Plus, who really buys a new PC without a new monitor? If they're in the Apple store, a monitor will set them back $1299 or more since they moved to only top-of-the-line LCD screens. Might as well go for the iMac.
Yes, the slashdot hordes who have three PCs in their bedroom will always have a spare monitor laying around, but notice the sub-$1000 entry-level offerings from Dell include a monitor.
Not to mention that those ready for a PC upgrade probably have a cheezy old bulky CRT monitor, and they realize that the new thing is a sleek flat-panel display.
On-demand cable transmission does not resemble iTunes, at least to my mind.
There's no question that movies & TV are already being affected by the digital revolution. I was just opposing the assumption people have been making that there is some magical reason Steve Jobs + iPod + iTunes will be just as successful in movies/TV as they have been in music.
I don't get how TV is a must-have for a PC. Nice-to-have, perhaps.
The main thing that gets me is that TV is a huge mess of schedule, ownership, and channel chaos, not to mention the analog/digital/HD variety. The iPod/iTunes combo is all a much more controlled case: a stack of personal CDs and one online store.
One *might* be able to do this for movies. But it isn't going to be easy for television. I can already play DVDs on my Mac. What more do I really need?
You know, it's not clear that movies will follow the same track as music.
* Record labels make money by selling albums over a relatively long period of time. * You only need one or two good singles to sell an album. * They push singles through the commercial, but not-for-pay radio [does MTV actually show music videos anymore?] * Actual concert performances profit the band, not the record label * They have *always* had to deal with the possibility of taping-off-radio and taping-off-CD
* Movies are much more expensive to make than albums. And probably riskier creatively. * Movie studios make a bunch of money on live performances in movie theatres. They will hesitate to dissipate that by releasing simultaneously to consumer digital. (Although there are huge advantages to digital transmission to theaters.) * They make a second chunk of money selling hard copy DVDs *once the first run revenue* is tapped out. * Finally, once the DVD stream is largely tapped out, they'll make a chunk of money selling the TV broadcast rights. * For now, the primary medium is heavy, bulky, film prints on reels, which are hard to pirate, except through sucky camcorder taping.
The whole rhythm of release and commercial structure is different for these two industries. Probably, they'll make the transition to digital quite differently.
I think your LDEF example may have the cart before the horse.
Is there really a demand for LDEF capabilities, or was LDEF simply advertisement for the Shuttle's retrieval ability---we have this big box of *stuff* we leave in orbit, and we bring it back to Earth later, because we *can*?
I think the original poster's point was that the composite chemicals are just as much a "pre-existing" thing as the "natural" stuff that gets ground up to make "all-natural" foods.
The coal or petroleum or minerals which were the chemical feedstock for the "synthetic" chemicals were dug out of the natural ground. In that sense, they are just as natural as a carrot. In fact, the carrot was only there because a farmer "artificially" planted it, while the oil has been in the ground since long before man.
That said, when I see a Twinkie, I am hesitant to call it "food" as opposed to "chemical concoction."
The Soviet Union's GDP was hardly improved by it being a nuclear superpower, or having satellite countries. Compare western and eastern Germany, for instance, even this long after reunification.
As for innovation, who invented the transistor, the television, the laser, Fortran, COBOL, Lisp, C, multi-tasking operating systems, UNIX, the integrated circuit, the microprocessor, the personal computer, the cell phone, the handheld calculator, and who brought them to commercial fruition?
Brazilian inventors and companies? Operating under the Brazilian patent rules? Soviet inventors and companies forcing their satellites to accept their innovative technologies? Or American inventors, operating under American patent rules?
Your emphasis of the US being a nuclear power is irrelevant, and simply marks you as some kind of non-aligned apologist. What, you think if Brazilians or Indians had invented the integrated circuit, the US would have nuked them?
The economic dominance of the US during the twentieth century was absolutely independent of the development of nuclear weapons. If anything, it was the other way around; the US could focus an enormous amount of economic resources on the Manhattan project, simply because the US economy was so large and productive, fueled by a huge industrial base that preceded the development of nuclear power. Germany could not afford a similar effort.
You have yet to offer ONE CONCRETE EXAMPLE of significant Brazilian innovation. Just looking around my office, I came up with a dozen American innovations, all of which have US patent numbers on the bottom, and none of which have anything to do with nuclear arms.
