Why not use the 8 inch hard sectored floppy disks? Of course there's nowhere you could either read or write them... but they were a bit thinner, and I think they stored all of 100KB.
Not really. There's no available evidence either way, so assuming it either is or is not a simulation is equally parsimonious. There are some features of the observable universe that could most easily be explained if it were a simulation, and had been simplified in places, but then you get the problem of the containing universe.
OTOH, the idea isn't original with him (or at least he isn't the first to publish). And the evidence in favor isn't sufficient that Occam's razor would say it's the reasonable assumption. But the ONLY argument against it is the argument from incredulity, which is a pretty weak argument. (Yeah, I don't believe it either. Now try to prove it.)
And yet... Transport is only one arm of the problem, and blaming everything on the developers of automated transport is missing the point as much as claiming that there will be reasonable replacement jobs.
Automation *IS* coming. Fast...but not fast enough. There will be 2-3 more decades of transition, at the end of which almost nobody will have a job. Whether they will still be alive hasn't yet been decided. The choices we make today shape the form of the civilization that we can't really see coming. And during the transition there will be a lot of needed jobs, but nowhere near as many as there are people who need jobs. This is already happening, and has been in process since at least the 1980's. For awhile new jobs were created as fast as the old jobs were destroyed, but that had stopped happening by 1985. Perhaps earlier, because you can't trust the government records. They keep redefining "unemployment" to make the current administration look good. It wasn't, however, significant before around 1960, because before then the Supreme Court hadn't decided that cities couldn't have a residency requirement for general assistance, so there were reasonable estimates (though they were seasonal, and differed in different parts of the country).
Actually, there's a fair chance that one of the known models would be good enough if you ran the simulation at the cc level and used ray-tracing to evaluate radiation heating and cooling.
Of course, that much computer power is totally unreasonable, as well as that fine a detailed collection of data.
The thing is, it may well not be a problem of the models, but a problem of chaotic environment at a fine level. You can't do atomic motions out of temperature and thermodynamics, either. In principle you could, but the requirements for detailed knowledge of atomic speed and positions of the initial state are unworkable. This may be that kind of problem in a more complex area. You can get general rules, like hot gases expand, and by how much, and even use thermodynamics to design a 4-cycle engine, but you can't get detailed predictions of the motions of particular atoms.
Mind you, I don't know how you'd go about proving that this was an insoluble problem, or what bounds of accuracy you could put around things. I'm sure that we could do better than the current models allow. But we still don't know all the variables. E.g., how important is it how many of what kind of bacteria are living in that cloud? It's well possible that the answer to that question could be used to tell you whether or not to expect rain. But this isn't certain, and if it were certain we wouldn't know which kinds of bacteria were important, and how many it took. (And, as I said at the start, this might not be important if you did find enough scale modeling using the proper one of the current models.)
This is a wild guess, but they may have done it by counting the width of tree rings with a separate method for measuring rainfall. This would give a bit of a wide error bar compared to a thermometer, of course, but it's the approximate method used in many places before other measures were possible. (What they're really measuring, of course, is how good a year it was for tree growth, but that's frequently a good [as in, best available] stand-in. If you don't mind the wider error-bars.)
If you had done that, would you have wanted to let people know who you were, or who you worked for?
I'm not saying you should believe the story, but pick legitimate points. There seems to be an improbable level of stupidity. (OTOH, I'm known some managers...let's just say most of the one's I'm thinking of wouldn't have thought to ask for help.)
You had made a block copy backup via direct connection first, right? (I'm just assuming you can still do that, since it's been several decades since I had someone do it for me.)
Backups are the best answer, but if you don't have backups, the first step should be to make a sector-by-sector backup. And I, personally, have had more success having a hardware guy do it for me in hardware. OTOH, I'm not sure I could even edit a disk anymore. The last time I needed to to that I was using MSWind95.
I was thinking of using a laser/light-sail combination, so the laser is going to need to be on a LONG time. If you're using ablative thrust, or some such, then yes, the laser would only be on a short time, and that might be a reasonable way to start, as it would give you much higher thrust than the lightsail, but at the cost of ejection mass. So after a bit switching to a light-sail would be better. And that means a long laser thrust.
Also, while dust does drift, I believe the average velocity (relative to the velocity of the sun) is small, so there shouldn't be a lot of dust drifting in from the side.
