nope. There exist no vehicles you can buy as a private citizen that can drive on normal roads to an on-ramp, then continue on rail to a off-ramp near the destination, then on normal roads for the last part.
The single-vehicle is key, if you drive to the station and park there, then to wait for a train and go to the destination-station, then get a *third* vehicle of some sort near the destination, then it becomes hopelessly impractical for short and medium trips. (for very long trips, it may be ok)
Trains have huge problems competing with just plain moving the goods by truck. The re-loading-problem is a significant part of the reason for that. Moving containers onto a train, and have that train depart at the right time etc is not free.
Trains -do- compete for some types of jobs, but for most, they're not competitive with plain trucks.
It works, but is only cost-effective for REALLY long distances, like inter-continental-transport, for shorter hauls, the re-loading would kill any wins, even if the rope-way was FREE, it might not be worthwhile using it.
Cars are less efficient in energy/utility terms, yes. But they are without comparison most efficient in practical/versatile terms. Which in practice is what people prioritize:
Your car always goes to the *time* you want to go. It starts *precisely* at your house and goes *precisely* to your destination. The road-network is magnitudes more fine-masked than any other existing transport-network. You have your own private car. You can store things, like groceries in the car while visiting a second shop. Your car is always available for your exclusive use.
Imagine that cars could drive electrically and driverless on a railnetwork, and additionally could drive on roads. It's win-win:
You get all the advantages of todays cars (advantages that explain the popularity of cars!) and in addition to that:
Full-electric mode even on long-hauls. (power from the rail). Less noise (rubber-on-steel-rail-electric-engine is SILENT) Less local pollution (still pollution at site of electricity-generation though). Higher capacity (cars on the rail can form "trains" with zero inter-vehicle-distance, so a single rail has the capacity of 3-4 lanes) Free time: you can read your email or watch the news while heading to work. Long-range-electric. Currently electric lacks range. If you need range only to get to the nearest rail-onramp and from offramp to destination, then this concern is significantly lessened. Significantly safer (elevated monorail-vehicles don't tend to hit many kids, and you don't skid off them even on icy conditions)
"get everyone to lower their practicality and comfort-requirements to what we're able to offer by bus or train" isn't realistic. Infact the opposite will happen: as people get richer, they consider wasted time and inconvenience LESS acceptable, not more.
Yes. But many bombings (suicide or not) do fall atleast close to this scenario.
Occupying forces in Afghanistan and Iraq (i.e. our forces!) have an express plan of gradually handing power over to the local government and it's security-forces/police/military. Would you say, blowing up, say the recruitment-lines for a school educating new police-officers in Afghanistan, is attacking a civillian target ? It's fairly grey, if you ask me - these schools are directly supported by the occupying forces, and their explicitly stated purpose, is to educate locals to take over some of the tasks that are presently handled by occupying soldiers.
Sure, it's easy to find examples that are *clearly* attacks on unrelated civillians, but a fairly large percentage of attacks are also on softer-targets on the outskirts, "civillian" institutions that form part of the military strategy and/or support-mechanism.
In short, you're right. It's tricky to say exactly what the proper definition of civillian is. Is a petrol-station-owner a civilian if he delivers petrol to Blackwater ? I honestly don't know.
The principal problem of this, and most other similar suggestions, is that while such a system is good for a-b transport, reality is a network. Such a system helps you not at all, unless the goods to be delivered are already at the start-station, and are being transported to the end-station.
If not, you need to *first* load it on one mode of transport (typically some kind of car) -then- drive to the nearest "station" where the goods are repackaged, then near the destination, repeat.
It turns out the delays and costs of reloading cargo, frequently makes the economy such that it's better to simply go the entire distance by lorry. The advantage of the lorry is that it goes from where your goods are, to where you want them, with zero intermediary re-loads. (typically anyway, sure there's exceptions)
The lack of a robust network, also makes the system vulnerable. When (not if!) one ropeway breaks down, what do you do ? Reroute onto roads ? Wait ?
