The relevancy was pretty clear: for the innumerate general public to have a meaningful dialogue with a data scientist without the data scientist having to bend over backwards into trite, kindergarten narratives. Find me a working data scientist who can't do the vast majority of these things upside down and underwater, and I'll show you a horse that can add by stamping its feet.
I basically didn't put anything on that list I didn't already grasp in a deep way by the time I reached grade 8 or 9.
The premise of this Slashdot story was that data scientists need to talk to the 60% (and that it's their fault if they can't). And I thought someone needed to point out the immense difficulty involved in meeting halfway.
It becomes an issue when you have, say, a very intelligent data scientist who has put together an analysis, but doesn't know how to communicate it.
For the most part, that's a complete canard.
60% of the adult American population belongs to the arithmetic Special Olympics: * couldn't solve a quadratic equation * couldn't integrate x * couldn't differentiate x * couldn't explain why anyone would ever add two logarithms * couldn't factor 1050 into primes without several mulligans * couldn't check a calculation by casting out nines * couldn't explain the significance of the law of large numbers * think that the "Bell" curve was invented by Alexander Graham Bell * think that "Bay's" rule concerns the golden ratio of cuts to cut-offs * think that tariffs aren't paid for by the end consumer * and don't even get them started on randomness or correlation.
And it's the cossetted research scientist who can't communicate?
The innumeracy gap is real, and it's spectacular.
But sure, you can add a few extra courses to their already intense course load to help them best explain the paintings of M. C. Escher to a congenitally blind man.
Data scientist: "You see, it's about perspective... "
Now the blind person believes that he or she has perspective, only in no way does it resemble the "perspective" under discussion.
Yeah, Microsoft used to pack lead foil into their hockey gloves. Best case, you get a concussion; worst case, you shuffle off this mortal coil.
Google raps the unprotected part of your arm above the glove with their stick, when they think the referee is looking away. You get a bruise, and continue to play with your head up.
Brought to you by the Encyclopedia of False Equivalency (and a stick tap from a butter-soft fast path that might very nearly trip over its own skates if they so much as repainted the blue lines—in fairness, though, that can actually change the texture of the ice for a few hours afterwards, and speed daemons are finely tuned).
It's hypocrisy. If women are attracted to something on men, well that's okay to show (everything but the genitals!), but if men are attracted to something on women, why, that must be hidden!
No, it isn't hypocrisy.
Adult human males get a potent shot of dopamine when gazing upon female breasts of optimal nubility (this recalibrates itself depending on circumstance). It's the breast, not the nipple in particular. The nipple conventionally designates the sharp transition from hydrocodone to fentanyl; women expose portions of their breasts to signal availability, in a general sense, but are notoriously coy about exposing their actual nipples, which signals availability right here and right now.
The perky-nippled apparition of Ms Right Now drives men who have difficulty gaining feminine attention by virtue of being successful, or competent, or cultured, or debonair, into paroxysms of boorish behaviour, which easily tilts into degrading oneupmanship and outright misogyny.
The male nipple, lacking the underlying breast (did I have to explain this?) does not invoke this potent hormonal and cultural response.
This is the paradox of SJW vulgaris. First they dope their own biological cluestick with a Quaalude frappuccino, then they run around noisily "waking" other people up (typical Quaalude culture: users spend most of their time assuring each other that they're still fully in the "woke" condition, because their perceptual vigilance hangs by a knife edge).
Of course, the difficulties here are not to be underestimated. Men and woman both have primary genitals, for which there is not much of a double standard; men and women both have secondary sexual characteristics (men being hairier—and mostly terrified of hot wax—find themselves restricted to swimsuits exposing far less butt cheek); but men and women do not both have primary reproductive organs wagging or sagging over and above their genitals. Observe carefully, and you might see this too.
SJW vulgaris: "Waaaa! Biology sucks."
Which is sad, because there's so legitimate crap out there to complain about. Indeed, it's a target-rich environment.
But no, job number one is to pretend that biology has no legitimate role in the human condition, because then the dialog would require nuance, and you couldn't just run around with your "woke" machine gun, spraying everything in sight like some solitary, embattled Doom marine (albeit, a solitary Doom marine who has mastered the fine art of laying down a backpedalling strafe-turn while dangling a fallback chainsaw from one side of his tool belt and a fallback rocket launcher from the other side, a feat of ultimate macho dexterity which throws serious shade on bounding over sewer gratings in stiletto heels, if you ask me).
Which is one reason why I stick to Firefox, until it becomes entirely unusable. We've had this problem before with IE and we didn't learn from it?
Absolutely. So often, the masses jumping onto the latest and greatest often turns into a form of collective free riding—ultimately a very expensive form of collective free riding.
