Why would the CIA/NSA want to bother with pretending to be an advertiser? They can just buy up the information from a real advertiser for less effort. Private enterprise for the win (and humankind for the loss)!
Sorry for the cynicism. I agree that stripping out all the junk is a great idea. The question is where to do this. Working through a third-party proxy as described above is great if the proxy is trustworthy. Unfortunately, it just adds another link in the chain that, if the idea takes off, would be attractive to scumsucking privacy invaders to exploit with their own deceptive variants. Working towards privacy-by-default on the browser side seems to me a better approach. Wouldn't it be cool if a default Firefox install would require the user to add a bunch of plugins if they wanted to unblock ads and tracking? Better browser privacy design to prevent "data leaks" (like what the EFF is trying to study with Panopticlick) can provide much of the benefit of proxies without requiring extra layers of trust (and costs for proxy operation).
Great idea! You could even raise additional funds by collecting and reselling info about what your users are browsing. Maybe even insert some relevant product-based sponsored informational links into the proxied pages?
I'm not denying that the factories churning out American mass-produced food-substitute and beverage-substitute products aren't marvels of industrial engineering and process control. Unfortunately, not just in beer, but pretty much every food product, convenience for centralized mass production to meet corporate desires completely trumps every other aspect of quality. Products are designed for shelf life, easy transport, uniformity, cheap monoculture ingredients, minimum manufacturing cost --- not for taste, nutrition, ecological sustainability, craftsmanship, or variety.
Our American Megacorporate Overlords have been pushing shitty beer (and every other kind of quasi-food product) on Americans since well before consolidating into global holdings conglomerates.
Becoming part of a globalized mega holding company, which doesn't actually give a shit for America, Americans, or the products you're making, but loves raking in money, is a hallowed part of the American Megacorporate Tradition.
Why not spending a couple of extra cents on quality ingredients to make a quality beer instead of blowing money on cooling?
Because that wouldn't be the American Megacorporate Way. Why spend more on product quality, when you could spend half as much on ubiquitous ad campaigns to redefine the country's understanding of what "beer" even is?
So what's the difference in that analogy between theoretical physics and mathematics within any established branch?
You pretty much answer your own question:
Admittedly with mathematics you can go completely wild and come up with a system that works however you want it to
and, even when you do the same under the guise of "science" (creating mathematical formalisms for "invented universes" entirely detached from experimental results), you've stopped doing science and started doing math. There's nothing wrong with math! I love math! I have a degree in math! But, if one is looking for the key technical distinction between "science" and other branches of philosophy, then connection to empirical observation is it --- whether or not you personally feel this is a big deal.
You also keep saying 'monopoly' like there is something wrong with being a monopoly - there is not.
Where do you see me saying this? You're the one making that inference. I'm using the word in its correct technical sense, to indicate that the "long term research" work of IBM correlates with breakdowns in idealized competitive free-market conditions.
In your world, the only meaning of 'competition' is offering the same thing at a lower price.
Or a better thing at the same price --- whatever it takes to undercut the competition while still making non-negative profits. I'm sticking to the plain technical sense of the word.
whereas your '0 profit' version of competition is of value to nobody.
'0 profit' is only directly detrimental to those who make money from profit, and spend money on speculation. If you make money from wages, and spend money on goods, then "0 profit" might work out just fine for you (you're still earning wages at the same market rates of a more "profitable" system, and getting goods at the minimum prices for production). How about a simple introduction to the relation between "profit" and "competition" from Wikipedia:
Economic profit does not occur in perfect competition in long run equilibrium; if it did, there would be an incentive for new firms to enter the industry, aided by a lack of barriers to entry until there was no longer any profit.[4] As new firms enter the industry, they increase the supply of the product available in the market, and these new firms are forced to charge a lower price to entice consumers to buy the additional supply these new firms are supplying (they compete for customers).
You are the one tossing around terms in plain contradiction to their basic meanings in economics, to support your whimsical corporatism.
However it is not true that IBM is a monopoly, it is an economy of scale that is successful in its niche.
