Obviously things like the magnetic pathfinding of birds would be useless in space.
Why? There's plenty of mag field in space - especially near-Earth space. If the habitat keeps a constant orientation with respect to the mag field and is built to allow it to penetrate the birds will have no problem. If not, a habitat large enough for it to matter, where magnetic-navigating birds are intended to fly free, will have a deliberately-generated field to keep them from becoming confused.
Yes. Michigan is enormous. It's farther from the Detroit to the straits of Macinac than it is from Detroit to New York City, and the straits are only a little over halfway to the border.
It also has an unusually high-reliability power grid. (It had to be designed for some severe storms and icing.) During the great northeast blackout the problem propagated to the Detroit Windsor boundary, Detroit Edison's equipment detected it, and cut off from the east coast. Pick a spot (like the west side of Ann Arbor) where Detroit Edison and Consolidated meet and you can get redundant feeds from both company's grids (as Compuserve did long ago), in addition to your backup UPS and generator. (Ann Arbor is also a good spot for communication connectivity, too.)
Michigan's topography breaks up the weather patterns enough that even a few tens of miles of separation often make the difference between a heavy storm and clear skies.
I went with just WWII because it's on my horizon (I'm a boomer with a dad who fought in it shortly before my birth), it was adequate to make the point, and including the hiatus would have complicated finding the numbers.
What if they could make a pill that cured the source of your pain, relieving you of both your pain AND your financial burden to pay them?
Do you think they would?
If they could charge enough for the one-shot cure to make at least as much expected value as a lifetime subscription to the pills? In a New York minute!
So the issue for Canonical is whether FSF might use the terms of GPL3 to force disclosure of Canonical's key? And Canonical won't take their word that they can't or won't? Then there's a simple solution on FSF's part.
FSF is also the holder of the copyright on GRUB 2. All they have to do is to double-license it, adding a second license that is the same as the GPL3 except for explicitly granting the right to NOT be subject to forced key disclosure. This would make Canonical safe in a legally binding way, as long as any modifications they make to GRUB 2 don't merge in other GPL software that doesn't carry the extra license term.
Even with psychopaths and criminal gangs preying on innocent victims, the crime, injury, and death rates are substantially lower where the gun-toting is higher and potential victims are often armed. And when laws are changed to encourage or discourage victim self-armament the crime, injury, and death rates change accordingly. So it seems to be a real effect and driven from laws to carry to crime, not some artifact. (See _More Guns, Less Crime_ by John Lott for research supporting this side of the interminable argument.)
This is what you expect if you think about it more than superficially. When it comes to behavior, especially with something known to be dangerous, people tend to be mostly rational actors and only occasionally random number generators. Predators avoid victims that are likely to injure or kill them. The few that don't are likely to have a much shorter career when victims are armed than when they can continue unchecked (or until caught by police).
In international level, treating governments as if they were people, you have a small neighborhood. And you have no police at that level (despite the posturing of international organizations), just competing gangs. The same real-world effects can be expected to rule at this scale that rule when a few people are interacting.
The 2.25 million people that died in the Korean War, and the ~ 2 million people that died in the Vietnam War would beg to differ.
Too true.
But compared to 73 million in eight years (averaging over nine million per year for about half a generation) some people believe that's a substantial improvement.
The reason we use xray at the dentist for example is to look at the density of bones for various lesion and holes, or how the root grow etc....
And the reason that's what we look for is because we CAN. That's what X-rays do that's useful.
Nothing the terrahertz technology can really replace.
Though it does it somewhat differently, it may be able to do at least some of those. Perhaps even something good enough to replace all of them. And it may be able to do other useful things that X-rays don't.
Right now they just got a transmitter and receiver working on a chip. Let's hang in there until they get some systems built up and see what can be practically engineered.
Terahertz EM radiation should have similar wavelengths to Ultrasound, which only penetrates a few inches and lacks resolution.
Actually, resolution using synthetic aperture techniques is very high. In principle (if I understand this correctly) it is an analytical solution and limited only by things like the sampling resolution, timing accuracy, and uncertainty principle, while practical equipment can resolve to far less than a wavelength.
It's INcoherent illumination that is suffers mightily from diffraction limits. Coherent processing preserves phase information and can compute out such artifacts.
All the TSA agent has to do is point at you (well, with some reason of course), and the cops would arrest you.
Also, once you come into view of the checkpoint you are not allowed to leave without permission and may be detained if you try.
