual grinding mill came into existence, grain was merely shelled or husked by pounding. This simple kind of a "first break" was effected by spreading the grain upon a slab or block of stone and beating it with a hand stone; a subsequent development of this rude apparatus being a hollow mortar and an improved hand stone. The original hand pounder was used on a flat block... Such relics are found throughout both hemispheres, having been used by all primitive nations throughout the world; but eventually they were universally discarded for more perfect apparatus, which really ground the grain into meal.
That's about the present state of machine learning, the hand-crafting of "features" playing the role of the recently discarded flat blocks.
Wheat is an incredible dietary resource, with the starch being light enough to transport over long distances, if only one can find a way to remove it (contrast potatoes, only ever transported downhill, if at all, until the invention of steam power). Once upon a time, all food was local, as, too, was starvation (fear the blight).
A better method to mill the world's vast stores of accumulated data is a big deal, even if we remain in the relatively crude era of water-powered stone grinding wheels.
Data is a bit like wheat, it doesn't give up its curvature easily. Too much applied force creates heat and destroys the end product. The applied force must have exactly the right ratio of compressive to shear stress, which only an expert miller can judge. Deep learning is nothing more than a slightly better mill than the one we had before, and it ranks right up there beside becoming slightly better at milling wheat.
The economic value of the curvature we can now hope to unlock is quite large. And probably there's a lot of curvature yet to find that remains inaccessible to current methodology.
Data is oil. Data is also wheat.
By way of contrast, unstructured tetrahedral finite element mesh generation shaves 5% of the metal mass off a milling apparatus that already worked just fine, being just one of ten thousand noisy specializations in the great roil of small improvements where a penny shaved is a penny earned.
In the entire world, fewer than 1000 people have the skills necessary to do unstructured tetrahedral finite element mesh generation.
Nevertheless, apparently a great career option for the metaphorically challenged.
That New York Times book review suggests that Bill Gates and Nathan Myhrvold were deliberately engaged in fraud, and deliberately eliminated anything of value from the book before it was printed.
That was good for a chuckle. The author of that piece is Joe Nocera, way back in 1995, while Nocera was still promoting his new book, A Piece of the Action: How the Middle Class Joined the Money Class (from the same year).
Holy dotcom relic, Batman.
Here's as close as the piece comes to hinting at fraud:
Whatever genuinely interesting thoughts Mr. Gates has about the coming technologies — and I have no doubt that he has plenty of them — he has managed to write an entire book without divulging one.
Probably true enough, but far from fraud, in a buyer-have-the-least-clue world (about three pay grades below buyer beware).
Nocera actually makes some astute points:
What does come through, inferentially at least, is the extent to which Microsoft has been built on Mr. Gates's insights into business rather than into technology. Though he has the pallid, slightly dishevelled appearance of a classic computer nerd, he is nothing of the sort and never has been. He has always been a shrewd and calculating businessman.
I've often said that 80% of Microsoft's innovation was business methods (at least), and 20% technological (at best).
Quite apart from Bill's track record (far from sterling), anyone making this kind of investment faces an almost insuperable problem in demonstrating net benefit. Any sufficiently advanced economy is indistinguishable from a random walk. No, not quite, but to a reasonable first approximation.
Like everything else in life, you tend to get what you measure, so almost any reform that obsesses over demonstrated benefit is tilted toward the technocratic (easier to measure) and away from the historical ideal of liberal education (harder to measure).
In theory, based on compelling research, we should be paying the best teachers in early K12 more than double what we presently pay, but the problem persists that there's an enormously risky net-present-value proposition: that the world will still work the same way fifteen to twenty years later, when these blessed children finally hit their earning stride.
Who wants to mortgage the farm for 15 years in the hopes that robots haven't taken all the lucrative jobs? (With so many people competing for so few jobs left over, what pricing power to labour?)
Apparently—to judge by people voting with their wallets—only a very slender group—for whom wealth is already of secondary concern—is willing to mortgage the farm (which is basically just a third vacation home) to pay for the very best elementary school teachers.
Some people win, therefore no people lose. Nice argument you got there, shame if something should happen to it.
You don't even seem to realize that families are multi-generational affairs.
In a free market, anyone can change their location at any time in response to changing economic conditions. But just try to take your spouse or your social network with you at the same time.
