The population of urban and suburban areas are liberal even in the Reddest of states, and in rural areas people are conservative even in the Bluest of states. Urban and suburban areas in Red of states vote Democrat, rural areas in Blue states vote Republican.
Ah, good to see yet another person who Has Never Been To Texas.
HP will steal it from them claiming that since they (the employees) already work for HP, that HP owns anything they might tend to create and thusly, will kill the project and notify all HP employees that similar skunkworks will get them fired.
Well that's just it, C++ is designed to be as general a language as you can manage
No, that's Lisp.
C++ is designed to be the fastest runtime language as BS can manage, pushing all of the optimization work to the compiler and allowing people to pay the performance price for only what they need.
Thus, your blanket assumption that any given "entire logging infrastructure" will only give or take "a whopping 0.1 CPU-seconds per hour" is... pretty dumb.
I wasn't actually assuming that but I suppose you could see it that way. I actually said "IF your entire logging..."
I must have been somehow spoiled by my logging infrastructure. We could specify log levels for individual subsystems (down to the class level) and unless logging was actually turned on for a particular class nothing was generated to be thrown away at a different layer. And it DIDN'T take a hash lookup on each logging statement (that would have been stupid indeed), instead logger flags were changed just once across the (clustered) application when the user changed the trace spec. This consumes only a couple MB more memory having all those separate logging flags, but performance was fine even with a static synchronized output function -- though it helped that it cached the string representation of System.currentTimeMillis(). And yes, enabling a lot of logging would slow things down considerably -- up to 99% slowdown in some cases -- but when logging was off there was essentially no measurable difference between having the check-then-log-statement and not having log-statements at all. (And yes, I had to verify that myself because of other performance bottlenecks being investigated.) Logging messages would appear in stdout/stderr and also in the appserver cluster console for severe errors.
These days I wouldn't even bother with any logging framework that doesn't have at least these basic features.
Of course it matters. If your entire logging infrastructure when turned off adds up to a whopping 0.1 seconds per hour of useful work, but can be turned on when needed to locate real problems, then not having it is pretty dumb.
People just don't realize the costs of logging everything.
Are you operating on embedded systems? Because anything faster than about 500MHz can easily do a few million "if (logging_level >= LOGGING_LEVEL_CONSTANT) {... log_something... }" type checks per second.
There is no code in the world that I can think of that needs a log line after every two steps in a procedure.
Any code in which timeouts can affect the result would require this kind of logging, which includes networking code or code that handshakes between multiple threads/processors. Example: debugging something like a new x/y/zmodem implementation is nigh impossible within a debugger because your side must respond within 10 seconds or the other side will start acting differently.
I tend to notice more the people than the individual posts, and what I've seen many times is one liberal (in the objective sense, not the American "anyone to the left of Reagan is a left-wing radical sense) defending a left-ish interpretation of some fact against a much larger number of pot-shots from conservative / libertarian types. That one person may have several of their posts upmodded, but in general the total number of right-wing posts exceeds theirs.
Does Obama really think he'll force us into Alternative Energy technologies by giving us no other choices?
No, I think he knows that we'll be forced into alternative energy technologies by the Lomonosov-Lavoisier law. Our only real choice is to either slowly adjust our lifestyles now (by developing the alternative energy technologies quickly enough to handle our energy needs) or rapidly adjust our lifestyles later (in an economically catastrophic way).
I suppose I could have just sat around and whined until now crying about how insurmountable things are, I am sure I would be better off right now. let me know how it turns out for you.
Dude, I didn't sit around and whine, I changed my life and am now doing great thankyouverymuch. If you had READ what I wrote YOU GODDAMN MORON you would have known that.
yeah I came up in the 80's right after obama's buddy jimmy demolished the economy with 13% mortgages double digit unemployment and regan's wonderful tinkle on the poor economics.
I'd trade today for 1980 in a heartbeat. I'll gladly take double-digit unemployment in a manufacturing economy -- where the remaining jobs are still decently-paid enough to raise a whole family on -- rather than single-digit unemployment in today's service economy where it takes two jobs to get 2/3 as far as one job from 1980. And I'd also gladly take 13% mortgages since it kept the housing sector way more solid than today's subprime meltdown. Most of all, I'd take university tuition that was less than 1/5 of today's cost.
