Working together is a requirement for civilization. Taxes are a fine example of doing this, especially for societies that are based upon a monetary system. Most of the time they are necessary. Of course, they can be abused, and of course countries with large outside incomes (e.g. Switzerland with its banks, or Monaco with its casinos) may have little to no taxes. But basically, although not strictly a requirement, you're going to have to deal with taxes within most societies.
Doesn't work. He will still have friends that benefited from his actions that would happily let him live in their expensive villa. Besides that, how much of the xxx billion would he pay back by doing any kind of job?
"At least the laws about copyright infringement here were passed by a popularly elected, semi-almost-functioning legislature. We should be blaming ourselves for electing politicians that pass laws mandating criminal penalties inthis instant."
Aw, yes, you might want to try that in other countries, though even in the Netherlands, I don't see this happen, and we have a multi-party democracy. In the US with their two party system: not a chance in hell.
You choose the party that comes most closely to your ideals, and about none of them won't be influenced by the record industry once in power. We should be focusing on influencing the sitting officials instead - and making sure the industry has less power over them.
In other words: theoretically you are right, but this is just not practical - it's not how democracy works.
Yes, they are. They are going to because of the billions of pounds that go into the "war" on drugs. This would be brilliant for spotting drug trafficking. Other uses will slip in in time (intentionally or not).
Um, yes, and they provide ad-free IMAP and POP services. They only cost bandwidth, but they will lure people (such as me) to their ad based web-interfaces, if only to make changes and/or to click on the update button for the external POP interface. Besides that they weaken competition such as hotmail, meaning that M$ will get even less money from the ads.
"Suddenly a word that had a well known meaning in the real world (i.e. libels are lies) has a different meaning in law."
If they were, we would call them lies. Libels and lies may be close but they are not the same.
"The plaintiff is complaining about a situation in which they were the ones doing something fundamentally wrong."
You mean he was sentenced by a court? Who is deciding that he was doing something fundamentally wrong?
"The truth seems to be less important than the ability to use weasel words and slippery logic."
Come now, you just made that up on the spot didn't you? In the article I did not see any weasel words or slippery logic.
"It encourages bad behavior e.g. in this case sales people with expense accounts who feel they don't need to keep records, and should suffer no adverse effects if they get caught."
You don't need to shame anyone by name to make that clear to your employees. He did suffer adverse effects already without being shamed: he did get fired and I doubt that he will get any recommendations from the company. And the rumor circuit will probably same him anyway to his direct colleagues, the other salesmen.
I think it was a necessary and good legal decision. One that does not bring the US even closer to the middle ages (where the ruling class can do what they want without being responsible for anything).
And what if he isn't? You're a thief when it is proven in court you are a thief. They should have just fired him. If you at the same time explain in your newsletter that employees are not to steal from the company, risking a direct pink slip if they do, ok. But damaging an employee this way is simply not done.
Even my MSX computer did real time raytracing like a champ, providing that all the pixels were produced from a 2D non-reflective surface using a 90 degree angle. Of course you had a limited color space, but otherwise everything run just smoothly.
Kidding aside, I suppose it's how far you want to take it. Amiga or MSX are not interesting anymore for about 90% of the things they did. The one exception is probably playing retro games.
Why not start with the heating of the fat itself? Make a frying pan that cleans itself and uses the leftover/bad oil to heat the new/filtered oil. One machine that powers itself. You can of course use some of the energy for the electronics/pumps and maybe even power a few other devices. Or would this not be technical feasible? I like self sufficient devices.
Can somebody confirm that the explosions of Li-Ion batteries *is* the release of all of the energy at once? In that case, it does not matter if you have this fast discharging battery or one of the older ones; what matters is the amount of energy stored within the battery.
A very fast DVD reader would be brilliant for my current way of using CD's and DVD's : copy them to drive (removing any constraints) and playing the media later on. I have a hick up free experience and I safe battery life. They could even make one that was very fast but could not do searching; just rip the whole thing to drive already, I'll use a virtual DVD or CD-ROM if required...
