> May I recommend not bringing your weapons with you?
You may suggest this. But the metal tools I normally carry are considered weapons by most security checkpoints, and I do want them everywhere. It's amazing what a small folding knife or a multitool can do to help in an emergency. Many Americans have taken to discarding personal knives, and my tools have been very helpful to them in numerous public instances.
> If you cannot trust handing them off to a sworn officer when you enter a secured area, maybe you should keep them at ho
If it's local police officer, or a military personnel whose name and ID number I've recorded, I might and have. But in such a large area, they're likely to be TSA personnel, not police. Those personnel are consistently undertrained and overworked, and are rapidly on their way in major airports to becoming the same harried staff who let the 9/11 bombers through with utility knives.
Logan airport was infamous for poor security. It's why they the 9/11 attackers selected that airport.
And if you think surrendering your weapons at a door to a public area is safety enhancing, you've perhaps not thought out how those weapons will be stored and released only to the original owner. They _will_ be stolen.
> an alternative format that can properly serve the same purpose: to be able to distribute documents in a way that is rendered identical
You mean LaTeX, or its modern descendant tetex, I think. Or the original Postscript standard, which has been effectively replaced by the open source tool ghostscript in most environments due to some outrageous licensing fees from Adobe.
One reason to use PDF is that it is a de factor standard, not becuase it actually renders more consistently than those older standards. Another is that it is possible to get commercial support for it, and a third is that it supports some useful "fill-in-blanks" formats. But consistent document formatting is not a reason to prefer PDF over LaTeX. Another is its very tight integration with most powerful web browsers, which does tend to make things faster than loading up the separate view application.
Because you, and your overpriced set of AD admins, will be spending all your time every day tweaking and overriding those settings. Most developers and systems people I know will revolt, actively or passively, against the necessary web of policies necessary to lock down Windows servers in large environments. They can, and will circulate, workarounds to get past IT's top down policies in such environments.
Desktops sitting around locally do provide large control over VLAN based security, firewalls, proxies, and the AD account management itself. Similar resistance happens when any new email system, network storage, backup, or authentication system is brought into play.
I'm afraid I've seen similar systems used elsewhere. The idea is too easy to promote and make a procedure when a company is large and has many layers of management. And the middle management, accustomed to such a system, will fall back into using it with a new name as swiftly as they can. Discarding such an embedded system means essentially replacing the entire hierarchy and especially middle management bureaucracy procedures.
I'd expect it to return to normal use within a year, once again without the rank and file employees being informed of the genuine nature of the new system.
How many Windows users do _not_ typically run as an "Administrator", to ease software management? And especially for Active Directory or Exchange administrators, re-authenticating every time they need to escalate to manage their resources during the day becomes burdensome. Most of the Windows admins I know do this as a matter of course, to ease the strain on their typing hands and improve their response time to requests.
> Firstly, the rootkit: 'implant' an agent (monolithic or multipartite) which stays as persistent as possible, maintaining control of the system. The most extreme case I've seen writes new firmware to the NIC, which is loaded by the BIOS or UEFI code
That is _nasty_. May I safely assume that virtual machines, with their hosting server based software NIC's, are immune from this vector? And do you have a reference to this mode of attack?
Various leaks about inappropriate surveillance have been happening in every nation since the beginning of nations and spies. The NSA effort is particularly large scale, and is part of its international collaborative efforts with other nation's monitoring agencies. Make no mistake, these agencies trade intelligence on a frequent basis so that each can gather data from the other agencies that they lack the local resources, or the legal ability, to gather themselves. The "Echelon" monitoring system is built on this basis, and hte UK and USA and other signatories have certainly gathered data that would be illegal for individual nations to gather, but nominally legal for them to collect from other nation's monitoring.
Since the SIGINT of the Netherlands was also participating in Echelon, it's not as if other northern European nations were not active in this network and covering Holland traffic.
