FICA finances social security and medicare programs. The "common man" pays the FICA portion of their wages (I believe around 8 percent) for these benefits.
Actually, since 1983, FICA revenue is treated just like the other two sources of revenue (corporate tax and personal income tax) as far as the budget is concerned. This was a fundamental change in how the government keeps its books, and it made obsolete an assumption that I falsely believed for the past 21 years.
Also, don't forget that (1) all workers pay FICA on the first $70K of their income only, so this is a highly regressive tax, and (2) employers match this contribution, so FICA acts as an even more direct drag on the creation of middle-class jobs than what supply-siders claim for the capital gains tax.
I'm not trying to be disagreeable here. Just spend an afternoon with the federal budget, available in PDF form from the Executive branch website. It is filled with historical data on revenues/taxes stretching back one century. Do avoid the partisan websites. For example, before finding the budget itself, I found (through Google) a Heritage Foundation analysis based on a carefully chosen ten-year period. The general lesson it drew did not hold up for the immediately preceding and immediately followinng decade -- among other flaws.
The top 50% pay ~99% of the tax. The top 20% pay 80%. So, strangely, when tax cuts are taken across the board, the people who make the most, get the most back. The "rich" still have the highest tax bracket.
This is false as stated. It may be true of income tax alone, but that is not your claim. You are neglecting FICA. All workers pay the same tax rate on the first $70K of their earnings, and $0 thereafter, so this is a highly regressive tax. Roughly 44% of the federal government's revenue is through FICA. Moreover, if you look, this contribution is roughly constant (in inflation-adjusted dollars) since Reagan took office.
Roughly 10% of the revenue is from corporate taxes and roughly 46% from personal income tax. These numbers fluctuate somewhat based on (1) tax increases/decreases and (2) the state of the economy, for obviously reasons. Given the current tax rates and economy, the worker is bearing a larger burden than anytime in the last 24 years.
It is merely a Republican talking point to claim that most federal tax is paid by the top 50% (or 20%). When you include FICA, you see that the common man, the worker, pays more than his or her fair share. Really, the FICA revenue is the backbone of the federal budget, especially since 1983 when it was first tapped to reduce (but not eliminate!) the huge deficits caused by Reagan's tax cuts. It appears to be the plan of neo-cons to keep these revenues in place while decreasing those collected through corporate and personal income taxes. Guess who this benefits and who this hurts?
Furthermore, remember that FICA contributions are matched by employers. So these taxes -- this fundamental source of revenue for the federal government -- acts as a drag on the hiring of American workers, especially those with middle-class (and lower) salaries. Gues who this benefits and who this hurts?
I found this information in the federal budget itself, which is available at the Executive branch website. I went looking for it after a friendly debate with a conservative stockbroker friend of mine who argued the supply-side logic for cutting taxes on the wealthy. What I found surprised him and surprised me. I'd recommend that anyone who believes tax policy is a philosophical debate just go look at the data and run it through Excel. I know I learned a hell of a lot in just an afternoon.
By the way, I'm open to intelligent critiques of what I wrote because I'm truly trying to cut through the BS and see what the numbers are saying. If I've gotten something wrong, please let me know.
You can also go straight to the source of above 100,000 and 10 year info: chech out Herbert A. Simon's Sciences of the Artificial.
You're right, I should have referenced Simon. I thought that might be too obscure and was aiming for something that was a click away -- the link I tried to include discusses Simon's work, and I thought this would be an indirect pointer to the truly interested.
For those who care, the two seminal articles on this topic are:
Chase, W. G., & Simon, H. A. (1973). Perception in chess. Cognitive Psychology, 4, 55-81.
Ericsson, K. A., Chase, W. G., & Faloon, S. (1980). Acquisition of a memory skill. Science, 208, 1181-1182.
The first article is on the 1000s of perceptual chunks chess experts use to instantly parse the strategic information in a game position. Although this article is brilliant, the second one is even better! Chase and Ericsson had Steve Faloon, an undergraduate at Carnegie Mellon and accomplished distance runner, practice his memory span for digits. You know: How many digits can you remember? For most people, the answer is in the range 7+/-2 (i.e., 5 to 9). Faloon was able to stretch his digit span to over 80 digits. I have an audiotape of him doing this for 25 digits and it's mind-bending to listen to it. The digits are read to him at a rate of about 1 per second. He pauses, strains, and struggles, and then spits them back out, perfectly.