Face it: American military dominance is a result of economic development, not the other way around. The results of the American patent system are demonstrated by looking at the American stock markets and seeing companies like IBM, DuPont, Intel, Texas Instruments, Microsoft, Motorola, and on and on, all driven by massive technological innovation, and prospering under the American patent regime. Sixty years ago, before the nuclear age, we would have seen a similar list.
When I see start hearing about Sao Paolo stock exchange results on the BBC World Service business report, then you might be able to begin talking about the fruits of Brazilian innovation.
I see. Your obviously coherent and rational view of economic development speaks strongly for your views on patent law.
Not.
You have not provided any objective evidence that the Brazilian patent system is producing superior results to the American one. Getting slashdotters' panties in a bunch does not qualify as a significant flaw in the US patent system.
I'm not sure you are the same AC who posted the original message, but in any case, I am confused by your message.
In particular, you seem to have a very strange view of what "the US" is. Whether US naval vessels are powered by nuclear reactors, whether nuclear power plants are built for US civilian power generation, and what to do about foreign nuclear development, are determined by totally different and independent decision-making processes.
The fact is that the US government doesn't have any problem with Canada or Japan or China or France or Germany building all the nuclear power plants they want. The US does have a problem with Iran or highly militarized countries like North Korea or free agents like A Q Khan developing nuclear technology which is suitable for weapons manufacture.
To clarify slightly, multiple dispatch means you have
(method object object object object...)
Where of course, some of the objects can be of unspecified class, so that the actual implementation of the method need not depend on specifying every possible combination. And, unlike some OO languages, every built-in type has a corresponding class to dispatch on.
Show me the GDP to back up the usefulness of the innovation and get back to me. Even using purchasing-power-parity figures, the per capita GDP of Brazil was US$7250 in 2002 (US$2630 nominal). Compared to US$37000 for the U.S.
That factor of something between 5 and 15 might have something to do with the patent system.
Naval reactors by no means prove that nuclear energy doesn't produce waste. Just that the cost of handling the waste, and the environmental standards that military operations are held to, are better than other military solutions. The military does all sorts of things that would not be acceptable for civilian purposes. Look at all the environmental problems the former Soviet Union is having with their rusting nuclear fleet.
The U.S. government is, in theory, working against nuclear power solutions that are not proliferation resistant. The current international treaty regime makes that very hard; basically you have to trust countries (like India) that say their nuclear research is peaceful only, until their first weapons test proves they were less than honest.
It is the environmental activists who chain themselves to railroad tracks to prevent nuclear fuel and waste transport, and general civilian fear that prevents new reactors from being proposed. Not official government policy.
Fusion in the sun has happened for 15 billion years. Controlled nuclear fusion has been around since the 50's, in the H-bomb. Controlled laboratory fusion has been around since the 70s(?) Laboratory fusion that has more than breakeven power generation has been demonstrated in the 90s (??)
ITER will provide another step toward demonstrating nuclear fusion above break-even on a commercial scale.
There are many steps still remaining:
- demonstrating commercial-scale electric power extraction using a fusion reactor vessel - commercial-scale nuclear fusion that can be sustained for the life of a power plant (decades) - commercial scale support for the nuclear fuel and nuclear waste issues for fusion
It is quite optimistic to claim that in 15 years we would be able to build a commercial-scale fusion power plant which would have better return on capital than a coal-, gas-, or fission-fired power plant. *That* is what is far away. ITER will not change the commercial reality, just the technological knowledge; i.e., knowing how to build a power plant does not make it commercially viable.
You really believe that within 15 years one could plug a computer into a home wall socket and be running on fusion-supplied electricity for the same or less cost than gas- or coal-fired electricity?
Would you care to explain how programming with setf, incf, rotatef is functional programming?
Most functional programming languages make you jump through special hoops to do I/O to avoid side-effects.
Lisp has many of the aspects of functional programming fully and properly supported, such as
- functions as first-class objects (can be assigned, passed, and returned, just like objects of other types) - apply/funcall/map... to allow function application under program control - closures allowing functions to be constructed at run time
(It lacks good built-in support for higher level functional operations like currying.)
But it also has things like set, nconc, eval, etc., that allow function calls to destructively modify the environment, i.e., have side-effects. That is completely antithetical to modern functional programming.
Hell, Lisp even has GO, so you can write 1950's spaghetti Fortran if you want.
