The oldest file that I don't need to restore (but can obviously restore in case my disk crashes) dates from April 1, 1992.
The oldest file that I can restore easily, will most likely be from the summer of 1989. That would be the original source file of a public domain program that I wrote at that time (released first in the early nineties, but I have the complete history all the way back to 1989 on backup). But I also might still have a LaTeX copy of my master's thesis from earlier that year hidden somewhere.
The oldest file that I can restore at all, must date from about 1987. This, however, requires me to dig up a working DOS machine (doable, because my father still has one of those dinosaurs - now unused, but still booting if needed).
Well,... I learned to read when entering school at age 6 - the usual age here in Belgium back in the early 70s. But within a few months I thought my sister how to read as well. She was only 4 at that time
I'm an adult (45) and I send and receive tens of SMS per day from myself to myself as part of a product testing effort that I'm involved in. Does that count as well?;-)
Actually, the Belgian Order of the Crown is a nice example of what I meant. I'm a knight in that order myself, but that does not imply a "title of nobility".
For clarity for those who do not get it yet: The word "knight" has a double meaning. When used on its own, it is indeed a title of nobility and in some countries - such as Belgium - each year a few people are still made knights or higher (mostly with a non-hereditary title) as a sign of recognition for exceptional services. When used in an expression such as "Knight of ", it is nothing more than one of several ranks that exist withing that order, but does not imply nobility. And as I mentioned in my earlier mail, these "knight of..." awards are much easier to get than real nobility titles.
For starters, he's a German national and hence the US Constitution is not relevant.
Next, he was not granted a title of nobility. He was given a medal that in other certain countries would rank equivalent to the knight level in a typical order of chivalry.
On top of that, being a member of an order of chivalry is not equivalent to being a member of the nobility. I should know, as I'm myself a knight in a Belgian order of chivalry. Just about any Belgian army officer (reserves included) who has been active for long enough is a knight in one or more such orders. That does not make them nobility.
If you consider unknown Belgian awards to be too obscure to be a reference, look at the famous "Knights Cross of the Iron Cross", so well known that it is often just referred to as "The Knights Cross". This award - in its various gradations - is very well known for being the most desirable award anyone in the WWII German armed forces could receive. But rest assured that winning one did not imply any title of nobility being granted by mr. H. who, in fact, despised the old German nobility. Besides, from a legal point of view nobility was abolished in Germany in 1919.
Because some problems interact. For instance because they affect the same code modules and fixing them one by one would actually be require more work overall - possibly involving additional throwaway temporary work. This could even delay getting them both fixed compared to fixing them in one go.
And on top of that, other GPS-like systems are being built. Yes, Galileo has been delayed, but it will eventually be launched nonetheless. And it's not the only one.
That is correct. And on top of that, recent satellites no longer have the technical capability to implement selective availability anyway (see http://www.defense.gov/releases/release.aspx?releaseid=11335). It is no longer needed for anything after having been turned off several years ago. That's because civilian users had developed multiple techniques - e.g. differential GPS - to get better accuracy even when back SA was still on. In short, SA is dead and buried forever.
Actually, MRSA does have something to do with cleanliness of hospitals. But not in the way one might think: one of the ways to keep MRSA under control, is to make sure it is not the only type of bacteria that are around.
For more information, see: http://www.ultimatewash.co.uk/pip/PIP%20Press%20Release.pdf, which describes an experiment done with a cleaning agent that, instead of trying to kill all bacteria, actually introduces new ones that act as competition to the bad ones. This approach is, in fact, nothing more than a targeted "formalisation" of one of the things the Norwegian hospitals do: keeping MSRA low by making sure it's not the only strand that is competing for all the nutrients.
Indeed, things happened sooner than Jan 1, 2000 and they also happened at the stroke of midnight. I encountered my first unexpected Y2K bug (I'd already fixed several ones that we knew of in our own systems) a few minutes after midnight in Jan 1, 1999. More in particular, SCCS on HP-UX was unable to check in a file after midnight on that day because for some reason that I never understood it calculated a date one year into the future while doing so. Fortunately, HP already had done their homework as well and they had an update readily available.
