If you want me to work an "emergency" double shift these days then fine but be advised I won't be in tomorrow, and I'm too financially comfortable to be unnerved by a hollow threat of unemployment.
The problem is that many don't have that level of financial comfort, especially if they're young. It is difficult to have that level of comfort if you're married either, unless it is to somebody who also has a fairly high income and without a desire to go spending it.
but it is a scary idea that a $10K missile can take out a $3.5B ship.
There isn't an army anywhere in the world that can't be completely annihilated by about $1k worth of bullets. However, it really isn't much of a concern unless the army just lies in its bunks while the bad buys systematically travel from base to base shooting everybody once in the head.
In order for the missile to take out the ship, it has to hit the ship. For that matter, it also needs to be fired on a course where it will encounter the ship, which isn't a small problem either (this requires knowing about where the target is, getting the launching platform within range, and doing so in a way that doesn't result in the launching platform being destroyed if it isn't expendible).
The modern method of taking down a well defensed naval target is to hide the true number of inbounds and over-saturate the defenses when its beyond the point of making any other defensive efforts. This requires hiding a rather large attack and attacking single targets at a disadvantage to defend themselves.
Doesn't seem terribly practical. So, if your Aegis-equipped destroyer has mechanical problems and has to turn back for Norfolk the bad guys can send a squadron of 100 bombers halfway across the Atlantic (with 50 tankers in support) to launch 300 missiles at it. That seems a bit akin to deploying half the eastern front to go after a single panzer that had a tread break during a retreat.
It's a national security threat. There are antitoxins to regular botulism. This is something else. Maybe readers will like to see a few million dead? Probably. Readers who think all info should be free are fools.
So, why publish it at all? Seems like the real problem is with the academic system that seems to believe that only published research is useful. Published research that doesn't actually publish anything isn't actually useful for anything but bragging...
The touch screen on the car is an unusable piece of crap, unfortunately you are forced to use it for just about everything. Whoever designed it designed it to look cool but not to be useable in a car.
This seems like a really bad trend in modern software/etc. The UIs are made to look cool, not to be functional. The software to control the home thermostat from my phone insists on running full screen, which means that using my password manager is a pain since the status bar is obscured (yes, there are hacks to get around this, but they shouldn't be necessary - really the only things that should consider being full screen are media apps). Cars have fancy touchscreens for functions that would be better accomplished with buttons. Applications and websites stick animation all over the place which looks cute the first 75 times, but becomes a real drag once you're trying to get work done, and so on.
Reminds me of when we were testing tablets at work. They're really cool for the first 20 minutes, and then you start to realize just how long it takes you to get anything done. For certain types of tasks they're GREAT, but as a general-purpose tool not so much (unless you basically turn them into desktops with keyboards/etc).
There is no such thing as a good UI concept or a bad UI concept. There is just fit-for-purpose, and not. Everything has its place.
The problem with this is that there is no guarantee that just because the flight is supposed to be on a 737-800 that it actually will be on a 737-800. They might sub out an A320 at the last minute, or whatever.
I always use Seatguru, but it is only of limited usefulness. Better than nothing...
Re:Yeah, but it does depend on the area of science
on
How Science Goes Wrong
·
· Score: 1
If you have accurately and fully described the requirements of that department, she's better off not getting a degree from such a sham institution. I have never seen requirements like that.
I've never seen them written down like that, but certainly that was my impression of how graduate research works. If you sunk 5 years into a research project and somebody else published basically the same work first, you wasted your time. Likewise if you sink 5 years into a project and basically come up with nothing exciting, you have also wasted your time. There is a lot of incentive to try to create discoveries, instead of discovering them.
Much of this depends on the motives of the researchers. They don't always align with their sponsors either.
Let's take a big pharmaceutical company. If they're doing research on a marketed product they have a HUGE incentive to show some benefit since they sunk lots of money into it and they want to make more money from it. If they're doing basic research in an area where they don't have a product then they probably have wishful thinking, but they really don't have an investment. In fact, smart business should drive them to pursue the truth because otherwise they're going to invest a lot of money in something that will eventually not pan out. However, the actual investigator might be an expert in some technology and they want their company to invest more in that technology so that they get to keep their job, so their own incentive might even be contrary to their sponsor's.
Conflict of interest is really messy, and the less likely the data is to be reproduced, the bigger the problem it is. When the cost of replicating an experiment is low it is less of an issue. Medical research tends to be expensive, and hence it doesn't get replicated much.
