"News.com didn't tell you that the number of teams competing has grown nearly sevenfold from 1994 through 2005. In other words, for a team to finish at, say, third place, in 1994 would be equivalent to finishing 21st this year. So a hypothetical team that News.com would have lauded in 1994 would now be dismissed as having badly "slipped" in 2005, even though it would be of the same quality."
From this I guess the author means that it's OK to be at the same level they were eight years ago. It doesn't matter that the American teams didn't improve at the same rate at the rest of the world. And in his statistical argument he ignores that although team numbers might have increased so did the number of American teams.
Next comes my absolute favorite argument:
"Long before Olympic athletes from all countries became quasiprofessionals, the Eastern European countries were seeing to it that training for the Games was their athletes' full-time job, giving them a major advantage over other nations' athletes."
OMG, it's not fair, they trained harder! Well hello! Is it cheating to produce programmers who can actually solve problems and write code? What exactly is coursework for if it isn't preparation for the kinds of problems you solve in programming contests? I've done a couple - it's the same thing, you just have to be faster and more accurate, compared to a programming assignment.
"the hidden agenda behind the shrill shortage claims was to push Congress to increase the yearly cap on the H-1B work visa program, which enabled industry to import cut-rate engineers from abroad."
I was a H1-B worker - I made great rates (thanks very much) and so did all the other H1-B's I know. It's convenient for Norm's flawed argument to repeat this myth, propagated by programmers who think they should have had my job because it was their birthright, not because they could have done it better.
"How can American engineers compete with cheap, imported labor?"
Too much time in academia Norm. If you can't do the job right it really doesn't matter how cheap you come. The way to compete is to be the best, there is no other way. Shopping for programmers is not like shopping for socks. Remember, computer-related thingys are digital. At the end of the day it is usually pretty obvious whether they work or do not work. "Almost works" is not good enough for anyone, except perhaps a professor who grades CS101 papers.
When Chinese (or Indian, or anyone else) programmers turn out to cost less AND be better programmers we'll be able to thank guys like Norm, who wanted to deny there was ever a problem.
What's Norm's issue with devoting more to education - is it just that he wants to be able to say "It wasn't MY fault?"
From this I guess the author means that it's OK to be at the same level they were eight years ago. It doesn't matter that the American teams didn't improve at the same rate at the rest of the world. And in his statistical argument he ignores that although team numbers might have increased so did the number of American teams.
While your statistical point is valid, your improvement one is not. He's saying that there's a large number of new entries, not that existing entries got better.
No, always be the best. When you're the best you get the same job done but you get it done quicker
Clearly you're not a programmer.;)
The "best" guy will get it done in the right way. Doing it "right" means doing it with an eye for maintainability and making things simpler to do later. Usually this takes *longer* at the beginning. Often, much, much longer.
When a mediocre programmer gets something "done" and their manager or peer reviewer picks it apart and they have to keep doing it until they get it right (or involved someone who can help get it right), that's taking longer.
90% of the programming jobs out there don't have managers that can understand or indeed read code, much less actual peer review.
In a perfect world, maybe, but the world is not perfect. Most code out there is not peer reviewed and barely tested.
Too much time in academia Norm. If you can't do the job right it really doesn't matter how cheap you come. The way to compete is to be the best, there is no other way. Shopping for programmers is not like shopping for socks. Remember, computer-related thingys are digital. At the end of the day it is usually pretty obvious whether they work or do not work. "Almost works" is not good enough for anyone, except perhaps a professor who grades CS101 papers.
Your experience may be atypical (I say to be polite), be
OMG, it's not fair, they trained harder! Well hello! Is it cheating to produce programmers who can actually solve problems and write code?
He doesn't say it isn't fair. He says it is not fair to take the results of the contest and extend them to "American CS students can't compete." Have you really done these programming contests? Are you seriously implying that dynamic programming with memoization is something you are even remotely likely to need in the average IT software project? Bipartite matching?
I don't think you understood his arguements at all:
"From this I guess the author means that it's OK to be at the same level they were eight years ago."
He is just talking about how percentiles work, and that the news isn't as alarming as it sounds. He isn't saying America shouldn't be motivated to be better, he is just speaking out against the alarmists.
"OMG, it's not fair, they trained harder! Well hello! Is it cheating to produce programmers who can actually solve problems and write code?"
No, he is
If what you say is true, then why would companies be hiring you? You think American companies outsource and hire H1-Bs out of some idealistic belief in a "global economy"? Of course not. They do it because it is cheaper. That said, there are valid arguements for allowing H1-Bs, but don't claim that the obvious is a "myth".
As a former U.S. work visa holder, it's not because we are "Cheaper" it's because there are lots of unqualified people out there and companies want to cast as wide a net as possible.
>>Long before Olympic athletes from all countries became quasiprofessionals, the Eastern European countries were seeing >>to it that training for the Games was their athletes' full-time job, giving them a major advantage over other nations' >>athletes."
>OMG, it's not fair, they trained harder! Well hello! Is it cheating to produce programmers who can actually solve problems >and write code? What exactly is coursework for if it isn't preparation for the kinds of problems you solve in
"How can American engineers compete with cheap, imported labor?" Too much time in academia Norm. If you can't do the job right it really doesn't matter how cheap you come.
