A bug is a failure on the part of the programmer. You are supposed to recover from occasional failures, not have them as a constant and mandatory presence in your work.
Translation: "Why should we be smart if we can fake our way through everything even if we are stupid?"
Kill yourself. Really, kill yourself. I need less militantly stupid people around. EVERYBODY except militantly stupid people needs less militantly stupid people around.
Science does not work the way you think it does. Also scientific research is the opposite of engineering -- research produces knowledge out of interaction with reality, engineering uses knowledge to produce interaction with reality.
And if you dont run it thru the debugger and STEP thru it you are just guessing what it will do.
If you are not right about behavior of your code, you are not qualified to write it in the first place.
Many time I step thru my code to find some assumption I was making that is invalid.
Then go kill yourself. People like you are the reason why there are bugs everywhere.
You can write code that compiles with 0 warnings on the highest levels, can get thru the most stringent of lint checks, passed dozens of code reviews, pair wise coded, etc, etc etc.
Compiler warnings are about things you are supposed to know -- a good programmer only gets them on typos or after removing things thus leaving something unused in the code.
But until you run it and step thru and see you will never know.
LISTEN, EVERYONE!
This is what is wrong with those people. They think, they can write random shit, single-step through it, do more random changes, and repeat until it seems to run. Their code only works by accident. Get them out of programming.
Same with Saddam and WMD's. Before the gulf War he did NOT allow inspectors in to check for WMD's and obstructed all the way. Which led the opposition to believe that he had something to hide. We know where that went.
He was stupid. If he allowed unfettered access he would still be in power.
US would just claim that the weapons were somewhere else. When US government wants to go to war, US goes to war -- otherwise military-industrial complex would need a bailout like one the financial companies got, and in 2003 such a bailout was not invented yet.
Oh, it is ideal. This is how it was implemented in USSR, and it worked -- corruption was lower than in Czarist Russia and lower than in post-USSR Russia.
'Well-designed operating systems do not have any "hardware abstraction layer"' No. Its a basic choice OS designers make when creating their operating system.
And the choice that is based entirely on analogy is usually a bad one.
Microsoft believes they should be able to change their kernel willynilly without having binary drivers fail after every update.
Those two are completely unrelated. Keeping a binary compatibility and having the interface tied to a per-hardware-device model are two separate ideas, both of different degrees of awfulness.
Linus is ideologically opposed to that so Linux requires the method you describe. It is not "well designed" its *ideologically driven* so that companies can't release binary blobs easily. Linus believes if you aren't willing to share your source, gtfo. I can respect that, but when someone like you comes along spouting it as a superior *technical* design it's like someone going on about how great and objective Fox News is.
Just because someone can make all kinds of choices without exploding, it does not mean that some of those choices are not idiotic.
"A decade later, Unix-like systems have vastly superior GUI". I'm sorry, but no, maybe on a single monitor compared to *XP*, but I use Win7, gui design is a moving target and Unix still lags behind Microsoft which lags behind Apple.
Now THAT is a subjective opinion. Still wrong because KDE does everything Windows 7 ever could, and Apple has the prettiest but otherwise unremarkable user interface.
Also, good luck getting 4+ monitors working on *nix without tweaking a single thing, windows? no tweaking needed beyond simply dragging the monitor around so it mimics the physical layout.
Actually I am doing just that, with Nvidia drivers and utilities out of the box (plus Synergy to expand multi-monitor configuration for multiple machines).
I love linux, but it is far inferior to windows as a desktop OS unless you're using it for ideological reasons
Oh, the signature of (usually paid) Microsoft apologists everywhere. No wonder, you defend virtualization -- Windows can't compartmentalize the host on its own, and Windows virtual memory, scheduler, and networking can't get any more awful even if it has layers upon layers of virtualization all the way down.
The thing is, it's not your language, my language, or any other one person's language. English is constantly evolving, and insisting on using outdated definitions of words limits your potential audience. In order to efficiently convey ideas, it's important to use words that everyone understands; this is the information age, and scientific, political, and social debate isn't limited to the elite anymore.
