It would be better if Trump just tweeted by sending text messages to his staff, who can then post it. That would also add an extra level of protection against making a fool of himself.
Kids from families with high incomes have significantly higher test scores; highly competitive universities will therefore overwhelmingly select from high-income families even if they exclusively select based on test scores. So, there is nothing particularly surprising about this result, nor does it demonstrate any kind of discrimination of selective colleges against low income kids.
You can now debate about whether high income causes kids to have high test scores, i.e., if you only gave kids from poor families more money, they'd be doing just as well. That is true to some very limited degree: kids who lack essentials (food, clean water, etc.) are held back by that, but fixing those problems can't increase their intelligence beyond their potential.
Most of the correlation is likely primarily caused by the fact that smart parents tend to have smart kids (through a combination of nature and nurture), and that high test scores and high incomes simply result from that.
I didn't say "all languages based on the JVM are useless/toys", I said "languages targeting the JVM get bogged down by the limitations of the JVM and the get entangled in the Java libraries."
That theory leads to a prediction: once the law is passed, the gap should start to shrink. So you wait a while and sure enough, it does
Again, you are making an elementary logic error. The theory says "If P then Q" (P="CRA helped", Q="gap shrinking"). You reason "Q is true therefore P", but that's a logical fallacy. The theory also says "If P then Q" (P="CRA helped", Q="high school graduate rates improve faster after CRA"). I reason "Q is false therefore P is false", and that is a correct logical inference.
You're the one who cited that graph
Correct. I did that because (1) you had linked to ed.gov pages that made claims about graduation rates and (2) because it falsifies the theory that the CRA helped at all, which in particular falsifies the premise that it helped through outlawing private discrimination.
I'd point you to Thomas Sowell's analysis of the (lack of) effects of the CRA again, but since you don't seem to understand elementary logic, there really isn't much point.
So what you're saying is that transportation and energy use policies should be based upon a pretty infrequent set of scenarios.
People rationally make the decision that if they spend $40k on a car, it better serve all their transportation needs. You have to be significantly more wealthy to be able to afford separate cars for commuting and long distance driving.
With that logic, why not build thirty lane highways to wine country, or fuck it, have a helicopter standing by?
I wouldn't know how to validate the "Almost all" claim. ClojureScript, which is not run on the JVM, but only compiled by it, is becoming extremely popular.
Again, you're missing the point. A JavaScript backend is easy to do for just about any language because it doesn't have to be very good.
Lux might conceivably replace Clojure or Scala (though I doubt it); it's in the same JVM-derived language family.
But Lux will likely never compete with general purpose ("universal") languages like Go, Swift, C#, C++, etc. No language that started out on the JVM ever made the jump, and I gave some reasons why.
Java itself - it got ported to.NET vm and renamed to C# 1.0...
It didn't just get "ported", it addressed the major problems of the JVM: poor native code interfaces, lack of generics, bad semantics for built-in types, and lack of value types.
The differences between the.NET VM and the JVM mean that the.NET VM is actually a reasonable target for a general purpose language that later gets an LLVM backend, while the JVM is not.
I believe Clojure, itself, in the form of ClojureScript, is an example of a language that made the leap from the JVM to other platforms.
AFAIK, there are no widely-used backends for Clojure other than JVM and JavaScript (there have been two attempts at LLVM backends and another one at a CLR backend). Almost all Clojure code is run on the JVM in practice.
Programs trapping Ctrl-C as an exception are exceptionally lazy - there should be a more "front end" way to quit. Originally Ctrl-C was just to kill, not to gracefully shut-down.
Control-C is the usual way of stopping a Linux command line program. Users expect it to leave the system in a reasonable state and programs to clean up after themselves.
Java did. Besides running in the JVM it is also quite successful in Android's Dalvik and now ART.
Native implementations of Java have pretty much failed. And Dalvik and ART are just incremental enhancements of JVM.
I'm sorry if it still isn't obvious: as far as I'm concerned, a "universal cross-platform language" needs to have a decent native code implementation, otherwise it's not "universal".
The biggest problem with expensive professional services is in health care. So, instead of having librarians teach kids how to code, why don't we have them teach kids how to treat patients? Librarians are smart, aren't they? Surely the could teach anything from GP diagnosis to pathology, radiology, and brain surgery, right? They are librarians! And by increasing the supply of medically trained kids, we could then better satisfy the demand for doctors!