Look, the reason Lisp has the syntax it does is exactly *because* it allows programs to parse it easily.
That's not so exciting for someone who programs in C/C++/Java, because only one program needs to parse your program, namely, the compiler. (OK, also your IDE to do syntax highlighting and name completion.)
But in Lisp, almost everyone writes programs that depend on parsing and rewriting Lisp programs. Informally, we call these program-writing programs "Lisp macros," but that always gives outsiders the wrong idea. They are basically nothing like #define in C. Templates in C++ are closer, but are infinitely less flexible.
Now, if the template definitions let you write explicit code, using the STL, for instance, which let you fully specify the resulting C++ code from the template expansion, you'd be getting somewhere. But, then, you quickly realize that the syntax of C++ is in fact too complicated, non-uniform, and context-dependent to make that at all easy.
So, from a Lisp programmer's point of view, C++'s syntax ranks pretty much number one on its list of problems.
Keep in mind that Lisp programmers way-back-when deliberately chose to keep the S-expression (i.e., parenthesized prefix notation) syntax around even though the language designer intended the S-expressions for meta-discussions about the language, and M-expressions (more like Fortran/Algol) for the actual programming.
I think the question is not the use of click data to directly represent the value to the advertiser so much as it is for Google to represent the value of the advertising channel.
Think of Nielsen ratings which allow TV networks to set their ad rates for time slots, into which advertisers can stick basically whatever good or annoying ad they want into the space. Ad rates don't tell you that your ad is going to be successful, but they do tell you that your ad is going to be broadcast to approximately N million people who, just seconds prior, were sitting slackjawed in their couches and recliners with their eyes open and pointing toward the glowing box. At that point, it is up to your ad agency to do something worthwhile with that opportunity.
TV networks make money when their programming provides the Nielsen rating that helps them make the sales pitch for ad time. They don't actually care whether your ads work or not.
Google wants to show click data to prove that targeted ads are a more cost-effective *channel* for advertising than some other channel (that, in addition, probably allows you to put Flash in your ads.) If the click data is obviously bogus, then Google's leverage is decreased. If the click data is demonstrably better than someone else's click data because the clicks are better validated, then Google's leverage is increased.
The issue with performance of things like the FFT is not really with the language, per se, but with the compilers.
In fact, a naive implementation of the FFT is going to suck in C also. Because FFT performance is not determined by the expression of the algorithm so much as it is determined by the actual bits-meets-metal details, such as instruction pipelining and memory architectures.
The way you get efficient FFT algorithms is not just by writing them in C, but having a combination of a library writer and compiler writer who both know how to get the absolute peak performance out of a particular processor (and cache hardware.)
It just happens that C compilers have included enough hooks to low-level processor primitives, and C serves as a universal "high-level" assembler, so that you can create efficient FFTs that link well to C, and might even be written in C. (Although you better make sure to use the right C compiler or your performance will suffer.)
The reality is that we are just slightly advanced from the need to write performance bottlenecks in assembler. Somebody sufficiently skilled has to write the routines in carefully tweaked C, and use the best available C compiler to generated architecture-specific libraries.
If Intel (or even the gcc architects) built OCaml compilers that knew as much about IA-86 as their C compilers do, then you wouldn't be complaining.
*I* never said you couldn't do it. It just doesn't convince me that C++ is easier. In fact, it tells me that C++ is becoming ever more baroque and more biased toward the gurus over the years.
I consider myself quite comfortable and experienced with C. I only have about six months of serious Fortran learning in my past, based on Fortran 77. I had about six months of serious C++ learning, but back before Bjarne and company realized that RTTI, STL, and C++ casting actually belonged in the language, and when templates were new on the scene, and barely worked in available compilers.
Given a similar preparation in Fortran and C++, I find the Fortran example much clearer and less intimidating than the C++ one. If I had a problem with the C++ example, because of a typo, let's say, or because I forgot one of the casts, I'm not sure I would be able to debug it. If I had a problem with the Fortran one, I'm sure I could make it work the way it is supposed to.
In fact, your use of the term "well-programmed" is a giveaway. Even a mediocre Fortran programmer would be able to use the Fortran example. The average C++-in-30-days type might not get the C++ example right.
