The other unrealistic assumption is that the overall hardware performance increases with CPU performance. While many hardware capabilities have increased dramatically over recent years, disk drive access and throughput speeds have taken about 5-7 years to double, and main memory speeds have taken about the same time.
So unless you're running a program that fits inside the CPU cache, a bloated program will still be a lot slower and inefficient, even if you have scads of memory and a multi-core CPU.
The other side of assuming that your programmers are competent is a parallel assumption that your HR team is competent and your hiring process is adequate. If you have a set of low-grade developers and engineers on your team, either you aren't paying enough to attract the people you need or the problem is in your hiring process.
And apart from the unrealistic assumptions on both sides of the equation, now that most gains in CPU performance come from adding cores, you need programmers who understand multi-threading and multi-processing. A poorly threaded program isn't a problem you can throw hardware at.
The Muslim violence we see today is comparable to the troubles in Northern Ireland, scaled up proportionally to a larger population and land area. The difference is that the violence in N.I. was directed locally and at the UK, while the U.S. was a funding source for various factions. For the Muslim violence, the U.S. is the target, as well as the funding source (if you track the oil revenues back to the consumer).
If you want to see the epitome of religious violence look at the period starting with the rise of Islam, through the Crusades, the Catholic/Protestant reformation and counter-reformation that only wound down after the Holocaust. After the 18th century's "Age of Reason" and the American proclamation of religious liberty, the violence continued, but with rare exceptions (pogroms and genocides) it wasn't covered with a religious fig-leaf.
Humans have a natural tendency for war and violence. It has nothing to do with which god we worship (or don't worship). Religion is just a tool that war hawks use to justify their methods.
Plus, we'll have USB 3.0 soon, so the interface can be sped up for the SSDs, and if memristor technology takes off, in 4-5 years we could see terabyte SSDs.
Actually, the "old" style of programming is new again. You just have to go back farther in time. The PC programming paradigm of just coding things with generic performance tweaks, and ensuring that the logic flow works won't hack it. Now you'll actually have to understand the architecture of the system that you're compiling to and optimize your algorithms for the number of cores and the memory architecture.
This gives the advantage to the old programmers who had to do things like fit a BASIC compiler into a 4k memory space.
At least most OS's already have a task management API to provide a crude level of management.
Well, actually that's only true because the CMOS semi-conductor architecture only disperses substantial amounts of power during the clock transition state (voltage going up or down). So power dissipation is proportional to clock frequency (the number of transitions per second). There are other historic semi-conductor architectures that essentially have constant power requirements (like TTL), but these were essentially abandoned in the 1980's because at the clock speeds then prevalent, CMOS was incredibly more power-effecient. TTL probably wouldn't hack it nowadays even at high frequencies, but perhaps it's time to take another look at how the actual basic circuits are designed to see if they can be implemented more efficiently for high clock rates.
Actually that is part of the problem. Most of the architectures have core-specific L1 cache, and unless a particular thread has its affinity mapped to a particular core, a thread can jump from a core where its data is in the L1 cache to a core where its data is not present, and is forced to undergo a cache refresh from memory.
Also, regardless of whether a system is multi-processing within a chip (multi-core) or on a board (multi-CPU), the number of communication channels required to avoid communication bottlenecks goes up as O(n^2) the number of cores.
So yes, we are probably seeing the beginning of the end of performance gains using general-purpose CPU interconnects and have to go back to vector processing. Unless we are somehow able to jump the heat dissipation barrier and start raising GHz again.
Teacher salaries vary wildly by district, so blanket statements that teachers are way underpaid for their educational level are bogus. In my area, teachers earn pretty close to the median income for bachelors/Masters degree holders $50k-$70k.
Regarding teachers doing work during their "vacation" time: At least they get the time off to do that work. How many doctors, software developers and engineers nowadays are given 2-3 months a year of time off to further their education. Particularly those that work for corporations. It ain't soft for teachers, but the stressors are different.
Uh, No child left behind is mostly trying to teach kids reading, writing and math. This is pretty much a pre-requisite for vocational education anyhoo.
