And his point was that his grandfather didn't like cows, because it was simpler to get more bang for your buck in tight conditions with other livestock. Of course, if you're keeping sheep around for other purposes (the love that dares not speak its name) then you'll certainly find a way to make it work. Love finds a way, as you continue to point out.

There's no such thing as parity with coal, because coal has minimal siting constraints. Whether you site a coal plant in Ecuador or Anchorage, the same physical plant generates the same power on the same schedule.

This isn't true for solar. The cost structure for solar depends intensely on siting factors. This determines how much power you get, and when you get it. Not all time is created equal, not unless you have a battery warehouse the size of Azathoth's nutsack, and that doesn't come for free.

Plus solar is land hungry, requiring about 8–10 kilo-acres/GWac. (See Land-Use Requirements for Solar Power Plants in the United States 2013.)

Which has a bigger footprint, a coal plant or a solar farm?

California's proposed Blythe plant will require a whopping 7,000 acres of Mohave Desert in order to deliver 2,100 GWh per year. The area of a coal plant producing the same output will typically be one square mile (640 acres) or less.

Then it goes on to an entirely brain-damaged comparison of the size of the coal mine required to support the generation capacity of the coal plant. But coal mines are generally found in places not much use for anything else, often far away from urban centers, and once again, it produces a mostly consistent coal output 365 days of the year (and you don't need a battery, because coal has this amazing capacity to sit there in giant piles, as piles were understood long before Volta).

So yes, the environmental disruption might have a large footprint, but we're actually talking about direct economic cost structures, and it doesn't in any way close the overall siting gap. So what this amounts to is that there are now many places in the world that can purchase a solar plant rather than a coal plant, on direct economic cost comparisons, without the addition of any green fairy dust.

But this is not your father's technology curve, where once a price threshold is crossed, it stays crossed. No, it's the other kind of crossed—as in "crossed off"—as the best sites for solar are exploited, and you progress ever further down the list, to less desirable princes and princesses.

It might be the case that progress in solar technology is able to keep pace with the degradation of the situation list. But it's a stupid assumption to make without even noticing that you haven't zipped up your fly, and your chuff is showing.

Such attacks can lead to serious security leaks. For example, finding out that someone's medical data was used to train a model associated with a disease might reveal that this person has that disease.

For fifteen years I visited slashdot twice a day, sometimes more often. In recent years I slowed down to twice a month, as slashdot culture become more like reddit culture. It's hard to put my finger on precisely what this x-factor is, but I know it when I see it.

Am I missing something here, or is the above passage total geek fail?

I mean it's possible that I've reached my silver-sourcer best-before date, and a trillion brain cells died over night, and I've crossed over the Dunning Krueger horizon into the final sunset, which is quite possibly not so different than falling into a gentle black hole of galactic mass, with no real tides at the event horizon to tap you with a clue stick by tearing your body asunder, limb from limb.

According to the brain cells I still retain, whatever their final number, it seems to me that to find out by this method that someone's medical data is inside the model, you need to test this by already having said medical data in hand. If so you can't by this method learn whether the person had the disease. What you can learn is whether they signed the consent form to participate in lending their personal data to the model builders.

While we're stuck here at the level of pushing back against reddit-worthy brain farts, we're not discussing any of the deeper issues. I can get that from any politician anywhere — modern supply now being as near to infinite as to make no difference.

Et tu, News for Nerds? This is so depressing.

I'm still able to find what I want with Google, but only because I switched to 50 results per page as my default search mode. Once you ignore the top 25 results, you find a lot of the stuff that used to come up from the old Google. It helps to have a pair of 23" monitors in portrait mode, alongside my YouTube and Google maps monitor in landscape mode.

No question that Google has been on the down slope for the last five years compared to what it used to be. What will it take after the next "upgrade"? 100 results per page. I shudder to contemplate the Google future these days.

I can't think of a single thing about my own computing environment that I prefer over what was available circa 2009. I hate my browser (Firefox), I hate my Android phone, I hate my wife's iMac, and I only barely tolerate my own BSD box.

The entire industry is no longer built to support fully empowered users. There are five companies now that flirt with a combined $10 trillion market cap. They have shaped computing to their needs and not ours.