You still don't understand the second law. The universe may or may not be closed in the thermodynamic sense (because universal expansion may add energy), but that is beside the point. The earth or an organism is just a small part. The second law of thermodynamics does not say anything about these small parts if they are connected to larger parts.
Are you living and breathing as you read this? Your metabolism is a local decrease in entropy, unless you remember to count all of the chemical inputs and outputs (food, oxygen, carbon dioxide). The second law doesn't say this is impossible. But if I put you in a box without letting *anything* in or out, you will eventually keel over, die, and disintegrate into a less ordered system than you are now. But you aren't in such a box, are you?Yes, eventually, the solar system will not be able to support life. For that reason, don't expect to live more than 10 billion years or so. But you have plenty of thermodynamic room to live a very long life without violating the second law.
This kind of profound ignorance is what is most frustrating about creationists. Anybody who takes the time to honestly understand thermodynamics would see that the second law does not prohibit the Darwinian theory. Yet, you people keep trotting it out. At some point, one has to believe this kind of ignorance is just a deliberate attempt at self-delusion to avoid having to believe the truth.
Scientific theories are not meant to make you all warm and comfortable so that you can say "oh, my vision of the universe is both pleasing to my psychological need for a personal God and in accordance with scientific observation." Feeling vaguely uncomfortable about the implications of a theory is not a sign that you will easily be able to find evidence against it. Trying to collect half-baked thermodynamic arguments absolutely and completely destroys the credibility of your claim to be open-minded.
I don't understand at all what you mean by "cross-over" species. Species form an incredibly dense family tree with lots of branches. At any point in history, there are species that are succeeding, species that are marginal and perhaps are dying out, species that are becoming more specialized in their niches, and species whose niche is disappearing, but are able to find a new niche.
The most plausible story of the evolution of birds is something like this: dinosaurs developing feather-like scales, and also dinosaurs that begin to develop structures that support gliding behaviors. The idea is that feathers have advantages over scales for thermal regulation. Gliding is obviously useful for quickly getting to food on the ground from a high observation point, or being able to safely escape a heavier predator coming after you in a tree. The order doesn't really matter too much. Once you get gliding dinosaurs, the evolution toward reduced bone density makes sense: lighter gliders can glide further and longer. Then, there is evolution toward powered flight, as the gliders develop body structures that allow for strong flapping.
My point on the ostrich and the flying squirrel was to show that historical intermediate forms are not at all unimaginable, and, in fact, would look pretty much like animals that exist today. They are not "in-between" anything, except their parents and their children, just as the historical intermediate forms were in-between their parents and their children.
The fact that you can't make that connection shows to me that you are not interested in giving evolutionary models an honest attempt at understanding. If I try to explain multiplication to a five-year old, and they don't try, they'll give up, and think I'm just making up this stuff. But that doesn't mean multiplication isn't a rigorous theory, it means they haven't had the preparation to understand it. For a five-year old, it is a matter of mental development. At some point, someone who doesn't believe 6 times 9 is 73, and refuses to be convinced, is either mentally damaged, or phobic. If he nonetheles
Perhaps I missed something, but I think your C++ example proved his point. The Fortran technique is about 100 times easier for a mid-level programmer to understand, or anyone who hasn't programmed C++ since they actually made templates work.
Plus, I don't think you customized the formatting of the output as he did. And you aren't printing doubles.
The only one I've used very much under Mac OS X is OpenMCL running under Emacs. For the stuff I do on my Mac, mostly noodling around with hobby-level projects, debugging at the read-eval-print loop, it's all I need.
For someone who wants to do Mac OS X, it is advertised to support the Cocoa/Objective C model of development.
I used Digitool's OS 9 environment for some technical programming, and found it to be excellent. I have not tried it under Mac OS X. Franz & Xanalys didn't really play on classic Mac as I recall.
For delivery of applications, I would expect the commercial versions to be attractive for portability over multiple platforms, and integrated development environments. I downloaded the trial edition of LispWorks; it felt heavyweight for my needs.
I have not tried any Lisp (or any language platform) for what I would consider serious software development. I work on technical projects and mostly use software for little one-off tools.
Slashdot Mirror
Plus, who really buys a new PC without a new monitor? If they're in the Apple store, a monitor will set them back $1299 or more since they moved to only top-of-the-line LCD screens. Might as well go for the iMac.
Yes, the slashdot hordes who have three PCs in their bedroom will always have a spare monitor laying around, but notice the sub-$1000 entry-level offerings from Dell include a monitor.