I think if you were just now coming out of school you'd be a lot less "take it like a man!" and a lot more "whine whine whine" yourself.
I have been in the "biz" over 20 years. My first jobs I got because I could spell computer. I started life as a programmer.
Us young whippersnappers didn't come out in the same economy youse did. Most of us couldn't get decent jobs without a 4-year degree (1997-2001 era excluded), most of us didn't get through college without 10-25 grand in school debt, and most of us didn't get the opportunity to work during the tail end of the largest economic expansion in the Western world's history. Our gas was never less than $1.20 per gallon, though I remember $0.79/gal as a kid. Our houses were never less than $80,000, and they won't magically quintuple in price over the next 15-year span to make our retirements easier.
I'm glad you've had such a successful life. I started switching careers four years ago and made some very serious commitments to living my life on a much smaller scale than I initially wanted in order to make way for similar success. In ten years at my new job I'll feel a bit more secure and might have your outlook, but right now I empathize a lot with everyone else who was in my spot but didn't have the kind of unique circumstances I did to make that change.
That's your problem right there: we do not now have, nor ever have had, a free market for skilled jobs.
One big barrier is college: 99% of corporate HR departments will not hire without a 4-year degree related to the field, period. That means for most entry-level new hires walking into the job with some debt and thus being unable to face their employer as real equals at the bargaining table.
The other big barrier is location: for some crazy reason, 99% of the population does not want to uproot and move across the country every few years for the best available job that matches their skills. It's almost as if people have some other priorities in their life besides salary. Go figure!
When most of the jobs out there are staffed with people who have debts and can't move around at whim, you've got a situation not unlike the sharecropping system. And the CXO's treat it accordingly.
It's a tough market. The jobs are there, but the applicants unfortunately aren't.
The market lost me five years ago when I returned to school to seek even bigger dollars as a real (as in chemical) engineer. The main reason I left was that I realized (at 27) that even with my skills (which are pretty decent) I would be nigh unemployable after 45 when my hypothetical children might be entering high school. I didn't want to have to move all over the country just to keep a decent salary either.
Perhaps you would have better luck finding your developer(s) if you specifically searched for older 40+ workers.
If girls have an average score, then no matter how many of them take the test that average should remain the same. That's what *average* means practically by definition.
That's true only if there is a statistically random chance for any girl or boy to take the test. I believe the argument is actually that the test participants are pre-selected such that those more likely to score low don't take it at all, AND that more of the would-have-scored-lower boys didn't take the test skewing the boys average higher than the girls.
Not only are they wrong, but the College Board practically told them they were. They collected data, performed flawed analysis, and then ignored the fact that the people who administer the bloody test disagree with their results.
Given the history of SAT scores, I don't think the analyses from College Board are even worth the paper they are printed on.
that nobody (outside MS) has that kind of skill wrt windows, at all. And that complaining rarely helps, if ever.
No shit. I started working with Access 2007 and there is STILL a bug in exporting tables to CSV from Access 95-ish (it assumes that doubles are currency and truncates them to 2 decimal places).
Sure there is. Download a windows program and run it. It just works.
Right, it just works. That is, if it's a standalone.exe and has no serious dependencies like.Net or Java or VB(3-6). Or if it isn't a standalone.exe (requires installation) and it can install into a user directory. Or it installs to a global directory and you're an admin.
Just as in Linux with particular libraries, Windows users also sometimes have to scour the 'net for those special dll's that stopped shipping with various releases of Windows.
I believe we need more women in science not for brilliant outliers (though I'll happily accept any who come along), but because women do think differently than men and tend to work more cooperatively and synergistically.
Agreed. In my major (Chemical Engineering), the gender ratio of the 20-somethings at the couple of universities I attended and at my current job is close to 1:1, and it's making for a much more productive (and far less stressful) environment for everyone.
Slashdot Mirror
The population of urban and suburban areas are liberal even in the Reddest of states, and in rural areas people are conservative even in the Bluest of states. Urban and suburban areas in Red of states vote Democrat, rural areas in Blue states vote Republican.
Ah, good to see yet another person who Has Never Been To Texas.
What will happen, however, is that American's will start using more American oil, reducing it's imports of oil,
And just how is that supposed to happen? Nationalize Exxon? Legislate that oil from American fields must be sold only to American refineries?