Virtualization in the modern era of usage covers hardware and process virtualization.
No. Hardware virtualization is an obsolete technology, worthless for any purpose other than simulated environment for testing and reverse engineering. Virtual hosts and isolated environment support in modern operating systems is a modern, useful technology that serves a legitimate purpose -- establishing a layer that provides additional separation between processes and their environments, but exists only in resource handling mechanisms where such separation is desired. Just because the same goal can be achieved through a massive kludge of hardware virtualization, does not legitimize such kludges.
It's common for vendors of an inferior product to make an effort to conflate it with something modern, and it's also common for vendors of superior but new and little-known product to downplay its fundamental differences from the technology it replaces. This does not mean that a car is a mechanical horse, and in the same way it does not turn OpenVZ into a virtualization platform.
The hardware abstraction layer is its own thing now and no one would consider it 'virtualization', hell I'm going to guess you don't even though it pretty much sums up what you're saying with only some theoretical differences, practically its exactly what you described.
"Hardware abstraction layer" is an obsolete technology, too. Well-designed operating systems do not have any "hardware abstraction layer" -- they do not present hardware as an accessible resource to the applications to begin with. In those systems, only drivers are meant to deal with hardware itself, and it's a driver's responsibility to provide whatever abstraction fits the device's purpose. OS treats drivers with the same interface as interchangeable, and can have its own interfaces and resources visible to applications, thus allowing applications to treat all resources without any direct connection to the details of the underlying hardware. This allows software to remain compatible across multiple generations of hardware, easily access local and remote resources, all without software knowing about it, and without any layers of emulation. Linux, all modern BSD and likely most commercial Unix systems are firmly in this camp -- even if you disagree with this design, this is what they have, and this is what you have to accommodate when you intend to run them.
Virtualization re-establishes the need for presenting hardware as pseudo-hardware because it establishes a whole layer under the OS drivers. Additional layers, especially additional layers with functionality identical to other layers, are always a bad idea. It is a bad solution except for one case -- when single host must run operating systems with incompatible driver models (what is never a good design decision in the first place, and is always driven by hardware or software vendors refusing to support the OS that the user actually intends to run). It's an unwanted burden for OS developers to write drivers for such pseudo-hardware, knowing that everything they do will be second-guessed by some other "super-driver" written by VM vendors, people not in the least concerned with efficiency and reliability of the developer's OS.
The best (as in "least bad") virtualization solutions avoid thick layers by utilizing hardware's native access-partitioning capabilities (such as IOMMU), however it's still a massive and completely unnecessary kludge when applied to an OS that already handles such access through its drivers, and has all necessary access-partitioning capabilities built in.
Perhaps those of us who lump them together are wrong and we should get off your lawn... or perhaps you should adapt as that's how the industry uses it.
If people did not resist massive idiocies and sabotage of technology development perpetrated by "the industry", we would be all running Windows now.
Or, the VM does neither, and simply boots directly to the VM image, which is what I believe was being discussed. There is also no requirement that the VM be hosted on Windows at all. If that OS is never booted, then how does its security issues factor in at all?
Virtualization does not work this way. To "boot from image" you need a whole fake environment that creates a fake drive, fed through fake BIOS to bootloader, and then without BIOS to the real SATA driver from the OS running in fake RAM. If you are lucky, network adapter access is real (through iommu), but the drive most likely is not only fake but accessed through fully booted Windows running its NTFS filesystem.
The difference is fundamental. Virtualization is an obsolete mechanism that pre-dates all modern operating systems. It is used on mainframes because their hardware was specifically designed for it, before the idea of multitasking OS and kernel-userspace interface was invented. It carefully imitates hardware (real or specially devised "virtual" one optimized for such use) and creates layers of virtual memory, scheduler, I/O manager, networking subsystem, and usually other kernel components above those exact components of the real OS. It only works well if those components of the OS are done so poorly (or are nonexistent in early OSes), another level of those things doesn't make the system any worse.
