Big mistake IMO. By making Netflix bigger, they get more leverage against the studios, who are the real bad guys that are driving prices up. With more leverage, they can put more pressure on the studios to push prices down.
Lets assume for a moment that you are not some special and magical higher power whose word we take for this. What is the basis for your claim?
Because the C library is part of the definition of the Unix operating system, and POSIX-compatible systems. It's further part of every Linux system currently deployed.
In addition to clear, direct, consistent, and self supporting logic I've provided references to solid third party computer science experts with particular specialty in operating system design and theory. Do you have anything new to add other than "nu uh?"
Excuse me, but you've provided no such thing. You've linked to Tanenbaum's textbook and claimed that it supported your definitions, but you didn't cite any of the content to support your claim. Are we supposed to just take your word that these experts agree with you? Because frankly, I've been active in the microkernel community for over 10 years (search the EROS lists and you'll find me), so I'll take my direct reading of the research literature and direct interaction with these researchers as stronger evidence than your unsupported claims.
I mean sure, the application can't function without that code
Except it can by linking in a different but compatible library.
But wasn't it GNU who claimed that linked libraries ARE part of the application and therefore derivatives subject to the GPL?
Legal and technical definitions are completely distinct. Sometimes they overlap, and sometimes they don't. What constitutes derivative works according to copyright is an argument that is completely separate from a technical argument about modularity.
and that code would function in the same way if placed in the same source file as the rest of the application
Except it's not linked directly with your code, or even exclusively with your code, it's dynamically linked with every other program in a running system, and is a necessary component for any fully portable application that can run on any platform the OS can run on. The only platform-specific code are in the kernel and in the supporting OS userland library, which is why they collectively form the OS.
For which we have only your word, despite every internet indexed source agreeing with my use. I think you're the one with the credibility issues frankly.
Unix is not an operating system, Unix is a trademark first and second a set of specifications used by similar yet distinct operating system distributions.
Unix was an operating system first released in 1969.
On the contrary, it provides the facilities needed to load a console, disk, networking, and many other goodies. Everything you need to build a feature rich operating system distribution.
This is basic logic. If you take X and add Y to make Z, that does not mean X is Z. The kernel is defined by what executes in privileged mode, and an operating system is defined by the kernel AND whatever executes in user space.
That Minix defines the interfaces needed to build and run consoles, disks and networking makes it a kernel. When combined with actual services that provide disks, networking and consoles, then it is an operating system.
In the case of monolithic kernels like Linux, the same argument applies to its drivers. Without drivers, Linux is not an operating system. Further, a userland must almost always be present to accomplish anything useful and thus qualify as an operating system. You can certainly run everything in kernel mode, even applications, and then the kernel happens to be the operating system, but this is almost never the case, and certainly isn't the default.
We have standard terminology to make meaningful distinctions and if you're not willing to make distinctions, terminology is meaningless. These are standard terms.
I already acknowledged this in my post. If you ship an alternate C library with Linux instead of glibc, it's no longer GNU/Linux, it's BusyBox/Linux, or what have you.
The library is part of the userland application, along with every other library used by said app.
Except it's not. The standard C library is part of the operating system used to interface with the kernel. Just because it's linked into the same address space does not mean it's somehow part of the application.
Actually in the case of a Macro kernel, the entire OS is the kernel and it is properly referred to as Linux.
This is not correct. The Linux kernel defines hardware abstractions, and you have to invoke its services using system calls by placing certain values in hardware registers, and certain values on the stack, and executing an assembly instruction to transfer control to the kernel. But apps never actually do this, they instead call the kernel services via a C library which does this for them. The Linux operating system is the Linux kernel + the userland C library. The kernel itself is never standalone, and always comes with a userland library to interface with it, and when that library is glibc, and particularly when the other userland tools are the GNU tools, Stallman is adamant in calling it GNU/Linux. He's not technically wrong either.