Of course, this is just thumbnail figuring (not even back of the envelope). But presumably this would be covered during the research project. Probably rather early in the project.
Well, the laser beam is going to be a lot wider than the craft mirror when you get to any distance, and there will also be some bending around the edge of the sail. So I don't think taking down the sail will be needed.
Also, a few holes in the sail aren't going to be important. All that matters is the payload, which will be a lot smaller.
Now if they were flying into a cloud of bbs, or even something 1/1000th that weight, I'd be thinking very differently, but most interstellar dust is going to be lighter than the dust that floats in the air, and that's light enough to be shoved around by a laser.
Perhaps you could explain how black holes prevent loss of information when the swallow something, then.
He's talking out of his area of expertise, but that's a different matter. And since what he's proposing is a research project, it sounds quite reasonable to me....Of course, I may have been trolled.
I think the estimate of dust particles is too high, because a bright laser has just been shining along that column of space. That should move or evaporate much of the dust.
Also, the spacecraft itself shouldn't be a mesh, but rather, as you said, a net. you need tensional stability. You probably even want it to be rotating (slowly, you don't need *much* tension). But communication between the nodes could well be via a mesh network.
I'm sure this kind of thing would be addressed during the research phase of this project. And we probably aren't up to building something like this yet anyway...not using nanotechnology. Give us 5 or 10 years thought and...
When are you assuming the aliens put the devices in place? If it was before around 1900 there's no way we would have noticed. For many frequencies there's no way we would have noticed before around 1970. Even now noticing them might not be easy. Wide scan surveys don't cover the full field, so most of the time nobody's looking in any particular direction, and most frequencies are essentially blocked by the atmosphere. (You wouldn't use radio for propulsion.)
Maybe the laser propulsion system would automatically clean the dust out of the way. Dust shouldn't take much to give it a shove, or possibly to evaporate it.
Some designs call for the sail to also be used in decelerating the craft. It does mean discarding the sail on arrival, of course, since you slow down based on reflected light from the sail hitting the craft, leading to the sail be accelerated and the craft slowed. And it requires that you trust that the propulsion lasers will still be working when you need to slow down. (Receivers are generally a lot easier to fund.)
Actually, there are many indications that relativity is wrong. They aren't convincing, but there are indications.
HOWEVER: General Relativity is clearly incomplete. So is Quantum Theory. And they appear to disagree in certain areas when they make predictions. This may, however, but an illusion. They use such different math to make their predictions that one can't quite be certain.
E.G.: Do black holes destroy information when they swallow things? No fully consistent answer appears to exist.
Unfortunately(?), this doesn't lead to a prediction that FTL communication is possible.
IIUC the "Privacy Shield" is intended to replace the current data-sharing arrangement between the US and the EU. An EU court said that the current arrangement violated the rights of EU citizens, but gave them some time to craft a replacement program.
Since the US is in favor of "Privacy Shield" one may guess that it's a bit pervious, but that's not proof, and I'm no lawyer.
OTOH, my cynicism is such that if MS in in favor of it, I expect that it would be bad for me.
I don't know the PDP-8, it might be a good choice. But in covering the assembler the goal is to teach basic principles, how integers are represented, what overflow means, etc. so "a bit more practical" is beside the point, and likely to be actively detrimental. Real CPUs tend to be complex, and you don't want complexity to hide the basics you're trying to convey.
Certainly, anything that could run Linux would be inappropriate, as Linux requires a memory management unit, and that's complexity beyond the desirable level.
C has it's points, but it doesn't give you the same understanding that a nice simple assembler does.
Please note: I'm not talking about an assembler for current chips. That's more complex than C, and harder to wrap your mind around. I'm talking about one that is SIMPLE. Which is why I mentioned MIX and the IBM 7090. Even the Apple ][ assembler is more complex than is ideal. And the M6800 assembler was simpler than the Z80 assembler, but I'm not sure about the i8080, it might have been just as simple.
But NO paging, and it's desirable to only have couple of registers. Those are advanced concepts, and most people don't get any advantage to compensate for the additional complexity.
Please note, I'm not talking about a full semester class in assembler. Probably only a month of the first class. Then transition to C as the next simpler language. (The complexity of C comes when you start using libraries, it's not in the basic language.) After a couple of months of that you can head in nearly any direction. You've got the foundations in place.
Security *is* a very hard problem, and if you insist on perfection impossible. This doesn't mean it isn't worth trying for.