Host compartmentalization (a.k.a. containers, jails) is a modern mechanism that imitates userspace interface of multiple hosts under a single kernel. It's elegant and efficient, the only thing it can't do is to run a completely different OS. Therefore, unless there is Windows involved, it's a vastly superior mechanism.
He intends to keep Windows usable and install Linux along with it, not under it. He even mentioned that he wants to shrink Windows partition, so Linux will be independent from it.
Since there are solutions from Citrix that do exactly this over a corporate WAN with far more nodes, I don't see how there would be too much trouble getting it to work on a local LAN with a 1 Gig switch. If you have something that would dispute that idea, then by all means, offer that rather than your little ad hominem attack.
1. Citrix. Sane people don't use Citrix products. 2. Modern laptops. As in, portable devices with ridiculous amount of resources in them.
I don't recall anyone saying anything about it running under the control of Windows, first of all; Windows itself would not actually launch, the machines would simply boot to the virtual image.
How, do you think, VM runs on anything other than high-end VMWare "solutions", and how come, an image is a file on a filesystem? Either, everything runs under the host OS, or with "primary" OS providing management and services from one of VMs while user's image is running from another.
Second, while I agree that Linux is still inherently more secure than Windows, that gap has drastically narrowed over the last 10 years
Right, from "fucking insecure" to "insecure". If you run Linux under VM under Windows, you are vulnerable to:
1. All security holes in Linux, 2. Plus all security holes in Windows (including ones that only give access to the user running VM or user that can write to VM files). 3. Plus all security holes in VM (yes, there are plenty of those, too).
As a bonus, you can't even update those components without suspending or shutting down your OS in VM.
, and you can actually deploy Windows securely. That line is getting really old.
By definition, if something is not secure, it can not be "deployed securely" unless you are going to deploy it without any intention to use it.
Socialism in USSR lacked the will to live. It was not energetic enough to compete on the international scene. Messages from abroad, with pictures of NYC city streets paved with gold, were not helpful.
I agree with that, however I don't see Capitalism as having "will to live" at this point, either. There just isn't much of an opposition, and its propaganda is still effective.
Many problems could be prevented by stopping Gorbachev (you may want to follow this link.) There would be no Chechnya, for example.
As much as I would be happy if such a thing happened, it's not that simple. I can't claim that I know Gorbachev's motivation, but I have fairly good idea about what Communists, at least sane and smart ones, were thinking.
Let me elaborate on this, putting some things in modern terms and context.
Communists' theory claims that government, laws, social system, and the structure of society in general inevitably adjusts to reflect its foundation in economy. A major change, such as revolution, happens when the economic model forced upon the society by its current structure is inconsistent with the development of its economy, in particular when large segments of populations defined by their role in economy, have their interests in a conflict that can be resolved by changing the rules for their roles.
Most famous is the conflict between capitalists and workers in its original 19th century form -- society recognizes capitalists as exclusive owners of all means of production, workers can only support their lives by working for capitalists, getting only basic life support in return. At the very beginning of Capitalism, worker's efficiency was close to the efficiency of subsistence farmer, and only by pooling combined, organized efforts, they can sustain production and give the capitalist a modest amount of surplus product that he can sell -- most of the product had to be sold for food to feed the workers.
As the efficiency of the workers grew and their salaries did not, capitalists end up with greater and greater amount of surplus product. If workers for some reason happened to be in a position of effective collective bargaining, capitalists would have to increase salaries, thus reducing the conflict between themselves and workers. If capitalists for some reason happened to be in a position to enforce their will upon workers, they would be in a position to keep workers poor, or even keep workers in conditions that are certain to cause disease and early death. High concentration of workers and increased literacy can cause the former condition (workers organize strikes, form unions), but increasing efficiency cause the latter (capitalists become so rich, they can take over cities, control the only source of employment in the area reachable by a worker, control various forms of private armies, force politicians to pass laws, etc.) The second trend eventually outpaces the former, and workers are in a situation when the only way to improve their condition beyond basic sustenance is to strip capitalists of their control over means of production (a.k.a. "have nothing to lose but their chains"), therefore eventually there is a revolution that preserves workers' collective work but removes capitalists' ability to take away the products of their work.