Fortunately, there's still 9% of us who can read the Middle Earth chess board, see the monster pawn-break boding future ills, and are willing to batten the circular hatches by stubbornly sticking it out with second best (which wouldn't be second best, but for all the rest, easily duped into smacking over a chromium-tainted apple turnover).
Maintaining diversity is often a big deal in machine learning. Your algorithm is less than entirely impressive if it stockpiles the seedcake pantry of the perfectly round hobhole door with a million shotgun-disrupted copies of GMO monoculture microdiversity.
On the other side of this, I pretty much only use YouTube for video. It's the only service that reliably plays on my BSD desktop systems under Firefox (without my faffing around in tetchy AV plumbing I would prefer to never visit again).
Now isn't that an irony worth smoking over on your happy hobbit hill?—where never the unadventuresome masses welcome Saruman's pastry wagon to Hobbiton with linked-arms to partake in his latest hasty scheme.
"Hobbit sense," sighs Gandalf, we're the worse off without it.
Not to sound too disparaging. But Open Source is often the waste bin of dead technology.
Interesting place to put a period. And a weirdly bioinformatics slant on the technological life cycle.
I don't know about yours, but my copy of the Pythagorean theorem remains as rust-free as it ever was.
There are good ways, however, to incorporate intrinsic vice into your code base, to make sure that once it goes into the trash heap, it stays on the trash heap. One approach to this problem is to write your application on top of MFC. Windows citizenship badges are not printed on acid-free paper like all the best computer-science reference works.
A good ISA that's well tuned to its transistor budget is simply not the same beast.
Apparently, Microsoft's goal was to create a fast rendering path that only worked on the one site they wished to brag about, and only if that one site never changed. "Good grief," they all whispered among themselves, "if we're forced to make our fast path robust we'll never climb this mountain fast enough to overtake competent competition".
If Google inserted custom code into Chrome with the only function of ignoring a hidden empty div, then I might enlarge my tiny violin to the manly scale of Schroeder's baby grand. After DR-DOS, it shrunk so small that my personal Jiminy Cricket hauls it out only when he needs a good mosquito repellent. I've got one earlobe that hasn't been bitten, yet.
Netscape had to contend with random and erratically documented behaviour from the entire operating system they ran on top of. One suspects that just one of those old Netscape greybeards from the 1990s could log roll the entire Edge team all by himself. While drinking scalding hot tea from bone china with those dainty handles—and not spilling a drop.
"Sayonara", of course, but I guess my spelling checker (i.e. typo extractor) doesn't reliably activate in the subject line.
I hesitate to call it a spelling checker straight up, because my typo extractor sure doesn't spell any better than I do, if you count half the valid words it still doesn't know. I've been patiently training it for ten years, and just now I had to add "irreproducibility" despite it knowing both "reproducibility" and "irreproducible". Dumb as a bag of hammers, truth be told.
File under things you spot in your rear view mirror, while hastening out the door.
In Amazonian lore, the long tail is the killer Godzilla of lost leaders.
You come for the long tail, you leave with a flying carton filled with All the Usual Retail Suspects (AuRS).
And now here comes Bezos all in a huff, treading on his own tail after a sharp 180, having finally nosed his way to the realization that dragging a long, flashy appendage along in the dirt behind the poop orifice was never a genius design in the first place.
Most of my typos are full-word substitutions: "what else must also change" turned into "what else much also change" when my "time to eat your yummy freshly baked bread" oven-timer went off mid-sentence, causing the ch from 'change' to subconsciously channel David Bowie, by the all-too-alluring lizard logic.
The easiest way to debunk this kind of naive futurism is to postulate what else much also change.
Right now we're in a time of tremendous asymmetry, where the vast majority of computer serves against the explicit interests of the end user. You know, you've got a life plan to make something of yourself, and the Internet says "hey, dude, why don't you click on these artfully extended boobies instead (we know you want to)". But you don't want to, just a tiny little bit of your lizard brain craves a short-term dopamine hit. The less you feed your lizard, the easier it becomes to tune out distraction and make something out of your life.
Until this dynamic is fixed, ambient computing is for schmucks only. Guaranteed, the further you fall into the ambience, the more your lizard brain is carved up by the ad auction of least customer thriving give-a-shit.
Wake me up when ambient computing serves to manage unwanted provocations of our lizard brains, like a good Jiminy Jarvis.
Because the smart money won't be voluntarily boarding good ship Ubiquitous Titillation on present terms.
Microsoft's culture has historically been several grades beyond mere NIH, something more akin to "not warheaded here syndrome".
