This reminds me of a similar result reported a few years back by Raymond Chiao on superluminal tunneling of light (Phys. Rev. Lett., v.71, p708, 1993). Although it is not exactly the same thing, I thought it might be relevant to the conversation here - especially since Ray was quoted in the NY Times article and cited as having laid some of the groundwork for the current experiment.
The question was: how long does it take a photon (or any other particle for that matter) to quantum mechanically "tunnel" through a classically forbidden region? Although superficially simple, this problem has a lot of depth because it strikes directly at the kinetic energy of a tunneling particle. In classical physics, a particle cannot enter a region in which its potential enegy would exceed its total energy (which is the sum of its kinetic and potential energies). However, in quantum mechanics, there is a non-zero probability for finding a particle in such a classically forbidden region. This implies a negative kinetic energy, since the potential energy exceeds the total energy, which in turn implies that the particle's momentum, which is proportional to the square root of the kinetic energy, is imaginary. Thus, the question becomes, how "fast" does a particle with imaginary momentum move? In all fairness, this is a very quasi-classical way of looking at the problem. A better way is to ask how long can the particle be in the forbidden region for a given energy uncertainty consistent with Heisenberg's uncertainty principle (HUP). But at the end of the day, they both reduce to the same problem.
Now, in practice, optics experiments of this type are performed with wave packets - typically gaussian envelopes of narrowband electromagnetic energy. The frequency spectrum of such a packet is also gaussian with a bandwidth proportional to 1/(pulse length). It is a well-known result of quantum mechanics that the phase velocity of the individual spectral components of the pulse can exceed the speed of light but that the aggregate "group velocity" of the center of the pulse cannot. This leads to dispersive effects such as pulse spreading as the packet propagates consistent with the HUP (which really has less to do with physics than with the relationship between temporal resolution and bandwidth of any signal). This turns out to be significant here.
If I recall correctly, the punchline in Chiao's superluminal tunneling experiments was that the _PEAK_ of the tunneling packet emerged from the forbidden region faster than the _PEAK_ of the original packet could have traversed the same distance at the speed of light. However, the _LEADING EDGE_ of the tunneling packet did NOT arrive any earlier than the _LEADING EDGE_ of the original pulse could have traversed the forbidden region at the speed of light. If you read the NY Times article of the Nature blurb you probably see where I am going with this. It was strongly implied in those articles that the peak of the superliminal pulse does not arrive any faster than the leading edge of the luminal pulse. Futhermore, it was pointed out that phase velocity of the light used in the experiment greatly exceeded the group velocity of the wave packet. This implies that when the leading edge of the luminal pulse enters the chamber, its spectral components can travel faster than the speed of light through the cesium and reconstruct an apparently superluminal pulse at the far end of the chamber (while simulataneously cancelling the input pulse). But, I suspect from Chiao's earlier work, that the the leading edges of the input and output pulses WILL maintain their luminal relationship and, thank heavens, causality is preserved.
So, the moral of the story is: a feature of a signal can arrive at its destination superluminally as long as the signal as a whole does not. In this case, I suspect that the leading edge, which represents the arrival of the signal, travels luminally or, more probably, sub-luminally. However, the peak, which is just a feature of the signal's envelope, appears to travel superluminally. This implies that the peak has moved closer to the leading edge of the pulse and all this is but a feature of the change in phase velocity for the spectral components in the electromagnetic pulse upon entering the cesium-filled chamber.
Please realize that this all conjecture since I have not actually seen the article being discussed (it is still in peer review). However, I believe this is a reasonable interpretation based on information available and is consistent with similar prior results. So, ultimately, I think that this is less an issue with relativity than an issue with quantum mechanics and that nobody need worry that their disgruntled grandchildren are about to travel backwards in time to kill them in a universe-shattering temporal paradox!
I agree - PDF is not PostScipt... but they are clearly kin. As Adobe says...
