Really? Because in my opinion, the truth is that you should just use the device as the manufacturer recommends, which is properly shut down the device before removing its power.
Inaccurate.
The system unmounts, and then disconnects the device. This is so that the device can finish up any tasks it may be doing before the user violently removes its power source.
Oh, I hope it does more than sync and umount, because that isn't enough to guarantee the disk is in a safe state to remove. It must also disconnect it.
The back-end storage chips on most USB sticks have rather complicated behavior to make those flash chips behave anything like a disk drive, and that behavior often requires the controller keep working in the background. Only a complete moron pulls the power from a USB stick before safely ejecting it. Sure, he may dodge the bullet most of the time, but you only have to zip up your pecker once to realize you were taking a stupid risk not checking in the past.
No, you simply don't understand what it's reporting. And that's ok, you're a generally ignorant human.
You see, kiddo, the modern OS has many layers stacked upon each other.
Your application submits I/O to a syscall, the syscall hands it off to the VFS subsystem, which then hands it off (maybe) to the filesystem, who then hands it off to the I/O scheduler, who then hands it off to the device driver, who then hands it off to the device. Here's where it gets extra cool- your device may be ultra dumb (floppy, single FIFO, blocking) to a TCQ SATA/SCSI device, to an eMMC performing wear leveling, sector size emulation, and coagulation in the background.
In the middle, we have all kinds of potential transforms, reordering, and write-combining. It's a mess, really. The software machinery involved for high-speed I/O would blow your mind, I think.
We try very hard to make sure people smarter than you are designing the systems so that this whole thing doesn't go awry. We do understand though that you can be easily confused by this UI widget saying it's done, when that eMMC chip in that USB flash drive is erasing a sector to concatenate your last bit of data with the incomplete sector write of the preceding bits, and so we have offered you a mechanism by which to safely eject these devices. If you're too stupid to use that feature, we can't help you further.
Your ignorance is terrifying. As if whether or not the OS cache is write-through or write-back is the only factor in play. Keep yanking away. If you wait long enough, where long enough is some completely unknown amount of time, you'll even avoid getting your dick stuck in your zipper.
I bet you're one of those assholes who doesn't think they need a blinker.
It's not a case of not figuring it out, it's a case of not reviewing a 20 year old assumption.
And this is a case of you reviewing a 20 year old assumption based on the incorrect assumption that you are qualified to review it. Let me give you a tip- you're not.
And if you cut out in the middle of an access-time-write, the filesystem at worst has an incorrect access time.
And herein lies our problem. Blind, proud, and dangerous ignorance.
There are a lot of layers in a file write. Even more layers when it's to a flash drive with built in wear-leveling.
You presume to know what the underlying filesystem, vfs cache, block scheduler, device driver, and the device itself are doing. In short, you are insane.
You do *not* want to pull power off that device simply because it's not receiving data to write from the PC anymore.
You are otherwise correct that atimes aren't commonly in use.
I used it plenty.... It definitely came off sometimes, like if I'd bump the cable near the connection point against my leg or something, but I don't recall it ever leaping off of the laptop of its own accord. The alternative, is that force is transferred to a connector on the mainboard of your laptop. I'll go with the disconnecting.
I think it's super innovative and awesome. I wish my HP had one.
Where I think they completely failed in critical thinking was making it replace the damn function keys.
As a PC guy myself, I concur. My girlfriend's MagSafe connector was the only part of her Mac that gave me envy. I had a higher resolution display and better processor than you could even get in a MBP, but those MagSafes really were a brilliant idea.
Made in USA, America, as well as Europe or Africa, are continents.
America is not a continent. North America is a continent. The Americas are continents.
As far as North America is concerned, our satellite nations to the north and south can suck this one up. We were the first independent nation in both Americas, Native American sovereignty notwithstanding, and our country is literally named for that fact.
Canada isn't called the Canadian Republic of America. Mexico isn't called the Mexican Republic of America.
It's well established, even over seas, that "America" refers to the USA. Now, when we start using North America to refer to the USA, you can bitch.
What you do is like French decided one day that they are the Europeans, call their country Europe, and call the rest of the continent "The Europas".
Na, it'd be more like if The United Kingdom of Great Britain were simply called Great Britain sometimes... Oh wait, whoops.
Jackass.
Dozens of events per frame, each multiple bytes long
Ya, no. Big no.
You have to factor in the action caused by the event. And when you're throwing pixbufs over that pipe as fast as you can, 60 or 100 mouse moves per second build up a significant backlog of events. Mouse down? Send pixbuf for pressed button. Mouse up? Send pixbuf for unpressed button. Wait- you're already sending 30 other pixbufs because there's crap going on in your application? Well, shit.
Oh yes, another more contemporary problem is that X hasn't really been network-transparent for a long time, yet still assumes that it is.
No, it's fully network transparent. The problem is that there are extensions to the protocol in use that are relied upon to make the models performance not suck, liked mapped pixbufs instead of pixbufs thrown over the socket to the X server. You take away all the things that remove network transparency on a local machine, and you're left with what the network transparent model can actually do.