There's nothing any of us can do about this now. Firefox, bless their misguided hearts, are pissing into the wind against the trend just like everyone else who dares to think different than what the tech titans have decreed.

You simply don't know that the human brain can not be fooled by single pixels, because we have no way of backing out the error term from the human neocortex to directly construct the pathological example.

Additionally, human recognition does have some surprising failure boundaries, but it tends not to happen at the pixel level. Human vision at the bottom is more oriented toward edges and motion.

Today science news flash: Pixel-based visual system has trivial pixel-based failure boundaries.

Future science news flash: Edge/motion-based visual system has trivial edge/motion-based failure boundaries.

In some alternate physical universe where the beings live in multi-thousand dimensional physical space (counting macroscopic dimensions only), the ultimate Olympic sport is walking a straight line. Damn hard in 2000 dimensions. I personally sometimes lose track of my compass points when I'm walking in the woods. Either that, or the sun really is conveyed by a chariot with nothing better to do than made a mad rush for a different sky quadrant whenever it senses that I'm hiking through thick brush under mild to moderate sky cover.

Imagine being lost in the woods when your 2000-dimensional GPS warbles at you like the chromatic miasma analyzed by Rick Beato: The Most COMPLEX Pop Song of All Time — 23 June 2021

Winning comment in YouTube thread: "Never gonna choose a key, gonna modulate forever." (People who grew up on Name That Tune have no doubt already figured out from this small hint the identity of the "complex" malt syrup in question.)

In 2000-dimensional space, all GPS coordinates resemble befuddled tunes brought to you by Sergio Mendes. Straight lines are vertiginous, and it might actually be better for your stomach if you shamble a bit from side to side.

Take any two gestalts in a high dimensional space. If you can construct a straight line between them, the intermediate value theorem from first year calculus pretty much guarantees a pathological case balanced on a single pixel.

Straight lines are easily enough constructed with large, deterministic matrices. Not so easily constructed from 100 trillion synapses, never frozen in place. This doesn't even begin to factor in any actual additional robustness of the human mind.

You're doing it wrong. Wind power kills birds. It's some substantial number of birds per GW/hour no matter how you slice it.

But there's a simple solution: more nuclear, less wind.

Nuclear dumps a lot of waste heat into the local environment (river, atmosphere). I'm sure this has many effects on wildlife, but we barely know what they are.

Fewer marked bird graves. And we all lived happily ever after.

Actually it just might, in a lot of cases those persons are the bad influences that got them into so much trouble in the first place.

I wonder about the original bad influence who inclined you toward thinking about bad influence in such a hopelessly reductive way.

Or perhaps there was more than one. Or perhaps the sum of your formative influences was greater than the parts. Or perhaps you came to this position on your own steam, and there's no original bad influence to excuse your present behaviour.

From what I've read, among the worst of all bad influences is the absent father: the one who disappears into the sunset, pays no support, and never shows up to offer fatherly support and advice. But we can fix this by arranging the absence of every other family tie at the same time.

The way I see it, if you want to argue over dinner with your crazy uncle, it's your life. Every family has its black sheep. Not that this couldn't be instantly cured with a mass interventionist shunning, to hear you tell the tale. You can't cut the criminal off from his family without also cutting the family off from the criminal. But no worries—by the logic of social justice warriors, taint by association is fair game.

I'm unable to read you post on any other level than the rantings of a crypto electric-chair fetishist (not without seeking secret subtext using wink solvent).

Dead men tell no tales. Starving men tilt no scales.

Dead men commit no crimes. Starving men steal bread because some other nearby bag of bones exerted negative karma.

Moral of the story: If only starved men never conversed among themselves, your apple pie cooling on the window sill would remain safe and unmolested for all eternity.

The fundamental, unspeakable confound in this space is that light skin is more reflective than dark skin, and the camera fundamentally receives more photons on average (in the same photo) from faces with light skin rather than dark skin.

Look up the definition of light and dark, as it was on the Lord's seventh day, long before the invention of grievance studies, and the modern interrogation into whether photons are the primary physical conduit of misinformation (yes, they are—for every distinguished slice of the pie—the primary conduit of information, value judgement aside).