I presume that the rocket and the boom are not that closely related. Certainly you could fire rocket + detonating mechanism for practice and do a (computer generated) boom separately? It's not like they have to explode on impact or anything. Sending out a life rocket into space does not seem to be the smartest thing to do...
This all assuming that you use the flash only for the build. Also note that many IDE's now use real time parsing, basically making hundreds of small changes. Only on a few files normally, but it does distort your mathematics. The mathematics are not that unreasonable, but keep in mind that real life is slightly more complicated than this.
It seems to work for the JAR problem, at least in 64 bit Linux. Note that this is only for consecutive builds. A restart will require all the data to be cached again - basically letting you start over all again. I am using suspend to RAM, which basically leaves the MB in cache. At work, where I only have 2 GB on 32 bit Windows (meaning: lots of restarts as well) this scheme won't work. The other thing there is that I rely on 3rd party, buggy, alpha/beta state drivers quite a lot, so restarting is more common for that reason alone.
Another thing: copying library files is much faster than randomly looking into them (at least on my Windows system - I'm amazed about the zipping speed on my Linux system). So copying the library files into RAM first may significantly speed things up, at least initially. Of course that does means that you have to make changes to your environment.
Another advantage for SSD is that you may not have to change the configuration for your project or OS, you just receive a speedup, period. Of course you could easily combine forces, have an SSD + a sizable amount of RAM. If you make sure you don't write too fast to SSD you would get a reasonable amount of speedup.
About a logging file system: I was thinking the same thing. It seems that e.g. the ZFS filesystem is not directly the best match for SSD, because it still writes a lot of writes at the same place. This is just a rumor I found on internet, but it seemed at the time to be convincing enough. So you may consider it fine, but do research to make sure that it really works. It's very hard to find reliable information about these issues.
"What happens if you exceed the limit of writes?" You may loose a block for writing. It will be tagged, but the data won't be lost.
"How does usage degrade the disks?" Doing many writes, and especially many small writes may hurt performance - a lot. You can reset some SSD's and reinstall. This will reset the counters, but they should be spreading the writes in the first place.
"Is heat bad?" No, up to a certain temperature of course, but SSD's are probably better resistant to heat than disks. Flash is pretty hard to kill.
"Does using the SSD as virtual memory degrade the disk fast?" Faster, yes, but it depends on the amount of swaps, obviously. Setting a high amount of swap space on my 8 GB of RAM with swappiness low won't do a thing to the life time of any hardware, it could just as easily be a tape drive. Personally I would rather worry about "temporary internet files".
"What about bad sectors, how do they compare with HDDs?" Better. You *know* when you have a bad sector. Most of the time you will still be able to read the block. Otherwise, you should not compare sectors with HDD's (sorry, couldn't resist).
"Are SSDs generally more sturdy(longer lifespans) than HDDs?" For a desktop system this may be the case. But not all applications will have the same characteristics.
Note that not all controllers are created equal; you *need* to do research on that before buying an SSD, since most use the bad ones. Bad controllers can slow down writes to a crawl, way worse than HDD's. This depends on the number of writes you perform (not the amount of free space).
PS. I haven't bought an SSD yet, this is from a few days of research into the matter.
You can also reset the entire drive once in a while and put a previously saved image back (e.g. the one you were just working on). Most performance problems would go away, and it only costs you a single clear for each sector.
Or just use a (proprietary) program in Windows to work with your local RAM or use tmpfs under linux. I've just compiled the OpenJDK in no time at all to a RAM drive. My 8 GB of RAM already had the openjdk itself cached, but the output files still had to be written (all ~ 1.1 GB of them).
I'm trying to see if I could load my Eclipse workspaces in RAM as well. I could save using version control and/or an automated backup script (e.g. rsync once in a while). Eclipse itself stays on the drive. Once you've started it once, it is in memory anyway, and copying it into RAM would only take time (locking the RAM for other uses as well, cache may be reused when running out of memory).
You can buy motherboards for 16 GB of RAM. But even 8 GB of RAM is plenty for even the biggest workspace.
Yes, but you first have to build the infrastructure in a safe way. You cannot just start to communicate. The problem with Quantum cryptography [for key distribution using light] is that it only solves the part where you transfer the key. It's definitely cheaper and arguably more secure to transfer a bunch of disks (for redundancy, you don't want one hdd that can crash) than to build yourself system for Quantum Key Distribution. And it is infinitely more secure to put that money into other security properties of the system.