With Java, Microsoft was neither slow nor reluctant to slip in the payload package, and pretty soon Java was reduced to a "write once, debug everywhere" programming language that Microsoft could truly count as one their own.
The Intel 80386 is a 32-bit microprocessor introduced in 1985.
The first versions had 275,000 transistors and were the CPU of many workstations and high-end personal computers of the time.
Direct comparable:
The Motorola 68020 is a 32-bit microprocessor from Motorola, released in 1984, with approximately 190,000 transistors.
The 68020 had 32-bit internal and external data and address buses, compared to the early 680x0 models with 16-bit data and 24-bit address buses.
The 68020's ALU was also natively 32-bit, so could perform 32-bit operations in one clock, whereas the 68000 took two clocks minimum due to its 16-bit ALU.
The 68EC020 lowered cost through a 24-bit address bus.
The 68020 was produced at speeds ranging from 12 MHz to 33 MHz.
People I ran with back then from the Waterloo Computer Systems Group (to later become famous for the Watcom C/C++ compiler) felt that Nat Semi had the cleanest ISA at the time.
In 1985, National Semi introduced the NS32332, a much improved version of the 32032.
From the datasheet, the enhancements include "the addition of new dedicated addressing hardware (consisting of a high speed ALU, a barrel shifter and an address register), a very efficient increased (20 bytes) instruction prefetch queue, a new system/memory bus interface/protocol, increased efficiency slave processor protocol and finally enhancements of microcode."
There was also a new MMU, FPU and the (very rare) NS32310 interface to a Weitek floating point accelerator.
The aggregate performance boost of the NS32332 from these enhancements only made it 50 percent faster than the original NS32032, and therefore less than that of the main competitor, the MC68020.
But it didn't really matter at the end of the day, because the screens were cramped, disk drives were tiny, memory cost a fortune, C89 didn't exist yet, the modems were dog slow, and there were few places to dial (not on long distance) unless you lived right in the thick of things.
On that account, pretty much every personal computer prior to 1990 was ahead of its time.
But you could invest in a magnificently clicky keyboard and keep it for thirty years.
Personally, I'd give up my smartphone before I'd give up my third 23" desktop monitor (two in portrait mode).
I use SMS to organize the use of shared transportation resources with my wife, and to intermingle social errands into my monthly shopping day. I could probably get by quite comfortably on 100 mobile SMS messages a month, no mobile voice service at all, and some kind of VoIP thing at home.
My third desktop monitor is in active use at least 70 hours per week, and heavy use about half of that time (where its sacrifice would severely cramp my work style).
Does Facebook's genie-stuffing operation also extend to Facebook partners whose own security melted down while they were in possession of illicit private-image contraband (and their partner's partners, too, et al and sundry)?
If so, they might want to maintain the CDC on a warm and cozy legal retainer (and the CDC might want to base itself in a larger home city—there are some things Atlantis just can't do).
When you <scarequote>pirate</scarequote> Blade Runner, you're not hurting Ridley Scott or Harrison Ford.
Nice self-serving, incomplete logic.
Do you really think that the negotiating power of the A-list stars and directors is entirely independent of the profitability of the industry as a whole?
As it happens, the stars of yesterday were compensated based on the profitability of yesteryear. Not every economic compensation loop is forward-biased.
It's the young up-and-coming stars who are presently deprived of negotiating power because the industry is less profitable than it might have been if piracy were less of an intrinsic problem. This causal relationship exists in the large as surely as long-term starvation causes weight loss (if you can't afford to pay for a medically trained doctor to perform bariatric surgery, wait a little longer and the worms will do it for free).
Short-term causality is complex, and is riddled with moguls, humps and fluctuations. It's therefore a trivial exercise in motivated reasoning to assert that the broad outer box is far too large to matter, and that any smaller box is far too erratic to understand.
(We absolutely know that the modern food environment is instrumental in the modern obesity epidemic, but to slice this picture any finer is the domain of a thousand rabid hucksters; what passes for consensus is limited to tobacco, trans-fats, chronic indolence, and excessive refined sugar intake—everything else has devolved into a three-ring macro-nutrient circus of differently hued social-media clowns).
So this kind of argument is just completely ridiculous. You can't honestly argue that shrinking the pie doesn't hurt specific individuals who claim a stake in that pie. Even if there are such individuals who are getting shafted at the pie table, that's theirs to judge and juggle, and not yours.
(Turns out, Mortdecaiwas a heist movie, only Charlie Mortdecai was merely a plant, and it was actually Johnny Depp who rode off into the sunset in sole possession of the Brinks truck.)