"Bore on" is probably how most HS students describe chemistry anyway. Maybe that's what they were complaining about.
I see two Filk songs out of this: 1) "Bore On" as a send-up of Buddy Holly's "Rave On", 2) The text of the test as a parody of "Shaft" - "What is the gas with 10.8 atomic mass? - Boron"
You talk about deductions and rates for federal income taxes, but most education funding comes from state property (and income) taxes.
There are 3 systemic problems in public education that need to be addressed, but funding is only an indirect issue: if we're taking in more money per student than most other countries but not seeing results, the problem is on the consumption side, not on the funding side. 1) Teacher competency and tenure - tenure should be granted based on merit over a period of several years, not merely on time served, which makes it too easy to slip through the cracks. 2) Administrative overhead. One big problem with public education as a government institution is that the administration accretes new functions over time to address such issues as corrupt hiring/contracting, discrimination and so on. A similar accretion process occurs in large corporations, and they respond by going through re-orgs and revamping their processes. Administrative functions need to be swept clean on a zero-budget and zero-process-baseline basis every 10 years or so to give a fresh start. 3) Parents/home environment - I don't think I need to reiterate the impact that this has on student performance. The public education system has to take everybody, and everybody should at least be functionally literate and able to manage their daily financial tasks when they graduate. There's not a lot that the schools can do to directly change the home environment, but they can do a better job of identifying students whose achievement suffers because of home issues and guide those students to sources of help.
Some additional random thoughts: One administrative activity to look at might be districting students according to developmental needs, rather than merely by location. Make the schools for the developmentally (or environmentally) challenged and exceptional students as small facilities that can function as centers of expertise for the type of education they are performing. This will provide a stronger support network for the faculty and staff which can reduce the stress of constantly dealing with challenged students.
Privately funded education cherry picks their students, so they always look better on paper.
We often concentrate the debate on urban school performance, but in some ways the rural schools with very small enrollments are suffering the worst. This can really only be addressed by federal funding.
Turnitin.com couldn't index the works unless the teachers illegally redistributed the works to them in the first place. While I can possibly buy the argument that turnitin isn't redistributing the works themselves, the teachers and institutions that submit student works to turnitin.com ARE redistributing copyrighted works. Of course the easy way for the institutions to get around this is to put a clause in the student code of conduct that any work turned in as a classroom assignment gives the institution the right to redistribute it for purposes x, y and z.
Well in some areas (like California), non-competes are considered unenforceable. But on the other hand, California has very strict trade secret laws. So they would be unlikely to get you for violating the non-compete, but if your new product/corporation uses any material at all from the old company (document templates, change management processes, software build environment architectures), you'll get your asses sued.
That's why it's cheaper to pay a lawyer up front to analyze the legal risks in their business plan vs. any contractual or IP obligations they owe to their previous employer.
IANAL, but two big areas that would require detailed checking would be non-compete clauses in your contracts and trade secret laws. Have the lawyer look at everything though. You also could be making some invalid assumptions about how your technical plans insulate you from copyright/patent infringement.
In the old days, companies used separate black box/white box reverse engineering teams to isolate the authors of the new product from the people who knew about the original product. The white box team would write the design specs describing the external behavior of the program, but the black box team would write the actual code. You and your friends could end up having the company but not be allowed to do any of the coding on the new product in order to avoid IP violations. Worse yet, you may not be able to control the details of how the black box team implements the design, so all of your great ideas for re-architecting the system may go by the wayside.
Somehow I think you get paid a bit more than the average CSR at a phone, Cable or credit card company.
I agree there's no excuse for that type of dishonesty, but there's also no excuse for expecting professional ethics from your employees if you're paying minimum wage. At least until the minimum wage is enough for a full time employee to live on without being desperately short of cash.
Either that or it's a "give them enough rope to hang themselves" test. You tell the employees the policy and the consequences, and then wait to see if they're honest or not.
But, we are all going to have to get over seeing them as ugly or migratory-bird killers for this program to work. I truly want a future where we use very little foreign energy, and we harness renewable energy sources.