As the history of 20th century shows, this scenario can be staved off by increasing the salaries to the level when workers do not have sufficient motivation to fight, guaranteeing legal protection of unions, etc., however as capitalists (or, now, companies as individual ownership became an obstacle for growth) compete with each other, their collective interest is still to keep workers as close to the poverty line as possible. Except before the problem was with chronically terrible conditions, and now it is with constant threat of unemployment that inevitably results with homelessness and death. So conflict is still alive, it is just bei
This is false. There have been dozens and dozens of fixes which have never been released before any exploit was found.
Silently, without announcing it as a bug or ever getting near CVE.
"The first fact to face is that UNIX was not developed with security, in any realistic sense, in mind; this fact alone guarantees a vast number of holes."
- Dennis Ritchie
Haha.. . hes right.. nobody sane would ever choose this shitty design if they wanted a secure system.
Even Dennis Ritchie occasionally had some incredibly stupid ideas. STREAMS and Plan 9 come as two most prominent examples.
There was nothing about "underlying" OS. It was the originally installed Windows, there was no requirement to keep using it WHILE THE SYSTEM IS BOOTED INTO LINUX.
Slashdot Mirror
A bug is a failure on the part of the programmer. You are supposed to recover from occasional failures, not have them as a constant and mandatory presence in your work.
Translation: "Why should we be smart if we can fake our way through everything even if we are stupid?"
Kill yourself. Really, kill yourself. I need less militantly stupid people around. EVERYBODY except militantly stupid people needs less militantly stupid people around.
Your expression does not work you think it does if there is more than one colon in the input... ...bitch!
Science does not work the way you think it does.
Also scientific research is the opposite of engineering -- research produces knowledge out of interaction with reality, engineering uses knowledge to produce interaction with reality.
Now go, punch yourself in the face.
And if you dont run it thru the debugger and STEP thru it you are just guessing what it will do.
If you are not right about behavior of your code, you are not qualified to write it in the first place.
Many time I step thru my code to find some assumption I was making that is invalid.
Then go kill yourself. People like you are the reason why there are bugs everywhere.
You can write code that compiles with 0 warnings on the highest levels, can get thru the most stringent of lint checks, passed dozens of code reviews, pair wise coded, etc, etc etc.
Compiler warnings are about things you are supposed to know -- a good programmer only gets them on typos or after removing things thus leaving something unused in the code.
But until you run it and step thru and see you will never know.
LISTEN, EVERYONE!
This is what is wrong with those people. They think, they can write random shit, single-step through it, do more random changes, and repeat until it seems to run. Their code only works by accident. Get them out of programming.
They water it down a lot.
Same with Saddam and WMD's. Before the gulf War he did NOT allow inspectors in to check for WMD's and obstructed all the way. Which led the opposition to believe that he had something to hide. We know where that went.
He was stupid. If he allowed unfettered access he would still be in power.
US would just claim that the weapons were somewhere else. When US government wants to go to war, US goes to war -- otherwise military-industrial complex would need a bailout like one the financial companies got, and in 2003 such a bailout was not invented yet.
Oh, it is ideal. This is how it was implemented in USSR, and it worked -- corruption was lower than in Czarist Russia and lower than in post-USSR Russia.
Less private money to politicians => less bribes to do things I am opposed to.
No, it means that different employees have different levels of trust and different responsibilities.
I mean, they do it already.
Thatcher, Thatcher, Thatcher, Thatcher,
Thatcher, Thatcher, Thatcher, Thatcher,
Thatcher, Thatcher, Thatcher, Thatcher,
Reagan, Reagan!
