If you rely on safe mode to protect your system from your users, and your users from each other, then you have already failed.
The only way to get this kind of security is to rely on the operating system to provide it for you; this is done by running PHP interpreters belonging to different security contexts as seperate users.
With such a setup, the worst the user can do is screw up their own files (boo hoo!).:)
Surely in a real-world situation you would run the C# app in a separate process and proxy requests from your frontend web server to the separate process via HTTP, FastCGI or SCGI.
Or perhaps even build an interpreter into the web server itself (mod_mono).
So, you admit that if you sing "Happy Birthday" in public you will affect the profits of the person or persons that owns or own the various rights to the work. This has an effect on the flow of money within the United States and therefore may be regulated by congress due to the interstate commerce clause of the US Constitution.
This is a moot point. They are not claiming that they own the number. They are claiming that it is, a "technology, product, service, device, component, or part thereof that that is primarily designed or produced for the purpose of circumventing a technological measure that effectively controls access to a work protected under this title". Such technologies, products, services, devices, components or parts thereof that are illegal to possess/distribute in the United States due to the DMCA.
FOR THE LOVE OF GOD, people, do thirty seconds of basic research before getting all high and preachy about how numbers cannot be owned (as if the post that you just made, and to which you own the copyright, is not just a number...
Interesting... but tracking the revisions of a file by name has some limitations. What happens if I rename a file (also to another directory)? What happens if I rename a directory itself? Is the file metadata (owner, access permissions, modification times, extended attributes (including selinux labels, ACLs and user extended attributes)) versioned?
I guess some of this info is on the project's home page, which is down at the moment...
It's certainly high enough to go over most people's heads...
Re:This is (now) a famous number-theory integer!
on
Censoring a Number
·
· Score: 1
You clearly can copyright a number; a verse or a song *is* just a number. I guess the question is, what is the threshold for a number being copyrightable?
Anyway, whether or not the number can be copyrighted is not an issue here, as no one is alleging that their copyrights are being infringed; the AACS authority is sending takedown notices based on the fact that the number forms part of a method that bypasses the effective copy protection mechanism that is AACS; such information is illegal to have/distribute in the United States.
Nice, though I can't quite see how you encoded it...
I chose to make mine as square as possible. The image and source code to generate it are at http://robots.org.uk/src/tiil/. In fact, the source will encode any arbitrary number into a square (or almost square) image, with pixels representing bits in increasing order of significance.
The result would make a nice web site icon for those 'in the know'.:)
More like, thank you Randall311 for causing the malware vendors to continue to fund Microsoft (and thereby all their other anti-competitive, anti-freedom efforts).
I think Symantec and the other malware vendors would object to this practice, since the whole point of the exercise is to get the software installed before the user uses their PC for the first time.:)
Freeview is an initiative that aims to discourage us from watching television, specifically by making it really annoying to channel-hop. Let me explain. With our existing, analogue equipment, changing channels is nice and fast; it takes perhaps half a second. But the analogue transmitters are going to be shut off in a few years time. Once that happens, the only way to recieve television signals will be with a cable or sky box, or a Freeview box, each of which takes between two and five seconds to do whatever it is that they do to change channels. The result is that I abandon television out of frustration.
It seems slashdot munged one of my questions. I wanted to know what the equivalent of the Location and Directory directives were. It's been a while since I had to use IIS (thank god), but I don't remember any way to implement the full functionality of Location and Directory, for instance how they can overlap and their directives combine.
I certainly wouldn't want to be in the business of configuring web servers if I couldn't use mod_macro or mod_perl!
I'm pretty sure that IIS can be made to do many of the cool things that are only an apt-get install away with Apache. But you're then going to have to hunt down loads of code from disparate places accross the web, watch out for security vulnerabilities in each thing you donwload, keep them all updated separately; it's Windows in a nutshell. But with Apache, all the cool features (and those are just some of the one's that I am familiar with) are just an apt-get away. And if you want to extend the server yourself, the source code is only an 'apt-get source' away... with IIS, you are stuck with the API Microsoft provides you with, and you have to deal with all the other annoyances that make Windows a crap platform for development.