OTOH, some "security" practices are just stupid. E.g., change your password every month to a new alphabetic string longer than 8 characters containing at least one punctuation character and at least one digit. And no repetition. That's a guaranteed recipe for work-arounds that break security.
Which is why I often think that the first class in university computing should be assembler. Possibly MIX or some other really simple virtual machine. (What I'd really like is a virtual IBM 7090, or possibly a bit earlier in that series, but I've never seen one. I virtual Apple ][ would also be good, the i6502 was a nice simple machine. But Apple would probably complain, while the IBM 7090 is out of trademark and patent protection.)
I don't know. When I took computer science, it was algorithm design and numerical analysis. Security wasn't even mentioned. But that was before public access to the Internet, so perhaps things are different now.
However, my expectation would be that security wouldn't be handled under Computer Science. And since Computer Engineering was a major under Electrical Engineering, and included things like designing half-adders, that doesn't sound like the right place either.
Perhaps there needs to be an Information Technologies major that WOULD be an appropriate home for computer security. The OS specific parts, though, would be a bugger even for that major. Linux seems pretty stable, but what would you put in a course on MSWindows security? That wouldn't change with the next version? (OK, I haven't used MSWind for over a decade...perhaps there is something reasonable.) It seems as if OS specific security should be covered under multiple "short courses" about a month long as a second or third year class in the IT major.
Besides, the effect on the distribution of thermal energy of the described motion would be difficult to measure. It's too small. I do expect it might result in a slightly longer day as water flowed down towards the equator. But microseconds might be too coarse a measure of time.
Actually, that's one of the benefits of the moon. It contributes to gyroscopic stability. There's even a school that claims any planet that's going to have successful life needs to have a large moon to prevent gyroscopic tumbling killing everything off. I don't know how good their claims are, and I've never tried to check their math, but it's not totally unreasonable.
Depends on what you're measuring. If you're measuring rainfall, it's climate change. If you're measuring temperature it's global warming. And, of course, newspapers are notorious for abusing the language. Consider the term "hacker".
Actually, it could always be climate change, but calling it global warming when you're talking about rainfall is just confusing.
Slashdot Mirror
Why not use the 8 inch hard sectored floppy disks? Of course there's nowhere you could either read or write them... but they were a bit thinner, and I think they stored all of 100KB.
Not really. There's no available evidence either way, so assuming it either is or is not a simulation is equally parsimonious. There are some features of the observable universe that could most easily be explained if it were a simulation, and had been simplified in places, but then you get the problem of the containing universe.
OTOH, the idea isn't original with him (or at least he isn't the first to publish). And the evidence in favor isn't sufficient that Occam's razor would say it's the reasonable assumption. But the ONLY argument against it is the argument from incredulity, which is a pretty weak argument. (Yeah, I don't believe it either. Now try to prove it.)
And yet...
Transport is only one arm of the problem, and blaming everything on the developers of automated transport is missing the point as much as claiming that there will be reasonable replacement jobs.
Automation *IS* coming. Fast...but not fast enough. There will be 2-3 more decades of transition, at the end of which almost nobody will have a job. Whether they will still be alive hasn't yet been decided. The choices we make today shape the form of the civilization that we can't really see coming. And during the transition there will be a lot of needed jobs, but nowhere near as many as there are people who need jobs. This is already happening, and has been in process since at least the 1980's. For awhile new jobs were created as fast as the old jobs were destroyed, but that had stopped happening by 1985. Perhaps earlier, because you can't trust the government records. They keep redefining "unemployment" to make the current administration look good. It wasn't, however, significant before around 1960, because before then the Supreme Court hadn't decided that cities couldn't have a residency requirement for general assistance, so there were reasonable estimates (though they were seasonal, and differed in different parts of the country).
Actually, there's a fair chance that one of the known models would be good enough if you ran the simulation at the cc level and used ray-tracing to evaluate radiation heating and cooling.
Of course, that much computer power is totally unreasonable, as well as that fine a detailed collection of data.
The thing is, it may well not be a problem of the models, but a problem of chaotic environment at a fine level. You can't do atomic motions out of temperature and thermodynamics, either. In principle you could, but the requirements for detailed knowledge of atomic speed and positions of the initial state are unworkable. This may be that kind of problem in a more complex area. You can get general rules, like hot gases expand, and by how much, and even use thermodynamics to design a 4-cycle engine, but you can't get detailed predictions of the motions of particular atoms.
