Because we were always told that losing the low-end jobs was no big deal because we'd create more high-end jobs with increased trade. Americans would move up, not out. Turns out that's a big lie, but they keep trying to sell it to us: "Go get more education!" It's like our politicians and trade negotiators never knew there were educated people in other countries too.
However let's consider the other options we might have instead of outsourcing? IBM moving all their employees out of America? Why have a company in America when you can get cheaper work in India?
Because there are advantages to hiring developers from the same culture as the intended customer market. Because intellectual property laws in third world countries are a joke. Because for some work, particularly defense work, it is actually illegal to use foreign developers. Because even though managers are happy hiring workers from overseas, those same managers don't want to move to the third world themselves, and there are certain advantages to having your development staff close to management and marketing instead of half a world away.
Most of the work that could go overseas is already overseas because it is so much cheaper. The jobs that are now left in the US is work that is harder or impossible to ship overseas. We can decide to fill those positions with US workers or foreign workers. If we decide to fill the jobs with foreign workers, then we are training our future foreign competition while telling US college students not to enter CS or IT. If we decide to fill the jobs with US workers, then we are going to keep high-paying jobs here in the US while telling US college students that they will have a bright future in either CS or IT.
Our point is, changing from one corporation to another is always possible (and usually quite easy). Changing the government, however -- and I don't mean electing a different President or lawmaker, but revamping the government bureaucracy -- is quite impossible...
Oh bullshit. First off, you're not even talking apples to apples. You certainly can change from one government to another - leave the country. It is quite easy. Changing a single corporation is just as hard or harder than changing the government. With government, you have countless rights to information and to petition, while the government has limitations on their power. Corporations are increasingly accountable to no one.
Fact is, the government still has the obligation to manage the data, they are just lazy and are putting their jobs off on the contractor.
The contractor has taken over the responsibility for the entire lab. The contractor was there to take over the lab's functions. This was one of the functions, which they're refusing to do. Yes, the government should step in and get the contractor under compliance, but it is the contractor's fault for being out of compliance in the first place.
At which point I have to say, if a corporate bureaucracy is going to be worse than a government bureaucracy, we shouldn't be bothering with privatization.
You have a public, government lab that was turned over to a private, corporate entity. That is the definition of privatization.
Libertarians like to think that eliminating government agencies and turning them all over to private corporations will magically make those businesses more efficient and cost less money. That has often been shown not to be true, but Libertarians still believe it on faith.
I'm sorry, but the Libertarian efficiency dogma is not always true. And in some cases like this, it has the side effect of making the agencies worse.
It's not entirely clear whether this is the contractor doing this on its own initiative, or, more likely, the contractor legitimately concerned about being accused by the government of giving someone improper access. So, LANS seems to be playing it safe by directing everyone to FOIA (Freedom of Information Act) which is of course all but useless for students and many archival researchers.
The researchers had access before the privatization of the lab. This whole article is the result of the privatization stopping the flow of information, which had normally been filled without FOIA requests up until this time. Presumably, the government would not care since they previously released this information when they ran this lab. This would not be a big deal otherwise.
A very unfortunate state of affairs, but I'm not certain privatization is exactly what's to blame.
Interesting that people will not hesitate to call government bureaucracy for what it is. But, when the same thing happens in the private sector, people excuse the behavior.
This is corporate bureaucracy.
Libertarians like to point out all the positives of privatization without going into the negatives. Here is one glowing example. The bureaucracy is now worse under corporate operation because the private businesses don't have to follow the same laws as the government. More than likely, we're also paying more money as well. At least government limits the amount that executives are compensated.
When I got married in 2004, I found a professional photographer that covered my wedding and provided the negatives for only $2000. Normally whenever you say "wedding" to any professional, they multiply their estimate by ten times. Certainly, we could have paid more for the photos and negatives, and probably would have with any other photographer. However, $2000 for eight hours of easy work is great by any standard.
If you gave $1b to the right people, then they could come up with a working product. However, the problem is finding and selecting the right people to privatize the project to. The fact is that privatization does not necessarily save the government money. Business promises are made which are meant to be broken. Business contracts are signed and then ignored. Cost overruns are considered normal. It's ironic that we don't trust the government to do the job itself, but we do trust it enough to do the job of finding the contractor.
Take a look around buddy. You live in a house full of American inventions built by American companies that are still designing and improving those inventions. Sure, we may actually build those devices in China, but that's only because China is currently one of the cheapest places to build. It takes no great ingenuity to slap together parts someone else designed. Soon as China gets too expense, we'll move to the next country and China will be another also-ran.