Winner: affluence. Loser: social cohesion.
There, was that actually so difficult to figure out?
They also gave Google kudos where that was deserved, but that doesn't make for very good headlines.
Empirically, one can only conclude that nerds like headlines and summaries that suck moonshite, it's what gives our puny, breathless existence meaning and purpose.
I tend to judge by the worse thing a person or organization won't fix. Unicode is beyond annoying, but the weedy quality of story summaries here (not all of them, but a sizeable proportion) is far and away the worst thing Slashdot won't fix.
Yet Slashdot persists in running under the banner of "news for nerds" so I was ultimate forced to concede my prior conception of a nerd as being someone who hews too close to material factuality for his or her social betterment as a false idol.
That should make people more, not less, likely to buy the original.
Proof by first derivative. Works every time. No dilemma ever.
Er, um, hold the presses.
No Google books: authors control most revenue, no soup for Google.
Partial Google books: one author rats out the other (economically) by signing up. Author who signs up wins, author who holds out loses. Plenty of canned alphabet soup for Google.
Full Google books: Google creams almost the whole of the economic surplus due to better consumption matching, authors left in roughly the same place (though a smaller piece of the whole pie). Cream of truffle soup for Google.
Society usually ends up deciding these matter in the large by a process of fait accompli.
The chief executive officer of Sun Microsystems said Monday that consumer privacy issues are a "red herring."
"You have zero privacy anyway," Scott McNealy told a group of reporters and analysts Monday night at an event to launch his company's new Jini technology.
"Get over it."
McNealy's comments came only hours after competitor Intel reversed course under pressure and disabled identification features in its forthcoming Pentium III chip.
It's so routine that McNealy completely forgot himself in his rush to get their before the fait accompli paint was dry.
The judge decides that the authors have already lost the power game, crosses that cell off the game theory matrix (out of superficial prudence), and then—Lo and Behold—corporate America wins again.
We shoot ourselves in the foot by claiming victory for network effects that aren't network effects.
This is a power heuristic, make no mistake about it. With enough power, no network required (though of course, actually having a network does tend to boost power, as well).
Now if only I had a nickel for every time I've had to suffer through "640K should be enough for anyone" comment I could have been rich for the rest of my life, instead of amused for five minutes.
Give a man some fire, and he'll be warm for a day. Set a man on fire, and he'll be warm for the rest of his life.
It was only a few years ago people were saying that the best Go computers would never beat human players because the game was so much more complex.
And if the person went on to demonstrate that the number of reachable game states in Go vastly exceeds the same in chess, said person was speaking out of his or her butt hole. There's pretty much nothing stupider than a penis fight over the greater vastness, when the smaller vastness already exceeds your accessible light cone.
The curvature of viable game play in chess was fairly well understood, with a dominant term coming from piece assets. The curvature of viable game play in Go was not amenable to the same analysis (Go lacks a piece asset term as such).
Now that we have found a useful curvature, one suspects it will turn out to be a form of curvature that human players will never fully command. Perhaps worse than chess, judging by how fast the computers went from chumps to champs.
We really know very little about the kinds of curvature still hiding in the haystack. The chess curvature was painstakingly hand-constructed over many decades by human experts, typically hamstrung by ludicrously limited machines.
Recently, the vision field was much the same, dominated by painstakingly hand-constructed "feature" recognizers.
Now that computers are powerful enough to extract their own features automatically, we're going to go from expensive haystack exploratory surgery directly to haystack MRI.
This is precisely the reason K12 needs to drop calculus and mandate instead statistical reasoning.
And there should be an entire test along the way where every question involves reading some promotional brochure, preferably sourced from the certification industry (if the documents remain identifiable, even better—perhaps supplied by those entities convicted of fraudulent representation, as part of their penance).
Time for the old 3Rs to step aside, in favour of factual, figurative, and fraudulent—the three modes of cognition essential to modern life.
What do you expect from an industry that lives off gambling?
I expect that an industry offering a product called "gambling" offers a product that resembles gambling, which this does not. This is heads I win, tails you lose. The only uncertainty is the position of the minute hand on the wall clock marking your (rare) upticks in an otherwise constantly downward march.
You're in prison. Everyday, you get to pick a number from 1 to 3. If you guess right, you get fed a nice meal. If you guess wrong, you don't get fed, and you get a punch in the head.
