I am not the original poster, but my understanding is that a lot of the complexity of the site comes from a draconianly complex law that simply can't easily be implemented officially.
That is absolutely false. Tax preparation software that actually works does exist, and no one can claim our tax code is simple. I think it is safe to say that anything which can be written into law can be written into software. It may be very complex and hard to write software, but I guess that is why they spent $150-300 million (depending on which source you believe).
While he probably does just have more innate talent at math than you do, your anecdote does not give enough information to determine that. You may have spent more time studying math as a sophomore in college than he did, but that doesn't mean you had spent more time up to that point. Perhaps he spent more time studying in the 18-20 years before that. Maybe he had intellectually stimulating hobbies like chess while you liked fishing.
I never took Masters level math courses in college, but I am sure I would do better with much less effort than the average person. While that may be in part because of innate ability, it is also in large part because the last 30 years of my life have been spent in above average scholarly pursuits. I have no idea how to do rigorous proofs, but I have a better background than most which would help me learn quicker than average.
Considering we already do separate kids into different tracks, and this does give an advantage to those put in these tracks, it is a noble goal to identify as many students as possible who may not have been selected for advanced tracks but should have been. Early signs of aptitude can be hard to identify.
Texas will not only terminate a corporation, they charge a $40 fee for the paperwork processing. Such inhumane treatment. Next they will start charging the inmate's estate for the injections.
Are you telling me that all I have to do to get away with murder is form "Murder Inc." -- sorry, I meant "Loving Hands, Inc."? And then as chairman I can just hire a hit man and shield myself from liability? Shweet!!
Executives and board members can still be charged with actual crimes they personally committed or are an accessory to. But being bad at your job is not a crime. If they can show that executives or board members intentionally defrauded someone, then they can go to prison like anyone else. But if it was just their negligence, arrogance, lack of caring, gaming the system, etc. that caused the problem, they didn't commit any crimes.
what is it that corrupts them and causes the disastrous short term thinking epidemic in companies these days?
As the parent mentioned, they are immediately given incentives that reward short term thinking. If they don't grasp for short term solutions, they don't reach a VP position by the time they are 40. It is an endemic condition found throughout our entire economic system.
Only lazy developers. I've worked on plenty of legacy software over the years that other people wouldn't touch, and the common thread was they were either too lazy or just not smart enough. It takes determination, patience, and a lot of detail to work on it. But it was all far from unmaintainable.
When someone says that code is unmaintainable, they obviously don't mean that it is impossible to maintain. You can even disassemble binaries if absolutely necessary. If the effort to maintain the code exceeds the benefit of maintaining the code, it has effectively become unmaintainable in any practical sense.
I do agree that maintaining code is more difficult than creating new code, and on top of that it is a less desirable task for most developers. I had to leave one company because they kept putting me projects with legacy software because I performed better on those projects than my peers. They even bumped my pay because they knew it was important. The problem was I didn't want to do it, and finding a new job is easy for anyone with real skills. I wasn't lazy, I just had better options.
If you are someone who likes working with legacy software and are good at it, then you are valuable indeed. Most companies don't have someone like this, so they have to not only worry about the effort to maintain software but also the morale of the developers forced to work on projects with excessive technical debt.
I can't believe so many people agree with this post. We are talking about game systems that cost around $500 that are probably played on TVs that cost $500-$1000 with perhaps a dozens games costing $60 each. And you are saying that a $250 headset is excessive? I don't find sound quality very important either, but who am I to judge what others find important enough to spend their money on? A $250 expenditure on a hobby is still pretty tame compared to most. I spent close to that on just one of two tickets to the musical Wicked, even though a movie ticket would have been $15.
Spending $250 on a headset instead of perhaps $50-100 is really no different than buying a 55" TV instead of a 40" one.
In the states, we'd have a higher salary, pay less taxes, but pay more to live in a safe neighbourhood, work more, and still not have real heathcare. At least, that's my stance.
Remember that America gets poor grades on its health care because of unequal distribution and the reliance on life expectancy to measure health. When you rule out poor lifestyle choices (McDonalds, 8 hours/day of TV), homicides, and car accidents, the US has the highest life expectancy of any Western country (although other countries may argue they have some country specific non-healthcare related problems that should be ruled out). Also if you look at actual healthcare results, such as cancer survival rates, the United States is #1. So if you have a professional career (which basically guarantees a top 10% income), you will not get better health care than in the US.
But if you have kids the same is not true of our educational system. You have to pay through the nose to get into the very few decent school districts here.
