Ah, thank you! I stand corrected. It's not my field, but I've spent a couple hours reading those and other related papers, it's quite interesting stuff!
What happens if you don't publish under an institutional affiliation? Do the publishers enforce IRB participation? How? What sort of oversight they have for international authors? After all, everyone and their dog can incorporate a "research institute", probably even in Vladivostok if one wants to.
This is about as good as advice from Dr. Bob. Stop it. Please. The "bag" will do nothing for fiberglass fibers. What it will do, though, is remove larger, easily visible particles, and give you a false sense of safety. The "vacuum bag" on circular power saws is designed for benign materials like wood and certain kinds of plywood (stuff that's not resin impregnated).
I somewhat agree. About the only truly insightful thing to ever come directly out of the Shuttle program was Feynman's analysis of how NASA's modus operandi is seriously broken. And it was promptly forgotten, so Columbia happened. And I'm sure they'll forget it again in whatever the next big program is going to be.
Of course Shuttle has contributed to many useful missions -- it's hard to tell, for example, what would have happened to Hubble mission without Shuttle. The reasonable thing to do would have been to launch another one with a fixed mirror, since a Shuttle launch costs way more, but people don't always to the economically sensible thing (sigh).
As far as I'm concerned, recent CPUs and GPUs are no less of a technological achievment than, say, a Shuttle launch.
Not exactly the same kind of risk involved or the same type of experience -- one is a spectacle and the other is not:
To watch a launch of a vehicle breaking the bonds of earth's gravity to venture out into an inhospitable environment where those on board risk their lives is on a somewhat different scale than, say, pushing the power button on my desktop.
You fuck up a run of a few million CPUs like Penryn, you're out of a billion bucks. That's in the same ballpark as losing a Shuttle. Certainly way more than monetary value of any human life that's lost in a Shuttle accident.
As far as inhospitable environments go: you should visit a sub-50nm fab once, and see the whole manufacturing process start to finish. Kinda makes Shuttle look dumb -- IMHO.
I don't get the whole romanticization of risk and frontier. Watching, say, an electron force microscope work gives me way more chills than any chemical rocket launch ever could. Small things are no less complex than big things. In fact, they can be way more complex, it's just hard to see with your naked eye. The exponents involved in the physical dimensions are irrelevant. Yeah, a Shuttle launch is as physically big, noisy thing, with plenty of audiovisuals to go with it. Knowing me, I'd probably fall asleep at T-20, just like I often do during 4th of July firework celebrations. The boom gets old real quick, you've got only one pair of ears, better keep good care of them.
Things fail spectacularly in the silicon world, too.
Really? The same way they can in Space Exploration? Really?
Just the fact that when you blow up a few billion 45nm transistors it doesn't cause billions worth of damage doesn't mean that it's any less spectacular. Semiconductor failure modes are quite fascinating things, and the accumulated knowledge in this area is well on par with accumulated knowledge about aerospace snafus. I'd tend to think that semiconductor knowledge has been surpassing aerospace knowledge simply due to rather fast progress in the former. About the only place where you'd use the exact same CPU for 3 decades is a nuclear plant control room, and that's just because certifying new stuff costs money that utilities are not willing to spend. No major performance indicators in the aerospace world have improved anywhere near the major indicators in the semiconductor world. None. Specific impulse, fuel-to-dry-weight ratio, etc., are all well in the same order of magnitude as they were 40 years ago. About the only major thing that has happened was SpaceX's pioneering vertical integration and resulting cost and time-to-market savings that are pretty much unheard of in the contemporary trenches. The last time rockets were designed so quickly and at such a relatively low cost was in Nazi Germany...
The shuttle launch is something that is likely never going to happen again, and those who have not had the opportunity should be jealous.
All of the shit that happens around us is unique and is never going to happen again. IOW: not much of an argument. It's all a matter of what one values in life. It's important to you: fine. Important to my Dad, who saw a Shuttle launch in the 80s. Not all that important to me -- certainly less important than, say, working on my house.
As far as I'm concerned, recent CPUs and GPUs are no less of a technological achievment than, say, a Shuttle launch. They are all immensely complex technical systems, even if the Shuttle is "just" a spaceplane strapped to a rocket, and, say Penryn is "just" a CPU on a piece of silicon wafer. Whether the parts are mechanical or not doesn't matter much, IMHO. Things fail spectacularly in the silicon world, too.
Doing "Something", to me, definitely wouldn't be watching a Shuttle launch.
Japanese is quite a different language (as far as I can tell, I don't speak it)
Anecdote: At least when it comes to writing Hiragana phonetic characters, they seem not to be hard at all to copy. I don't remember all of them yet, but I can write them all reasonably well while looking, say, at a children's book. Took a few weeks of practice. Kanji is a whole different story, and of course the characters are originally Chinese, but the way they are used is apparently vastly different (and much simpler).
I have a Chinese friend. She can't really read Japanse in Kanji. I asked her a few times to do a best effort, pretending it was some oulandishly bad Chinese shoolkids' writing. It came out hilarious and made no sense at all. So yes, Japanese seems to be very, very different from Chinese, even if ideographic writing uses characters of same origin...