Does this resemble gambling?
I've said this before, but the truth about the modern gambling industry is that it functions as a form of psychology money laundering in the war between the responsible and irresponsible mental systems.
Suppose a gambler found a house where the house take only covered keeping the lights on (a necessary overhead). He'll show up and lose a $100/week for many weeks, before his "lucky" night where he pots a $2500 windfall (which he immediately spends on a round of drinks for all his friends, some blow, and then a pair of Swedish bombshells, which he's already too bombed to fully enjoy).
The alternate plan is to invest $100/week into his retirement savings account. Guess which plan he'll choose? And he'll call it breaking even, more or less. Without the casino, he deposits $100/week in his debauchery fund, which he then blows sky high every six months.
Few debauchers can tolerate having a slow-and-steady fund so named (in an ever-fragile piggy bank), and so they willingly pay a giant house fee for a socially acceptable cover story, and for never having access to equity they can call their own, or available on a schedule anyone prone to busting their balls can predict in advance.
Once the cameras are ubiquitous, the police legacy of he said/she said system abuse (from either side) is permanently put onto a better track.
It's always the case that the most effective deterrents are the ones so effective, they never get used.
Contrary to displays of mind-numbing stupidity (intentional or inadvertent) that one sometimes encounters, one can not cross these "inactive" rows out of the game theory matrix without changing the equilibrium solution.
As for early adoption, probably the first rat onto the floating ship is the rat who determines there isn't going to an immediate differential that basically demonstrates how completely full of shit he was, not all that long ago. Every police force is going to clean up their act over the five year period before body cameras become standard issue, precisely to avoid this acute political embarrassment. Huh. Who would have guessed? Yet another robust game theoretic matrix, reporting for active duty: fear of foreseeable future embarrassment.
Police chiefs are notorious for weathering the kind of embarrassment that they can ultimate succeed in sending to hell.
That demonstrates power.
"We just lie to achieve our conviction rate" lies, however, closer to Putin's pay grade, than to your average Miller's Crossing-era chief of police.
Making a mistake once is nothing. Making a mistake twice, wake up call. Making a mistake three times, hey idiot what are you doing?
The underlying mistake here is classifying wrapping a wide error bar around an inflection point as a "mistake" in the first place, a mistake which you are apparently here to interminably repeat.
The very first time I see the government correctly estimate an inflection point, I'm going into full-on Dick the Butcher mode: "the first thing we do, let's kill all the neoliberals", because surely once government perfects centralized estimation, the neoliberals among us have survived well beyond their Best Before date.
Chances are, not going to happen, confidence level 99.999 etc. %.
Slashdot Mirror
In years past, I often expressed a similar sentiment on threads trashing C++. The strawmen of the haters are the rarely the authentic bogie men of the proficient. Part of this is the self-selection effect, which is endlessly underestimated.
Half of all my problems with C++ in that phase of my career were due to not having this talk fully codified: CppCon 2014: Jon Kalb on exception-safe code, part I. Few of the oft-mooted C++ boogie men concern me the least. The segment in this Kalb video beginning at 13:20 explains precisely why your father's C++ still sucked, as perceived by an actual user.
For the matter at hand, as it happens, I do have fairly limited copy & paste skills using vi, which I actively dislike, but often find myself using for quick and dirty jobs at the command line.
For quick and dirty hacks at the command line to installed packages I usually align all my code insertions to column 1. This makes them easier to spot while I'm thrashing around, and easier to back out again, later, once the dust settles. Except, of course, in Python, which uniquely rejects my long-established survival toolkit.
Typical scenario: you're installing an unfamiliar piece of software, written in some scripting language, and you can't even figure out which sub-module is failing (due to a WTF error message, which is probably some lazy-ass stack dump), so you hack the source of the officially installed scripts a bit to add print statements, or remove subsystem calls, etc. in a diagnostic capacity, until you've figured out the lay of the land well enough to begin to Google an answer.
Yes, there are other ways to structure this work—all of which tend to elevate thrashing around to the level of actual software development. Man, some people out there must have extremely slight contact with the unfamiliar, who can afford to elevate every minor obstacle or setback to a formal deployment process. (So you wanted a quick solution, and now you're diving into Ant or Maven or Grunt or Ninja or Waf or Gulp or Jenkins or CruiseControl.)