HP will steal it from them claiming that since they (the employees) already work for HP, that HP owns anything they might tend to create and thusly, will kill the project and notify all HP employees that similar skunkworks will get them fired.
Fixed that for you.
Make the list password protected, and make the feature require an explicit activation step.
User wants to enable private browsing feature:
1. Start browser.
2. Click the private mode button.
3. Password prompt: correct password loads custom settings (including white/black lists), or no password loads default settings.
4. Continue browsing.
Well that's just it, C++ is designed to be as general a language as you can manage
No, that's Lisp.
C++ is designed to be the fastest runtime language as BS can manage, pushing all of the optimization work to the compiler and allowing people to pay the performance price for only what they need.
Thus, your blanket assumption that any given "entire logging infrastructure" will only give or take "a whopping 0.1 CPU-seconds per hour" is... pretty dumb.
I wasn't actually assuming that but I suppose you could see it that way. I actually said "IF your entire logging..."
I must have been somehow spoiled by my logging infrastructure. We could specify log levels for individual subsystems (down to the class level) and unless logging was actually turned on for a particular class nothing was generated to be thrown away at a different layer. And it DIDN'T take a hash lookup on each logging statement (that would have been stupid indeed), instead logger flags were changed just once across the (clustered) application when the user changed the trace spec. This consumes only a couple MB more memory having all those separate logging flags, but performance was fine even with a static synchronized output function -- though it helped that it cached the string representation of System.currentTimeMillis(). And yes, enabling a lot of logging would slow things down considerably -- up to 99% slowdown in some cases -- but when logging was off there was essentially no measurable difference between having the check-then-log-statement and not having log-statements at all. (And yes, I had to verify that myself because of other performance bottlenecks being investigated.) Logging messages would appear in stdout/stderr and also in the appserver cluster console for severe errors.
These days I wouldn't even bother with any logging framework that doesn't have at least these basic features.
That is a fantastic idea! (goes to add feature to my logger classes...)
Of course it matters. If your entire logging infrastructure when turned off adds up to a whopping 0.1 seconds per hour of useful work, but can be turned on when needed to locate real problems, then not having it is pretty dumb.
One feature I like is a built in String.format function that won't evaluate if logging is disabled.
That's what I love about Lisp: the ability to make a macro that defers evaluation for precisely this case.
People just don't realize the costs of logging everything.
Are you operating on embedded systems? Because anything faster than about 500MHz can easily do a few million "if (logging_level >= LOGGING_LEVEL_CONSTANT) { ... log_something ... }" type checks per second.
There is no code in the world that I can think of that needs a log line after every two steps in a procedure.
Any code in which timeouts can affect the result would require this kind of logging, which includes networking code or code that handshakes between multiple threads/processors. Example: debugging something like a new x/y/zmodem implementation is nigh impossible within a debugger because your side must respond within 10 seconds or the other side will start acting differently.
I tend to notice more the people than the individual posts, and what I've seen many times is one liberal (in the objective sense, not the American "anyone to the left of Reagan is a left-wing radical sense) defending a left-ish interpretation of some fact against a much larger number of pot-shots from conservative / libertarian types. That one person may have several of their posts upmodded, but in general the total number of right-wing posts exceeds theirs.
Does Obama really think he'll force us into Alternative Energy technologies by giving us no other choices?
No, I think he knows that we'll be forced into alternative energy technologies by the Lomonosov-Lavoisier law. Our only real choice is to either slowly adjust our lifestyles now (by developing the alternative energy technologies quickly enough to handle our energy needs) or rapidly adjust our lifestyles later (in an economically catastrophic way).
I suppose I could have just sat around and whined until now crying about how insurmountable things are, I am sure I would be better off right now. let me know how it turns out for you.
Dude, I didn't sit around and whine, I changed my life and am now doing great thankyouverymuch. If you had READ what I wrote YOU GODDAMN MORON you would have known that.
yeah I came up in the 80's right after obama's buddy jimmy demolished the economy with 13% mortgages double digit unemployment and regan's wonderful tinkle on the poor economics.
I'd trade today for 1980 in a heartbeat. I'll gladly take double-digit unemployment in a manufacturing economy -- where the remaining jobs are still decently-paid enough to raise a whole family on -- rather than single-digit unemployment in today's service economy where it takes two jobs to get 2/3 as far as one job from 1980. And I'd also gladly take 13% mortgages since it kept the housing sector way more solid than today's subprime meltdown. Most of all, I'd take university tuition that was less than 1/5 of today's cost.