That's not true. The real world is not like programming contests for the most part. They are an interesting sport, but not highly practical.
Second, one can throw boddies at the problem, so that wages make a difference. Have more code reviews, more testers, etc.
"Home life as we understand it is no more natural to us than a cage is to a
cockatoo."
-- George Bernard Shaw
A blinkered view from the ivory tower of UC Davis. (Score:4, Informative)
From the article:
"News.com didn't tell you that the number of teams competing has grown nearly sevenfold from 1994 through 2005. In other words, for a team to finish at, say, third place, in 1994 would be equivalent to finishing 21st this year. So a hypothetical team that News.com would have lauded in 1994 would now be dismissed as having badly "slipped" in 2005, even though it would be of the same quality."
From this I guess the author means that it's OK to be at the same level they were eight years ago. It doesn't matter that the American teams didn't improve at the same rate at the rest of the world. And in his statistical argument he ignores that although team numbers might have increased so did the number of American teams.
Next comes my absolute favorite argument:
"Long before Olympic athletes from all countries became quasiprofessionals, the Eastern European countries were seeing to it that training for the Games was their athletes' full-time job, giving them a major advantage over other nations' athletes."
OMG, it's not fair, they trained harder! Well hello! Is it cheating to produce programmers who can actually solve problems and write code? What exactly is coursework for if it isn't preparation for the kinds of problems you solve in programming contests? I've done a couple - it's the same thing, you just have to be faster and more accurate, compared to a programming assignment.
"the hidden agenda behind the shrill shortage claims was to push Congress to increase the yearly cap on the H-1B work visa program, which enabled industry to import cut-rate engineers from abroad."
I was a H1-B worker - I made great rates (thanks very much) and so did all the other H1-B's I know. It's convenient for Norm's flawed argument to repeat this myth, propagated by programmers who think they should have had my job because it was their birthright, not because they could have done it better.
"How can American engineers compete with cheap, imported labor?"
Too much time in academia Norm. If you can't do the job right it really doesn't matter how cheap you come. The way to compete is to be the best, there is no other way. Shopping for programmers is not like shopping for socks. Remember, computer-related thingys are digital. At the end of the day it is usually pretty obvious whether they work or do not work. "Almost works" is not good enough for anyone, except perhaps a professor who grades CS101 papers.
When Chinese (or Indian, or anyone else) programmers turn out to cost less AND be better programmers we'll be able to thank guys like Norm, who wanted to deny there was ever a problem.
What's Norm's issue with devoting more to education - is it just that he wants to be able to say "It wasn't MY fault?"
Re:A blinkered view from the ivory tower of UC Dav (Score:2, Informative)
While your statistical point is valid, your improvement one is not. He's saying that there's a large number of new entries, not that existing entries got better.
OMG, it's not fair, t
Re:A blinkered view from the ivory tower of UC Dav (Score:1)
Clearly you're not a programmer.
The "best" guy will get it done in the right way. Doing it "right" means doing it with an eye for maintainability and making things simpler to do later. Usually this takes *longer* at the beginning. Often, much, much longer.
Faster != better.
Re:A blinkered view from the ivory tower of UC Dav (Score:1)
90% of the programming jobs out there don't have managers that can understand or indeed read code, much less actual peer review.
In a perfect world, maybe, but the world is not perfect. Most code out there is not peer reviewed and barely tested.
This is not the case in my current job (
Re:A blinkered view from the ivory tower of UC Dav (Score:1, Insightful)
Your experience may be atypical (I say to be polite), be
Re:A blinkered view from the ivory tower of UC Dav (Score:1)
we're quite happy being mediocre, thanks.
Re:A blinkered view from the ivory tower of UC Dav (Score:3, Insightful)
He doesn't say it isn't fair. He says it is not fair to take the results of the contest and extend them to "American CS students can't compete." Have you really done these programming contests? Are you seriously implying that dynamic programming with memoization is something you are even remotely likely to need in the average IT software project? Bipartite matching?
Re:A blinkered view from the ivory tower of UC Dav (Score:2)
"From this I guess the author means that it's OK to be at the same level they were eight years ago."
He is just talking about how percentiles work, and that the news isn't as alarming as it sounds. He isn't saying America shouldn't be motivated to be better, he is just speaking out against the alarmists.
"OMG, it's not fair, they trained harder! Well hello! Is it cheating to produce programmers who can actually solve problems and write code?"
No, he is
no (Score:2)
As a former U.S. work visa holder, it's not because we are "Cheaper" it's because there are lots of unqualified people out there and companies want to cast as wide a net as possible.
Mos
Re:A blinkered view from the ivory tower of UC Dav (Score:1)
>OMG, it's not fair, they trained harder! Well hello! Is it cheating to produce programmers who can actually solve problems >and write code? What exactly is coursework for if it isn't preparation for the kinds of problems you solve in
Re:A blinkered view from the ivory tower of UC Dav (Score:1)
That's not true. The real world is not like programming contests for the most part. They are an interesting sport, but not highly practical.
Second, one can throw boddies at the problem, so that wages make a difference. Have more code reviews, more testers, etc.