Selection bias. When you're poor and burnt out, people stop being friends with you. I had lots of friends when I was earning $80k/year as an analyst, now I count maybe 3, and those are friendships out of pity that will soon wear out. Maybe you had some immigrant friends who are no longer friends, and you've written off. They're probably not doing so well.
I thought I was pretty clear that the experience of immigrants depends on where they fall in the socioeconomic ladder when I said:
The experience of immigrants really depends on where they fall in the socioeconomic ladder.
I was clear that coming to America is a great opportunity for people who are truly at the top of their professions. And it is fairly easy to know if this is true even at an early age. Did you go to the top universities in your country and finish at the top of your class? If yes, then moving to America will probably work out quite well for you. If not, don't expect roads paved with gold. This experience isn't much different than native born Americans.
I'm gonna show this post to my girlfriend, who still sometimes thinks we should move to the states.
The experience of immigrants really depends on where they fall in the socioeconomic ladder. All of my immigrant friends are in their late 20s to mid 30s with IT careers, and are well into the top 5-10% of earners in the US. Without exception they say their life is better here than anyone they know back home.
I am sure there are plenty of immigrants you come here and fail though, and their experiences are not nearly as good. The unfortunate thing about the US is that it probably is the best place in the world if you are highly skilled, but it is much worse for the other 80+% of the population.
But let's address other libertarian anarcho-capitalist concerns. Without big buck potential nobody would start the next Microsoft. Personally I think a $200,000 salary would make a lot of people really happy since 99.5% of the population would see a HUGE pay increase. The number of people making six figure salaries would increase. The number of people buying cars would increase. The number of people upgrading their ipad every year would dramatically increase.
But you still need to motivate that top 5-10% to do the innovative work that moved us from an agricultural society to the information age.
Based on my performance a while back in school (as a severe slacker at the time), I have the innate ability be in the top 10% of performers in my field with virtually no work. I am sure people like Bill Gates, Steve Jobs, etc. could do the same. But in today's society, I still have to work very hard to make over $100k. I do this work because I would probably only make around $60k or so if I was coasting through life.
But if I made $150-200k per year while coasting through life (which I probably would in your model), I surely would not be as productive. I like software development, but I still like watching football, playing video games, and spending time with friends/family more. But considering the number of metaphorical fires I put out for my clients, I think they are very happy that I work as hard as I do. If it was easy to find people to replace me, I would be paid far less, and I would find far less projects that were failing for years before people like me came along.
If I wasn't working that hard, more projects would fail. And I am not that unique; this would be true for a huge number of highly skilled workers. I have a few lawyer and doctor friends with far higher incomes than mine, and none of them would put the same hours if they could make six figures with minimal effort. And none of this even approaches the top producers in our society whose accomplishes dwarf mine.
The very source you posted claimed that their model represents 1% of the human brain. By just using Moore's law we would reach a computer about 100x faster in just under 10 years.
Just like computation goes back to the Babylonians, yet it would be ridiculous to attribute the IBM PC to them.
Machine learning of today is very similar to the machine learning of the 50s, or at least far more similar than todays computers are to ancient computing. The parent post even provided concrete examples of machine learning techniques that were widely used by the 50s (which you obviously didn't even read, or perhaps didn't understand):
Says the guy who uses a Computer World article as a reference.
As I said, I'm referring to a well known failure, so much so that it has its own entry in Wikipedia.
Am I reading this correctly? First you criticize him for using Computer World as a reference, and then you go on to use Wikipedia to prove your point?
He is worried that the assessments themselves will be very expensive. It is not the specialized classes that would cost extra, but the assessment that determines which classes to take make be more thorough if you can spend money for private testing. I am not commenting on whether I agree with him, but that is his contention.
Who supervises these students to 1) give them the curriculum and 2) make sure they follow it?
Once again, I never said teachers will ever be completely out of the loop. At the absolute minimum you will always still need adults there for simple crowd control. It is not much different than modern factories. There are still factory workers, just less of them and they do different tasks than they did 50 years ago. Although for teaching I hope there aren't many reductions in the number of teachers, just a shift of what their common tasks are.
What you are advocating is turning our children into guinea pigs. Let's throw hundreds of thousands of math problems at them and see which ones work and which ones don't. Let's throw hundreds of different lectures at them and see which ones work and which ones don't. If your child is the unlucky one to get the problems and lectures that didn't work, well, sorry, maybe your next child in line will get the education your first one missed out on.