Assaulting a TSA agent would not be considered assaulting a police office[r]...
Agreed, "assaulting a peace officer" wouldn't apply. But don't they also have a separate special charge for screwing around with TSA people and operations, similar to the post-9/11 "interfering with a flight crew" prohibition?
As another poster has already mentioned: The discoverers of X-rays warned from the beginning that there might be harm, and reports of damage from exposure were in the literature within the first couple years.
(Now some people may have said, somewhere along the way, that some level of X-rays is safe. Manufacturers and users of X-ray equipment, for example. B-) )
The individual photons of terahertz radiation are at far too low an energy to ionize an atom or molecule or otherwise substantially affect a molecular bond, most foldings, and other molecular behavior. (That's a vast improvement on x-rays, where each photon has enough energy to ionize anything it interacts with, demolishing molecular bonds and spraying reactive species throughout highly-organized structures.)
But terahertz radiation is is COHERENT. That means billions of photons gang up, with their electric and magnetic fields aligned. Then these gangs march by in-step for timescales that are enormous at molecular scales. So biological molecules and structures may still be strongly affected.
Conductive structures whose (electrical) length is related to a half-wavelength of the frequency in use, along with structures that vibrate at that frequency and have unbalanced charges, can absorb energy from multiple cycles of the signal, "pumping up" the resonance until enough energy is accumulated to cause damage.
Even without such resonance, the field of a single half-cycle may be strong enough to affect a molecule as strongly as a single photon of a considerably higher energy.
So using lower energy photons may turn out to be harmless, and almost certainly will be less harmful than using ionizing radiation. But some biological structures may turn out to be vulnerable to it. It needs to be tested. Further, EACH FREQUENCY to be used needs to be tested. With resonance phenomena and fixed molecular sizes you can't generalize from one frequency to another.
On the other hand, if tuned molecular effects ARE significant there is an opportunity to discover them and use them to create new beneficial interventions. B-)
A blood vessel isn't just empty space though. It has walls that expand and contract in response to various conditions. I assume normal vessels also resist intrusion.
As I read it this process does one of the following:
- reserves the space for the blood vessels to grow into a bit later (and perhaps marks it with growth factors to encourage vascularization along the paths.)
- provides a scaffold around which the blood vessels form, then dissolves away to leave the resulting vessels open.
But in an urban environment these are not accurate signal strength is only loosely proportional to inverse square of the distance so any accuracy will utterly break down.
There's a bit more to it than that.
First: Inverse square falls off FAST near the transmitter. For distant transmitters, like TV stations and (rural) cellphone towers, you have the problem you describe. But for WiFi in nearby houses or other apartments in your own you can get a pretty good idea where you are just from strength.
Second: There's also timing signals.
TV stations (especially since the changeover to digital and the use of multiple synchronized transmitters sending the same signal on the same frequency) and cell towers all send fantastically accurate timing references. They are often deliberately synchronized to a common timebase, or inadvertently synchronized by referencing a common master clock (typically GPS) so they can be compared to like obtain path-distance differences between transmitters, creating loran-style hyperbolic loci. (If they aren't explicitly synchronized the database can track their current offsets by measuring and recording it - and updating as it responds to queries in its neighborhood.
In the synchronized TV signal case the distance difference on the path between different transmitters shows up as a pattern of relative signal strengths among the many carriers of the OFDM signal. You can get one loran-curve by analyzing just one station's signal.
Both TV and cellphone base station measurements can be confused somewhat by multipath reflections. On the other hand this can be small and distorted signals dropped in favor of others with better paths. Or it can even be used to get MORE information about location.
Distance to WiFI devices can be measured by actively interacting with them in various low-level ways that are a part of the standard and a legitimate thing for any other WiFi device to do.
I'm not saying that this is what the service does. But these are things it clearly COULD be doing. And they can produce an accuracy at least as good as what is claimed.
~Oh, disaster! The sea level rise on the east coast might be TRIPLE that of the world average in our previous prediction. It might rise 14 to 20 inches over the next century! That's as much as a whole 2 tenths of an inch per year! They're all going to drown.~
Somehow I'm not impressed. While it might be nice to see New York and Washington become awash, this is a number of orders of magnitude too low to be useful.
Compare this to Venice (which still seems to be doing very well, thank you.)
Another translated genomic data into musicâ"because when it comes to data-crunching and neuroscience, you can't be deadly serious all the time.