I'm afraid that's not true. While shooting your car with such a device will probably trigger a car alarm on a modern, high value car, the ability to implant such a device remotely without ever being seen near the car is invaluable for both legitimate and illegitimate surveillance. The ability for an officer, 50 yards away, to implant 2 or 3 tracking devices in only a few seconds and get away unseen, without surveillance cameras showing them near the vehicle, is invaluable. Outside a loud bar frequented by gang members, it could be a much safer way for the officer to attach tracking devices to select vehicles.
The potential for abuse is also certainly there: if the device is small enough to be effectively fired as a bullet, it's also small enough to be effectively dropped in the vehicle or snuck into someone's bag or purse.
A great deal about drivers _is_ cross platform. Many of the same libraries, used by programs to manage the actual behavior of the card, are OpenGL based which is indeed cross-platform. The binary drivers do require _compilation_ for a particular graphical environment, and that does take thoughtful development to manage OS-specific function calls.
My experience of.NET is that it's very fast for developing programs that are unusably slow because they are bloated. This is not a good trade-off.
The "War on Terror" has not had the jingoistic power in the UK it enjoyed in the USA. The UK was dealing with domestic terror attacks by the IRA in recent decades, and learned harsh lessons on domestic terror involving small weapons or personal explosives. Their civilian security is generally no-nonsense, and has had centuries of dealing with violent protest by under-armed civilians from occupied territories. They have certainly not always _won_ such conflicts: the USA itself was once just such a remote territory, first engaged in guerrilla warfare, later in open revolt, and certainly including what would not be called "terrorist attacks".
I've done security work as part of systems engineering, and helped other companies with it, for decades. It would be difficult to pay me enough to take that as a primary role. Many projects think of security as something that can be painted on after a project is done: others have managers or core developers who think of every moment spent thinking about security as wasted, non-profit-generating work.and actively discourage any attention to security implications. Others rely on external firewalls to say "we trust the people we work with" and "if they can get to our network, we have much bigger problems" and proceed to ignore _all_ security concerns, especially those of angry former employees or zombied laptops.
Getting people to agree to, and follow, even the most basic security practices is nightmarish managerial and political work, and new security employees will not have any of the necessary political authority or acumen to get the changes done. The constant compromising, especially compromising for employees who are fundamentally stupid but work for someone important enough to protect them, can be soul draining and professionally devastating. It's also very difficult to get recommendations from former employers for security work: doing it well often means aggravating people who just want things to be easy. Those people _will_ complain to your supervisors, and get you labeled as "not a team player" in performance reviews.
That's why I prefer to get in, do our work, do our best with the security concerns, ttry to resolve the trade-offs as best we can, document the remaining issues, and _get out_.
I'm afraid that "non-isolated chargers" are common in poorly designed devices. Isolated chargers cost more to build, and many first-time designers have not been mentored in basic power system design, or in when to argue with the business department about the need for proper fusing capacitors or higher rated components. than the expected output or input voltages.
Even high voltage, low current sparks can hit the internal electronics before any reasonable overvoltage protection or fuse can take effect. It's a serious problem with industrial equipment.
It's quite understandable. Since a badly built commercial or home device can destroy the USB port on a computer or even feed back enough energy to destroy other components, making the "USB compliant" certification freely available without some trace of contractual responsibility is dangerousl. We went through this with Microsoft and their "Java" labels on their box. It would be too easy for those "magically freed" vendors to make, and sell, incompatible or even destructive hardware.
> He also says there is nothing wrong with paedophilia as long as it's consensual... If you don't understand why this is *very* wrong, then you must not have children.
Checking Wikipedia, I find this quote"
> [P]rostitution, adultery, necrophilia, bestiality, possession of child pornography, and even incest and pedophilia... should be legal as long as no one is coerced. They are illegal only because of prejudice and narrowmindedness.
By simplifying his statement, you can make it sound extremely dangerous. But RMS is very careful with the details of his ideas and claims. Notice that "as long as no one is coerced". That covers most of the abusive situations, and leaves strange legal processes out of the weird cases such as siblings who are adopted and later marry, discovering only later that they're related. RMS is a strange person, but actually thinks about the consequences of his beliefs in some detail. do read the rest of that thread over on Wikipedia.