The idea that you need to drop $100 to see if there's any likelihood that your kid will eventually contract Alzheimers is ludicrous.
The $100 is for a fairly comprehensive SNP analysis, not just an Alzheimer's test. Also, you have to read a big disclaimer before they'll even tell you whether you have a risk for Alzheimer's for all the reasons that are coming up here. By default they won't tell you anything about it.
Actually, there are MANY doctors who question the value of prostate screening. If you get a positive result, you aren't necessarily going to get cancer, but you WILL end up getting treatment. The treatment is not without its own risks. So you have the low risk of getting cancer and dying on the one hand, and the much higher risk of having other problems as a result of treatment on the other hand.
It isn't a clear-cut decision, and most patients probably aren't really given all the facts. You'd be amazed at how few modern treatments actually have clinical evidence to back them in terms of better outcomes. Sure, somebody who gets a bypass might live 10 more years, but what would have happened if they didn't get the bypass, and how many die from complications/etc? Bypass surgery probably isn't a good example there, but there are many of modern treatments that lack any proof that they actually result in any meaningful improvement of life. That doesn't mean that they don't - often nobody bothers to ask the question.
My understanding is that the course of action if there is a positive screening result is worse on average for a patient than getting prostate cancer, when adjusted for the various risks. The treatment has risks of complications, and the risk of actually getting prostate cancer is low, so you're actually better off ignoring it.
It isn't about ignoring cancer - it is about ignoring a likely false positive when the reaction to the test is likely to cause other problems.
Yup - I remember writing a routine in BASIC back in the 90s to pull data from an A/D card (using INP/OUTP in a loop). The thing would take 10 seconds to read in 128k of samples, even compiled/optimized. I then wrote a function in about 8 lines of assembly language and it took less than 1/28th of a second to read in the array (I didn't bother trying to program a timer chip to measure it more accurately than that), despite having to manipulate the samples to stuff it into a data structure compatible with the 64k segment model of the DOS era. High-level languages/etc can add incredible amounts of overhead.
A modern CPU can run through about 90 million pages (printed) of assembly code per core in a second, assuming no loops. Do you think the average CPU really does that much useful work every second?
Yup. People who think that China is loaning all this money to the US to somehow gain an advantage over it aren't really thinking about things from a common-sense standpoint. If I walk up to you and give you a piece of paper in exchange for hard cash, how is it exactly that you have me over a barrel?
Strategically China's main interest is keeping its currency low compared to the dollar.
Not sure I really agree. If they really were such poor investments, nobody but the Fed would be buying them, and the US buying its own debt would probably not make for a very stable currency.
People buy treasures as an alternative to holding cash, or to manipulate currencies (on the China scale).
The US would never cancel a bond. The US enjoys the lowest interest rates by far of any nation on earth precisely because it doesn't play those games. Treasury bonds are boring - you buy one for $99.878667 and in a year they give you $100.
The US doesn't even set the rates - they're sold at auction. A 1-yr bill sells for $99.88 because somebody was willing to pay that, and not enough people were willing to pay a 100,000th of a cent more, apparently. The fact that people are willing to shell out so much only to make 12 cents in a year tells you how popular they are. Folks were willing to shell out $99.972778 last week for bills that are due in early November - the rates are actually pretty high on those by recent historical standards (0.35%), but clearly they don't reflect a lot of panic over a default.
China is just making noise though. Why do you think they buy T-Bills in the first place? They aren't interested in making low-interest donations to furthering the US government's mission. Their goal is simple - to keep their currency value low relative to the dollar, so that everybody manufactures their widgets there. If they stopped buying T-Bills their currency would rise in value relative to ours, and the price of EVERYTHING in your local Walmart would go up. Walmart doesn't have some kind of love for China - the second their spreadsheets tell them that the goods could be made elsewhere cheaper you'll see a HUGE shift in manufacturing.
But you're not saying we need a warrant to wiretap a known foreign terrorist in another country, right? So what if it turns out he's talking to a US citizen in the US? I don't think the discovery of that information, without a warrant, precludes pursuing the US citizen from then on, with warrants.
Honestly, I'm not a fan of foreign data collected by the NSA without a warrant ever being used in US courts of law.
First, when this happens nobody wants to talk about the collection system used, which means the methods aren't subject to judicial review. How does a jury evaluate whether the evidence was properly collected and preserved if the defendant isn't allowed to ask questions about it?
