To get even more economy, you need to reduce weight.
Utterly false.
replace need to with can, and add , amongst other things. to the end.
Take away weight and you ultimately take away strength
Cars aren't about strength. They're about safety.
I'm thinking that you're part of the camp that believes a 4000 pound tank from the 60s is safer than a modern car.
Sure, one one hand, it has a whole lot of momentum to put into the argument of who's going to push who. On the other hand, 100% of that momentum is going to be transferred to you in that rigid ass body. Hope you enjoy being, well, dead.
wtf?
let's go with 1.5g/cm^3 as a good middle density for dry ice.
we have to replace 3kt, or 3,000,000kg, or 3,000,000,000g, or 2,000,000,000 cm^3, or a cube 1259cm on all sides. 1259cm is also known as 12.59 meters. Which is a lot fucking closer to parent's 15m^3 than your 2km 2-dimensional "block"
I don't know about that.
Convection has quite literally completely stopped on that planet most likely. It's just cooling down and slowly solidifying.
To create enough "cracks" to get the heat flux for convection to really start again might take hitting it with something the size of... well, itself, and then blowing away all of its atmosphere so that there's somewhere for that heat to actually go.
There are also other possibilities for what we have observed that give no hope whatsoever of ever getting that dynamo running- ie, the sucker is already solidified.
No, that analogy is completely stupid. An ant walking past the Empire State Building does not steal orbital momentum from it and forever alter its motion through the cosmos.
I get that you were trying to say it's a small amount, but that doesn't excuse that bone headed analogy.
I don't see where you get 'it's very limited' from.
That depends entirely on its use.
If, for example, the world's consumption of gold were to trickle down to just electronics (desire for gold as jewelry is an entirely cultural factor. some cultures valued salt more than gold, by weight, due to the abundance and lack of use for gold), then the amount of gold left in the crust is enough for millennia without the need to recycle.
Current actual goods consumption is driven largely by the Chinese, probably due to their rapid expanding middle class. People love them some glittery jewelry when they start having money for the first time.
US consumption of gold for reasons other than investment has only declined, and probably will continue to decline, as the average Joe the millennial has less and less of a shit to give about it. It's investment value will drop like a rock as soon as the Chinese have had their fill of glittery crap made of the stuff.
Yes, but water makes sense based on basic physics (H2O at that pressure and temperature equals water), as well as the specific reflective properties of the return signal.
You are right that it might not be water, but there are many indications that its water, and none that it is not.
No fucking way, chief.
Gold is just another commodity. Its value is based on A) it's use as a non-corroding conductor, and B) rich people wanting glittery shit made out of it, and poor people wanting glittery shit made of it so that they can feel more like those rich people.
It's no longer the reserve currency of the world, by any means.
An entire asteroid worth of gold would be worthless if it costed 10 cents more per troy ounce to bring it back than it cost terrestrially to mine. And what's worse, is that value would quickly drop the more of it you brought in if the operation was too large. Gold mining today is already limited by the value of gold. The vast majority of known gold in the dirt is left there, because it's not economically viable to pull up.
I know a few amateur prospectors who really overestimate the worth of gold. I don't really understand why. Bretton-Woods is dead. There is no gold standard. Gold wasn't a better standard than the more modern basket-of-goods. It was a useful one for its time, back when economies were relatively simple.
Invasive species only outcompete native species when the environment is similar.
Sure, I guess. Where "similar" is loosely defined as "area capable of sustained aerobic metabolism within the thermal requirements of the organism"
Invasive species decimate ecosystems in environments that are nothing like their own by terrestrial standards.
Since we're talking about bacteria here, it gets even scarier, since there are plenty of bacteria on this planet, right now, that wouldn't be too deterred by conditions on Mars. Or do you think Lake Vostok isn't full of life?
You put giant land snails in the arctic, they won't survive.
Sure. And if you put a Vibrio bacterium in your gut, well outside of its happy place, it will still thrive and kill you.
You put snakehead carp in the desert they won't survive.
Very true. And if you put a butterfly in space, it won't survive either.
But there are species of cyanobacteria that call the driest place on this planet home. In fact, Chroococcidiopsis has been singled out as an ideal invasive species for Mars, should we ever decide we want to terraform the place.
We're too different.
We, where we means all the phyla native to Planet Earth, are definitely not too different.
You can measure the activity of our biosphere from space. You start throwing Earth biota, organisms that have been playing the game of life in hypercompetitive conditions for 4 billion years, on that barren rock to face a small and relatively-speaking static biome, we're going to win. We will destroy them.