Of course Apple products are not U.S. made, yet they are designed here, and all I know is I have seen plenty enough of iDevices riding on Tokyo's metro. Heck, it seems that Japanese (of all people!) crave U.S. designs -- they'd love to drive american cars, for example! Must be marketing or something, since U.S. cars are supposedly crap?
You may find interesting my anecdote then: in most "budget" stores around here (Walmart, Dollar General, Dollar Tree), we started running into plenty of seemingly "cheap and easy" products that were made in U.S.A. Stuff that "screams" that it should be made in a sweatshop somewhere since the margins are so thin, like simple molded or extruded plastic household goods, is often made in the U.S. -- I even bothered to verify that some of those manufacturers existed and were, in fact, registered as a proper kind of a business (manufacturer, not importer).
So tell us, is this true (as per one of the videos), that in China you'll get approved for 10:1 leverage (property value vs income), compared to "typical" U.S. 3:1? Is that "saner", then? Wasn't overleveraging one of the reasons the U.S. housing bubble burst?
As for the high speed rail: it seems like a yet another argument that China is spending for sake of increasing GDP and nothing else. For high speed rail to make any sense at Europe-to-China distances, it'd have to go about as fast as planes fly. I fail to see how it could be any cheaper to operate such a system on the ground than in the air. Even at Shinkansen speeds, it'd make no sense to travel, say, from Beijing to Berlin. That's 7500 km -- at 300 km/h (top Shinkansen speed, thus unrealistic), it'd take 25 hours. Now compare this to a direct flight taking about 9.5 hours on average. Even with connecting travel the factor doesn't improve.
No, it's almost impossible. Not with everyone and their dog getting spambot malware on their systems. It'd be very, very unlikely that one's family and friends are all extremely tech-literate, to the extent that they get no malware at all, over a course of more than a decade. I find it pretty much an unbelievable scenario.
What is it with people who don't know or don't care how to set up their email clients/readers so that stuff gets automatically filtered? I've seen plenty of subscribers complain of excessive traffic on mailing list "X", and I always think: WTF? This stuff should go to its own folder automatically, and get organized by threads, too. I mean this functionality has been available for more than a decade... I use email as my own private archive of many mailing lists, it's very convenient. It is all locally indexed, too!
#1 is fine and dandy until someone in your family gets infected by malware, then it'll show up on dozens of email lists within days. Or until data gets siphoned out from your financial institution. BTDT on both counts, worse -- multiple times in both scenarios.
So educate me. A U.S. President has nothing much to do with economics of U.S.A., or the world at large, for that matter. Their net worth is usually very low, and the influence they have in the matters of economical "ecosystem" is I'd say apocryphal at best. It's demonstrated over and over when they can't get shit done.
Their core competency is to get shit done, even when it means reengineering everything from scratch. Not much different than what was called for in JWST. Of course JWST called for stuff that was never done before, but hey, SpaceX's feat was pretty much never done before either: a completely integrated launch vehicle supplier.
Hmm, I don't know, I'm working with an electrical engineer "close to retirement" with plenty of previous big project / corporate experience, and the guy is a joy to work with and there's a shitload of stuff I'm learning from him. Of course it's just an anecdote. Even the "bureaucracy" aspects of his experience are useful. We seriously needed some insight into product lifecycle management, and the guy just came knowing all this, knowing what worked well and what was a pain in the butt, etc., all from front-line experience in a corporate environment. YMMV...
I have to agree. Let's see at how you could spend $200k per year (for 20 kids), in a low cost of living state like Ohio:
- a Ph.D. to teach the kids: $90k - classroom rental in an office complex somewhere: $25k - transportation contract with a small business: $25k - classroom supplies: $40k - roadtrips: $10k - surplus/cushion: $10k
I'm assuming that healthcare can be taken care of by the parents. Heck, they have to have a pediatrician and yearly checkups anyway, I don't think a "school nurse" is really necessary. Have the teacher go through a CPR and first aid training and it should be fine.
Those are all very realistic numbers, and I can't but see an amazing experience for the kids that could go through such a program, and for the teacher, too. Yes, you could easily get a group of very, very good kids after 12 years of this.
I'm wondering how much would it cost if SpaceX's engineering and management could be used to run it. They'd probably pull it off for $750M or somesuch, and end up launching it on their own freaking rocket to boot. Heck, they probably could launch a technology testbed/prototype (with a smaller mirror) to check things out, too.
Slashdot Mirror
Ah, thank you! I stand corrected. It's not my field, but I've spent a couple hours reading those and other related papers, it's quite interesting stuff!
How is it abuse when the data is supposedly collected in an anonymizing fashion?
What happens if you don't publish under an institutional affiliation? Do the publishers enforce IRB participation? How? What sort of oversight they have for international authors? After all, everyone and their dog can incorporate a "research institute", probably even in Vladivostok if one wants to.
This is about as good as advice from Dr. Bob. Stop it. Please. The "bag" will do nothing for fiberglass fibers. What it will do, though, is remove larger, easily visible particles, and give you a false sense of safety. The "vacuum bag" on circular power saws is designed for benign materials like wood and certain kinds of plywood (stuff that's not resin impregnated).