What you can argue is that there's shenanigans all around. The studios perpetrate all kinds of shit they shouldn't be doing (e.g. Hollywood accounting). Cable companies and networks are some of the brashest corporate oligopolies known to man. Copyright term extension is cynical and destructive of what had started out as a workable economic, cultural, and social compromise. Retroactive copyright extension is beyond farce.
I would be conceptually prepared to slap YouTube's wrists over this, but for this: wake me up when retroactive copyright extension is well and truly off the books (and everything that had been slated to expire at some past date actually has expired).
Meanwhile, my governing attitude amounts to: karma gonna karma.
While any single artistic work remains in a state of retroactive copyright protection (protection that would otherwise have expired), the studios and the networks—and everyone else feeding from this pie, no matter how small their canary cage—have the least and last and lowest claim on progressive cultural outrage.
But I'm not going to lie about the economics. Those marginal impacts surely exist.
You know, Douglas Adams was infamous for fiddling with his language for 50 drafts, but it just dawned on me that instance of "accidentally" really should have been "inadvertently".
It wasn't his finest hour, nor his best quote, either, but even his seconds are not bad.
Netcraft has Slashdot ranked around 50,500th, a bit ahead of jp.match.com and a bit behind linkedin.fr.
My overactive imagination presumes that linkedin.jp and fr.match.com would both have enormously larger traffic shares (but what do I know about the relative power dynamics of wives and mistresses, here and abroad?)
Even so, given that there are now on the order of 100 million registered web sites, the Slashdot effect remains as potent as ever.
The mighty ships tore across the empty wastes of space and finally dived screaming on to the first planet they came across — which happened to be the Earth — where due to a terrible miscalculation of scale the entire battle fleet was accidentally swallowed by a small dog.
— Douglas Adams, The Hitchhiker's Guide to the Galaxy
As intergalactic dog's breakfasts go, though, 50000th is not exactly chump change.
Even in our heyday, we always had an implicit alignment with The Mouse That Roared (and a blood feud with the Mickey Mouse Copyright Term Extension Act, which goes to show, boys and girls, that while can beat juggernaut America from the 1950s, but you can't beat multinational Disney, nighty-night sweet dreams).
Which is exactly why some want all of us so poor that the only thing we can do is spend what little money we get.
If I'm reading you correctly between the lines, that's a fine policy proscription for small minds: anything that smacks on the surface of sticking it to The Man is inherently good, therefore any nuanced understanding of systems theory can go fly a kite.
It's certainly the case that the haves in any political order stick it to the have nots (who generally prove twice as quick to stick it back should the tables turn, but I digress).
The have nots usually have more viable options than they think they have, and get themselves all hung up fighting the wrong battles.
Now that Democrats are poised to take charge in January, there is nothing to stop them from inserting earmarks in the fiscal 2020 spending bills next year, and lawmakers said they are hoping to soon get a piece of the federal spending pie for specific needs in their districts.
"I hope they come back," Rep. Jim Clyburn, D-S.C., who will take over as majority whip next year, told the Washington Examiner. "I was against them ever leaving."
Instead of fawning over Grover Norquist's ridiculous anti-government ideology, we could actually fight a specific battle worth winning.
Personally, its my considered opinion that clue helps.
This paper effectively supports the regulatory intervention of government to ensure accurate public labelling of remarketware of all stripes and sizes.
It's not by any means always a bad thing for flourishing private commerce that government maintains certain forms of caveat emptor in their fiat-powered gun sights. Who, precisely, wants a mode of private commerce where everyone sensible runs around with permanently cinched purse strings?
Moral of the story: be careful what you drown in the bathtub if you value liquid enterprise.
Subtracting t_first_adherent from t_second_adherent and taking the reciprocal to compute the Borgesian uptake of Global Illumination is a violation of the Nyquist–Shannon sampling theorem.
Somewhere in there, the math completely falls apart for year-over-year sample sets of two eager, bounteous measurements, bursting at the seams. (Who's going to knock 100% knowledge inflation? Bah! Humbug! Nyquist–Shannon is all wet.)
This explains one of modern society's growth obsessions. At the second observation, you fit a linear model (that slices through your twin observations with the clingy perfection of Seven of Nine's skin-tight body suit), or no model at all.
If your data set is still monotonic at the third observation, you fit x^k or e^x.
If your data set is non-monotonic after the third observation (drat), you fit ax+b (with error bars) or C*sin(ax+b) or, more likely, D*sin(ax+b) (with no error bars — yay! Seven of Nine's cuspy-cupped dimples of delight).
By here we've already exceeded the mental bandwidth of your struggling, low-status "human interest" journalist, so neither of these elementary (but pleasing) harmonic forms are even considered.