He's right, a kernel is not an operating system. For instance, consider the Unix OS. There is no "Unix" kernel, there are only compatible implementations of the Unix operating system interface, which is defined entirely at user-level. Some of the Unix interface may be provided in the kernel, like Linux, Solaris, AIX, and some implementations are purely in userland, like Unix binary compatibility layers for microkernel operating systems like KeyKOS, L4, and Minix.
The kernel merely forms the core part of an operating system, which includes the drivers (which may or may not ship as part of the kernel), and the userland for operating system services. Minix drivers are entirely outside the kernel for instance, so if the Minix kernel itself were a complete operating system, all you could do was boot to a blank screen. No console, no disk, no networking. Not much of an operating system.
Finally, we wouldn't invent separate terms in computing for "kernel" and "operating system" if they were synonyms. The kernel is exactly that, the "kernel/core" part of an operating system, but not synonymous with it.
Here's a thought: why not use a plant with a higher ethanol output than corn and which grows on land we can't use for corn and other food, like switchgrass. Personally, I'd rather see the fossil fuel subsidies abolished so we can get some actual competition in the energy market.
That'd drive up the cost of software development. People write buggy, insecure code because it's fast and cheap, and that's all the end user is willing to pay for.
I doubt very much that this cost would be less than creating legislation, enforcing it via criminal investigations, trying the accused in our overburdened courts, and housing these criminals in overflowing prisons. Legislation should always be the *last* recourse, not the first one.
It still amazes me how people seek legislative solutions to what are purely technical problems. Hey politicians: you're doing it wrong. If you're going to legislate something, then legislate the use of memory safe programming languages and proof carrying code. Security problems would be mostly solved, and software would have fewer bugs overall to boot.
American citizens have rights under the Constitution so they cannot be treated as "enemy combatants". The only way the American government gets away with Guantanamo is because none of the detainees are citizens, and there is no internationally recognized constitution that would protect their rights and guarantee due process.
Citizens can be charged with treason however, whch is just as bad, but the requirements are strict and established by precedent.
Cert Patrol lets you know when a cert is new to your browser. Surely that's of some value, is it not?
It depends, certs get updated all the time and you don't want to pester the user with messages he doesn't understand. That makes it easy to spoof or phish the user because they'll just get used to clicking through. As long as the cert upgrade is valid, you shouldn't see anything change IMO.
I looked at the add-ons page comments and saw the following. Do you know anything about this?
If the cert was properly upgraded, then it shouldn't give you a warning. The author of the tool is a security researcher who knows what he's doing. Read up on his petname tool paper if you're interested in further information.
Their claim that OOP is "anti-modular" is of course absurd.
No it's not. You're biased because you're used to the cognitive load associated with reasoning about open recursion, higher order programming, runtime type checking, covariance/contravariance and so on.
Not at all. For instance, interfaces can't declare type members. Well, they can in Scala, but Scala was designed with ML modules in mind. Every mainstream OO language was not.
Agreed... but aren't most modern OS's OO based? In most cases students need OO programming in order to become employable.
You're employable if you can program well in *any* language IMO (speaking as a developer who has hired others). In fact, the more wild the language, the more impressed I am.:-)
And OS's may be object-based, but not object-oriented. Big difference.
Neither of those is 100% reliable. Cert Patrol provides the user with meaningless messages that they are not equipped to understand, and so their choices reduces to "get work done" or "give up on doing any work". I also doubt that Perspectives has the critical mass needed to make its data reliable, and it's far too easy to miss a single invalid cert. The shared vulnerability is simply too large.
The only truly secure option is a petname system used to build a web of trust. At least here the window of vulnerability is small, and phishing and spoofing attacks are impossible.
Not true. This whole thread proves that CA signed certs ARE vulnerable to MITM attacks, because the CA itself is a single shared point of vulnerability.
You would have self-signed certs presented as "semi-secure", which they are not.