How did he do it? By "chunking" the digits into running times running times that were meaningful to him (e.g., "4-1-2: my mile time in meet X back in 1978") and then by learning to chunk these chunks (and chunk these chunks of chunks (and...)).
When John Watson, the father of American behaviorism, was kicked out of the academy for marrying his graduate student, he went to Madison Avenue and made a fortune as an advertising executive.
I agree, Kurweil does sound like a crazy man. And I don't appreciate the rather inflated list of achievements attributed to him.
However, I think I know where the 100,000 number came from. Research on expertise suggests that one must learn about 100,000 "chunks" of information to achieve expert status. Given the rate at which humans learn new chunks, this translates into 10 years of concerted practice. If you look at the time it takes people to attain world class status in various endeavours (e.g., chess, tennis), these numbers fit pretty well.
Bobby Fischer is commonly held up as an exception as he achieved a grandmaster rating 9 or so years after being introduced to chess. Prodigies like Mozart are not considered exceptions because apparently people distinguish between the works he produced as a child and the true masterworks of his adulthood.
Java is not necessarily borrowing from C(++). There are a lot of other programming languages out there which implement these features (e.g., varargs), some of which the Java developers (e.g., Guy Steele) designed decades ago (e.g., Common Lisp).
The scandal isn't that Java is borrowing from C++. It's that it has taken so long for "mainstream" languages to provide constructs that should have been there a long time ago.
Yet if I want to do the same thing with marijuana I'm the criminal? Yes, you are. That's the law. You can't sell marijuana. [...]
For the record, selling bootlegs is wrong And so is everything the RIAA does. No, thats provably false. The RIAA is validated by the massess. The music industry sells millions of records and is rather profitable. Logically that means something they are doing is filling a marketable demand.
Drug Dealer: I am validated by the masses. I sell lots of drugs and am rather profitable. Logically that means that something I am doing is filling a marketable demand.
(I think you should forget about "marketable demand" and simply say the RIAA's practices are allowed by the law.)
However, Marois and Vogel (the authors of the current papers) are very sharp guys, so I'm not going to second guess their conclusions before reading their paper. However, I am going to ding the science editor writing for Nature for bring the fuzzy, baggage-laden concept of "intelligence" into the mix. They seem to have found the visual working memory bottleneck...what this says about "intelligence" is pure speculation and a red herring.
Cogent post overall. I just want to offer a slightly different take. Vogel and Marois are certainly sharp guys. I don't consider them working memory researchers though. They study visual attention (and its capacity), and in this regard they sometimes brush up against working memory issues. And given the current fascination in cognitive science with low-level forms of cognition and the brain, many equate these two topics. I personally don't. There are researchers who study the role of working memory in high-level and ecologically interesting forms of cognition, such as Ericsson (sp?) and Engle. There are others who actually care about the intersection of measures of fluid intelligence and working memory, including Kyllonen and Just, and even Duncan, who's quoted on the Nature website. It's easy to look at all these intertwined threads and conclude that posterior parietal cortex is central to intelligence or whatever the article says. To me, this is just drawing an overly broad conclusion from what are, at their heart, simple tests of visual attention.
The article even quotes Pat Langley as saying this is nothing new. He mentions some system from the early 1990s as the true pioneer here, but he's just being modest. Langley, Simon, and colleagues published in Science back in the early 1980s on their scientific discovery programs.
This is an incremental advance perhaps, but not worthy of this kind of attention. Just goes to show that Nature and the like are as much about PR as they are about the genuinely new.
I mean, did the Elves make Glamdring and Sting and Orcrist and then FORGET what they just did?
Ever try to maintain old-but-working code written in a crufty language? It can be damn near impossible; you just use it as is and hope it keeps doing its magic.
I have a pet theory that the many software engineers who beat their chests about process and discipline have no idea what *real* engineers -- EEs, MEs, ChemEs, etc. -- know and what they do on a day to day basis.
Let me give you a hint: Grady Booch, he ain't no Newton.
I have never heard a credible medical professional assert that the output of an MRI provides any insight into the feelings or thoughts of the patient. It's made for detecting tumors and stuff like that.