Not to mention that those ready for a PC upgrade probably have a cheezy old bulky CRT monitor, and they realize that the new thing is a sleek flat-panel display.
Interesting; but that supports my point, I think.
On-demand cable transmission does not resemble iTunes, at least to my mind.
There's no question that movies & TV are already being affected by the digital revolution. I was just opposing the assumption people have been making that there is some magical reason Steve Jobs + iPod + iTunes will be just as successful in movies/TV as they have been in music.
I don't get how TV is a must-have for a PC. Nice-to-have, perhaps.
The main thing that gets me is that TV is a huge mess of schedule, ownership, and channel chaos, not to mention the analog/digital/HD variety. The iPod/iTunes combo is all a much more controlled case: a stack of personal CDs and one online store.
One *might* be able to do this for movies. But it isn't going to be easy for television. I can already play DVDs on my Mac. What more do I really need?
You know, it's not clear that movies will follow the same track as music.
* Record labels make money by selling albums over a relatively long period of time.
* You only need one or two good singles to sell an album.
* They push singles through the commercial, but not-for-pay radio [does MTV actually show music videos anymore?]
* Actual concert performances profit the band, not the record label
* They have *always* had to deal with the possibility of taping-off-radio and taping-off-CD
* Movies are much more expensive to make than albums. And probably riskier creatively.
* Movie studios make a bunch of money on live performances in movie theatres. They will hesitate to dissipate that by releasing simultaneously to consumer digital. (Although there are huge advantages to digital transmission to theaters.)
* They make a second chunk of money selling hard copy DVDs *once the first run revenue* is tapped out.
* Finally, once the DVD stream is largely tapped out, they'll make a chunk of money selling the TV broadcast rights.
* For now, the primary medium is heavy, bulky, film prints on reels, which are hard to pirate, except through sucky camcorder taping.
The whole rhythm of release and commercial structure is different for these two industries. Probably, they'll make the transition to digital quite differently.
Actually, the contract manufacturer has almost certainly executed an NDA as part of the contract.
I think your LDEF example may have the cart before the horse.
Is there really a demand for LDEF capabilities, or was LDEF simply advertisement for the Shuttle's retrieval ability---we have this big box of *stuff* we leave in orbit, and we bring it back to Earth later, because we *can*?
I think the original poster's point was that the composite chemicals are just as much a "pre-existing" thing as the "natural" stuff that gets ground up to make "all-natural" foods.
The coal or petroleum or minerals which were the chemical feedstock for the "synthetic" chemicals were dug out of the natural ground. In that sense, they are just as natural as a carrot. In fact, the carrot was only there because a farmer "artificially" planted it, while the oil has been in the ground since long before man.
That said, when I see a Twinkie, I am hesitant to call it "food" as opposed to "chemical concoction."
I think you mean to refer to OOG, the OPEN SOURCE CAVEMAN. Although I sometimes enjoy OOGG.
The Soviet Union's GDP was hardly improved by it being a nuclear superpower, or having satellite countries. Compare western and eastern Germany, for instance, even this long after reunification.
As for innovation, who invented the transistor, the television, the laser, Fortran, COBOL, Lisp, C, multi-tasking operating systems, UNIX, the integrated circuit, the microprocessor, the personal computer, the cell phone, the handheld calculator, and who brought them to commercial fruition?
Brazilian inventors and companies? Operating under the Brazilian patent rules? Soviet inventors and companies forcing their satellites to accept their innovative technologies? Or American inventors, operating under American patent rules?
Your emphasis of the US being a nuclear power is irrelevant, and simply marks you as some kind of non-aligned apologist. What, you think if Brazilians or Indians had invented the integrated circuit, the US would have nuked them?
The economic dominance of the US during the twentieth century was absolutely independent of the development of nuclear weapons. If anything, it was the other way around; the US could focus an enormous amount of economic resources on the Manhattan project, simply because the US economy was so large and productive, fueled by a huge industrial base that preceded the development of nuclear power. Germany could not afford a similar effort.
You have yet to offer ONE CONCRETE EXAMPLE of significant Brazilian innovation. Just looking around my office, I came up with a dozen American innovations, all of which have US patent numbers on the bottom, and none of which have anything to do with nuclear arms.
