You might want to read up on the Ares I and note that is has a liquid second stage.
I'm curious, by what mechanism would the liquid propellant in the *second stage* have any positive feedback (not damping) to thrust oscillations in the *first stage*?
The reason longitudinal oscillations (call them what you like) can be really huge in a liquid is because of the positive feedback loop in which a thrust oscillation causes a reinforcing change in the inlet pressure of the pumps, this mechanism does not exist in a solid; that was my only point.
I guess I just disagree that 'any longitudinal vibration is pogo'. YMMV.
How is parent offtopic?
The topic seems to be an error caused by a single bit flip. Most implicit algorithms do have as part of them an 'error smoothing' component, either as a preconditioner or as the whole algorithm. So an error at a single point would get smoothed out...
The vibrations that are commonly called 'pogo' in big rockets are caused by a feedback / resonance of thrust oscillations with inlet pressure of the turbopumps, see this extensive discussion.
Pogo is fixed by adding dampers to the propellant lines. Ares I, like every big solid, has combustion instabilities that cause thrust oscillations, but there's no feedback like in a liquid rocket. Only danger is hitting one of the structural resonances and ringing the rocket like a bell (and possibly causing the structure to 'diverge').
You can't tack in a magnetic field. Well, you could, but it would require an actual change in the magnetic field.
I think you could 'tack', sort of; the force is in the direction mutually perpendicular to the tether and the field (I x B), like you said you can't change the orientation of the field, but you could change the orientation of the tether with respect to the field lines and the direction of the current flow.
Remember, you can't make a net gain of energy in a closed system.
There's no net change, an electrodynamic tether basically uses the whole earth (through the magnetic field) as the reaction mass, any change in the tether momentum is balanced by an equal and opposite change in the earth's momentum.
Can't this "drag" be used for some near earth maneuvering... ?
Yep, just need a power source to 'push' the current the other way; check out electrodynamic tether, pretty neat.
Like other electric propulsion concepts it is power limited, so it's kind of so-so with solar panels and batteries, but would probably really kick ass with nukes.
Absolutely. Allowing consensus to substitute for good experimental design, solid data analysis and independent reproducibility leaves the door wide open to this sort of abuse. "Consensus Science" is not science. Charlie Rose interview with Michael Crichton, where he makes this point in reference to global warming.
This is really a reinvention and extension of 1940's British technology.... It was used with devastating effect against the German U Boat pens, canals, bridges and viaducts where the "earthquake" effect of a deep explosion undermined foundations.
That happens at the Radio Shacks in the US too; I was looking for a little momentary switch for a project a while back and the dude walks over and asks me what I'm looking for and tries to pretend he's interested in my project (which is just awkward because he's clueless), but he doesn't really know what they carry or what might work instead in a pinch.
So all he accomplished was making sure I didn't pilfer a few pennies worth of potentiometers and make me feel awkward because he was such a no-help-giving goober. Thanks. Waiting a couple days for the mail order keeps looking better.
Parent is not flaimbait; the summary and the article are bullshit and cite no relevant sources (they obviously don't understand how academics work).
No one has to 'redo pretty much the identical experiment' because they published some results and analysis based on some measurements. The journal owns the copyright on the *article* not the *data*...
And citing previous academic works is the definition of fair use...
Especially since the topic is a controlled experiment, where it is actually possible (and proper) to infer a causal relationship. This isn't survey data on browsers, f'in slashtards.
3D imaging, especially MRI, is hideously complicated and indirect. It's almost inconceivable that it could yield results with any physical significance.
...and yet, it does. It's become so routine, so reliable, so well-understood and well-controlled, that doctors and researchers know they can rely on it as a matter of course. They still have to be aware of the errors and distortions that can arise, but that's true of every imaging or monitoring system, all the way down to the stethoscope and the fever thermometer.
Modelling the measurement is fundamental to physical science, even something as simple as measuring the temperature of air, and these are much more complex measurements to model.
The assumptions underlying the choice of forward model and measurement process that allow you to use the FFT approach turn an ill-posed problem into a well-posed one, true.
Making different assumptions leads to different reconstruction techniques. *Any* scan like this (CT, MRI, ultrasound, infrared, microwave) involves dealing with an ill-posed problem in some way. It may be dealt with before any of the researchers in question touch the machine or the subjects, but it was dealt with by someone at some point.
I think (based on the admittedly slim article) their criticism is based on many researchers not properly accounting for pseudoreplication, which tends to be a problem when you have tons and tons of data like in an fMRI scan, but it is only on a few individuals, or it is on one individual measured a couple different times. The way to properly treat this is with split-plot or mixed effects models, this is a bit harder to get right than just fitting a linear model and looking at error bars.