Actually today we are treating children as guinea pigs but with no positive benefits. Today there are about 90,000 elementary teachers giving 90,000 different lectures each day and giving 90,000 different sets of test questions and homework assignments to their students each chapter/unit/etc. Standardized testings and curriculum at the federal, state, district, and school level reduce those numbers somewhat, but students are still getting millions of unique lectures/homework/tests each year.
This is incredibly inefficient and it is very difficult to measure what works and what doesn't. I am advocating that we reduce the number of different lectures and try to determine which ones are working best.
As for "lecture providers", that's already gone. You've got half of what you want, let's not throw the baby out with the bathwater, ok?
I am not aware of any such change. Teachers still spend quite a bit of time lecturing in modern classrooms.
Wouldn't it be nice if there were enough money in the pool so that teachers could be personal tutors instead of being in charge of a whole class.
It would be nice, but it doesn't necessarily take more money or more teachers. Automating and consolidating common tasks such as lecturing, creating homework / tests, and basic Q&A could free them to spend more time on other tasks. I used to write software for pharmacists, and our software significantly reduced the number of pharmacists it took to complete their work. But pharmacists were rarely fired; instead the hospitals found other work for them that improved the quality of patient care even more.
Should your experience be the guide for everyone, or should we admit that planning on the exceptional is not a good plan?
I think using the exceptional for a guide is a great idea. Educational powerhouses like Korea and Finland show that if you expect more from students they usually rise to the challenge. My experience wasn't someone who defied the odds and self-taught despite some horrible environment fighting against me (a "lottery ticket" type scenario). I was simply brought up in an environment that encouraged self-discovery. That is not an environment that schools are incapable of emulating. (even though very few do this today)
Anyone can learn just what they want to learn without supervision. Getting an education means sometimes being forced to learn things that you don't think are fun or exciting.
That can still be done with self-learning. A student can still be given a curriculum that includes topics they aren't interested in. And testing is a pretty good way of determining if the students are actually paying attention. In fact, new technology will make creating metrics even easier. Eye tracking software can determine if students are paying attention when watching a video lecture, questions that determine comprehension can be delivered throughout the lecture to test comprehension. Supplemental lectures could be shown to go over areas that the student is struggling in so every single student has their own "private tutor". And like I said in another post, the real breakthroughs will be techniques I can't even imagine right now.
You don't need hundreds of thousands of math problems to teach math. In fact, if you need to have that many you are doing something wrong. The principles can be taught with far fewer.
I never said you need that many questions per student. But how do you determine which questions are working and which ones aren't? Without computers we are limited to the hundred questions sitting in some textbook, with no scientific rigor in selecting those questions. In the near future we can rely on innovators that create new ways to educate and can easily test every hypothesis on tens of thousands of students in thousands of different permutations to see which work and which don't.
There could even be hundreds of different lectures on every topic, each given to thousands of different students. Tests would then be able to determine which lectures are working and which aren't. Some students may benefit from different types of instruction than other students. Each student could watch their own individually tailored lectures and have their own individually tailored projects and be given their own individually tailored homework and test questions.
There are so many options once human manpower is no longer the limiting factor.
There are a large number of students who would sit reading comic books all day if they were left to their own devices. Before the demise of the pinball parlor, that's where a lot of them would spend their time.
I never said to do away with school. Just the reliance of human teachers to relay most information to students. I led off my original post with saying that teachers are still important to the education process. But teachers are not the only way to provide education, and certain tasks such as creating curriculums, lecturing, creating homework / test questions, and tracking performance are not what we should be using them for in the near future.
I look forward to a day where teachers are more like personal tutors than the lecture providers that they are today.
Speaking as someone who has started and sold a couple of small businesses, I can promise you that Obamacare will NOT make it easier. There is even a tax for every employee, whether you have health insurance or not.
They were talking about tech jobs, which are going to have health insurance anyway. You may be right that Obamacare will hurt new restaurants and similar retail jobs, but not high tech startups.
That is a pretty silly argument. Employees don't have to worry about losing their employer provided auto insurance or house. That doesn't even take into account that your health is much more important than you car or house.
For the majority of students, your experience is irrelevant. Were we to base our educational system on it today, we'd have a large number of people who are unqualified to be cashiers at a burger joint, and a few who manage to take advantage of them because they know how.
I don't see how my experience is irrelevant. I, and others I know, are examples that the kind of direct supervision that teachers provide is not necessary to learn. That doesn't mean that there is no need for institutions of education, just that teachers aren't the only source of instruction.
If it wasn't for the people writing textbooks and computer manuals I wouldn't have learned anything on my own. If I didn't have people motivating me to learn (my parents in this case) I wouldn't have learned anything on my own. If I didn't have real-world projects to work on that required me to learn the parts of programming that I didn't like at the time (database queries for example), I wouldn't have had a very well rounded independent programming education.