Second, half the time nobody wants to even share the evidence, since even that is classified. So, you end up with government agents telling the jury that there is data linking the defendant to terrorists, and they'll just have to take their word on it.
The problem with terrorism is that it blurs the line between warfare and justice. In warfare the targets aren't people, but states - the people just happen to be in the way. In justice the targets are definitely people, and states sometimes just happen to get in the way. The methods developed by the NSA were originally intended for warfare with other states - when we tapped Russian undersea cables they weren't interested in catching criminals - they wanted data on the Russian government.
Presumably the email account and/or client could/would have the email, and email servers along the path would have a record of the message.
If he was under surveillance they would probably be capturing the email.
Since the jury has access to none of those servers, I'm not sure how that helps.
If the government were to obtain a warrant to access them and present their methods as well as the evidence in court, allowing both to be subject to cross-examination, then that would help. The issue isn't that the government isn't able to get the email - they just don't want to disclose how they did it.
Mencken notwithstanding, it is not defending freedom to claim that phone conversations with foreign terrorists are free speech, or deserve privacy.
Regardless, they certainly should require a warrant to intercept - there is no reason that one couldn't at least be required within 24 hours of the start of interception (it could even be retroactive).
If somebody has a phone conversation with a know terrorist that sounds like probably cause. If there is probable cause you can get a warrant.
That's my biggest beef with these intelligence dragnets - they rely on collecting huge volumes of data from innocent parties to try to spot people who might be suspicious. The way the 4th amendment is written you need to first establish who is suspicious and then target them for search/seizure.
We know (now) how the evidence was collected. We also know that most of it was collected without probable cause. The issue isn't the method of collection, but the justification for it.
Actually, the method is also important. If the government claims they intercepted an email sent by the defendant, how does a jury know whether he actually sent that email if the defense can't subject the government's methods to scrutiny. For all they know the government mixed up the email addresses/etc.
You can't just present evidence against somebody in court without defending the methods under which the evidence was collected.
If they wanted a ring they could just buy Merck's headquarters - they're trying to sell it, actually. Granted, North Jersey isn't all that much better than California, but traffic-wise it has to be at least a moderate improvement.
In fact, Merck is trying to sell the place (stock not doing quite as well as AAPL). I'm sure Apple could clean up the place for considerably less than $5B.
If you're not comfortable with this, then don't connect the account to the service. Period.
Why does it need to be this way? Why not give the user granular access to permissions? Platforms like Twitter/Android/etc give way too much control to apps and not enough to the user - the user shouldn't be given all-or-nothing choices like this.
They didn't "ask" for permission. They inferred it from people providing their twitter account info. There wasn't even an "opt-out" option because people didn't know this was going to happen.
When you grant a third party access to sent Tweets on your behalf, don't you click through a warning telling you that? Why would you give a convention permission to send Tweets as you, and if you do, why would you be surprised when they do?
The problem is that there is a growing trend towards letting apps request permissions, and then giving the user two choices - accept all the permissions the app requests, or don't use the app at all. That is true of many online services, and it is true of Android as well (and likely other mobile OSes).
The better solution is to allow the application to request a default list of permissions, and then give the user the opportunity to accept or modify them. The application would still work if the permissions are modified, though with limited functionality. I'd probably go a step further and not make it possible for the application to know what permissions were granted, so that app authors don't just force the all-or-nothing situation back on users by refusing to run if full permissions are not granted. 99% of the time partial permissions only cause failure modes that the application has to handle gracefully anyway (no access to contacts is no different than a user who has no contacts, no access to location/network is no different than a user in a building, etc).
The all-or-nothing approach just gives app authors a club to hit users with - it puts the app author in control of the device, and not the user. Not running mobile apps really isn't an acceptable alternative.
Fair enough - I live about 75 miles somewhat west of you most likely, and while I don't mind the occasional foray into North Jersey I definitely would not want to live there.
Maybe 10min from a farm is stretching it, but the suburbs of many cities in the US are not nearly so congested as the NYC metro area. I also generally prefer the outer suburbs - not the areas just outside the city limits.
While I usually work from home, if I need to drive into the office it only takes 20min during rush hour. Even a commute of that length is annoying, but it is nothing compared to my coworkers who live in Jersey.
