I wouldn't hold my breath waiting for strong link encryption. The last time I looked at wireless LAN transceivers, the products that had link encryption were limited to NSA friendly 40-bit keys. Digital wireless phones are even worse. The NSA and FBI say "jump", the vendors ask "how high?".
Don't you think that the folks who wrote such reliable software have the skills to update their applications to support four-digit numbers? I do.
You mean the programmers that haven't died, retired, quit or been downsized or outsourced.
I know of many systems where none of the original engineers and programmers are still around, not to mention source code, compilers and development systems.
Fortunately, most of these systems are being replaced by newer systems.
The scary thing is that there is no manual backup for most of these systems. The people who had that knowledge were eliminated as part of the cost savings of the new, automated systems.
Lockheed-Martin has a web page for their Silent Sentry system. Not much in the way of technical description but it can't be that secret if they put it on the web.
Graphite was used as a moderator. The control rods were boron carbide, most were inserted from the top of the reactor. The fuel assemblies were inside water-filled pressure tubes. The reactor core was contained in a concrete line cavity with a steel cap.
The main problem with the reactor design was that it was unstable at low power levels. Steam bubbles would result in increased power output, creating more steam bubbles. The control rods also displaced water from the lower part of the reactor core as they were inserted, which increased the power output.
Deuterium and Tritium are often used in fission weapons to boost their explosive yield. Deuterium-Tritium gas is injected into the center of the fissile core. During the detonation of the weapon, the Deuterium and Tritium fuse, releasing a burst of neutrons that accelerate the fission process and improve the efficiency of the weapon.
And about a week later, the background radiation count in the U.S. spiked, to levels not seen since the late 1950's. It wasn't Chernobyl, it was the DOE "secretly" cracking open two bad missle silos and a melted reactor, hoping that the newspaper headlines about fallout arriving from Chernobyl would cover their actions.
That must have been the same week that aliens abducted the President of the United States and the CIA stole your tin foil hat.
I wouldn't bother, make the code readable and obvious.
I used to do source code optimization when C compilers were not good at optimization and the code could run much faster if the source code was tweaked to effectively use registers and avoid slow address computations.
Modern compilers have much better optimizers and it has become more difficult to optimize code for the CPUs of today. Counting cycles in a program is much more difficult than it used to be. When you have families of processors, the optimal code for one processor may be slow on another.
I used to use pointers whenever possible since experience had taught me that array references were always slower than pointer references. Then I discovered that array references were substantially faster than pointer references on some code that I was writing for a Pentium. The code was cleaner, easier to understand, smaller and faster when I used arrays.
If you look at the output of a good compiler, you will often be surprised at the optimization tricks that you would have never thought of.
If that same argument was applied to the First Amendment (Freedom of Speech), the federal government could censor any speech that used television, radio or high-speed assault printing presses.
In my opinion, the Second Amendment should protect my right to possess any weapon that the U.S. Army issues to an infantry soldier, including assault rifles, pistols, grenades and light machine guns. These are the modern equivalent of the musket.
Anyone who sends unencrypted text or audio over satellite transponders or radio/microwave links should not be surprised that other people are listening.
The NSA, KGB, GCHQ and other intelligence agencies have been doing this for decades. They are not going to stop just because some people have the illusion that their email and conversations are private.
European countries have been intercepting and reading paper mail in "black chambers" for hundreds of years.
Computers don't normally store numeric data as digits.
That wasn't true in the old days. Not only were memory sizes small but CPUs were slow. Many did not have multiply/divide hardware. Conversions between ASCII/EBCDIC character strings and binary integers were very slow and avoided whenever possible. Most business oriented systems used some variant of BCD to store numbers. BCD can be converted to/from character codes very quickly.
Every time they publish a review on something (cars, stereo equipment, etc.), every enthusiast of that group just laughs. CR's reviews are oversimplified and cater to the mainstream (aka uninformed) population.
Well, that is the point. Their audience is the average consumer, not the enthusiast.
I recently bought a car and used CR's reviews to help select the make/model. I'm sure Car & Driver would have hated it and told me to buy some overpriced, unreliable muscle car. That's fine, I was looking for reliable, safe transportation, not a sports car.
The bit about 1972 is correct. The calendar has a 28 year cycle before it repeats. If you set the computer's clock to 1972, assuming the operating system doesn't have a later epoch date, the days, dates and leap years will be the same as in 2000. Some people have used this to "fix" their Y2K computer problems. The only problem is that you have to subtract 28 before inputting a year and add 28 to any year in the output.
And "the emissions sensors might be misreporting the date" - Umm, emissions sensors don't report the date, they have no need to. They report what they're supposed to be sensing, nothing more.
