Finally! A posting that knows something about this. It is indeed Clairvoyante that provides the technology underpinning this.
Suggestion to future posters: read the papers on the Clairvoyante site. They're tough, because they *assume* you know something about the human visual system. But that's where the magic lies: understanding how to trick the HVS into thinking it's getting a better image than it really is.
WiMAX has the *potential* to be important. But that's it for now. That's why so many posters ask 'what's the range?' and 'what's the throughput?' Nobody knows, because there aren't any chipsets yet. First mature chipsets -- from Fujitsu and Intel and others -- are in 1Q05. This is important because nature is analog. There's no simulation tool, because nobody has any real relevant experience on how OFDM + all the other 802.16 subtechnologies work... in this area, with trees, or in this other area, with rain, or... And it's important because nobody knows how good the receivers will be. For example: will we have anemic receivers built into devices, 'centrino-style'? Or will we have window-mounted antennae? Or antennae aligned by skilled craftspeople? And it's important because the antennae are still being cooked: MIMO? MISO? SIMO? Difficult all of them. (MIMO = multiple input, multiple output...) key, perhaps, for getting the extra few dB of receiver performance.
Given these, I wonder why the book exists. You can't write a book on business cases for wireless unless you know what the ranges, throughputs, device costs (and availability) are... at least to within a factor where the model makes some sense. OK, you can write the book, but it wouldn't be worth reading.
Most of the responses to this assume that it's the terrorists that we want to catch and that there is not enough probability that we'll succeed in snaring terrorists to make the (real or imagined, new or renewed) infringement on liberty worth while. At least one post notes the availability of encryption, IPSEC,... and assumes, perhaps correctly, that a crafty terrorist lot will use some of these tools. I'd like to make a comment or two that are explicitly separate from considerations of the "Patriot Act", but are in the context of prior use of CALEA.
First: The majority of those who are likely to get caught through CALEA are not terrorists: they're your average scumbag felon -- murderers, rapists, etc. While at least some in the terrorist line o' business might be smart (hmmm check the want ads for IT support for Big Al Qaeda?) the typical criminal makes mistakes... it's a dumb line of work after all. And here, as more than one poster has noted, the trail of phone calls is itself a great evidence trail "called xxx from yyy at 2214;..."
Second, and not less important... even the most evil, smartest terrorist will make an occasional mistake. Stuff on a hard drive; an unencrypted phone call; the passcode taped to a desk as someone noted. Get just ONE of these... while the Bad Guys still think they're safe, and... It's like when the Brits & Poles cracked Enigma in WW2 -- from then on they could read ALL German intelligence.
Third, on average, any nation that expends enough energy (read: money) that it can spy on tinfoils is doomed. But that's on average, and it's a time-averaged series. The Patriot Act has gotta go.
The general who wins a battle makes many calculations in his temple ere the battle is fought (Sun Tzu, 'Art of War')
Plasmas: problems include lousy lifetimes (two or three years before serious image degradation, intensity down by 3dB), high power consumption, image-burn-in, pixel-flicker... and v. high prices -- which will not decline much. Also: plasmas are not even sold in some areas (e.g. states such as Colorado, I'm told, because the elevation means low atmospheric pressure... and so the individual cells can "pop".)
Projectors: problems include v. poor color/hue fidelity (multiple reasons, including a white/dark illumination cycle, not just RGB, as a way of increasing brightness); fan noise. All front projectors have image visibility problems... the SNR is quite low, unless the room is darkened.
While it is easy (and reasonable, in some ways) to get bent out of shape about this, it really is not much of a story. Why? Because all voice switching equipment already has to have backdoors for electronic wiretaps. The requirements are known as CALEA: Communications Access for Law Enforcement Agencies, or some such. All access under CALEA has to be authorized by court order, at least in theory. Cisco is, at least in part, responding to a symmetric need. The thing to worry about is NOT whether/how this is implemented at the engineering level, but how the court orders are implemented.
I recall, and here am digging into memories AND am out of my depth, that St. John Ashcroft, the beloved, has eroded the levels of permission needed to get wiretaps. Now, if true, THAT's something to lose sleep about.
Which opens up the wonderfully complex questions: what to expect from 802.16 and 802.11... it would be neat to think that.11 would be a homeLAN or short drop solution, but it's unlicensed, so there's no control. And 802.16 is far from being cost effective.
Is what you're saying that 802.16 is the future for broadband?