Obviously all of this is a matter of preference. Many of my coworkers doubtless consider me a hick.:)
Slashdot Mirror
If you want me to work an "emergency" double shift these days then fine but be advised I won't be in tomorrow, and I'm too financially comfortable to be unnerved by a hollow threat of unemployment.
The problem is that many don't have that level of financial comfort, especially if they're young. It is difficult to have that level of comfort if you're married either, unless it is to somebody who also has a fairly high income and without a desire to go spending it.
but it is a scary idea that a $10K missile can take out a $3.5B ship.
There isn't an army anywhere in the world that can't be completely annihilated by about $1k worth of bullets. However, it really isn't much of a concern unless the army just lies in its bunks while the bad buys systematically travel from base to base shooting everybody once in the head.
In order for the missile to take out the ship, it has to hit the ship. For that matter, it also needs to be fired on a course where it will encounter the ship, which isn't a small problem either (this requires knowing about where the target is, getting the launching platform within range, and doing so in a way that doesn't result in the launching platform being destroyed if it isn't expendible).
The modern method of taking down a well defensed naval target is to hide the true number of inbounds and over-saturate the defenses when its beyond the point of making any other defensive efforts. This requires hiding a rather large attack and attacking single targets at a disadvantage to defend themselves.
Doesn't seem terribly practical. So, if your Aegis-equipped destroyer has mechanical problems and has to turn back for Norfolk the bad guys can send a squadron of 100 bombers halfway across the Atlantic (with 50 tankers in support) to launch 300 missiles at it. That seems a bit akin to deploying half the eastern front to go after a single panzer that had a tread break during a retreat.
It's a national security threat. There are antitoxins to regular botulism. This is something else. Maybe readers will like to see a few million dead? Probably. Readers who think all info should be free are fools.
So, why publish it at all? Seems like the real problem is with the academic system that seems to believe that only published research is useful. Published research that doesn't actually publish anything isn't actually useful for anything but bragging...
The touch screen on the car is an unusable piece of crap, unfortunately you are forced to use it for just about everything. Whoever designed it designed it to look cool but not to be useable in a car.
This seems like a really bad trend in modern software/etc. The UIs are made to look cool, not to be functional. The software to control the home thermostat from my phone insists on running full screen, which means that using my password manager is a pain since the status bar is obscured (yes, there are hacks to get around this, but they shouldn't be necessary - really the only things that should consider being full screen are media apps). Cars have fancy touchscreens for functions that would be better accomplished with buttons. Applications and websites stick animation all over the place which looks cute the first 75 times, but becomes a real drag once you're trying to get work done, and so on.
Reminds me of when we were testing tablets at work. They're really cool for the first 20 minutes, and then you start to realize just how long it takes you to get anything done. For certain types of tasks they're GREAT, but as a general-purpose tool not so much (unless you basically turn them into desktops with keyboards/etc).
There is no such thing as a good UI concept or a bad UI concept. There is just fit-for-purpose, and not. Everything has its place.
The problem with this is that there is no guarantee that just because the flight is supposed to be on a 737-800 that it actually will be on a 737-800. They might sub out an A320 at the last minute, or whatever.
I always use Seatguru, but it is only of limited usefulness. Better than nothing...
$500? Try $9000 on long-haul flights.
If you have accurately and fully described the requirements of that department, she's better off not getting a degree from such a sham institution. I have never seen requirements like that.
I've never seen them written down like that, but certainly that was my impression of how graduate research works. If you sunk 5 years into a research project and somebody else published basically the same work first, you wasted your time. Likewise if you sink 5 years into a project and basically come up with nothing exciting, you have also wasted your time. There is a lot of incentive to try to create discoveries, instead of discovering them.
Much of this depends on the motives of the researchers. They don't always align with their sponsors either.
Let's take a big pharmaceutical company. If they're doing research on a marketed product they have a HUGE incentive to show some benefit since they sunk lots of money into it and they want to make more money from it. If they're doing basic research in an area where they don't have a product then they probably have wishful thinking, but they really don't have an investment. In fact, smart business should drive them to pursue the truth because otherwise they're going to invest a lot of money in something that will eventually not pan out. However, the actual investigator might be an expert in some technology and they want their company to invest more in that technology so that they get to keep their job, so their own incentive might even be contrary to their sponsor's.
Conflict of interest is really messy, and the less likely the data is to be reproduced, the bigger the problem it is. When the cost of replicating an experiment is low it is less of an issue. Medical research tends to be expensive, and hence it doesn't get replicated much.