Better yet, lets work on some turbine designs that don't involve large blades sweeping through the air. With this "transmission" design, it should be possible to build some wind generators that can work efficiently without requiring the amount of torque that is needed to make the current designs work. If we can efficiently generate wind-power without the large torque requirements, no more bird-killers needed.
Nahhh, that's just California. you left out Slashdot, strange sports not played in other countries ("football"), country "music" and political correctness.
BTW, Porn goes way back before American culture. We just commercialized it with mass production.
On the other side, the internet propagates new cultural memes and outlets so fast, it's hard to tell where the new ones are coming from. Is blogging originally an artifact of American culture that spread like TV and music?
IBM has a pretty decent-sized Linux operation. Odds are the AIX workstations are being ditched as redundant and non-competitive (price-wise) with Linux-based stations. Is there really any workstation software that is available on AIX that isn't available on Linux?
Surely this is a problem that has already been addressed by CalTech, MIT and Stanford (among others). Each of those colleges has had dozens of student works that spun off into businesses.
Math is the basis for so much, but no matter what profession she chooses, most of her time is going to be spent writing. Reports, design documents, advertising brochures, processes and procedures, presentation materials.
Every book in the US has it's bibliographic information and much of it's cataloging information on the copyright page (I forget the official publishing term for this, but you know the one I mean). If they make access and use rights to the database too onerous, the libraries can simply hire typists (keyboardists?) to enter the data by hand. Then there ain't an damn thing they can do about it since it wasn't sourced from their database.
This is where PHBs come from, too. People who erroneously believe they are competent do more damage than competent people who do evil seeking their own gain. These people are so incompetent they don't even realize that they ARE incompetent, whether it's due to lack of knowledge or just lack of ability.
Slashdot Mirror
The other unrealistic assumption is that the overall hardware performance increases with CPU performance. While many hardware capabilities have increased dramatically over recent years, disk drive access and throughput speeds have taken about 5-7 years to double, and main memory speeds have taken about the same time.
So unless you're running a program that fits inside the CPU cache, a bloated program will still be a lot slower and inefficient, even if you have scads of memory and a multi-core CPU.
The other side of assuming that your programmers are competent is a parallel assumption that your HR team is competent and your hiring process is adequate. If you have a set of low-grade developers and engineers on your team, either you aren't paying enough to attract the people you need or the problem is in your hiring process.
And apart from the unrealistic assumptions on both sides of the equation, now that most gains in CPU performance come from adding cores, you need programmers who understand multi-threading and multi-processing. A poorly threaded program isn't a problem you can throw hardware at.
"Muslims are more likely to kill..."
Fail.
The Muslim violence we see today is comparable to the troubles in Northern Ireland, scaled up proportionally to a larger population and land area. The difference is that the violence in N.I. was directed locally and at the UK, while the U.S. was a funding source for various factions. For the Muslim violence, the U.S. is the target, as well as the funding source (if you track the oil revenues back to the consumer).
If you want to see the epitome of religious violence look at the period starting with the rise of Islam, through the Crusades, the Catholic/Protestant reformation and counter-reformation that only wound down after the Holocaust. After the 18th century's "Age of Reason" and the American proclamation of religious liberty, the violence continued, but with rare exceptions (pogroms and genocides) it wasn't covered with a religious fig-leaf.
Humans have a natural tendency for war and violence. It has nothing to do with which god we worship (or don't worship). Religion is just a tool that war hawks use to justify their methods.
at http://neurohack.com/transcendence/ This is by George Moromisato, the same guy who wrote Anacreon, back in the DOS 3.x days.
Anacreon is available there http://neurohack.com/ too, if you're taking character-mode programs. And where it used to be shareware, it's now free.
And I also second votes for FreeCiv, Vega Strike, and Wesnoth, among others.
Plus, we'll have USB 3.0 soon, so the interface can be sped up for the SSDs, and if memristor technology takes off, in 4-5 years we could see terabyte SSDs.