Dev with root (unix): yes, because - with very few exceptions - development on a unix system does not require root; Dev with local admin (windows): no, because without it almost nothing can get done efficiently.
That's actually how things were in my previous company. In the very beginning (think before 1990) there were some devs with root, basically because they were also part-time admins. As the unix community grew, all those devs did the wise thing and refused to retain their root privileges: they did not need them and did not want to risk breaking things. Many moons later, Windows entered the picture and all devs had to be given local admin, even the most junior ones. Even so, on the Linux machines that we also introduced nobody outside IT had root. As a Linux dev (at the time) who had an excellent relationship with the IT group and did a lot of admin work for them, I never had root (even if IT once told me they'd trust me with it). Today, I'm a project manager (not a dev!) in an other company, and guess what: I need local admin on my Windows laptop in order to actually get my work done. (OK, I'm not one of those "manage by excel" managers, but even so...)
It would still be 150 on *our* scale, which they are using because it is the only way for today's human audience to get some sort of feeling for what it means. (Note: I was diagnosed with 150 on our scale as well. If I were to use myself and my even smarter "twin brother" as the defining reference population, I'd sure be close to 100. But by that scale most of the population would be far below 100. Willing to used to that?;-))
My real problem with this summary/story, is that it uses little more than brain size to estimate something that is much more complex than the primitive "bigger is better" idea. Quoting the actual story: "if brain size accounts for just 10 to 20 percent of an IQ test score... we can readily calculate... 149". Sounds like a lot of unfounded conjecture to me. The real key to IQ is in the connections, of which we can know nothing at all.
Actually, I forgot a fifth problem: because the two antennae are only 379m apart, they would both face the same atmospheric/ionospheric errors. So it is highly unlikely that the antenna on the bow of the ship would see a worst case error in one direction and the one on the stern would see a worst case error in exactly the opposite direction. More likely, they would both be offset by roughly the same distance in roughly the same direction. This means that while the reported position of each might be off be a few meters, the ship's heading calculated from comparing both positions would be much closer to the true heading.
The first "problem" with your calculation is that you are assuming a 10m worst case error, while for the kind of application you describe more expensive and accurate (e.g. dual-frequency) receivers would be used. Also, SBAS would be used whenever available, making the worst case errors even smaller and less likely to occur (good DGPS systems can reach 0.1m accuracy). Not to mention that at sea one always has open sky conditions, so no canopy or multipath problems (provided the antennae are mounted in a suitable location on the ship, obviously).
The second problem is that you calculate the average error as being half of your calculated maximum error. This is wrong in two ways: first, your maximum error calculation is symmetrical, meaning that even by your crude method of averaging, the average would be 0; second, the calculation of the average error needs to take the probability of each value into account and the smaller errors are more likely than the extreme ones.
The third problem is that you assume the worst case error to be constant over the duration of that "voyage of thousands of kilometers", which is totally unrealistic. A real-world GPS system maintains a history of past positions and uses this to correct for instantaneous errors.
The fourth problem is that in the application you describe, the two antennae are located at a known and unchangeable distance from each other. This means that their physical movements are correlated and hence that the system has extra information at its disposal that it can use to detect physically impossible combinations of positional errors. The more extreme the errors, the more likely they are to be flagged as being suspect. This again can be used as input to the GPS system's calculations.
Anyway, no sailor worth his or her salt relies on a single source of positioning data. Certainly not for a "voyage of thousands of kilometers".
One common misconception is that the guys who blow up planes are mindless idiots not able to figure anything out. This is very wrong. Firstly, while they may be mindless religious zealots, that does not mean they can't think at all (just one example: Mohammed Atta was studying towards a PhD in engineering; Yasser Arafat - to name just one other example - also was an engineer). Secondly, and much more importantly, the guys that plot how to blow up planes are smart enough not to be on board when the bang occurs.
The biggest mistake one can make in warfare is to underestimate the opposing party.
Quoting you: "No liquids is a good policy", I'd say: Indeed, Except that "no liquids" is not at all practical (and hence not good after all). So the replacement policy is "at most 100ml", which is as bad a policy as any, for the exact reason you mention: if I need 500ml in total in order to blow up the plane, all need to do is to arrange for 5 suckers to board it. It's not like that hasn't been done before either.
