IPSec is primarily intended for creating a point-to-point tunnel.
No, IPSEC is primarily intended for securing internet connections, tunneling is just one way of using it.
Establishing arbitrary IPSec connections on demand to each endpoint you want to contact would be extremely difficult.
Unfornutely true for most cases. IIRC, the plan was to have DNS listings for host list the public key for the host which would then be used to initiate the secure connection. This would require secure connections to the DNS servers...
Magnetic track brakes use regenerated current to energize a magnet that applies the brake clasps to the wheels. There is a bit of extra braking effort due to force of attraction between the magnet and the rail below it. The technology probably dates back to before 1910.
With 8X the sampling rate and a bit of dithering, you will get 3 bits more dynamic range (18 dB) at audio frequencies. Increasing the sample rate will allow use of gentler filtering which should improve reproduction of high-frequency transients. The dithering can be accomplished by allowing a few bits of high frequency (say 100kHz) noise during recording.
What you see on the scope is not how the magnetic domains respond. With no signal, the bias drives the domains from full magnetization in one direction to full magnetization in the other with a duty cycle of 50%. Adding a signal affects how much time the domains are driven in one direction versus another, hence changes the duty cycle. This sounds pretty much the same way PWM works on many Class-D amplifiers.
Actually the statement about Nyquist's theorem is poppycock. This a mathematical fact, not some weird subjective result open to interpretation. Saying that Nyquist's theorem is wrong is equivalent to stating that the value of pi is really 6.
There's a lot of subtleties involved in going from Nyquist's theorem to actual practice. Some are related to problems of numerical analysis and others relate to how close you want the upper frequency cut-off to approach the Nyquist limit. The numerical analysis aspect is that the digital (discrete) representation is never exact, having said that, it is close enough most of the time (e.g. bass in mid-range). Getting usable frequency response to be close to the Nyquist limit requires use of 'brick-wall' filters which do bad things to time domain response - probably the worst case being an instrument like the triangle.
Some of this is covered on the design and implementation of direct digital synthesizers.
Compression is the removal of dynamic range, and is actually REQUIRED for vinyl to get the low volume sounds out of the vinyl surface noise to make them audible.
BS. What's required is pre-emphasis (e.g. the 'RIAA curve' created ca 1950, back when the RIAA was doing something useful). To get a decent amount of recording time on vinyl, you don't want a consistently high recording level (requires larger spacing between grooves and may burn out the cutting head) - which argues against using compression.
While a properly made CD will typically sound better than a vinyl recording, the article was correct in stating that CD's lend themselves more to overcompressing than vinyl and that has to do with the process of cutting the record (see points about groove spacing and burning out the cutting head).
You're reply is pretty much spot on - having a battery will reduce the amount of wires needed. You're also correct in pointing out this would be even better for a bus - note that some work was being done in the 1960's on flywheel powered buses with recharging stations at the bus stops.
Battery operated streetcars are nothing new - see The Time of the Trolley by William D. Middleton. Battery cars were used in such diverse places as New York City and Billings, Montana.
What makes this new is improvements in motor control circuitry making regeneration a lot more practical for streetcar use and improvements in battery technology - the old battery cars typically used Edison cells.
Ever hear of bias? Almost all 'direct' analog tape recorders use an AC bias signal to overcome the inherently non-linear magnetization of the tape (think B-H curves and hysterisis) which effectively creates a PWM signal.
This statement is true, but completely irrelevant. The fact that a recording medium is analog does not mean that it is better at accurately recording and reproducing a sound than a digital medium. Magnetic tapes are also analog recordings.
The advantage of vinyl would be most evident at high frequencies and I've heard some people claim they can hear the difference.
'Analog' tapes are a PWM format - analog by virtue of continuously variable pulse width.
Compression is a way of altering a sound wave, and has nothing to do with the final recording medium. Overcompression is a problem, but this is not an argument for vinyl over CD--it's just a comment on postprocessing techniques.
Overcompression is considerably more difficult to achieve with vinyl than with other media - unless you're willing to cut back on peak recording levels on vinyl - otherwise you're likely to burn out the cutting head.
