"Teach Yourself Programming in Ten Years" by Peter Norvig (http://norvig.com/21-days.html) is still an excellent discussion on how long it will take to learn programming to a degree that is actually adequate to do professional-level work.
In addition, I expect that most people will never get beyond "fair", even with this amount of training and experience, just as most people will not become much better than "fair" at any other task, unless they have some real talent/potential/gift/deal-with-the-devil/etc. whether it be cooking, mathematics, or playing the harp. That is not to discourage people. "Fair" would be a huge improvements over the skill-level of many of the people that write software today. "Fair" is level on which you can depend on people to get regular stuff right and to know what is beyond their skills ans ask for help.
Of course, "fair" is also the level where the expected salary needs to be very reasonable (say, at least enough for a 4 people family to live in modest comfort off it) in order to make people go though the long process of becoming good at it. In particular the latter is lacking today.
A fast surge will create a huge inrush current into a device running at 110V when the surge protector is dimensioned for 220V. That has a good change to damage or kill the device. On a slow/weak surge, it will offer reasonable protection.
If you have a 100-240V "compatible" surge protector, then you have been cheated, as that is not possible to do reliably without significantly more effort than goes into these devices. These will always only offer limited protection at 110V.
Indeed. Even when you have a 100-240V device connected to 110V, you must use an 110V surge protector. The problem is that while a 220V surge protector would clamp at a voltage the device can survive, it can only survive reliably if it has been running at 220V because of the way these devices are designed. They have a rectifier and filter capacitor. If the filter capacitor gets charged up from 110V by a surge clamped for 220V (which clamps at around 400...500V), the inrush-current will likely blow the last-ditch fuse in the device and may well damage other components.
So, sorry, what you want is not possible. You must get both.
... are called children, and in the regular age range (which certainly does not go beyond 15 or so) they have a valid excuse, namely lack of maturity. These overgrown children that presume to call themselves "students", when they are anything but have no such excuse. On the other hand, they are destroying the change to grow and learn anything relevant in their sanitized campuses, so they will get what they deserve. The only problem are all those that do not have good opportunities to get an actual education someplace else.
For variable values of "best". If I look at systemd, which would at least need another few years of time and some lead developers with an actual clue about Unix to be ready for prime time, I highly doubt that Red Hat is a good choice for anything.
What do you mean "lock down the hardware"? I expect my hardware to be locked down and not change its settings unless I go to the BIOS set-up. My take is that systemd is very much at fault here for not protecting the user.
My guess is that they have cash-flow problems and they now think pissing-off potential customers is the way to go. You know, like the music and movie industries.
On the side of solid engineering practices, they can refuse to talk to that counterfeit device by not detecting it or giving an out-of-band error on detection, but that is it. Breaking the hardware intentionally is sabotage and exceptionally unethical. Being willing to work with the device but then injecting data into the data-stream intentionally is the same. If anything bad happens as a result, this is the step that comes after gross negligence: It is called "intent".
While I do expect they will have had this cleared from a legal perspective and will be hard or impossible to attack, from an engineering perspective there is only one valid way to deal with this: To not ever use their products until they have credible sworn off their evil ways.
Well, possibly. Most people also do not have actual intelligence at their disposal, but just copy what others do without understanding what they are doing.
Your argument is insofar a good one as machines may be able to get to that level eventually. That is why I usually write "smart human beings" when illustrating to what level machines will likely never get. But if you are right, machines will, for example, never write meaningful code, do translation of more than superficial conversations, do meaningful research, etc., just as most human beings.
I do disagree on the proportions though: About 10-20% of humans are able to do more than the others can. They are capable of independent thought and insight. They are capable of identifying bullshit when presented with it and they can understand things from scratch that they never have heard about before. Incidentally, there seems to be only a very lose connection to intelligence. These people are hot candidates for actually having consciousness and that could be exactly why they are capable of more.
The "sum" part describes the configuration. In physics, the function of the whole is fully determined by the function of all its parts and the microscopic relations between them. There are no unexpected or "magic" properties that suddenly manifest. In physics there are no "emergent properties" at all.
My brain could be considered a biological machine and it implements human-equivalent intelligence (or at least the people around me seem to think so, perhaps I have them all fooled).
Only if you assume physicalism is right. If you do that, you can of course demonstrate (by a few intermediate step) that physicalism is right. But that reasoning is circular and meaningless. Most physicalists fall for it though, but this is not science. It is more in line with religion.
Your position is certainly not a scientific one and you are rather severely misinformed about what software can do (most people are).
Can you point out a specific problem with any part of what I've stated or demonstrate with sufficient proof that it won't work or cannot be done, not just now, but also at no point in the future?
You have not pointed out anything with sufficient proof. You are arguing for a restricted model with insufficient proof. That is my whole point.
You have no idea how the real world works. The only people that will decline for that reason are those with huge egos and typically rather mediocre skills. These are the people that like to be described as "rock-stars". Hint: None of them are.
However, even with an NDA you can get a lot of interesting insights when doing such a job. You can just not use them directly or use them to compete with the business where you had them. But that is the deal any employer with interesting technology has to make. As soon as people are working on it, they are going to learn and improve their skills and insights. There is no way around that.
Slashdot Mirror
"Teach Yourself Programming in Ten Years" by Peter Norvig (http://norvig.com/21-days.html) is still an excellent discussion on how long it will take to learn programming to a degree that is actually adequate to do professional-level work.
In addition, I expect that most people will never get beyond "fair", even with this amount of training and experience, just as most people will not become much better than "fair" at any other task, unless they have some real talent/potential/gift/deal-with-the-devil/etc. whether it be cooking, mathematics, or playing the harp. That is not to discourage people. "Fair" would be a huge improvements over the skill-level of many of the people that write software today. "Fair" is level on which you can depend on people to get regular stuff right and to know what is beyond their skills ans ask for help.