You're half right. MRI is used for probing the static structure of the brain, e.g., in search of tumors. functional MRI (fMRI) is used for measuring the dynamics of blood flow in the brain during performance of a task. It can, for example, indicate which areas of the brain are active when processing "emotional" stimuli, and therefore may be useful for advertisers who want an emotional, not logical reaction (e.g., the makers of weight loss supplements).
The idea is that areas that are being actively used will get increased blood flow, but this happens on the order of tens of seconds, so it doesn't really provide a detailed picture of brain activity.
There are two kinds of experimental designs one can employ with fMRI. You are referring to "block" designs, where in fact the neural response is aggregated over tens of seconds. However, "event-related" designs permit temporal resolutions of 1.5 seconds, and I have seen some studies that clam 0.5 seconds. So, fMRI can do a bit better time-wise than you think.
The other dimension of interest is spatial resolution. Just like a computer screen is composed of 2D pixels, fMRI partitions brains into 3D voxels. The smallest voxels sizes that current technologies allow is on the order of 10s of cubic mms. However, Logothetis and colleagues have pioneered more invasive techniques that allow voxels that are several orders of magnitude smaller. The only drawback is that they can't be used in humans.
However, advertisers can gain minute knowledge of how animals respond to different ads! Perhaps Alpo, Purina, and the like should use neuromarketing.
By using MRI, scientists can know what parts of the brain are / may be stimulated by ads, so what kind of feelings we got when seeing / hearing it..
They can't tell what "feelings" a person is experiencing. Emotions aren't individually localizable to regions of the brain.
They can tell whether an area of the brain generally responsible for emotions (e.g., orbitofrontal cortex) is being engaged versus one generally responsible for deliberate reasoning (e.g., the dorsolateral prefrontal cortex). This could reveal whether an ad prompts emotional or logical responses, generally speaking.
But that's about it given the limitations of the technology and cognitive neuroscience.
Disclaimer: The site was borderline slashdotted so I couldn't read the source article.
Marketplace on CBC, that's a Canadian station for you Americans...
It's not just Americans who haven't heard of CBC.
No one outside Canada cares about the call letters of your pissant stations.
Re:What desktop users want to know..
on
AMD's 64-bit Plot
·
· Score: 2
And what does it get you? [...] Maybe you're getting a 20% across the board improvement, but _man_ are you paying for it, both in cost and power consumption. And was it worth it, for 27% faster than "more speed than I know what to do with?" Probably not (though I realize that all hardware site weenies will absolutely insist that they can feel the difference when browsing the web on such a machine).
Tubes versus solid-state... Beta versus VHS... Vinyl records versus CDs... Air-cooled versus water-cooled...
You and several other people in this thread have it dead wrong. Microsoft hired Charles Simonyi from Xerox PARC in the early 1980s. Word was his crappy attempt to deliver a more modern word processor to the PC, whose users at the time were willing to put up with shit like the "dot commands" of WordStar.
(No flames from the old timers. I'm an old timer myself. But the facts is the facts. And if you don't like 'em, I've got an Exidy Sorcerer to sell you.)
Anyway, Word was trounced in the market by WordPerfect and even marginal programs like XYWrite. In the late 1980s, Microsoft turned it into a usable product on the Macintosh, good enough that Apple/Claris let their own MacWrite blend into the background. (This was back when people were convinced Apple wanted to control both Mac hardware and software, and so Apple was willing to de-emphasize its office productivity products -- MacWrite, MacPaint, MacDraw, etc.)
When Windows 3.0 came out, Microsoft had a leg up on their PC competition making GUI-based word processors from their Mac experience. They beat their competitors to the market. This is when Word and Excel really trounced their WordPerfect, Lotus, and Quattro.
(And to the people lamenting the loss of Ami Pro: It was only good if you were stuck on stone age PCs.)
Slashdot Mirror
FICA finances social security and medicare programs. The "common man" pays the FICA portion of their wages (I believe around 8 percent) for these benefits.
Actually, since 1983, FICA revenue is treated just like the other two sources of revenue (corporate tax and personal income tax) as far as the budget is concerned. This was a fundamental change in how the government keeps its books, and it made obsolete an assumption that I falsely believed for the past 21 years.