I think the best hopes are for dual-mode-transport, that is, vehicles that can drive both on normal roads -and- on special-purpose tracks of some sort. Doing this, gives you the best of two worlds. Have a look at http://www.ruf.dk/ for an example system.
I guess it could be rationalized if the civillians where, say, delivering supplies to a military base. But then again, if someone is willingly working to directly support military operations, I guess it's questionable if they count as civillians at all.
That is, if your country is occupied by a foreign power, I find it not-too-hard to rationalize blowing up for example a diesel-truck on the way to one of the bases of the occupying forces, even if the driver ain't a soldier.
Not that that has anything to do with what you're talking of, offcourse.
It's ambigious, sure. But you can help. A sequence that consists of a product-of-two-primes-symbols, can only be split one way.
If 35 bits (for example) are to be split up at all, then 5*7 (or 7*5) is the only way to do it. And you can give hints to that interpretation by framing, it doesn't take a genious to figure out how to split:
the 7-groups of 1s and 0s, connected with the fact that the *entire* string is cleanly splittable in chunks-of-7 is a fairly strong hint that this is the thing to do.
Getting people to arrange the groups in a 2-dimensional grid is more difficult, but mutiplication-as-area is fairly inituitive, the area of a 5x7 rectangle is 35. 5 groups of 7 objects, can be arranged in two dimensions cleanly. (but NOT in 3, because 5 and 7 are primes)
It still does not matter if the random string is large enough. 122 bits is large enough.
Simplified, the birthday-paradox says that you can expect a collision (on the average!) when you've got aprox the square root of the keyspace participants. Further simplification says you get the square root by simply chopping a number in half. (this ain't accurate, but good enough for back-of-the-envelope)
Thus, you'd expect to see collisions with APROXIMATELY 2^61 vehicles, if and only if, every vehicle has been within range of every other vehicle. 2**61 is 2305843009213693952 or 384307168 cars for every human being on the planet.
Buy a car for EVERY human. Place them ALL in range of eachother. Have them each change id every minute.
And you'd expect, on average, to get the first collision after aproximately 1000 years.
In real life, that's off by a factor of 5 (fraction who owns a car), times a factor of 60 (hourly id, not every minute!), times a factor of 1000+ (fraction of cars which are in range of oneanother at any given time)
In short:
You're wrong. A collision really genuinely is exceedingly unlikely. Even given VERY liberal assumptions.
It's a dilemma actually. What to do about a political party in a democracy who says they want to dismantle democracy.
"Elect us ! We'll enact sharia, and never hold an election again !"
Even if the people do, I would argue that their claim to power expires, there's a reason elections are held regularily - even if the people support a certain policy now, it doesn't mean that they will indefinitely. Furthermore if a dictatorship rules for a couple of decades, a significant fraction of the population will *never* have been asked because at the time of last election, they where too young to vote, or wheren't even born.
It depends of how many random bits you have, and how often you test. The birthday paradox does not really apply here, because you don't have a problem if -somewhere- there's a car with your id, but instead only if there is a car with your ID that ever gets within (say) 10 miles of yours.
The standard java-uuid-thing has 122 random bits. That's rather a lot. Even if the birthday-paradox applied fully (it does not), you could have every human on the planet own a car, and every car on the planet create a new uuid every minute, and the expected time to first collision would still be in the millions-of-years category.
You contradict yourself. First, you say that the article does NOT claim that self-control causes success (but merely that it predicts it) - then you go on to say that instilling self-control will likely help them in life (i.e. that self-control likely causes success.)
The theory is plausible, but uncertain. It's entirely possible that self-control is merely a proxy for some underlying mechanism that is the real cause of success, and that increased self-control-training as such has no effect.
Could be tested offcourse. You could -give- self-control training to half of a group of kids, then compare how the two groups do later in life, to ascertain if the training had any effect.
I don't know if that study has been done - but atleast it's not done in the study linked from this story.
The burst-disc prevents overfiling. If the tank is rated 3000psi, it'll have a burst-disc that ruptures in the 125% to 140% range (i.e. somewhere between 3750 and 4200 psi.
This is a great safety-device, particularily the newer ones that vent gas to both sides of the neck, thus reducing or eliminating the out-of-control-spin that could be dangerous with the older systems.