One of the reasons that the 6600 succeeded was that Cray was in charge of the project - he had a relatively unique understanding of hardware, software and had 'good taste'. The 6000 series ISA is one of the cleanest that I've seen and is one of the first RISC machines. Cray knew how to get the right people to work on the project and backed them up when people outside his group criticized them. On the other hand, the 6600 was a bitch to manufacture.
I suspect that Apple, Samsung, Micron and Sony have quite a few patents of their own that could be used against SanDisk if SanDisk sued them, and supect the companies they are suing don't have much of a patent portfolio.
Another possibility is that the companies not being sued have cross-licensing agreements in place with SanDisk.
I think what he is saying is to do a 'clean sheet' hardware design and then use a VM to emulate running on a PC. One of Theo's points is that the PC architecture is a bloody mess as it has to look like a early 1980's PC-AT plus add-ons.
A couple of examples of design eff-ups in the PC include:
Using interupts below 20H when Intel specifically said not to use those interrupts
Using the NMI for the 8087/80287 when Intel specifically said not to use the NMI for the co-processor.
The BIOS itself is a hell of a mess, Open Boot firmware is a much nicer interface to work with.
The city of San Diego is in the process of setting up a website for registering cell phones and e-maill for reverse 911 messages. The County of San Diego is considering a similar move.
All the reports I've heard so far indicate that the Reverse 911 system worked very well during the fires - there were only a few minor glitches, which is to be expected.
One of the great technology success stories of the ongoing fires is the number of people told to evacuate via reverse 911 calls. The bad news is that the calls only work for standard landline (and presumably cable_co phones where the number is tied to a specific address). In my case, I went to bed last night with a wired phone next to the bed - didn't want to depend on the power being on for the wireless phones.
One sign of the success of the program is that only one fatality has been reported so far.
Kudo's to 'Craig' for posting the information to Google Maps Sunday evening - that was the most informative source for info on the fire Sunday evening - pretty clear by 11PM that I wasn't going to work the next day (work was in a mandatory evac zone declared Monday morning).
Some of the technology that hasn't worked has been the local '211' website (absolutely worthless) and the San Diego Union-Tribune website yesterday afternoon - they finally fixed that by dumping a lot of the flash and hosting the news updates on Blogspot. The local TV sites had too much flash to be useful.
You can get pretty much the functionality you want with a point to point IPSEC connection, probably the most difficult aspect would be securely exchanging the keys. Another possible means is to implement BitTorrent to use UDP instead of TCP, the downside is the app would than have to handle all the error handling now handled by the TCP stack.
They're basically doing this with a "man in the middle" attack by sending false messages to both parties in the communication, pretending to be the other. This is why all net traffic needs to be encrypted and signed.
This is exactly what the Authentication Header (AH) is suppoed to do - preventing "man in the middle" attacks. Use of the ESP feature of IPSEC can prevent snoopers from learning what TCP ports are being used.
Your solution would work if you had no plans for doing business in California. I don't think Target is in the position to give up on the California market and their presence in California is what gives state laws jurisdiction. Things might be different for an outfit with no physical presence in the state.
Corporations are formed to serve in the public interest, convenience or necessity - might be interesting to challenge some charters on that ground.
Target should have no problem continuing to serve their existing customer base -- just without the Federal Government intrusion of the benefits and protections of being 'incorporated'.
The biggest benefit is not directly to the corporation, but to the stockholders. The act of incorporating limits the liability of the stockholders to the value of the stock (i.e. stock becomes worthless). The tradeoff is that a corporation is not as free as an individual to do what they want.
USB 1.1 maxes out somewhere between 6 to 8 Mbps of throughput, which would be adequate for surfing but would limit a router throughput to maybe 3 Mbps (bits have to go both ways). Emphatically not in the same league as the fit-PC. Contrast this to some embedded boards coming with gigabit ethernet adapters.
One of my co-workers is working on putting a couple of webcams in his car to record what happened in case he was involved in an accident (I won't get into his being a former demolition derby driver). What's ironic about this article is that we were talking about mounting small cameras in the wing mirrors for aiding in parking - he drives a Prius and it is darn hard to judge where the corners of the car are.
