I can tell that, from personal experience, that a big hunk of programmers out there bailed on math ASAP in school. Many of them can't even put an upper bound on many algorithms.
I worked for one company where one of the senior people was asking candidates to give the big-O running time of reading a character from a file and putting it in a buffer repeatedly until EOF. When the buffer is full, you double the size of the buffer.
The guy *ASKING* the question had the wrong answer (which I'll leave as an exercise for the reader). Leaving the fact that this guy was being a tool aside, he was trying to find people who could work through the problem. Most of the candidates couldn't even get started, let alone make a coherent argument that he was incorrect.
As to why the skills are rare, I don't think that it's all that rare it's just that the HR filtering does not increase the likelyhood that a particular screened candidate is a fit. So, there's a lot of chaff for the hiring manager to sort through.
It's why most of the people I know hire past coworkers first then from trusted headhunters then from the pool of HR resumes last.
I think that it's a system Vs. application thing. We have problem finding good Java folks who are comfortable writing complex multi-threaded software.
On the other hand, I think that doing business applications is easier *IF* you just shuffle code. However (and maybe this is what the article is talking about) if you have to really understand the business applications can actually be harder.
Let's reword it this way. Paul send all his correspondance through a courier service. The courier service charges the recipient of any message 5 cents. I notice that 9 out a 10 times I accept a message from the courier, it's of no value to me. 1 out of 200 times, it's a mailbomb. So, I decide that I'm not going to accept messages from Paul's courier service anymore. Seems okay to me.
Now, let's assume that I'm a Mailboxes, Etc. Is it okay for me to refuse delivery from Paul's courier service for my clients? Well, I happen to think so, after all there's a chance there's a mailbomb in there. Also, the 5 cent charge is a problem because Paul's courier service wants to deliver 300,000 messages to my clients a day.
The IP that they are trying to protect is 1) largely in the driver software and 2) any driver that doesn't use the technique is likely to suck. So, either these companies give up trade secret protection (not likely), they provide closed source drivers (nvidia), or the flip OSS the bird.
It's not like this is some great mystery it's just people trying to what they *THINK* is best for themselves.
You might argue that they're wrong about the conclusion but it's no great mystery as to how they arrived at the conclusion.
Although the above-described virtual server setup (virtual server via NAT) can meet the performance requirements of many servers, the design is limited by the load balancer, which is a single point of failure for the whole cluster. However, you can eliminate this bottleneck by having multiple virtual servers, each connected to its own cluster of real servers, grouped together at a single domain name by round robin DNS.
Do two machine with failover before you do round-robin DNS.
Okay, so it's called Monad -- I've only heard this name in one other context, and that's Haskell (http://www.haskell.org./ The interesting thing is that Simon Peyton Jones went to MSR a few years ago.
So, it seems that, either the name is unrealated, and that would suck. Or, that somehow, this is related. *IF* it is realated, I'm not sure how adoption will go. Functional Programming can be a little odd.
As a single comlainant, you are unlikely to find it financially worthwhile to sue Apple over misleading advertising.
So, these lawyer provide a service in that the provide a strong incentive for companies to be truthful in their dealings with customers.
So, in absence of class action lawsuits, there's a zone in which companies can mislead consumers without fear of loss. With class action this gap is closed.
If you are opposed to class action lawsuits, you should be opposed to venture capitalists as well.
Lawyers who take these cases fund the cases, including all time (which would otherwise be billable) and the cost of expert witnesses, transportation for witnesses, discovery costs, everything, which in a case like this can be a pretty penny -- $1M isn't an unreasonable guess.
So, these lawyers get $1.78M in profit. Well, if you consider that they're going to win some and loose some, then getting the occasionaly $1.78M payday is no worse than VCs getting a 10X payday on IPO.
There's a movie out call "A Civil Action" (yeah it was a book first) that shows how class action suits can bankrupt firms (and individuals).
I'm not saying that this isn't a good payday, I'm just saying it's a high-risk business that has correspondingly high rewards.
Like any such business the key is to reduce idiosyncratic risk by diversifying. That's how VCs get and stay rich.
My sources at MS have always said that C is still the dominant language over there but I must admit my contacts into the Office group aren't very strong at this point.
