The broken window fallacy does refer to a broken window, but the analogy is not limited to destruction. It can be expanded to any expenditure you were forced to make that shouldn't have been necessary. The 'forcing the expenditure' is a negative, and can't be re-cast as a positive because of the economic flow that resulted from it. While property destruction and loss of capital exacerbates the situation, lack of literal destruction does not invalidate the analogy.
What would invalidate the analogy is if the forced expenditure is filling a need that would have existed without that particular forcing function. The previous poster's statement was that spending all this money was not the answer, not being there was. Part of your response was that 'yeah, but look at all the good stuff coming out of these expenditures that go into our economy'. Using that statement as a rebuttal to the statement that the money shouldn't have been needed to be spent is very close to the broken window fallacy, and I still stand by it as a valid logical counter to that part of your argument.
That said, don't make assumptions about my bias. If it wasn't for said defense expenditures, I'd be out of a job. Us "engineers with advanced understandings of calculus, physics, and chemistry" know perfectly well that weapons ain't easy to design. We also tend to have decent understandings of logical progressions of thought, and would rather if you didn't use fallacious arguments to defend your points when legitimate ones are readily available.
Scientists and Engineers. 2006 reported average engineering starting salaries: $45-55K
from College Journal / Engineering
That's for people that just got their BS. Add 1-2 years experience, add $2-3K, or about equal to someone fresh out with an MS. MS with 1-2 years exp, add on another $2 (total, +$4-5K). Finally PhD and 1-2 yrs experience, add $10k (total +$14-15K).
My guess is NASA will be able to pick from a pool of MS/PhD candidates, each with many years of technical experience. But relative to what they'd be making in other employment, they'd be taking a paycut. Could be a significant paycut. Get a guy with a wife and a few kids, a mortgage, car payments, and looming kids' college bills, and any paycut can hurt quite a bit.
Then again, if you're going to go through a mid-life crisis, this is a hell of an option. Most people settle for the $50K sports car, but a paycut to go into space would count, too.
close very close.p>
the limiting factor for silicon is loss of junction action as more carriers get thermally promoted to the conduction band. typically right around 200C, the intrinsic carrier concentration overtakes the typical doped carrier density. But, you start getting increased leakage currents and higher current gain well below that. Depending on the type of transistors used, latch-up failure becomes more likely. the prime factor affecting what temp things start going bad is the amount of doping used. Increase doping, you can run a little hotter. But, it decreases the voltage breakdown limit, forcing you to de-rate.
yields are the big limiter right now. MOS devices are attacking oxide trap problems similar to silicon research in the 70's. manufacturers are demoing 1200V, 50A JFET's. they're getting there, but the device technology is a lot further along than other power device options (GaN, diamond, etc.) But those others are getting a lot of attention as well, especially in Japan.
not sure where they pulled that number from. 350F is ~175C. That's about right. harsh applications, certain silicon devices can go a bit above 150. maybe with heavy doping a bit higher. my guess is they meant F on the site.
wow. a 3 inch silicon carbide wafer costs a couple grand. a whole ship? maybe a small sloop or something. something like the titanic would require the GDP of a small country. or not so small.
but yes. it's expensive. until we figure out the processing control, costs will stay high too. still too expensive for most commercial app's, but once that changes SiC will replace a lot of Si power electronic devices.
hmmm.. too bad the PhD's never thought of that problem. oh, wait. they did. that's why they're using Silicon Carbide.
Thermal noise is typically related to the random promotion of carriers from the valence band to the conduction band, which gets worse at higher temperatures because the electrons get more energy. THE primary electronic difference between Silicon Carbide and Silicon is that SiC is a wide-bandgap material. It takes a lot more energy to promote carriers than in silicon. Also, SiC has ~0 thermally promoted carriers at room temperature, vs 10^10 for silicon. So, as temperature increases, it takes a while to get to the point that thermal noise becomes as much of a problem as it would be for silicon. Yes, if I needed a super precise amp, that could still be too much noise for some applications. Of course, if you needed one, not sure what better amp you're going to get to operate in a 600C environment.
no interconnects. this is not a processor. it's a chip. likely a single transistor the size of the silicon in your CPU. wirebonding typically platinum for very high temp packages. al-al or au-au if not quite as severe. other options available below that.
congratulations. you have no idea what you're talking about.
high temperature boards are ceramic (AlN, Al2O3, HTCC, DBC, etc.) seeing as how they're fired from 1-2000C, they'll be ok.
silver-glass die attaches are okay up to 400-450C. Beyond that, you have high-temp brazes, AuIn, AgAuGe, AgCu, oh and AuNi ok up to 950C.
Circuit!= computer. Chip != microprocessor. SiC chips = power electronics switch or sensor components. sure, you could build a processor out of these, but you could also just go back and build a Pentium out of vac.tubes.