> While you think you have grand plans for that sandbox, you MUST respect those who set the original rules, or you will not be welcomed in their sandbox.
Gong further with that metaphor, you also must not act like a feral cat in that sandbox. We want that sandbox not to have unburied, unclaimed bundles in it that may bear disease or simply sully the clean workspace we're using for sand castles. To use one of your own examples, the NVidia drivers replace the OpenGL library and effectively halt local development of that library.
Perhaps not. The subtle differences in software licenses left people like me free to write original tools under GPL, and publish them, with the result that patches get published as well. But I've worked with companies that didn't want to publish their "secret sauce" that they add to open source projects to make them commercially usable, even though they send the binaries to their clients. They're active for most parts of the software project, but they believe they're protecting their own business by concealing parts of their code or requiring commercial licenses to do modify it.
Sun did this with Java for years. The question then is not whether they'd have been able to run their business, but whether they _believed_ they could succeed with their business and were willing to write the software. The full set of Java development includes huge amounts of invested development in Java based utilities. Would Sun have invested in that project if they couldn't sell commercial licenses for it? I think not.
Fortunately, when Oracle bought Sun, they switched development efforts to "openjdk" and are gradually switching it to GPL. I'm seeing benefits already. Simply improving the Oracle installer is now possible, in free software, in ways that the Sun binary installer made quite difficult.
I've done some review of Canonical's license agreements for a Debian compatible software tool. Their licensing is peculiar. While individual components are being published as GPLv3, they're requesting, and getting, written permission from some contributors to re-publish the code under alternative licenses, at Canonical's whim. That is releasing licensing rights to someone else. Even if Canonical proves trustworthy (and they've not, due to their strange browser collection data practices), that goes far beyond most open source or freeware licenses.
Paranoia about open source licensing, for authors, has repeatedly proven justified. Projects released under older licenses have had their licenses carefully skirted, and software effectively encumbered with additional requirements that prevented open development. Examples have included NVidia drivers, which proprietized the OpenGL libraries, and Sun's encumbered licensing for Java. Ubuntu is doing reasonably well riding on the shoulders of the Debian upstream developers, and have been contributing back to the open source world. But this is not the first time Mr. Shuttleworth has made licensing, clearly to Ubuntu's commercial advantage and with the potential for abuse, at the expense of the open source community's safety.
You can't get rid of Outlook. This was a source of monopoly abuse lawsuits with Microsoft, it's woven into the OS too deeply to gracefully remove and replace completely. You can't keep Java updated for it. Hardening the remainder is infeasible if you have other computers inside the same home network or if the host is exposed to the Internet, because another host that is infected even temporarily inside your local network can spread worms and viruses to such an old host quite easily.
I've worked with people to try to secure such environments, but it becomes so labor intensive and so limiting so quickly,and so vulnerable to becoming a zombie and a security risk inside even a small network, the time and work are better spent setting up a new host in almost all cases. Even VM;ing it in a way that offers a "golden snapshot" that can be re-activated at whim and hae the active system deleted on a weekly basis can help, but the time and work exceed the cost of a new license or switching the host to a handy small Linux server for something useful.
Wars also bring technological development. This does not mean that the technical advancements justify starting a war of genocide.
Similarly, the space shuttle did have its uses with its large cargo bay and human crew. But the funds wasted in deciding, after the design was complete, that the fuel solid fuel boosters should not be recovered and re-used, and the decisions to manufacture the shuttle in one state by one contractor, different booster parts in other states for other contractors, and the nightmares of mismatched components from designing most of its systems form scratch rather than re-using consistent technologies, led to incredible costs and system failures as mismatched components were damaged in transport and _did not work_ when comibined with the other components developed in isolation.