In other words, IBM is protected from competition by high barriers to entry (which include "economies of scale"), allowing it to the 9th most profitable corporation in the US according to Forbe's 2012 rankings. When competition is working, lots of other people would be interested in getting their chunk of that 9th most profitable spot (every investor with money in #10 and below), and would allocate their own investments to compete (hence drive down prices and profits) in the sector. However, barriers to entry are sufficiently high that IBM gets to keep its lucrative position to itself. While your simpleminded worldview may only have two binary categories for either "pure monopoly" or "pure competition," real-world economic behavior is better described by a continuous spectrum, with IBM very close to the "monopoly" side.
I'm not Feynman (sorry!), but I can offer my interpretation of what this means beyond cheap-shot vulgarity.
So... maths is something you do when you're in the mood for physics but don't have anyone else to do it with?
That actually pretty much sums it up. Physics requires you to interact with your "partner" --- the real world --- to carry out the act. Mathematics occurs in the solitary world of your head --- you manipulate symbolic systems without the need to touch the real world and move in responsive dialog to its motions. In short, mathematics fucks with your mind, physics fucks with the world.
My version of 'ideal' competition, straight out of the basic Econ 101 definitions, is considered the basic motivation for why people want competitive markets in the first place (squeezing out all inefficient waste to "efficiently" produce products at the bare minimum cost of production). And yes, I think "market solutions" are a fine way to produce "hell on earth". "Profit" is a measure of inefficiency in markets, since it indicates the same product could have been sold cheaper (without the profit margins) if there was enough competition (where everyone is struggling to reap profits, but never gets ahead for more than brief flashes until someone else comes in to eat their lunch). Note that "zero profits" isn't terrible for everyone --- profit is money after you've paid off all costs, including labor, so if you actually work for a living (instead of expecting to be handed free money as a reward for previously having money), you're still getting paid your wages in a profit-free perfectly competitive market (and have plenty of motive to do things, like keeping your job and wages).
Someday, when new snazzy technology catches up to the point of actually improving over older systems, I'll sign up too. For a tiny fraction of the population, perhaps the bleeding-edge whizzies already do offer something useful (...at least points on the conspicuous consumption fashion scoreboard). However, I see a lot of people leaping two steps back to grab that one step forward promised by the latest-and-greatest micro-doodad.
Hey, I'm gonna burn six hours of my wages per month on a fancy smartphone contract, so I can twit my facebook angrybirds in the grocery line! Timesaver! And now I've got way better things to do while driving than watch the boring road!
What's that? "Theological rhetorical ethics"? Damn, those humanities departments are getting completely out of control. No idea how you'd debate things from a theorhethical standpoint, but it sounds messy.
Anyway, from a theoretical standpoint, see my post above --- sub-wavelength structures are handy for holographic manipulation of light wavefronts. While there are diminishing returns from going way smaller than viewing light wavelengths (atomic scale would be overkill), control over feature scales in the lambda/4 to lambda/10 range would be a nice start.
I said "monopolistic," which doesn't require an absolutely pure monopoly, but for a lot of "big data," mainframe, and supercomputing applications, the market is certainly an oligopoly with only a handful of big players (which produces economic results very similar to pure monopoly). Anyway, in ideal pure competition, profit margins are reduced to zero --- everyone is scrabbling to be immediately responsive to market forces, not able to raise prices so far above production costs that they have oodles of cash lying around for pie-in-the-sky research (because someone else could skip the costly long-horizon research and eat their business with cheaper products today). Research divisions like those in IBM, or the old Bell Labs, only exist where surplus profit (nonexistent in (nonexistent) competitive markets) is used to create little internal "communist paradises" for researchers free from responding to market forces. There's no "stop screwing around with atoms; we need to knock 0.7 cents off the cost of power supply wiring insulation to meet quarterly goals" in this IBM research division.