It's a wide-bandgap semiconductor material that is being extensively developed for specific power or harsh environment applications. There currently are no MOS devices (used in your PC). Switching speeds typcially in the kilohertz range, for power conditioning. That chip is a single transistor, about the size of the piece of silicon in your PC. Finally, silicon's only okay to 150-200C. The article should have said 350F, not 350C.
that may be a good point, and it's a question I've always wondered.
The FF "memory leak" feature is defended as 'making full use of the system resources (which would otherwise be sitting idle) in order to provide a better, faster experience'
This sounds perfectly valid IFF: when anything else needs that memory, it can get it with negligible delay or mishap.
If you're saying the same for Opera, I'd be interested to know how valid that assumption is for either browser, or memory management and handling in general. From my experience windows tends to barf once memory utilization gets really high, but that's because most programs grab and keep their memory. How would a browser 'instantly give it back' when it's needed, without sending the system into a pagefile caching frenzy?
just curious, if you did try to scale up something like this (parallel distributed network, etc.) into the real Supercomputer flop range (what are they up to now, petaflops?) how would the costs scale?
by the way the protocol is designed, if you have an open router, that router is designed to provide access to anyone. Inherent to the protocol is a sign saying 'this is open, connect and use it.' i.e., the "Linksys" ssid issue. I've been in a neighborhood where I've seen 3 of them at the same time on different channels. 802.11 is a permission is inherently granted and must be specifically and actively denied to be considered otherwise. Implementing a technology with an inherent 'use me' sign, and not removing the sign, is an implicit agreement to provide access whether or not you understand what you are or aren't doing. If I bought a door to my house that came with a sign on it saying "come on in rest a while", and the sign was easily removable, I'd have a hard time making a trespassing case against someone. Even if it took a little technical knowledge to remove the sign, it's still there. Even if it wasn't obvious to me that the sign was there but everyone else could see it, the presence of the sign appears as an invitation to everyone else, and implies permission.
FOUO is specifically designated to NOT be used as a way of keeping Unclass info away from FOIA inquiries. It's for things that aren't government secrets, but shouldn't be shared with the general public. You would likely agree with many of these. Examples:
Privacy Information, Social security numbers, medical, etc.
Company Trade Secrets
Legal documents, law enforcement documents, with limits
And there are others, some discretionary. Full definition in Chapter 4 here (~100 page PDF): http://www.dtra.mil/documents/be/5400.7-R.pdf
BUT, from Chapter 4:
C4.1.1. General. Information that has not been given a security classification pursuant to the criteria of an Executive Order, but which may be withheld from the public because disclosure would cause a foreseeable harm to an interest protected by one or more FOIA Exemptions 2 through 9 (see Chapter C3.) shall be considered as being for official use only (FOUO). No other material shall be considered FOUO and FOUO is not authorized as an anemic form of classification to protect national security interests..
you have just described inforocket.com Archive.org mirror. (something from the 2000-2001 timeframe. The defunct site now redirects to an online psychic network. Get paid to answer burning questions... with a referral network! and your first $5 in questions is free! Dot-com at its best. Guess what happened? Lots of people signing up, blowing through $5 worth of questions, and done. Lots of people signing up, answering lots of those people's $5 worth of questions, then cashing out. Once real cash is involved, it's all about gaming the system. I refer to cashwars.com (Archive.org mirror) and prizegames.com (Archive.org mirror) as further evidence. Yes, I speak from experiece. I pulled a total of about.... $275 from those guys. $350 if you include Archive.org mirror)>funbets.com.
And this is currently how it should be. If the spammer had to pick up a portion of your ISP costs based on the amount of spam they were sending, that would be fine. They're incurred business costs would be payed for out of their own pockets instead of being passed on to unwilling consumers, and the consumers would get a net benefit for the inconvenience. Unfortunately, the current situation enables the spammer to shift the majority of the costs of his advertising onto the consumer. If done 'illegitimately', nearly all costs are passed onto a 3rd party. Supporting the spam industry is now a significant portion of your ISP bill (ISP's should list this line item separately on the bill), it provides you no net benefit, and you have no way to remove yourself from the 'service'.
ahh, the good old days. I remember CuteHTML being one of my favorites (SciTE supposedly not bad either). Haven't done any significant web design in about 7-9 years though. (CSS? what's that?:) ) Anyway, while you obviously CAN do it all it directly in text, with all the wackiness on the web these days, is it still straightforward and productive to do so?
Slashdot Mirror
The broken window fallacy does refer to a broken window, but the analogy is not limited to destruction. It can be expanded to any expenditure you were forced to make that shouldn't have been necessary. The 'forcing the expenditure' is a negative, and can't be re-cast as a positive because of the economic flow that resulted from it. While property destruction and loss of capital exacerbates the situation, lack of literal destruction does not invalidate the analogy.