Harmonic structure? Never heard of it. Now get the f(x) out of my way, you're cramping my click flow.
I should note that the improperly maligned P6 was also trashed by a second camp, the assembly language power optimizers, such as Michael Abrash (though I don't recall his complaints, specifically).
The superscalar Pentium was deterministic. You always got the same clock count from the same initial conditions.
But on the P6, the OOO pipeline has it's own complex internal history, and it inserted random bubbles into the pipeline that no-one ever explained.
The problem with a bubble is that it can knock your instruction decode cadence into a different alignment and that could change dispatch order, and then you'd get weird, fluctuating benchmark scores that would be 2.7 IPC on one pass through the loop, then 2.9 IPC on the next pass through the loop.
People who naturally go into this line of work were almost uniformly more irritated that 2.7 != 2.9 than they were impressed that 2.7 >> 2 (the best IPC the Pentium ever achieved).
Daniel Kahneman could have studied this and included it in Thinking Fast and Slow. It's not just Israeli parole judges suffering from an empty stomach who defy rational comprehension, turns out our own tribe is also far from immune.
Slashdot Mirror
The relevancy was pretty clear: for the innumerate general public to have a meaningful dialogue with a data scientist without the data scientist having to bend over backwards into trite, kindergarten narratives. Find me a working data scientist who can't do the vast majority of these things upside down and underwater, and I'll show you a horse that can add by stamping its feet.
I basically didn't put anything on that list I didn't already grasp in a deep way by the time I reached grade 8 or 9.
The premise of this Slashdot story was that data scientists need to talk to the 60% (and that it's their fault if they can't). And I thought someone needed to point out the immense difficulty involved in meeting halfway.
For the most part, that's a complete canard.
60% of the adult American population belongs to the arithmetic Special Olympics:
* couldn't solve a quadratic equation
* couldn't integrate x
* couldn't differentiate x
* couldn't explain why anyone would ever add two logarithms
* couldn't factor 1050 into primes without several mulligans
* couldn't check a calculation by casting out nines
* couldn't explain the significance of the law of large numbers
* think that the "Bell" curve was invented by Alexander Graham Bell
* think that "Bay's" rule concerns the golden ratio of cuts to cut-offs
* think that tariffs aren't paid for by the end consumer
* and don't even get them started on randomness or correlation.
And it's the cossetted research scientist who can't communicate?
The innumeracy gap is real, and it's spectacular.
But sure, you can add a few extra courses to their already intense course load to help them best explain the paintings of M. C. Escher to a congenitally blind man.
Data scientist: "You see, it's about perspective ... "
Now the blind person believes that he or she has perspective, only in no way does it resemble the "perspective" under discussion.
And we're unlucky to have missed Jupiter's rings, which were far more impressive.
Yeah, Microsoft used to pack lead foil into their hockey gloves. Best case, you get a concussion; worst case, you shuffle off this mortal coil.
Google raps the unprotected part of your arm above the glove with their stick, when they think the referee is looking away. You get a bruise, and continue to play with your head up.
Brought to you by the Encyclopedia of False Equivalency (and a stick tap from a butter-soft fast path that might very nearly trip over its own skates if they so much as repainted the blue lines—in fairness, though, that can actually change the texture of the ice for a few hours afterwards, and speed daemons are finely tuned).
Typos happen. That one was obvious, at least to me.
No, it isn't hypocrisy.
Adult human males get a potent shot of dopamine when gazing upon female breasts of optimal nubility (this recalibrates itself depending on circumstance). It's the breast, not the nipple in particular. The nipple conventionally designates the sharp transition from hydrocodone to fentanyl; women expose portions of their breasts to signal availability, in a general sense, but are notoriously coy about exposing their actual nipples, which signals availability right here and right now.
The perky-nippled apparition of Ms Right Now drives men who have difficulty gaining feminine attention by virtue of being successful, or competent, or cultured, or debonair, into paroxysms of boorish behaviour, which easily tilts into degrading oneupmanship and outright misogyny.
The male nipple, lacking the underlying breast (did I have to explain this?) does not invoke this potent hormonal and cultural response.
This is the paradox of SJW vulgaris. First they dope their own biological cluestick with a Quaalude frappuccino, then they run around noisily "waking" other people up (typical Quaalude culture: users spend most of their time assuring each other that they're still fully in the "woke" condition, because their perceptual vigilance hangs by a knife edge).
Of course, the difficulties here are not to be underestimated. Men and woman both have primary genitals, for which there is not much of a double standard; men and women both have secondary sexual characteristics (men being hairier—and mostly terrified of hot wax—find themselves restricted to swimsuits exposing far less butt cheek); but men and women do not both have primary reproductive organs wagging or sagging over and above their genitals. Observe carefully, and you might see this too.