Of course, "fair" is also the level where the expected salary needs to be very reasonable (say, at least enough for a 4 people family to live in modest comfort off it) in order to make people go though the long process of becoming good at it. In particular the latter is lacking today.
Most clueless posting so far, USB power is DC, no "waveform" present. The actual problem is that they are often too weak and can heat up dangerously.
A fast surge will create a huge inrush current into a device running at 110V when the surge protector is dimensioned for 220V. That has a good change to damage or kill the device. On a slow/weak surge, it will offer reasonable protection.
If you have a 100-240V "compatible" surge protector, then you have been cheated, as that is not possible to do reliably without significantly more effort than goes into these devices. These will always only offer limited protection at 110V.
No. Because of the way surges, surge-protectors work and wide-range PSUs work, this is not true at all.
Indeed. Even when you have a 100-240V device connected to 110V, you must use an 110V surge protector. The problem is that while a 220V surge protector would clamp at a voltage the device can survive, it can only survive reliably if it has been running at 220V because of the way these devices are designed. They have a rectifier and filter capacitor. If the filter capacitor gets charged up from 110V by a surge clamped for 220V (which clamps at around 400...500V), the inrush-current will likely blow the last-ditch fuse in the device and may well damage other components.
So, sorry, what you want is not possible. You must get both.
Google is an ad-company (and maybe an NSA front, who knows), while Apple is a devices company.
You forget that many slashdotters are functionally illiterate: They can read the words, but the meaning escapes them.
Spoken like a true cave-man. It is surprising you can even read and write.
... are called children, and in the regular age range (which certainly does not go beyond 15 or so) they have a valid excuse, namely lack of maturity. These overgrown children that presume to call themselves "students", when they are anything but have no such excuse. On the other hand, they are destroying the change to grow and learn anything relevant in their sanitized campuses, so they will get what they deserve. The only problem are all those that do not have good opportunities to get an actual education someplace else.
John Cleese is insightful, as always.
Serves to show that on average, statistics are not a very good model of reality.
For variable values of "best". If I look at systemd, which would at least need another few years of time and some lead developers with an actual clue about Unix to be ready for prime time, I highly doubt that Red Hat is a good choice for anything.
Because, you know, that would mean systemd protecting users from some bug that is not theirs. They don't do that and screw the user.
What do you mean "lock down the hardware"? I expect my hardware to be locked down and not change its settings unless I go to the BIOS set-up. My take is that systemd is very much at fault here for not protecting the user.
That is the way the systemd malware gets developed. Bugs are always other peoples faults and protecting users is not an option.
Possibly. I would love to see them sued into the ground for this. Would make others think twice before trying these practices.
My guess is that they have cash-flow problems and they now think pissing-off potential customers is the way to go. You know, like the music and movie industries.
On the side of solid engineering practices, they can refuse to talk to that counterfeit device by not detecting it or giving an out-of-band error on detection, but that is it. Breaking the hardware intentionally is sabotage and exceptionally unethical. Being willing to work with the device but then injecting data into the data-stream intentionally is the same. If anything bad happens as a result, this is the step that comes after gross negligence: It is called "intent".
While I do expect they will have had this cleared from a legal perspective and will be hard or impossible to attack, from an engineering perspective there is only one valid way to deal with this: To not ever use their products until they have credible sworn off their evil ways.
If you kook at the individual parts, you also get all the ways they can form relationships.
Indeed.
Well, possibly. Most people also do not have actual intelligence at their disposal, but just copy what others do without understanding what they are doing.
Your argument is insofar a good one as machines may be able to get to that level eventually. That is why I usually write "smart human beings" when illustrating to what level machines will likely never get. But if you are right, machines will, for example, never write meaningful code, do translation of more than superficial conversations, do meaningful research, etc., just as most human beings.
I do disagree on the proportions though: About 10-20% of humans are able to do more than the others can. They are capable of independent thought and insight. They are capable of identifying bullshit when presented with it and they can understand things from scratch that they never have heard about before. Incidentally, there seems to be only a very lose connection to intelligence. These people are hot candidates for actually having consciousness and that could be exactly why they are capable of more.
You obviously have no clue about modern physics.
The "sum" part describes the configuration. In physics, the function of the whole is fully determined by the function of all its parts and the microscopic relations between them. There are no unexpected or "magic" properties that suddenly manifest. In physics there are no "emergent properties" at all.
My brain could be considered a biological machine and it implements human-equivalent intelligence (or at least the people around me seem to think so, perhaps I have them all fooled).
Only if you assume physicalism is right. If you do that, you can of course demonstrate (by a few intermediate step) that physicalism is right. But that reasoning is circular and meaningless. Most physicalists fall for it though, but this is not science. It is more in line with religion.
Your position is certainly not a scientific one and you are rather severely misinformed about what software can do (most people are).
Can you point out a specific problem with any part of what I've stated or demonstrate with sufficient proof that it won't work or cannot be done, not just now, but also at no point in the future?
You have not pointed out anything with sufficient proof. You are arguing for a restricted model with insufficient proof. That is my whole point.
You have no idea how the real world works. The only people that will decline for that reason are those with huge egos and typically rather mediocre skills. These are the people that like to be described as "rock-stars". Hint: None of them are.
However, even with an NDA you can get a lot of interesting insights when doing such a job. You can just not use them directly or use them to compete with the business where you had them. But that is the deal any employer with interesting technology has to make. As soon as people are working on it, they are going to learn and improve their skills and insights. There is no way around that.