The idea that you need to drop $100 to see if there's any likelihood that your kid will eventually contract Alzheimers is ludicrous.
The $100 is for a fairly comprehensive SNP analysis, not just an Alzheimer's test. Also, you have to read a big disclaimer before they'll even tell you whether you have a risk for Alzheimer's for all the reasons that are coming up here. By default they won't tell you anything about it.
1. your doctor is an idiot
Actually, there are MANY doctors who question the value of prostate screening. If you get a positive result, you aren't necessarily going to get cancer, but you WILL end up getting treatment. The treatment is not without its own risks. So you have the low risk of getting cancer and dying on the one hand, and the much higher risk of having other problems as a result of treatment on the other hand.
It isn't a clear-cut decision, and most patients probably aren't really given all the facts. You'd be amazed at how few modern treatments actually have clinical evidence to back them in terms of better outcomes. Sure, somebody who gets a bypass might live 10 more years, but what would have happened if they didn't get the bypass, and how many die from complications/etc? Bypass surgery probably isn't a good example there, but there are many of modern treatments that lack any proof that they actually result in any meaningful improvement of life. That doesn't mean that they don't - often nobody bothers to ask the question.
My understanding is that the course of action if there is a positive screening result is worse on average for a patient than getting prostate cancer, when adjusted for the various risks. The treatment has risks of complications, and the risk of actually getting prostate cancer is low, so you're actually better off ignoring it.
It isn't about ignoring cancer - it is about ignoring a likely false positive when the reaction to the test is likely to cause other problems.
Yup - I remember writing a routine in BASIC back in the 90s to pull data from an A/D card (using INP/OUTP in a loop). The thing would take 10 seconds to read in 128k of samples, even compiled/optimized. I then wrote a function in about 8 lines of assembly language and it took less than 1/28th of a second to read in the array (I didn't bother trying to program a timer chip to measure it more accurately than that), despite having to manipulate the samples to stuff it into a data structure compatible with the 64k segment model of the DOS era. High-level languages/etc can add incredible amounts of overhead.
A modern CPU can run through about 90 million pages (printed) of assembly code per core in a second, assuming no loops. Do you think the average CPU really does that much useful work every second?
Yup. People who think that China is loaning all this money to the US to somehow gain an advantage over it aren't really thinking about things from a common-sense standpoint. If I walk up to you and give you a piece of paper in exchange for hard cash, how is it exactly that you have me over a barrel?
Strategically China's main interest is keeping its currency low compared to the dollar.
Not sure I really agree. If they really were such poor investments, nobody but the Fed would be buying them, and the US buying its own debt would probably not make for a very stable currency.
People buy treasures as an alternative to holding cash, or to manipulate currencies (on the China scale).
The US would never cancel a bond. The US enjoys the lowest interest rates by far of any nation on earth precisely because it doesn't play those games. Treasury bonds are boring - you buy one for $99.878667 and in a year they give you $100.
The US doesn't even set the rates - they're sold at auction. A 1-yr bill sells for $99.88 because somebody was willing to pay that, and not enough people were willing to pay a 100,000th of a cent more, apparently. The fact that people are willing to shell out so much only to make 12 cents in a year tells you how popular they are. Folks were willing to shell out $99.972778 last week for bills that are due in early November - the rates are actually pretty high on those by recent historical standards (0.35%), but clearly they don't reflect a lot of panic over a default.
China is just making noise though. Why do you think they buy T-Bills in the first place? They aren't interested in making low-interest donations to furthering the US government's mission. Their goal is simple - to keep their currency value low relative to the dollar, so that everybody manufactures their widgets there. If they stopped buying T-Bills their currency would rise in value relative to ours, and the price of EVERYTHING in your local Walmart would go up. Walmart doesn't have some kind of love for China - the second their spreadsheets tell them that the goods could be made elsewhere cheaper you'll see a HUGE shift in manufacturing.
But you're not saying we need a warrant to wiretap a known foreign terrorist in another country, right? So what if it turns out he's talking to a US citizen in the US? I don't think the discovery of that information, without a warrant, precludes pursuing the US citizen from then on, with warrants.
Honestly, I'm not a fan of foreign data collected by the NSA without a warrant ever being used in US courts of law.
First, when this happens nobody wants to talk about the collection system used, which means the methods aren't subject to judicial review. How does a jury evaluate whether the evidence was properly collected and preserved if the defendant isn't allowed to ask questions about it?