Looks like the Technologic Systems TS-7300 with the battery back-up PC-104 board - including the T.S. logo on the front of the case. T.S. has just started shipping a board with a 500 MHz ARM-9, which would be about 3X faster than the 7300, but would need the VGA add-on board.
Another poster points out that it would be even better with three ethernet ports, for a DMZ.
Shouldn't be that hard to plug in a USB to ethernet adapter. If you really want three hard wired ethernet ports, you're probably better off getting a Soekris.
Think the FitPC would be a much better fit to what I want in a small form factor X86 box for a bit less than what a similar Mini-ITX system goes for. The FitPC appears to consume less power than the lowest power Mini-ITX systems - should be able to get several hours of run-time off of a motorcycle battery.
Slashdot Mirror
No, IPSEC is primarily intended for securing internet connections, tunneling is just one way of using it.
Establishing arbitrary IPSec connections on demand to each endpoint you want to contact would be extremely difficult.
Unfornutely true for most cases. IIRC, the plan was to have DNS listings for host list the public key for the host which would then be used to initiate the secure connection. This would require secure connections to the DNS servers...Magnetic track brakes use regenerated current to energize a magnet that applies the brake clasps to the wheels. There is a bit of extra braking effort due to force of attraction between the magnet and the rail below it. The technology probably dates back to before 1910.
With 8X the sampling rate and a bit of dithering, you will get 3 bits more dynamic range (18 dB) at audio frequencies. Increasing the sample rate will allow use of gentler filtering which should improve reproduction of high-frequency transients. The dithering can be accomplished by allowing a few bits of high frequency (say 100kHz) noise during recording.
What you see on the scope is not how the magnetic domains respond. With no signal, the bias drives the domains from full magnetization in one direction to full magnetization in the other with a duty cycle of 50%. Adding a signal affects how much time the domains are driven in one direction versus another, hence changes the duty cycle. This sounds pretty much the same way PWM works on many Class-D amplifiers.
There's a lot of subtleties involved in going from Nyquist's theorem to actual practice. Some are related to problems of numerical analysis and others relate to how close you want the upper frequency cut-off to approach the Nyquist limit. The numerical analysis aspect is that the digital (discrete) representation is never exact, having said that, it is close enough most of the time (e.g. bass in mid-range). Getting usable frequency response to be close to the Nyquist limit requires use of 'brick-wall' filters which do bad things to time domain response - probably the worst case being an instrument like the triangle.Some of this is covered on the design and implementation of direct digital synthesizers.
Compression is the removal of dynamic range, and is actually REQUIRED for vinyl to get the low volume sounds out of the vinyl surface noise to make them audible.
BS. What's required is pre-emphasis (e.g. the 'RIAA curve' created ca 1950, back when the RIAA was doing something useful). To get a decent amount of recording time on vinyl, you don't want a consistently high recording level (requires larger spacing between grooves and may burn out the cutting head) - which argues against using compression.While a properly made CD will typically sound better than a vinyl recording, the article was correct in stating that CD's lend themselves more to overcompressing than vinyl and that has to do with the process of cutting the record (see points about groove spacing and burning out the cutting head).
You're reply is pretty much spot on - having a battery will reduce the amount of wires needed. You're also correct in pointing out this would be even better for a bus - note that some work was being done in the 1960's on flywheel powered buses with recharging stations at the bus stops.
What makes this new is improvements in motor control circuitry making regeneration a lot more practical for streetcar use and improvements in battery technology - the old battery cars typically used Edison cells.
Ever hear of bias? Almost all 'direct' analog tape recorders use an AC bias signal to overcome the inherently non-linear magnetization of the tape (think B-H curves and hysterisis) which effectively creates a PWM signal.
The advantage of vinyl would be most evident at high frequencies and I've heard some people claim they can hear the difference.'Analog' tapes are a PWM format - analog by virtue of continuously variable pulse width.
Compression is a way of altering a sound wave, and has nothing to do with the final recording medium. Overcompression is a problem, but this is not an argument for vinyl over CD--it's just a comment on postprocessing techniques.