Your point is very valid, but it does not take away the fact that in real life the grandparent poster is way too often right. In many/most companies, a middle manager's job is exactly what he describes: make the impossible happen with not enough people and/or resources and without real formal authority to change those things. Yes this is stupid. So are most big companies.
Yes, they are tracking IP addresses and mixing them with other sources of info.
I've disabled Google cookies on all computers that I use (and I delete unwanted cookies from just about any site on an automatic basis once a week), but even so I've observed a bias in search & advertisement results "leaking" from my private machines to my office one. That's despite having a clear policy that non-work stuff on the office machine is limited to the absolute minimum and that the exceptions to that rule are limited to occasionally viewing a very small list of sites that cannot possibly reveal the information needed to generate the biases that I've observed.
One simple question regarding spelling/grammar mistakes: What about non-native speakers? Is their code bad just because he or she tends to write "recieve" instead of "receive" (just one example mistake that is quite common for native Dutch speakers due to bias by his/her mother tongue)? I would hope not... TFA goes as far as to claim that non-natives actually tend to have better English language skills. First of all, I'm not convinced of that as a general rule. But even if it were true, there will always be traps for even those people to fall into unknowingly. What's more, I myself most certainly know about "receive", but even so I get it wrong at least 75% of the time. I also know that due to "selective blindness" reading the comment afterwards is not enough to catch all of those.
At the other extreme: Back when I was still writing code on a daily basis, I tended to spell check it every so often and fix any mistakes found. Talk about bitf*cking...:-) But I fail to see why wasting time on doing that would be a guarantee of proper design.
Slashdot Mirror
The oldest file that I don't need to restore (but can obviously restore in case my disk crashes) dates from April 1, 1992.
The oldest file that I can restore easily, will most likely be from the summer of 1989. That would be the original source file of a public domain program that I wrote at that time (released first in the early nineties, but I have the complete history all the way back to 1989 on backup). But I also might still have a LaTeX copy of my master's thesis from earlier that year hidden somewhere.
The oldest file that I can restore at all, must date from about 1987. This, however, requires me to dig up a working DOS machine (doable, because my father still has one of those dinosaurs - now unused, but still booting if needed).
So what? Lots of people have had operations. Are you going to open up all of them on the spot, just to see what's inside?
Well,... I learned to read when entering school at age 6 - the usual age here in Belgium back in the early 70s. But within a few months I thought my sister how to read as well. She was only 4 at that time
I'm an adult (45) and I send and receive tens of SMS per day from myself to myself as part of a product testing effort that I'm involved in. Does that count as well? ;-)
Nah. Only 3 digits matter. Especially those below 510.
Easy & simple, you say? It shows that you don't have a girlfriend... ;-)
Actually, the Belgian Order of the Crown is a nice example of what I meant. I'm a knight in that order myself, but that does not imply a "title of nobility".
For clarity for those who do not get it yet: The word "knight" has a double meaning. When used on its own, it is indeed a title of nobility and in some countries - such as Belgium - each year a few people are still made knights or higher (mostly with a non-hereditary title) as a sign of recognition for exceptional services. When used in an expression such as "Knight of ", it is nothing more than one of several ranks that exist withing that order, but does not imply nobility. And as I mentioned in my earlier mail, these "knight of ..." awards are much easier to get than real nobility titles.
For starters, he's a German national and hence the US Constitution is not relevant.
Next, he was not granted a title of nobility. He was given a medal that in other certain countries would rank equivalent to the knight level in a typical order of chivalry.
On top of that, being a member of an order of chivalry is not equivalent to being a member of the nobility. I should know, as I'm myself a knight in a Belgian order of chivalry. Just about any Belgian army officer (reserves included) who has been active for long enough is a knight in one or more such orders. That does not make them nobility.