But none of these things requires teachers. A well done curriculum can handle all of it. Teachers will still be very useful in the education process, but using them as a primary form of instruction is incredibly wasteful without today's technology.
One of the big problems is that we teach with technology today in a very similar way that we teach without technology, but just replace paper with a screen. But computers can revolutionize how education is done. Databases of hundreds of thousands of math questions can be created, and textbooks could have access to all of them. Algorithms could track which students answer which questions, and determine which assignments have the largest impact on comprehension. Truthfully the biggest improvements will be things I can't even think about until some far more knowledgeable, creative, and ambitious people come up with them.
instead, people have bought the line that teachers are "overpaid" and don't bother to realize that teachers earn incredibly low salaries for the education and professional level of their work [epi.org]. that is insane.
They buy the line because it is accurate. The study you linked to is so ridiculous it must have been written by some union reps.
First it says that teachers are paid less than similar professionals in private industry. But they don't say how they determine who similar professionals are. They don't mention if they take into account that the average teacher had below average SAT scores, or that only 1 in 5 teachers finish in the top 30% of their high school class. So teachers aren't just below average compared to other college bound students, they are below average when you count students who aren't even college bound. Based on the quality of students who want to become teachers, it is not surprising that the average teacher makes 12% less than the average non-teacher professional.
Then they say that after factoring in benefits, teachers make 9% less instead of 12% less. That is ridiculous. That would mean teacher's benefits are about $150 per month for someone making $60k per year. I would contend that teacher's benefits are worth more like $1500 per month. And that doesn't count increased vacation days, which probably bring their benefits' value up to $2000-2500 per month (for someone making $60k/yr). So by my calculations the portion of a teacher's benefits that are above and beyond what private employees get account for about 30% of their actual compensation.
As far as I'm concerned, if teacher's salaries are only 12% less than similar professionals, their actual income is at least 15% above similar professionals after factoring in benefits. And considering I seriously doubt that teachers' salaries are actually higher than private professionals of similar ability / intelligence / ambition, I contend that teachers make about 30-40% more than similar professionals in private industry.
Perhaps being a teacher is such a crummy job that you need to pay that much. One of my good friends is a garbage collector and he makes over $100k after including benefits, and they still have high turnover. But I think the real reason more qualified people don't become teachers is because of the public perception that teachers are underpaid, not because of the actual data.
A professionally trained, well-paid *human* teacher is absolutely the only thing that educates a child.
HAHAHA. That joke made my morning.
While I'm not saying teachers are unimportant, but are you actually saying that children are unable to learn without supervision? I remember writing my own computer games half a decade before my first programming class in college. And I am pretty sure no one taught me math from the fourth grade (when my parents started buying me textbooks) until sophomore year in college. Our teaching colleges are so poor that I couldn't even ask questions to enhance my independent learning because even recently graduated high school math teachers rarely know calculus (and I'm talking basic stuff, I wasn't a genius or anything).
I did learn a great deal from teachers in other subjects, such as rhetoric, but they were the exception to the rule.
Well tell that PM that new procedures are more time over all and more man power if you are getting done / ready for finale / RC QA testing just days before go live.
Oh that is cute, thinking that a PM with such ridiculous opinions can be reasoned with. Many people did try, but eventually it was only fixed when higher level management got fed up and fired the PMs in question.
But now that I work as a consultant, I find these kinds of PMs all over the place. Luckily now it is easier to just not work with these kinds of people, or at least keep racking up billable hours fixing the problems caused by their ridiculousness.
Rationing is a horrible idea because it just reproduces that problem, in addition to putting you on long ass waiting lists for even basic operations, and making the medical field less attractive as a career choice. It's already bad enough that we have waiting lists for organ transplants (Which by the way this problem is very solvable - have a look at how Iran does transplants. With as much shit as that country gets wrong, they shockingly nailed that one better than anybody else.)
That is only if you have a central body such as the government doing the rationing. You could also ration care the free market way by making poor people unable to pay for certain care. Regardless of the method, we need to understand that as we continue to make new medical breakthroughs we simply cannot spend all of our resources on keeping people alive forever.
If we finally start making breakthroughs that increase our healthy age (the age where you can work) then it isn't as much of a problem. But currently all we are doing is keeping people barely alive and spending a fortune to do it.
Slashdot Mirror
I am not the original poster, but my understanding is that a lot of the complexity of the site comes from a draconianly complex law that simply can't easily be implemented officially.