(with people in Guy Fawkes masks dancing, of course)
'Well-designed operating systems do not have any "hardware abstraction layer"' No. Its a basic choice OS designers make when creating their operating system.
And the choice that is based entirely on analogy is usually a bad one.
Microsoft believes they should be able to change their kernel willynilly without having binary drivers fail after every update.
Those two are completely unrelated. Keeping a binary compatibility and having the interface tied to a per-hardware-device model are two separate ideas, both of different degrees of awfulness.
Linus is ideologically opposed to that so Linux requires the method you describe. It is not "well designed" its *ideologically driven* so that companies can't release binary blobs easily. Linus believes if you aren't willing to share your source, gtfo. I can respect that, but when someone like you comes along spouting it as a superior *technical* design it's like someone going on about how great and objective Fox News is.
Just because someone can make all kinds of choices without exploding, it does not mean that some of those choices are not idiotic.
"A decade later, Unix-like systems have vastly superior GUI". I'm sorry, but no, maybe on a single monitor compared to *XP*, but I use Win7, gui design is a moving target and Unix still lags behind Microsoft which lags behind Apple.
Now THAT is a subjective opinion. Still wrong because KDE does everything Windows 7 ever could, and Apple has the prettiest but otherwise unremarkable user interface.
Also, good luck getting 4+ monitors working on *nix without tweaking a single thing, windows? no tweaking needed beyond simply dragging the monitor around so it mimics the physical layout.
Actually I am doing just that, with Nvidia drivers and utilities out of the box (plus Synergy to expand multi-monitor configuration for multiple machines).
I love linux, but it is far inferior to windows as a desktop OS unless you're using it for ideological reasons
Oh, the signature of (usually paid) Microsoft apologists everywhere. No wonder, you defend virtualization -- Windows can't compartmentalize the host on its own, and Windows virtual memory, scheduler, and networking can't get any more awful even if it has layers upon layers of virtualization all the way down.
The thing is, it's not your language, my language, or any other one person's language. English is constantly evolving, and insisting on using outdated definitions of words limits your potential audience. In order to efficiently convey ideas, it's important to use words that everyone understands; this is the information age, and scientific, political, and social debate isn't limited to the elite anymore.
u nggaz shd bettr be rdy4 txt msg spk in ur books
Virtualization in the modern era of usage covers hardware and process virtualization.
No. Hardware virtualization is an obsolete technology, worthless for any purpose other than simulated environment for testing and reverse engineering. Virtual hosts and isolated environment support in modern operating systems is a modern, useful technology that serves a legitimate purpose -- establishing a layer that provides additional separation between processes and their environments, but exists only in resource handling mechanisms where such separation is desired. Just because the same goal can be achieved through a massive kludge of hardware virtualization, does not legitimize such kludges.
It's common for vendors of an inferior product to make an effort to conflate it with something modern, and it's also common for vendors of superior but new and little-known product to downplay its fundamental differences from the technology it replaces. This does not mean that a car is a mechanical horse, and in the same way it does not turn OpenVZ into a virtualization platform.
The hardware abstraction layer is its own thing now and no one would consider it 'virtualization', hell I'm going to guess you don't even though it pretty much sums up what you're saying with only some theoretical differences, practically its exactly what you described.
"Hardware abstraction layer" is an obsolete technology, too. Well-designed operating systems do not have any "hardware abstraction layer" -- they do not present hardware as an accessible resource to the applications to begin with. In those systems, only drivers are meant to deal with hardware itself, and it's a driver's responsibility to provide whatever abstraction fits the device's purpose. OS treats drivers with the same interface as interchangeable, and can have its own interfaces and resources visible to applications, thus allowing applications to treat all resources without any direct connection to the details of the underlying hardware. This allows software to remain compatible across multiple generations of hardware, easily access local and remote resources, all without software knowing about it, and without any layers of emulation. Linux, all modern BSD and likely most commercial Unix systems are firmly in this camp -- even if you disagree with this design, this is what they have, and this is what you have to accommodate when you intend to run them.