If you consider unknown Belgian awards to be too obscure to be a reference, look at the famous "Knights Cross of the Iron Cross", so well known that it is often just referred to as "The Knights Cross". This award - in its various gradations - is very well known for being the most desirable award anyone in the WWII German armed forces could receive. But rest assured that winning one did not imply any title of nobility being granted by mr. H. who, in fact, despised the old German nobility. Besides, from a legal point of view nobility was abolished in Germany in 1919.
Because some problems interact. For instance because they affect the same code modules and fixing them one by one would actually be require more work overall - possibly involving additional throwaway temporary work. This could even delay getting them both fixed compared to fixing them in one go.
And on top of that, other GPS-like systems are being built. Yes, Galileo has been delayed, but it will eventually be launched nonetheless. And it's not the only one.
That is correct. And on top of that, recent satellites no longer have the technical capability to implement selective availability anyway (see http://www.defense.gov/releases/release.aspx?releaseid=11335). It is no longer needed for anything after having been turned off several years ago. That's because civilian users had developed multiple techniques - e.g. differential GPS - to get better accuracy even when back SA was still on. In short, SA is dead and buried forever.
Actually, MRSA does have something to do with cleanliness of hospitals. But not in the way one might think: one of the ways to keep MRSA under control, is to make sure it is not the only type of bacteria that are around.
For more information, see: http://www.ultimatewash.co.uk/pip/PIP%20Press%20Release.pdf, which describes an experiment done with a cleaning agent that, instead of trying to kill all bacteria, actually introduces new ones that act as competition to the bad ones. This approach is, in fact, nothing more than a targeted "formalisation" of one of the things the Norwegian hospitals do: keeping MSRA low by making sure it's not the only strand that is competing for all the nutrients.
Indeed, things happened sooner than Jan 1, 2000 and they also happened at the stroke of midnight. I encountered my first unexpected Y2K bug (I'd already fixed several ones that we knew of in our own systems) a few minutes after midnight in Jan 1, 1999. More in particular, SCCS on HP-UX was unable to check in a file after midnight on that day because for some reason that I never understood it calculated a date one year into the future while doing so. Fortunately, HP already had done their homework as well and they had an update readily available.
Dev with root (unix): yes, because - with very few exceptions - development on a unix system does not require root; Dev with local admin (windows): no, because without it almost nothing can get done efficiently.
That's actually how things were in my previous company. In the very beginning (think before 1990) there were some devs with root, basically because they were also part-time admins. As the unix community grew, all those devs did the wise thing and refused to retain their root privileges: they did not need them and did not want to risk breaking things. Many moons later, Windows entered the picture and all devs had to be given local admin, even the most junior ones. Even so, on the Linux machines that we also introduced nobody outside IT had root. As a Linux dev (at the time) who had an excellent relationship with the IT group and did a lot of admin work for them, I never had root (even if IT once told me they'd trust me with it). Today, I'm a project manager (not a dev!) in an other company, and guess what: I need local admin on my Windows laptop in order to actually get my work done. (OK, I'm not one of those "manage by excel" managers, but even so...)
It would still be 150 on *our* scale, which they are using because it is the only way for today's human audience to get some sort of feeling for what it means. (Note: I was diagnosed with 150 on our scale as well. If I were to use myself and my even smarter "twin brother" as the defining reference population, I'd sure be close to 100. But by that scale most of the population would be far below 100. Willing to used to that? ;-))
My real problem with this summary/story, is that it uses little more than brain size to estimate something that is much more complex than the primitive "bigger is better" idea. Quoting the actual story: "if brain size accounts for just 10 to 20 percent of an IQ test score ... we can readily calculate ... 149". Sounds like a lot of unfounded conjecture to me. The real key to IQ is in the connections, of which we can know nothing at all.
Actually, I forgot a fifth problem: because the two antennae are only 379m apart, they would both face the same atmospheric/ionospheric errors. So it is highly unlikely that the antenna on the bow of the ship would see a worst case error in one direction and the one on the stern would see a worst case error in exactly the opposite direction. More likely, they would both be offset by roughly the same distance in roughly the same direction. This means that while the reported position of each might be off be a few meters, the ship's heading calculated from comparing both positions would be much closer to the true heading.
There, fixed that for you.