"PDF relies on the imaging model of the PostScript ® language to describe text and graphics in a device-independent and resolution-independent manner. [...] A PDF file is not a PostScript language program and cannot be directly interpreted by a PostScript interpreter. However, the page descriptions in a PDF file can be converted into a PostScript language program."
So, basically, PDF is a more structured version of PostScript with the programming components removed. For a more complete comparison of PDF & PostScript, please refer to section 2.4 of the Portable Document Format Reference Manual Version 1.3.
IMHO, the advantage to Apple is that, if they ever offer a Cocoa environment for Intel, a port of Quartz would provide a cross platform imaging model for developers to target and Apple wouldn't need to rewrite the interface APIs - just recompile. I'll grant that cross-compiling is a far cry from open sourcing. However, given Apple's trouble in convincing developers to migrate to Cocoa (hence the need for Carbon), this might just be the kind of move needed to get the ball rolling. If Apple is harboring any aspirations of being a player on the Intel side of the fence, it's going to take a lot more than the Cocoa APIs by themselves to make it happen.
Several people have (rightly) pointed out that DP4 != Aqua. Nevertheless, as the screenshots indicatre, Aqua is coming along nicely. The use of transparency and global antialiasing is delightful - even if some of the widgets are excessively gaudy.
Now that Darwin has been ported to Intel with support for X11, there has been much talk about if/when any of Aqua's tasty goodness will be available in that context. It has been suggested that this will never happen b/c Apple is unlikely to give away as important a crown jewel as its much ballyhooed GUI.
It is important to remember that Aqua is essentially just a widget set - the real power behind the interface is Quartz, the new PDF based rendering engine. If you refer to Wilfredo Sanchez's diary he indicates the possibility of a port of X11 to Quartz! The addition of Quartz's advanced features to X11 could be quite a treat. Although this is mentioned as a means of supporting X11 apps on MacOS X, it is possible that some support of this type might pave the way to supporting graphical Cocoa apps on Intel (it is unlikely that Classic or even Carbon apps will ever be supported on Intel but Cocoa is a whole 'nother kettle of fish).
Given the well known limitations of X11, esp. wrt to antialiasing, opening Quartz would be a much greater gift to the community than Aqua which is just another, albeit pretty, set of interface elements. I suspect that the real roadblock in opening Quartz is not Apple but, rather, Adobe which maintains a pretty tight leash on PostScript (for good reason).
I implore everyone who's been crying out for more open source code from Apple to focus their efforts on Quartz and to extend their cajoling to Adobe. Out of all of MacOS X's new goodies, I think Quartz is the pick of the litter - not Aqua.
I think that the rise in compelling female characters in videogames may be related to a similar rise in female action heroes in the mainstream media. Consider the following examples:
- Sigourney Weaver in Aliens - Linda Hamilton in Terminator 2 - Sarah Michelle Gellar in Buffy the Vampire Slayer - Hillary Rodham Clinton in the New York Senatorial Race;-)
Seriously, our culture has become more accepting of strong female characters - even/especially when there is some butt to be kicked. And I haven't even mentioned the prevalence of bad-ass women in cyberpunk novels!
How many women action heroes can you name from the popular media (film, TV, literature, videogames, etc.)?
Absolutely. Katz mentions that the information ecosystem is driven by memes but then fails to realize the extent of the biological analogy.
Memes, like genes, compete for limited resources - mindshare, as XNormal astutely points out. Just like biological resources such as food and air, mindshare is limited - primarily by the number of minds in the information ecosystem.
Memes that fail to capture people's imaginations, the requirement for meme reproduction, go the way of the Macarena or Latin - i.e. they either reach a point of ecologically limited niche status or become extinct.
The infosphere has its own Darwinian regulatory mechanisms just like the biosphere. That's why the meme analogy was invented in the first place.
Maybe Apple is hoping that they can use the settlement from iMac wannabes like Emachines & Daewoo to pay their settlement to Microware for using "OS 9"
Slashdot Mirror
This reminds me of a similar result reported a few years back by Raymond Chiao on superluminal tunneling of light (Phys. Rev. Lett., v.71, p708, 1993). Although it is not exactly the same thing, I thought it might be relevant to the conversation here - especially since Ray was quoted in the NY Times article and cited as having laid some of the groundwork for the current experiment.