Virtualization re-establishes the need for presenting hardware as pseudo-hardware because it establishes a whole layer under the OS drivers. Additional layers, especially additional layers with functionality identical to other layers, are always a bad idea. It is a bad solution except for one case -- when single host must run operating systems with incompatible driver models (what is never a good design decision in the first place, and is always driven by hardware or software vendors refusing to support the OS that the user actually intends to run). It's an unwanted burden for OS developers to write drivers for such pseudo-hardware, knowing that everything they do will be second-guessed by some other "super-driver" written by VM vendors, people not in the least concerned with efficiency and reliability of the developer's OS.
The best (as in "least bad") virtualization solutions avoid thick layers by utilizing hardware's native access-partitioning capabilities (such as IOMMU), however it's still a massive and completely unnecessary kludge when applied to an OS that already handles such access through its drivers, and has all necessary access-partitioning capabilities built in.
Perhaps those of us who lump them together are wrong and we should get off your lawn ... or perhaps you should adapt as that's how the industry uses it.
If people did not resist massive idiocies and sabotage of technology development perpetrated by "the industry", we would be all running Windows now.
There are companies that offer MS Exchange server hosting -- so what?
Ok, it actually meant "IT'S AWWRIGHT".
Or, the VM does neither, and simply boots directly to the VM image, which is what I believe was being discussed. There is also no requirement that the VM be hosted on Windows at all. If that OS is never booted, then how does its security issues factor in at all?
Virtualization does not work this way. To "boot from image" you need a whole fake environment that creates a fake drive, fed through fake BIOS to bootloader, and then without BIOS to the real SATA driver from the OS running in fake RAM. If you are lucky, network adapter access is real (through iommu), but the drive most likely is not only fake but accessed through fully booted Windows running its NTFS filesystem.
* * * BARF!!! * * *
The difference is fundamental. Virtualization is an obsolete mechanism that pre-dates all modern operating systems. It is used on mainframes because their hardware was specifically designed for it, before the idea of multitasking OS and kernel-userspace interface was invented. It carefully imitates hardware (real or specially devised "virtual" one optimized for such use) and creates layers of virtual memory, scheduler, I/O manager, networking subsystem, and usually other kernel components above those exact components of the real OS. It only works well if those components of the OS are done so poorly (or are nonexistent in early OSes), another level of those things doesn't make the system any worse.
Host compartmentalization (a.k.a. containers, jails) is a modern mechanism that imitates userspace interface of multiple hosts under a single kernel. It's elegant and efficient, the only thing it can't do is to run a completely different OS. Therefore, unless there is Windows involved, it's a vastly superior mechanism.
Are you a moron?
He intends to keep Windows usable and install Linux along with it, not under it. He even mentioned that he wants to shrink Windows partition, so Linux will be independent from it.
Since there are solutions from Citrix that do exactly this over a corporate WAN with far more nodes, I don't see how there would be too much trouble getting it to work on a local LAN with a 1 Gig switch. If you have something that would dispute that idea, then by all means, offer that rather than your little ad hominem attack.
1. Citrix. Sane people don't use Citrix products.
2. Modern laptops. As in, portable devices with ridiculous amount of resources in them.
I don't recall anyone saying anything about it running under the control of Windows, first of all; Windows itself would not actually launch, the machines would simply boot to the virtual image.
How, do you think, VM runs on anything other than high-end VMWare "solutions", and how come, an image is a file on a filesystem? Either, everything runs under the host OS, or with "primary" OS providing management and services from one of VMs while user's image is running from another.
Second, while I agree that Linux is still inherently more secure than Windows, that gap has drastically narrowed over the last 10 years
Right, from "fucking insecure" to "insecure".
If you run Linux under VM under Windows, you are vulnerable to:
1. All security holes in Linux,
2. Plus all security holes in Windows (including ones that only give access to the user running VM or user that can write to VM files).