At the start, there's a large gap between the two lines. And they're moving exactly in parallel. There's no sign of the gap shrinking. Then in the early 70s, less than a decade after the Civil Rights Act was passed, it does start to shrink. And it continues shrinking steadily, so by the right side of the graph, the gap is only a tiny fraction what it was at the start.
There is no significant change in the rate of improvement of black high school graduation rates until the late 80's, when improvements slow down. That is, the Civil Rights Act obviously didn't help improve black graduation rates. That observation is independent of whatever the effects of the Civil Rights Act may have been on other groups.
Now, the gap is narrowing because white graduation rates at each point in time improve slower than black graduation rates, though that is (hopefully) unrelated to the Civil Rights Act. In fact, both graphs follow a typical logistic growth curve, and such curves mathematically result in narrowing gaps for populations that start growing at different times as they reach saturation.
That is exactly what you would expect to see if the law were working.
You're commiting the fallacy of the converse. It is true that if the act works, then the gap narrows. But observing a narrowing gap doesn't tell you that the act is working because there are many other reasons why the gap might narrow.
Yet somehow, you try to cite it as evidence that it isn't working.
And I've explained to you why. That isn't just my opinion, that's the conclusion of economists who have actually looked at the data. I have repeatedly given you references.
And, again, even if the Civil Rights Act had helped improve black graduation rates, it would have done so through the elimination of government racial discrimination, not any laws related to private businesses, since private businesses have little influence on high school graduation rates. So, your data both fails to show that the Civil Rights Act as a whole helped blacks and that the provisions related to private businesses helped blacks.
Wow, what a *ton* of bullshit here. Honestly you should be proud - it's not so easy to be so completely wrong.
By all means, feel free to show any significant language that has managed to make the leap from the JVM to other platforms.
The most interesting and successful new languages on the JVM, Scala and Clojure, have tried for years to create non-JVM backends and failed to deliver anything other than toys.
Therefore I'll simply quote the relevant portion from one of those links [ed.gov] to save you the trouble of having to click on it.
The question we are discussing is whether the anti-discrimination laws imposed on private businesses had any positive effect. That page isn't relevant to our discussions because it addresses improvements in public and publicly funded education. Public schools, public universities, and public funding of education was racist and discriminatory until the 1960's (thanks to Democrats and progressives), so it wouldn't be surprising if reversing those racist government policies led to improvements (note that it was predominantly Republicans that pushed for this). But any benefits of reversing racist government policies in public education tell you nothing about the benefits of imposing anti-discrimination laws on private employers.
In addition, even if it were relevant, that page lists improvements after the passage of the civil rights act. But the situation of African Americans was already improving before the passage of the civil rights act. In order to show that the civil rights act had significant practical benefits, you need to show that the situation of African Americans improved faster after the passage of the civil rights act than before. Thomas Sowell's research (among others) showed that it did not, and that conclusion is also pretty obvious simply from the graphs that I linked to.
So, as you can see, the OCR page is misleading and it contradicts research in economics. Far from supporting your point, what it actually supports is the idea that the OCR is politically biased and not trustworthy.
Now, you can continue to mindlessly point to political propaganda pieces, or you can actually start using your head, read the economic literature, and understand the data.
first targeted the Java Virtual Machine, but will be a universal, cross-platform language
This may be nice for Java developers, but I can't think of any significant language that started off targeting the JVM and then successfully moved to another platform. That's because languages targeting the JVM get bogged down by the limitations of the JVM and the get entangled in the Java libraries.
If you want to develop a new language these days, start by targeting the LLVM.
Ah, a European "scientist" chimes in on US tax policy. Why is the European intellectual class so terribly obsessed with the US and with trying to turn the US into a replica of Europe?
Yes, there are some studies that purport to show that; they are misleading. You can't just look at net influx/outflux of federal dollars to see whether a state or city benefits from federal spending.
In addition, even in the Bay Area, only a tiny fraction of people benefits from HSR or light rail anyway; they have no significant impact on congestion, and they benefit only a negligible fraction of commuters.
Mostly, these big public projects are government handouts to unions, developers, and construction companies; simple crony capitalism. And at the same time, they tend to drive up housing prices even further and increase inequality.