I have always loved the network transparency of X, at least in concept. In use, it's only viable on high bandwidth low latency networks.
In my experience, even a full-blown RDP session outperforms X forwarding in most cases.
When you think about it, it's not really surprising as to why.
Forwarding all interaction with the GUI, redraw events, inputs (a lot of events are generated as that mouse moves across that window), is a lot more data than a fixed-rate rasterized product sent with compression.
I thought I was a reasonably intelligent human until I encountered PowerShell. 5 minutes with it, and I feel retarded. I haven't determined if it's because I'm too stupid to digest that syntax, if my brain cells are forcefully refusing to, or they're outright committing ritual suicide.
but only a moron (well, lots of them) would assume that he didn't have a good reason.
Man, that's some bad fucking logic.
We're to give every asshole on the internet flinging accusations without a shred of evidence the benefit of the doubt?
Our President would love that.
It was completely unacceptable and so very unbecoming of someone in his position.
Unfortunately, this kind of shit has been made mainstream by our President.
It hasn't cost him anything, why would other powerful men think it would cost them anything?
There is *definitely* some cult of personality shit going on with Tesla/Musk and some of the cheerleaders who own his products.
That being said, I've owned and driven a lot of cars, from a $600 Ford Probe, to a $220,000 SL65. Teslas are amongst the best vehicles I have ever driven.
I think there's also a cult of people who hate on anything Tesla because they represent some kind of affront to some kind of conviction they hold or something. I kinda think you might be one of those people.
He may be overexaggerating with "withing the same model year"
He wasn't. Lots of changes are made to vehicles mid-year. It's not literally every vehicle every year or anything, but if the manufacturer needs to, they do, and it's common.
That's why every other car manufacturer settles on a design and sticks with it for a full model year before releasing a new revision.
LOL. You're clearly not someone who has ever worked on the cars you've owned.
It's not rare for there to be 2 or 3 different sets of power train designs for a single model year on a car. Sometimes, you need to know which damn factory your car came from, and the month it rolled off the line in order to know exactly which motor you've got.
Car manufacturers generally don't give two squirts of pass about third-party repair needs. They're considered competition to their own overpriced services, and their services could care less what changes happened when, because they can give an exact accounting of vehicle equipment with nothing but the VIN.
Slashdot Mirror
Bingo. Mod this coward up. He's got more knowledge than 80% of the registered dotards commenting on this article.
Really? Because in my opinion, the truth is that you should just use the device as the manufacturer recommends, which is properly shut down the device before removing its power.
Inaccurate.
The system unmounts, and then disconnects the device. This is so that the device can finish up any tasks it may be doing before the user violently removes its power source.
Fortunately, it would appear that the people in charge of such decisions are a lot smarter than you.
Oh, I hope it does more than sync and umount, because that isn't enough to guarantee the disk is in a safe state to remove. It must also disconnect it.
The back-end storage chips on most USB sticks have rather complicated behavior to make those flash chips behave anything like a disk drive, and that behavior often requires the controller keep working in the background. Only a complete moron pulls the power from a USB stick before safely ejecting it. Sure, he may dodge the bullet most of the time, but you only have to zip up your pecker once to realize you were taking a stupid risk not checking in the past.
No, you simply don't understand what it's reporting. And that's ok, you're a generally ignorant human.
You see, kiddo, the modern OS has many layers stacked upon each other.
Your application submits I/O to a syscall, the syscall hands it off to the VFS subsystem, which then hands it off (maybe) to the filesystem, who then hands it off to the I/O scheduler, who then hands it off to the device driver, who then hands it off to the device. Here's where it gets extra cool- your device may be ultra dumb (floppy, single FIFO, blocking) to a TCQ SATA/SCSI device, to an eMMC performing wear leveling, sector size emulation, and coagulation in the background.
In the middle, we have all kinds of potential transforms, reordering, and write-combining. It's a mess, really. The software machinery involved for high-speed I/O would blow your mind, I think.
We try very hard to make sure people smarter than you are designing the systems so that this whole thing doesn't go awry. We do understand though that you can be easily confused by this UI widget saying it's done, when that eMMC chip in that USB flash drive is erasing a sector to concatenate your last bit of data with the incomplete sector write of the preceding bits, and so we have offered you a mechanism by which to safely eject these devices. If you're too stupid to use that feature, we can't help you further.
I bet you're one of those assholes who doesn't think they need a blinker.
It's not a case of not figuring it out, it's a case of not reviewing a 20 year old assumption.
And this is a case of you reviewing a 20 year old assumption based on the incorrect assumption that you are qualified to review it. Let me give you a tip- you're not.
And if you cut out in the middle of an access-time-write, the filesystem at worst has an incorrect access time.
And herein lies our problem. Blind, proud, and dangerous ignorance.
There are a lot of layers in a file write. Even more layers when it's to a flash drive with built in wear-leveling.