The only advances the Chinese are making are all from what their spies are currently stealing from the US. That doesn't take any great ingenuity, which the people who came up with those ideas still have, who live in the US. China isn't coming up with anything new itself.
The worlds oldest civilization has to catch up with the good ol' USA. I think if the USA had anything good for China to copy, they would have embraced it by now. They've had thousands of years of experience in political struggle, power changes and population explosions to make the USA look like a newborn.
Right. That's why China has been leading the world in science and technology and human rights for the last 60 years while the US was an ignorant backwater that brutalized its citizens. Oh wait... I got that backwards.
China may be older, but they're certainly not better.
It depends on how the knowledge is acquired. Learning engineering, creating companies, and out-competing like the Japanese did is one thing. Planting spies across the world, stealing trade and military secrets, and strong-arming foreign companies and countries like the Chinese do is another thing.
More than that, Toyota copied Ford, took Deming's alterations, and ran with that. Toyota had a thirty year head start by the time Ford got its head out of its ass. Even then, Ford didn't get behind Deming like Toyota did. Hard to make those changes when you're already enormous, rather than growing your company that way in the first place.
The poor throughout history are working enormous numbers of hours to barely pay their bills. The poor need their jobs, so they focus on those jobs and keeping their bosses happy, who generally tend to be the wealthy they would be agitating against. Further, poor children are generally put to work early, don't get educated, and don't value education, which would give them the tools to see the problems around them. In short, the poor don't have time nor the inclination nor the ability to start revolutions.
Only when people reach the middle class do they have the free time to sit back, look around, and reflect upon the conditions around them. They have time to read books and get educated on the world around them. They have time to see the misery in peoples' lives and can imagine a better world. They can look at their poor friends and family and empathize with their plight since they either came from that group or could easily be part of that group.
It's much harder for the middle class to empathize with the rich since the disparity tends to be enormous. Further, the rich often flaunt their money and insult the people around them, making them despicable in the eyes of just about everyone not in that social group.
Yes, money does allow freedom to flourish. And that money needs to be spread around to a large group of people in the middle, not focused in a small group at the top.
They started with making inferior copies cheaply, figured out how to improve the quality without substantially increasing the cost
Actually, the Japanese also copied their quality improvement program from an American, W. Edwards Deming. We handed them everything they're currently using to put us out of business.
Japan started upon Deming's quality improvement path in 1950. Ford Motor, in contrast, didn't start until 1981. Those facts alone can explain the last 40 years of automotive history.
India didn't invent this car. Read to the end of the article and it says, "MDI is a small, family-controlled company located at Carros, near Nice (Southern France) where Guy and Cyril Negre and their technical team have developed the engine technology and the technologically advanced car it powers." That's right, this car was developed in France.
50 minutes? What makes you think you have 50 minutes? Try 10. You've got an hour interview and this problem is the first in the list they'll ask you.
And who said it was a linked list? Pick from any of the basic data types.
Good luck trying to figure out how to implement a hash map considering it's been 10 years since the last time you implemented one by hand in college.
Further, once you get done, they are going to bitch about syntax and any specific implementation problems - despite the fact that they originally said otherwise. Further, they'll completely switch the requirements on you and then bitch that your code doesn't meet those either. Then, you'll be expected to fix those issues. At that point, they'll bitch about something else in your code and claim it doesn't address those issues either.
Oh, and you better keep your cool about doing all this on the whiteboard, the physical IDE equivalent of Notepad. You don't want to show your growing anger or nervousness, that could cost you the job you need from being out of work.
Yeah, I did a Microsoft contract. What a joke. They bust your ass on the interview and then give you idiot work when you get in. I was amazed that most of the full time employees in the test department were there for more than five years. God, I would shoot myself.
MS employees are not smarter than everyone else. And no, their interview process does not pick the best and the brightest.
All the questions asked... test one of the following a) critical thinking b) creative thinking c) solid CS foundations. If you can't manage one or more of the above, why would you consider yourself qualified?
No, they only test the ability to think on your feet in a high stress situation that is the exact opposite of the way code is actually written in reality.
I don't understand why some people think they can go be ``coding monkeys'' without care or thought... You mean you can't tell the guy how to implement a list? Or how strcpy works? Or how to implement a tree? Wtf? If you fail to show your knowledge through these basic things, why should the interviewer trust your skill set? Because you said so on your resume? Give me a break...