Slashdot Mirror
Finally! A posting that knows something about this. It is indeed Clairvoyante that provides the technology underpinning this.
Suggestion to future posters: read the papers on the Clairvoyante site. They're tough, because they *assume* you know something about the human visual system. But that's where the magic lies: understanding how to trick the HVS into thinking it's getting a better image than it really is.
WiMAX has the *potential* to be important. But that's it for now. That's why so many posters ask 'what's the range?' and 'what's the throughput?' Nobody knows, because there aren't any chipsets yet. First mature chipsets -- from Fujitsu and Intel and others -- are in 1Q05. ... in this area, with trees, or in this other area, with rain, or ... ...) key, perhaps, for getting the extra few dB of receiver performance.
... at least to within a factor where the model makes some sense. OK, you can write the book, but it wouldn't be worth reading.
This is important because nature is analog. There's no simulation tool, because nobody has any real relevant experience on how OFDM + all the other 802.16 subtechnologies work
And it's important because nobody knows how good the receivers will be. For example: will we have anemic receivers built into devices, 'centrino-style'? Or will we have window-mounted antennae? Or antennae aligned by skilled craftspeople?
And it's important because the antennae are still being cooked: MIMO? MISO? SIMO? Difficult all of them. (MIMO = multiple input, multiple output
Given these, I wonder why the book exists. You can't write a book on business cases for wireless unless you know what the ranges, throughputs, device costs (and availability) are
Most of the responses to this assume that it's the terrorists that we want to catch and that there is not enough probability that we'll succeed in snaring terrorists to make the (real or imagined, new or renewed) infringement on liberty worth while. At least one post notes the availability of encryption, IPSEC, ... and assumes, perhaps correctly, that a crafty terrorist lot will use some of these tools. I'd like to make a comment or two that are explicitly separate from considerations of the "Patriot Act", but are in the context of prior use of CALEA.
... it's a dumb line of work after all. And here, as more than one poster has noted, the trail of phone calls is itself a great evidence trail "called xxx from yyy at 2214; ..."
... even the most evil, smartest terrorist will make an occasional mistake. Stuff on a hard drive; an unencrypted phone call; the passcode taped to a desk as someone noted. Get just ONE of these ... while the Bad Guys still think they're safe, and ... It's like when the Brits & Poles cracked Enigma in WW2 -- from then on they could read ALL German intelligence.
First: The majority of those who are likely to get caught through CALEA are not terrorists: they're your average scumbag felon -- murderers, rapists, etc. While at least some in the terrorist line o' business might be smart (hmmm check the want ads for IT support for Big Al Qaeda?) the typical criminal makes mistakes
Second, and not less important
Third, on average, any nation that expends enough energy (read: money) that it can spy on tinfoils is doomed. But that's on average, and it's a time-averaged series. The Patriot Act has gotta go.
The general who wins a battle makes many calculations in his temple ere the battle is fought (Sun Tzu, 'Art of War')
Plasmas: problems include lousy lifetimes (two or three years before serious image degradation, intensity down by 3dB), high power consumption, image-burn-in, pixel-flicker ... and v. high prices -- which will not decline much. Also: plasmas are not even sold in some areas (e.g. states such as Colorado, I'm told, because the elevation means low atmospheric pressure ... and so the individual cells can "pop".)
... the SNR is quite low, unless the room is darkened.
Projectors: problems include v. poor color/hue fidelity (multiple reasons, including a white/dark illumination cycle, not just RGB, as a way of increasing brightness); fan noise. All front projectors have image visibility problems
While it is easy (and reasonable, in some ways) to get bent out of shape about this, it really is not much of a story. Why? Because all voice switching equipment already has to have backdoors for electronic wiretaps. The requirements are known as CALEA: Communications Access for Law Enforcement Agencies, or some such. All access under CALEA has to be authorized by court order, at least in theory. Cisco is, at least in part, responding to a symmetric need. The thing to worry about is NOT whether/how this is implemented at the engineering level, but how the court orders are implemented.
I recall, and here am digging into memories AND am out of my depth, that St. John Ashcroft, the beloved, has eroded the levels of permission needed to get wiretaps. Now, if true, THAT's something to lose sleep about.
Which opens up the wonderfully complex questions: what to expect from 802.16 and 802.11 ... it would be neat to think that .11 would be a homeLAN or short drop solution, but it's unlicensed, so there's no control. And 802.16 is far from being cost effective.
Is what you're saying that 802.16 is the future for broadband?