Second, half the time nobody wants to even share the evidence, since even that is classified. So, you end up with government agents telling the jury that there is data linking the defendant to terrorists, and they'll just have to take their word on it.
The problem with terrorism is that it blurs the line between warfare and justice. In warfare the targets aren't people, but states - the people just happen to be in the way. In justice the targets are definitely people, and states sometimes just happen to get in the way. The methods developed by the NSA were originally intended for warfare with other states - when we tapped Russian undersea cables they weren't interested in catching criminals - they wanted data on the Russian government.
Presumably the email account and/or client could/would have the email, and email servers along the path would have a record of the message.
If he was under surveillance they would probably be capturing the email.
Since the jury has access to none of those servers, I'm not sure how that helps.
If the government were to obtain a warrant to access them and present their methods as well as the evidence in court, allowing both to be subject to cross-examination, then that would help. The issue isn't that the government isn't able to get the email - they just don't want to disclose how they did it.
Mencken notwithstanding, it is not defending freedom to claim that phone conversations with foreign terrorists are free speech, or deserve privacy.
Regardless, they certainly should require a warrant to intercept - there is no reason that one couldn't at least be required within 24 hours of the start of interception (it could even be retroactive).
If somebody has a phone conversation with a know terrorist that sounds like probably cause. If there is probable cause you can get a warrant.
That's my biggest beef with these intelligence dragnets - they rely on collecting huge volumes of data from innocent parties to try to spot people who might be suspicious. The way the 4th amendment is written you need to first establish who is suspicious and then target them for search/seizure.
We know (now) how the evidence was collected. We also know that most of it was collected without probable cause. The issue isn't the method of collection, but the justification for it.
Actually, the method is also important. If the government claims they intercepted an email sent by the defendant, how does a jury know whether he actually sent that email if the defense can't subject the government's methods to scrutiny. For all they know the government mixed up the email addresses/etc.
You can't just present evidence against somebody in court without defending the methods under which the evidence was collected.
If they wanted a ring they could just buy Merck's headquarters - they're trying to sell it, actually. Granted, North Jersey isn't all that much better than California, but traffic-wise it has to be at least a moderate improvement.
Looks like Merck as well.
In fact, Merck is trying to sell the place (stock not doing quite as well as AAPL). I'm sure Apple could clean up the place for considerably less than $5B.
If you're not comfortable with this, then don't connect the account to the service. Period.
Why does it need to be this way? Why not give the user granular access to permissions? Platforms like Twitter/Android/etc give way too much control to apps and not enough to the user - the user shouldn't be given all-or-nothing choices like this.
They didn't "ask" for permission. They inferred it from people providing their twitter account info. There wasn't even an "opt-out" option because people didn't know this was going to happen.
When you grant a third party access to sent Tweets on your behalf, don't you click through a warning telling you that? Why would you give a convention permission to send Tweets as you, and if you do, why would you be surprised when they do?
The problem is that there is a growing trend towards letting apps request permissions, and then giving the user two choices - accept all the permissions the app requests, or don't use the app at all. That is true of many online services, and it is true of Android as well (and likely other mobile OSes).
The better solution is to allow the application to request a default list of permissions, and then give the user the opportunity to accept or modify them. The application would still work if the permissions are modified, though with limited functionality. I'd probably go a step further and not make it possible for the application to know what permissions were granted, so that app authors don't just force the all-or-nothing situation back on users by refusing to run if full permissions are not granted. 99% of the time partial permissions only cause failure modes that the application has to handle gracefully anyway (no access to contacts is no different than a user who has no contacts, no access to location/network is no different than a user in a building, etc).
The all-or-nothing approach just gives app authors a club to hit users with - it puts the app author in control of the device, and not the user. Not running mobile apps really isn't an acceptable alternative.
Fair enough - I live about 75 miles somewhat west of you most likely, and while I don't mind the occasional foray into North Jersey I definitely would not want to live there.
Maybe 10min from a farm is stretching it, but the suburbs of many cities in the US are not nearly so congested as the NYC metro area. I also generally prefer the outer suburbs - not the areas just outside the city limits.
While I usually work from home, if I need to drive into the office it only takes 20min during rush hour. Even a commute of that length is annoying, but it is nothing compared to my coworkers who live in Jersey.
Obviously all of this is a matter of preference. Many of my coworkers doubtless consider me a hick. :)