Even real certs are not secure as this thread demonstrates. So presenting certs as secure in the browser is committing the exact same sin that you are objecting against for self-signed certs. The only secure means to verify is secure introduction, or out of band verification.
CAs are not a valid out of band means of verification, they are a global trusted computing base (TCB), and thus are a global point of vulnerability.
Rather, I think the opposite should happen: once classified material becomes public, it should automatically become unclassified. You cannot put a genie back in a bottle, and trying beat it back in by punishing otherwise law-abiding citizens is plain stupid.
Slashdot Mirror
Big mistake IMO. By making Netflix bigger, they get more leverage against the studios, who are the real bad guys that are driving prices up. With more leverage, they can put more pressure on the studios to push prices down.
Amen. Businesses should be driven by domain experts informed by business/economics experts on staff, and not vice versa.
Sure you can, by the police.
Because the C library is part of the definition of the Unix operating system, and POSIX-compatible systems. It's further part of every Linux system currently deployed.
Excuse me, but you've provided no such thing. You've linked to Tanenbaum's textbook and claimed that it supported your definitions, but you didn't cite any of the content to support your claim. Are we supposed to just take your word that these experts agree with you? Because frankly, I've been active in the microkernel community for over 10 years (search the EROS lists and you'll find me), so I'll take my direct reading of the research literature and direct interaction with these researchers as stronger evidence than your unsupported claims.
Except it can by linking in a different but compatible library.
Legal and technical definitions are completely distinct. Sometimes they overlap, and sometimes they don't. What constitutes derivative works according to copyright is an argument that is completely separate from a technical argument about modularity.
Except it's not linked directly with your code, or even exclusively with your code, it's dynamically linked with every other program in a running system, and is a necessary component for any fully portable application that can run on any platform the OS can run on. The only platform-specific code are in the kernel and in the supporting OS userland library, which is why they collectively form the OS.
For which we have only your word, despite every internet indexed source agreeing with my use. I think you're the one with the credibility issues frankly.
Unix was an operating system first released in 1969.
This is basic logic. If you take X and add Y to make Z, that does not mean X is Z. The kernel is defined by what executes in privileged mode, and an operating system is defined by the kernel AND whatever executes in user space.
That Minix defines the interfaces needed to build and run consoles, disks and networking makes it a kernel. When combined with actual services that provide disks, networking and consoles, then it is an operating system.
In the case of monolithic kernels like Linux, the same argument applies to its drivers. Without drivers, Linux is not an operating system. Further, a userland must almost always be present to accomplish anything useful and thus qualify as an operating system. You can certainly run everything in kernel mode, even applications, and then the kernel happens to be the operating system, but this is almost never the case, and certainly isn't the default.
We have standard terminology to make meaningful distinctions and if you're not willing to make distinctions, terminology is meaningless. These are standard terms.
I already acknowledged this in my post. If you ship an alternate C library with Linux instead of glibc, it's no longer GNU/Linux, it's BusyBox/Linux, or what have you.
Except it's not. The standard C library is part of the operating system used to interface with the kernel. Just because it's linked into the same address space does not mean it's somehow part of the application.
This is not correct. The Linux kernel defines hardware abstractions, and you have to invoke its services using system calls by placing certain values in hardware registers, and certain values on the stack, and executing an assembly instruction to transfer control to the kernel. But apps never actually do this, they instead call the kernel services via a C library which does this for them. The Linux operating system is the Linux kernel + the userland C library. The kernel itself is never standalone, and always comes with a userland library to interface with it, and when that library is glibc, and particularly when the other userland tools are the GNU tools, Stallman is adamant in calling it GNU/Linux. He's not technically wrong either.
He's right, a kernel is not an operating system. For instance, consider the Unix OS. There is no "Unix" kernel, there are only compatible implementations of the Unix operating system interface, which is defined entirely at user-level. Some of the Unix interface may be provided in the kernel, like Linux, Solaris, AIX, and some implementations are purely in userland, like Unix binary compatibility layers for microkernel operating systems like KeyKOS, L4, and Minix.