A holographic display like this would likely require a coherent laser backlight, but there's no fundamental reason you couldn't just do that (with similar efficiency to solid-state LED lighting currently in use). In 3D tech available today like the Nintendo 3DS, your eyes focus at a fixed distance on a screen, while receiving different images to provide stereo depth cues. This is mismatched from how you see real-life 3D objects, where your eye also changes focus to see closer or farther details, and the perspective changes as you move your head (this part can be done for one viewer with head tracking). A hologram approximates the ideal of "exactly how light reaches your eyes from real-world objects," eliminating the subtle headache/eyestrain-inducing effects of decoupling focus from stereo cues, and allowing natural "depth of field" blur (without everything unnaturally in focus, or focus restricted to where the device chooses instead of where your eye looks).
Darn, I'd like to type this message up, but my computer terminal is all the way over in Sector W. It'll take me 8 minutes of walking, just for 2 minutes to type and send the message.
New hotness:
I can send a message right from where I am, on my watch! All I have to do is fiddle with my wrist for 12 minutes, and voila, instant gratification!
All the advances in wristwatch IO are fixing problems created by insisting on doing stupid shit on a wristwatch that a wristwatch isn't any good for. Looking forward to future technological advances, like:
Shaving by burning my hair off with a hot clothes iron used to be incredibly painful... but with advances in rapid temperature control, my new techno-iron can singe off my hair with 78% less third-degree-burn scars, while reading me live updates from my twitfeed!
Yep, nothing like a large monopolistic megacorporation, flush with cash from big lucrative government contracts, to highlight the success of free enterprise research. Like the old Bell Labs during the telco monopoly days, when a corporation gets so big and rich that it can afford to have whole divisions entirely isolated from worrying about market competition, it can do far-forward-looking work (just like all the directly government-funded research labs, also free from dealing with "the markets" to set their goals). So long as you are sufficiently protected from having your priorities shaped by market forces, you might be able to do long-term useful stuff beyond the usual race-for-the-bottom, rich-get-richer grind.
In all chemical bonds the ground state has non-zero energy which results in a vibration of the two atoms.
Ground-state "vibration" of a quantum harmonic oscillator isn't exactly like what you might think of based on the "classical" limit like a swinging pendulum. Quantum harmonic oscillator energy level states (including the ground state) are time stationary: the particle has some probability distribution of being in various locations which does not change as a function of time. Only when you mix different "pure" energy eigenstates together do you get a time-varying probability distribution that "sloshes back and forth" like you'd expect from a classical pendulum. So, what this experiment is trying to do differently is produce a ground state system with periodic, time-dependent variation, which you don't get from simple particle-in-a-well oscillator systems.
Spatially interleaving color pixels would produce issues similar to viewing the world through a window screen --- yes, there are diffraction artifacts, but it's not a completely horrible unrecognizable view. Furthermore, you could move to non-periodic tilings for the colors which would eliminate obvious diffraction spikes. Temporal interleaving would indeed work too (and better in some aspects); we'll have to get a lot closer to production-ready technology than we currently are to assess what particular approach will work.
Yes, there are many other engineering/technology hurdles to cross to usefully generate dynamic holograms. However, I don't think there's anything particularly impassable in this case.
With such display technology, I doubt you'd think in terms of addressing/storing individual "pixels" in one "centralized" place; instead, you might have hardware DSPs behind "macropixels" ("normal pixel"-sized arrays of the light-wavelength-scale micropixels) which would receive 3D models and calculate the transform into the interference pattern for the "macropixel." A "supercomputer-in-a-box," indeed, but for a highly parallelizable task. You'd never store the (highly redundant) "pixel representation" of a hologram, but rather the 3D model that produces it, and re-create the "pixels" on the fly in hardware as needed. By interleaving colored "macropixels" (just like interleaved colored pixels on regular displays) you could produce "full color" holograms.
The "modern" input equivalent is chording keyboards --- instead of a "single bit per tap," you use combinations of several fingers to produce even higher data rates. Steep learning curve, indeed, but people who learn to use them can type blazingly fast on a rather small input device.
Isn't it amazing how great the advancement of technology is for device usability! Now people can type at an amazing *9.3* words per minute! It's hard to imagine how we ever got by back in the old days, when a casual typist could only achieve 30-60 WPM --- uphill through the snow both ways barefoot. Progress!
Slashdot Mirror
Why would the CIA/NSA want to bother with pretending to be an advertiser? They can just buy up the information from a real advertiser for less effort. Private enterprise for the win (and humankind for the loss)!