Also, don't forget that (1) all workers pay FICA on the first $70K of their income only, so this is a highly regressive tax, and (2) employers match this contribution, so FICA acts as an even more direct drag on the creation of middle-class jobs than what supply-siders claim for the capital gains tax.
I'm not trying to be disagreeable here. Just spend an afternoon with the federal budget, available in PDF form from the Executive branch website. It is filled with historical data on revenues/taxes stretching back one century. Do avoid the partisan websites. For example, before finding the budget itself, I found (through Google) a Heritage Foundation analysis based on a carefully chosen ten-year period. The general lesson it drew did not hold up for the immediately preceding and immediately followinng decade -- among other flaws.
The top 50% pay ~99% of the tax. The top 20% pay 80%. So, strangely, when tax cuts are taken across the board, the people who make the most, get the most back. The "rich" still have the highest tax bracket.
This is false as stated. It may be true of income tax alone, but that is not your claim. You are neglecting FICA. All workers pay the same tax rate on the first $70K of their earnings, and $0 thereafter, so this is a highly regressive tax. Roughly 44% of the federal government's revenue is through FICA. Moreover, if you look, this contribution is roughly constant (in inflation-adjusted dollars) since Reagan took office.
Roughly 10% of the revenue is from corporate taxes and roughly 46% from personal income tax. These numbers fluctuate somewhat based on (1) tax increases/decreases and (2) the state of the economy, for obviously reasons. Given the current tax rates and economy, the worker is bearing a larger burden than anytime in the last 24 years.
It is merely a Republican talking point to claim that most federal tax is paid by the top 50% (or 20%). When you include FICA, you see that the common man, the worker, pays more than his or her fair share. Really, the FICA revenue is the backbone of the federal budget, especially since 1983 when it was first tapped to reduce (but not eliminate!) the huge deficits caused by Reagan's tax cuts. It appears to be the plan of neo-cons to keep these revenues in place while decreasing those collected through corporate and personal income taxes. Guess who this benefits and who this hurts?
Furthermore, remember that FICA contributions are matched by employers. So these taxes -- this fundamental source of revenue for the federal government -- acts as a drag on the hiring of American workers, especially those with middle-class (and lower) salaries. Gues who this benefits and who this hurts?
I found this information in the federal budget itself, which is available at the Executive branch website. I went looking for it after a friendly debate with a conservative stockbroker friend of mine who argued the supply-side logic for cutting taxes on the wealthy. What I found surprised him and surprised me. I'd recommend that anyone who believes tax policy is a philosophical debate just go look at the data and run it through Excel. I know I learned a hell of a lot in just an afternoon.
By the way, I'm open to intelligent critiques of what I wrote because I'm truly trying to cut through the BS and see what the numbers are saying. If I've gotten something wrong, please let me know.
You can also go straight to the source of above 100,000 and 10 year info: chech out Herbert A. Simon's Sciences of the Artificial.
...)).
You're right, I should have referenced Simon. I thought that might be too obscure and was aiming for something that was a click away -- the link I tried to include discusses Simon's work, and I thought this would be an indirect pointer to the truly interested.
For those who care, the two seminal articles on this topic are:
Chase, W. G., & Simon, H. A. (1973). Perception in chess. Cognitive Psychology, 4, 55-81.
Ericsson, K. A., Chase, W. G., & Faloon, S. (1980). Acquisition of a memory skill. Science, 208, 1181-1182.
The first article is on the 1000s of perceptual chunks chess experts use to instantly parse the strategic information in a game position. Although this article is brilliant, the second one is even better! Chase and Ericsson had Steve Faloon, an undergraduate at Carnegie Mellon and accomplished distance runner, practice his memory span for digits. You know: How many digits can you remember? For most people, the answer is in the range 7+/-2 (i.e., 5 to 9). Faloon was able to stretch his digit span to over 80 digits. I have an audiotape of him doing this for 25 digits and it's mind-bending to listen to it. The digits are read to him at a rate of about 1 per second. He pauses, strains, and struggles, and then spits them back out, perfectly.
How did he do it? By "chunking" the digits into running times running times that were meaningful to him (e.g., "4-1-2: my mile time in meet X back in 1978") and then by learning to chunk these chunks (and chunk these chunks of chunks (and
When John Watson, the father of American behaviorism, was kicked out of the academy for marrying his graduate student, he went to Madison Avenue and made a fortune as an advertising executive.