SJW vulgaris: "Waaaa! Biology sucks."
Which is sad, because there's so legitimate crap out there to complain about. Indeed, it's a target-rich environment.
But no, job number one is to pretend that biology has no legitimate role in the human condition, because then the dialog would require nuance, and you couldn't just run around with your "woke" machine gun, spraying everything in sight like some solitary, embattled Doom marine (albeit, a solitary Doom marine who has mastered the fine art of laying down a backpedalling strafe-turn while dangling a fallback chainsaw from one side of his tool belt and a fallback rocket launcher from the other side, a feat of ultimate macho dexterity which throws serious shade on bounding over sewer gratings in stiletto heels, if you ask me).
Absolutely. So often, the masses jumping onto the latest and greatest often turns into a form of collective free riding—ultimately a very expensive form of collective free riding.
Fortunately, there's still 9% of us who can read the Middle Earth chess board, see the monster pawn-break boding future ills, and are willing to batten the circular hatches by stubbornly sticking it out with second best (which wouldn't be second best, but for all the rest, easily duped into smacking over a chromium-tainted apple turnover).
Maintaining diversity is often a big deal in machine learning. Your algorithm is less than entirely impressive if it stockpiles the seedcake pantry of the perfectly round hobhole door with a million shotgun-disrupted copies of GMO monoculture microdiversity.
On the other side of this, I pretty much only use YouTube for video. It's the only service that reliably plays on my BSD desktop systems under Firefox (without my faffing around in tetchy AV plumbing I would prefer to never visit again).
Now isn't that an irony worth smoking over on your happy hobbit hill?—where never the unadventuresome masses welcome Saruman's pastry wagon to Hobbiton with linked-arms to partake in his latest hasty scheme.
"Hobbit sense," sighs Gandalf, we're the worse off without it.
Interesting place to put a period. And a weirdly bioinformatics slant on the technological life cycle.
I don't know about yours, but my copy of the Pythagorean theorem remains as rust-free as it ever was.
There are good ways, however, to incorporate intrinsic vice into your code base, to make sure that once it goes into the trash heap, it stays on the trash heap. One approach to this problem is to write your application on top of MFC. Windows citizenship badges are not printed on acid-free paper like all the best computer-science reference works.
A good ISA that's well tuned to its transistor budget is simply not the same beast.
Microsoft can't keep up with a hidden empty div?
Apparently, Microsoft's goal was to create a fast rendering path that only worked on the one site they wished to brag about, and only if that one site never changed. "Good grief," they all whispered among themselves, "if we're forced to make our fast path robust we'll never climb this mountain fast enough to overtake competent competition".
If Google inserted custom code into Chrome with the only function of ignoring a hidden empty div, then I might enlarge my tiny violin to the manly scale of Schroeder's baby grand. After DR-DOS, it shrunk so small that my personal Jiminy Cricket hauls it out only when he needs a good mosquito repellent. I've got one earlobe that hasn't been bitten, yet.
Netscape had to contend with random and erratically documented behaviour from the entire operating system they ran on top of. One suspects that just one of those old Netscape greybeards from the 1990s could log roll the entire Edge team all by himself. While drinking scalding hot tea from bone china with those dainty handles—and not spilling a drop.
"Sayonara", of course, but I guess my spelling checker (i.e. typo extractor) doesn't reliably activate in the subject line.
I hesitate to call it a spelling checker straight up, because my typo extractor sure doesn't spell any better than I do, if you count half the valid words it still doesn't know. I've been patiently training it for ten years, and just now I had to add "irreproducibility" despite it knowing both "reproducibility" and "irreproducible". Dumb as a bag of hammers, truth be told.
File under things you spot in your rear view mirror, while hastening out the door.
In Amazonian lore, the long tail is the killer Godzilla of lost leaders.
You come for the long tail, you leave with a flying carton filled with All the Usual Retail Suspects (AuRS).
And now here comes Bezos all in a huff, treading on his own tail after a sharp 180, having finally nosed his way to the realization that dragging a long, flashy appendage along in the dirt behind the poop orifice was never a genius design in the first place.
Most of my typos are full-word substitutions: "what else must also change" turned into "what else much also change" when my "time to eat your yummy freshly baked bread" oven-timer went off mid-sentence, causing the ch from 'change' to subconsciously channel David Bowie, by the all-too-alluring lizard logic.
The easiest way to debunk this kind of naive futurism is to postulate what else much also change.