But it does nothing if the problem isn't over-pressure. If for example the valve-threads or the tank itself is damaged by corrosion the valve or tank could fail, even though the pressure is not higher than nominal.
Mechanical impact to the valve, can and has caused the same thing.
Burst-disc reduce risk from overfilling or overheating the tank (heating the tank causes pressure to rise)
It does very little, or nothing, to reduce the risk from tank-failure, valve-failure or impacts-causing-failure.
Claim: Statistically speaking, those with a degree today, are less intelligent than those with a degree a decade, or two, or 3, ago.
Rationale: Back then, a small fraction of the population had a degree, there's less weeding going on when a larger fraction of the population have higher degrees.
Thus, you have a widening salary-gap despite a closing intelligence-gap.
It's *very* common with speed/price/size tradeoffs in engineering, regardless of which technology choosen. Usually, "pick any 2" is what it boils down to.
Sure, in principle, if you invent something that scores well on all 3 axes, then no trade-off is nessecary and all the older systems that score poorly on atleast one of the 3 axis, are obsolete. But I'm not holding my breath.
For storing large amounts of data for a long time, you want huge and cheap, but need little speed. For the registers in your CPU, the opposite is the case, you want all the speed you can possibly get, yet the amount of storage is very limited.
3000 psi is equivalent to 2000N for every square cm. The neck of a scuba-bottle has a cross-section of perhaps 2cm^2, so a catastrophic failure of the valve would subject the tank to about 4000N of force.
A full 10l tank weighs about 20kg, thus 4000N gives it an acceleration of about 200m/s^2, i.e. if the pressure didn't drop (it does offcourse!), it'd reach 500mph in a second, give or take.
A 40 pound projectile moving at 500mph is not "just air", if it gets lose in a confined area, the question ain't if it'll go trough the window, but if it'll go trough the *wall*.
You do -not- want to be close to a 3000psi pressurised tank that ruptures.
While your critique of Wikipedia has -some- merit, cost really isn't it.
Show me a website with atleast 10% of wikipedias activity-level, that doesn't have atleast ten times the budget. $20M/year is about the same amount we as a society use on rubber-bands, it's an utterly insignificant sum.
(yes, I get that Wikipedia is only cheap 'cos the contributions are free)
That's actually very close to irrelevant for an offsite-backup.
Given the cost-structure you CERTAINLY have atleast one on-site backup, simply 'cos it's essentially free. (a usb-terabyte-disc ain't precisely pricey)
The off-site backup is thus your secondary backup - existing at all to safeguard against stuff like a fire or a burglar.
The odds that your house burn down - AND your online backup-provider (say dropbox) goes out of business the same week (before you can re-download the data and reestablish a local copy) is astronomically low.
If this is something that worries you (my house could burn the same week Dropbox closes it's door suddenly and unexpectedly), then you're asking for a level of assurance you won't get from ask slashdot - you're basically asking for -multiple- independent offsite backups.
Sure, go ahead, sign up for 2 different services, make sure they don't use the same physical facility. It'll cost double in both money and bandwith, and make very little difference, but if you're really paranoid, feel free.
It's aproximately doable, but by no means "trivial". Because to estimate, you need to know the deductions, which include fairly big ones such as mortgage-interest and childcare-costs (both of which are movable between married partners)
This doesn't help you a lot. It does give you a *minimum* as my real gross is certainly atleast the taxable-income plus the minimum deduction that everyone gets. so when this says 281.597, you can look up the minimum deduction and conclude I earned *atleast* 350.000 in 2009.
But you don't know if I got this all from one job, or from several. You don't know if some of it was investment-income. You don't know if my deductions for interest paid, childcare and other stuff was 50K or 150K.
In short, you can make a guesstimate, but the error-bars are large enough to make it fairly useless for salary-estimation. You'd basically end up guessing that my monthly gross in 2009 was probably in the 30K to 45K bracket. This is correct -- but it's much too imprecise to be useful. If your job is similar to mine, you have similar skills, and you're paid 37K - do I earn 7K more than you, 7K less than you -- or the same thing you do ?
