Software driven radio. While this sound cool you are getting nowhere without some hardware that will transform analog signals (In cellular with a dynamic range of more than 100 dB (1e10)) to
digital. While it might be possible to build
such an A/D converter it will use much to much current unless you use some trickery in getting rid
of the dynamic range before sampling.
This transformation is what the hardware in modern radios do. This hardware is what is not covered in the gnu project.
What is covered in the gnu project is digital modulator and demodulator at baseband. While this is very nice it is not revolutionary.
There are companies that make and market "secure phones". They use the circuit switched data function in GSM. They make their own speech codec and once the speach is crammed into a 9600 b/s or similar data stream they encypt it using DES,AES or whatever and transmits the stuff to the phone in the other end that decrypts it.
The CCC stunt makes it possible to clone your
SIM-card in the case where your operator have
chosen to use an algorithm called COMP128.
(It enables you to extract Ki from the SIM-card) but requires you know the PIN-code and have access to the card since it is a chosen plain-text attack that requires in average 100000 16 byte data words.
Ki is a 16 byte secret key known to the SIM-card and to your home operator. In the GSM system session keys are transferred from the operator to the SIM-card in the handheld at regular intervals and each time the phone is tiurned on. The session keys are encrypted using an algorithm that is in effect a hash function of the Ki and the session key generating a set of 12 byte encrypted session keys. The operator is free to chose the hash algorithm but originally this comp128 was passed along as a demonstration.
Data encryption is using the encrypted session keys and another algorithm.
It is left as an exercise to the interested reader to figure out why the SIM-card hash function f(16 bytes key,16 bytes Ki)->12 bytes data, is not usefull for data encryption:-).
This is just as wrong as if a country mandated 80% Microsoft. Mandate open file formats and protocols, but don't mandate people or agencies MUST use a specific type of software.
They mandate that they will only use software for which they do not get the source/control for 20% of their computers. They do not include or exclude any software, they just state a requirement for what they want for the majority of their software.
No companies are excluded. If M$, Oracle or anybody else wants to bid for contracts in Brazil they wil be welcome along side anybody else, as long as they have a product that conforms to the requirements of the Brazilian government. It is not that any different than when the DoD demands software to be written in ADA and hardware specified in VHDL.
Since both Linux and Unixware are supposed to be POSIX compliant it is not so strange that the system calls are similar. That is actually the whole purpose of POSIX compliance. The thing is how the system calls are implemented.
Most likely both the guys writing for SCO and the guys writing for Linux would use the variable names used in the POSIX spec which are most likely enherited from the BSD/SysV implementation.
Clearly, if Linksys got it working on their own access point...., it is very likely that they would have to change the source to accomodate the difference in hardware.
If they are making their own hardware it should not be to difficult to design the hardware so Linux could run on it without modifications. Hardware is often more flexible than software, especially if you do not have the full overview of the software you are trying to change.
I find it very probable that they could have made the box without changing any GPLed code and then added their own applications/drivers in usermode/module form. Before jumping the gun on them it might be worth for an owner of their product to ask for a copy of whatever GPLed source they are using and in case they refuse I would be surprised if RSM&Co were not ready with a lawsuit.
There is no reason to debase oneself the SCO way and sue someone without first asking for an amiable solution.
I like your style pal. RSA and Diffie Helman would also have been my secret candidates but I dont see how Rivest Shamir etc profited from their US patent. Could they not have made the same impact without a patent?
BTW, the moderators here have a sick sense of humour. How do you get a +5 rating for Funny with you post?:-)
Greeting from the hood. I am located in Ulm so happy trails Schulz;-)
Granted some (most) software patents are ridiculous (like the Amazon 1-Click patent - that's akin to patenting the doorknob), but some aren't.
Hmm., can anybody give a concrete example of a software patent that actually makes sense? In the EU they are currently lobbying for the introduction of software patents because not having software patents are presumably giving american companies an advantage(?strange argument, neither European nor American companies can patent software in Europe, but either can patent the most obvious goblelygook in the US)
At BBC, theregister etc there is a story about Sendo suing Orange over patent infringement on the Orange SPV Smartphony. The link is at Sendo vs Orange
It seems M$ should have bought a licence to Sendos patents before giving it away to HTC and others.
I hate to say it but WINE will newer fly in any meaning-full way. The only reason anybody would want to run WINE would be to run M$ applications. Non M$ applications can just as well be compiled for Linux or OS-X.
