Why teach History? Few people need that in their daily life or jobs.
I agree in particular about history. I think it's placement in the curriculum is probably the worst choice of all the subjects. I remember myself and most of my classmates being particularly uninterested in the doings of the French, the British and the Native Americans. I think history is important, but it's targeted at totally the wrong age group.
I think they should focus primarily on practical science that engages students at younger ages, together with lots of reading and writing to expand their comprehension, analytical and articulation, with the basic math they do now. Then they should phase in other subjects like history in later grades as they get to more sophisticated reading material which require understanding more of the wider world and how it came to be the way it is.
It might make people realize that population growth, resource consumption, etc. can't keep increasing at current levels without severe corrections in the somewhat close future.
They've been saying that for a few hundred years, but there's little evidence or reason to suggest we cannot handle any such problems as they arise. Population growth is self-correcting, since social and economic progress is strongly correlated with lower growth rates (if IIRC, infant mortality is the strongest predictor, ie. low infant mortality = low to negative population growth).
By the time resources become scarce, a situation which seems quite far away at this point, there's little reason that we couldn't mine the moon and asteroids. That technology is almost already within our grasp. Beyond that, for the far, far future, there's transmutation. Only energy matters, and technically there's zero total energy in the entire universe. If we can have everything we see around us on a nil energy budget, I'd put my money on our descendants finding some exploitable loopholes in the laws of physics for anything we'll ever need.
In any case, we certainly have no shortage of energy in the near future, renewable or otherwise. I have no faith in the doctrine that you've cited here, because the models by you which you describe them are simply inaccurate, ie. population growth != exponential growth.
Depends what you mean by "test-driven". It's a well known fact that the more you give students, the more they retain the knowledge. Testing their recall and reasoning skills under pressure improves the retention of that knowledge and those skills.
BUT let's think about that hard disk business. Hard disk for IBM 1620 did exist, the 1311 with two megabytes. Put that on a 50KHz driven 1620 and good luck with your jvm. Maybe you could prompt the user to halt and power off the machine and change hard disks.
Do you have a point? You've essentially acknowledged that there's nothing intrinsically limiting the IBM 1620's ability to execute arbitrary programs, except the hardware limitations at the time. In fact, there's nothing intrinsically limiting any current computer from being universal Turing machine, except the infinite tape requirements of said machine, ie. a true universal Turing machine is physically unrealizable.
These are all well known facts, and they in no way disprove any of my statements, in particular my original one which started this whole thread.
So if you have an actual, useful point, I'd like to hear it.
BUT let's think about that hard disk business. Hard disk for IBM 1620 did exist, the 1311 with two megabytes. Put that on a 50KHz driven 1620 and good luck with your jvm. Maybe you could prompt the user to halt and power off the machine and change hard disks.
Do you have a point? You've essentially acknowledged that there's nothing intrinsically limiting the IBM 1620's ability to execute arbitrary programs, except the hardware limitations at the time. In fact, there's nothing intrinsically limiting any machine from becoming a real universal Turing machine, except the infinite tape requirements of said machine, ie. a true universal Turing machine is physically unrealizable.
These are all well known facts, and they in no way disprove any of my statements, in particular my original one which started this whole thread.
So if you have an actual, useful point, I'd like to hear it.
theoretical ivory tower B.S. Real computers have limits on resources and capability that make the statement false in some cases.
That's because no Turing machine is physically realizable due to the infinite tape, but that doesn't negate the truth of the principle. All physically realizable machines are finite state machines.
For example, you can't port the jvm to a Z80 or an IBM 1620 in this universe.
Sure you could, it would just be brutally slow, and you'll have to live with the fact that some programs will simply throw out of memory exceptions. Java or the JVM make no guarantees about the amount of available physical resources.
Realistically, you would use an external I/O interface to address a sufficiently large enough address space to run these programs.
So again, this doesn't negate the point that JVM opcodes can be compiled to any other abstract or real machine and executed. There's noting special about Dalvik doing this.
The British public pays to maintain these sites, and an awful lot of money at that, so why should some company be allowed to step in and enjoy the benefits of the public's investment?
Because you paid an entrance fee to visit this site, and you're also paying to maintain the site via taxes, so why should you pay even more? Will I now have to pay the government every time I take a picture of a road? I'm sorry, but it's ludicrous.
I posted the calculations of the odds of another earth-moon system - in this galaxy, somewhere in my journal, but I'll give you the executive summary: we're IT. Unique.
I find it hard to believe we have enough data to even begin to estimate these sorts of odds, particularly since this is the first planet we've detected that's even close to Earth-sized.