I suggest a little civil disobedience. Whenever asked for our ID we should repeat the phrase "Leeloo Dallas, multi-pass." in response to any further questions.
I already have a girl, so apparently I need to have 1.4 boys. The whole boy is gonna be fine but what am I gonna do with 40% of a boy? I mean, aside from encouraging him to be a high school social studies teacher.
Well, I've been thinking about this a lot and the one question I keep coming back to is whether the quality problem in outsourcing is actually the result of shoddy work or is it a function of the fundamental complexity of software development process. I suspect that it's not shoddy work.
This idea has its root in Brook's "There is no silver bullet." (Which I admit that I need to reread)
However, *IF*, the poor quality is not the result of shoddy coding but a result of the difficulties in running an outsourced project then those difficulties effectively work as a tarrif on imported labor hours. This would suggest that the big surge in outsourcing is the result of "transportation" costs essentially going to zero. (Imagine what would happen to some physical industries if transportation costs dropped to zero in the course of 2-3 years!)
Anyway, I guess the question I'm currently wrestling with is the problem with outsourcing a fundamental complexity or and accidental one. I'm beginning to think that it's a fundamental one.
So, what about outsource companies developing and marketing competing products directly? Well, at this point, I suspect that's a bigger threat to american programmers than outsourcing. I do eventually think it will happen in fairly large scale.
Slashdot Mirror
I can tell that, from personal experience, that a big hunk of programmers out there bailed on math ASAP in school. Many of them can't even put an upper bound on many algorithms.
I worked for one company where one of the senior people was asking candidates to give the big-O running time of reading a character from a file and putting it in a buffer repeatedly until EOF. When the buffer is full, you double the size of the buffer.
The guy *ASKING* the question had the wrong answer (which I'll leave as an exercise for the reader). Leaving the fact that this guy was being a tool aside, he was trying to find people who could work through the problem. Most of the candidates couldn't even get started, let alone make a coherent argument that he was incorrect.
As to why the skills are rare, I don't think that it's all that rare it's just that the HR filtering does not increase the likelyhood that a particular screened candidate is a fit. So, there's a lot of chaff for the hiring manager to sort through.
It's why most of the people I know hire past coworkers first then from trusted headhunters then from the pool of HR resumes last.
I think that it's a system Vs. application thing. We have problem finding good Java folks who are comfortable writing complex multi-threaded software.
On the other hand, I think that doing business applications is easier *IF* you just shuffle code. However (and maybe this is what the article is talking about) if you have to really understand the business applications can actually be harder.
Just to reinforce what you're saying. I've had an offer this last month along with 2 others that wanted to interview. I declined them all.
The company I work for is seeing demand for its products pick up.
The recruiters I know are busy.
I have a friend who can't fill C++ positions at his company.
It's my impression that VC money is starting to trickle back into the market.
Let's reword it this way. Paul send all his correspondance through a courier service. The courier service charges the recipient of any message 5 cents. I notice that 9 out a 10 times I accept a message from the courier, it's of no value to me. 1 out of 200 times, it's a mailbomb. So, I decide that I'm not going to accept messages from Paul's courier service anymore. Seems okay to me.
Now, let's assume that I'm a Mailboxes, Etc. Is it okay for me to refuse delivery from Paul's courier service for my clients? Well, I happen to think so, after all there's a chance there's a mailbomb in there. Also, the 5 cent charge is a problem because Paul's courier service wants to deliver 300,000 messages to my clients a day.
The IP that they are trying to protect is 1) largely in the driver software and 2) any driver that doesn't use the technique is likely to suck. So, either these companies give up trade secret protection (not likely), they provide closed source drivers (nvidia), or the flip OSS the bird.
It's not like this is some great mystery it's just people trying to what they *THINK* is best for themselves.
You might argue that they're wrong about the conclusion but it's no great mystery as to how they arrived at the conclusion.
There was an officially declared recession for the first 3 quarters of 2001.
Here for more.
Not only am I an F5 customer and have many friends who work there but I once worked with a company that occupied space formerly occupied by F5.
Get this, the system that controlled card-key access was a *VERY* early F5 box that they had installed Windows on.