What's non-serious about that? Translating the data into music makes it accessible to the powerful rhythm, tonal sequence, and echo-imaging processing of the auditory system, which might identify interesting and useful features in the data.
It used to be fun to analyze public speaking in the frequency domain and find the 8 to 12 hertz oscillations. The sound equipment used in the stages at the White House appear to remove these in real time these days, though.
Or you may be observing that pathological liars aren't stressed when they lie.
If you look at what classes compile to, it's just structs and some syntactic sugar to automatically do what can be easily done manually to manipulate them.
Also some const vectors of pointers to functions. Like what you see in kernel driver tables.
I suspect, though, due to the way you are writing and asking the question, that you've never worked on a large C code base (which if they know what they are doing are already taking the good ideas from OOP methodology - but don't need the "class" keyword as a mental crutch).
Hear hear!
If you look at what classes compile to, it's just structs and some syntactic sugar to automatically do what can be easily done manually to manipulate them.
The main advantage of using the C++ compiler, rather than doing it yourself, is not that the compiler automates the code generation for the manipulation. It's that the compiler always gets it right on what it does for you and can check much of what YOU do to be sure you don't violate intended encapsulation boundaries or point to the wrong thing (unless you use casts to tell it you really meant to do something risky). So there's less opportunity to write a bug and have the compiler accept it.
ANSI C incorporates some of this with strong type checking (which it got from C++). But a real C++ compiler can make-right-or-detect-error in more ways.
Objective C, like Smalltalk, has an issue with construction and destruction: Overridden methods use the sub/derived version if called during the execution of the super/base class constructor. This means writing a sub/derived class has the potential to break the already-debugged super/base class constructruction, which must be rechecked for every sub/derived class.
C++ gets this ALMOST exactly right: Base class constructors get the base class version, derived class constructors get the derived class version. Iterate down the entire hierarchy. Equivalently with DEstructors as the successive levels are torn down.
I say ALMOST because the spec, and most compilers have what IMHO is a bug: During construction and destruction of member objects, if the member object constructor or destructor (or something it uses) gets hold of a pointer to the containing object and calls an overridden method it SHOULD get the base class version. (On construction the support for the derived class version is not initialized. On destruction it's already torn down.) But whether it actually gets the base or derived version is problematic.
If you take the binary combination of what the constructor and destructor get there are three "wrong" and one "right" combinations. In the early days cfront and cfront-derived compilers got it "wrong" one way, three compilers for PCs got it "wrong" another way, and gcc got it "wrong" the third way. So if (as I asked them on each of the first two standards efforts) they had specified using the "right" behavior, everybody would have had to make a minor change and nobody would have had an advantage.
The first ANSI standard explicitly left this undefined and the second said essentially "don't write code that lets this happen". B-b (I haven't really looked at the recent update to see what it says.)
This was important to the project I was working on back then: We had rolled our own garbage-collection and exception-handling systems, using preprocessor-defined overridden virtual function bodies to track the level of construction. (Catch and throw were still just reserved words and the politically correct way to handle heap management was with user-written explicit allocation and deallocation.) With this "bug" we couldn't count on correctly handling an exception thrown or garbage-collection triggered by a member object constructor or destructor. So we couldn't use them. Instead of building compact composite heap-allocated objects we had to use smartpointers, separately heap-allocate the "contained" objects, and knit together a cluster of separate objects, Smalltalk-style. B-b
Microsoft has publicly declared that they have no intention of supporting anything past C95
I gave up on Microsoft back when Byte magazine was in, or recently beyond, single-digit issue numbers.
A letter to the editor complained about Microsoft's support of their FORTRAN compiler: There was a bug in the floating point format handling that a customer needed to use. After several iterations of bug reports and fix requests, Microsoft had told the guy that not only had they not fixed it yet, but they were never going to fix it. IMHO that meant Microsoft had an institutional issue with customer support and adherence to standards.
Avoiding Microsoft software and products has saved me immeasurable grief over the several decades since.
Slashdot Mirror
Obviously things like the magnetic pathfinding of birds would be useless in space.
Why? There's plenty of mag field in space - especially near-Earth space. If the habitat keeps a constant orientation with respect to the mag field and is built to allow it to penetrate the birds will have no problem. If not, a habitat large enough for it to matter, where magnetic-navigating birds are intended to fly free, will have a deliberately-generated field to keep them from becoming confused.