I'm also not totally convinced by your arguments that this planet would simply be another Venus, since whether the greenhouse effect is detrimental depends entirely upon the intensity of incident radiation, which is dependent on the brightness its local sun and the distance of the planet from that sun. Greenhouse on Mars would be great, greenhouse on Earth not so much.
I think companies should stop trying to dictate what their users do and do not want. Having fart apps hardly devalue a device, and the moment you censor one app for one reason, is the moment you're going to have other people beating down your door trying to legitimize their own reasons for barring other apps, and you'll have no legitimate excuse to refuse them. Better to just be impartial, and let the user decide.
Furthermore, this is solving the wrong problem. Don't try to reduce the scope of your app selection in an indirect attempt to improve user experience when finding and installing apps, just improve the user experience directly. Improve search, improve categorization, make app searching interactive instead of by keyword, ie. interactive wizard-style interfaces like 20 questions perhaps. You know, be innovative instead of restrictive.
Assuming companies properly follow all safety procedures when developing and testing drugs, and assuming they disclose known problems within an appropriate window, companies should not be held liable for their products. While I hate arguments by analogy, this is truly akin to holding a construction company liable for a bridge collapsing after it was struck by a missile, or an earthquake rated 9.9 on the Richter scale. The company had no control over these unforeseeable events, and to make them liable makes all new product development a much riskier and costlier endeavor, to the detriment of society as a whole.
Dalvik still wouldn't run JVM bytecodes, so I don't think it would a conforming implementation regardless. I haven't read the spec however, so I don't know if the bytecode is specified there or separately.
Even *if* that were true (hint it isn't since the Android platform does not 'run Java' that would leave Google up #### creek in a major way. Check the Java conformity patent pledge - it only refers to desktop implementations. Mobile Java is not in any way pledged or even FOSS. Google would have been attacked then for using an unlicensed J2ME implementation
You can make any technical argument you like, but the fact is, Java is the primary development language for Dalvik and Google markets it as such. Thus, for all intents and purposes, Dalvik is a Java virtual machine, but not a strictly conforming JVM according to the spec, thus they are not protected. Even if Java were only one of many languages, and not even the primary one, Oracle could still sue them, because they could still infringe on these patents.
It's not a matter of Google not wanting to make a "fully conforming" JVM, Google doesn't make a JVM, so the issue of fully conforming doesn't come into any equation.
It does, because the Java license grants patent protection if you are conforming. The fact is, Google built and markets a platform that runs Java. If it were 100% conformant, Google wouldn't have this problem. Because it's not conformant, they are vulnerable to Java patents.
Because Microsoft had a contract with Sun to create a certified VM and they broke the conditions plus they called it Java when it wasn't they got hit with the judicial hammer...
Which is what I said: the same compliance clause is being used in both cases, just differently. MS was supposed to provide a conforming JVM but didn't, and Google didn't want to provide a fully conforming JVM, but is still being hit by this same clause.
Guess what? That same compliance clause is exactly what's being used to now sue Google over Android. The Java license provides a patent grant assuming you have a conforming implementation. Android is not fully conforming, nor did they ever claim to be, thus they are vulnerable to patent suits based on Java technology.
Honestly, I wish Via had the resources AMD and Intel have. Their Nano CPU is pretty nice, but it's languishing. They're only just now coming out with a dual core version. The Nano's on-die crypto extensions, low power use, and higher performance per watt would otherwise make it ideal for server applications, particularly SSL front-ends.
I suspect that this will be the sequence of events with the attendant burden of proof required before serious action is taken:
1. Global warming trend is happening.
2. The warming trend is anthropogenic, ie. human-caused, and specifically caused by factors A, B, C, etc.
3. The warming is actually disastrous, and will lead to loss of life and/or livelihood.
4. Addressing the symptoms of the warming after the fact will cost more and be less effective than attempting to curb the warming in the first place.
At this point, #1 is definitely irrefutable. There are some hold outs for #2, but I think the evidence is sufficient to conclude we're causing warming. #3 is pure speculation, but based on reasonable arguments; still, far from a given. #4 is almost beyond our ability to speculate, though there are some proposals to reverse some of the expected warming effects.
I think there are plenty of other good reasons to change the way we regulate environmental impact, eg. poor air quality has been killing thousands of people a year for decades. But anything less than the irrefutable evidence on all 4 of the above points will fail to convince someone. I suspect at least the first 3 will be required to galvanize enough people to make definitive political moves to addressing these issues.
Dude, you need a reality adjustment. It is estimated that there is enough surface-mineable thorium alone to power us for hundreds of thousands of years to come. In fact, just the thorium discarded from our surface-mined coal could power us for thousands of years.