The compounding problem with any sort of CT scan is that you have to make assumptions / regularizations to create the image from the measurements in the first place, so the bias introduced by this process is probably not treated well either (it's not easy).
These sort of images are pretty familiar to me and I must admit I was never skeptical of research [...] it's a shame that one of the few tools used to determine the hows and whys of it is being called into question.
I don't think this 'uncertainty' is anything new. Computing a tomographic reconstruction is an ill-posed problem, you can do least squares, you can be a Bayesian, but in the end you have to introduce information or assumptions to fill out the "null space" of your measurements.
I think there's been a lack of understanding on the part of many folks in the medical community about just what kinds of assumptions go into making those pretty CT and MRI results. Treating spurious features in reconstructions due to the measurement and regularization technique was an intense area of early research in this field.
Would someone who is technically competent "push his pet project," as you say, in the face of solid analysis that says his bar napkin rocket wasn't really that good an idea after all?
Technical competence means being willing to walk away from your own sexy ideas when they don't stand up to the cold light of reality. It means not using your positional authority to support your ego / biases.
I would argue that he was incompetent despite his many degrees. Falling in love with your own ideas breaks Feynman's first principle.
I'll agree with you about lack of vision/leadership, but that is a symptom not a cause. The system rewards and promotes a certain type of manager, right now NASA's system promotes and rewards bureaucrats (like most gov orgs, I'm not just picking on NASA) rather than technically competent leaders. You get the leadership that the system gives you, this has nothing to do with funding levels. I've seen really great leaders do awesome stuff on a shoestring.
I've consistently been impressed by the technical competence of NASA engineers in my field (especially out of Langley), but as a taxpayer I've been consistently disappointed by their management (you're right, many other orgs have the same problems).
Buzz is a smart guy. Robert Zubrin's books (Case for Mars, Entering Space ) also address why the Moon is *not* a good stepping stone to Mars. He has, *gasp*, actual rocket science to back up his argument (basic orbital mechanics and the rocket equation) rather than silly rah-rah from NASA management about reliving the glory days by returning to the moon. The really frustrating thing about it is that if you've had a high-school physics class and a calculus course you can understand the argument, no need for NASA to dumb it down (too much) for public debate (I know, I know, I must be new here).
That's certainly the party line, but it's not consistent with reality. The problem with NASA is NASA. See Rogers commission report or the Columbia accident investigation report, Tufte's or Feynman's criticisms. NASA is a bureaucracy that's sicker than most, and the criticisms haven't changed much over the last few decades. They aren't quite terminal yet, but they are circling the drain. COTS is a bright spot, firm fixed price for actual accomplishments rather than cost-plus for power point slides and paralysis by analysis (Lock-Mart and Boeing I'm looking at you), that's a win no matter how you look at it.
Check out the Direct slides (it's unfortunately bloated compared to the rest of the files by some superfluous vids), they recommend an approach that uses a propellant depot. Direct also gives you one modular family of vehicles, like the EELV and unlike current Constellation, this is an indication of the very poor fundamental design choices made by NASA technocrats using powerpoint engineering (seeTufte).
Completely scrapping the shuttle would be wasteful when it already *is* a Saturn V class launch vehicle if you don't have to lug the orbiter around all the time. Plus it is more efficient, fuel and structure-wise than the Saturn V.
Going the EELV route has lots more unknown unknowns than a shuttle derivative.
I think the Direct + COTS approach is a win. Direct leverages existing infrastructure/technology/workforce, SpaceX is lean and actually throwing mass.
Hey Thanks!
Here's a good survey of 'symplectic integrators':
http://math.berkeley.edu/~alanw/242papers99/markiewicz.pdf
Basically, choosing the parameters of your integration scheme smartly (based on the system you're integrating) so you do a better job at conserving energy for long time integrations. The example they give in that survey article is exactly this problem (solar system orbit simulation).
One of the other articles (on CNET I think) says they used a 'unique time integration scheme', anybody know which one? I don't have a subscription to Nature, and it isn't addressed in their abstract. Thanks!
Except I'm pretty sure that there are modern languages and libraries that can handle this without Fortran. I don't have much experience with it myself, but I'm pretty sure that's exactly what MATLAB is for, for one.
Dude, what do you think the libraries Matlab uses are written in?
Check out netlib to get an idea, ATLAS, BLAS, LAPACK, LINPACK, etc. Matlab stands on the shoulders of the giants of scientific computing (implemented for the most part in Fortran).
Slashdot Mirror
The space shuttle's solids have a small amount of thrust vector control.
You might want to read up on the Ares I and note that is has a liquid second stage.
I'm curious, by what mechanism would the liquid propellant in the *second stage* have any positive feedback (not damping) to thrust oscillations in the *first stage*?