Yes. Michigan is enormous. It's farther from the Detroit to the straits of Macinac than it is from Detroit to New York City, and the straits are only a little over halfway to the border.
It also has an unusually high-reliability power grid. (It had to be designed for some severe storms and icing.) During the great northeast blackout the problem propagated to the Detroit Windsor boundary, Detroit Edison's equipment detected it, and cut off from the east coast. Pick a spot (like the west side of Ann Arbor) where Detroit Edison and Consolidated meet and you can get redundant feeds from both company's grids (as Compuserve did long ago), in addition to your backup UPS and generator. (Ann Arbor is also a good spot for communication connectivity, too.)
Michigan's topography breaks up the weather patterns enough that even a few tens of miles of separation often make the difference between a heavy storm and clear skies.
I agree with essentially all your points.
I went with just WWII because it's on my horizon (I'm a boomer with a dad who fought in it shortly before my birth), it was adequate to make the point, and including the hiatus would have complicated finding the numbers.
What if they could make a pill that cured the source of your pain, relieving you of both your pain AND your financial burden to pay them?
Do you think they would?
If they could charge enough for the one-shot cure to make at least as much expected value as a lifetime subscription to the pills? In a New York minute!
So the issue for Canonical is whether FSF might use the terms of GPL3 to force disclosure of Canonical's key? And Canonical won't take their word that they can't or won't? Then there's a simple solution on FSF's part.
FSF is also the holder of the copyright on GRUB 2. All they have to do is to double-license it, adding a second license that is the same as the GPL3 except for explicitly granting the right to NOT be subject to forced key disclosure. This would make Canonical safe in a legally binding way, as long as any modifications they make to GRUB 2 don't merge in other GPL software that doesn't carry the extra license term.
Problem solved.
It works for civilians and pistols.
Even with psychopaths and criminal gangs preying on innocent victims, the crime, injury, and death rates are substantially lower where the gun-toting is higher and potential victims are often armed. And when laws are changed to encourage or discourage victim self-armament the crime, injury, and death rates change accordingly. So it seems to be a real effect and driven from laws to carry to crime, not some artifact. (See _More Guns, Less Crime_ by John Lott for research supporting this side of the interminable argument.)
This is what you expect if you think about it more than superficially. When it comes to behavior, especially with something known to be dangerous, people tend to be mostly rational actors and only occasionally random number generators. Predators avoid victims that are likely to injure or kill them. The few that don't are likely to have a much shorter career when victims are armed than when they can continue unchecked (or until caught by police).
In international level, treating governments as if they were people, you have a small neighborhood. And you have no police at that level (despite the posturing of international organizations), just competing gangs. The same real-world effects can be expected to rule at this scale that rule when a few people are interacting.
The 2.25 million people that died in the Korean War, and the ~ 2 million people that died in the Vietnam War would beg to differ.
Too true.
But compared to 73 million in eight years (averaging over nine million per year for about half a generation) some people believe that's a substantial improvement.
The reason we use xray at the dentist for example is to look at the density of bones for various lesion and holes, or how the root grow etc....
And the reason that's what we look for is because we CAN. That's what X-rays do that's useful.
Nothing the terrahertz technology can really replace.
Though it does it somewhat differently, it may be able to do at least some of those. Perhaps even something good enough to replace all of them. And it may be able to do other useful things that X-rays don't.
Right now they just got a transmitter and receiver working on a chip. Let's hang in there until they get some systems built up and see what can be practically engineered.
Terahertz EM radiation should have similar wavelengths to Ultrasound, which only penetrates a few inches and lacks resolution.
Actually, resolution using synthetic aperture techniques is very high. In principle (if I understand this correctly) it is an analytical solution and limited only by things like the sampling resolution, timing accuracy, and uncertainty principle, while practical equipment can resolve to far less than a wavelength.
It's INcoherent illumination that is suffers mightily from diffraction limits. Coherent processing preserves phase information and can compute out such artifacts.
All the TSA agent has to do is point at you (well, with some reason of course), and the cops would arrest you.
Also, once you come into view of the checkpoint you are not allowed to leave without permission and may be detained if you try.
Assaulting a TSA agent would not be considered assaulting a police office[r]...
Agreed, "assaulting a peace officer" wouldn't apply. But don't they also have a separate special charge for screwing around with TSA people and operations, similar to the post-9/11 "interfering with a flight crew" prohibition?
Didn't these used to say that X-rays were safe?
No.