But employees should do something different: talk to oneanother, including about salary. Your employer knows everyones salary anyway, it's to his advantage that he has all the knowledge, and you have none of it.
Besides, even the well-paid ones have a interest in seeing the poorly-paid ones catch up, because being well-paid relatively speaking, has an associated risk, namely that the company starts considering you "expensive" relative to the others.
It's truly bizarre connecting sharing or not of a number to what kind of a number it is.
I don't know USA, but here in Norway, you can generally find most cell-numbers the same place you find most regular numbers; http://tlf.no/
It's a choice - when you sign up for a number (either sort!) you get to *choose* if you want to be listed or not. What a concept ! Oh yeah, and precisely the same thing applies if you've got a VoIP line. (there's actually three levels; "listed, unlisted, secret" but that's details.
Furthermore, especially for technical workers like programmers, decent equipment is a must.
It would be idiotic of a company to pay ~$100K/year in salary + $20K in attached costs + ~$10K for offices, training, food and so on -- and then give me sub-standard equipment. An *excellent* computer with two good screens for programming is what ? $2500 ? And has a practical use-time of 2-3 years. That's not even 1% of the total cost.
If I get -1%- more done in a year with good equipment, then it's worth it. Should be a nobrainer.
Some companies don't have brains. It's better not to work for them.
Thing is, even in bad job-markets SOME of your employeers has a choice - typically the most valuable ones.
So pull bullshit like this, and what happens is that those who CAN leave, do it. In other words, the result is that those who are the most valuable in the job-market, typically the ones who are the most valuable to you too, leave.
Slashdot Mirror
nope. There exist no vehicles you can buy as a private citizen that can drive on normal roads to an on-ramp, then continue on rail to a off-ramp near the destination, then on normal roads for the last part.
The single-vehicle is key, if you drive to the station and park there, then to wait for a train and go to the destination-station, then get a *third* vehicle of some sort near the destination, then it becomes hopelessly impractical for short and medium trips. (for very long trips, it may be ok)
This is what trains do.
Trains have huge problems competing with just plain moving the goods by truck. The re-loading-problem is a significant part of the reason for that. Moving containers onto a train, and have that train depart at the right time etc is not free.
Trains -do- compete for some types of jobs, but for most, they're not competitive with plain trucks.
re-loading containers is ALSO time and cost-intensive. This is what we do with ships today for many goods.
Container-truck - container-ship - container-truck.
It works, but is only cost-effective for REALLY long distances, like inter-continental-transport, for shorter hauls, the re-loading would kill any wins, even if the rope-way was FREE, it might not be worthwhile using it.
true, you can do that, or atleast use standard containers.
You can't reasonably drive a lorry onto a ropeway though :)
Cars are less efficient in energy/utility terms, yes. But they are without comparison most efficient in practical/versatile terms. Which in practice is what people prioritize:
Your car always goes to the *time* you want to go. It starts *precisely* at your house and goes *precisely* to your destination. The road-network is magnitudes more fine-masked than any other existing transport-network. You have your own private car. You can store things, like groceries in the car while visiting a second shop. Your car is always available for your exclusive use.
Imagine that cars could drive electrically and driverless on a railnetwork, and additionally could drive on roads. It's win-win:
You get all the advantages of todays cars (advantages that explain the popularity of cars!) and in addition to that:
Full-electric mode even on long-hauls. (power from the rail). Less noise (rubber-on-steel-rail-electric-engine is SILENT) Less local pollution (still pollution at site of electricity-generation though). Higher capacity (cars on the rail can form "trains" with zero inter-vehicle-distance, so a single rail has the capacity of 3-4 lanes) Free time: you can read your email or watch the news while heading to work. Long-range-electric. Currently electric lacks range. If you need range only to get to the nearest rail-onramp and from offramp to destination, then this concern is significantly lessened. Significantly safer (elevated monorail-vehicles don't tend to hit many kids, and you don't skid off them even on icy conditions)
"get everyone to lower their practicality and comfort-requirements to what we're able to offer by bus or train" isn't realistic. Infact the opposite will happen: as people get richer, they consider wasted time and inconvenience LESS acceptable, not more.