Yeah, I can tell you how to do any of that shit. The problem comes when you ask for details and want me to show you specifics. Telling someone how to implement a list or a tree is different from writing it on a whiteboard, from the top of your head, without any hints, in less than 10 minutes, with your future job on the line, getting the syntax straight, and not making any major mistakes. As I mentioned, that is the exact opposite of the way code is actually written in reality. And it doesn't tell you anything about the candidate other than their ability to think fast. The ability to think fast does not necessarily indicate a good developer.
How many software engineers actually implement any of that shit on a regular basis? Not many. If you do, you should be fired for wasting time and resources. I haven't implemented a list in years since I left college and became a productive developer, so excuse me if I think your reasoning is lousy. My skill set is not untrustworthy if I can't think fast, remember my college days, and write a complete list implementation without errors under those bullshit conditions.
Me: But, all the MS teams do is ask bullshit interview questions. Why are manhole covers round? How do I implement a list?... Those questions don't find good candidates.
You: They eliminate bad ones, though.
No. They eliminate people who can't or don't think fast at answering bullshit questions. The ability to think fast does not necessarily indicate a good developer.
The point is to test your ability to think on your feet.
No, the point is to get a good candidate for the job. The ability to think fast does not necessarily indicate a good developer.
If you can code it off the cuff in the allowed time, you've exhibited a basic level of competence.
And you've just pissed off the 20 year veteran asking him to prove that he knows what he's been doing for the last 20 years. Even if you ask a harder question, it's still insulting and tells you nothing other than his ability to think fast.
The resume is enough to exhibit a basic level of competence unless you're hiring straight out of college. You want to know whether they really worked at their previous company, at that position, doing what they wrote they did. A background check will catch those first two items. Less insulting and more useful questions can determine the third.
And BTW, I've heard your argument a hundred times.
This is not about _remembering_, it's about deriving. If someone knows the question off the top of their head, we try something different. If someone cannot derive an implementation of a string function, they're not an interesting candidate.
Oh bullshit. Good development and good code isn't about reinventing the wheel when good algorithms and code already exists. It's about recombining what you know and what you have to produce new functionality that works in the minimal amount of time. Developers don't just go out and create all new code without first looking at what's already there.
No wonder MS is the way it is.
"Ripping apart" answers isn't something we do.
So, you call it "follow-on" while I call it ripping apart - same thing.
Rarely does someone issue a perfect answer on their first try
Certainly in interviews, yes. However, it certainly is possible to come up with good code if you have time to think about the solution and look at previous solutions.
For almost any answer someone gives, there is some possible drawback or "gocha". What is the memory consumption of your routine? How many conditional branch statements would it require? Asking these follow-on questions are what makes it a less-worthless question, and seeing how someone thinks about the implications of their decisions and describes the tradeoffs is what makes it worthwhile.
And that is what I was getting to. Yes, that "gotcha" is the "ripping apart" of someone's logic. You ask them to solve one problem in a completely bullshit manner in a bullshit amount of time under intense pressure from not having a job and needing one. And then, you twist things around and say that the solution doesn't address this other problem which was never mentioned previously.
And no, asking them to solve that second question based on the first doesn't make the first question more meaningful. You're just piling on the shit, requiring more bullshit from the candidate on top of the bullshit they've already produced. And then, there are other gotchas that you pull after, further ripping apart the solution, and further requiring more bullshit. You hold back on the actual requirements of the bullshit problem. You're asking the candidate to do the role of developer, project manager, and business analyst in the middle of an interview when you only first asked for a "simple C function". You make something appear deceptively simple and then ream them for not writing a bulletproof answer the first time.
That's a fine response to have, but i'd ask you to justify it. Why is it a stupid question? Obviously, i'd ask it as an allegorical question to the problem of how to test software.
Well let's see, because "allegorical" or not, it actually has nothing to do with software. If you want to know about software testing, then ask about software testing. My dad could answer this toaster question and he can barely turn on a computer. He could bullshit his way through this answer and actually convince you he'd make a great tester if he wanted.
Although i'm not sure about "liking them".
Wow. That's another huge problem with MS. I can't believe they would actually hire people that don't get along with each other. Ballmer's tantrums seem less and less surprising now.
I find that the opposite is true -- people that are unwilling to delve into the details of an answer.. people that keep things "high level" are bullshit artists. The saying "The devil is in the details" is a saying for a _reason_.