The first "problem" with your calculation is that you are assuming a 10m worst case error, while for the kind of application you describe more expensive and accurate (e.g. dual-frequency) receivers would be used. Also, SBAS would be used whenever available, making the worst case errors even smaller and less likely to occur (good DGPS systems can reach 0.1m accuracy). Not to mention that at sea one always has open sky conditions, so no canopy or multipath problems (provided the antennae are mounted in a suitable location on the ship, obviously).
The second problem is that you calculate the average error as being half of your calculated maximum error. This is wrong in two ways: first, your maximum error calculation is symmetrical, meaning that even by your crude method of averaging, the average would be 0; second, the calculation of the average error needs to take the probability of each value into account and the smaller errors are more likely than the extreme ones.
The third problem is that you assume the worst case error to be constant over the duration of that "voyage of thousands of kilometers", which is totally unrealistic. A real-world GPS system maintains a history of past positions and uses this to correct for instantaneous errors.
The fourth problem is that in the application you describe, the two antennae are located at a known and unchangeable distance from each other. This means that their physical movements are correlated and hence that the system has extra information at its disposal that it can use to detect physically impossible combinations of positional errors. The more extreme the errors, the more likely they are to be flagged as being suspect. This again can be used as input to the GPS system's calculations.
Anyway, no sailor worth his or her salt relies on a single source of positioning data. Certainly not for a "voyage of thousands of kilometers".
See my other post elsewhere in this discussion: http://slashdot.org/comments.pl?sid=1490372&cid=30562402
One common misconception is that the guys who blow up planes are mindless idiots not able to figure anything out. This is very wrong. Firstly, while they may be mindless religious zealots, that does not mean they can't think at all (just one example: Mohammed Atta was studying towards a PhD in engineering; Yasser Arafat - to name just one other example - also was an engineer). Secondly, and much more importantly, the guys that plot how to blow up planes are smart enough not to be on board when the bang occurs.
The biggest mistake one can make in warfare is to underestimate the opposing party.
Quoting you: "No liquids is a good policy", I'd say: Indeed, Except that "no liquids" is not at all practical (and hence not good after all). So the replacement policy is "at most 100ml", which is as bad a policy as any, for the exact reason you mention: if I need 500ml in total in order to blow up the plane, all need to do is to arrange for 5 suckers to board it. It's not like that hasn't been done before either.
And most importantly, you are STILL persecuting a group of people, the vast, VAST majority of whom are innocent, in an attempt to find a few bad guys.
Which is exactly one of the things the real terrorists want, as it fuels their cause/business.
Your point is very valid, but it does not take away the fact that in real life the grandparent poster is way too often right. In many/most companies, a middle manager's job is exactly what he describes: make the impossible happen with not enough people and/or resources and without real formal authority to change those things. Yes this is stupid. So are most big companies.
Yes, they are tracking IP addresses and mixing them with other sources of info.
I've disabled Google cookies on all computers that I use (and I delete unwanted cookies from just about any site on an automatic basis once a week), but even so I've observed a bias in search & advertisement results "leaking" from my private machines to my office one. That's despite having a clear policy that non-work stuff on the office machine is limited to the absolute minimum and that the exceptions to that rule are limited to occasionally viewing a very small list of sites that cannot possibly reveal the information needed to generate the biases that I've observed.
One simple question regarding spelling/grammar mistakes: What about non-native speakers? Is their code bad just because he or she tends to write "recieve" instead of "receive" (just one example mistake that is quite common for native Dutch speakers due to bias by his/her mother tongue)? I would hope not... TFA goes as far as to claim that non-natives actually tend to have better English language skills. First of all, I'm not convinced of that as a general rule. But even if it were true, there will always be traps for even those people to fall into unknowingly. What's more, I myself most certainly know about "receive", but even so I get it wrong at least 75% of the time. I also know that due to "selective blindness" reading the comment afterwards is not enough to catch all of those.
At the other extreme: Back when I was still writing code on a daily basis, I tended to spell check it every so often and fix any mistakes found. Talk about bitf*cking... :-) But I fail to see why wasting time on doing that would be a guarantee of proper design.