The kernel merely forms the core part of an operating system, which includes the drivers (which may or may not ship as part of the kernel), and the userland for operating system services. Minix drivers are entirely outside the kernel for instance, so if the Minix kernel itself were a complete operating system, all you could do was boot to a blank screen. No console, no disk, no networking. Not much of an operating system.
Finally, we wouldn't invent separate terms in computing for "kernel" and "operating system" if they were synonyms. The kernel is exactly that, the "kernel/core" part of an operating system, but not synonymous with it.
Here's a thought: why not use a plant with a higher ethanol output than corn and which grows on land we can't use for corn and other food, like switchgrass. Personally, I'd rather see the fossil fuel subsidies abolished so we can get some actual competition in the energy market.
I doubt very much that this cost would be less than creating legislation, enforcing it via criminal investigations, trying the accused in our overburdened courts, and housing these criminals in overflowing prisons. Legislation should always be the *last* recourse, not the first one.
It still amazes me how people seek legislative solutions to what are purely technical problems. Hey politicians: you're doing it wrong. If you're going to legislate something, then legislate the use of memory safe programming languages and proof carrying code. Security problems would be mostly solved, and software would have fewer bugs overall to boot.
American citizens have rights under the Constitution so they cannot be treated as "enemy combatants". The only way the American government gets away with Guantanamo is because none of the detainees are citizens, and there is no internationally recognized constitution that would protect their rights and guarantee due process.
Citizens can be charged with treason however, whch is just as bad, but the requirements are strict and established by precedent.
It depends, certs get updated all the time and you don't want to pester the user with messages he doesn't understand. That makes it easy to spoof or phish the user because they'll just get used to clicking through. As long as the cert upgrade is valid, you shouldn't see anything change IMO.
If the cert was properly upgraded, then it shouldn't give you a warning. The author of the tool is a security researcher who knows what he's doing. Read up on his petname tool paper if you're interested in further information.
No it's not. You're biased because you're used to the cognitive load associated with reasoning about open recursion, higher order programming, runtime type checking, covariance/contravariance and so on.
Not at all. For instance, interfaces can't declare type members. Well, they can in Scala, but Scala was designed with ML modules in mind. Every mainstream OO language was not.
No, they are making the point that properly grasping OO principles requires understanding some very advanced concepts in computer science, concepts which are not appropriate for freshman.
Not it doesn't! Why do so many people think this?
You're employable if you can program well in *any* language IMO (speaking as a developer who has hired others). In fact, the more wild the language, the more impressed I am. :-)
And OS's may be object-based, but not object-oriented. Big difference.
Neither of those is 100% reliable. Cert Patrol provides the user with meaningless messages that they are not equipped to understand, and so their choices reduces to "get work done" or "give up on doing any work". I also doubt that Perspectives has the critical mass needed to make its data reliable, and it's far too easy to miss a single invalid cert. The shared vulnerability is simply too large.
The only truly secure option is a petname system used to build a web of trust. At least here the window of vulnerability is small, and phishing and spoofing attacks are impossible.
That's not necessary. Provide the cert as a download on their own https site.
Not true. This whole thread proves that CA signed certs ARE vulnerable to MITM attacks, because the CA itself is a single shared point of vulnerability.
Even real certs are not secure as this thread demonstrates. So presenting certs as secure in the browser is committing the exact same sin that you are objecting against for self-signed certs. The only secure means to verify is secure introduction, or out of band verification.
CAs are not a valid out of band means of verification, they are a global trusted computing base (TCB), and thus are a global point of vulnerability.
Rather, I think the opposite should happen: once classified material becomes public, it should automatically become unclassified. You cannot put a genie back in a bottle, and trying beat it back in by punishing otherwise law-abiding citizens is plain stupid.
The Android brand.