This was in the first half of the 20th century.
I agree, Kurweil does sound like a crazy man. And I don't appreciate the rather inflated list of achievements attributed to him.
However, I think I know where the 100,000 number came from. Research on expertise suggests that one must learn about 100,000 "chunks" of information to achieve expert status. Given the rate at which humans learn new chunks, this translates into 10 years of concerted practice. If you look at the time it takes people to attain world class status in various endeavours (e.g., chess, tennis), these numbers fit pretty well.
Bobby Fischer is commonly held up as an exception as he achieved a grandmaster rating 9 or so years after being introduced to chess. Prodigies like Mozart are not considered exceptions because apparently people distinguish between the works he produced as a child and the true masterworks of his adulthood.
Google "expertise" and "chunks" for more information, such as http://www.admin.upm.edu.my/~mzbd/expert.html.
Java is not necessarily borrowing from C(++). There are a lot of other programming languages out there which implement these features (e.g., varargs), some of which the Java developers (e.g., Guy Steele) designed decades ago (e.g., Common Lisp).
The scandal isn't that Java is borrowing from C++. It's that it has taken so long for "mainstream" languages to provide constructs that should have been there a long time ago.
This will not do. No one will buy records from a happy gangsta rapper.
Ob. quote: those who can't do, teach.
Those who can't do, teach.
Those who can't teach, teach gym.
Drug Dealer: I am validated by the masses. I sell lots of drugs and am rather profitable. Logically that means that something I am doing is filling a marketable demand.
(I think you should forget about "marketable demand" and simply say the RIAA's practices are allowed by the law.)
I was going to write a long reply but then you wouldn't be able to read it all as it wouldn't fit on your screen.
Don't worry, Andrew Wiles is working on a fix. Should be available in about ten years.
However, Marois and Vogel (the authors of the current papers) are very sharp guys, so I'm not going to second guess their conclusions before reading their paper. However, I am going to ding the science editor writing for Nature for bring the fuzzy, baggage-laden concept of "intelligence" into the mix. They seem to have found the visual working memory bottleneck...what this says about "intelligence" is pure speculation and a red herring.
Cogent post overall. I just want to offer a slightly different take. Vogel and Marois are certainly sharp guys. I don't consider them working memory researchers though. They study visual attention (and its capacity), and in this regard they sometimes brush up against working memory issues. And given the current fascination in cognitive science with low-level forms of cognition and the brain, many equate these two topics. I personally don't. There are researchers who study the role of working memory in high-level and ecologically interesting forms of cognition, such as Ericsson (sp?) and Engle. There are others who actually care about the intersection of measures of fluid intelligence and working memory, including Kyllonen and Just, and even Duncan, who's quoted on the Nature website. It's easy to look at all these intertwined threads and conclude that posterior parietal cortex is central to intelligence or whatever the article says. To me, this is just drawing an overly broad conclusion from what are, at their heart, simple tests of visual attention.
Mod parent up. She knows of which she speaks.
There are several reasons why a competent manager would refuse to promote you.
I find it humorous that management's perspective is being given by someone whose username is fingerfucker.
The article even quotes Pat Langley as saying this is nothing new. He mentions some system from the early 1990s as the true pioneer here, but he's just being modest. Langley, Simon, and colleagues published in Science back in the early 1980s on their scientific discovery programs.
This is an incremental advance perhaps, but not worthy of this kind of attention. Just goes to show that Nature and the like are as much about PR as they are about the genuinely new.
I mean, did the Elves make Glamdring and Sting and Orcrist and then FORGET what they just did?
Ever try to maintain old-but-working code written in a crufty language? It can be damn near impossible; you just use it as is and hope it keeps doing its magic.
Well said.
;-) We think alike.
Is it any coincidence that I've seen your name off-and-on on comp.lang.lisp over the years?
This is not real engineering.
I have a pet theory that the many software engineers who beat their chests about process and discipline have no idea what *real* engineers -- EEs, MEs, ChemEs, etc. -- know and what they do on a day to day basis.
Let me give you a hint: Grady Booch, he ain't no Newton.