Right now we're in a time of tremendous asymmetry, where the vast majority of computer serves against the explicit interests of the end user. You know, you've got a life plan to make something of yourself, and the Internet says "hey, dude, why don't you click on these artfully extended boobies instead (we know you want to)". But you don't want to, just a tiny little bit of your lizard brain craves a short-term dopamine hit. The less you feed your lizard, the easier it becomes to tune out distraction and make something out of your life.
Until this dynamic is fixed, ambient computing is for schmucks only. Guaranteed, the further you fall into the ambience, the more your lizard brain is carved up by the ad auction of least customer thriving give-a-shit.
Wake me up when ambient computing serves to manage unwanted provocations of our lizard brains, like a good Jiminy Jarvis.
Because the smart money won't be voluntarily boarding good ship Ubiquitous Titillation on present terms.
Microsoft's culture has historically been several grades beyond mere NIH, something more akin to "not warheaded here syndrome".
With Java, Microsoft was neither slow nor reluctant to slip in the payload package, and pretty soon Java was reduced to a "write once, debug everywhere" programming language that Microsoft could truly count as one their own.
Nope.
Direct comparable:
People I ran with back then from the Waterloo Computer Systems Group (to later become famous for the Watcom C/C++ compiler) felt that Nat Semi had the cleanest ISA at the time.
But it didn't really matter at the end of the day, because the screens were cramped, disk drives were tiny, memory cost a fortune, C89 didn't exist yet, the modems were dog slow, and there were few places to dial (not on long distance) unless you lived right in the thick of things.
On that account, pretty much every personal computer prior to 1990 was ahead of its time.
But you could invest in a magnificently clicky keyboard and keep it for thirty years.
Personally, I'd give up my smartphone before I'd give up my third 23" desktop monitor (two in portrait mode).
I use SMS to organize the use of shared transportation resources with my wife, and to intermingle social errands into my monthly shopping day. I could probably get by quite comfortably on 100 mobile SMS messages a month, no mobile voice service at all, and some kind of VoIP thing at home.
My third desktop monitor is in active use at least 70 hours per week, and heavy use about half of that time (where its sacrifice would severely cramp my work style).
Does Facebook's genie-stuffing operation also extend to Facebook partners whose own security melted down while they were in possession of illicit private-image contraband (and their partner's partners, too, et al and sundry)?
If so, they might want to maintain the CDC on a warm and cozy legal retainer (and the CDC might want to base itself in a larger home city—there are some things Atlantis just can't do).
Nice self-serving, incomplete logic.
Do you really think that the negotiating power of the A-list stars and directors is entirely independent of the profitability of the industry as a whole?
As it happens, the stars of yesterday were compensated based on the profitability of yesteryear. Not every economic compensation loop is forward-biased.
It's the young up-and-coming stars who are presently deprived of negotiating power because the industry is less profitable than it might have been if piracy were less of an intrinsic problem. This causal relationship exists in the large as surely as long-term starvation causes weight loss (if you can't afford to pay for a medically trained doctor to perform bariatric surgery, wait a little longer and the worms will do it for free).
Short-term causality is complex, and is riddled with moguls, humps and fluctuations. It's therefore a trivial exercise in motivated reasoning to assert that the broad outer box is far too large to matter, and that any smaller box is far too erratic to understand.
(We absolutely know that the modern food environment is instrumental in the modern obesity epidemic, but to slice this picture any finer is the domain of a thousand rabid hucksters; what passes for consensus is limited to tobacco, trans-fats, chronic indolence, and excessive refined sugar intake—everything else has devolved into a three-ring macro-nutrient circus of differently hued social-media clowns).
So this kind of argument is just completely ridiculous. You can't honestly argue that shrinking the pie doesn't hurt specific individuals who claim a stake in that pie. Even if there are such individuals who are getting shafted at the pie table, that's theirs to judge and juggle, and not yours.
(Turns out, Mortdecai was a heist movie, only Charlie Mortdecai was merely a plant, and it was actually Johnny Depp who rode off into the sunset in sole possession of the Brinks truck.)
What you can argue is that there's shenanigans all around. The studios perpetrate all kinds of shit they shouldn't be doing (e.g. Hollywood accounting). Cable companies and networks are some of the brashest corporate oligopolies known to man. Copyright term extension is cynical and destructive of what had started out as a workable economic, cultural, and social compromise. Retroactive copyright extension is beyond farce.
I would be conceptually prepared to slap YouTube's wrists over this, but for this: wake me up when retroactive copyright extension is well and truly off the books (and everything that had been slated to expire at some past date actually has expired).
Meanwhile, my governing attitude amounts to: karma gonna karma.