As another poster has already mentioned: The discoverers of X-rays warned from the beginning that there might be harm, and reports of damage from exposure were in the literature within the first couple years.
(Now some people may have said, somewhere along the way, that some level of X-rays is safe. Manufacturers and users of X-ray equipment, for example. B-) )
Unfortunately, that misses a point:
The individual photons of terahertz radiation are at far too low an energy to ionize an atom or molecule or otherwise substantially affect a molecular bond, most foldings, and other molecular behavior. (That's a vast improvement on x-rays, where each photon has enough energy to ionize anything it interacts with, demolishing molecular bonds and spraying reactive species throughout highly-organized structures.)
But terahertz radiation is is COHERENT. That means billions of photons gang up, with their electric and magnetic fields aligned. Then these gangs march by in-step for timescales that are enormous at molecular scales. So biological molecules and structures may still be strongly affected.
Conductive structures whose (electrical) length is related to a half-wavelength of the frequency in use, along with structures that vibrate at that frequency and have unbalanced charges, can absorb energy from multiple cycles of the signal, "pumping up" the resonance until enough energy is accumulated to cause damage.
Even without such resonance, the field of a single half-cycle may be strong enough to affect a molecule as strongly as a single photon of a considerably higher energy.
So using lower energy photons may turn out to be harmless, and almost certainly will be less harmful than using ionizing radiation. But some biological structures may turn out to be vulnerable to it. It needs to be tested. Further, EACH FREQUENCY to be used needs to be tested. With resonance phenomena and fixed molecular sizes you can't generalize from one frequency to another.
On the other hand, if tuned molecular effects ARE significant there is an opportunity to discover them and use them to create new beneficial interventions. B-)
A blood vessel isn't just empty space though. It has walls that expand and contract in response to various conditions. I assume normal vessels also resist intrusion.
As I read it this process does one of the following:
- reserves the space for the blood vessels to grow into a bit later (and perhaps marks it with growth factors to encourage vascularization along the paths.)
- provides a scaffold around which the blood vessels form, then dissolves away to leave the resulting vessels open.
I've just been running through the Babylon 5 DVDs.
A running theme was that the shadows (and their agents) were always asking "What do you want?", then using your wants to tempt you into their schemes.
Facebook seems to keep aligning with SF cycles' dark sides.
But in an urban environment these are not accurate signal strength is only loosely proportional to inverse square of the distance so any accuracy will utterly break down.
There's a bit more to it than that.
First: Inverse square falls off FAST near the transmitter. For distant transmitters, like TV stations and (rural) cellphone towers, you have the problem you describe. But for WiFi in nearby houses or other apartments in your own you can get a pretty good idea where you are just from strength.
Second: There's also timing signals.
TV stations (especially since the changeover to digital and the use of multiple synchronized transmitters sending the same signal on the same frequency) and cell towers all send fantastically accurate timing references. They are often deliberately synchronized to a common timebase, or inadvertently synchronized by referencing a common master clock (typically GPS) so they can be compared to like obtain path-distance differences between transmitters, creating loran-style hyperbolic loci. (If they aren't explicitly synchronized the database can track their current offsets by measuring and recording it - and updating as it responds to queries in its neighborhood.
In the synchronized TV signal case the distance difference on the path between different transmitters shows up as a pattern of relative signal strengths among the many carriers of the OFDM signal. You can get one loran-curve by analyzing just one station's signal.
Both TV and cellphone base station measurements can be confused somewhat by multipath reflections. On the other hand this can be small and distorted signals dropped in favor of others with better paths. Or it can even be used to get MORE information about location.
Distance to WiFI devices can be measured by actively interacting with them in various low-level ways that are a part of the standard and a legitimate thing for any other WiFi device to do.
I'm not saying that this is what the service does. But these are things it clearly COULD be doing. And they can produce an accuracy at least as good as what is claimed.
~Oh, disaster! The sea level rise on the east coast might be TRIPLE that of the world average in our previous prediction. It might rise 14 to 20 inches over the next century! That's as much as a whole 2 tenths of an inch per year! They're all going to drown.~
Somehow I'm not impressed. While it might be nice to see New York and Washington become awash, this is a number of orders of magnitude too low to be useful.
Compare this to Venice (which still seems to be doing very well, thank you.)
Another translated genomic data into musicâ"because when it comes to data-crunching and neuroscience, you can't be deadly serious all the time.