I know I could get nailed to a nearby cross for this—all places are ever and always nearby in virtual space—but it's entirely possible that a picture of a light skinned face is worth 1000 words, but a picture of a dark skinned face is only worth 950 words, simply because fewer photons arrives (holding typical illumination and exposure levels constant).

Deep learning is the ultimate black box, at present much closer to a black hole (steadfastly anal retentive) than a black body (so highly emissive as to be credited with the original rainbow). What we presently don't know about deep learning would fill an entire Library of Congress on the other side of an event horizon. Oh yes, but if you trained on the non-STEM side of campus, fifteen seconds worth of contemplation is enough to shriek "bias" at the top of your lungs (in truth, you were really only waiting to fully inhale again after the last shriek).

There are substantial advantages to a dark complexion. It's not obvious there should be any fair-skinned people in the modern world, given the large problem of skin cancer, and the small problem of vitamin D enriched diets. One could even argue that day one of the Gattaca Revolution should commence with implanting melanin genes in all the world's people, so that sunburn and premature aging of the skin is relegated to a dull roar.

You read about it first on Bloom County back in the mid-1980s.

Oliver learns of the Apartheid system in South Africa. He invents a "pigmentizer", which will temporarily turn a white person black.

Cutter John and Opus are dispatched to Washington to zap the South African ambassador, but their balloon-powered wheelchair crashes into the Atlantic Ocean and they disappear. Though officially listed as "Eaten By Squid", Opus reappears some time later, suffering from such strong amnesia that he initially has no idea he is even a penguin. Eventually the fake news of a secret wedding between Eddie Murphy and Diane Sawyer, Opus' longtime crush, shocks him into recalling what happened.

After drifting for a while between lost islands, using the wheelchair as a raft, Cutter John and Opus were rescued by a Soviet submarine and arrested as spies. In order to rescue him, Steve Dallas meets with Russian envoys to trade Cutter John in for the one thing they want from Bloom County: Bill the Cat.

The pigmentizer is a small hand-held ray gun.

Curiously, the next entry down on the Wikipedia list of Bloom County highlights is the following:

Donald Trump is accidentally and fatally injured by the anchor of his own yacht. Incredibly, surgeons turn to Bill the Cat as a donor body in which to insert Trump's still-living brain.

Trapped in Bill's body, Trump finds himself disinherited from his financial empire and estranged from his wife Ivana. With nowhere else to turn, he takes Bill's place in the Bloom County boarding house, making unsuccessful attempts to start from scratch and occasionally being given equally unsuccessful lessons on the value of life by Opus. This eventually culminates in Trump regaining power and using it to buy out Bloom County, firing the entire staff of characters in the process.

Personally, I'd happily ray gun the entire human population outside the tropics with an mRNA pigmentizer in the water supply, and put legions of dermatologists everywhere out of business tomorrow.

As an intended bonus, this would also fix the Twitter algorithm.

However, one might first pause to inquire also about unintended effects. It could actually turn out to be the case, that with the entire population of Sweden suddenly converted to a fetching shade of ebony, that their blue eyes have difficulty reading each other's facial expressions in bright sunlight. Bereft of the social cohesion accruing to accurate perception of group body language, Sweden could degenerate into a Heart of Darkness political quagmire overnight. We just don't know.

Dark skin: less cancer, less social cohesion due to not so quick recognition of facial expressions.

Light skin: more cancer, more social cohesion due to quick recognition of facial expressions.

Not in the least an unusual evolutionary trade-off.

Can we rule this outlandish, unspeakable hypothesis out a priori? Yes, but only from the grievance studies side of campus. On the STEM side, we stubbornly persist with our flawed humility of not knowing what we actually don't know.

A human doing the task wouldn't have trouble, so neither should an AI.

I took a long Slashdot hiatus recently. Eventually I ran out of groans. It seems I will again soon.

Man, way to undervalue your own brain. You have 100 trillion synapses that form an extremely sophisticated Bayesian prior on everything ordinary under the sun (for phenomena illuminated by precision instruments, the Bayesian prior from 90% of the daylight human population is complete batshit, but we'll ignore this for the moment).

The human brain trains in novel recognition tasks against the background of its pre-existing vast model of the world. An AI system that trains ab initio should not be expected to achieve the same result at a deep level, in no way, shape or form.