Of course, once your Python script is fully imported into your favourite IDE, and wired up to the project's customary build script, and source controlled in Git, there would be no further reason for me to use my old column 1, quick-and-dirty hack at the command line fu. (If one supposes that, while installing Gulp for the first time ever, my original problem—too much balky, unfamiliar software—doesn't recurse into hyperspace.)
So I ask myself, why does this only happen to me? Do I live wrong? Did I commit some atrocity in a past life? Has gaining proficiency with C++ ruined my karma for life and beyond? Clearly, happy, shiny Pythonistas are doing something different than I'm doing, I just can't figure out what it is, or make it a go (or a go to) in my environment.
Didn't use Perl much, never found it much of a challenge, either to read or write, at my low level of usage. Those sigils, however, were always a thirty minute brainfuck while I switched gears. (LISP and APL and Erlang and JavaScript have similar gear-switching overhead.) I switch gears a lot. That bit became truly painful, when added to all the rest. I tend to forgive R, because I can bang out some incredible plots really damn fast after throwing a mental coin at the toll booth.
Has anyone else ever had a Swiss Army knife where your favorite blades change over time, or am I all alone on this axis, too?
Too often the standard of evidence in matters of bias boils down to this:
A circus promises to reform itself, to eliminate the grotesques: the fat woman, the thin man, the midget, the geek, etc.
Rube Sneergasm promptly visits reformed circus, and later reports back to townies that Barnum is smoking crack: quite obviously, the lion tamer is pudgy and diabetic, one of the acrobats still has a distinctively anorexic pallor, and there was an incriminating pile of rough-hewn chicken heads found in a heap behind the mess tent.
Less keenly observed: the nearby kitchen axe hasn't been properly sharpened since the geek was handed his walking papers.
The bottom line: black people, get yourself some more middle names (which you can continue to fill in, long after everyone else gets a head start on the actual exam questions—what, were you expecting a free lunch?)
But consider the upside: Uvuvwevwevwe Onyetenyevwe Ugwemuhwem Osas votes every damn time.
As does anyone named Covfefe before June 2017 (but Covfefe will probably get through the application line-up a lot faster).
Artificial Intelligence Can Now Predict Suicide With Remarkable Accuracy — 16 June 2017
All we need now is an algorithm capable of picking out Winona Ryder (as goth-girl Lydia) from a Beetlejuice police line-up, and that about wraps it up for suicide prediction.
So if the layout were performed by random coin flips, by your logic, it would still be biased? By this standard, scaling a photograph from large to medium also commits an act of bias. But then, if a summary is not discernibly different than what it summarizes, it's not a summary, is it?
Without compression, comprehension a mile wide and an inch deep is all you have left. Good luck with that.
When is the last time you participated in a discussion thread by offering a constructive path forward?
You entire post amounts to "security is difficult". Seriously? That's the dumbest thing I've heard since Trump say "who knew health care could be so complicated?"
So I guess your path forward is to repeal and replace whatever is presently being done, only you're 99% vague on what "replace" would finally amount to (spoiler alert: something else difficult and imperfect), and I'm not sure you're 100% up to speed on the repeal part, either.
You don't even get around to completing your thought. Security is difficult therefore ...? Therefore what? Therefore we should all pick up our toys and go home?
Another field of human endeavour with no perfect solutions: romance. Romance is supposed to cure this kind of shallow moaning. Unfortunately, since we invented the Internet, too many people are not even getting to first base. You wouldn't think that romance was something that young adults could find their way around (in large numbers), but evolution has apparently been proved wrong. What can evolution possibly do? It could double the voltage, but then how could we possibly build enough prisons to hold all the sex offenders that would surely result.
Maybe evolution should simply admit that this is an impossible problem and give up. I'm not so sure that's a great idea myself, but to its credit it does have the ring of an idea you've never found personally unsatisfactory, by what I can glean from your post.
Yeah, every young, asthmatic brainiac dreams of going into science for the money. Yeah, every Jewish mother pressures her firstborn to become the next Dr Botany Weedkill.
Unfortunately, even the most highly motivated shirker of the golddigger rat race ultimately learns that money is instrumental to opportunity.
The Faraday inflation of big science. It's a bitch. The old glory days of "hey, Mom, can I have some lemon juice?" are long gone.