Actually, I think you need the adjustment. See The Coming Nuclear Crisis. Our current nuclear power is expensive with the easily mined Uranium sources, and we're running out of those. Now picture how expensive nuclear will be when we try to tap those less dense Uranium sources, like the oceans.
You are correct however that nuclear power is currently so expensive because they're inefficient. Deploying faster reactors and thorium reactors, which as you say, are orders of magnitude more efficient, will stretch out the Uranium supply significantly longer.
Nuclear isn't competitive because they're using outdated, inefficient reactor technology, and not using all the fissionable material they have at their disposal. They need to deploy Thorium reactors to boost the efficiency and safety.
However, distributed power generation via solar is absolutely a necessary step forward as well. We're just wasting all the ambient energy by absorbing all that light into our rooftops and parking lots.
Slashdot Mirror
I agree in particular about history. I think it's placement in the curriculum is probably the worst choice of all the subjects. I remember myself and most of my classmates being particularly uninterested in the doings of the French, the British and the Native Americans. I think history is important, but it's targeted at totally the wrong age group.
I think they should focus primarily on practical science that engages students at younger ages, together with lots of reading and writing to expand their comprehension, analytical and articulation, with the basic math they do now. Then they should phase in other subjects like history in later grades as they get to more sophisticated reading material which require understanding more of the wider world and how it came to be the way it is.
They've been saying that for a few hundred years, but there's little evidence or reason to suggest we cannot handle any such problems as they arise. Population growth is self-correcting, since social and economic progress is strongly correlated with lower growth rates (if IIRC, infant mortality is the strongest predictor, ie. low infant mortality = low to negative population growth).
By the time resources become scarce, a situation which seems quite far away at this point, there's little reason that we couldn't mine the moon and asteroids. That technology is almost already within our grasp. Beyond that, for the far, far future, there's transmutation. Only energy matters, and technically there's zero total energy in the entire universe. If we can have everything we see around us on a nil energy budget, I'd put my money on our descendants finding some exploitable loopholes in the laws of physics for anything we'll ever need.
In any case, we certainly have no shortage of energy in the near future, renewable or otherwise. I have no faith in the doctrine that you've cited here, because the models by you which you describe them are simply inaccurate, ie. population growth != exponential growth.
Depends what you mean by "test-driven". It's a well known fact that the more you give students, the more they retain the knowledge. Testing their recall and reasoning skills under pressure improves the retention of that knowledge and those skills.
Do you have a point? You've essentially acknowledged that there's nothing intrinsically limiting the IBM 1620's ability to execute arbitrary programs, except the hardware limitations at the time. In fact, there's nothing intrinsically limiting any current computer from being universal Turing machine, except the infinite tape requirements of said machine, ie. a true universal Turing machine is physically unrealizable.
These are all well known facts, and they in no way disprove any of my statements, in particular my original one which started this whole thread.
So if you have an actual, useful point, I'd like to hear it.
Do you have a point? You've essentially acknowledged that there's nothing intrinsically limiting the IBM 1620's ability to execute arbitrary programs, except the hardware limitations at the time. In fact, there's nothing intrinsically limiting any machine from becoming a real universal Turing machine, except the infinite tape requirements of said machine, ie. a true universal Turing machine is physically unrealizable.
These are all well known facts, and they in no way disprove any of my statements, in particular my original one which started this whole thread.
So if you have an actual, useful point, I'd like to hear it.
Some engineer you are. It's called a hard disk. C's read()/write()/seek() are all you need.
Being an engineer myself, I sure hope you aren't an accredited one, because if so, it shows a distinct decline in the quality of education.
That's because no Turing machine is physically realizable due to the infinite tape, but that doesn't negate the truth of the principle. All physically realizable machines are finite state machines.
Sure you could, it would just be brutally slow, and you'll have to live with the fact that some programs will simply throw out of memory exceptions. Java or the JVM make no guarantees about the amount of available physical resources.
Realistically, you would use an external I/O interface to address a sufficiently large enough address space to run these programs.
So again, this doesn't negate the point that JVM opcodes can be compiled to any other abstract or real machine and executed. There's noting special about Dalvik doing this.
You can translate anything to run on any computer. It's called the Turing Tarpit.
Because you paid an entrance fee to visit this site, and you're also paying to maintain the site via taxes, so why should you pay even more? Will I now have to pay the government every time I take a picture of a road? I'm sorry, but it's ludicrous.
I'm truly surprised that every TV doesn't do this by now.
I find it hard to believe we have enough data to even begin to estimate these sorts of odds, particularly since this is the first planet we've detected that's even close to Earth-sized.
I'm also not totally convinced by your arguments that this planet would simply be another Venus, since whether the greenhouse effect is detrimental depends entirely upon the intensity of incident radiation, which is dependent on the brightness its local sun and the distance of the planet from that sun. Greenhouse on Mars would be great, greenhouse on Earth not so much.