Face it: American military dominance is a result of economic development, not the other way around. The results of the American patent system are demonstrated by looking at the American stock markets and seeing companies like IBM, DuPont, Intel, Texas Instruments, Microsoft, Motorola, and on and on, all driven by massive technological innovation, and prospering under the American patent regime. Sixty years ago, before the nuclear age, we would have seen a similar list.
When I see start hearing about Sao Paolo stock exchange results on the BBC World Service business report, then you might be able to begin talking about the fruits of Brazilian innovation.
I see. Your obviously coherent and rational view of economic development speaks strongly for your views on patent law.
Not.
You have not provided any objective evidence that the Brazilian patent system is producing superior results to the American one. Getting slashdotters' panties in a bunch does not qualify as a significant flaw in the US patent system.
I'm not sure you are the same AC who posted the original message, but in any case, I am confused by your message.
In particular, you seem to have a very strange view of what "the US" is. Whether US naval vessels are powered by nuclear reactors, whether nuclear power plants are built for US civilian power generation, and what to do about foreign nuclear development, are determined by totally different and independent decision-making processes.
The fact is that the US government doesn't have any problem with Canada or Japan or China or France or Germany building all the nuclear power plants they want. The US does have a problem with Iran or highly militarized countries like North Korea or free agents like A Q Khan developing nuclear technology which is suitable for weapons manufacture.
By the way, most people who talk about Lisp mean Common Lisp. A useful on-line spec is here.
Consider, for instance, the section on iteration.
To clarify slightly, multiple dispatch means you have
...)
(method object object object object
Where of course, some of the objects can be of unspecified class, so that the actual implementation of the method need not depend on specifying every possible combination. And, unlike some OO languages, every built-in type has a corresponding class to dispatch on.
Show me the GDP to back up the usefulness of the innovation and get back to me. Even using purchasing-power-parity figures, the per capita GDP of Brazil was US$7250 in 2002 (US$2630 nominal). Compared to US$37000 for the U.S.
That factor of something between 5 and 15 might have something to do with the patent system.
You are totally out to lunch.
Naval reactors by no means prove that nuclear energy doesn't produce waste. Just that the cost of handling the waste, and the environmental standards that military operations are held to, are better than other military solutions. The military does all sorts of things that would not be acceptable for civilian purposes. Look at all the environmental problems the former Soviet Union is having with their rusting nuclear fleet.
The U.S. government is, in theory, working against nuclear power solutions that are not proliferation resistant. The current international treaty regime makes that very hard; basically you have to trust countries (like India) that say their nuclear research is peaceful only, until their first weapons test proves they were less than honest.
It is the environmental activists who chain themselves to railroad tracks to prevent nuclear fuel and waste transport, and general civilian fear that prevents new reactors from being proposed. Not official government policy.
Depends very much on what you mean by "fusion."
Fusion in the sun has happened for 15 billion years.
Controlled nuclear fusion has been around since the 50's, in the H-bomb.
Controlled laboratory fusion has been around since the 70s(?)
Laboratory fusion that has more than breakeven power generation has been demonstrated in the 90s (??)
ITER will provide another step toward demonstrating nuclear fusion above break-even on a commercial scale.
There are many steps still remaining:
- demonstrating commercial-scale electric power extraction using a fusion reactor vessel
- commercial-scale nuclear fusion that can be sustained for the life of a power plant (decades)
- commercial scale support for the nuclear fuel and nuclear waste issues for fusion
It is quite optimistic to claim that in 15 years we would be able to build a commercial-scale fusion power plant which would have better return on capital than a coal-, gas-, or fission-fired power plant. *That* is what is far away. ITER will not change the commercial reality, just the technological knowledge; i.e., knowing how to build a power plant does not make it commercially viable.
You really believe that within 15 years one could plug a computer into a home wall socket and be running on fusion-supplied electricity for the same or less cost than gas- or coal-fired electricity?
Yeah, because Brazil's patent system has led to its domination in high-tech innovation, leading to trillions of dollars of economic benefit.
Not.
Would you care to explain how programming with setf, incf, rotatef is functional programming?
Most functional programming languages make you jump through special hoops to do I/O to avoid side-effects.
Lisp has many of the aspects of functional programming fully and properly supported, such as
- functions as first-class objects (can be assigned, passed, and returned, just like objects of other types)
- apply/funcall/map... to allow function application under program control
- closures allowing functions to be constructed at run time
(It lacks good built-in support for higher level functional operations like currying.)