What would invalidate the analogy is if the forced expenditure is filling a need that would have existed without that particular forcing function. The previous poster's statement was that spending all this money was not the answer, not being there was. Part of your response was that 'yeah, but look at all the good stuff coming out of these expenditures that go into our economy'. Using that statement as a rebuttal to the statement that the money shouldn't have been needed to be spent is very close to the broken window fallacy, and I still stand by it as a valid logical counter to that part of your argument.
That said, don't make assumptions about my bias. If it wasn't for said defense expenditures, I'd be out of a job. Us "engineers with advanced understandings of calculus, physics, and chemistry" know perfectly well that weapons ain't easy to design. We also tend to have decent understandings of logical progressions of thought, and would rather if you didn't use fallacious arguments to defend your points when legitimate ones are readily available.
Broken window fallacy. see here:
http://en.wikipedia.org/wiki/Parable_of_the_broken_window
Economically always better for the money to be spent in the first place, rather than having to spend it on something you didn't want.
ack... have to fight the lameness filter. that was yelling, apparently.
I think Memron was a much better documentary. http://www.imdb.com/title/tt0392439/
Scientists and Engineers. 2006 reported average engineering starting salaries: $45-55K
from College Journal / Engineering That's for people that just got their BS. Add 1-2 years experience, add $2-3K, or about equal to someone fresh out with an MS. MS with 1-2 years exp, add on another $2 (total, +$4-5K). Finally PhD and 1-2 yrs experience, add $10k (total +$14-15K).
My guess is NASA will be able to pick from a pool of MS/PhD candidates, each with many years of technical experience. But relative to what they'd be making in other employment, they'd be taking a paycut. Could be a significant paycut. Get a guy with a wife and a few kids, a mortgage, car payments, and looming kids' college bills, and any paycut can hurt quite a bit.
Then again, if you're going to go through a mid-life crisis, this is a hell of an option. Most people settle for the $50K sports car, but a paycut to go into space would count, too.
yields are the big limiter right now. MOS devices are attacking oxide trap problems similar to silicon research in the 70's. manufacturers are demoing 1200V, 50A JFET's. they're getting there, but the device technology is a lot further along than other power device options (GaN, diamond, etc.) But those others are getting a lot of attention as well, especially in Japan.
there are many packaging techniques available to take advantage of high SiC temperatures. None cheap and commonly used, since Silicon can't make use of them, but ceramic boards, non-alloy wirebonds, eutectic die attaches, high temp brazes. they all exist, and have been used, but they're expensive. not-so-aggressive options can get you to 250 without too much trouble. e.g. http://www.honeywell.com/sites/portal?smap=aerospace&page=High-Temp-Electonics3&theme=T5&catID=C82A27CF1-C0F1-76E9-6B52-2C477FB52FF7&id=H5E761CAC-F16E-40AF-B54E-3DFBA7F0A988&sel=1
no. see comments above. specifically http://science.slashdot.org/comments.pl?sid=294871&cid=20567843
not sure where they pulled that number from. 350F is ~175C. That's about right. harsh applications, certain silicon devices can go a bit above 150. maybe with heavy doping a bit higher. my guess is they meant F on the site.
but yes. it's expensive. until we figure out the processing control, costs will stay high too. still too expensive for most commercial app's, but once that changes SiC will replace a lot of Si power electronic devices.
Thermal noise is typically related to the random promotion of carriers from the valence band to the conduction band, which gets worse at higher temperatures because the electrons get more energy. THE primary electronic difference between Silicon Carbide and Silicon is that SiC is a wide-bandgap material. It takes a lot more energy to promote carriers than in silicon. Also, SiC has ~0 thermally promoted carriers at room temperature, vs 10^10 for silicon. So, as temperature increases, it takes a while to get to the point that thermal noise becomes as much of a problem as it would be for silicon. Yes, if I needed a super precise amp, that could still be too much noise for some applications. Of course, if you needed one, not sure what better amp you're going to get to operate in a 600C environment.
not for a very long time. it's not a processor. it's likely a single transistor. see http://science.slashdot.org/comments.pl?sid=294871&cid=20567843
no interconnects. this is not a processor. it's a chip. likely a single transistor the size of the silicon in your CPU. wirebonding typically platinum for very high temp packages. al-al or au-au if not quite as severe. other options available below that.
high temperature boards are ceramic (AlN, Al2O3, HTCC, DBC, etc.) seeing as how they're fired from 1-2000C, they'll be ok.
silver-glass die attaches are okay up to 400-450C. Beyond that, you have high-temp brazes, AuIn, AgAuGe, AgCu, oh and AuNi ok up to 950C.