RfC 3514. Why re-invent the wheel with new standards?
Sorry for the cynicism. I agree that stripping out all the junk is a great idea. The question is where to do this. Working through a third-party proxy as described above is great if the proxy is trustworthy. Unfortunately, it just adds another link in the chain that, if the idea takes off, would be attractive to scumsucking privacy invaders to exploit with their own deceptive variants. Working towards privacy-by-default on the browser side seems to me a better approach. Wouldn't it be cool if a default Firefox install would require the user to add a bunch of plugins if they wanted to unblock ads and tracking? Better browser privacy design to prevent "data leaks" (like what the EFF is trying to study with Panopticlick) can provide much of the benefit of proxies without requiring extra layers of trust (and costs for proxy operation).
Great idea! You could even raise additional funds by collecting and reselling info about what your users are browsing. Maybe even insert some relevant product-based sponsored informational links into the proxied pages?
I'm not denying that the factories churning out American mass-produced food-substitute and beverage-substitute products aren't marvels of industrial engineering and process control. Unfortunately, not just in beer, but pretty much every food product, convenience for centralized mass production to meet corporate desires completely trumps every other aspect of quality. Products are designed for shelf life, easy transport, uniformity, cheap monoculture ingredients, minimum manufacturing cost --- not for taste, nutrition, ecological sustainability, craftsmanship, or variety.
Our American Megacorporate Overlords have been pushing shitty beer (and every other kind of quasi-food product) on Americans since well before consolidating into global holdings conglomerates.
Becoming part of a globalized mega holding company, which doesn't actually give a shit for America, Americans, or the products you're making, but loves raking in money, is a hallowed part of the American Megacorporate Tradition.
Why not spending a couple of extra cents on quality ingredients to make a quality beer instead of blowing money on cooling?
Because that wouldn't be the American Megacorporate Way. Why spend more on product quality, when you could spend half as much on ubiquitous ad campaigns to redefine the country's understanding of what "beer" even is?
So what's the difference in that analogy between theoretical physics and mathematics within any established branch?
You pretty much answer your own question:
Admittedly with mathematics you can go completely wild and come up with a system that works however you want it to
and, even when you do the same under the guise of "science" (creating mathematical formalisms for "invented universes" entirely detached from experimental results), you've stopped doing science and started doing math. There's nothing wrong with math! I love math! I have a degree in math! But, if one is looking for the key technical distinction between "science" and other branches of philosophy, then connection to empirical observation is it --- whether or not you personally feel this is a big deal.
You also keep saying 'monopoly' like there is something wrong with being a monopoly - there is not.
Where do you see me saying this? You're the one making that inference. I'm using the word in its correct technical sense, to indicate that the "long term research" work of IBM correlates with breakdowns in idealized competitive free-market conditions.
In your world, the only meaning of 'competition' is offering the same thing at a lower price.
Or a better thing at the same price --- whatever it takes to undercut the competition while still making non-negative profits. I'm sticking to the plain technical sense of the word.
whereas your '0 profit' version of competition is of value to nobody.
'0 profit' is only directly detrimental to those who make money from profit, and spend money on speculation. If you make money from wages, and spend money on goods, then "0 profit" might work out just fine for you (you're still earning wages at the same market rates of a more "profitable" system, and getting goods at the minimum prices for production). How about a simple introduction to the relation between "profit" and "competition" from Wikipedia:
Economic profit does not occur in perfect competition in long run equilibrium; if it did, there would be an incentive for new firms to enter the industry, aided by a lack of barriers to entry until there was no longer any profit.[4] As new firms enter the industry, they increase the supply of the product available in the market, and these new firms are forced to charge a lower price to entice consumers to buy the additional supply these new firms are supplying (they compete for customers).
You are the one tossing around terms in plain contradiction to their basic meanings in economics, to support your whimsical corporatism.
However it is not true that IBM is a monopoly, it is an economy of scale that is successful in its niche.