Yes. But many bombings (suicide or not) do fall atleast close to this scenario.
Occupying forces in Afghanistan and Iraq (i.e. our forces!) have an express plan of gradually handing power over to the local government and it's security-forces/police/military. Would you say, blowing up, say the recruitment-lines for a school educating new police-officers in Afghanistan, is attacking a civillian target ? It's fairly grey, if you ask me - these schools are directly supported by the occupying forces, and their explicitly stated purpose, is to educate locals to take over some of the tasks that are presently handled by occupying soldiers.
Sure, it's easy to find examples that are *clearly* attacks on unrelated civillians, but a fairly large percentage of attacks are also on softer-targets on the outskirts, "civillian" institutions that form part of the military strategy and/or support-mechanism.
In short, you're right. It's tricky to say exactly what the proper definition of civillian is. Is a petrol-station-owner a civilian if he delivers petrol to Blackwater ? I honestly don't know.
The principal problem of this, and most other similar suggestions, is that while such a system is good for a-b transport, reality is a network. Such a system helps you not at all, unless the goods to be delivered are already at the start-station, and are being transported to the end-station.
If not, you need to *first* load it on one mode of transport (typically some kind of car) -then- drive to the nearest "station" where the goods are repackaged, then near the destination, repeat.
It turns out the delays and costs of reloading cargo, frequently makes the economy such that it's better to simply go the entire distance by lorry. The advantage of the lorry is that it goes from where your goods are, to where you want them, with zero intermediary re-loads. (typically anyway, sure there's exceptions)
The lack of a robust network, also makes the system vulnerable. When (not if!) one ropeway breaks down, what do you do ? Reroute onto roads ? Wait ?
I think the best hopes are for dual-mode-transport, that is, vehicles that can drive both on normal roads -and- on special-purpose tracks of some sort. Doing this, gives you the best of two worlds. Have a look at http://www.ruf.dk/ for an example system.
Keep that in mind, when the US government suggest implementing an "Internet Kill Switch", please.
I guess it could be rationalized if the civillians where, say, delivering supplies to a military base. But then again, if someone is willingly working to directly support military operations, I guess it's questionable if they count as civillians at all.
That is, if your country is occupied by a foreign power, I find it not-too-hard to rationalize blowing up for example a diesel-truck on the way to one of the bases of the occupying forces, even if the driver ain't a soldier.
Not that that has anything to do with what you're talking of, offcourse.
It's ambigious, sure. But you can help. A sequence that consists of a product-of-two-primes-symbols, can only be split one way.
If 35 bits (for example) are to be split up at all, then 5*7 (or 7*5) is the only way to do it. And you can give hints to that interpretation by framing, it doesn't take a genious to figure out how to split:
11111110000000[data-in-groups-of7-here]00000001111111
the 7-groups of 1s and 0s, connected with the fact that the *entire* string is cleanly splittable in chunks-of-7 is a fairly strong hint that this is the thing to do.
Getting people to arrange the groups in a 2-dimensional grid is more difficult, but mutiplication-as-area is fairly inituitive, the area of a 5x7 rectangle is 35. 5 groups of 7 objects, can be arranged in two dimensions cleanly. (but NOT in 3, because 5 and 7 are primes)
It still does not matter if the random string is large enough. 122 bits is large enough.
Simplified, the birthday-paradox says that you can expect a collision (on the average!) when you've got aprox the square root of the keyspace participants. Further simplification says you get the square root by simply chopping a number in half. (this ain't accurate, but good enough for back-of-the-envelope)
Thus, you'd expect to see collisions with APROXIMATELY 2^61 vehicles, if and only if, every vehicle has been within range of every other vehicle. 2**61 is 2305843009213693952 or 384307168 cars for every human being on the planet.
Buy a car for EVERY human. Place them ALL in range of eachother. Have them each change id every minute.
And you'd expect, on average, to get the first collision after aproximately 1000 years.