The question was: how long does it take a photon (or any other particle for that matter) to quantum mechanically "tunnel" through a classically forbidden region? Although superficially simple, this problem has a lot of depth because it strikes directly at the kinetic energy of a tunneling particle. In classical physics, a particle cannot enter a region in which its potential enegy would exceed its total energy (which is the sum of its kinetic and potential energies). However, in quantum mechanics, there is a non-zero probability for finding a particle in such a classically forbidden region. This implies a negative kinetic energy, since the potential energy exceeds the total energy, which in turn implies that the particle's momentum, which is proportional to the square root of the kinetic energy, is imaginary. Thus, the question becomes, how "fast" does a particle with imaginary momentum move? In all fairness, this is a very quasi-classical way of looking at the problem. A better way is to ask how long can the particle be in the forbidden region for a given energy uncertainty consistent with Heisenberg's uncertainty principle (HUP). But at the end of the day, they both reduce to the same problem.
Now, in practice, optics experiments of this type are performed with wave packets - typically gaussian envelopes of narrowband electromagnetic energy. The frequency spectrum of such a packet is also gaussian with a bandwidth proportional to 1/(pulse length). It is a well-known result of quantum mechanics that the phase velocity of the individual spectral components of the pulse can exceed the speed of light but that the aggregate "group velocity" of the center of the pulse cannot. This leads to dispersive effects such as pulse spreading as the packet propagates consistent with the HUP (which really has less to do with physics than with the relationship between temporal resolution and bandwidth of any signal). This turns out to be significant here.
If I recall correctly, the punchline in Chiao's superluminal tunneling experiments was that the _PEAK_ of the tunneling packet emerged from the forbidden region faster than the _PEAK_ of the original packet could have traversed the same distance at the speed of light. However, the _LEADING EDGE_ of the tunneling packet did NOT arrive any earlier than the _LEADING EDGE_ of the original pulse could have traversed the forbidden region at the speed of light. If you read the NY Times article of the Nature blurb you probably see where I am going with this. It was strongly implied in those articles that the peak of the superliminal pulse does not arrive any faster than the leading edge of the luminal pulse. Futhermore, it was pointed out that phase velocity of the light used in the experiment greatly exceeded the group velocity of the wave packet. This implies that when the leading edge of the luminal pulse enters the chamber, its spectral components can travel faster than the speed of light through the cesium and reconstruct an apparently superluminal pulse at the far end of the chamber (while simulataneously cancelling the input pulse). But, I suspect from Chiao's earlier work, that the the leading edges of the input and output pulses WILL maintain their luminal relationship and, thank heavens, causality is preserved.
So, the moral of the story is: a feature of a signal can arrive at its destination superluminally as long as the signal as a whole does not. In this case, I suspect that the leading edge, which represents the arrival of the signal, travels luminally or, more probably, sub-luminally. However, the peak, which is just a feature of the signal's envelope, appears to travel superluminally. This implies that the peak has moved closer to the leading edge of the pulse and all this is but a feature of the change in phase velocity for the spectral components in the electromagnetic pulse upon entering the cesium-filled chamber.
Please realize that this all conjecture since I have not actually seen the article being discussed (it is still in peer review). However, I believe this is a reasonable interpretation based on information available and is consistent with similar prior results. So, ultimately, I think that this is less an issue with relativity than an issue with quantum mechanics and that nobody need worry that their disgruntled grandchildren are about to travel backwards in time to kill them in a universe-shattering temporal paradox!
Thanks - I was not aware of that.
... that means we can focus our efforts on Apple! ;-)
I assumed that PDF carried much of the same baggage as PostScript and I am happy to hear that is not the case.
I retract my call to arms against Adobe
I agree - PDF is not PostScipt ... but they are clearly kin. As Adobe says ...