Of course I'm willing to make generalizations that earth species couldn't compete
You shouldn't be, because you're wrong.
with perhaps a few extremophile exceptions
No. The places we're talking about life existing on Mars aren't extremophilic places.
We're talking about deep in the soil, and deep under ice in cold lakes. There are literally millions of species on this planet that thrive in those conditions.
If a species is evolved to survive Mars's chemistry- temperature- radiation levels... etc... yes... it will outcompete any species that we would accidentally take there.
Well, most scientists on the planet disagree with you, and for good reason.
You seem to have fallen prey to the idea that evolution produces the best suited resident of an environment.
It doesn't.
Any organisms on humans are not likely to be able to compete in Mars against species that evolved in that environment. If you stick gut bacteria in a hydrothermal vent it won't outcompete it. Terran bacteria will only overrun martian bacteria if none are still alive.
Eek.
Your general assumption, that an invasive species will automatically be less suited to compete in an environment against native species is flat out wrong.
You then back it up with a very strictly limited scenario where it is not.
I wouldn't underestimate the ability of Earth organisms, living in a hyper-competitive environment, to be able to completely dominate a small and constrained ecosystem if one exists.
Earth organisms would be rats on their island.
Nope. It was a lot of words to say, 'you don't understand what it's reporting'.
It is reporting the progress of that userspace transfer to the kernel, nothing more, nothing less. It doesn't account for any OS cachine, device cache, or asynchronous device behavior.
That makes it somewhat confusing, it's true, but that's what it reports. And it does not lie in that reporting.
Beyond that, the devices are emulating behavior that who's behavior doesn't translate linearly to the operating system.
Flash devices do not operate atomically on 512 byte sectors. USB drives do.
The flash devices typically have 4K (or larger) write sector, and 16K (or larger) erase blocks.
A lot of magic must be performed in order to write a 512 byte sector to the middle of a 4K already-written flash sector that can only be erased along with 16K worth of sectors around it.
These are still being done when the controller returns and says 'I can accept more data, now.'
Right back at you. Those "background tasks" finish in milliseconds and do not sit there in some mythical long idle queue waiting for the OS to signal that the USB driver has been unloaded.
It isn't some mythical idle queue. The interface to the backend flash controller chips is almost always asynchronous. You're also writing 512 byte blocks to devices with 4K or larger sectors. The full delay for the transmission of data to the flash controller and it completing all the tasks can be in the order of seconds, not milliseconds.
Actually I'm using the device entirely as designed. "Optimised for quick disconnect" is such for a reason. No need to wait 10 seconds. As I said with write-caching disabled the time between being given the 100% complete signal and you actually yanking is milliseconds for the device to actually finish what its doing. About the only risk you still have is that an application actively has a file open. At that point yanking will net you a lovely error message.
Nope. You are still operating at the wrong layer. Like I said, you're not even aware of what you don't know. Disabling write-back caching at the OS is only half of the problem. Some devices, like disk drives, have extra commands that can be sent to ask their media controllers to behave purely synchronously. Backend flash controller devices often do not.
You see, in the early days this shit was actually a problem. Slow wear leveling algorithms, write caching, poor USB drivers, generally poor (in the windows world) handling of hot-swappable hardware, filesystems that would fall over if you look at them the wrong way, it all was very real 20 years ago. Then people realised that something needed to be done, and did something about it.
You're entirely out of your depth.
Let us present the simple scenario of modifying a single 512 byte sector.
The flash drive cannot do this. There are multiple ways it can accomplish this, depending on what the media looks like.
It can simply write that 512 byte blocks to an erased sector that it has available if it knows the surrounding 512 byte blocks are not used. If they are, we enter a new branch. What's the best way forward, do we read the flash sector, erase it, and then re-write it? Or if we can, do we read, and write somewhere to an available erased sector?
All of this simply does not trickle back all the way up to the application layer synchronous writing calls, even with all OS caches disabled.
The modern manufacturer intends a device to be connected to a fallible bus by idiots.
Objection: Speculation.
I am using the device exactly as the manufacturer intended
No, you're not. As evidenced by the fact that the first person to decry this article was someone from *SanDisk*. You may know them as the chaps who make these fucking devices.
Sorry buddy, you're an idiot and you're too dim-witted to see it.
What you do may be safe for your workloads, but it's not safe in general, and isn't recommended because the people who understand how it all works are quite simply smarter than you. I know that hurts, but suck it the fuck up and move on.
It is the dominant factor for any write that isn't presented to the user. Any other write is typically shown via a progress bar and if those writes fail you get an error message.
Wrong.
The devices you are talking to are little computers in unto themselves. They are designed to be hot swappable- *after* they've been told to stop doing the work they have queued up, and to finish up any background tasks required for the medium. There's a reason clicking that little eject button de-enumerates the USB device.