In real life, that's off by a factor of 5 (fraction who owns a car), times a factor of 60 (hourly id, not every minute!), times a factor of 1000+ (fraction of cars which are in range of oneanother at any given time)
In short:
You're wrong. A collision really genuinely is exceedingly unlikely. Even given VERY liberal assumptions.
It's a dilemma actually. What to do about a political party in a democracy who says they want to dismantle democracy.
"Elect us ! We'll enact sharia, and never hold an election again !"
Even if the people do, I would argue that their claim to power expires, there's a reason elections are held regularily - even if the people support a certain policy now, it doesn't mean that they will indefinitely. Furthermore if a dictatorship rules for a couple of decades, a significant fraction of the population will *never* have been asked because at the time of last election, they where too young to vote, or wheren't even born.
It depends of how many random bits you have, and how often you test. The birthday paradox does not really apply here, because you don't have a problem if -somewhere- there's a car with your id, but instead only if there is a car with your ID that ever gets within (say) 10 miles of yours.
The standard java-uuid-thing has 122 random bits. That's rather a lot. Even if the birthday-paradox applied fully (it does not), you could have every human on the planet own a car, and every car on the planet create a new uuid every minute, and the expected time to first collision would still be in the millions-of-years category.
You contradict yourself. First, you say that the article does NOT claim that self-control causes success (but merely that it predicts it) - then you go on to say that instilling self-control will likely help them in life (i.e. that self-control likely causes success.)
The theory is plausible, but uncertain. It's entirely possible that self-control is merely a proxy for some underlying mechanism that is the real cause of success, and that increased self-control-training as such has no effect.
Could be tested offcourse. You could -give- self-control training to half of a group of kids, then compare how the two groups do later in life, to ascertain if the training had any effect.
I don't know if that study has been done - but atleast it's not done in the study linked from this story.
The burst-disc prevents overfiling. If the tank is rated 3000psi, it'll have a burst-disc that ruptures in the 125% to 140% range (i.e. somewhere between 3750 and 4200 psi.
This is a great safety-device, particularily the newer ones that vent gas to both sides of the neck, thus reducing or eliminating the out-of-control-spin that could be dangerous with the older systems.
But it does nothing if the problem isn't over-pressure. If for example the valve-threads or the tank itself is damaged by corrosion the valve or tank could fail, even though the pressure is not higher than nominal.
Mechanical impact to the valve, can and has caused the same thing.
Burst-disc reduce risk from overfilling or overheating the tank (heating the tank causes pressure to rise)
It does very little, or nothing, to reduce the risk from tank-failure, valve-failure or impacts-causing-failure.
But the steps are bullshit, because they have as one of the early ones, to admit defeat, and say we can only be saved by, in essence, God.
i.e. if you're atheist you're doomed.
But that's nonsense, atheists addicts have comparable chances to believing addicts, at quitting.
Claim: Statistically speaking, those with a degree today, are less intelligent than those with a degree a decade, or two, or 3, ago.
Rationale: Back then, a small fraction of the population had a degree, there's less weeding going on when a larger fraction of the population have higher degrees.
Thus, you have a widening salary-gap despite a closing intelligence-gap.
If - but that seems a pretty gigantic if to me.
It's *very* common with speed/price/size tradeoffs in engineering, regardless of which technology choosen. Usually, "pick any 2" is what it boils down to.
Sure, in principle, if you invent something that scores well on all 3 axes, then no trade-off is nessecary and all the older systems that score poorly on atleast one of the 3 axis, are obsolete. But I'm not holding my breath.
For storing large amounts of data for a long time, you want huge and cheap, but need little speed. For the registers in your CPU, the opposite is the case, you want all the speed you can possibly get, yet the amount of storage is very limited.
"only" air - surely you're joking ?
3000 psi is equivalent to 2000N for every square cm. The neck of a scuba-bottle has a cross-section of perhaps 2cm^2, so a catastrophic failure of the valve would subject the tank to about 4000N of force.
A full 10l tank weighs about 20kg, thus 4000N gives it an acceleration of about 200m/s^2, i.e. if the pressure didn't drop (it does offcourse!), it'd reach 500mph in a second, give or take.