The "details" you can infer from their resume. If they were with a company for any significant amount of time, especially during the last few years during the tech downturn, then you can bet they know their shit. You want to know whether they really worked at that company, at that position, doing what they wrote they did. Your background check will catch those first two items.
Nurses get licensed, and at that salary range, he should be chock full of licenses and highly specialized. IT folks aren't licensed.
Uh, yes they are. The difference is that licenses aren't required in IT fields. Someone with his qualifications in IT would also be chock full of licenses and be highly specialized. But, there aren't many non-managerial people making $240,000 or even $100,000 in IT.
There's another aspect to nursing: the hours.
What about it? The hours in IT shops are just as bad or worse.
Me: However, most CS grads are not good with communications skills, so they couldn't convince you of their skills. Wow, big surprise.
You: They should be able to talk tech. If they can't even communicate on purely technical issues then what good are they when they need to parse complex requirements, or, gasp, track down a requirements issue with a business user? Communications skills are part of the package when looking for people who can drive projects. No skills? No job.
Should they be able to sell you on their skills?... No. The people with that type of communications skills are salesmen. You shouldn't expect an IT job candidate to convince you that they are the one for the job. And you shouldn't think of everyone else as shit because they can't.
Yeah, IBM was stupid in giving away a good chunk of their business to Bill Gates after Gates ripped off CP/M from Gary Killdall.
Even if there isn't a shortage of 100,000 IT workers, if we import more, all of the sudden you have more computer savvy people in the marketplace buying Microsoft products.
Not necessarily. Supply and demand works both ways. If you import a bunch of workers in one field, then a lot of workers in that field will leave. That's exactly what we've been seeing in CS over the last few years.
And further, the more technically savvy people are, the less likely they are to buy MS products because they know they can get better products for free from open source projects.
Also, more cheap newbie workers trained in the latest stuff keep costs down and provides a powerful incentive for companies to keep upgrading.
Yes, but keeping salaries down also means that students have no incentive to learning CS and entering IT.
Slashdot Mirror
Because we were always told that losing the low-end jobs was no big deal because we'd create more high-end jobs with increased trade. Americans would move up, not out. Turns out that's a big lie, but they keep trying to sell it to us: "Go get more education!" It's like our politicians and trade negotiators never knew there were educated people in other countries too.
However let's consider the other options we might have instead of outsourcing? IBM moving all their employees out of America? Why have a company in America when you can get cheaper work in India?
Because there are advantages to hiring developers from the same culture as the intended customer market. Because intellectual property laws in third world countries are a joke. Because for some work, particularly defense work, it is actually illegal to use foreign developers. Because even though managers are happy hiring workers from overseas, those same managers don't want to move to the third world themselves, and there are certain advantages to having your development staff close to management and marketing instead of half a world away.
Most of the work that could go overseas is already overseas because it is so much cheaper. The jobs that are now left in the US is work that is harder or impossible to ship overseas. We can decide to fill those positions with US workers or foreign workers. If we decide to fill the jobs with foreign workers, then we are training our future foreign competition while telling US college students not to enter CS or IT. If we decide to fill the jobs with US workers, then we are going to keep high-paying jobs here in the US while telling US college students that they will have a bright future in either CS or IT.
Our point is, changing from one corporation to another is always possible (and usually quite easy). Changing the government, however -- and I don't mean electing a different President or lawmaker, but revamping the government bureaucracy -- is quite impossible...
Oh bullshit. First off, you're not even talking apples to apples. You certainly can change from one government to another - leave the country. It is quite easy. Changing a single corporation is just as hard or harder than changing the government. With government, you have countless rights to information and to petition, while the government has limitations on their power. Corporations are increasingly accountable to no one.
Fact is, the government still has the obligation to manage the data, they are just lazy and are putting their jobs off on the contractor.
The contractor has taken over the responsibility for the entire lab. The contractor was there to take over the lab's functions. This was one of the functions, which they're refusing to do. Yes, the government should step in and get the contractor under compliance, but it is the contractor's fault for being out of compliance in the first place.
At which point I have to say, if a corporate bureaucracy is going to be worse than a government bureaucracy, we shouldn't be bothering with privatization.
You have a public, government lab that was turned over to a private, corporate entity. That is the definition of privatization.
Libertarians like to think that eliminating government agencies and turning them all over to private corporations will magically make those businesses more efficient and cost less money. That has often been shown not to be true, but Libertarians still believe it on faith.