Just create two 512 or 521 bit ECC keys and swap the public keys. Make sure you've got your parties set up to handle any problems when they have been deployed, that's much more important. Use ECDH or simple ECC encryption when setting up 256 bit AES session keys, make sure that side channel attacks are not possible, and you've got yourself one heck of a secure line. No need for a one time pad at all, one time pads are simply not needed *at all*.
I've got mixed results with Google Chrome and FF3 regarding Java Script performance. Not that it matters because I'm typing this on Linux, and the platform support promise of Google is getting *very* long in the tooth.
About the bugs: sometimes you need to change the underlying architecture to really get rid of bugs. Not all bugs can be fixed that easily. It's not a 3.x upgrade for nothing. And if you really change things, you need to test them. See how long it takes IE to do something about fundamental flaws. Yes, this is frustrating and not like it is supposed to go, but sometimes it takes more time and effort.
Not that I am completely at odds with your argument; I do think the author got a bit carried away. He definitely was raving a bit.
I can't say it is completely stable, but I was rather surprised when I found flash working in Ubuntu 64 bit (8.10) at least 90% of the time. That said, the open source variants are still playing catch up so far - a shame because I would rather use files from those projects - if only to have the automated update work for all of my browsing habits as well.
Nope, too many people like it too much. Some things that are prominent features should be in the base install, otherwise you will see that the FF3 uptake will be less. If people are going to have to look through the plugins (hoping to find anything, you can hardly tell them to look for the "awesomebar") then it's already too late.
Maybe they could make an advanced "custom" install that figures this option, but the savvy users will find about:config anyway. Or someone could host a stripped firefox version with all the recent additions switched off.
Slashdot Mirror
Working together is a requirement for civilization. Taxes are a fine example of doing this, especially for societies that are based upon a monetary system. Most of the time they are necessary. Of course, they can be abused, and of course countries with large outside incomes (e.g. Switzerland with its banks, or Monaco with its casinos) may have little to no taxes. But basically, although not strictly a requirement, you're going to have to deal with taxes within most societies.
Doesn't work. He will still have friends that benefited from his actions that would happily let him live in their expensive villa. Besides that, how much of the xxx billion would he pay back by doing any kind of job?
"At least the laws about copyright infringement here were passed by a popularly elected, semi-almost-functioning legislature. We should be blaming ourselves for electing politicians that pass laws mandating criminal penalties inthis instant."
Aw, yes, you might want to try that in other countries, though even in the Netherlands, I don't see this happen, and we have a multi-party democracy. In the US with their two party system: not a chance in hell.
You choose the party that comes most closely to your ideals, and about none of them won't be influenced by the record industry once in power. We should be focusing on influencing the sitting officials instead - and making sure the industry has less power over them.
In other words: theoretically you are right, but this is just not practical - it's not how democracy works.
Yes, they are. They are going to because of the billions of pounds that go into the "war" on drugs. This would be brilliant for spotting drug trafficking. Other uses will slip in in time (intentionally or not).
Um, yes, and they provide ad-free IMAP and POP services. They only cost bandwidth, but they will lure people (such as me) to their ad based web-interfaces, if only to make changes and/or to click on the update button for the external POP interface. Besides that they weaken competition such as hotmail, meaning that M$ will get even less money from the ads.
"Suddenly a word that had a well known meaning in the real world (i.e. libels are lies) has a different meaning in law."
If they were, we would call them lies. Libels and lies may be close but they are not the same.
"The plaintiff is complaining about a situation in which they were the ones doing something fundamentally wrong."
You mean he was sentenced by a court? Who is deciding that he was doing something fundamentally wrong?
"The truth seems to be less important than the ability to use weasel words and slippery logic."
Come now, you just made that up on the spot didn't you? In the article I did not see any weasel words or slippery logic.
"It encourages bad behavior e.g. in this case sales people with expense accounts who feel they don't need to keep records, and should suffer no adverse effects if they get caught."