A 40 pound projectile moving at 500mph is not "just air", if it gets lose in a confined area, the question ain't if it'll go trough the window, but if it'll go trough the *wall*.
You do -not- want to be close to a 3000psi pressurised tank that ruptures.
While your critique of Wikipedia has -some- merit, cost really isn't it.
Show me a website with atleast 10% of wikipedias activity-level, that doesn't have atleast ten times the budget. $20M/year is about the same amount we as a society use on rubber-bands, it's an utterly insignificant sum.
(yes, I get that Wikipedia is only cheap 'cos the contributions are free)
That's actually very close to irrelevant for an offsite-backup.
Given the cost-structure you CERTAINLY have atleast one on-site backup, simply 'cos it's essentially free. (a usb-terabyte-disc ain't precisely pricey)
The off-site backup is thus your secondary backup - existing at all to safeguard against stuff like a fire or a burglar.
The odds that your house burn down - AND your online backup-provider (say dropbox) goes out of business the same week (before you can re-download the data and reestablish a local copy) is astronomically low.
If this is something that worries you (my house could burn the same week Dropbox closes it's door suddenly and unexpectedly), then you're asking for a level of assurance you won't get from ask slashdot - you're basically asking for -multiple- independent offsite backups.
Sure, go ahead, sign up for 2 different services, make sure they don't use the same physical facility. It'll cost double in both money and bandwith, and make very little difference, but if you're really paranoid, feel free.
It's aproximately doable, but by no means "trivial". Because to estimate, you need to know the deductions, which include fairly big ones such as mortgage-interest and childcare-costs (both of which are movable between married partners)
For example, here is the data on me: http://skattelister.no/skatt/profil/eivind-kjorstad-31230853/
This doesn't help you a lot. It does give you a *minimum* as my real gross is certainly atleast the taxable-income plus the minimum deduction that everyone gets. so when this says 281.597, you can look up the minimum deduction and conclude I earned *atleast* 350.000 in 2009.
But you don't know if I got this all from one job, or from several. You don't know if some of it was investment-income. You don't know if my deductions for interest paid, childcare and other stuff was 50K or 150K.
In short, you can make a guesstimate, but the error-bars are large enough to make it fairly useless for salary-estimation. You'd basically end up guessing that my monthly gross in 2009 was probably in the 30K to 45K bracket. This is correct -- but it's much too imprecise to be useful. If your job is similar to mine, you have similar skills, and you're paid 37K - do I earn 7K more than you, 7K less than you -- or the same thing you do ?
But employees should do something different: talk to oneanother, including about salary. Your employer knows everyones salary anyway, it's to his advantage that he has all the knowledge, and you have none of it.
Besides, even the well-paid ones have a interest in seeing the poorly-paid ones catch up, because being well-paid relatively speaking, has an associated risk, namely that the company starts considering you "expensive" relative to the others.
It's truly bizarre connecting sharing or not of a number to what kind of a number it is.
I don't know USA, but here in Norway, you can generally find most cell-numbers the same place you find most regular numbers; http://tlf.no/
It's a choice - when you sign up for a number (either sort!) you get to *choose* if you want to be listed or not. What a concept ! Oh yeah, and precisely the same thing applies if you've got a VoIP line. (there's actually three levels; "listed, unlisted, secret" but that's details.
Furthermore, especially for technical workers like programmers, decent equipment is a must.
It would be idiotic of a company to pay ~$100K/year in salary + $20K in attached costs + ~$10K for offices, training, food and so on -- and then give me sub-standard equipment. An *excellent* computer with two good screens for programming is what ? $2500 ? And has a practical use-time of 2-3 years. That's not even 1% of the total cost.
If I get -1%- more done in a year with good equipment, then it's worth it. Should be a nobrainer.
Some companies don't have brains. It's better not to work for them.
Thing is, even in bad job-markets SOME of your employeers has a choice - typically the most valuable ones.
So pull bullshit like this, and what happens is that those who CAN leave, do it. In other words, the result is that those who are the most valuable in the job-market, typically the ones who are the most valuable to you too, leave.
Good luck with that !