I'm sorry, but the Libertarian efficiency dogma is not always true. And in some cases like this, it has the side effect of making the agencies worse.
Or just save more money and not contract out at all.
It's not entirely clear whether this is the contractor doing this on its own initiative, or, more likely, the contractor legitimately concerned about being accused by the government of giving someone improper access. So, LANS seems to be playing it safe by directing everyone to FOIA (Freedom of Information Act) which is of course all but useless for students and many archival researchers.
The researchers had access before the privatization of the lab. This whole article is the result of the privatization stopping the flow of information, which had normally been filled without FOIA requests up until this time. Presumably, the government would not care since they previously released this information when they ran this lab. This would not be a big deal otherwise.
A very unfortunate state of affairs, but I'm not certain privatization is exactly what's to blame.
Interesting that people will not hesitate to call government bureaucracy for what it is. But, when the same thing happens in the private sector, people excuse the behavior.
This is corporate bureaucracy.
Libertarians like to point out all the positives of privatization without going into the negatives. Here is one glowing example. The bureaucracy is now worse under corporate operation because the private businesses don't have to follow the same laws as the government. More than likely, we're also paying more money as well. At least government limits the amount that executives are compensated.
When I got married in 2004, I found a professional photographer that covered my wedding and provided the negatives for only $2000. Normally whenever you say "wedding" to any professional, they multiply their estimate by ten times. Certainly, we could have paid more for the photos and negatives, and probably would have with any other photographer. However, $2000 for eight hours of easy work is great by any standard.
If you gave $1b to the right people, then they could come up with a working product. However, the problem is finding and selecting the right people to privatize the project to. The fact is that privatization does not necessarily save the government money. Business promises are made which are meant to be broken. Business contracts are signed and then ignored. Cost overruns are considered normal. It's ironic that we don't trust the government to do the job itself, but we do trust it enough to do the job of finding the contractor.
You have to go back more than a hundred years to find a China that was comparable to anything in the west.
Take a look around buddy. You live in a house full of American inventions built by American companies that are still designing and improving those inventions. Sure, we may actually build those devices in China, but that's only because China is currently one of the cheapest places to build. It takes no great ingenuity to slap together parts someone else designed. Soon as China gets too expense, we'll move to the next country and China will be another also-ran.
The only advances the Chinese are making are all from what their spies are currently stealing from the US. That doesn't take any great ingenuity, which the people who came up with those ideas still have, who live in the US. China isn't coming up with anything new itself.
The worlds oldest civilization has to catch up with the good ol' USA. I think if the USA had anything good for China to copy, they would have embraced it by now. They've had thousands of years of experience in political struggle, power changes and population explosions to make the USA look like a newborn.
Right. That's why China has been leading the world in science and technology and human rights for the last 60 years while the US was an ignorant backwater that brutalized its citizens. Oh wait... I got that backwards. China may be older, but they're certainly not better.
Entitlement? Yes, I suppose you can use that to justify theft. The rich deserve to have their trade and military secrets plundered.
It depends on how the knowledge is acquired. Learning engineering, creating companies, and out-competing like the Japanese did is one thing. Planting spies across the world, stealing trade and military secrets, and strong-arming foreign companies and countries like the Chinese do is another thing.
More than that, Toyota copied Ford, took Deming's alterations, and ran with that. Toyota had a thirty year head start by the time Ford got its head out of its ass. Even then, Ford didn't get behind Deming like Toyota did. Hard to make those changes when you're already enormous, rather than growing your company that way in the first place.
The poor throughout history are working enormous numbers of hours to barely pay their bills. The poor need their jobs, so they focus on those jobs and keeping their bosses happy, who generally tend to be the wealthy they would be agitating against. Further, poor children are generally put to work early, don't get educated, and don't value education, which would give them the tools to see the problems around them. In short, the poor don't have time nor the inclination nor the ability to start revolutions.
Only when people reach the middle class do they have the free time to sit back, look around, and reflect upon the conditions around them. They have time to read books and get educated on the world around them. They have time to see the misery in peoples' lives and can imagine a better world. They can look at their poor friends and family and empathize with their plight since they either came from that group or could easily be part of that group.
It's much harder for the middle class to empathize with the rich since the disparity tends to be enormous. Further, the rich often flaunt their money and insult the people around them, making them despicable in the eyes of just about everyone not in that social group.