In other words, IBM is protected from competition by high barriers to entry (which include "economies of scale"), allowing it to the 9th most profitable corporation in the US according to Forbe's 2012 rankings. When competition is working, lots of other people would be interested in getting their chunk of that 9th most profitable spot (every investor with money in #10 and below), and would allocate their own investments to compete (hence drive down prices and profits) in the sector. However, barriers to entry are sufficiently high that IBM gets to keep its lucrative position to itself. While your simpleminded worldview may only have two binary categories for either "pure monopoly" or "pure competition," real-world economic behavior is better described by a continuous spectrum, with IBM very close to the "monopoly" side.
I'm not Feynman (sorry!), but I can offer my interpretation of what this means beyond cheap-shot vulgarity.
So... maths is something you do when you're in the mood for physics but don't have anyone else to do it with?
That actually pretty much sums it up. Physics requires you to interact with your "partner" --- the real world --- to carry out the act. Mathematics occurs in the solitary world of your head --- you manipulate symbolic systems without the need to touch the real world and move in responsive dialog to its motions. In short, mathematics fucks with your mind, physics fucks with the world.
My version of 'ideal' competition, straight out of the basic Econ 101 definitions, is considered the basic motivation for why people want competitive markets in the first place (squeezing out all inefficient waste to "efficiently" produce products at the bare minimum cost of production). And yes, I think "market solutions" are a fine way to produce "hell on earth". "Profit" is a measure of inefficiency in markets, since it indicates the same product could have been sold cheaper (without the profit margins) if there was enough competition (where everyone is struggling to reap profits, but never gets ahead for more than brief flashes until someone else comes in to eat their lunch). Note that "zero profits" isn't terrible for everyone --- profit is money after you've paid off all costs, including labor, so if you actually work for a living (instead of expecting to be handed free money as a reward for previously having money), you're still getting paid your wages in a profit-free perfectly competitive market (and have plenty of motive to do things, like keeping your job and wages).
Someday, when new snazzy technology catches up to the point of actually improving over older systems, I'll sign up too. For a tiny fraction of the population, perhaps the bleeding-edge whizzies already do offer something useful (...at least points on the conspicuous consumption fashion scoreboard). However, I see a lot of people leaping two steps back to grab that one step forward promised by the latest-and-greatest micro-doodad.
Hey, I'm gonna burn six hours of my wages per month on a fancy smartphone contract, so I can twit my facebook angrybirds in the grocery line! Timesaver! And now I've got way better things to do while driving than watch the boring road!
theorhethical
What's that? "Theological rhetorical ethics"? Damn, those humanities departments are getting completely out of control. No idea how you'd debate things from a theorhethical standpoint, but it sounds messy.
Anyway, from a theoretical standpoint, see my post above --- sub-wavelength structures are handy for holographic manipulation of light wavefronts. While there are diminishing returns from going way smaller than viewing light wavelengths (atomic scale would be overkill), control over feature scales in the lambda/4 to lambda/10 range would be a nice start.
I said "monopolistic," which doesn't require an absolutely pure monopoly, but for a lot of "big data," mainframe, and supercomputing applications, the market is certainly an oligopoly with only a handful of big players (which produces economic results very similar to pure monopoly). Anyway, in ideal pure competition, profit margins are reduced to zero --- everyone is scrabbling to be immediately responsive to market forces, not able to raise prices so far above production costs that they have oodles of cash lying around for pie-in-the-sky research (because someone else could skip the costly long-horizon research and eat their business with cheaper products today). Research divisions like those in IBM, or the old Bell Labs, only exist where surplus profit (nonexistent in (nonexistent) competitive markets) is used to create little internal "communist paradises" for researchers free from responding to market forces. There's no "stop screwing around with atoms; we need to knock 0.7 cents off the cost of power supply wiring insulation to meet quarterly goals" in this IBM research division.