The reason longitudinal oscillations (call them what you like) can be really huge in a liquid is because of the positive feedback loop in which a thrust oscillation causes a reinforcing change in the inlet pressure of the pumps, this mechanism does not exist in a solid; that was my only point.
I guess I just disagree that 'any longitudinal vibration is pogo'. YMMV.
How is parent offtopic? The topic seems to be an error caused by a single bit flip. Most implicit algorithms do have as part of them an 'error smoothing' component, either as a preconditioner or as the whole algorithm. So an error at a single point would get smoothed out ...
The vibrations that are commonly called 'pogo' in big rockets are caused by a feedback / resonance of thrust oscillations with inlet pressure of the turbopumps, see this extensive discussion. Pogo is fixed by adding dampers to the propellant lines. Ares I, like every big solid, has combustion instabilities that cause thrust oscillations, but there's no feedback like in a liquid rocket. Only danger is hitting one of the structural resonances and ringing the rocket like a bell (and possibly causing the structure to 'diverge').
You can't tack in a magnetic field. Well, you could, but it would require an actual change in the magnetic field.
I think you could 'tack', sort of; the force is in the direction mutually perpendicular to the tether and the field (I x B), like you said you can't change the orientation of the field, but you could change the orientation of the tether with respect to the field lines and the direction of the current flow.
Remember, you can't make a net gain of energy in a closed system.
There's no net change, an electrodynamic tether basically uses the whole earth (through the magnetic field) as the reaction mass, any change in the tether momentum is balanced by an equal and opposite change in the earth's momentum.
Can't this "drag" be used for some near earth maneuvering ... ?
Yep, just need a power source to 'push' the current the other way; check out electrodynamic tether, pretty neat. Like other electric propulsion concepts it is power limited, so it's kind of so-so with solar panels and batteries, but would probably really kick ass with nukes.
In real science, no consensus is required.
Absolutely. Allowing consensus to substitute for good experimental design, solid data analysis and independent reproducibility leaves the door wide open to this sort of abuse. "Consensus Science" is not science. Charlie Rose interview with Michael Crichton, where he makes this point in reference to global warming.
This is really a reinvention and extension of 1940's British technology. ... It was used with devastating effect against the German U Boat pens, canals, bridges and viaducts where the "earthquake" effect of a deep explosion undermined foundations.
Those U-boat pens had nothing on the massive, layered earth and concrete protection of modern 'hard and deep' targets: http://www.globalsecurity.org/wmd/world/iran/natanz-imagery.htm
That happens at the Radio Shacks in the US too; I was looking for a little momentary switch for a project a while back and the dude walks over and asks me what I'm looking for and tries to pretend he's interested in my project (which is just awkward because he's clueless), but he doesn't really know what they carry or what might work instead in a pinch.
So all he accomplished was making sure I didn't pilfer a few pennies worth of potentiometers and make me feel awkward because he was such a no-help-giving goober. Thanks. Waiting a couple days for the mail order keeps looking better.
Parent is not flaimbait; the summary and the article are bullshit and cite no relevant sources (they obviously don't understand how academics work).
No one has to 'redo pretty much the identical experiment' because they published some results and analysis based on some measurements. The journal owns the copyright on the *article* not the *data* ...
And citing previous academic works is the definition of fair use...
Fucking slashtards
I think you mean Bilski... http://www.mondaq.com/article.asp?articleid=69582
Especially since the topic is a controlled experiment, where it is actually possible (and proper) to infer a causal relationship. This isn't survey data on browsers, f'in slashtards.
Mod parent up.
3D imaging, especially MRI, is hideously complicated and indirect. It's almost inconceivable that it could yield results with any physical significance.
...and yet, it does. It's become so routine, so reliable, so well-understood and well-controlled, that doctors and researchers know they can rely on it as a matter of course. They still have to be aware of the errors and distortions that can arise, but that's true of every imaging or monitoring system, all the way down to the stethoscope and the fever thermometer.
Modelling the measurement is fundamental to physical science, even something as simple as measuring the temperature of air, and these are much more complex measurements to model.
Making different assumptions leads to different reconstruction techniques. *Any* scan like this (CT, MRI, ultrasound, infrared, microwave) involves dealing with an ill-posed problem in some way. It may be dealt with before any of the researchers in question touch the machine or the subjects, but it was dealt with by someone at some point.
I think (based on the admittedly slim article) their criticism is based on many researchers not properly accounting for pseudoreplication, which tends to be a problem when you have tons and tons of data like in an fMRI scan, but it is only on a few individuals, or it is on one individual measured a couple different times. The way to properly treat this is with split-plot or mixed effects models, this is a bit harder to get right than just fitting a linear model and looking at error bars.