Yes, money does allow freedom to flourish. And that money needs to be spread around to a large group of people in the middle, not focused in a small group at the top.
They started with making inferior copies cheaply, figured out how to improve the quality without substantially increasing the cost
Actually, the Japanese also copied their quality improvement program from an American, W. Edwards Deming. We handed them everything they're currently using to put us out of business.
Japan started upon Deming's quality improvement path in 1950. Ford Motor, in contrast, didn't start until 1981. Those facts alone can explain the last 40 years of automotive history.
India didn't invent this car. Read to the end of the article and it says, "MDI is a small, family-controlled company located at Carros, near Nice (Southern France) where Guy and Cyril Negre and their technical team have developed the engine technology and the technologically advanced car it powers." That's right, this car was developed in France.
50 minutes? What makes you think you have 50 minutes? Try 10. You've got an hour interview and this problem is the first in the list they'll ask you.
And who said it was a linked list? Pick from any of the basic data types.
Good luck trying to figure out how to implement a hash map considering it's been 10 years since the last time you implemented one by hand in college.
Further, once you get done, they are going to bitch about syntax and any specific implementation problems - despite the fact that they originally said otherwise. Further, they'll completely switch the requirements on you and then bitch that your code doesn't meet those either. Then, you'll be expected to fix those issues. At that point, they'll bitch about something else in your code and claim it doesn't address those issues either.
Oh, and you better keep your cool about doing all this on the whiteboard, the physical IDE equivalent of Notepad. You don't want to show your growing anger or nervousness, that could cost you the job you need from being out of work.
Yeah, I did a Microsoft contract. What a joke. They bust your ass on the interview and then give you idiot work when you get in. I was amazed that most of the full time employees in the test department were there for more than five years. God, I would shoot myself.
MS employees are not smarter than everyone else. And no, their interview process does not pick the best and the brightest.
All the questions asked... test one of the following a) critical thinking b) creative thinking c) solid CS foundations. If you can't manage one or more of the above, why would you consider yourself qualified?
No, they only test the ability to think on your feet in a high stress situation that is the exact opposite of the way code is actually written in reality.
I don't understand why some people think they can go be ``coding monkeys'' without care or thought... You mean you can't tell the guy how to implement a list? Or how strcpy works? Or how to implement a tree? Wtf? If you fail to show your knowledge through these basic things, why should the interviewer trust your skill set? Because you said so on your resume? Give me a break...
Yeah, I can tell you how to do any of that shit. The problem comes when you ask for details and want me to show you specifics. Telling someone how to implement a list or a tree is different from writing it on a whiteboard, from the top of your head, without any hints, in less than 10 minutes, with your future job on the line, getting the syntax straight, and not making any major mistakes. As I mentioned, that is the exact opposite of the way code is actually written in reality. And it doesn't tell you anything about the candidate other than their ability to think fast. The ability to think fast does not necessarily indicate a good developer.
How many software engineers actually implement any of that shit on a regular basis? Not many. If you do, you should be fired for wasting time and resources. I haven't implemented a list in years since I left college and became a productive developer, so excuse me if I think your reasoning is lousy. My skill set is not untrustworthy if I can't think fast, remember my college days, and write a complete list implementation without errors under those bullshit conditions.
Me: But, all the MS teams do is ask bullshit interview questions. Why are manhole covers round? How do I implement a list?... Those questions don't find good candidates.
You: They eliminate bad ones, though.
No. They eliminate people who can't or don't think fast at answering bullshit questions. The ability to think fast does not necessarily indicate a good developer.
The point is to test your ability to think on your feet.
No, the point is to get a good candidate for the job. The ability to think fast does not necessarily indicate a good developer.
If you can code it off the cuff in the allowed time, you've exhibited a basic level of competence.
And you've just pissed off the 20 year veteran asking him to prove that he knows what he's been doing for the last 20 years. Even if you ask a harder question, it's still insulting and tells you nothing other than his ability to think fast.
The resume is enough to exhibit a basic level of competence unless you're hiring straight out of college. You want to know whether they really worked at their previous company, at that position, doing what they wrote they did. A background check will catch those first two items. Less insulting and more useful questions can determine the third.
And BTW, I've heard your argument a hundred times.
Well, aren't you special...
This is not about _remembering_, it's about deriving. If someone knows the question off the top of their head, we try something different. If someone cannot derive an implementation of a string function, they're not an interesting candidate.