That is absolutely false. Tax preparation software that actually works does exist, and no one can claim our tax code is simple. I think it is safe to say that anything which can be written into law can be written into software. It may be very complex and hard to write software, but I guess that is why they spent $150-300 million (depending on which source you believe).
While he probably does just have more innate talent at math than you do, your anecdote does not give enough information to determine that. You may have spent more time studying math as a sophomore in college than he did, but that doesn't mean you had spent more time up to that point. Perhaps he spent more time studying in the 18-20 years before that. Maybe he had intellectually stimulating hobbies like chess while you liked fishing.
I never took Masters level math courses in college, but I am sure I would do better with much less effort than the average person. While that may be in part because of innate ability, it is also in large part because the last 30 years of my life have been spent in above average scholarly pursuits. I have no idea how to do rigorous proofs, but I have a better background than most which would help me learn quicker than average.
And only study them, instead doing a broad survey and looking for outliers in the data. Great fucking science, folks.
I just assumed that we already have many genomes sequenced that came from the general population that they could compare their results to.
Considering we already do separate kids into different tracks, and this does give an advantage to those put in these tracks, it is a noble goal to identify as many students as possible who may not have been selected for advanced tracks but should have been. Early signs of aptitude can be hard to identify.
Texas will not only terminate a corporation, they charge a $40 fee for the paperwork processing. Such inhumane treatment. Next they will start charging the inmate's estate for the injections.
Are you telling me that all I have to do to get away with murder is form "Murder Inc." -- sorry, I meant "Loving Hands, Inc."? And then as chairman I can just hire a hit man and shield myself from liability? Shweet!!
Executives and board members can still be charged with actual crimes they personally committed or are an accessory to. But being bad at your job is not a crime. If they can show that executives or board members intentionally defrauded someone, then they can go to prison like anyone else. But if it was just their negligence, arrogance, lack of caring, gaming the system, etc. that caused the problem, they didn't commit any crimes.
what is it that corrupts them and causes the disastrous short term thinking epidemic in companies these days?
As the parent mentioned, they are immediately given incentives that reward short term thinking. If they don't grasp for short term solutions, they don't reach a VP position by the time they are 40. It is an endemic condition found throughout our entire economic system.
Only lazy developers. I've worked on plenty of legacy software over the years that other people wouldn't touch, and the common thread was they were either too lazy or just not smart enough. It takes determination, patience, and a lot of detail to work on it. But it was all far from unmaintainable.
When someone says that code is unmaintainable, they obviously don't mean that it is impossible to maintain. You can even disassemble binaries if absolutely necessary. If the effort to maintain the code exceeds the benefit of maintaining the code, it has effectively become unmaintainable in any practical sense.
I do agree that maintaining code is more difficult than creating new code, and on top of that it is a less desirable task for most developers. I had to leave one company because they kept putting me projects with legacy software because I performed better on those projects than my peers. They even bumped my pay because they knew it was important. The problem was I didn't want to do it, and finding a new job is easy for anyone with real skills. I wasn't lazy, I just had better options.
If you are someone who likes working with legacy software and are good at it, then you are valuable indeed. Most companies don't have someone like this, so they have to not only worry about the effort to maintain software but also the morale of the developers forced to work on projects with excessive technical debt.
I can't believe so many people agree with this post. We are talking about game systems that cost around $500 that are probably played on TVs that cost $500-$1000 with perhaps a dozens games costing $60 each. And you are saying that a $250 headset is excessive? I don't find sound quality very important either, but who am I to judge what others find important enough to spend their money on? A $250 expenditure on a hobby is still pretty tame compared to most. I spent close to that on just one of two tickets to the musical Wicked, even though a movie ticket would have been $15.
Spending $250 on a headset instead of perhaps $50-100 is really no different than buying a 55" TV instead of a 40" one.
In the states, we'd have a higher salary, pay less taxes, but pay more to live in a safe neighbourhood, work more, and still not have real heathcare. At least, that's my stance.
Remember that America gets poor grades on its health care because of unequal distribution and the reliance on life expectancy to measure health. When you rule out poor lifestyle choices (McDonalds, 8 hours/day of TV), homicides, and car accidents, the US has the highest life expectancy of any Western country (although other countries may argue they have some country specific non-healthcare related problems that should be ruled out). Also if you look at actual healthcare results, such as cancer survival rates, the United States is #1. So if you have a professional career (which basically guarantees a top 10% income), you will not get better health care than in the US.
But if you have kids the same is not true of our educational system. You have to pay through the nose to get into the very few decent school districts here.