That was the only semi-realistic bit about the power plant problems. There was a Y2K bug that shut down a power plant during testing that was caused by a problem with the flue gas sensor. The date discontinuity screwed up the algorithm that computed the flue gas temperature rate-of-change and produced a bogus value. This would have lead to an automatic shutdown of the plant if it hadn't been fixed.
I was pissed off by the anti-nuclear propaganda implicit in the assumption that a cooling failure was guaranteed to result in a meltdown, breech of the containment and large plumes of Plutonium wafting over the country. Also, reactors are contained in pressure vessels containg water at high temperature and pressure. So why do the show control rods sticking out of the surface of something that looks like a big swimming pool while people are running around in the same area. If the core temperature was really as high as shown, all of that water would be flashed to steam, which would ruin someone's day.
I was impressed that they got that sort of performance out of the Windows NT 2000 TCP/IP stack. The Windows NT 4.0 TCP/IP stack was infamous for running like a crippled pig on 100Base-T. They must have made some major changes to the TCP/IP stack.
Does anyone have any performance figures for Linux or *BSD on 1000Base-T?
Why must each BIT be encrypted? What's wrong with encrypting streams on a "packet" level? Or at least step up to the byte level.
There may not be any packets to encrypt. The data may already be encrypted or may not have any structure. For example, delta modulation converts an audio signal to a stream of bits, not bytes or packets.
A stream encryption device allows you to encrypt an arbitrary bit stream at high speed and minimum delay. This is ideal for link encryption, such as point-to-point data lines.
The fact is, single-bit encryption Cannot be done securly.
I would disagree. Each cipher text bit can be computed as the exclusive-or of the plain text bit and a complex function of the preceding N bits (plain or cipher or key).
If the stream encryption algorithm is being used as a key generator, the algorithm's output xor'd with the plain text, the receiver is not going to be able to strip the key generator stream from the received data. A key generator is going to have a finite length session key that must be common to the transmitter and receiver. The data transmission is going to be variable length. There must be an out-of-band and secure method of sending the session key to the receiver.
I've worked with systems that use convolutional encoding to protect data streams against noise. The encoding and decoding algorithms are very different. The problem is that the encoding algorithm is much simpler than the decoding algorithm. The generator polynomial could be viewed as a private key.
Re:It's all how you look at it
on
Copyright!
·
· Score: 1
I would argue that the company or individual has no "right" to their creation, other than property rights to the physical instances that they create. Patents and copyrights are better viewed as privileges granted by the state for the purpose of improving the public welfare, not as natural rights.
The NSA also has some facilities in Texas and other locations. See this for a list of NSA facilities. Just because you haven't heard of it doesn't mean that it doesn't exist.
The US encryption export policies used to make sense, about 30 years ago, when encryption devices were mainly used by the military and government. The invention of the microprocessor, DES and the proliferation of data networks changed the situation.
The NSA's nightmare would probably be widespread distribution and adoption of software that would automatically encrypt all data communications, without any effort on the part of the user.
1.The most powerful nuclear devices yet invented are fusion-based; they get the majority of their energy by assembling atoms of helium from isotopes of hydrogen.
Are you sure? I thought that most of the yield from large thermonuclear devices (fission-fusion-fission) came from the fission of the U238 jacket in the fusion stage. A "clean" but much lower yield device can be built by using a lead jacket.
I wouldn't hold my breath waiting for strong link encryption. The last time I looked at wireless LAN transceivers, the products that had link encryption were limited to NSA friendly 40-bit keys. Digital wireless phones are even worse. The NSA and FBI say "jump", the vendors ask "how high?".
You mean the programmers that haven't died, retired, quit or been downsized or outsourced.
I know of many systems where none of the original engineers and programmers are still around, not to mention source code, compilers and development systems.
Fortunately, most of these systems are being replaced by newer systems.
The scary thing is that there is no manual backup for most of these systems. The people who had that knowledge were eliminated as part of the cost savings of the new, automated systems.
Lockheed-Martin has a web page for their Silent Sentry system. Not much in the way of technical description but it can't be that secret if they put it on the web.
Graphite was used as a moderator. The control rods were boron carbide, most were inserted from the top of the reactor. The fuel assemblies were inside water-filled pressure tubes. The reactor core was contained in a concrete line cavity with a steel cap.
The main problem with the reactor design was that it was unstable at low power levels. Steam bubbles would result in increased power output, creating more steam bubbles. The control rods also displaced water from the lower part of the reactor core as they were inserted, which increased the power output.
Deuterium and Tritium are often used in fission weapons to boost their explosive yield. Deuterium-Tritium gas is injected into the center of the fissile core. During the detonation of the weapon, the Deuterium and Tritium fuse, releasing a burst of neutrons that accelerate the fission process and improve the efficiency of the weapon.
That must have been the same week that aliens abducted the President of the United States and the CIA stole your tin foil hat.
