Also, I suspect with a high degree of certainty that your available credit far exceeds what you are capable of paying off in one month
Not really. The available credit is about equal to one months paycheck.
Sooner or later you'll use it to finance something because it's the most convenient option
I don't think so. I've never felt the urge to spend more than I can easily afford, and I doubt that will ever change.
The bank still gets paid through the credit card fee though. I'll admit I keep the credit card because it's convenient to use it for renting a car, booking hotels, or ordering stuff on-line. But at least the fee is constant, and I consider it worth it.
Great plan, as long as you don't mind that most people wouldn't survive a full blown revolution. Our society has become too fragile to keep the daily flow of necessities going in the middle of a revolution.
This assumes the tech is reliable enough that it can be made to work in a comfortable (loose hanging) shirt, that it has a good power source that will last for a long time, and that it can go through a washer and dryer without breaking, and that it's cheap enough that I can buy a wardrobe full of clothes that all have this technology.
I would think a device that's worn like a wristwatch has a better chance of working.
It depends on the specifics of the case. If somebody owns the code, or has been granted a special license that allows closed source development, the GPL doesn't apply to their fork, so they can keep developing it, without any obligation to put their changes also in the GPL version.
If they take patches from the GPL tree, and apply it to their own (closed source) tree, it depends on the agreements. If contributors to the project agree to transfer the copyright to them, or they allow patches to be copied into the closed tree, I don't see a problem.
If a contributor doesn't transfer the copyright, and doesn't grant the owner of the closed fork a license to copy, it would not be legal to put those patches in the closed tree. However, as a copyright holder of GPL patches you would have to show this actually happened before you can make any claim.
If code was produced with a dual license, one GPL, and one closed, then it's perfectly legal to abandon the GPL version, and keep working on the closed version.
Eventually there is some number of elements (as the element count grows) at which it goes so much slower that it vastly outweighs the flipping time.
First of all, you never mentioned that "for sufficiently big N" in your first post.
Secondly, finding a solution that requires a maximum of 2N flips can be done with O(N*log N) comparisons, not O(N^2).
Thirdly, for N sufficiently big that O(N*log N) comparisons on a computer takes more time than O(N) actual flips, you need have such a large stack of pancakes, that the flipping time can no longer be considered O(1).
Maybe you should spend some more time on the problem, and a little less effort on patronizing.
Sure, but you're comparing apples and oranges. The O(N^2) refers to the actual flipping time, while NP refers to the algorithm used for determining the minimum number of flips.
Now, an interesting question is: given a certain time per computation, and a certain time per actual flip, what is the total time you'd need to sort the stack, and does this approach O(N^2) for large enough N ?
Not necessarily. Maybe it could take 1 billion calculations to figure out the optimal algorithm with an NP algorithm. But if a computer can do those calculations in 1 second total, while it takes you 2 seconds per flip, it's still worth it to spend all those CPU cycles.
Plenty of other adjustments are done. People have moved measuring stations over the years. People also changed when and how they recorded the temperature. For instance, it could be that the recorded daily minimum/maximum between 1900 and 1950, and then switched to hourly, or maybe the thermometer broke, and they replaced it, or they replaced it with a more accurate model.
These changes can mostly be found automatically by looking at sudden discontinuities compared to neighboring stations.
one should simply discard all the urban data all the time since it's been forced to match the rural data
You could, but it would be a waste of useful data. It's better to collect data over several decades, determine the necessary adjustment (which should vary slowly over the years), and then use all the corrected data in a total average.
You have to realize that very few of the stations have perfect records spanning 100+ years. The solution is to combine all the imperfect records in such a way to extract maximum useful data.
But, people have tried using only rural data. The differences are minimal (and show slightly more warming).
PWM is only easy in software if you don't mind it being slow and jittery.
Of course, you can add-on an ADC, but that's messy. You'd be better off going with a simple microcontroller that has them included. You might even hook up that microcontroller to the Pi so you can show the results on the monitor, or control it through a web server.
If you want to do low-level hacking, like on Arduino, but you'd like some more speed and memory, and have ethernet and USB, it's probably better to look for a simpler ARM based microcontroller.
The Pi doesn't have much user programmable I/O, and it's also harder to access SoC features from the application, or get really accurate timing (Linux isn't very good at real-time stuff).
I was a bit too quick to assume that the date format in the POSIX touch was based on the POSIX getdate() function, which isn't true.
The GNU touch does use a getdate() function that is similar in complexity. This function is then implemented using a YACC grammar, adding to the bloat.
If someone cared, it would be easy to write a much smaller 'touch' program, without all these fancy options. Of course, nobody cares.
Slashdot Mirror
Not really. The available credit is about equal to one months paycheck.
I don't think so. I've never felt the urge to spend more than I can easily afford, and I doubt that will ever change.
The bank still gets paid through the credit card fee though. I'll admit I keep the credit card because it's convenient to use it for renting a car, booking hotels, or ordering stuff on-line. But at least the fee is constant, and I consider it worth it.
So, it's like a direct deposit, but printed out on a piece of paper ? Sounds very cumbersome and archaic.
I pay 0% on my credit cards. The credit card bills are automatically taken out of my checking account at the end of the month.