I see this kind of thing every month in software and hardware projects whose developers and engineers are telecommuting or remotely located. This was coupled with congressional needs to distribute the manufacturing tasks across as many states as possible, in pure "pork barrel" American politics. The results were repeated returns to the drawing boards to redesign the most critical components, such as radar, heat shields, and life support, when the combined components simply did not work as planned.
Waveform "cancellation" is limited by a number of factors. There is always interference, and it takes as much energy to create the canceling waveform as the original, and unless the "inverted" waveform comes from precisely the same source, and orientation, as the original waveform, it cannot cancel the waveforms out everywhere, Moreover, if the canceling waveform is is being generated based on receiving the original waveform, simple lightspeed limitations preclude being able to completely detect and counter-generate a closely matching signal. For stable signals, a reflection counter generated signal is possible: holographic images rely on the light of the same color creating meaningful images based on the recorded interference patterns from the original holographic recording.
As you've noticed, sound cancellation can work well in a limited, well controlled environment (such as inside the ear of someone wearing noise cancelling headphones). But sound waves are also quite _long_ compared to most light waves, so producing them in phase to to cancel out another signal is relatively easy, and can be done dynamically. And there's the separate mater of "quantum" electromagnetic behavior. Energy is typically "quantized", existing in discreet bundles, for various reasons. You can think of it as sound coming from individual molecules of air, working together. For light, the very small wavelength makes it particularly noticeable, just as a stunningly high sound would have wavelengths on the order of the size of the space occupied by a single air molecule. At those frequencies, attempting to cancel the waves gets caught up in the behavior of individual "photons" of light or "molecules" of air, and controlling the behavior of the whole wave to get complete cancellation gets quite odd.
Their original funding, designers, and pilots were absolutely military. Even more recently, military satellites make up a large portion of their launches and any craft capable of bringing a launch to LEO must be considered capable of military payloads.
Slashdot Mirror
> May I recommend not bringing your weapons with you?
You may suggest this. But the metal tools I normally carry are considered weapons by most security checkpoints, and I do want them everywhere. It's amazing what a small folding knife or a multitool can do to help in an emergency. Many Americans have taken to discarding personal knives, and my tools have been very helpful to them in numerous public instances.
> If you cannot trust handing them off to a sworn officer when you enter a secured area, maybe you should keep them at ho
If it's local police officer, or a military personnel whose name and ID number I've recorded, I might and have. But in such a large area, they're likely to be TSA personnel, not police. Those personnel are consistently undertrained and overworked, and are rapidly on their way in major airports to becoming the same harried staff who let the 9/11 bombers through with utility knives.
Logan airport was infamous for poor security. It's why they the 9/11 attackers selected that airport.
And if you think surrendering your weapons at a door to a public area is safety enhancing, you've perhaps not thought out how those weapons will be stored and released only to the original owner. They _will_ be stolen.
Music scores don't tend to have the complex, MS Word driven document overformatting that is directly visible if you ever read the PDF.
> an alternative format that can properly serve the same purpose: to be able to distribute documents in a way that is rendered identical
You mean LaTeX, or its modern descendant tetex, I think. Or the original Postscript standard, which has been effectively replaced by the open source tool ghostscript in most environments due to some outrageous licensing fees from Adobe.
One reason to use PDF is that it is a de factor standard, not becuase it actually renders more consistently than those older standards. Another is that it is possible to get commercial support for it, and a third is that it supports some useful "fill-in-blanks" formats. But consistent document formatting is not a reason to prefer PDF over LaTeX. Another is its very tight integration with most powerful web browsers, which does tend to make things faster than loading up the separate view application.
Because you, and your overpriced set of AD admins, will be spending all your time every day tweaking and overriding those settings. Most developers and systems people I know will revolt, actively or passively, against the necessary web of policies necessary to lock down Windows servers in large environments. They can, and will circulate, workarounds to get past IT's top down policies in such environments.
Desktops sitting around locally do provide large control over VLAN based security, firewalls, proxies, and the AD account management itself. Similar resistance happens when any new email system, network storage, backup, or authentication system is brought into play.