While any single artistic work remains in a state of retroactive copyright protection (protection that would otherwise have expired), the studios and the networks—and everyone else feeding from this pie, no matter how small their canary cage—have the least and last and lowest claim on progressive cultural outrage.
But I'm not going to lie about the economics. Those marginal impacts surely exist.
You know, Douglas Adams was infamous for fiddling with his language for 50 drafts, but it just dawned on me that instance of "accidentally" really should have been "inadvertently".
It wasn't his finest hour, nor his best quote, either, but even his seconds are not bad.
Netcraft has Slashdot ranked around 50,500th, a bit ahead of jp.match.com and a bit behind linkedin.fr.
My overactive imagination presumes that linkedin.jp and fr.match.com would both have enormously larger traffic shares (but what do I know about the relative power dynamics of wives and mistresses, here and abroad?)
Even so, given that there are now on the order of 100 million registered web sites, the Slashdot effect remains as potent as ever.
— Douglas Adams, The Hitchhiker's Guide to the Galaxy
As intergalactic dog's breakfasts go, though, 50000th is not exactly chump change.
Even in our heyday, we always had an implicit alignment with The Mouse That Roared (and a blood feud with the Mickey Mouse Copyright Term Extension Act, which goes to show, boys and girls, that while can beat juggernaut America from the 1950s, but you can't beat multinational Disney, nighty-night sweet dreams).
If I'm reading you correctly between the lines, that's a fine policy proscription for small minds: anything that smacks on the surface of sticking it to The Man is inherently good, therefore any nuanced understanding of systems theory can go fly a kite.
It's certainly the case that the haves in any political order stick it to the have nots (who generally prove twice as quick to stick it back should the tables turn, but I digress).
The have nots usually have more viable options than they think they have, and get themselves all hung up fighting the wrong battles.
Return of pork barrel politics? Democrats plot to revive earmarks — 13 Democrats 2018
Instead of fawning over Grover Norquist's ridiculous anti-government ideology, we could actually fight a specific battle worth winning.
Personally, its my considered opinion that clue helps.
The Market for Lemons (1970) never gets old.
This paper effectively supports the regulatory intervention of government to ensure accurate public labelling of remarketware of all stripes and sizes.
It's not by any means always a bad thing for flourishing private commerce that government maintains certain forms of caveat emptor in their fiat-powered gun sights. Who, precisely, wants a mode of private commerce where everyone sensible runs around with permanently cinched purse strings?
Moral of the story: be careful what you drown in the bathtub if you value liquid enterprise.
Subtracting t_first_adherent from t_second_adherent and taking the reciprocal to compute the Borgesian uptake of Global Illumination is a violation of the Nyquist–Shannon sampling theorem.
Somewhere in there, the math completely falls apart for year-over-year sample sets of two eager, bounteous measurements, bursting at the seams. (Who's going to knock 100% knowledge inflation? Bah! Humbug! Nyquist–Shannon is all wet.)
This explains one of modern society's growth obsessions. At the second observation, you fit a linear model (that slices through your twin observations with the clingy perfection of Seven of Nine's skin-tight body suit), or no model at all.
If your data set is still monotonic at the third observation, you fit x^k or e^x.
If your data set is non-monotonic after the third observation (drat), you fit ax+b (with error bars) or C*sin(ax+b) or, more likely, D*sin(ax+b) (with no error bars — yay! Seven of Nine's cuspy-cupped dimples of delight).
By here we've already exceeded the mental bandwidth of your struggling, low-status "human interest" journalist, so neither of these elementary (but pleasing) harmonic forms are even considered.
Harmonic structure? Never heard of it. Now get the f(x) out of my way, you're cramping my click flow.
Awesome post. Best drool-slap ever.
I should note that the improperly maligned P6 was also trashed by a second camp, the assembly language power optimizers, such as Michael Abrash (though I don't recall his complaints, specifically).
The superscalar Pentium was deterministic. You always got the same clock count from the same initial conditions.
But on the P6, the OOO pipeline has it's own complex internal history, and it inserted random bubbles into the pipeline that no-one ever explained.
The problem with a bubble is that it can knock your instruction decode cadence into a different alignment and that could change dispatch order, and then you'd get weird, fluctuating benchmark scores that would be 2.7 IPC on one pass through the loop, then 2.9 IPC on the next pass through the loop.
People who naturally go into this line of work were almost uniformly more irritated that 2.7 != 2.9 than they were impressed that 2.7 >> 2 (the best IPC the Pentium ever achieved).
Daniel Kahneman could have studied this and included it in Thinking Fast and Slow. It's not just Israeli parole judges suffering from an empty stomach who defy rational comprehension, turns out our own tribe is also far from immune.