Actually, the "old" style of programming is new again. You just have to go back farther in time. The PC programming paradigm of just coding things with generic performance tweaks, and ensuring that the logic flow works won't hack it. Now you'll actually have to understand the architecture of the system that you're compiling to and optimize your algorithms for the number of cores and the memory architecture.
This gives the advantage to the old programmers who had to do things like fit a BASIC compiler into a 4k memory space.
At least most OS's already have a task management API to provide a crude level of management.
Well, actually that's only true because the CMOS semi-conductor architecture only disperses substantial amounts of power during the clock transition state (voltage going up or down). So power dissipation is proportional to clock frequency (the number of transitions per second). There are other historic semi-conductor architectures that essentially have constant power requirements (like TTL), but these were essentially abandoned in the 1980's because at the clock speeds then prevalent, CMOS was incredibly more power-effecient. TTL probably wouldn't hack it nowadays even at high frequencies, but perhaps it's time to take another look at how the actual basic circuits are designed to see if they can be implemented more efficiently for high clock rates.
Actually that is part of the problem. Most of the architectures have core-specific L1 cache, and unless a particular thread has its affinity mapped to a particular core, a thread can jump from a core where its data is in the L1 cache to a core where its data is not present, and is forced to undergo a cache refresh from memory.
Also, regardless of whether a system is multi-processing within a chip (multi-core) or on a board (multi-CPU), the number of communication channels required to avoid communication bottlenecks goes up as O(n^2) the number of cores.
So yes, we are probably seeing the beginning of the end of performance gains using general-purpose CPU interconnects and have to go back to vector processing. Unless we are somehow able to jump the heat dissipation barrier and start raising GHz again.
Teacher salaries vary wildly by district, so blanket statements that teachers are way underpaid for their educational level are bogus. In my area, teachers earn pretty close to the median income for bachelors/Masters degree holders $50k-$70k.
Regarding teachers doing work during their "vacation" time: At least they get the time off to do that work. How many doctors, software developers and engineers nowadays are given 2-3 months a year of time off to further their education. Particularly those that work for corporations. It ain't soft for teachers, but the stressors are different.
Uh, No child left behind is mostly trying to teach kids reading, writing and math. This is pretty much a pre-requisite for vocational education anyhoo.
"Bore on" is probably how most HS students describe chemistry anyway. Maybe that's what they were complaining about.
I see two Filk songs out of this: 1) "Bore On" as a send-up of Buddy Holly's "Rave On", 2) The text of the test as a parody of "Shaft" - "What is the gas with 10.8 atomic mass? - Boron"
You talk about deductions and rates for federal income taxes, but most education funding comes from state property (and income) taxes.
There are 3 systemic problems in public education that need to be addressed, but funding is only an indirect issue: if we're taking in more money per student than most other countries but not seeing results, the problem is on the consumption side, not on the funding side.
1) Teacher competency and tenure - tenure should be granted based on merit over a period of several years, not merely on time served, which makes it too easy to slip through the cracks.
2) Administrative overhead. One big problem with public education as a government institution is that the administration accretes new functions over time to address such issues as corrupt hiring/contracting, discrimination and so on. A similar accretion process occurs in large corporations, and they respond by going through re-orgs and revamping their processes. Administrative functions need to be swept clean on a zero-budget and zero-process-baseline basis every 10 years or so to give a fresh start.
3) Parents/home environment - I don't think I need to reiterate the impact that this has on student performance. The public education system has to take everybody, and everybody should at least be functionally literate and able to manage their daily financial tasks when they graduate. There's not a lot that the schools can do to directly change the home environment, but they can do a better job of identifying students whose achievement suffers because of home issues and guide those students to sources of help.
Some additional random thoughts:
One administrative activity to look at might be districting students according to developmental needs, rather than merely by location. Make the schools for the developmentally (or environmentally) challenged and exceptional students as small facilities that can function as centers of expertise for the type of education they are performing. This will provide a stronger support network for the faculty and staff which can reduce the stress of constantly dealing with challenged students.
Privately funded education cherry picks their students, so they always look better on paper.