You don't need to shame anyone by name to make that clear to your employees. He did suffer adverse effects already without being shamed: he did get fired and I doubt that he will get any recommendations from the company. And the rumor circuit will probably same him anyway to his direct colleagues, the other salesmen.
I think it was a necessary and good legal decision. One that does not bring the US even closer to the middle ages (where the ruling class can do what they want without being responsible for anything).
And what if he isn't? You're a thief when it is proven in court you are a thief. They should have just fired him. If you at the same time explain in your newsletter that employees are not to steal from the company, risking a direct pink slip if they do, ok. But damaging an employee this way is simply not done.
Even my MSX computer did real time raytracing like a champ, providing that all the pixels were produced from a 2D non-reflective surface using a 90 degree angle. Of course you had a limited color space, but otherwise everything run just smoothly.
Kidding aside, I suppose it's how far you want to take it. Amiga or MSX are not interesting anymore for about 90% of the things they did. The one exception is probably playing retro games.
"while the police is not allowed to shoot even after they get shot at and are injured"
They can only shoot their side arm after being buried 6 feet under? Or do they have to go to the arcade like the rest of us?
Thank you, it took me precisely 5 seconds to install using the package manager :). That's one thing to like in Ubuntu...
Why not start with the heating of the fat itself? Make a frying pan that cleans itself and uses the leftover/bad oil to heat the new/filtered oil. One machine that powers itself. You can of course use some of the energy for the electronics/pumps and maybe even power a few other devices. Or would this not be technical feasible? I like self sufficient devices.
Can somebody confirm that the explosions of Li-Ion batteries *is* the release of all of the energy at once? In that case, it does not matter if you have this fast discharging battery or one of the older ones; what matters is the amount of energy stored within the battery.
A very fast DVD reader would be brilliant for my current way of using CD's and DVD's : copy them to drive (removing any constraints) and playing the media later on. I have a hick up free experience and I safe battery life. They could even make one that was very fast but could not do searching; just rip the whole thing to drive already, I'll use a virtual DVD or CD-ROM if required...
I presume that the rocket and the boom are not that closely related. Certainly you could fire rocket + detonating mechanism for practice and do a (computer generated) boom separately? It's not like they have to explode on impact or anything. Sending out a life rocket into space does not seem to be the smartest thing to do...
This all assuming that you use the flash only for the build. Also note that many IDE's now use real time parsing, basically making hundreds of small changes. Only on a few files normally, but it does distort your mathematics. The mathematics are not that unreasonable, but keep in mind that real life is slightly more complicated than this.
It seems to work for the JAR problem, at least in 64 bit Linux. Note that this is only for consecutive builds. A restart will require all the data to be cached again - basically letting you start over all again. I am using suspend to RAM, which basically leaves the MB in cache. At work, where I only have 2 GB on 32 bit Windows (meaning: lots of restarts as well) this scheme won't work. The other thing there is that I rely on 3rd party, buggy, alpha/beta state drivers quite a lot, so restarting is more common for that reason alone.
Another thing: copying library files is much faster than randomly looking into them (at least on my Windows system - I'm amazed about the zipping speed on my Linux system). So copying the library files into RAM first may significantly speed things up, at least initially. Of course that does means that you have to make changes to your environment.
Another advantage for SSD is that you may not have to change the configuration for your project or OS, you just receive a speedup, period. Of course you could easily combine forces, have an SSD + a sizable amount of RAM. If you make sure you don't write too fast to SSD you would get a reasonable amount of speedup.
About a logging file system: I was thinking the same thing. It seems that e.g. the ZFS filesystem is not directly the best match for SSD, because it still writes a lot of writes at the same place. This is just a rumor I found on internet, but it seemed at the time to be convincing enough. So you may consider it fine, but do research to make sure that it really works. It's very hard to find reliable information about these issues.
Sorry, but I would prefer using tmpfs (I've already done this). Tweaking development is not the same as tweaking the whole computer.
PS. we're talking Linux here
"What happens if you exceed the limit of writes?"
You may loose a block for writing. It will be tagged, but the data won't be lost.
"How does usage degrade the disks?"