Great plan, as long as you don't mind that most people wouldn't survive a full blown revolution. Our society has become too fragile to keep the daily flow of necessities going in the middle of a revolution.
This assumes the tech is reliable enough that it can be made to work in a comfortable (loose hanging) shirt, that it has a good power source that will last for a long time, and that it can go through a washer and dryer without breaking, and that it's cheap enough that I can buy a wardrobe full of clothes that all have this technology.
I would think a device that's worn like a wristwatch has a better chance of working.
It depends on the specifics of the case. If somebody owns the code, or has been granted a special license that allows closed source development, the GPL doesn't apply to their fork, so they can keep developing it, without any obligation to put their changes also in the GPL version.
If they take patches from the GPL tree, and apply it to their own (closed source) tree, it depends on the agreements. If contributors to the project agree to transfer the copyright to them, or they allow patches to be copied into the closed tree, I don't see a problem.
If a contributor doesn't transfer the copyright, and doesn't grant the owner of the closed fork a license to copy, it would not be legal to put those patches in the closed tree. However, as a copyright holder of GPL patches you would have to show this actually happened before you can make any claim.
Looks like they are distributing their own binary, together with complete sources of the Linux kernel + patches.
Whether this is a GPL violation depends on how these two parts interact, exactly.
If code was produced with a dual license, one GPL, and one closed, then it's perfectly legal to abandon the GPL version, and keep working on the closed version.
Ah, your pancakes were probably all similarly sized, seeing that you ordered them at a professional facility.
The sorting problem (and especially the variation with burned pancakes) is more typical for nerds who attempt to bake their own.
First of all, you never mentioned that "for sufficiently big N" in your first post.
Secondly, finding a solution that requires a maximum of 2N flips can be done with O(N*log N) comparisons, not O(N^2).
Thirdly, for N sufficiently big that O(N*log N) comparisons on a computer takes more time than O(N) actual flips, you need have such a large stack of pancakes, that the flipping time can no longer be considered O(1).
Maybe you should spend some more time on the problem, and a little less effort on patronizing.
Nerds do. Feel free to turn in your membership card.
Sure, but you're comparing apples and oranges. The O(N^2) refers to the actual flipping time, while NP refers to the algorithm used for determining the minimum number of flips.
Now, an interesting question is: given a certain time per computation, and a certain time per actual flip, what is the total time you'd need to sort the stack, and does this approach O(N^2) for large enough N ?
That's easy. You flip the pancake when the top is just dry enough to allow flipping without making a mess.
The more important question is knowing what temperature the skillet should be.
Not necessarily. Maybe it could take 1 billion calculations to figure out the optimal algorithm with an NP algorithm. But if a computer can do those calculations in 1 second total, while it takes you 2 seconds per flip, it's still worth it to spend all those CPU cycles.
Plenty of other adjustments are done. People have moved measuring stations over the years. People also changed when and how they recorded the temperature. For instance, it could be that the recorded daily minimum/maximum between 1900 and 1950, and then switched to hourly, or maybe the thermometer broke, and they replaced it, or they replaced it with a more accurate model.
These changes can mostly be found automatically by looking at sudden discontinuities compared to neighboring stations.
You could, but it would be a waste of useful data. It's better to collect data over several decades, determine the necessary adjustment (which should vary slowly over the years), and then use all the corrected data in a total average.
You have to realize that very few of the stations have perfect records spanning 100+ years. The solution is to combine all the imperfect records in such a way to extract maximum useful data.
But, people have tried using only rural data. The differences are minimal (and show slightly more warming).
The hard part is finding the optimal solution with the fewest number of flips.
Except that a car battery couldn't deliver the instantaneous current required for this project.
So, the question about storage is still a valid and interesting one.
PWM is only easy in software if you don't mind it being slow and jittery.
Of course, you can add-on an ADC, but that's messy. You'd be better off going with a simple microcontroller that has them included. You might even hook up that microcontroller to the Pi so you can show the results on the monitor, or control it through a web server.
If you want to do low-level hacking, like on Arduino, but you'd like some more speed and memory, and have ethernet and USB, it's probably better to look for a simpler ARM based microcontroller.
These guys have some nice boards (click ARM on the left): http://olimex.com/dev/index.html
The Pi doesn't have much user programmable I/O, and it's also harder to access SoC features from the application, or get really accurate timing (Linux isn't very good at real-time stuff).
It's probably that big because it uses the gettext library to translate the help/version messages in local languages.
The core of the program is just an exit() call, which compiles to about 20 bytes of real code.
They better not have filenames with spaces in them, like ", ~"
At least he won't be around to care.
As an embedded programmer, I shun C++.
I originally meant to say that a lot of this stuff comes from POSIX, so it's not just GNU.
Here is the description of the POSIX getdate() function:
http://pubs.opengroup.org/onlinepubs/007904875/functions/getdate.html
I was a bit too quick to assume that the date format in the POSIX touch was based on the POSIX getdate() function, which isn't true.
The GNU touch does use a getdate() function that is similar in complexity. This function is then implemented using a YACC grammar, adding to the bloat.
If someone cared, it would be easy to write a much smaller 'touch' program, without all these fancy options. Of course, nobody cares.
Asking a lawyer to change your will every time you sign up for some website sounds like an expensive proposal.