IMO it is the top-down, x.500-inspired PKI that needs to change to become more like PGP's system. The current system is fairly useless at the moment, but it is so, so profitable for the lucky few companies who are blessed by having their CA certificates included by operating system and application vendors.
It's been called various things over its lifetime... it started off as the GRE. These days it's called xulrunner/libxul possibly. Eventually Firefox and Thunderbird, etc. will hopefully both link against a shared copy. This hasn't happened already because it's not a priority for the developers, and because managing any kind of shared resource is a huge pain on most operating systems, but systems like Debian are already linking third-party apps against a shared xulrunner (in the form of libxul0d, libnss3-0d and libnspr4-0d).
You should really install Debian. Just think of all the extra productivity you will have now that you're not recompiling every time you want to enable a feature!
Your cute little straw man is more telling that you realise. What are IIS' equivalents of the and directives? How can I use TLS SNI with IIS? How do I use IIS as a backend for a subversion server? How do I enable AccessFileName functionality on IIS? How can I set up filters that deny requests based on the occurance of suspicious strings in their headers? How can I have an IIS server publish its virtual hosts via DNS-SD? How can I set it up to fire off requests for different virtual hosts as different system users, so that I can grant different priviliges to different web applications? How can I have IIS get the value that it uses for the Content-Type header of its responses from an extended attribute on the file being served, rather than a dumb lookup in a table of file extension (god, don't get me STARTED on file extensions!) to media type? How can I embed arbitrary Perl code into my IIS config that is run when the server starts up?
Slashdot Mirror
No, you're right. ;)
If you rely on safe mode to protect your system from your users, and your users from each other, then you have already failed.
:)
The only way to get this kind of security is to rely on the operating system to provide it for you; this is done by running PHP interpreters belonging to different security contexts as seperate users.
With such a setup, the worst the user can do is screw up their own files (boo hoo!).
Surely in a real-world situation you would run the C# app in a separate process and proxy requests from your frontend web server to the separate process via HTTP, FastCGI or SCGI.
Or perhaps even build an interpreter into the web server itself (mod_mono).
Spped problems eliminated.
Current interpretations of copyright law are based on the interstate commerce clause, not the 'progress of arts & science' clause.
So, you admit that if you sing "Happy Birthday" in public you will affect the profits of the person or persons that owns or own the various rights to the work. This has an effect on the flow of money within the United States and therefore may be regulated by congress due to the interstate commerce clause of the US Constitution.
This is a moot point. They are not claiming that they own the number. They are claiming that it is, a "technology, product, service, device, component, or part thereof that that is primarily designed or produced for the purpose of circumventing a technological measure that effectively controls access to a work protected under this title". Such technologies, products, services, devices, components or parts thereof that are illegal to possess/distribute in the United States due to the DMCA.
Well said. Shame about the interstate commerce clause.
AFAIK, AACS is just AES. So they key is just 128 bits of random data; it does not have to have any other special qualities.
HAND!
No one is claiming that they own the number!
FOR THE LOVE OF GOD, people, do thirty seconds of basic research before getting all high and preachy about how numbers cannot be owned (as if the post that you just made, and to which you own the copyright, is not just a number...
Like http://www.chillingeffects.org/notice.cgi?sID=3218 : note the second URL in the takedown notice...
Oh I see, you have the most significant digit at the top left... I put the least significant digit there.
"Theft" how?
And of what IP?
Make a specific allegation or stop trolling, please.
Interesting... but tracking the revisions of a file by name has some limitations. What happens if I rename a file (also to another directory)? What happens if I rename a directory itself? Is the file metadata (owner, access permissions, modification times, extended attributes (including selinux labels, ACLs and user extended attributes)) versioned?
I guess some of this info is on the project's home page, which is down at the moment...
It's certainly high enough to go over most people's heads...