I think companies should stop trying to dictate what their users do and do not want. Having fart apps hardly devalue a device, and the moment you censor one app for one reason, is the moment you're going to have other people beating down your door trying to legitimize their own reasons for barring other apps, and you'll have no legitimate excuse to refuse them. Better to just be impartial, and let the user decide.
Furthermore, this is solving the wrong problem. Don't try to reduce the scope of your app selection in an indirect attempt to improve user experience when finding and installing apps, just improve the user experience directly. Improve search, improve categorization, make app searching interactive instead of by keyword, ie. interactive wizard-style interfaces like 20 questions perhaps. You know, be innovative instead of restrictive.
Assuming companies properly follow all safety procedures when developing and testing drugs, and assuming they disclose known problems within an appropriate window, companies should not be held liable for their products. While I hate arguments by analogy, this is truly akin to holding a construction company liable for a bridge collapsing after it was struck by a missile, or an earthquake rated 9.9 on the Richter scale. The company had no control over these unforeseeable events, and to make them liable makes all new product development a much riskier and costlier endeavor, to the detriment of society as a whole.
I'm failing to see how software and books differ in this case. How does paying for a book entitle you to ownership, but paying for software does not?
The Evolution of a Haskell Programmer.
Indeed, they already do have that right, by not using Google's free services they ensure Google can't use any personal information.
Dalvik still wouldn't run JVM bytecodes, so I don't think it would a conforming implementation regardless. I haven't read the spec however, so I don't know if the bytecode is specified there or separately.
You can make any technical argument you like, but the fact is, Java is the primary development language for Dalvik and Google markets it as such. Thus, for all intents and purposes, Dalvik is a Java virtual machine, but not a strictly conforming JVM according to the spec, thus they are not protected. Even if Java were only one of many languages, and not even the primary one, Oracle could still sue them, because they could still infringe on these patents.
It does, because the Java license grants patent protection if you are conforming. The fact is, Google built and markets a platform that runs Java. If it were 100% conformant, Google wouldn't have this problem. Because it's not conformant, they are vulnerable to Java patents.
Which is what I said: the same compliance clause is being used in both cases, just differently. MS was supposed to provide a conforming JVM but didn't, and Google didn't want to provide a fully conforming JVM, but is still being hit by this same clause.
Guess what? That same compliance clause is exactly what's being used to now sue Google over Android. The Java license provides a patent grant assuming you have a conforming implementation. Android is not fully conforming, nor did they ever claim to be, thus they are vulnerable to patent suits based on Java technology.
Honestly, I wish Via had the resources AMD and Intel have. Their Nano CPU is pretty nice, but it's languishing. They're only just now coming out with a dual core version. The Nano's on-die crypto extensions, low power use, and higher performance per watt would otherwise make it ideal for server applications, particularly SSL front-ends.
I suspect that this will be the sequence of events with the attendant burden of proof required before serious action is taken:
1. Global warming trend is happening.
2. The warming trend is anthropogenic, ie. human-caused, and specifically caused by factors A, B, C, etc.
3. The warming is actually disastrous, and will lead to loss of life and/or livelihood.
4. Addressing the symptoms of the warming after the fact will cost more and be less effective than attempting to curb the warming in the first place.
At this point, #1 is definitely irrefutable. There are some hold outs for #2, but I think the evidence is sufficient to conclude we're causing warming. #3 is pure speculation, but based on reasonable arguments; still, far from a given. #4 is almost beyond our ability to speculate, though there are some proposals to reverse some of the expected warming effects.
I think there are plenty of other good reasons to change the way we regulate environmental impact, eg. poor air quality has been killing thousands of people a year for decades. But anything less than the irrefutable evidence on all 4 of the above points will fail to convince someone. I suspect at least the first 3 will be required to galvanize enough people to make definitive political moves to addressing these issues.
Actually, I think you need the adjustment. See The Coming Nuclear Crisis. Our current nuclear power is expensive with the easily mined Uranium sources, and we're running out of those. Now picture how expensive nuclear will be when we try to tap those less dense Uranium sources, like the oceans.
You are correct however that nuclear power is currently so expensive because they're inefficient. Deploying faster reactors and thorium reactors, which as you say, are orders of magnitude more efficient, will stretch out the Uranium supply significantly longer.
Nuclear isn't competitive because they're using outdated, inefficient reactor technology, and not using all the fissionable material they have at their disposal. They need to deploy Thorium reactors to boost the efficiency and safety.
However, distributed power generation via solar is absolutely a necessary step forward as well. We're just wasting all the ambient energy by absorbing all that light into our rooftops and parking lots.