Why would you think that?
See above. People often think blinkers are unnecessary. They're wrong. Their brains lack the critical thinking to fully understand the problem space, though they think they do.
LOL okay mate we'll go with your crazy doom and gloom scenarios.
It's not a doom and gloom scenario. I'm well aware you may never run into a problem. Certain types of writing heuristics are more dangerous for a 'write and yank' than others, and if yours is just copying an excel spreadsheet to the thing, waiting 10 seconds, and yanking, you're probably OK. You're still misusing the device as the manufacturer and protocols dictate, and you're still an idiot for suggesting that to other people without understanding why your specific scenarios are less problematic than scenarios that may cost people lost data.
No, I mean smarter.
The design is as such for reasons that are clearly beyond your comprehension. That doesn't make it bad. It makes you ignorant.
If you can't be bothered to use the product correctly, that's fine. But you're the idiot, not them.
int("1234foobar") equals 1234 and "foobar" will be ignored, as atrocious as it sounds
As evil as that sounds, i was never caught off guard by that behavior with my background as a C programmer. That has always been the behavior of atoi() and strto* in libc.
PHP7 is a pretty decent language to work with.
PHP4 and 5 are piles of fucking garbage. The linguistic gymnastics in common use to emulate reasonable behavior with that language are rage-inducing to work with.
I think he's referring to the loader. It's often called the 'interpreter' in OS parlance. In fact, the ELF header refers to the loader as the interpreter.
Developing in PHP 7.0 or higher in an OOP fashion, creating nice clean, testable and robust code is easy and painless.
I will concede that as of PHP7, the language is actually not a complete pile of shit used by piles of shit to construct bigger piles of shit. It's actually pretty nice.
A tuba is still an instrument that can make beautiful noises.
We're talking about PHP here. I'd like to see what John Lennon could do with the entrails of a sheep and some baboon excrement.
Slashdot Mirror
To get even more economy, you need to reduce weight.
Utterly false.
replace need to with can, and add , amongst other things. to the end.
Take away weight and you ultimately take away strength
Cars aren't about strength. They're about safety.
I'm thinking that you're part of the camp that believes a 4000 pound tank from the 60s is safer than a modern car.
Sure, one one hand, it has a whole lot of momentum to put into the argument of who's going to push who. On the other hand, 100% of that momentum is going to be transferred to you in that rigid ass body. Hope you enjoy being, well, dead.
I agree 100%... but also vice versa :(
wtf?
let's go with 1.5g/cm^3 as a good middle density for dry ice.
we have to replace 3kt, or 3,000,000kg, or 3,000,000,000g, or 2,000,000,000 cm^3, or a cube 1259cm on all sides. 1259cm is also known as 12.59 meters. Which is a lot fucking closer to parent's 15m^3 than your 2km 2-dimensional "block"
cracking the crust to start plate tectonics
I don't know about that.
Convection has quite literally completely stopped on that planet most likely. It's just cooling down and slowly solidifying.
To create enough "cracks" to get the heat flux for convection to really start again might take hitting it with something the size of... well, itself, and then blowing away all of its atmosphere so that there's somewhere for that heat to actually go.
There are also other possibilities for what we have observed that give no hope whatsoever of ever getting that dynamo running- ie, the sucker is already solidified.
No, that analogy is completely stupid. An ant walking past the Empire State Building does not steal orbital momentum from it and forever alter its motion through the cosmos.
I get that you were trying to say it's a small amount, but that doesn't excuse that bone headed analogy.
I don't see where you get 'it's very limited' from.
That depends entirely on its use.
If, for example, the world's consumption of gold were to trickle down to just electronics (desire for gold as jewelry is an entirely cultural factor. some cultures valued salt more than gold, by weight, due to the abundance and lack of use for gold), then the amount of gold left in the crust is enough for millennia without the need to recycle.
Current actual goods consumption is driven largely by the Chinese, probably due to their rapid expanding middle class. People love them some glittery jewelry when they start having money for the first time.
US consumption of gold for reasons other than investment has only declined, and probably will continue to decline, as the average Joe the millennial has less and less of a shit to give about it. It's investment value will drop like a rock as soon as the Chinese have had their fill of glittery crap made of the stuff.
Interest is not affiliation.
Yes, but water makes sense based on basic physics (H2O at that pressure and temperature equals water), as well as the specific reflective properties of the return signal.
You are right that it might not be water, but there are many indications that its water, and none that it is not.
No fucking way, chief.
Gold is just another commodity. Its value is based on A) it's use as a non-corroding conductor, and B) rich people wanting glittery shit made out of it, and poor people wanting glittery shit made of it so that they can feel more like those rich people.