I'm afraid I've seen similar systems used elsewhere. The idea is too easy to promote and make a procedure when a company is large and has many layers of management. And the middle management, accustomed to such a system, will fall back into using it with a new name as swiftly as they can. Discarding such an embedded system means essentially replacing the entire hierarchy and especially middle management bureaucracy procedures.
I'd expect it to return to normal use within a year, once again without the rank and file employees being informed of the genuine nature of the new system.
How many Windows users do _not_ typically run as an "Administrator", to ease software management? And especially for Active Directory or Exchange administrators, re-authenticating every time they need to escalate to manage their resources during the day becomes burdensome. Most of the Windows admins I know do this as a matter of course, to ease the strain on their typing hands and improve their response time to requests.
> Firstly, the rootkit: 'implant' an agent (monolithic or multipartite) which stays as persistent as possible, maintaining control of the system. The most extreme case I've seen writes new firmware to the NIC, which is loaded by the BIOS or UEFI code
That is _nasty_. May I safely assume that virtual machines, with their hosting server based software NIC's, are immune from this vector? And do you have a reference to this mode of attack?
The comic book was, indeed, much better. I've read a copy from a collector friend: it was poignant, and chilling.
The NSA got caught?
Various leaks about inappropriate surveillance have been happening in every nation since the beginning of nations and spies. The NSA effort is particularly large scale, and is part of its international collaborative efforts with other nation's monitoring agencies. Make no mistake, these agencies trade intelligence on a frequent basis so that each can gather data from the other agencies that they lack the local resources, or the legal ability, to gather themselves. The "Echelon" monitoring system is built on this basis, and hte UK and USA and other signatories have certainly gathered data that would be illegal for individual nations to gather, but nominally legal for them to collect from other nation's monitoring.
Since the SIGINT of the Netherlands was also participating in Echelon, it's not as if other northern European nations were not active in this network and covering Holland traffic.
> it's only useful in a high speed chase.
I'm afraid that's not true. While shooting your car with such a device will probably trigger a car alarm on a modern, high value car, the ability to implant such a device remotely without ever being seen near the car is invaluable for both legitimate and illegitimate surveillance. The ability for an officer, 50 yards away, to implant 2 or 3 tracking devices in only a few seconds and get away unseen, without surveillance cameras showing them near the vehicle, is invaluable. Outside a loud bar frequented by gang members, it could be a much safer way for the officer to attach tracking devices to select vehicles.
The potential for abuse is also certainly there: if the device is small enough to be effectively fired as a bullet, it's also small enough to be effectively dropped in the vehicle or snuck into someone's bag or purse.
A great deal about drivers _is_ cross platform. Many of the same libraries, used by programs to manage the actual behavior of the card, are OpenGL based which is indeed cross-platform. The binary drivers do require _compilation_ for a particular graphical environment, and that does take thoughtful development to manage OS-specific function calls.
My experience of .NET is that it's very fast for developing programs that are unusably slow because they are bloated. This is not a good trade-off.
The "War on Terror" has not had the jingoistic power in the UK it enjoyed in the USA. The UK was dealing with domestic terror attacks by the IRA in recent decades, and learned harsh lessons on domestic terror involving small weapons or personal explosives. Their civilian security is generally no-nonsense, and has had centuries of dealing with violent protest by under-armed civilians from occupied territories. They have certainly not always _won_ such conflicts: the USA itself was once just such a remote territory, first engaged in guerrilla warfare, later in open revolt, and certainly including what would not be called "terrorist attacks".
Because they're careful to write contracts that legally protect them from such lawsuits.
I've done security work as part of systems engineering, and helped other companies with it, for decades. It would be difficult to pay me enough to take that as a primary role. Many projects think of security as something that can be painted on after a project is done: others have managers or core developers who think of every moment spent thinking about security as wasted, non-profit-generating work.and actively discourage any attention to security implications. Others rely on external firewalls to say "we trust the people we work with" and "if they can get to our network, we have much bigger problems" and proceed to ignore _all_ security concerns, especially those of angry former employees or zombied laptops.