3. Plus all security holes in VM (yes, there are plenty of those, too).
As a bonus, you can't even update those components without suspending or shutting down your OS in VM.
, and you can actually deploy Windows securely. That line is getting really old.
By definition, if something is not secure, it can not be "deployed securely" unless you are going to deploy it without any intention to use it.
No, I just think, "here is a remote VM, go mess with things within it" kind of service is stupid.
Socialism in USSR lacked the will to live. It was not energetic enough to compete on the international scene. Messages from abroad, with pictures of NYC city streets paved with gold, were not helpful.
I agree with that, however I don't see Capitalism as having "will to live" at this point, either. There just isn't much of an opposition, and its propaganda is still effective.
Many problems could be prevented by stopping Gorbachev (you may want to follow this link.) There would be no Chechnya, for example.
As much as I would be happy if such a thing happened, it's not that simple. I can't claim that I know Gorbachev's motivation, but I have fairly good idea about what Communists, at least sane and smart ones, were thinking.
Let me elaborate on this, putting some things in modern terms and context.
Communists' theory claims that government, laws, social system, and the structure of society in general inevitably adjusts to reflect its foundation in economy. A major change, such as revolution, happens when the economic model forced upon the society by its current structure is inconsistent with the development of its economy, in particular when large segments of populations defined by their role in economy, have their interests in a conflict that can be resolved by changing the rules for their roles.
Most famous is the conflict between capitalists and workers in its original 19th century form -- society recognizes capitalists as exclusive owners of all means of production, workers can only support their lives by working for capitalists, getting only basic life support in return. At the very beginning of Capitalism, worker's efficiency was close to the efficiency of subsistence farmer, and only by pooling combined, organized efforts, they can sustain production and give the capitalist a modest amount of surplus product that he can sell -- most of the product had to be sold for food to feed the workers.
As the efficiency of the workers grew and their salaries did not, capitalists end up with greater and greater amount of surplus product. If workers for some reason happened to be in a position of effective collective bargaining, capitalists would have to increase salaries, thus reducing the conflict between themselves and workers. If capitalists for some reason happened to be in a position to enforce their will upon workers, they would be in a position to keep workers poor, or even keep workers in conditions that are certain to cause disease and early death. High concentration of workers and increased literacy can cause the former condition (workers organize strikes, form unions), but increasing efficiency cause the latter (capitalists become so rich, they can take over cities, control the only source of employment in the area reachable by a worker, control various forms of private armies, force politicians to pass laws, etc.) The second trend eventually outpaces the former, and workers are in a situation when the only way to improve their condition beyond basic sustenance is to strip capitalists of their control over means of production (a.k.a. "have nothing to lose but their chains"), therefore eventually there is a revolution that preserves workers' collective work but removes capitalists' ability to take away the products of their work.
As the history of 20th century shows, this scenario can be staved off by increasing the salaries to the level when workers do not have sufficient motivation to fight, guaranteeing legal protection of unions, etc., however as capitalists (or, now, companies as individual ownership became an obstacle for growth) compete with each other, their collective interest is still to keep workers as close to the poverty line as possible. Except before the problem was with chronically terrible conditions, and now it is with constant threat of unemployment that inevitably results with homelessness and death. So conflict is still alive, it is just bei
This is false. There have been dozens and dozens of fixes which have never been released before any exploit was found.
Silently, without announcing it as a bug or ever getting near CVE.
"The first fact to face is that UNIX was not developed with security, in any realistic sense, in mind; this fact alone guarantees a vast number of holes."
- Dennis Ritchie
Haha.. . hes right .. nobody sane would ever choose this shitty design if they wanted a secure system.
Even Dennis Ritchie occasionally had some incredibly stupid ideas. STREAMS and Plan 9 come as two most prominent examples.
There was nothing about "underlying" OS. It was the originally installed Windows, there was no requirement to keep using it WHILE THE SYSTEM IS BOOTED INTO LINUX.