I get the impression that the PM can do just about anything he/she wishes as long as it doesn't trigger a vote of "no confidence".
Is there anything in the British system that is similar to the U.S. system of "checks and balances"?
I used to do source code optimization when C compilers were not good at optimization and the code could run much faster if the source code was tweaked to effectively use registers and avoid slow address computations.
Modern compilers have much better optimizers and it has become more difficult to optimize code for the CPUs of today. Counting cycles in a program is much more difficult than it used to be. When you have families of processors, the optimal code for one processor may be slow on another.
I used to use pointers whenever possible since experience had taught me that array references were always slower than pointer references. Then I discovered that array references were substantially faster than pointer references on some code that I was writing for a Pentium. The code was cleaner, easier to understand, smaller and faster when I used arrays.
If you look at the output of a good compiler, you will often be surprised at the optimization tricks that you would have never thought of.
In my opinion, the Second Amendment should protect my right to possess any weapon that the U.S. Army issues to an infantry soldier, including assault rifles, pistols, grenades and light machine guns. These are the modern equivalent of the musket.
The NSA, KGB, GCHQ and other intelligence agencies have been doing this for decades. They are not going to stop just because some people have the illusion that their email and conversations are private.
European countries have been intercepting and reading paper mail in "black chambers" for hundreds of years.
My limited experience has been that large datasets will blow out the cache on PCs and send the performance into the crapper.
High-performance memory systems cost money. PC vendors don't seem to be interested in building balanced systems.
That wasn't true in the old days. Not only were memory sizes small but CPUs were slow. Many did not have multiply/divide hardware. Conversions between ASCII/EBCDIC character strings and binary integers were very slow and avoided whenever possible. Most business oriented systems used some variant of BCD to store numbers. BCD can be converted to/from character codes very quickly.
Well, that is the point. Their audience is the average consumer, not the enthusiast.
I recently bought a car and used CR's reviews to help select the make/model. I'm sure Car & Driver would have hated it and told me to buy some overpriced, unreliable muscle car. That's fine, I was looking for reliable, safe transportation, not a sports car.
The bit about 1972 is correct. The calendar has a 28 year cycle before it repeats. If you set the computer's clock to 1972, assuming the operating system doesn't have a later epoch date, the days, dates and leap years will be the same as in 2000. Some people have used this to "fix" their Y2K computer problems. The only problem is that you have to subtract 28 before inputting a year and add 28 to any year in the output.
That was the only semi-realistic bit about the power plant problems. There was a Y2K bug that shut down a power plant during testing that was caused by a problem with the flue gas sensor. The date discontinuity screwed up the algorithm that computed the flue gas temperature rate-of-change and produced a bogus value. This would have lead to an automatic shutdown of the plant if it hadn't been fixed.
I was pissed off by the anti-nuclear propaganda implicit in the assumption that a cooling failure was guaranteed to result in a meltdown, breech of the containment and large plumes of Plutonium wafting over the country. Also, reactors are contained in pressure vessels containg water at high temperature and pressure. So why do the show control rods sticking out of the surface of something that looks like a big swimming pool while people are running around in the same area. If the core temperature was really as high as shown, all of that water would be flashed to steam, which would ruin someone's day.
Does anyone have any performance figures for Linux or *BSD on 1000Base-T?
There may not be any packets to encrypt. The data may already be encrypted or may not have any structure. For example, delta modulation converts an audio signal to a stream of bits, not bytes or packets.
A stream encryption device allows you to encrypt an arbitrary bit stream at high speed and minimum delay. This is ideal for link encryption, such as point-to-point data lines.
I would disagree. Each cipher text bit can be computed as the exclusive-or of the plain text bit and a complex function of the preceding N bits (plain or cipher or key).
I've worked with systems that use convolutional encoding to protect data streams against noise. The encoding and decoding algorithms are very different. The problem is that the encoding algorithm is much simpler than the decoding algorithm. The generator polynomial could be viewed as a private key.
I would argue that the company or individual has no "right" to their creation, other than property rights to the physical instances that they create. Patents and copyrights are better viewed as privileges granted by the state for the purpose of improving the public welfare, not as natural rights.
The NSA also has some facilities in Texas and other locations. See this for a list of NSA facilities. Just because you haven't heard of it doesn't mean that it doesn't exist.
France has been doing this for a long time through the DGSE.
The NSA's nightmare would probably be widespread distribution and adoption of software that would automatically encrypt all data communications, without any effort on the part of the user.
All I have to do is count the corpses to make a judgement about communism. Communism excels at producing large numbers of corpses.
Are you sure? I thought that most of the yield from large thermonuclear devices (fission-fusion-fission) came from the fission of the U238 jacket in the fusion stage. A "clean" but much lower yield device can be built by using a lead jacket.