You clearly can copyright a number; a verse or a song *is* just a number. I guess the question is, what is the threshold for a number being copyrightable?
Anyway, whether or not the number can be copyrighted is not an issue here, as no one is alleging that their copyrights are being infringed; the AACS authority is sending takedown notices based on the fact that the number forms part of a method that bypasses the effective copy protection mechanism that is AACS; such information is illegal to have/distribute in the United States.
Nice, though I can't quite see how you encoded it...
:)
I chose to make mine as square as possible. The image and source code to generate it are at http://robots.org.uk/src/tiil/. In fact, the source will encode any arbitrary number into a square (or almost square) image, with pixels representing bits in increasing order of significance.
The result would make a nice web site icon for those 'in the know'.
Use some imagination!
More like, thank you Randall311 for causing the malware vendors to continue to fund Microsoft (and thereby all their other anti-competitive, anti-freedom efforts).
Real modems were not on the market? ;)
I think Symantec and the other malware vendors would object to this practice, since the whole point of the exercise is to get the software installed before the user uses their PC for the first time. :)
Freeview is an initiative that aims to discourage us from watching television, specifically by making it really annoying to channel-hop. Let me explain. With our existing, analogue equipment, changing channels is nice and fast; it takes perhaps half a second. But the analogue transmitters are going to be shut off in a few years time. Once that happens, the only way to recieve television signals will be with a cable or sky box, or a Freeview box, each of which takes between two and five seconds to do whatever it is that they do to change channels. The result is that I abandon television out of frustration.
It seems slashdot munged one of my questions. I wanted to know what the equivalent of the Location and Directory directives were. It's been a while since I had to use IIS (thank god), but I don't remember any way to implement the full functionality of Location and Directory, for instance how they can overlap and their directives combine.
I certainly wouldn't want to be in the business of configuring web servers if I couldn't use mod_macro or mod_perl!
I'm pretty sure that IIS can be made to do many of the cool things that are only an apt-get install away with Apache. But you're then going to have to hunt down loads of code from disparate places accross the web, watch out for security vulnerabilities in each thing you donwload, keep them all updated separately; it's Windows in a nutshell. But with Apache, all the cool features (and those are just some of the one's that I am familiar with) are just an apt-get away. And if you want to extend the server yourself, the source code is only an 'apt-get source' away... with IIS, you are stuck with the API Microsoft provides you with, and you have to deal with all the other annoyances that make Windows a crap platform for development.
IMO it is the top-down, x.500-inspired PKI that needs to change to become more like PGP's system. The current system is fairly useless at the moment, but it is so, so profitable for the lucky few companies who are blessed by having their CA certificates included by operating system and application vendors.
It's been called various things over its lifetime... it started off as the GRE. These days it's called xulrunner/libxul possibly. Eventually Firefox and Thunderbird, etc. will hopefully both link against a shared copy. This hasn't happened already because it's not a priority for the developers, and because managing any kind of shared resource is a huge pain on most operating systems, but systems like Debian are already linking third-party apps against a shared xulrunner (in the form of libxul0d, libnss3-0d and libnspr4-0d).
You should really install Debian. Just think of all the extra productivity you will have now that you're not recompiling every time you want to enable a feature!
Your cute little straw man is more telling that you realise. What are IIS' equivalents of the and directives? How can I use TLS SNI with IIS? How do I use IIS as a backend for a subversion server? How do I enable AccessFileName functionality on IIS? How can I set up filters that deny requests based on the occurance of suspicious strings in their headers? How can I have an IIS server publish its virtual hosts via DNS-SD? How can I set it up to fire off requests for different virtual hosts as different system users, so that I can grant different priviliges to different web applications? How can I have IIS get the value that it uses for the Content-Type header of its responses from an extended attribute on the file being served, rather than a dumb lookup in a table of file extension (god, don't get me STARTED on file extensions!) to media type? How can I embed arbitrary Perl code into my IIS config that is run when the server starts up?