But it also has things like set, nconc, eval, etc., that allow function calls to destructively modify the environment, i.e., have side-effects. That is completely antithetical to modern functional programming.
Hell, Lisp even has GO, so you can write 1950's spaghetti Fortran if you want.
Look, the reason Lisp has the syntax it does is exactly *because* it allows programs to parse it easily.
That's not so exciting for someone who programs in C/C++/Java, because only one program needs to parse your program, namely, the compiler. (OK, also your IDE to do syntax highlighting and name completion.)
But in Lisp, almost everyone writes programs that depend on parsing and rewriting Lisp programs. Informally, we call these program-writing programs "Lisp macros," but that always gives outsiders the wrong idea. They are basically nothing like #define in C. Templates in C++ are closer, but are infinitely less flexible.
Now, if the template definitions let you write explicit code, using the STL, for instance, which let you fully specify the resulting C++ code from the template expansion, you'd be getting somewhere. But, then, you quickly realize that the syntax of C++ is in fact too complicated, non-uniform, and context-dependent to make that at all easy.
So, from a Lisp programmer's point of view, C++'s syntax ranks pretty much number one on its list of problems.
Keep in mind that Lisp programmers way-back-when deliberately chose to keep the S-expression (i.e., parenthesized prefix notation) syntax around even though the language designer intended the S-expressions for meta-discussions about the language, and M-expressions (more like Fortran/Algol) for the actual programming.
I think the question is not the use of click data to directly represent the value to the advertiser so much as it is for Google to represent the value of the advertising channel.
Think of Nielsen ratings which allow TV networks to set their ad rates for time slots, into which advertisers can stick basically whatever good or annoying ad they want into the space. Ad rates don't tell you that your ad is going to be successful, but they do tell you that your ad is going to be broadcast to approximately N million people who, just seconds prior, were sitting slackjawed in their couches and recliners with their eyes open and pointing toward the glowing box. At that point, it is up to your ad agency to do something worthwhile with that opportunity.
TV networks make money when their programming provides the Nielsen rating that helps them make the sales pitch for ad time. They don't actually care whether your ads work or not.
Google wants to show click data to prove that targeted ads are a more cost-effective *channel* for advertising than some other channel (that, in addition, probably allows you to put Flash in your ads.) If the click data is obviously bogus, then Google's leverage is decreased. If the click data is demonstrably better than someone else's click data because the clicks are better validated, then Google's leverage is increased.
The issue with performance of things like the FFT is not really with the language, per se, but with the compilers.
In fact, a naive implementation of the FFT is going to suck in C also. Because FFT performance is not determined by the expression of the algorithm so much as it is determined by the actual bits-meets-metal details, such as instruction pipelining and memory architectures.
The way you get efficient FFT algorithms is not just by writing them in C, but having a combination of a library writer and compiler writer who both know how to get the absolute peak performance out of a particular processor (and cache hardware.)
It just happens that C compilers have included enough hooks to low-level processor primitives, and C serves as a universal "high-level" assembler, so that you can create efficient FFTs that link well to C, and might even be written in C. (Although you better make sure to use the right C compiler or your performance will suffer.)
The reality is that we are just slightly advanced from the need to write performance bottlenecks in assembler. Somebody sufficiently skilled has to write the routines in carefully tweaked C, and use the best available C compiler to generated architecture-specific libraries.
If Intel (or even the gcc architects) built OCaml compilers that knew as much about IA-86 as their C compilers do, then you wouldn't be complaining.
*I* never said you couldn't do it. It just doesn't convince me that C++ is easier. In fact, it tells me that C++ is becoming ever more baroque and more biased toward the gurus over the years.
I consider myself quite comfortable and experienced with C. I only have about six months of serious Fortran learning in my past, based on Fortran 77. I had about six months of serious C++ learning, but back before Bjarne and company realized that RTTI, STL, and C++ casting actually belonged in the language, and when templates were new on the scene, and barely worked in available compilers.
Given a similar preparation in Fortran and C++, I find the Fortran example much clearer and less intimidating than the C++ one. If I had a problem with the C++ example, because of a typo, let's say, or because I forgot one of the casts, I'm not sure I would be able to debug it. If I had a problem with the Fortran one, I'm sure I could make it work the way it is supposed to.