If WINE gets successfull, M$ can destroy it in 10 min. All they have to do is to issue a security warning and a patch to M$ office and win2k3 that would let office use a new undocumented system call. WINE would roll over and die.
So needing to explain the filesystem means it is not intuitive.
True, but in that case I doubt you will be able to design anything on a computer that can be used without prior knowledge. That the little plastic thingie besides the larger plastic thingie in front of the glass plate with images on, can be used to manipulate a group of dots on the glass plate - and that you, if you move the group of dots over another group of dots that looks like a small picture and presses an area of the little plastic thingie can make new images appear and that this has any relevance to the user, is also not self-evident.
A computer is a tool and in general you need some kind of introduction to an effective tool in order to use it. If you have never seen a wheel before, I doubt you would guess the usefullness of a wheel-barrow or a bicycle or a pair of roller-skates without a helping hand from a by-stander. The process of introducing one-another to new concepts is what we call culture.
I will actually claim that it is both clear and intuitive if you consider the scope.
/etc,/bin,..etc. are for the system admin to worry about and in his/my scope it is a fairly straight-forward way to organize things. For a system user, the home-dir is the main reference point. I have still to meet a person I cannot explain the concept of files and directories and the file-system tree structure in the matter of minutes.
And that's the point of this project and several other next-gen file systems in development now... Presenting users with a unified and organized interface that shows them their data in a way they can find it easily.
Exactly what everybody wants. I have a unified and organized interface. I call it the filesystem. It has a root directory and under this root I have files and sub-directories. It is clear and intuitive and it gives great flexibility in how to store and organize your data.
Then IBM (or SCO, or somebody) will have to define what those "certain aspects of the kernel" are, and they will be replaced by code written by people who have never worked for IBM or SCO. If IBM wants to maintain a "SCO-Fork" of the kernel, more power to them.
Except they cannot distribute a proprietary version since it contains GPLed code. The only way around this would be for them to release the fork under GPL or to re-write the proprietary parts as kernal-modules and only distribute these modules as binaries.
Telephony might just be where you see IPv6 being deployed first.
Telephony is by definition peer-to-peer so you are stuck if you are hidden behind a NAT. Even if you confined VoIP to a class A network like 10.255.255.255 you would only have a little more than 16 million available numbers.
IPv6 is also prepared for QOS which will be a good thing for telephony.
At some point we have to ask ourselves how much of a control system it is safe
to build. Power corrupts, and total power corrupts totally.
I dont like to play the part of a off-the-rocker libertarian, but how much information about individuals is it safe to allow the Ministry of Information Retrieval to collect?
Let us say we find it a good idea to have public places filmed and install automatic tracking of vehicles, why stop there. Let us do automatic tracking of the whereabouts of all cell-phones as well. When we get RFID tags in most stuff
we buy why not track the whereabouts of peoples wallets or underwear - Just to make sure we can locate criminals fast in case we miss them. When we know where
people are we also know who they meet.
Again, this is nothing that should worry law-abiding citizens.
Exactly what surveillance measures would people in general object against?
The problem is that services like SMS, MMS etc are part of the GSM/WCDMA standard and supported by their infrastructure. These services are not available using IS-95 and friends.
The situation is a little bit akin to the difficulty involved in having a message that originated as a paper christmas card delivered as an E-mail or TV broad-cast message.
Re:Cringley, Linus, and Christoph Hellwig
on
Today's SCO News
·
· Score: 1
??? Where should this UNIX support for main-frames come from? AIX is not available for main-frames, who else have made a UNIX version for IBM mainframes?
IBM might at some point have made an AIX version for research purposes to run on a 3090, but where would SCO have gotten hold of IBMs code, and how can an IBM extension to their own code become SCO property?
Yeah, but most people who have DVDs don't want to watch them on the computer.
But wont they need to watch the ripped of copies on a computer?
I am using ogle for watching DVDs and I have no complaints. I dont have a windows license I can dual-boot into and I guess neither can most Solaris/HP-UX/AIX/SCO^h^h^hCMS users.
One could add that having DeCSS would enable you to rip a DVD and make a un-encrypted copy. In this respect it could be used for circumventing the copy protection. Nevertheless I will risk the claim that 99.9% of the people using DeCSS is doing so to watch DVDs they have purchased for their hard earned cash.
If you wanted a pirated version it is easier to download one than to make one yourself;-)
Again, you dont need DeCSS to copy a DVD, you need it to be able to decode its content.