It's no longer the reserve currency of the world, by any means.
An entire asteroid worth of gold would be worthless if it costed 10 cents more per troy ounce to bring it back than it cost terrestrially to mine. And what's worse, is that value would quickly drop the more of it you brought in if the operation was too large. Gold mining today is already limited by the value of gold. The vast majority of known gold in the dirt is left there, because it's not economically viable to pull up.
I know a few amateur prospectors who really overestimate the worth of gold. I don't really understand why. Bretton-Woods is dead. There is no gold standard. Gold wasn't a better standard than the more modern basket-of-goods. It was a useful one for its time, back when economies were relatively simple.
Invasive species only outcompete native species when the environment is similar.
Sure, I guess. Where "similar" is loosely defined as "area capable of sustained aerobic metabolism within the thermal requirements of the organism"
Invasive species decimate ecosystems in environments that are nothing like their own by terrestrial standards.
Since we're talking about bacteria here, it gets even scarier, since there are plenty of bacteria on this planet, right now, that wouldn't be too deterred by conditions on Mars. Or do you think Lake Vostok isn't full of life?
You put giant land snails in the arctic, they won't survive.
Sure. And if you put a Vibrio bacterium in your gut, well outside of its happy place, it will still thrive and kill you.
You put snakehead carp in the desert they won't survive.
Very true. And if you put a butterfly in space, it won't survive either.
But there are species of cyanobacteria that call the driest place on this planet home. In fact, Chroococcidiopsis has been singled out as an ideal invasive species for Mars, should we ever decide we want to terraform the place.
We're too different.
We, where we means all the phyla native to Planet Earth, are definitely not too different.
You can measure the activity of our biosphere from space. You start throwing Earth biota, organisms that have been playing the game of life in hypercompetitive conditions for 4 billion years, on that barren rock to face a small and relatively-speaking static biome, we're going to win. We will destroy them.
Of course I'm willing to make generalizations that earth species couldn't compete
You shouldn't be, because you're wrong.
with perhaps a few extremophile exceptions
No. The places we're talking about life existing on Mars aren't extremophilic places. We're talking about deep in the soil, and deep under ice in cold lakes. There are literally millions of species on this planet that thrive in those conditions.
If a species is evolved to survive Mars's chemistry- temperature- radiation levels... etc... yes... it will outcompete any species that we would accidentally take there.
Well, most scientists on the planet disagree with you, and for good reason.
You seem to have fallen prey to the idea that evolution produces the best suited resident of an environment.
It doesn't.
Any organisms on humans are not likely to be able to compete in Mars against species that evolved in that environment. If you stick gut bacteria in a hydrothermal vent it won't outcompete it. Terran bacteria will only overrun martian bacteria if none are still alive.
Eek. Your general assumption, that an invasive species will automatically be less suited to compete in an environment against native species is flat out wrong.
You then back it up with a very strictly limited scenario where it is not.
I wouldn't underestimate the ability of Earth organisms, living in a hyper-competitive environment, to be able to completely dominate a small and constrained ecosystem if one exists.
Earth organisms would be rats on their island.
Nope. It was a lot of words to say, 'you don't understand what it's reporting'.
It is reporting the progress of that userspace transfer to the kernel, nothing more, nothing less. It doesn't account for any OS cachine, device cache, or asynchronous device behavior.
That makes it somewhat confusing, it's true, but that's what it reports. And it does not lie in that reporting.
Beyond that, the devices are emulating behavior that who's behavior doesn't translate linearly to the operating system.
Flash devices do not operate atomically on 512 byte sectors. USB drives do.
The flash devices typically have 4K (or larger) write sector, and 16K (or larger) erase blocks.
A lot of magic must be performed in order to write a 512 byte sector to the middle of a 4K already-written flash sector that can only be erased along with 16K worth of sectors around it.
These are still being done when the controller returns and says 'I can accept more data, now.'
Insanity is doing the same thing over and over again while expecting a different result from the ones that have occurred.
Nope. That's something said by ignorant humans who fill their brain with sound bites and cliches instead of actual knowledge.
Given your consistent horrible mis-use of the word
Given your apparent ignorance of the definition for consistent, as well as insane, you mean...
I'm going to say that by definition, you are the insane one.
And definition...
Don't you have a ditch to dig somewhere, man?
Right back at you. Those "background tasks" finish in milliseconds and do not sit there in some mythical long idle queue waiting for the OS to signal that the USB driver has been unloaded.