In fact, your use of the term "well-programmed" is a giveaway. Even a mediocre Fortran programmer would be able to use the Fortran example. The average C++-in-30-days type might not get the C++ example right.
You still don't understand the second law. The universe may or may not be closed in the thermodynamic sense (because universal expansion may add energy), but that is beside the point. The earth or an organism is just a small part. The second law of thermodynamics does not say anything about these small parts if they are connected to larger parts.
Are you living and breathing as you read this? Your metabolism is a local decrease in entropy, unless you remember to count all of the chemical inputs and outputs (food, oxygen, carbon dioxide). The second law doesn't say this is impossible. But if I put you in a box without letting *anything* in or out, you will eventually keel over, die, and disintegrate into a less ordered system than you are now. But you aren't in such a box, are you?Yes, eventually, the solar system will not be able to support life. For that reason, don't expect to live more than 10 billion years or so. But you have plenty of thermodynamic room to live a very long life without violating the second law.
This kind of profound ignorance is what is most frustrating about creationists. Anybody who takes the time to honestly understand thermodynamics would see that the second law does not prohibit the Darwinian theory. Yet, you people keep trotting it out. At some point, one has to believe this kind of ignorance is just a deliberate attempt at self-delusion to avoid having to believe the truth.
Scientific theories are not meant to make you all warm and comfortable so that you can say "oh, my vision of the universe is both pleasing to my psychological need for a personal God and in accordance with scientific observation." Feeling vaguely uncomfortable about the implications of a theory is not a sign that you will easily be able to find evidence against it. Trying to collect half-baked thermodynamic arguments absolutely and completely destroys the credibility of your claim to be open-minded.
I don't understand at all what you mean by "cross-over" species. Species form an incredibly dense family tree with lots of branches. At any point in history, there are species that are succeeding, species that are marginal and perhaps are dying out, species that are becoming more specialized in their niches, and species whose niche is disappearing, but are able to find a new niche.
The most plausible story of the evolution of birds is something like this: dinosaurs developing feather-like scales, and also dinosaurs that begin to develop structures that support gliding behaviors. The idea is that feathers have advantages over scales for thermal regulation. Gliding is obviously useful for quickly getting to food on the ground from a high observation point, or being able to safely escape a heavier predator coming after you in a tree. The order doesn't really matter too much. Once you get gliding dinosaurs, the evolution toward reduced bone density makes sense: lighter gliders can glide further and longer. Then, there is evolution toward powered flight, as the gliders develop body structures that allow for strong flapping.
My point on the ostrich and the flying squirrel was to show that historical intermediate forms are not at all unimaginable, and, in fact, would look pretty much like animals that exist today. They are not "in-between" anything, except their parents and their children, just as the historical intermediate forms were in-between their parents and their children.
The fact that you can't make that connection shows to me that you are not interested in giving evolutionary models an honest attempt at understanding. If I try to explain multiplication to a five-year old, and they don't try, they'll give up, and think I'm just making up this stuff. But that doesn't mean multiplication isn't a rigorous theory, it means they haven't had the preparation to understand it. For a five-year old, it is a matter of mental development. At some point, someone who doesn't believe 6 times 9 is 73, and refuses to be convinced, is either mentally damaged, or phobic. If he nonetheles
Perhaps I missed something, but I think your C++ example proved his point. The Fortran technique is about 100 times easier for a mid-level programmer to understand, or anyone who hasn't programmed C++ since they actually made templates work.
Plus, I don't think you customized the formatting of the output as he did. And you aren't printing doubles.
The only one I've used very much under Mac OS X is OpenMCL running under Emacs. For the stuff I do on my Mac, mostly noodling around with hobby-level projects, debugging at the read-eval-print loop, it's all I need.
For someone who wants to do Mac OS X, it is advertised to support the Cocoa/Objective C model of development.
I used Digitool's OS 9 environment for some technical programming, and found it to be excellent. I have not tried it under Mac OS X. Franz & Xanalys didn't really play on classic Mac as I recall.
For delivery of applications, I would expect the commercial versions to be attractive for portability over multiple platforms, and integrated development environments. I downloaded the trial edition of LispWorks; it felt heavyweight for my needs.
I have not tried any Lisp (or any language platform) for what I would consider serious software development. I work on technical projects and mostly use software for little one-off tools.