A holographic display like this would likely require a coherent laser backlight, but there's no fundamental reason you couldn't just do that (with similar efficiency to solid-state LED lighting currently in use). In 3D tech available today like the Nintendo 3DS, your eyes focus at a fixed distance on a screen, while receiving different images to provide stereo depth cues. This is mismatched from how you see real-life 3D objects, where your eye also changes focus to see closer or farther details, and the perspective changes as you move your head (this part can be done for one viewer with head tracking). A hologram approximates the ideal of "exactly how light reaches your eyes from real-world objects," eliminating the subtle headache/eyestrain-inducing effects of decoupling focus from stereo cues, and allowing natural "depth of field" blur (without everything unnaturally in focus, or focus restricted to where the device chooses instead of where your eye looks).
Old and broken:
Darn, I'd like to type this message up, but my computer terminal is all the way over in Sector W. It'll take me 8 minutes of walking, just for 2 minutes to type and send the message.
New hotness:
I can send a message right from where I am, on my watch! All I have to do is fiddle with my wrist for 12 minutes, and voila, instant gratification!
All the advances in wristwatch IO are fixing problems created by insisting on doing stupid shit on a wristwatch that a wristwatch isn't any good for. Looking forward to future technological advances, like:
Shaving by burning my hair off with a hot clothes iron used to be incredibly painful... but with advances in rapid temperature control, my new techno-iron can singe off my hair with 78% less third-degree-burn scars, while reading me live updates from my twitfeed!
Yep, nothing like a large monopolistic megacorporation, flush with cash from big lucrative government contracts, to highlight the success of free enterprise research. Like the old Bell Labs during the telco monopoly days, when a corporation gets so big and rich that it can afford to have whole divisions entirely isolated from worrying about market competition, it can do far-forward-looking work (just like all the directly government-funded research labs, also free from dealing with "the markets" to set their goals). So long as you are sufficiently protected from having your priorities shaped by market forces, you might be able to do long-term useful stuff beyond the usual race-for-the-bottom, rich-get-richer grind.
In all chemical bonds the ground state has non-zero energy which results in a vibration of the two atoms.
Ground-state "vibration" of a quantum harmonic oscillator isn't exactly like what you might think of based on the "classical" limit like a swinging pendulum. Quantum harmonic oscillator energy level states (including the ground state) are time stationary: the particle has some probability distribution of being in various locations which does not change as a function of time. Only when you mix different "pure" energy eigenstates together do you get a time-varying probability distribution that "sloshes back and forth" like you'd expect from a classical pendulum. So, what this experiment is trying to do differently is produce a ground state system with periodic, time-dependent variation, which you don't get from simple particle-in-a-well oscillator systems.
Yeah, it's not like people screwing around with impractical technology at IBM ever produced anything useful, like inventing the scanning-tunneling microscope .
Spatially interleaving color pixels would produce issues similar to viewing the world through a window screen --- yes, there are diffraction artifacts, but it's not a completely horrible unrecognizable view. Furthermore, you could move to non-periodic tilings for the colors which would eliminate obvious diffraction spikes. Temporal interleaving would indeed work too (and better in some aspects); we'll have to get a lot closer to production-ready technology than we currently are to assess what particular approach will work.
Yes, there are many other engineering/technology hurdles to cross to usefully generate dynamic holograms. However, I don't think there's anything particularly impassable in this case.
With such display technology, I doubt you'd think in terms of addressing/storing individual "pixels" in one "centralized" place; instead, you might have hardware DSPs behind "macropixels" ("normal pixel"-sized arrays of the light-wavelength-scale micropixels) which would receive 3D models and calculate the transform into the interference pattern for the "macropixel." A "supercomputer-in-a-box," indeed, but for a highly parallelizable task. You'd never store the (highly redundant) "pixel representation" of a hologram, but rather the 3D model that produces it, and re-create the "pixels" on the fly in hardware as needed. By interleaving colored "macropixels" (just like interleaved colored pixels on regular displays) you could produce "full color" holograms.
The "modern" input equivalent is chording keyboards --- instead of a "single bit per tap," you use combinations of several fingers to produce even higher data rates. Steep learning curve, indeed, but people who learn to use them can type blazingly fast on a rather small input device.
Isn't it amazing how great the advancement of technology is for device usability! Now people can type at an amazing *9.3* words per minute! It's hard to imagine how we ever got by back in the old days, when a casual typist could only achieve 30-60 WPM --- uphill through the snow both ways barefoot. Progress!