Would you rather those "young indonesian girls" made their $1.80 the old-fashioned way?
I have never heard a credible medical professional assert that the output of an MRI provides any insight into the feelings or thoughts of the patient. It's made for detecting tumors and stuff like that.
You're half right. MRI is used for probing the static structure of the brain, e.g., in search of tumors. functional MRI (fMRI) is used for measuring the dynamics of blood flow in the brain during performance of a task. It can, for example, indicate which areas of the brain are active when processing "emotional" stimuli, and therefore may be useful for advertisers who want an emotional, not logical reaction (e.g., the makers of weight loss supplements).
Good response. I just have two caveats.
The idea is that areas that are being actively used will get increased blood flow, but this happens on the order of tens of seconds, so it doesn't really provide a detailed picture of brain activity.
There are two kinds of experimental designs one can employ with fMRI. You are referring to "block" designs, where in fact the neural response is aggregated over tens of seconds. However, "event-related" designs permit temporal resolutions of 1.5 seconds, and I have seen some studies that clam 0.5 seconds. So, fMRI can do a bit better time-wise than you think.
The other dimension of interest is spatial resolution. Just like a computer screen is composed of 2D pixels, fMRI partitions brains into 3D voxels. The smallest voxels sizes that current technologies allow is on the order of 10s of cubic mms. However, Logothetis and colleagues have pioneered more invasive techniques that allow voxels that are several orders of magnitude smaller. The only drawback is that they can't be used in humans.
However, advertisers can gain minute knowledge of how animals respond to different ads! Perhaps Alpo, Purina, and the like should use neuromarketing.
By using MRI, scientists can know what parts of the brain are / may be stimulated by ads, so what kind of feelings we got when seeing / hearing it..
They can't tell what "feelings" a person is experiencing. Emotions aren't individually localizable to regions of the brain.
They can tell whether an area of the brain generally responsible for emotions (e.g., orbitofrontal cortex) is being engaged versus one generally responsible for deliberate reasoning (e.g., the dorsolateral prefrontal cortex). This could reveal whether an ad prompts emotional or logical responses, generally speaking.
But that's about it given the limitations of the technology and cognitive neuroscience.
Disclaimer: The site was borderline slashdotted so I couldn't read the source article.
Marketplace on CBC, that's a Canadian station for you Americans...
It's not just Americans who haven't heard of CBC.
No one outside Canada cares about the call letters of your pissant stations.
And what does it get you? [...] Maybe you're getting a 20% across the board improvement, but _man_ are you paying for it, both in cost and power consumption. And was it worth it, for 27% faster than "more speed than I know what to do with?" Probably not (though I realize that all hardware site weenies will absolutely insist that they can feel the difference when browsing the web on such a machine).
Tubes versus solid-state...
Beta versus VHS...
Vinyl records versus CDs...
Air-cooled versus water-cooled...
Word is a ripoff from WordPerfect
You and several other people in this thread have it dead wrong. Microsoft hired Charles Simonyi from Xerox PARC in the early 1980s. Word was his crappy attempt to deliver a more modern word processor to the PC, whose users at the time were willing to put up with shit like the "dot commands" of WordStar.
(No flames from the old timers. I'm an old timer myself. But the facts is the facts. And if you don't like 'em, I've got an Exidy Sorcerer to sell you.)
Anyway, Word was trounced in the market by WordPerfect and even marginal programs like XYWrite. In the late 1980s, Microsoft turned it into a usable product on the Macintosh, good enough that Apple/Claris let their own MacWrite blend into the background. (This was back when people were convinced Apple wanted to control both Mac hardware and software, and so Apple was willing to de-emphasize its office productivity products -- MacWrite, MacPaint, MacDraw, etc.)
When Windows 3.0 came out, Microsoft had a leg up on their PC competition making GUI-based word processors from their Mac experience. They beat their competitors to the market. This is when Word and Excel really trounced their WordPerfect, Lotus, and Quattro.
(And to the people lamenting the loss of Ami Pro: It was only good if you were stuck on stone age PCs.)
Remarkably, nuclear detonations are not a good option, as they would break the asteroid into many pieces and merely increase our odds of being hit.
Clearly, the pointdexter astrophysicists who offered this opinion have never seen Armageddon.