Overcompression is considerably more difficult to achieve with vinyl than with other media - unless you're willing to cut back on peak recording levels on vinyl - otherwise you're likely to burn out the cutting head.Glad I'm not the only one who remembers his campaign - has it really been 40 years now???
Then again, Nixon said "Sock it to me?" on Laugh-in.
One of the reasons that the 6600 succeeded was that Cray was in charge of the project - he had a relatively unique understanding of hardware, software and had 'good taste'. The 6000 series ISA is one of the cleanest that I've seen and is one of the first RISC machines. Cray knew how to get the right people to work on the project and backed them up when people outside his group criticized them. On the other hand, the 6600 was a bitch to manufacture.
Another possibility is that the companies not being sued have cross-licensing agreements in place with SanDisk.
A couple of examples of design eff-ups in the PC include:
Using interupts below 20H when Intel specifically said not to use those interrupts
Using the NMI for the 8087/80287 when Intel specifically said not to use the NMI for the co-processor.
The BIOS itself is a hell of a mess, Open Boot firmware is a much nicer interface to work with.
All the reports I've heard so far indicate that the Reverse 911 system worked very well during the fires - there were only a few minor glitches, which is to be expected.
In the case of Qualcomm stadium or the Del Mar fairgrounds, we're talking about thousands of people. Your point about morale is absolutely right on.
One sign of the success of the program is that only one fatality has been reported so far.
Kudo's to 'Craig' for posting the information to Google Maps Sunday evening - that was the most informative source for info on the fire Sunday evening - pretty clear by 11PM that I wasn't going to work the next day (work was in a mandatory evac zone declared Monday morning).
Some of the technology that hasn't worked has been the local '211' website (absolutely worthless) and the San Diego Union-Tribune website yesterday afternoon - they finally fixed that by dumping a lot of the flash and hosting the news updates on Blogspot. The local TV sites had too much flash to be useful.
You can get pretty much the functionality you want with a point to point IPSEC connection, probably the most difficult aspect would be securely exchanging the keys. Another possible means is to implement BitTorrent to use UDP instead of TCP, the downside is the app would than have to handle all the error handling now handled by the TCP stack.
This is exactly what the Authentication Header (AH) is suppoed to do - preventing "man in the middle" attacks. Use of the ESP feature of IPSEC can prevent snoopers from learning what TCP ports are being used.Your solution would work if you had no plans for doing business in California. I don't think Target is in the position to give up on the California market and their presence in California is what gives state laws jurisdiction. Things might be different for an outfit with no physical presence in the state.
Target should have no problem continuing to serve their existing customer base -- just without the Federal Government intrusion of the benefits and protections of being 'incorporated'.
The biggest benefit is not directly to the corporation, but to the stockholders. The act of incorporating limits the liability of the stockholders to the value of the stock (i.e. stock becomes worthless). The tradeoff is that a corporation is not as free as an individual to do what they want.USB 1.1 maxes out somewhere between 6 to 8 Mbps of throughput, which would be adequate for surfing but would limit a router throughput to maybe 3 Mbps (bits have to go both ways). Emphatically not in the same league as the fit-PC. Contrast this to some embedded boards coming with gigabit ethernet adapters.
One of my co-workers is working on putting a couple of webcams in his car to record what happened in case he was involved in an accident (I won't get into his being a former demolition derby driver). What's ironic about this article is that we were talking about mounting small cameras in the wing mirrors for aiding in parking - he drives a Prius and it is darn hard to judge where the corners of the car are.
Looks like the Technologic Systems TS-7300 with the battery back-up PC-104 board - including the T.S. logo on the front of the case. T.S. has just started shipping a board with a 500 MHz ARM-9, which would be about 3X faster than the 7300, but would need the VGA add-on board.
Shouldn't be that hard to plug in a USB to ethernet adapter. If you really want three hard wired ethernet ports, you're probably better off getting a Soekris.Think the FitPC would be a much better fit to what I want in a small form factor X86 box for a bit less than what a similar Mini-ITX system goes for. The FitPC appears to consume less power than the lowest power Mini-ITX systems - should be able to get several hours of run-time off of a motorcycle battery.