Doing many writes, and especially many small writes may hurt performance - a lot. You can reset some SSD's and reinstall. This will reset the counters, but they should be spreading the writes in the first place.
"Is heat bad?"
No, up to a certain temperature of course, but SSD's are probably better resistant to heat than disks. Flash is pretty hard to kill.
"Does using the SSD as virtual memory degrade the disk fast?"
Faster, yes, but it depends on the amount of swaps, obviously. Setting a high amount of swap space on my 8 GB of RAM with swappiness low won't do a thing to the life time of any hardware, it could just as easily be a tape drive. Personally I would rather worry about "temporary internet files".
"What about bad sectors, how do they compare with HDDs?"
Better. You *know* when you have a bad sector. Most of the time you will still be able to read the block. Otherwise, you should not compare sectors with HDD's (sorry, couldn't resist).
"Are SSDs generally more sturdy(longer lifespans) than HDDs?"
For a desktop system this may be the case. But not all applications will have the same characteristics.
Note that not all controllers are created equal; you *need* to do research on that before buying an SSD, since most use the bad ones. Bad controllers can slow down writes to a crawl, way worse than HDD's. This depends on the number of writes you perform (not the amount of free space).
PS. I haven't bought an SSD yet, this is from a few days of research into the matter.
You can also reset the entire drive once in a while and put a previously saved image back (e.g. the one you were just working on). Most performance problems would go away, and it only costs you a single clear for each sector.
Or just use a (proprietary) program in Windows to work with your local RAM or use tmpfs under linux. I've just compiled the OpenJDK in no time at all to a RAM drive. My 8 GB of RAM already had the openjdk itself cached, but the output files still had to be written (all ~ 1.1 GB of them).
I'm trying to see if I could load my Eclipse workspaces in RAM as well. I could save using version control and/or an automated backup script (e.g. rsync once in a while). Eclipse itself stays on the drive. Once you've started it once, it is in memory anyway, and copying it into RAM would only take time (locking the RAM for other uses as well, cache may be reused when running out of memory).
You can buy motherboards for 16 GB of RAM. But even 8 GB of RAM is plenty for even the biggest workspace.
Yes, but you first have to build the infrastructure in a safe way. You cannot just start to communicate. The problem with Quantum cryptography [for key distribution using light] is that it only solves the part where you transfer the key. It's definitely cheaper and arguably more secure to transfer a bunch of disks (for redundancy, you don't want one hdd that can crash) than to build yourself system for Quantum Key Distribution. And it is infinitely more secure to put that money into other security properties of the system.
Just create two 512 or 521 bit ECC keys and swap the public keys. Make sure you've got your parties set up to handle any problems when they have been deployed, that's much more important. Use ECDH or simple ECC encryption when setting up 256 bit AES session keys, make sure that side channel attacks are not possible, and you've got yourself one heck of a secure line. No need for a one time pad at all, one time pads are simply not needed *at all*.
I've got mixed results with Google Chrome and FF3 regarding Java Script performance. Not that it matters because I'm typing this on Linux, and the platform support promise of Google is getting *very* long in the tooth.
About the bugs: sometimes you need to change the underlying architecture to really get rid of bugs. Not all bugs can be fixed that easily. It's not a 3.x upgrade for nothing. And if you really change things, you need to test them. See how long it takes IE to do something about fundamental flaws. Yes, this is frustrating and not like it is supposed to go, but sometimes it takes more time and effort.
Not that I am completely at odds with your argument; I do think the author got a bit carried away. He definitely was raving a bit.
I can't say it is completely stable, but I was rather surprised when I found flash working in Ubuntu 64 bit (8.10) at least 90% of the time. That said, the open source variants are still playing catch up so far - a shame because I would rather use files from those projects - if only to have the automated update work for all of my browsing habits as well.
Nope, too many people like it too much. Some things that are prominent features should be in the base install, otherwise you will see that the FF3 uptake will be less. If people are going to have to look through the plugins (hoping to find anything, you can hardly tell them to look for the "awesomebar") then it's already too late.
Maybe they could make an advanced "custom" install that figures this option, but the savvy users will find about:config anyway. Or someone could host a stripped firefox version with all the recent additions switched off.