Selection bias.
When you're poor and burnt out, people stop being friends with you. I had lots of friends when I was earning $80k/year as an analyst, now I count maybe 3, and those are friendships out of pity that will soon wear out.
Maybe you had some immigrant friends who are no longer friends, and you've written off. They're probably not doing so well.
I thought I was pretty clear that the experience of immigrants depends on where they fall in the socioeconomic ladder when I said:
The experience of immigrants really depends on where they fall in the socioeconomic ladder.
I was clear that coming to America is a great opportunity for people who are truly at the top of their professions. And it is fairly easy to know if this is true even at an early age. Did you go to the top universities in your country and finish at the top of your class? If yes, then moving to America will probably work out quite well for you. If not, don't expect roads paved with gold. This experience isn't much different than native born Americans.
I'm gonna show this post to my girlfriend, who still sometimes thinks we should move to the states.
The experience of immigrants really depends on where they fall in the socioeconomic ladder. All of my immigrant friends are in their late 20s to mid 30s with IT careers, and are well into the top 5-10% of earners in the US. Without exception they say their life is better here than anyone they know back home.
I am sure there are plenty of immigrants you come here and fail though, and their experiences are not nearly as good. The unfortunate thing about the US is that it probably is the best place in the world if you are highly skilled, but it is much worse for the other 80+% of the population.
But let's address other libertarian anarcho-capitalist concerns. Without big buck potential nobody would start the next Microsoft. Personally I think a $200,000 salary would make a lot of people really happy since 99.5% of the population would see a HUGE pay increase. The number of people making six figure salaries would increase. The number of people buying cars would increase. The number of people upgrading their ipad every year would dramatically increase.
But you still need to motivate that top 5-10% to do the innovative work that moved us from an agricultural society to the information age.
Based on my performance a while back in school (as a severe slacker at the time), I have the innate ability be in the top 10% of performers in my field with virtually no work. I am sure people like Bill Gates, Steve Jobs, etc. could do the same. But in today's society, I still have to work very hard to make over $100k. I do this work because I would probably only make around $60k or so if I was coasting through life.
But if I made $150-200k per year while coasting through life (which I probably would in your model), I surely would not be as productive. I like software development, but I still like watching football, playing video games, and spending time with friends/family more. But considering the number of metaphorical fires I put out for my clients, I think they are very happy that I work as hard as I do. If it was easy to find people to replace me, I would be paid far less, and I would find far less projects that were failing for years before people like me came along.
If I wasn't working that hard, more projects would fail. And I am not that unique; this would be true for a huge number of highly skilled workers. I have a few lawyer and doctor friends with far higher incomes than mine, and none of them would put the same hours if they could make six figures with minimal effort. And none of this even approaches the top producers in our society whose accomplishes dwarf mine.
The very source you posted claimed that their model represents 1% of the human brain. By just using Moore's law we would reach a computer about 100x faster in just under 10 years.
Machine learning goes back to the 1950's
Just like computation goes back to the Babylonians, yet it would be ridiculous to attribute the IBM PC to them.
Machine learning of today is very similar to the machine learning of the 50s, or at least far more similar than todays computers are to ancient computing. The parent post even provided concrete examples of machine learning techniques that were widely used by the 50s (which you obviously didn't even read, or perhaps didn't understand):
Says the guy who uses a Computer World article as a reference.
As I said, I'm referring to a well known failure, so much so that it has its own entry in Wikipedia.
Am I reading this correctly? First you criticize him for using Computer World as a reference, and then you go on to use Wikipedia to prove your point?
He is worried that the assessments themselves will be very expensive. It is not the specialized classes that would cost extra, but the assessment that determines which classes to take make be more thorough if you can spend money for private testing. I am not commenting on whether I agree with him, but that is his contention.
Who supervises these students to 1) give them the curriculum and 2) make sure they follow it?
Once again, I never said teachers will ever be completely out of the loop. At the absolute minimum you will always still need adults there for simple crowd control. It is not much different than modern factories. There are still factory workers, just less of them and they do different tasks than they did 50 years ago. Although for teaching I hope there aren't many reductions in the number of teachers, just a shift of what their common tasks are.
What you are advocating is turning our children into guinea pigs. Let's throw hundreds of thousands of math problems at them and see which ones work and which ones don't. Let's throw hundreds of different lectures at them and see which ones work and which ones don't. If your child is the unlucky one to get the problems and lectures that didn't work, well, sorry, maybe your next child in line will get the education your first one missed out on.