I think if you were just now coming out of school you'd be a lot less "take it like a man!" and a lot more "whine whine whine" yourself.
I have been in the "biz" over 20 years. My first jobs I got because I could spell computer. I started life as a programmer.
Us young whippersnappers didn't come out in the same economy youse did. Most of us couldn't get decent jobs without a 4-year degree (1997-2001 era excluded), most of us didn't get through college without 10-25 grand in school debt, and most of us didn't get the opportunity to work during the tail end of the largest economic expansion in the Western world's history. Our gas was never less than $1.20 per gallon, though I remember $0.79/gal as a kid. Our houses were never less than $80,000, and they won't magically quintuple in price over the next 15-year span to make our retirements easier.
I'm glad you've had such a successful life. I started switching careers four years ago and made some very serious commitments to living my life on a much smaller scale than I initially wanted in order to make way for similar success. In ten years at my new job I'll feel a bit more secure and might have your outlook, but right now I empathize a lot with everyone else who was in my spot but didn't have the kind of unique circumstances I did to make that change.
Thing is, in truly free market place,
That's your problem right there: we do not now have, nor ever have had, a free market for skilled jobs.
One big barrier is college: 99% of corporate HR departments will not hire without a 4-year degree related to the field, period. That means for most entry-level new hires walking into the job with some debt and thus being unable to face their employer as real equals at the bargaining table.
The other big barrier is location: for some crazy reason, 99% of the population does not want to uproot and move across the country every few years for the best available job that matches their skills. It's almost as if people have some other priorities in their life besides salary. Go figure!
When most of the jobs out there are staffed with people who have debts and can't move around at whim, you've got a situation not unlike the sharecropping system. And the CXO's treat it accordingly.
There is only one truth about unions worth remembering: companies get the unions they deserve.
It's a tough market. The jobs are there, but the applicants unfortunately aren't.
The market lost me five years ago when I returned to school to seek even bigger dollars as a real (as in chemical) engineer. The main reason I left was that I realized (at 27) that even with my skills (which are pretty decent) I would be nigh unemployable after 45 when my hypothetical children might be entering high school. I didn't want to have to move all over the country just to keep a decent salary either.
Perhaps you would have better luck finding your developer(s) if you specifically searched for older 40+ workers.
My brother, who's an engineer, generally agrees.
I'm an engineer and I agree too. Lord knows I don't want millions more people out there competing for my well-paying job...
If girls have an average score, then no matter how many of them take the test that average should remain the same. That's what *average* means practically by definition.
That's true only if there is a statistically random chance for any girl or boy to take the test. I believe the argument is actually that the test participants are pre-selected such that those more likely to score low don't take it at all, AND that more of the would-have-scored-lower boys didn't take the test skewing the boys average higher than the girls.
Not only are they wrong, but the College Board practically told them they were. They collected data, performed flawed analysis, and then ignored the fact that the people who administer the bloody test disagree with their results.
Given the history of SAT scores, I don't think the analyses from College Board are even worth the paper they are printed on.
that nobody (outside MS) has that kind of skill wrt windows, at all. And that complaining rarely helps, if ever.
No shit. I started working with Access 2007 and there is STILL a bug in exporting tables to CSV from Access 95-ish (it assumes that doubles are currency and truncates them to 2 decimal places).
Sure there is. Download a windows program and run it. It just works.
Right, it just works. That is, if it's a standalone .exe and has no serious dependencies like .Net or Java or VB(3-6). Or if it isn't a standalone .exe (requires installation) and it can install into a user directory. Or it installs to a global directory and you're an admin.
Just as in Linux with particular libraries, Windows users also sometimes have to scour the 'net for those special dll's that stopped shipping with various releases of Windows.
I believe we need more women in science not for brilliant outliers (though I'll happily accept any who come along), but because women do think differently than men and tend to work more cooperatively and synergistically.
Agreed. In my major (Chemical Engineering), the gender ratio of the 20-somethings at the couple of universities I attended and at my current job is close to 1:1, and it's making for a much more productive (and far less stressful) environment for everyone.
Good. You act like a dildo, you reap the consequences.
Just how does one "act like a dildo" ?