Making a bit by bit copy of a DVD will play flawlessly in any DVD player, no problem. The problem comes when you want to build your own DVD player, then you need DeCSS.
Slashdot Mirror
Software driven radio. While this sound cool you are getting nowhere without some hardware that will transform analog signals (In cellular with a dynamic range of more than 100 dB (1e10)) to digital. While it might be possible to build such an A/D converter it will use much to much current unless you use some trickery in getting rid of the dynamic range before sampling.
This transformation is what the hardware in modern radios do. This hardware is what is not covered in the gnu project.
What is covered in the gnu project is digital modulator and demodulator at baseband. While this is very nice it is not revolutionary.
The counterpane announcement is about CDMA and US-TDMA phones, not GSM
Since GSM is growing fast in the US FBI would need some other approach.
There are companies that make and market "secure phones". They use the circuit switched data function in GSM. They make their own speech codec and once the speach is crammed into a 9600 b/s or similar data stream they encypt it using DES,AES or whatever and transmits the stuff to the phone in the other end that decrypts it.
Close but no cigar.
The CCC stunt makes it possible to clone your SIM-card in the case where your operator have chosen to use an algorithm called COMP128. (It enables you to extract Ki from the SIM-card) but requires you know the PIN-code and have access to the card since it is a chosen plain-text attack that requires in average 100000 16 byte data words.
Ki is a 16 byte secret key known to the SIM-card and to your home operator. In the GSM system session keys are transferred from the operator to the SIM-card in the handheld at regular intervals and each time the phone is tiurned on. The session keys are encrypted using an algorithm that is in effect a hash function of the Ki and the session key generating a set of 12 byte encrypted session keys. The operator is free to chose the hash algorithm but originally this comp128 was passed along as a demonstration.
Data encryption is using the encrypted session keys and another algorithm.
It is left as an exercise to the interested reader to figure out why the SIM-card hash function f(16 bytes key,16 bytes Ki)->12 bytes data, is not usefull for data encryption :-).
They mandate that they will only use software for which they do not get the source/control for 20% of their computers. They do not include or exclude any software, they just state a requirement for what they want for the majority of their software.
No companies are excluded. If M$, Oracle or anybody else wants to bid for contracts in Brazil they wil be welcome along side anybody else, as long as they have a product that conforms to the requirements of the Brazilian government. It is not that any different than when the DoD demands software to be written in ADA and hardware specified in VHDL.
Since both Linux and Unixware are supposed to be POSIX compliant it is not so strange that the system calls are similar. That is actually the whole purpose of POSIX compliance. The thing is how the system calls are implemented.
Most likely both the guys writing for SCO and the guys writing for Linux would use the variable names used in the POSIX spec which are most likely enherited from the BSD/SysV implementation.
If they are making their own hardware it should not be to difficult to design the hardware so Linux could run on it without modifications. Hardware is often more flexible than software, especially if you do not have the full overview of the software you are trying to change.
I find it very probable that they could have made the box without changing any GPLed code and then added their own applications/drivers in usermode/module form. Before jumping the gun on them it might be worth for an owner of their product to ask for a copy of whatever GPLed source they are using and in case they refuse I would be surprised if RSM&Co were not ready with a lawsuit.
There is no reason to debase oneself the SCO way and sue someone without first asking for an amiable solution.
I like your style pal. RSA and Diffie Helman would also have been my secret candidates but I dont see how Rivest Shamir etc profited from their US patent. Could they not have made the same impact without a patent?
BTW, the moderators here have a sick sense of humour. How do you get a +5 rating for Funny with you post? :-)
Greeting from the hood. I am located in Ulm so happy trails Schulz;-)
Hmm., can anybody give a concrete example of a software patent that actually makes sense? In the EU they are currently lobbying for the introduction of software patents because not having software patents are presumably giving american companies an advantage(?strange argument, neither European nor American companies can patent software in Europe, but either can patent the most obvious goblelygook in the US)
At BBC, theregister etc there is a story about Sendo suing Orange over patent infringement on the Orange SPV Smartphony. The link is at Sendo vs Orange
It seems M$ should have bought a licence to Sendos patents before giving it away to HTC and others.
I hate to say it but WINE will newer fly in any meaning-full way. The only reason anybody would want to run WINE would be to run M$ applications. Non M$ applications can just as well be compiled for Linux or OS-X.
If WINE gets successfull, M$ can destroy it in 10 min. All they have to do is to issue a security warning and a patch to M$ office and win2k3 that would let office use a new undocumented system call. WINE would roll over and die.