So you're saying that Californians should, on average, pay hundreds of dollars in extra taxes so that the privileged people living in the centers of SF and LA (mostly millionaires) have a convenient ride downtown to downtown?
The above post suggests that money should instead be spend on Bay Area or Southern California transit projects, but this is a false dichotomy (trichotomy?) -- the public benefits from spending money on all three of these areas
There is one group of people that benefits from this, namely developers, home owners, and commuters in the Bay Area.
There is another group of people that pays for this, namely state and federal tax payers.
These projects transfer money from the latter group to the former group. That's beneficial to the recipients, but harmful to the people actually paying for it.
Slashdot Mirror
It would be better if Trump just tweeted by sending text messages to his staff, who can then post it. That would also add an extra level of protection against making a fool of himself.
You're just the typical anti-science and anti-reason bigot. There really is no getting through to you.
Kids from families with high incomes have significantly higher test scores; highly competitive universities will therefore overwhelmingly select from high-income families even if they exclusively select based on test scores. So, there is nothing particularly surprising about this result, nor does it demonstrate any kind of discrimination of selective colleges against low income kids.
You can now debate about whether high income causes kids to have high test scores, i.e., if you only gave kids from poor families more money, they'd be doing just as well. That is true to some very limited degree: kids who lack essentials (food, clean water, etc.) are held back by that, but fixing those problems can't increase their intelligence beyond their potential.
Most of the correlation is likely primarily caused by the fact that smart parents tend to have smart kids (through a combination of nature and nurture), and that high test scores and high incomes simply result from that.
Both!
I didn't say "all languages based on the JVM are useless/toys", I said "languages targeting the JVM get bogged down by the limitations of the JVM and the get entangled in the Java libraries."
But, yeah, whatever.
Again, you are making an elementary logic error. The theory says "If P then Q" (P="CRA helped", Q="gap shrinking"). You reason "Q is true therefore P", but that's a logical fallacy. The theory also says "If P then Q" (P="CRA helped", Q="high school graduate rates improve faster after CRA"). I reason "Q is false therefore P is false", and that is a correct logical inference.
Correct. I did that because (1) you had linked to ed.gov pages that made claims about graduation rates and (2) because it falsifies the theory that the CRA helped at all, which in particular falsifies the premise that it helped through outlawing private discrimination.
I'd point you to Thomas Sowell's analysis of the (lack of) effects of the CRA again, but since you don't seem to understand elementary logic, there really isn't much point.
People rationally make the decision that if they spend $40k on a car, it better serve all their transportation needs. You have to be significantly more wealthy to be able to afford separate cars for commuting and long distance driving.
Because that would be economically irrational.
Again, you're missing the point. A JavaScript backend is easy to do for just about any language because it doesn't have to be very good.
Lux might conceivably replace Clojure or Scala (though I doubt it); it's in the same JVM-derived language family.
But Lux will likely never compete with general purpose ("universal") languages like Go, Swift, C#, C++, etc. No language that started out on the JVM ever made the jump, and I gave some reasons why.
It didn't just get "ported", it addressed the major problems of the JVM: poor native code interfaces, lack of generics, bad semantics for built-in types, and lack of value types.
The differences between the .NET VM and the JVM mean that the .NET VM is actually a reasonable target for a general purpose language that later gets an LLVM backend, while the JVM is not.
AFAIK, there are no widely-used backends for Clojure other than JVM and JavaScript (there have been two attempts at LLVM backends and another one at a CLR backend). Almost all Clojure code is run on the JVM in practice.
Control-C is the usual way of stopping a Linux command line program. Users expect it to leave the system in a reasonable state and programs to clean up after themselves.
Native implementations of Java have pretty much failed. And Dalvik and ART are just incremental enhancements of JVM.
I'm sorry if it still isn't obvious: as far as I'm concerned, a "universal cross-platform language" needs to have a decent native code implementation, otherwise it's not "universal".
And Gosu and Groovy :-)
The biggest problem with expensive professional services is in health care. So, instead of having librarians teach kids how to code, why don't we have them teach kids how to treat patients? Librarians are smart, aren't they? Surely the could teach anything from GP diagnosis to pathology, radiology, and brain surgery, right? They are librarians! And by increasing the supply of medically trained kids, we could then better satisfy the demand for doctors!