We often concentrate the debate on urban school performance, but in some ways the rural schools with very small enrollments are suffering the worst. This can really only be addressed by federal funding.
Actually, I think the teacher is engaging in a sort of civil disobedience to protest the lack of public funding for education.
I applaud this as a form of protest, not as an ongoing revenue source.
Turnitin.com couldn't index the works unless the teachers illegally redistributed the works to them in the first place. While I can possibly buy the argument that turnitin isn't redistributing the works themselves, the teachers and institutions that submit student works to turnitin.com ARE redistributing copyrighted works. Of course the easy way for the institutions to get around this is to put a clause in the student code of conduct that any work turned in as a classroom assignment gives the institution the right to redistribute it for purposes x, y and z.
Well in some areas (like California), non-competes are considered unenforceable. But on the other hand, California has very strict trade secret laws. So they would be unlikely to get you for violating the non-compete, but if your new product/corporation uses any material at all from the old company (document templates, change management processes, software build environment architectures), you'll get your asses sued.
That's why it's cheaper to pay a lawyer up front to analyze the legal risks in their business plan vs. any contractual or IP obligations they owe to their previous employer.
IANAL, but two big areas that would require detailed checking would be non-compete clauses in your contracts and trade secret laws. Have the lawyer look at everything though. You also could be making some invalid assumptions about how your technical plans insulate you from copyright/patent infringement.
In the old days, companies used separate black box/white box reverse engineering teams to isolate the authors of the new product from the people who knew about the original product. The white box team would write the design specs describing the external behavior of the program, but the black box team would write the actual code. You and your friends could end up having the company but not be allowed to do any of the coding on the new product in order to avoid IP violations. Worse yet, you may not be able to control the details of how the black box team implements the design, so all of your great ideas for re-architecting the system may go by the wayside.
Somehow I think you get paid a bit more than the average CSR at a phone, Cable or credit card company.
I agree there's no excuse for that type of dishonesty, but there's also no excuse for expecting professional ethics from your employees if you're paying minimum wage. At least until the minimum wage is enough for a full time employee to live on without being desperately short of cash.
Either that or it's a "give them enough rope to hang themselves" test. You tell the employees the policy and the consequences, and then wait to see if they're honest or not.
Better yet, lets work on some turbine designs that don't involve large blades sweeping through the air. With this "transmission" design, it should be possible to build some wind generators that can work efficiently without requiring the amount of torque that is needed to make the current designs work. If we can efficiently generate wind-power without the large torque requirements, no more bird-killers needed.
WOOT! Assuming it isn't just BS to get VC funding. The theory sounds reasonable, though.
Nahhh, that's just California. you left out Slashdot, strange sports not played in other countries ("football"), country "music" and political correctness.
BTW, Porn goes way back before American culture. We just commercialized it with mass production.
On the other side, the internet propagates new cultural memes and outlets so fast, it's hard to tell where the new ones are coming from. Is blogging originally an artifact of American culture that spread like TV and music?
IBM has a pretty decent-sized Linux operation. Odds are the AIX workstations are being ditched as redundant and non-competitive (price-wise) with Linux-based stations. Is there really any workstation software that is available on AIX that isn't available on Linux?
Surely this is a problem that has already been addressed by CalTech, MIT and Stanford (among others). Each of those colleges has had dozens of student works that spun off into businesses.
Math is the basis for so much, but no matter what profession she chooses, most of her time is going to be spent writing. Reports, design documents, advertising brochures, processes and procedures, presentation materials.
Every book in the US has it's bibliographic information and much of it's cataloging information on the copyright page (I forget the official publishing term for this, but you know the one I mean). If they make access and use rights to the database too onerous, the libraries can simply hire typists (keyboardists?) to enter the data by hand. Then there ain't an damn thing they can do about it since it wasn't sourced from their database.
This is where PHBs come from, too. People who erroneously believe they are competent do more damage than competent people who do evil seeking their own gain. These people are so incompetent they don't even realize that they ARE incompetent, whether it's due to lack of knowledge or just lack of ability.
I vote we start a divorce fund for the husband.