Actually today we are treating children as guinea pigs but with no positive benefits. Today there are about 90,000 elementary teachers giving 90,000 different lectures each day and giving 90,000 different sets of test questions and homework assignments to their students each chapter/unit/etc. Standardized testings and curriculum at the federal, state, district, and school level reduce those numbers somewhat, but students are still getting millions of unique lectures/homework/tests each year.
This is incredibly inefficient and it is very difficult to measure what works and what doesn't. I am advocating that we reduce the number of different lectures and try to determine which ones are working best.
As for "lecture providers", that's already gone. You've got half of what you want, let's not throw the baby out with the bathwater, ok?
I am not aware of any such change. Teachers still spend quite a bit of time lecturing in modern classrooms.
Wouldn't it be nice if there were enough money in the pool so that teachers could be personal tutors instead of being in charge of a whole class.
It would be nice, but it doesn't necessarily take more money or more teachers. Automating and consolidating common tasks such as lecturing, creating homework / tests, and basic Q&A could free them to spend more time on other tasks. I used to write software for pharmacists, and our software significantly reduced the number of pharmacists it took to complete their work. But pharmacists were rarely fired; instead the hospitals found other work for them that improved the quality of patient care even more.
The same can be done for teachers.
Should your experience be the guide for everyone, or should we admit that planning on the exceptional is not a good plan?
I think using the exceptional for a guide is a great idea. Educational powerhouses like Korea and Finland show that if you expect more from students they usually rise to the challenge. My experience wasn't someone who defied the odds and self-taught despite some horrible environment fighting against me (a "lottery ticket" type scenario). I was simply brought up in an environment that encouraged self-discovery. That is not an environment that schools are incapable of emulating. (even though very few do this today)
Anyone can learn just what they want to learn without supervision. Getting an education means sometimes being forced to learn things that you don't think are fun or exciting.
That can still be done with self-learning. A student can still be given a curriculum that includes topics they aren't interested in. And testing is a pretty good way of determining if the students are actually paying attention. In fact, new technology will make creating metrics even easier. Eye tracking software can determine if students are paying attention when watching a video lecture, questions that determine comprehension can be delivered throughout the lecture to test comprehension. Supplemental lectures could be shown to go over areas that the student is struggling in so every single student has their own "private tutor". And like I said in another post, the real breakthroughs will be techniques I can't even imagine right now.
You don't need hundreds of thousands of math problems to teach math. In fact, if you need to have that many you are doing something wrong. The principles can be taught with far fewer.
I never said you need that many questions per student. But how do you determine which questions are working and which ones aren't? Without computers we are limited to the hundred questions sitting in some textbook, with no scientific rigor in selecting those questions. In the near future we can rely on innovators that create new ways to educate and can easily test every hypothesis on tens of thousands of students in thousands of different permutations to see which work and which don't.
There could even be hundreds of different lectures on every topic, each given to thousands of different students. Tests would then be able to determine which lectures are working and which aren't. Some students may benefit from different types of instruction than other students. Each student could watch their own individually tailored lectures and have their own individually tailored projects and be given their own individually tailored homework and test questions.
There are so many options once human manpower is no longer the limiting factor.
There are a large number of students who would sit reading comic books all day if they were left to their own devices. Before the demise of the pinball parlor, that's where a lot of them would spend their time.
I never said to do away with school. Just the reliance of human teachers to relay most information to students. I led off my original post with saying that teachers are still important to the education process. But teachers are not the only way to provide education, and certain tasks such as creating curriculums, lecturing, creating homework / test questions, and tracking performance are not what we should be using them for in the near future.
I look forward to a day where teachers are more like personal tutors than the lecture providers that they are today.
Speaking as someone who has started and sold a couple of small businesses, I can promise you that Obamacare will NOT make it easier. There is even a tax for every employee, whether you have health insurance or not.
They were talking about tech jobs, which are going to have health insurance anyway. You may be right that Obamacare will hurt new restaurants and similar retail jobs, but not high tech startups.
That is a pretty silly argument. Employees don't have to worry about losing their employer provided auto insurance or house. That doesn't even take into account that your health is much more important than you car or house.
For the majority of students, your experience is irrelevant. Were we to base our educational system on it today, we'd have a large number of people who are unqualified to be cashiers at a burger joint, and a few who manage to take advantage of them because they know how.
I don't see how my experience is irrelevant. I, and others I know, are examples that the kind of direct supervision that teachers provide is not necessary to learn. That doesn't mean that there is no need for institutions of education, just that teachers aren't the only source of instruction.
If it wasn't for the people writing textbooks and computer manuals I wouldn't have learned anything on my own.