Mind you, I don't know how you'd go about proving that this was an insoluble problem, or what bounds of accuracy you could put around things. I'm sure that we could do better than the current models allow. But we still don't know all the variables. E.g., how important is it how many of what kind of bacteria are living in that cloud? It's well possible that the answer to that question could be used to tell you whether or not to expect rain. But this isn't certain, and if it were certain we wouldn't know which kinds of bacteria were important, and how many it took. (And, as I said at the start, this might not be important if you did find enough scale modeling using the proper one of the current models.)
This is a wild guess, but they may have done it by counting the width of tree rings with a separate method for measuring rainfall. This would give a bit of a wide error bar compared to a thermometer, of course, but it's the approximate method used in many places before other measures were possible. (What they're really measuring, of course, is how good a year it was for tree growth, but that's frequently a good [as in, best available] stand-in. If you don't mind the wider error-bars.)
If you had done that, would you have wanted to let people know who you were, or who you worked for?
I'm not saying you should believe the story, but pick legitimate points. There seems to be an improbable level of stupidity. (OTOH, I'm known some managers...let's just say most of the one's I'm thinking of wouldn't have thought to ask for help.)
You had made a block copy backup via direct connection first, right? (I'm just assuming you can still do that, since it's been several decades since I had someone do it for me.)
Backups are the best answer, but if you don't have backups, the first step should be to make a sector-by-sector backup. And I, personally, have had more success having a hardware guy do it for me in hardware. OTOH, I'm not sure I could even edit a disk anymore. The last time I needed to to that I was using MSWind95.
I was thinking of using a laser/light-sail combination, so the laser is going to need to be on a LONG time. If you're using ablative thrust, or some such, then yes, the laser would only be on a short time, and that might be a reasonable way to start, as it would give you much higher thrust than the lightsail, but at the cost of ejection mass. So after a bit switching to a light-sail would be better. And that means a long laser thrust.
Also, while dust does drift, I believe the average velocity (relative to the velocity of the sun) is small, so there shouldn't be a lot of dust drifting in from the side.
Of course, this is just thumbnail figuring (not even back of the envelope). But presumably this would be covered during the research project. Probably rather early in the project.
Well, the laser beam is going to be a lot wider than the craft mirror when you get to any distance, and there will also be some bending around the edge of the sail. So I don't think taking down the sail will be needed.
Also, a few holes in the sail aren't going to be important. All that matters is the payload, which will be a lot smaller.
Now if they were flying into a cloud of bbs, or even something 1/1000th that weight, I'd be thinking very differently, but most interstellar dust is going to be lighter than the dust that floats in the air, and that's light enough to be shoved around by a laser.
Perhaps you could explain how black holes prevent loss of information when the swallow something, then.
He's talking out of his area of expertise, but that's a different matter. And since what he's proposing is a research project, it sounds quite reasonable to me. ...Of course, I may have been trolled.
I think the estimate of dust particles is too high, because a bright laser has just been shining along that column of space. That should move or evaporate much of the dust.
Also, the spacecraft itself shouldn't be a mesh, but rather, as you said, a net. you need tensional stability. You probably even want it to be rotating (slowly, you don't need *much* tension). But communication between the nodes could well be via a mesh network.
I'm sure this kind of thing would be addressed during the research phase of this project. And we probably aren't up to building something like this yet anyway...not using nanotechnology. Give us 5 or 10 years thought and ...
When are you assuming the aliens put the devices in place? If it was before around 1900 there's no way we would have noticed. For many frequencies there's no way we would have noticed before around 1970. Even now noticing them might not be easy. Wide scan surveys don't cover the full field, so most of the time nobody's looking in any particular direction, and most frequencies are essentially blocked by the atmosphere. (You wouldn't use radio for propulsion.)
Maybe the laser propulsion system would automatically clean the dust out of the way. Dust shouldn't take much to give it a shove, or possibly to evaporate it.
Some designs call for the sail to also be used in decelerating the craft. It does mean discarding the sail on arrival, of course, since you slow down based on reflected light from the sail hitting the craft, leading to the sail be accelerated and the craft slowed. And it requires that you trust that the propulsion lasers will still be working when you need to slow down. (Receivers are generally a lot easier to fund.)
Actually, there are many indications that relativity is wrong. They aren't convincing, but there are indications.