We have the same conceptual problem in genetics with the evolutionary process. For the majority of genes in an organism, the evolutionary environment is not the outside world so much as it is all the other genes in the organism. This is where Dawkin's selfish gene falls short: "the" environment is not what you initially think it is, but in practice contains a large recursive term.

The Selfish Gene was written in 1976. If your brain is capable of advanced analogy (see Godel, Escher, Bach from 1979) you would recognize already that deep learning as performed in the neocortex also contains a highly recursive term. Forty years this basic notion has been swimming in the water supply (see Star Wars from 1976) for anyone astute enough to notice.

But if you're willing to multiply your 100 trillion pre-existing synapses by zero, a full forty years after the clue hammer dropped (see The Difference Engine from 1990 for a resounding exposition of an incipient clue-hammer detection field), then so am I–specially after reading your last comment.

Sheesh.

Geek sheeple.

Not the worst kind, but definitely causing the most violence to my esteemed geek prior.

Uh, no to no immune response. There are many immune responses. It's a complicated system. What is they mean is: no immune response of the kind they normally go looking for, ordinarily in normal people.

Perhaps being on immunosuppressants means that looking in all the normal places is the wrong way to skin the cat.

I give it 6 months before Google decides to stop supporting it.

If that makes you slap your forehead so hard you end up with a ruby red palm divot, don't even begin to consider what we've done here to debase +5 insightful, which is how this remark is presently rated.

New addition table:
* 2+2 = +4 funny
* 2+3 = +5 insightful
* 3+3 = space alien technology

I'd wager only C and C++ are older, and in the case of C++ not by much ...

The first Python release wasn't until 1991.

In 1985, C++ had its first official reference manual, and later that same year, it's first commercial compiler.

Maybe that's your definition of "not very much".

But I was there, and more happened in the world of computing during those six years than any six year period that followed.

Although the magazine replied to the reader's proposal with "Please say you're kidding about the bi-weekly schedule. Please?", after the December 1983 issue reached 800 pages in size, in 1984 PC Magazine began publishing new issues every two weeks, with each about 400 pages in size.

In fact, the reason Python succeeded in the 1990s is that C and C++ had by then done so much of the heavy lifting, it was possible to pour an inefficient interpreted language into the cracks.

Crack culture is different than heavy lifting culture. It soon becomes much larger, and the enlarged surface error becomes riddled with more flaws that all the king's eyeballs and all the king's men can isolate in computational ICU inside three layers of protective virtual hull.

Great; what's his idea for a replacement?

For the semantically inclined reader, Jamie is clear about this.

It begins by eliminating all the garbage that annoyed Michael Stonebraker since forever.

My biggest complaint about System R is that the team never stopped to clean up SQL. All the annoying features of the language have endured to this day SQL will be the COBOL of 2020.

It proceeds by implementing a portable core which people are actually willing to use as a portable core, eschewing premature optimization. (For a sufficiently large dataset, one will eventually bypass the portable core, but this doesn't need to be architecturally determined in the database design phase.)

Finally, it suggests dealing with the expressivity problem by incorporating some of the syntactic sugar we've managed to invent for other languages since COBOL roamed the Earth in rainbow coloured T-shirts, on COBOL pride day.

But apparently, you think the road to progress looks like this:
* step into phone booth
* motion-blur air guitar man-panty pole dance
* emerge triumphant
* serve up hors d'oeuvre finger food on a silver platter to every random stranger with the attention span of a mayfly

Solutionism is a form of greedy algorithm, isn't it?

Voice of God film trailer: In a world where no homonid can move the rock without defining the problem and enacting the solution in a single grand gesture, the bulk of civilization continues to languish barely a toe into the Bronze Age.

You've fallen into the pit of parochialism which plagues the entire field of hard AI. You see regularities that the program does not. The program sees regularities that you do not. But not to worry. You have a parochial type theory which places your regularities on a higher rung. No reason. But it feels good.

Egorithm:
* if the mechanical intelligence is similar enough to human biological intelligence, then it falls short of being entirely human
* if the mechanical intelligence is insufficiently similar to human biological intelligence, then it falls short of being entirely intelligent

Ergo, humans intelligence can not be replaced. QED.