If I didn't have people motivating me to learn (my parents in this case) I wouldn't have learned anything on my own.
If I didn't have real-world projects to work on that required me to learn the parts of programming that I didn't like at the time (database queries for example), I wouldn't have had a very well rounded independent programming education.
But none of these things requires teachers. A well done curriculum can handle all of it. Teachers will still be very useful in the education process, but using them as a primary form of instruction is incredibly wasteful without today's technology.
One of the big problems is that we teach with technology today in a very similar way that we teach without technology, but just replace paper with a screen. But computers can revolutionize how education is done. Databases of hundreds of thousands of math questions can be created, and textbooks could have access to all of them. Algorithms could track which students answer which questions, and determine which assignments have the largest impact on comprehension. Truthfully the biggest improvements will be things I can't even think about until some far more knowledgeable, creative, and ambitious people come up with them.
instead, people have bought the line that teachers are "overpaid" and don't bother to realize that teachers earn incredibly low salaries for the education and professional level of their work [epi.org]. that is insane.
They buy the line because it is accurate. The study you linked to is so ridiculous it must have been written by some union reps.
First it says that teachers are paid less than similar professionals in private industry. But they don't say how they determine who similar professionals are. They don't mention if they take into account that the average teacher had below average SAT scores, or that only 1 in 5 teachers finish in the top 30% of their high school class. So teachers aren't just below average compared to other college bound students, they are below average when you count students who aren't even college bound. Based on the quality of students who want to become teachers, it is not surprising that the average teacher makes 12% less than the average non-teacher professional.
Then they say that after factoring in benefits, teachers make 9% less instead of 12% less. That is ridiculous. That would mean teacher's benefits are about $150 per month for someone making $60k per year. I would contend that teacher's benefits are worth more like $1500 per month. And that doesn't count increased vacation days, which probably bring their benefits' value up to $2000-2500 per month (for someone making $60k/yr). So by my calculations the portion of a teacher's benefits that are above and beyond what private employees get account for about 30% of their actual compensation.
As far as I'm concerned, if teacher's salaries are only 12% less than similar professionals, their actual income is at least 15% above similar professionals after factoring in benefits. And considering I seriously doubt that teachers' salaries are actually higher than private professionals of similar ability / intelligence / ambition, I contend that teachers make about 30-40% more than similar professionals in private industry.
Perhaps being a teacher is such a crummy job that you need to pay that much. One of my good friends is a garbage collector and he makes over $100k after including benefits, and they still have high turnover. But I think the real reason more qualified people don't become teachers is because of the public perception that teachers are underpaid, not because of the actual data.
A professionally trained, well-paid *human* teacher is absolutely the only thing that educates a child.
HAHAHA. That joke made my morning.
While I'm not saying teachers are unimportant, but are you actually saying that children are unable to learn without supervision? I remember writing my own computer games half a decade before my first programming class in college. And I am pretty sure no one taught me math from the fourth grade (when my parents started buying me textbooks) until sophomore year in college. Our teaching colleges are so poor that I couldn't even ask questions to enhance my independent learning because even recently graduated high school math teachers rarely know calculus (and I'm talking basic stuff, I wasn't a genius or anything).
I did learn a great deal from teachers in other subjects, such as rhetoric, but they were the exception to the rule.
Well tell that PM that new procedures are more time over all and more man power if you are getting done / ready for finale / RC QA testing just days before go live.
Oh that is cute, thinking that a PM with such ridiculous opinions can be reasoned with. Many people did try, but eventually it was only fixed when higher level management got fed up and fired the PMs in question.
But now that I work as a consultant, I find these kinds of PMs all over the place. Luckily now it is easier to just not work with these kinds of people, or at least keep racking up billable hours fixing the problems caused by their ridiculousness.
Rationing is a horrible idea because it just reproduces that problem, in addition to putting you on long ass waiting lists for even basic operations, and making the medical field less attractive as a career choice. It's already bad enough that we have waiting lists for organ transplants (Which by the way this problem is very solvable - have a look at how Iran does transplants. With as much shit as that country gets wrong, they shockingly nailed that one better than anybody else.)
That is only if you have a central body such as the government doing the rationing. You could also ration care the free market way by making poor people unable to pay for certain care. Regardless of the method, we need to understand that as we continue to make new medical breakthroughs we simply cannot spend all of our resources on keeping people alive forever.
If we finally start making breakthroughs that increase our healthy age (the age where you can work) then it isn't as much of a problem. But currently all we are doing is keeping people barely alive and spending a fortune to do it.