The Guardian's house style must be interesting: straight quote for the contraction, curly quote for the plural possessive, and both in the same headline.
Or maybe the curly quote on "billionaires" was just a one-off to remind people of the three-comma asshole in Silicon Valley.
I've always felt the best use of CAPTCHAs was to motivate machine learning.
It has always been a dumb task to ask real people to do, beloved only by those whose business models involve learning something trivial about a small potential bias in a person's purchasing habits, without really knowing anything about the person at all.
Web scale: broad and oh so shallow.
Except for the big fish, who already know everything.
I live further up the west coast, and the probability of "the big one" we all fear is about 0.3% per year (which does not increase much as years go by without such an event happening; there's a whole chapter devoted to the Big Three statistical metaphors in Algorithms to Live By; fractal history approximates no history to a first order).
Furthermore, a big chunk of the "cost" that so worries you is bringing all of the damaged infrastructure into alignment with modern building code and building practice. Sure, it's an unplanned cash outlay (from the category of "unplanned" events that get incessant airplay), but a big chunk of that outlay amounts to capital investment, under duress though it may be.
One can even wonder whether a little creative destruction in the Bay area wouldn't help to ameliorate the Bay-area housing crisis.
Never waste a good crisis.
10 Good Things We Owe To The Black Death
The bottom line here is that ecologists have long recognized that humanities global footprint is a lot better with dense urbanization than without.
It does stack a lot of eggs into some fragile baskets (the damn things barely last a century or few), but this tends to go hand-in-hand—in the least brotherly sense—with an unimaginable concentration of wealth, much of it encoded in 1s and 0s, for which the building code is global redundancy, interconnected by such fat pipes that the restoration bandwidth required to salvage a large, badly shaken urban economy is on the order of a few Netflix-hours.
Methinks the lady has a soft spot for her disaster porn.
s/homologous/homozygous
Not an everyday typo. My fingers betray me in ever more clever ways.
About half of my typing errors are full word substitutions which preserve 80% of the desired keystrokes, though not usually involving words this abstract.
Shifting of the population expression rate of genes already present in the gene pool is such weak tea, I practically have to screen for homeopathy dilution when I read a story submission like this one.
An actual evolution product worth talking about is the capacity of a population to rapidly shift composition to match local conditions.
Ideally, the number of short beaks would remain compatible with the food best exacted with short beaks, while the number of long beaks increases to optimally extract the newly available food source. Or some blended matrix of similar effect. Then, when the feeders all go away (easy come, easy go), it all shifts back again. What an awesome survival skill, that evolution might in some mysterious way have favoured over long megamillennia.
When I adjust the dial on my coffee grinder to suit a new bean, I don't go around calling it evolution. Somehow my ego never got that particular genetic memo.
The grandiose gadflies have long coexisted with the bullshit busters, in fluctuating ratio as social norms evolve. It might be true that a fool is born every minute, but it won't help the grifters much if a grifter is born every thirty seconds. There's a fancy name for a long-term phenotypic equilibrium (ever drifting) which I presently forget. But look around, there are many, all around us.
Case in point: beach weakling has the pick of the marital litter if 90% of the local population is homologous for CAD (and women have any say in the matter at all).
Thus always both types, in some ratio.
Nice rhetoric—factual statement masquerading as metaphor, for any reader dumb enough to go along for the ride.
The Evolution of the Flour Mill from Prehistoric Ages to Modern Times — 1905
That's about the present state of machine learning, the hand-crafting of "features" playing the role of the recently discarded flat blocks.
Wheat is an incredible dietary resource, with the starch being light enough to transport over long distances, if only one can find a way to remove it (contrast potatoes, only ever transported downhill, if at all, until the invention of steam power). Once upon a time, all food was local, as, too, was starvation (fear the blight).
A better method to mill the world's vast stores of accumulated data is a big deal, even if we remain in the relatively crude era of water-powered stone grinding wheels.
Data is a bit like wheat, it doesn't give up its curvature easily. Too much applied force creates heat and destroys the end product. The applied force must have exactly the right ratio of compressive to shear stress, which only an expert miller can judge. Deep learning is nothing more than a slightly better mill than the one we had before, and it ranks right up there beside becoming slightly better at milling wheat.
The economic value of the curvature we can now hope to unlock is quite large. And probably there's a lot of curvature yet to find that remains inaccessible to current methodology.