True, but in that case I doubt you will be able to design anything on a computer that can be used without prior knowledge. That the little plastic thingie besides the larger plastic thingie in front of the glass plate with images on, can be used to manipulate a group of dots on the glass plate - and that you, if you move the group of dots over another group of dots that looks like a small picture and presses an area of the little plastic thingie can make new images appear and that this has any relevance to the user, is also not self-evident.
A computer is a tool and in general you need some kind of introduction to an effective tool in order to use it. If you have never seen a wheel before, I doubt you would guess the usefullness of a wheel-barrow or a bicycle or a pair of roller-skates without a helping hand from a by-stander. The process of introducing one-another to new concepts is what we call culture.
I will actually claim that it is both clear and intuitive if you consider the scope.
/etc, /bin,..etc. are for the system admin to worry about and in his/my scope it is a fairly straight-forward way to organize things. For a system user, the home-dir is the main reference point. I have still to meet a person I cannot explain the concept of files and directories and the file-system tree structure in the matter of minutes.
Exactly what everybody wants. I have a unified and organized interface. I call it the filesystem. It has a root directory and under this root I have files and sub-directories. It is clear and intuitive and it gives great flexibility in how to store and organize your data.
Except they cannot distribute a proprietary version since it contains GPLed code. The only way around this would be for them to release the fork under GPL or to re-write the proprietary parts as kernal-modules and only distribute these modules as binaries.
Telephony might just be where you see IPv6 being deployed first.
Telephony is by definition peer-to-peer so you are stuck if you are hidden behind a NAT. Even if you confined VoIP to a class A network like 10.255.255.255 you would only have a little more than 16 million available numbers.
IPv6 is also prepared for QOS which will be a good thing for telephony.
IPv6 support QOS.
At some point we have to ask ourselves how much of a control system it is safe to build. Power corrupts, and total power corrupts totally.
I dont like to play the part of a off-the-rocker libertarian, but how much information about individuals is it safe to allow the Ministry of Information Retrieval to collect?
Let us say we find it a good idea to have public places filmed and install automatic tracking of vehicles, why stop there. Let us do automatic tracking of the whereabouts of all cell-phones as well. When we get RFID tags in most stuff we buy why not track the whereabouts of peoples wallets or underwear - Just to make sure we can locate criminals fast in case we miss them. When we know where people are we also know who they meet.
Again, this is nothing that should worry law-abiding citizens.
Exactly what surveillance measures would people in general object against?
Monitoring all of us 24/7 will naturally make law enforcement so much easyer. Life in Oceania 2003.
Why should any law-abiding citizen object to a two-way TV monitor in their living rooms to help inform them on the war against terrorism.
The problem is that services like SMS, MMS etc are part of the GSM/WCDMA standard and supported by their infrastructure. These services are not available using IS-95 and friends.
The situation is a little bit akin to the difficulty involved in having a message that originated as a paper christmas card delivered as an E-mail or TV broad-cast message.
??? Where should this UNIX support for main-frames come from? AIX is not available for main-frames, who else have made a UNIX version for IBM mainframes?
IBM might at some point have made an AIX version for research purposes to run on a 3090, but where would SCO have gotten hold of IBMs code, and how can an IBM extension to their own code become SCO property?
But wont they need to watch the ripped of copies on a computer?
I am using ogle for watching DVDs and I have no complaints. I dont have a windows license I can dual-boot into and I guess neither can most Solaris/HP-UX/AIX/SCO^h^h^hCMS users.
I might be biased by the fact that without DeCSS I would not be able to use any of my DVDs for anything more elaborate than as small frisbees.
I would hazard a guess that maybe 20% of DeCSS usage is only for watching the disc. The remaining 80% are the people who create the DVD rips.I find it illogical that people should be more interested in ripping the DVDs than in watching them.
One could add that having DeCSS would enable you to rip a DVD and make a un-encrypted copy. In this respect it could be used for circumventing the copy protection. Nevertheless I will risk the claim that 99.9% of the people using DeCSS is doing so to watch DVDs they have purchased for their hard earned cash.
If you wanted a pirated version it is easier to download one than to make one yourself;-)
Again, you dont need DeCSS to copy a DVD, you need it to be able to decode its content.
Making a bit by bit copy of a DVD will play flawlessly in any DVD player, no problem. The problem comes when you want to build your own DVD player, then you need DeCSS.