What's non-serious about that? Translating the data into music makes it accessible to the powerful rhythm, tonal sequence, and echo-imaging processing of the auditory system, which might identify interesting and useful features in the data.
Now if only we had one of these for AIDS
Ha ha but seriously...
Other neurological and/or muscular syndromes (such as ALS or Demyelinating diseases like MS) might produce detectable, but distinct, signatures.
It used to be fun to analyze public speaking in the frequency domain and find the 8 to 12 hertz oscillations. The sound equipment used in the stages at the White House appear to remove these in real time these days, though.
Or you may be observing that pathological liars aren't stressed when they lie.
If you look at what classes compile to, it's just structs and some syntactic sugar to automatically do what can be easily done manually to manipulate them.
Also some const vectors of pointers to functions. Like what you see in kernel driver tables.
I suspect, though, due to the way you are writing and asking the question, that you've never worked on a large C code base (which if they know what they are doing are already taking the good ideas from OOP methodology - but don't need the "class" keyword as a mental crutch).
Hear hear!
If you look at what classes compile to, it's just structs and some syntactic sugar to automatically do what can be easily done manually to manipulate them.
The main advantage of using the C++ compiler, rather than doing it yourself, is not that the compiler automates the code generation for the manipulation. It's that the compiler always gets it right on what it does for you and can check much of what YOU do to be sure you don't violate intended encapsulation boundaries or point to the wrong thing (unless you use casts to tell it you really meant to do something risky). So there's less opportunity to write a bug and have the compiler accept it.
ANSI C incorporates some of this with strong type checking (which it got from C++). But a real C++ compiler can make-right-or-detect-error in more ways.
Objective C, like Smalltalk, has an issue with construction and destruction: Overridden methods use the sub/derived version if called during the execution of the super/base class constructor. This means writing a sub/derived class has the potential to break the already-debugged super/base class constructruction, which must be rechecked for every sub/derived class.
C++ gets this ALMOST exactly right: Base class constructors get the base class version, derived class constructors get the derived class version. Iterate down the entire hierarchy. Equivalently with DEstructors as the successive levels are torn down.
I say ALMOST because the spec, and most compilers have what IMHO is a bug: During construction and destruction of member objects, if the member object constructor or destructor (or something it uses) gets hold of a pointer to the containing object and calls an overridden method it SHOULD get the base class version. (On construction the support for the derived class version is not initialized. On destruction it's already torn down.) But whether it actually gets the base or derived version is problematic.
If you take the binary combination of what the constructor and destructor get there are three "wrong" and one "right" combinations. In the early days cfront and cfront-derived compilers got it "wrong" one way, three compilers for PCs got it "wrong" another way, and gcc got it "wrong" the third way. So if (as I asked them on each of the first two standards efforts) they had specified using the "right" behavior, everybody would have had to make a minor change and nobody would have had an advantage.
The first ANSI standard explicitly left this undefined and the second said essentially "don't write code that lets this happen". B-b (I haven't really looked at the recent update to see what it says.)
This was important to the project I was working on back then: We had rolled our own garbage-collection and exception-handling systems, using preprocessor-defined overridden virtual function bodies to track the level of construction. (Catch and throw were still just reserved words and the politically correct way to handle heap management was with user-written explicit allocation and deallocation.) With this "bug" we couldn't count on correctly handling an exception thrown or garbage-collection triggered by a member object constructor or destructor. So we couldn't use them. Instead of building compact composite heap-allocated objects we had to use smartpointers, separately heap-allocate the "contained" objects, and knit together a cluster of separate objects, Smalltalk-style. B-b
Microsoft has publicly declared that they have no intention of supporting anything past C95
I gave up on Microsoft back when Byte magazine was in, or recently beyond, single-digit issue numbers.
A letter to the editor complained about Microsoft's support of their FORTRAN compiler: There was a bug in the floating point format handling that a customer needed to use. After several iterations of bug reports and fix requests, Microsoft had told the guy that not only had they not fixed it yet, but they were never going to fix it. IMHO that meant Microsoft had an institutional issue with customer support and adherence to standards.
Avoiding Microsoft software and products has saved me immeasurable grief over the several decades since.
And the US will tax you for years as well.
That has nothing to do with Switzerland, though.
But it does have a bunch to do with whether it's practical to move to Switzerland - or ANYWHERE else - from the US.
Then move.
It's REALLY hard to get citizenship in Switzerland. (Though John Walker, a founder of Autodesk, did manage it.)
And the US will tax you for years as well.