Circuit!= computer. Chip != microprocessor. SiC chips = power electronics switch or sensor components. sure, you could build a processor out of these, but you could also just go back and build a Pentium out of vac.tubes.
It's a wide-bandgap semiconductor material that is being extensively developed for specific power or harsh environment applications. There currently are no MOS devices (used in your PC). Switching speeds typcially in the kilohertz range, for power conditioning. That chip is a single transistor, about the size of the piece of silicon in your PC. Finally, silicon's only okay to 150-200C. The article should have said 350F, not 350C.
read and learn. http://en.wikipedia.org/wiki/Power_semiconductor_device
that may be a good point, and it's a question I've always wondered. The FF "memory leak" feature is defended as 'making full use of the system resources (which would otherwise be sitting idle) in order to provide a better, faster experience' This sounds perfectly valid IFF: when anything else needs that memory, it can get it with negligible delay or mishap. If you're saying the same for Opera, I'd be interested to know how valid that assumption is for either browser, or memory management and handling in general. From my experience windows tends to barf once memory utilization gets really high, but that's because most programs grab and keep their memory. How would a browser 'instantly give it back' when it's needed, without sending the system into a pagefile caching frenzy?
just curious, if you did try to scale up something like this (parallel distributed network, etc.) into the real Supercomputer flop range (what are they up to now, petaflops?) how would the costs scale?
by the way the protocol is designed, if you have an open router, that router is designed to provide access to anyone. Inherent to the protocol is a sign saying 'this is open, connect and use it.' i.e., the "Linksys" ssid issue. I've been in a neighborhood where I've seen 3 of them at the same time on different channels. 802.11 is a permission is inherently granted and must be specifically and actively denied to be considered otherwise. Implementing a technology with an inherent 'use me' sign, and not removing the sign, is an implicit agreement to provide access whether or not you understand what you are or aren't doing. If I bought a door to my house that came with a sign on it saying "come on in rest a while", and the sign was easily removable, I'd have a hard time making a trespassing case against someone. Even if it took a little technical knowledge to remove the sign, it's still there. Even if it wasn't obvious to me that the sign was there but everyone else could see it, the presence of the sign appears as an invitation to everyone else, and implies permission.
hey, we all know any exploit can be fixed in 10 f-ing days!!!
but only if you consume 50 times your body weight over a course of 2 weeks...
I believe the limit is 11 stories... but it's been a while since I read about that experiment in my high school physics book ;)
Privacy Information, Social security numbers, medical, etc.
Company Trade Secrets
Legal documents, law enforcement documents, with limits
And there are others, some discretionary. Full definition in Chapter 4 here (~100 page PDF):
http://www.dtra.mil/documents/be/5400.7-R.pdf BUT, from Chapter 4:
C4.1.1. General. Information that has not been given a security classification pursuant to the criteria of an Executive Order, but which may be withheld from the public because disclosure would cause a foreseeable harm to an interest protected by one or more FOIA Exemptions 2 through 9 (see Chapter C3.) shall be considered as being for official use only (FOUO). No other material shall be considered FOUO and FOUO is not authorized as an anemic form of classification to protect national security interests..
please, please never say that again.
you have just described inforocket.com Archive.org mirror. (something from the 2000-2001 timeframe. The defunct site now redirects to an online psychic network. Get paid to answer burning questions... with a referral network! and your first $5 in questions is free! Dot-com at its best. Guess what happened? Lots of people signing up, blowing through $5 worth of questions, and done. Lots of people signing up, answering lots of those people's $5 worth of questions, then cashing out. Once real cash is involved, it's all about gaming the system. I refer to cashwars.com (Archive.org mirror) and prizegames.com (Archive.org mirror) as further evidence. Yes, I speak from experiece. I pulled a total of about.... $275 from those guys. $350 if you include Archive.org mirror)>funbets.com.
And this is currently how it should be. If the spammer had to pick up a portion of your ISP costs based on the amount of spam they were sending, that would be fine. They're incurred business costs would be payed for out of their own pockets instead of being passed on to unwilling consumers, and the consumers would get a net benefit for the inconvenience. Unfortunately, the current situation enables the spammer to shift the majority of the costs of his advertising onto the consumer. If done 'illegitimately', nearly all costs are passed onto a 3rd party. Supporting the spam industry is now a significant portion of your ISP bill (ISP's should list this line item separately on the bill), it provides you no net benefit, and you have no way to remove yourself from the 'service'.
ahh, the good old days. I remember CuteHTML being one of my favorites (SciTE supposedly not bad either). Haven't done any significant web design in about 7-9 years though. (CSS? what's that? :) ) Anyway, while you obviously CAN do it all it directly in text, with all the wackiness on the web these days, is it still straightforward and productive to do so?
You still pay for it one way or the other. If it still costs you, it hasn't been solved.