You presume to know what the underlying filesystem, vfs cache, block scheduler, device driver, and the device itself are doing. In short, you are insane.
You do *not* want to pull power off that device simply because it's not receiving data to write from the PC anymore.
You are otherwise correct that atimes aren't commonly in use.
And to be on-topic - i usually just `yank` my thumb drive out whenever i want.
Nevermind. You're fucking hopeless.
I used it plenty.... It definitely came off sometimes, like if I'd bump the cable near the connection point against my leg or something, but I don't recall it ever leaping off of the laptop of its own accord. The alternative, is that force is transferred to a connector on the mainboard of your laptop. I'll go with the disconnecting.
I hate the Touch Bar
I think it's super innovative and awesome. I wish my HP had one.
Where I think they completely failed in critical thinking was making it replace the damn function keys.
As a PC guy myself, I concur. My girlfriend's MagSafe connector was the only part of her Mac that gave me envy. I had a higher resolution display and better processor than you could even get in a MBP, but those MagSafes really were a brilliant idea.
Made in USA, America, as well as Europe or Africa, are continents.
America is not a continent. North America is a continent. The Americas are continents.
As far as North America is concerned, our satellite nations to the north and south can suck this one up. We were the first independent nation in both Americas, Native American sovereignty notwithstanding, and our country is literally named for that fact.
Canada isn't called the Canadian Republic of America. Mexico isn't called the Mexican Republic of America.
It's well established, even over seas, that "America" refers to the USA. Now, when we start using North America to refer to the USA, you can bitch.
What you do is like French decided one day that they are the Europeans, call their country Europe, and call the rest of the continent "The Europas".
Na, it'd be more like if The United Kingdom of Great Britain were simply called Great Britain sometimes... Oh wait, whoops.
Jackass.
Dozens of events per frame, each multiple bytes long
Ya, no. Big no. You have to factor in the action caused by the event. And when you're throwing pixbufs over that pipe as fast as you can, 60 or 100 mouse moves per second build up a significant backlog of events. Mouse down? Send pixbuf for pressed button. Mouse up? Send pixbuf for unpressed button. Wait- you're already sending 30 other pixbufs because there's crap going on in your application? Well, shit.
Oh yes, another more contemporary problem is that X hasn't really been network-transparent for a long time, yet still assumes that it is.
No, it's fully network transparent. The problem is that there are extensions to the protocol in use that are relied upon to make the models performance not suck, liked mapped pixbufs instead of pixbufs thrown over the socket to the X server. You take away all the things that remove network transparency on a local machine, and you're left with what the network transparent model can actually do.
I have always loved the network transparency of X, at least in concept. In use, it's only viable on high bandwidth low latency networks.
In my experience, even a full-blown RDP session outperforms X forwarding in most cases.
When you think about it, it's not really surprising as to why.
Forwarding all interaction with the GUI, redraw events, inputs (a lot of events are generated as that mouse moves across that window), is a lot more data than a fixed-rate rasterized product sent with compression.
You're right. Fascism is nothing if not effective.
I thought I was a reasonably intelligent human until I encountered PowerShell. 5 minutes with it, and I feel retarded. I haven't determined if it's because I'm too stupid to digest that syntax, if my brain cells are forcefully refusing to, or they're outright committing ritual suicide.
Isn't free-market capitalism amazing?!
'Murica!
but only a moron (well, lots of them) would assume that he didn't have a good reason.
Man, that's some bad fucking logic.
We're to give every asshole on the internet flinging accusations without a shred of evidence the benefit of the doubt?
Our President would love that.
It was completely unacceptable and so very unbecoming of someone in his position.
Unfortunately, this kind of shit has been made mainstream by our President.
It hasn't cost him anything, why would other powerful men think it would cost them anything?
^ This. The bang-for-buck cost analysis for Hyundais and Kias is hard to beat.
There is *definitely* some cult of personality shit going on with Tesla/Musk and some of the cheerleaders who own his products.
That being said, I've owned and driven a lot of cars, from a $600 Ford Probe, to a $220,000 SL65. Teslas are amongst the best vehicles I have ever driven.
I think there's also a cult of people who hate on anything Tesla because they represent some kind of affront to some kind of conviction they hold or something. I kinda think you might be one of those people.
He may be overexaggerating with "withing the same model year"
He wasn't. Lots of changes are made to vehicles mid-year. It's not literally every vehicle every year or anything, but if the manufacturer needs to, they do, and it's common.
*piss
That's why every other car manufacturer settles on a design and sticks with it for a full model year before releasing a new revision.
LOL. You're clearly not someone who has ever worked on the cars you've owned.
It's not rare for there to be 2 or 3 different sets of power train designs for a single model year on a car. Sometimes, you need to know which damn factory your car came from, and the month it rolled off the line in order to know exactly which motor you've got.
Car manufacturers generally don't give two squirts of pass about third-party repair needs. They're considered competition to their own overpriced services, and their services could care less what changes happened when, because they can give an exact accounting of vehicle equipment with nothing but the VIN.