It isn't some mythical idle queue. The interface to the backend flash controller chips is almost always asynchronous. You're also writing 512 byte blocks to devices with 4K or larger sectors. The full delay for the transmission of data to the flash controller and it completing all the tasks can be in the order of seconds, not milliseconds.
Actually I'm using the device entirely as designed. "Optimised for quick disconnect" is such for a reason. No need to wait 10 seconds. As I said with write-caching disabled the time between being given the 100% complete signal and you actually yanking is milliseconds for the device to actually finish what its doing. About the only risk you still have is that an application actively has a file open. At that point yanking will net you a lovely error message.
Nope. You are still operating at the wrong layer. Like I said, you're not even aware of what you don't know. Disabling write-back caching at the OS is only half of the problem. Some devices, like disk drives, have extra commands that can be sent to ask their media controllers to behave purely synchronously. Backend flash controller devices often do not.
You see, in the early days this shit was actually a problem. Slow wear leveling algorithms, write caching, poor USB drivers, generally poor (in the windows world) handling of hot-swappable hardware, filesystems that would fall over if you look at them the wrong way, it all was very real 20 years ago. Then people realised that something needed to be done, and did something about it.
You're entirely out of your depth.
Let us present the simple scenario of modifying a single 512 byte sector.
The flash drive cannot do this. There are multiple ways it can accomplish this, depending on what the media looks like. It can simply write that 512 byte blocks to an erased sector that it has available if it knows the surrounding 512 byte blocks are not used. If they are, we enter a new branch. What's the best way forward, do we read the flash sector, erase it, and then re-write it? Or if we can, do we read, and write somewhere to an available erased sector?
All of this simply does not trickle back all the way up to the application layer synchronous writing calls, even with all OS caches disabled.
The modern manufacturer intends a device to be connected to a fallible bus by idiots.
Objection: Speculation.
I am using the device exactly as the manufacturer intended
No, you're not. As evidenced by the fact that the first person to decry this article was someone from *SanDisk*. You may know them as the chaps who make these fucking devices.
Sorry buddy, you're an idiot and you're too dim-witted to see it.
What you do may be safe for your workloads, but it's not safe in general, and isn't recommended because the people who understand how it all works are quite simply smarter than you. I know that hurts, but suck it the fuck up and move on.
It is the dominant factor for any write that isn't presented to the user. Any other write is typically shown via a progress bar and if those writes fail you get an error message.
Wrong.
The devices you are talking to are little computers in unto themselves. They are designed to be hot swappable- *after* they've been told to stop doing the work they have queued up, and to finish up any background tasks required for the medium. There's a reason clicking that little eject button de-enumerates the USB device.
Why would you think that?
See above. People often think blinkers are unnecessary. They're wrong. Their brains lack the critical thinking to fully understand the problem space, though they think they do.
LOL okay mate we'll go with your crazy doom and gloom scenarios.
It's not a doom and gloom scenario. I'm well aware you may never run into a problem. Certain types of writing heuristics are more dangerous for a 'write and yank' than others, and if yours is just copying an excel spreadsheet to the thing, waiting 10 seconds, and yanking, you're probably OK. You're still misusing the device as the manufacturer and protocols dictate, and you're still an idiot for suggesting that to other people without understanding why your specific scenarios are less problematic than scenarios that may cost people lost data.
No, I mean smarter.
The design is as such for reasons that are clearly beyond your comprehension. That doesn't make it bad. It makes you ignorant.
If you can't be bothered to use the product correctly, that's fine. But you're the idiot, not them.
built-in == comparison operator whose semantics are byzantine
I prefer the term "fucking asinine"
Or do CentOS and RHEL have PHP 7 packages yet?
I use the remi repo.
int("1234foobar") equals 1234 and "foobar" will be ignored, as atrocious as it sounds
As evil as that sounds, i was never caught off guard by that behavior with my background as a C programmer. That has always been the behavior of atoi() and strto* in libc.
PHP7 is a pretty decent language to work with.
PHP4 and 5 are piles of fucking garbage. The linguistic gymnastics in common use to emulate reasonable behavior with that language are rage-inducing to work with.
I think he's referring to the loader. It's often called the 'interpreter' in OS parlance. In fact, the ELF header refers to the loader as the interpreter.
I had someone do that once. Only worse. He did it in ruby.
Developing in PHP 7.0 or higher in an OOP fashion, creating nice clean, testable and robust code is easy and painless.
I will concede that as of PHP7, the language is actually not a complete pile of shit used by piles of shit to construct bigger piles of shit. It's actually pretty nice.
A tuba is still an instrument that can make beautiful noises.
We're talking about PHP here. I'd like to see what John Lennon could do with the entrails of a sheep and some baboon excrement.
No shit. Stab stab stab.