Kotlin suffers from the same limitations as Clojure and Scala (both of which also have JavaScript backends), and in addition is pretty obscure.
All those languages are just rearranging deck chairs on the JVM Titanic.
There is no significant change in the rate of improvement of black high school graduation rates until the late 80's, when improvements slow down. That is, the Civil Rights Act obviously didn't help improve black graduation rates. That observation is independent of whatever the effects of the Civil Rights Act may have been on other groups.
Now, the gap is narrowing because white graduation rates at each point in time improve slower than black graduation rates, though that is (hopefully) unrelated to the Civil Rights Act. In fact, both graphs follow a typical logistic growth curve, and such curves mathematically result in narrowing gaps for populations that start growing at different times as they reach saturation.
You're commiting the fallacy of the converse. It is true that if the act works, then the gap narrows. But observing a narrowing gap doesn't tell you that the act is working because there are many other reasons why the gap might narrow.
And I've explained to you why. That isn't just my opinion, that's the conclusion of economists who have actually looked at the data. I have repeatedly given you references.
And, again, even if the Civil Rights Act had helped improve black graduation rates, it would have done so through the elimination of government racial discrimination, not any laws related to private businesses, since private businesses have little influence on high school graduation rates. So, your data both fails to show that the Civil Rights Act as a whole helped blacks and that the provisions related to private businesses helped blacks.
By all means, feel free to show any significant language that has managed to make the leap from the JVM to other platforms.
The most interesting and successful new languages on the JVM, Scala and Clojure, have tried for years to create non-JVM backends and failed to deliver anything other than toys.
The question we are discussing is whether the anti-discrimination laws imposed on private businesses had any positive effect. That page isn't relevant to our discussions because it addresses improvements in public and publicly funded education. Public schools, public universities, and public funding of education was racist and discriminatory until the 1960's (thanks to Democrats and progressives), so it wouldn't be surprising if reversing those racist government policies led to improvements (note that it was predominantly Republicans that pushed for this). But any benefits of reversing racist government policies in public education tell you nothing about the benefits of imposing anti-discrimination laws on private employers.
In addition, even if it were relevant, that page lists improvements after the passage of the civil rights act. But the situation of African Americans was already improving before the passage of the civil rights act. In order to show that the civil rights act had significant practical benefits, you need to show that the situation of African Americans improved faster after the passage of the civil rights act than before. Thomas Sowell's research (among others) showed that it did not, and that conclusion is also pretty obvious simply from the graphs that I linked to.
So, as you can see, the OCR page is misleading and it contradicts research in economics. Far from supporting your point, what it actually supports is the idea that the OCR is politically biased and not trustworthy.
Now, you can continue to mindlessly point to political propaganda pieces, or you can actually start using your head, read the economic literature, and understand the data.
This may be nice for Java developers, but I can't think of any significant language that started off targeting the JVM and then successfully moved to another platform. That's because languages targeting the JVM get bogged down by the limitations of the JVM and the get entangled in the Java libraries.
If you want to develop a new language these days, start by targeting the LLVM.
Ah, a European "scientist" chimes in on US tax policy. Why is the European intellectual class so terribly obsessed with the US and with trying to turn the US into a replica of Europe?
I do, but not one that someone as narrow minded, bigoted, and economically illiterate as you would understand.
Yes, there are some studies that purport to show that; they are misleading. You can't just look at net influx/outflux of federal dollars to see whether a state or city benefits from federal spending.
In addition, even in the Bay Area, only a tiny fraction of people benefits from HSR or light rail anyway; they have no significant impact on congestion, and they benefit only a negligible fraction of commuters.
Mostly, these big public projects are government handouts to unions, developers, and construction companies; simple crony capitalism. And at the same time, they tend to drive up housing prices even further and increase inequality.
Travel by high-speed rail is much more expensive than travel by bus everywhere I have ever lived. And it is usually also more expensive than flying.
So you're saying that Californians should, on average, pay hundreds of dollars in extra taxes so that the privileged people living in the centers of SF and LA (mostly millionaires) have a convenient ride downtown to downtown?
There is one group of people that benefits from this, namely developers, home owners, and commuters in the Bay Area.
There is another group of people that pays for this, namely state and federal tax payers.
These projects transfer money from the latter group to the former group. That's beneficial to the recipients, but harmful to the people actually paying for it.