Yeah, what he said. However, the TFA said:
Although the above-described virtual server setup (virtual server via NAT) can meet the performance requirements of many servers, the design is limited by the load balancer, which is a single point of failure for the whole cluster. However, you can eliminate this bottleneck by having multiple virtual servers, each connected to its own cluster of real servers, grouped together at a single domain name by round robin DNS.
Do two machine with failover before you do round-robin DNS.
Many F5 appliances were not much more than FreeBSD PCs
Okay, so it's called Monad -- I've only heard this name in one other context, and that's Haskell (http://www.haskell.org./ The interesting thing is that Simon Peyton Jones went to MSR a few years ago.
So, it seems that, either the name is unrealated, and that would suck. Or, that somehow, this is related. *IF* it is realated, I'm not sure how adoption will go. Functional Programming can be a little odd.
Anyone know?
And now I have to write some text here because some stupid 'tard decided to crapflood slashdot in the dim dark past.
As a single comlainant, you are unlikely to find it financially worthwhile to sue Apple over misleading advertising.
So, these lawyer provide a service in that the provide a strong incentive for companies to be truthful in their dealings with customers.
So, in absence of class action lawsuits, there's a zone in which companies can mislead consumers without fear of loss. With class action this gap is closed.
You have a point there. I was trying to speak to the idea of excessive compensation. I'm addled by impending math final -- sorry.
If you are opposed to class action lawsuits, you should be opposed to venture capitalists as well.
Lawyers who take these cases fund the cases, including all time (which would otherwise be billable) and the cost of expert witnesses, transportation for witnesses, discovery costs, everything, which in a case like this can be a pretty penny -- $1M isn't an unreasonable guess.
So, these lawyers get $1.78M in profit. Well, if you consider that they're going to win some and loose some, then getting the occasionaly $1.78M payday is no worse than VCs getting a 10X payday on IPO.
There's a movie out call "A Civil Action" (yeah it was a book first) that shows how class action suits can bankrupt firms (and individuals).
I'm not saying that this isn't a good payday, I'm just saying it's a high-risk business that has correspondingly high rewards.
Like any such business the key is to reduce idiosyncratic risk by diversifying. That's how VCs get and stay rich.
http://expat.sourceforge.net/
My sources at MS have always said that C is still the dominant language over there but I must admit my contacts into the Office group aren't very strong at this point.
I guess they were so used to English only being the native tongue for half the residents they just went with it.
Either that or they've had enough geektards like you (and me) come in and do it that it has become teh unfunny.
C++ will continue to be the 'crown jewels' of microsoft.
You misspelled C.
I suggest a little civil disobedience. Whenever asked for our ID we should repeat the phrase "Leeloo Dallas, multi-pass." in response to any further questions.
Aaaaaahhhhhhhhhhhhhhh!
Wow. Sorry you didn't take mine as a joke as well. That's what I was trying to convey the the little smiley at the end of my post.
What, you got real you in storage someplace? Maybe, one for everyday and one for Sunday best? I don't understand. ;-)
I already have a girl, so apparently I need to have 1.4 boys. The whole boy is gonna be fine but what am I gonna do with 40% of a boy? I mean, aside from encouraging him to be a high school social studies teacher.
Yep, I think he earned a good, long look. Someday, I hope to earn my peek.
Well, I've been thinking about this a lot and the one question I keep coming back to is whether the quality problem in outsourcing is actually the result of shoddy work or is it a function of the fundamental complexity of software development process. I suspect that it's not shoddy work.
This idea has its root in Brook's "There is no silver bullet." (Which I admit that I need to reread)
However, *IF*, the poor quality is not the result of shoddy coding but a result of the difficulties in running an outsourced project then those difficulties effectively work as a tarrif on imported labor hours. This would suggest that the big surge in outsourcing is the result of "transportation" costs essentially going to zero. (Imagine what would happen to some physical industries if transportation costs dropped to zero in the course of 2-3 years!)
Anyway, I guess the question I'm currently wrestling with is the problem with outsourcing a fundamental complexity or and accidental one. I'm beginning to think that it's a fundamental one.
So, what about outsource companies developing and marketing competing products directly? Well, at this point, I suspect that's a bigger threat to american programmers than outsourcing. I do eventually think it will happen in fairly large scale.