Data is oil. Data is also wheat.
By way of contrast, unstructured tetrahedral finite element mesh generation shaves 5% of the metal mass off a milling apparatus that already worked just fine, being just one of ten thousand noisy specializations in the great roil of small improvements where a penny shaved is a penny earned.
Nevertheless, apparently a great career option for the metaphorically challenged.
That was good for a chuckle. The author of that piece is Joe Nocera, way back in 1995, while Nocera was still promoting his new book, A Piece of the Action: How the Middle Class Joined the Money Class (from the same year).
Holy dotcom relic, Batman.
Here's as close as the piece comes to hinting at fraud:
Probably true enough, but far from fraud, in a buyer-have-the-least-clue world (about three pay grades below buyer beware).
Nocera actually makes some astute points:
I've often said that 80% of Microsoft's innovation was business methods (at least), and 20% technological (at best).
Quite apart from Bill's track record (far from sterling), anyone making this kind of investment faces an almost insuperable problem in demonstrating net benefit. Any sufficiently advanced economy is indistinguishable from a random walk. No, not quite, but to a reasonable first approximation.
Like everything else in life, you tend to get what you measure, so almost any reform that obsesses over demonstrated benefit is tilted toward the technocratic (easier to measure) and away from the historical ideal of liberal education (harder to measure).
In theory, based on compelling research, we should be paying the best teachers in early K12 more than double what we presently pay, but the problem persists that there's an enormously risky net-present-value proposition: that the world will still work the same way fifteen to twenty years later, when these blessed children finally hit their earning stride.
Who wants to mortgage the farm for 15 years in the hopes that robots haven't taken all the lucrative jobs? (With so many people competing for so few jobs left over, what pricing power to labour?)
Apparently—to judge by people voting with their wallets—only a very slender group—for whom wealth is already of secondary concern—is willing to mortgage the farm (which is basically just a third vacation home) to pay for the very best elementary school teachers.
Fight, flight, or fornicate: 4chan, Travelocity & match.com.
We thought we were building a WWW, but we ended up with an FFF.
Some people win, therefore no people lose. Nice argument you got there, shame if something should happen to it.
You don't even seem to realize that families are multi-generational affairs.
In a free market, anyone can change their location at any time in response to changing economic conditions. But just try to take your spouse or your social network with you at the same time.
Winner: affluence. Loser: social cohesion.
There, was that actually so difficult to figure out?
Empirically, one can only conclude that nerds like headlines and summaries that suck moonshite, it's what gives our puny, breathless existence meaning and purpose.
I tend to judge by the worse thing a person or organization won't fix. Unicode is beyond annoying, but the weedy quality of story summaries here (not all of them, but a sizeable proportion) is far and away the worst thing Slashdot won't fix.
Yet Slashdot persists in running under the banner of "news for nerds" so I was ultimate forced to concede my prior conception of a nerd as being someone who hews too close to material factuality for his or her social betterment as a false idol.
Material factuality remains a stretch aspiration.
Jason just called. He wants to know if you're his daddy.
Proof by first derivative. Works every time. No dilemma ever.
Er, um, hold the presses.
No Google books: authors control most revenue, no soup for Google.
Partial Google books: one author rats out the other (economically) by signing up. Author who signs up wins, author who holds out loses. Plenty of canned alphabet soup for Google.
Full Google books: Google creams almost the whole of the economic surplus due to better consumption matching, authors left in roughly the same place (though a smaller piece of the whole pie). Cream of truffle soup for Google.
Society usually ends up deciding these matter in the large by a process of fait accompli.
Sun on Privacy: 'Get Over It' — January 1999
It's so routine that McNealy completely forgot himself in his rush to get their before the fait accompli paint was dry.
The judge decides that the authors have already lost the power game, crosses that cell off the game theory matrix (out of superficial prudence), and then—Lo and Behold—corporate America wins again.
We shoot ourselves in the foot by claiming victory for network effects that aren't network effects.
This is a power heuristic, make no mistake about it. With enough power, no network required (though of course, actually having a network does tend to boost power, as well).
That was funny.
Now if only I had a nickel for every time I've had to suffer through "640K should be enough for anyone" comment I could have been rich for the rest of my life, instead of amused for five minutes.