Getting people to agree to, and follow, even the most basic security practices is nightmarish managerial and political work, and new security employees will not have any of the necessary political authority or acumen to get the changes done. The constant compromising, especially compromising for employees who are fundamentally stupid but work for someone important enough to protect them, can be soul draining and professionally devastating. It's also very difficult to get recommendations from former employers for security work: doing it well often means aggravating people who just want things to be easy. Those people _will_ complain to your supervisors, and get you labeled as "not a team player" in performance reviews.
That's why I prefer to get in, do our work, do our best with the security concerns, ttry to resolve the trade-offs as best we can, document the remaining issues, and _get out_.
I'm afraid that "non-isolated chargers" are common in poorly designed devices. Isolated chargers cost more to build, and many first-time designers have not been mentored in basic power system design, or in when to argue with the business department about the need for proper fusing capacitors or higher rated components. than the expected output or input voltages.
Even high voltage, low current sparks can hit the internal electronics before any reasonable overvoltage protection or fuse can take effect. It's a serious problem with industrial equipment.
It's quite understandable. Since a badly built commercial or home device can destroy the USB port on a computer or even feed back enough energy to destroy other components, making the "USB compliant" certification freely available without some trace of contractual responsibility is dangerousl. We went through this with Microsoft and their "Java" labels on their box. It would be too easy for those "magically freed" vendors to make, and sell, incompatible or even destructive hardware.
> He also says there is nothing wrong with paedophilia as long as it's consensual... If you don't understand why this is *very* wrong, then you must not have children.
Checking Wikipedia, I find this quote"
> [P]rostitution, adultery, necrophilia, bestiality, possession of child pornography, and even incest and pedophilia ... should be legal as long as no one is coerced. They are illegal only because of prejudice and narrowmindedness.
By simplifying his statement, you can make it sound extremely dangerous. But RMS is very careful with the details of his ideas and claims. Notice that "as long as no one is coerced". That covers most of the abusive situations, and leaves strange legal processes out of the weird cases such as siblings who are adopted and later marry, discovering only later that they're related. RMS is a strange person, but actually thinks about the consequences of his beliefs in some detail. do read the rest of that thread over on Wikipedia.
> While you think you have grand plans for that sandbox, you MUST respect those who set the original rules, or you will not be welcomed in their sandbox.
Gong further with that metaphor, you also must not act like a feral cat in that sandbox. We want that sandbox not to have unburied, unclaimed bundles in it that may bear disease or simply sully the clean workspace we're using for sand castles. To use one of your own examples, the NVidia drivers replace the OpenGL library and effectively halt local development of that library.
Perhaps not. The subtle differences in software licenses left people like me free to write original tools under GPL, and publish them, with the result that patches get published as well. But I've worked with companies that didn't want to publish their "secret sauce" that they add to open source projects to make them commercially usable, even though they send the binaries to their clients. They're active for most parts of the software project, but they believe they're protecting their own business by concealing parts of their code or requiring commercial licenses to do modify it.
Sun did this with Java for years. The question then is not whether they'd have been able to run their business, but whether they _believed_ they could succeed with their business and were willing to write the software. The full set of Java development includes huge amounts of invested development in Java based utilities. Would Sun have invested in that project if they couldn't sell commercial licenses for it? I think not.
Fortunately, when Oracle bought Sun, they switched development efforts to "openjdk" and are gradually switching it to GPL. I'm seeing benefits already. Simply improving the Oracle installer is now possible, in free software, in ways that the Sun binary installer made quite difficult.
I've done some review of Canonical's license agreements for a Debian compatible software tool. Their licensing is peculiar. While individual components are being published as GPLv3, they're requesting, and getting, written permission from some contributors to re-publish the code under alternative licenses, at Canonical's whim. That is releasing licensing rights to someone else. Even if Canonical proves trustworthy (and they've not, due to their strange browser collection data practices), that goes far beyond most open source or freeware licenses.