HOWEVER: General Relativity is clearly incomplete. So is Quantum Theory. And they appear to disagree in certain areas when they make predictions. This may, however, but an illusion. They use such different math to make their predictions that one can't quite be certain.
E.G.: Do black holes destroy information when they swallow things? No fully consistent answer appears to exist.
Unfortunately(?), this doesn't lead to a prediction that FTL communication is possible.
IIUC the "Privacy Shield" is intended to replace the current data-sharing arrangement between the US and the EU. An EU court said that the current arrangement violated the rights of EU citizens, but gave them some time to craft a replacement program.
Since the US is in favor of "Privacy Shield" one may guess that it's a bit pervious, but that's not proof, and I'm no lawyer.
OTOH, my cynicism is such that if MS in in favor of it, I expect that it would be bad for me.
I don't know the PDP-8, it might be a good choice. But in covering the assembler the goal is to teach basic principles, how integers are represented, what overflow means, etc. so "a bit more practical" is beside the point, and likely to be actively detrimental. Real CPUs tend to be complex, and you don't want complexity to hide the basics you're trying to convey.
Certainly, anything that could run Linux would be inappropriate, as Linux requires a memory management unit, and that's complexity beyond the desirable level.
C has it's points, but it doesn't give you the same understanding that a nice simple assembler does.
Please note: I'm not talking about an assembler for current chips. That's more complex than C, and harder to wrap your mind around. I'm talking about one that is SIMPLE. Which is why I mentioned MIX and the IBM 7090. Even the Apple ][ assembler is more complex than is ideal. And the M6800 assembler was simpler than the Z80 assembler, but I'm not sure about the i8080, it might have been just as simple.
But NO paging, and it's desirable to only have couple of registers. Those are advanced concepts, and most people don't get any advantage to compensate for the additional complexity.
Please note, I'm not talking about a full semester class in assembler. Probably only a month of the first class. Then transition to C as the next simpler language. (The complexity of C comes when you start using libraries, it's not in the basic language.) After a couple of months of that you can head in nearly any direction. You've got the foundations in place.
Security *is* a very hard problem, and if you insist on perfection impossible. This doesn't mean it isn't worth trying for.
OTOH, some "security" practices are just stupid. E.g., change your password every month to a new alphabetic string longer than 8 characters containing at least one punctuation character and at least one digit. And no repetition. That's a guaranteed recipe for work-arounds that break security.
Which is why I often think that the first class in university computing should be assembler. Possibly MIX or some other really simple virtual machine. (What I'd really like is a virtual IBM 7090, or possibly a bit earlier in that series, but I've never seen one. I virtual Apple ][ would also be good, the i6502 was a nice simple machine. But Apple would probably complain, while the IBM 7090 is out of trademark and patent protection.)
I don't know. When I took computer science, it was algorithm design and numerical analysis. Security wasn't even mentioned. But that was before public access to the Internet, so perhaps things are different now.
However, my expectation would be that security wouldn't be handled under Computer Science. And since Computer Engineering was a major under Electrical Engineering, and included things like designing half-adders, that doesn't sound like the right place either.
Perhaps there needs to be an Information Technologies major that WOULD be an appropriate home for computer security. The OS specific parts, though, would be a bugger even for that major. Linux seems pretty stable, but what would you put in a course on MSWindows security? That wouldn't change with the next version? (OK, I haven't used MSWind for over a decade...perhaps there is something reasonable.) It seems as if OS specific security should be covered under multiple "short courses" about a month long as a second or third year class in the IT major.
Besides, the effect on the distribution of thermal energy of the described motion would be difficult to measure. It's too small. I do expect it might result in a slightly longer day as water flowed down towards the equator. But microseconds might be too coarse a measure of time.
Actually, that's one of the benefits of the moon. It contributes to gyroscopic stability. There's even a school that claims any planet that's going to have successful life needs to have a large moon to prevent gyroscopic tumbling killing everything off. I don't know how good their claims are, and I've never tried to check their math, but it's not totally unreasonable.
If you want to see it really tilt, wait until Antarctica really melts. But you'd better check your hip boots for water-tightness first.
Depends on what you're measuring. If you're measuring rainfall, it's climate change. If you're measuring temperature it's global warming. And, of course, newspapers are notorious for abusing the language. Consider the term "hacker".
Actually, it could always be climate change, but calling it global warming when you're talking about rainfall is just confusing.