And if the person went on to demonstrate that the number of reachable game states in Go vastly exceeds the same in chess, said person was speaking out of his or her butt hole. There's pretty much nothing stupider than a penis fight over the greater vastness, when the smaller vastness already exceeds your accessible light cone.
The curvature of viable game play in chess was fairly well understood, with a dominant term coming from piece assets. The curvature of viable game play in Go was not amenable to the same analysis (Go lacks a piece asset term as such).
Now that we have found a useful curvature, one suspects it will turn out to be a form of curvature that human players will never fully command. Perhaps worse than chess, judging by how fast the computers went from chumps to champs.
We really know very little about the kinds of curvature still hiding in the haystack. The chess curvature was painstakingly hand-constructed over many decades by human experts, typically hamstrung by ludicrously limited machines.
Recently, the vision field was much the same, dominated by painstakingly hand-constructed "feature" recognizers.
Now that computers are powerful enough to extract their own features automatically, we're going to go from expensive haystack exploratory surgery directly to haystack MRI.
Paradigm shift. It's a thing.
This is precisely the reason K12 needs to drop calculus and mandate instead statistical reasoning.
And there should be an entire test along the way where every question involves reading some promotional brochure, preferably sourced from the certification industry (if the documents remain identifiable, even better—perhaps supplied by those entities convicted of fraudulent representation, as part of their penance).
Time for the old 3Rs to step aside, in favour of factual, figurative, and fraudulent—the three modes of cognition essential to modern life.
I expect that an industry offering a product called "gambling" offers a product that resembles gambling, which this does not. This is heads I win, tails you lose. The only uncertainty is the position of the minute hand on the wall clock marking your (rare) upticks in an otherwise constantly downward march.
You're in prison. Everyday, you get to pick a number from 1 to 3. If you guess right, you get fed a nice meal. If you guess wrong, you don't get fed, and you get a punch in the head.
Does this resemble gambling?
I've said this before, but the truth about the modern gambling industry is that it functions as a form of psychology money laundering in the war between the responsible and irresponsible mental systems.
Suppose a gambler found a house where the house take only covered keeping the lights on (a necessary overhead). He'll show up and lose a $100/week for many weeks, before his "lucky" night where he pots a $2500 windfall (which he immediately spends on a round of drinks for all his friends, some blow, and then a pair of Swedish bombshells, which he's already too bombed to fully enjoy).
The alternate plan is to invest $100/week into his retirement savings account. Guess which plan he'll choose? And he'll call it breaking even, more or less. Without the casino, he deposits $100/week in his debauchery fund, which he then blows sky high every six months.
Few debauchers can tolerate having a slow-and-steady fund so named (in an ever-fragile piggy bank), and so they willingly pay a giant house fee for a socially acceptable cover story, and for never having access to equity they can call their own, or available on a schedule anyone prone to busting their balls can predict in advance.
Once the cameras are ubiquitous, the police legacy of he said/she said system abuse (from either side) is permanently put onto a better track.
It's always the case that the most effective deterrents are the ones so effective, they never get used.
Contrary to displays of mind-numbing stupidity (intentional or inadvertent) that one sometimes encounters, one can not cross these "inactive" rows out of the game theory matrix without changing the equilibrium solution.
As for early adoption, probably the first rat onto the floating ship is the rat who determines there isn't going to an immediate differential that basically demonstrates how completely full of shit he was, not all that long ago. Every police force is going to clean up their act over the five year period before body cameras become standard issue, precisely to avoid this acute political embarrassment. Huh. Who would have guessed? Yet another robust game theoretic matrix, reporting for active duty: fear of foreseeable future embarrassment.
Police chiefs are notorious for weathering the kind of embarrassment that they can ultimate succeed in sending to hell.
That demonstrates power.
"We just lie to achieve our conviction rate" lies, however, closer to Putin's pay grade, than to your average Miller's Crossing-era chief of police.
The underlying mistake here is classifying wrapping a wide error bar around an inflection point as a "mistake" in the first place, a mistake which you are apparently here to interminably repeat.
The very first time I see the government correctly estimate an inflection point, I'm going into full-on Dick the Butcher mode: "the first thing we do, let's kill all the neoliberals", because surely once government perfects centralized estimation, the neoliberals among us have survived well beyond their Best Before date.
Chances are, not going to happen, confidence level 99.999 etc. %.