Paranoia about open source licensing, for authors, has repeatedly proven justified. Projects released under older licenses have had their licenses carefully skirted, and software effectively encumbered with additional requirements that prevented open development. Examples have included NVidia drivers, which proprietized the OpenGL libraries, and Sun's encumbered licensing for Java. Ubuntu is doing reasonably well riding on the shoulders of the Debian upstream developers, and have been contributing back to the open source world. But this is not the first time Mr. Shuttleworth has made licensing, clearly to Ubuntu's commercial advantage and with the potential for abuse, at the expense of the open source community's safety.
You can't get rid of Outlook. This was a source of monopoly abuse lawsuits with Microsoft, it's woven into the OS too deeply to gracefully remove and replace completely. You can't keep Java updated for it. Hardening the remainder is infeasible if you have other computers inside the same home network or if the host is exposed to the Internet, because another host that is infected even temporarily inside your local network can spread worms and viruses to such an old host quite easily.
I've worked with people to try to secure such environments, but it becomes so labor intensive and so limiting so quickly,and so vulnerable to becoming a zombie and a security risk inside even a small network, the time and work are better spent setting up a new host in almost all cases. Even VM;ing it in a way that offers a "golden snapshot" that can be re-activated at whim and hae the active system deleted on a weekly basis can help, but the time and work exceed the cost of a new license or switching the host to a handy small Linux server for something useful.
Wars also bring technological development. This does not mean that the technical advancements justify starting a war of genocide.
Similarly, the space shuttle did have its uses with its large cargo bay and human crew. But the funds wasted in deciding, after the design was complete, that the fuel solid fuel boosters should not be recovered and re-used, and the decisions to manufacture the shuttle in one state by one contractor, different booster parts in other states for other contractors, and the nightmares of mismatched components from designing most of its systems form scratch rather than re-using consistent technologies, led to incredible costs and system failures as mismatched components were damaged in transport and _did not work_ when comibined with the other components developed in isolation.
I see this kind of thing every month in software and hardware projects whose developers and engineers are telecommuting or remotely located. This was coupled with congressional needs to distribute the manufacturing tasks across as many states as possible, in pure "pork barrel" American politics. The results were repeated returns to the drawing boards to redesign the most critical components, such as radar, heat shields, and life support, when the combined components simply did not work as planned.
Waveform "cancellation" is limited by a number of factors. There is always interference, and it takes as much energy to create the canceling waveform as the original, and unless the "inverted" waveform comes from precisely the same source, and orientation, as the original waveform, it cannot cancel the waveforms out everywhere, Moreover, if the canceling waveform is is being generated based on receiving the original waveform, simple lightspeed limitations preclude being able to completely detect and counter-generate a closely matching signal. For stable signals, a reflection counter generated signal is possible: holographic images rely on the light of the same color creating meaningful images based on the recorded interference patterns from the original holographic recording.
As you've noticed, sound cancellation can work well in a limited, well controlled environment (such as inside the ear of someone wearing noise cancelling headphones). But sound waves are also quite _long_ compared to most light waves, so producing them in phase to to cancel out another signal is relatively easy, and can be done dynamically. And there's the separate mater of "quantum" electromagnetic behavior. Energy is typically "quantized", existing in discreet bundles, for various reasons. You can think of it as sound coming from individual molecules of air, working together. For light, the very small wavelength makes it particularly noticeable, just as a stunningly high sound would have wavelengths on the order of the size of the space occupied by a single air molecule. At those frequencies, attempting to cancel the waves gets caught up in the behavior of individual "photons" of light or "molecules" of air, and controlling the behavior of the whole wave to get complete cancellation gets quite odd.
Their original funding, designers, and pilots were absolutely military. Even more recently, military satellites make up a large portion of their launches and any craft capable of bringing a launch to LEO must be considered capable of military payloads.