"PDF relies on the imaging model of the PostScript ® language to describe text and graphics in a device-independent and resolution-independent manner. [...] A PDF file is not a PostScript language program and cannot be directly interpreted by a PostScript interpreter. However, the page descriptions in a PDF file can be
converted into a PostScript language program."
So, basically, PDF is a more structured version of PostScript with the programming components removed. For a more complete comparison of PDF & PostScript, please refer to section 2.4 of the Portable Document Format Reference Manual Version 1.3.
IMHO, the advantage to Apple is that, if they ever offer a Cocoa environment for Intel, a port of Quartz would provide a cross platform imaging model for developers to target and Apple wouldn't need to rewrite the interface APIs - just recompile. I'll grant that cross-compiling is a far cry from open sourcing. However, given Apple's trouble in convincing developers to migrate to Cocoa (hence the need for Carbon), this might just be the kind of move needed to get the ball rolling. If Apple is harboring any aspirations of being a player on the Intel side of the fence, it's going to take a lot more than the Cocoa APIs by themselves to make it happen.
Several people have (rightly) pointed out that DP4 != Aqua. Nevertheless, as the screenshots indicatre, Aqua is coming along nicely. The use of transparency and global antialiasing is delightful - even if some of the widgets are excessively gaudy.
Now that Darwin has been ported to Intel with support for X11, there has been much talk about if/when any of Aqua's tasty goodness will be available in that context. It has been suggested that this will never happen b/c Apple is unlikely to give away as important a crown jewel as its much ballyhooed GUI.
It is important to remember that Aqua is essentially just a widget set - the real power behind the interface is Quartz, the new PDF based rendering engine. If you refer to Wilfredo Sanchez's diary he indicates the possibility of a port of X11 to Quartz! The addition of Quartz's advanced features to X11 could be quite a treat. Although this is mentioned as a means of supporting X11 apps on MacOS X, it is possible that some support of this type might pave the way to supporting graphical Cocoa apps on Intel (it is unlikely that Classic or even Carbon apps will ever be supported on Intel but Cocoa is a whole 'nother kettle of fish).
Given the well known limitations of X11, esp. wrt to antialiasing, opening Quartz would be a much greater gift to the community than Aqua which is just another, albeit pretty, set of interface elements. I suspect that the real roadblock in opening Quartz is not Apple but, rather, Adobe which maintains a pretty tight leash on PostScript (for good reason).
I implore everyone who's been crying out for more open source code from Apple to focus their efforts on Quartz and to extend their cajoling to Adobe. Out of all of MacOS X's new goodies, I think Quartz is the pick of the litter - not Aqua.
I think that the rise in compelling female characters in videogames may be related to a similar rise in female action heroes in the mainstream media. Consider the following examples:
;-)
- Sigourney Weaver in Aliens
- Linda Hamilton in Terminator 2
- Sarah Michelle Gellar in Buffy the Vampire Slayer
- Hillary Rodham Clinton in the New York Senatorial Race
Seriously, our culture has become more accepting of strong female characters - even/especially when there is some butt to be kicked. And I haven't even mentioned the prevalence of bad-ass women in cyberpunk novels!
How many women action heroes can you name from the popular media (film, TV, literature, videogames, etc.)?
Absolutely. Katz mentions that the information ecosystem is driven by memes but then fails to realize the extent of the biological analogy.
Memes, like genes, compete for limited resources - mindshare, as XNormal astutely points out. Just like biological resources such as food and air, mindshare is limited - primarily by the number of minds in the information ecosystem.
Memes that fail to capture people's imaginations, the requirement for meme reproduction, go the way of the Macarena or Latin - i.e. they either reach a point of ecologically limited niche status or become extinct.
The infosphere has its own Darwinian regulatory mechanisms just like the biosphere. That's why the meme analogy was invented in the first place.
Maybe Apple is hoping that they can use the settlement from iMac wannabes like Emachines & Daewoo to pay their settlement to Microware for using "OS 9"