Oh bullshit. Good development and good code isn't about reinventing the wheel when good algorithms and code already exists. It's about recombining what you know and what you have to produce new functionality that works in the minimal amount of time. Developers don't just go out and create all new code without first looking at what's already there.
No wonder MS is the way it is.
"Ripping apart" answers isn't something we do.
So, you call it "follow-on" while I call it ripping apart - same thing.
Rarely does someone issue a perfect answer on their first try
Certainly in interviews, yes. However, it certainly is possible to come up with good code if you have time to think about the solution and look at previous solutions.
For almost any answer someone gives, there is some possible drawback or "gocha". What is the memory consumption of your routine? How many conditional branch statements would it require? Asking these follow-on questions are what makes it a less-worthless question, and seeing how someone thinks about the implications of their decisions and describes the tradeoffs is what makes it worthwhile.
And that is what I was getting to. Yes, that "gotcha" is the "ripping apart" of someone's logic. You ask them to solve one problem in a completely bullshit manner in a bullshit amount of time under intense pressure from not having a job and needing one. And then, you twist things around and say that the solution doesn't address this other problem which was never mentioned previously.
And no, asking them to solve that second question based on the first doesn't make the first question more meaningful. You're just piling on the shit, requiring more bullshit from the candidate on top of the bullshit they've already produced. And then, there are other gotchas that you pull after, further ripping apart the solution, and further requiring more bullshit. You hold back on the actual requirements of the bullshit problem. You're asking the candidate to do the role of developer, project manager, and business analyst in the middle of an interview when you only first asked for a "simple C function". You make something appear deceptively simple and then ream them for not writing a bulletproof answer the first time.
That's a fine response to have, but i'd ask you to justify it. Why is it a stupid question? Obviously, i'd ask it as an allegorical question to the problem of how to test software.
Well let's see, because "allegorical" or not, it actually has nothing to do with software. If you want to know about software testing, then ask about software testing. My dad could answer this toaster question and he can barely turn on a computer. He could bullshit his way through this answer and actually convince you he'd make a great tester if he wanted.
Although i'm not sure about "liking them".
Wow. That's another huge problem with MS. I can't believe they would actually hire people that don't get along with each other. Ballmer's tantrums seem less and less surprising now.
I find that the opposite is true -- people that are unwilling to delve into the details of an answer.. people that keep things "high level" are bullshit artists. The saying "The devil is in the details" is a saying for a _reason_.
The "details" you can infer from their resume. If they were with a company for any significant amount of time, especially during the last few years during the tech downturn, then you can bet they know their shit. You want to know whether they really worked at that company, at that position, doing what they wrote they did. Your background check will catch those first two items.
Nurses get licensed, and at that salary range, he should be chock full of licenses and highly specialized. IT folks aren't licensed.
Uh, yes they are. The difference is that licenses aren't required in IT fields. Someone with his qualifications in IT would also be chock full of licenses and be highly specialized. But, there aren't many non-managerial people making $240,000 or even $100,000 in IT.
There's another aspect to nursing: the hours.
What about it? The hours in IT shops are just as bad or worse.
Me: However, most CS grads are not good with communications skills, so they couldn't convince you of their skills. Wow, big surprise.
You: They should be able to talk tech. If they can't even communicate on purely technical issues then what good are they when they need to parse complex requirements, or, gasp, track down a requirements issue with a business user? Communications skills are part of the package when looking for people who can drive projects. No skills? No job.
Should they be able to sell you on their skills?... No. The people with that type of communications skills are salesmen. You shouldn't expect an IT job candidate to convince you that they are the one for the job. And you shouldn't think of everyone else as shit because they can't.
There's a reason why Bill is a multi-billionaire.
Yeah, IBM was stupid in giving away a good chunk of their business to Bill Gates after Gates ripped off CP/M from Gary Killdall.
Even if there isn't a shortage of 100,000 IT workers, if we import more, all of the sudden you have more computer savvy people in the marketplace buying Microsoft products.
Not necessarily. Supply and demand works both ways. If you import a bunch of workers in one field, then a lot of workers in that field will leave. That's exactly what we've been seeing in CS over the last few years.
And further, the more technically savvy people are, the less likely they are to buy MS products because they know they can get better products for free from open source projects.
Also, more cheap newbie workers trained in the latest stuff keep costs down and provides a powerful incentive for companies to keep upgrading.
Yes, but keeping salaries down also means that students have no incentive to learning CS and entering IT.