The compounding problem with any sort of CT scan is that you have to make assumptions / regularizations to create the image from the measurements in the first place, so the bias introduced by this process is probably not treated well either (it's not easy).
These sort of images are pretty familiar to me and I must admit I was never skeptical of research [...] it's a shame that one of the few tools used to determine the hows and whys of it is being called into question.
I don't think this 'uncertainty' is anything new. Computing a tomographic reconstruction is an ill-posed problem, you can do least squares, you can be a Bayesian, but in the end you have to introduce information or assumptions to fill out the "null space" of your measurements.
I think there's been a lack of understanding on the part of many folks in the medical community about just what kinds of assumptions go into making those pretty CT and MRI results. Treating spurious features in reconstructions due to the measurement and regularization technique was an intense area of early research in this field.
Computed Tomography
Tomographic Reconstruction
Would someone who is technically competent "push his pet project," as you say, in the face of solid analysis that says his bar napkin rocket wasn't really that good an idea after all?
Technical competence means being willing to walk away from your own sexy ideas when they don't stand up to the cold light of reality. It means not using your positional authority to support your ego / biases.
I would argue that he was incompetent despite his many degrees. Falling in love with your own ideas breaks Feynman's first principle.
I'll agree with you about lack of vision/leadership, but that is a symptom not a cause. The system rewards and promotes a certain type of manager, right now NASA's system promotes and rewards bureaucrats (like most gov orgs, I'm not just picking on NASA) rather than technically competent leaders. You get the leadership that the system gives you, this has nothing to do with funding levels. I've seen really great leaders do awesome stuff on a shoestring. I've consistently been impressed by the technical competence of NASA engineers in my field (especially out of Langley), but as a taxpayer I've been consistently disappointed by their management (you're right, many other orgs have the same problems).
Buzz is a smart guy. Robert Zubrin's books (Case for Mars, Entering Space ) also address why the Moon is *not* a good stepping stone to Mars. He has, *gasp*, actual rocket science to back up his argument (basic orbital mechanics and the rocket equation) rather than silly rah-rah from NASA management about reliving the glory days by returning to the moon. The really frustrating thing about it is that if you've had a high-school physics class and a calculus course you can understand the argument, no need for NASA to dumb it down (too much) for public debate (I know, I know, I must be new here).
That's certainly the party line, but it's not consistent with reality. The problem with NASA is NASA. See Rogers commission report or the Columbia accident investigation report, Tufte's or Feynman's criticisms. NASA is a bureaucracy that's sicker than most, and the criticisms haven't changed much over the last few decades. They aren't quite terminal yet, but they are circling the drain. COTS is a bright spot, firm fixed price for actual accomplishments rather than cost-plus for power point slides and paralysis by analysis (Lock-Mart and Boeing I'm looking at you), that's a win no matter how you look at it.
Check out the Direct slides (it's unfortunately bloated compared to the rest of the files by some superfluous vids), they recommend an approach that uses a propellant depot. Direct also gives you one modular family of vehicles, like the EELV and unlike current Constellation, this is an indication of the very poor fundamental design choices made by NASA technocrats using powerpoint engineering (see Tufte). Completely scrapping the shuttle would be wasteful when it already *is* a Saturn V class launch vehicle if you don't have to lug the orbiter around all the time. Plus it is more efficient, fuel and structure-wise than the Saturn V. Going the EELV route has lots more unknown unknowns than a shuttle derivative. I think the Direct + COTS approach is a win. Direct leverages existing infrastructure/technology/workforce, SpaceX is lean and actually throwing mass.
Hey Thanks! Here's a good survey of 'symplectic integrators': http://math.berkeley.edu/~alanw/242papers99/markiewicz.pdf Basically, choosing the parameters of your integration scheme smartly (based on the system you're integrating) so you do a better job at conserving energy for long time integrations. The example they give in that survey article is exactly this problem (solar system orbit simulation).
Nope, it was the Ars article: http://arstechnica.com/science/news/2009/06/kicking-a-planet-out-of-the-solar-system-physically.ars They mention a 'complex time integration scheme' fourth paragraph down. Though with the state of technical journalism on the intrawebs that could mean Euler or leap-frog.
One of the other articles (on CNET I think) says they used a 'unique time integration scheme', anybody know which one? I don't have a subscription to Nature, and it isn't addressed in their abstract. Thanks!
Except I'm pretty sure that there are modern languages and libraries that can handle this without Fortran. I don't have much experience with it myself, but I'm pretty sure that's exactly what MATLAB is for, for one.
Dude, what do you think the libraries Matlab uses are written in? Check out netlib to get an idea, ATLAS, BLAS, LAPACK, LINPACK, etc. Matlab stands on the shoulders of the giants of scientific computing (implemented for the most part in Fortran).