Have you been meaning to say that you can read Chinese well, though?
I somewhat agree. About the only truly insightful thing to ever come directly out of the Shuttle program was Feynman's analysis of how NASA's modus operandi is seriously broken. And it was promptly forgotten, so Columbia happened. And I'm sure they'll forget it again in whatever the next big program is going to be.
Of course Shuttle has contributed to many useful missions -- it's hard to tell, for example, what would have happened to Hubble mission without Shuttle. The reasonable thing to do would have been to launch another one with a fixed mirror, since a Shuttle launch costs way more, but people don't always to the economically sensible thing (sigh).
As far as I'm concerned, recent CPUs and GPUs are no less of a technological achievment than, say, a Shuttle launch.
Not exactly the same kind of risk involved or the same type of experience -- one is a spectacle and the other is not:
To watch a launch of a vehicle breaking the bonds of earth's gravity to venture out into an inhospitable environment where those on board risk their lives is on a somewhat different scale than, say, pushing the power button on my desktop.
You fuck up a run of a few million CPUs like Penryn, you're out of a billion bucks. That's in the same ballpark as losing a Shuttle. Certainly way more than monetary value of any human life that's lost in a Shuttle accident.
As far as inhospitable environments go: you should visit a sub-50nm fab once, and see the whole manufacturing process start to finish. Kinda makes Shuttle look dumb -- IMHO.
I don't get the whole romanticization of risk and frontier. Watching, say, an electron force microscope work gives me way more chills than any chemical rocket launch ever could. Small things are no less complex than big things. In fact, they can be way more complex, it's just hard to see with your naked eye. The exponents involved in the physical dimensions are irrelevant. Yeah, a Shuttle launch is as physically big, noisy thing, with plenty of audiovisuals to go with it. Knowing me, I'd probably fall asleep at T-20, just like I often do during 4th of July firework celebrations. The boom gets old real quick, you've got only one pair of ears, better keep good care of them.
Things fail spectacularly in the silicon world, too.
Really? The same way they can in Space Exploration? Really?
Just the fact that when you blow up a few billion 45nm transistors it doesn't cause billions worth of damage doesn't mean that it's any less spectacular. Semiconductor failure modes are quite fascinating things, and the accumulated knowledge in this area is well on par with accumulated knowledge about aerospace snafus. I'd tend to think that semiconductor knowledge has been surpassing aerospace knowledge simply due to rather fast progress in the former. About the only place where you'd use the exact same CPU for 3 decades is a nuclear plant control room, and that's just because certifying new stuff costs money that utilities are not willing to spend. No major performance indicators in the aerospace world have improved anywhere near the major indicators in the semiconductor world. None. Specific impulse, fuel-to-dry-weight ratio, etc., are all well in the same order of magnitude as they were 40 years ago. About the only major thing that has happened was SpaceX's pioneering vertical integration and resulting cost and time-to-market savings that are pretty much unheard of in the contemporary trenches. The last time rockets were designed so quickly and at such a relatively low cost was in Nazi Germany...
High speed rail for freight? Hmm. That would sort of make sense, I didn't think about it.
I agree about incorrectly using the term leverage, I meant debt-to-income ratio. So, what's the debt-to-income ratio on those loans in China? 10:1?
I think that synthetic kerosene is still cheaper over a span of a couple of decades than move to fast rail.
The shuttle launch is something that is likely never going to happen again, and those who have not had the opportunity should be jealous.
All of the shit that happens around us is unique and is never going to happen again. IOW: not much of an argument. It's all a matter of what one values in life. It's important to you: fine. Important to my Dad, who saw a Shuttle launch in the 80s. Not all that important to me -- certainly less important than, say, working on my house.
As far as I'm concerned, recent CPUs and GPUs are no less of a technological achievment than, say, a Shuttle launch. They are all immensely complex technical systems, even if the Shuttle is "just" a spaceplane strapped to a rocket, and, say Penryn is "just" a CPU on a piece of silicon wafer. Whether the parts are mechanical or not doesn't matter much, IMHO. Things fail spectacularly in the silicon world, too.
Doing "Something", to me, definitely wouldn't be watching a Shuttle launch.
Japanese is quite a different language (as far as I can tell, I don't speak it)
Anecdote: At least when it comes to writing Hiragana phonetic characters, they seem not to be hard at all to copy. I don't remember all of them yet, but I can write them all reasonably well while looking, say, at a children's book. Took a few weeks of practice. Kanji is a whole different story, and of course the characters are originally Chinese, but the way they are used is apparently vastly different (and much simpler).
I have a Chinese friend. She can't really read Japanse in Kanji. I asked her a few times to do a best effort, pretending it was some oulandishly bad Chinese shoolkids' writing. It came out hilarious and made no sense at all. So yes, Japanese seems to be very, very different from Chinese, even if ideographic writing uses characters of same origin...
Umm, a State cannot simply override Federal immigration and labor laws, can it now?
Of course Apple products are not U.S. made, yet they are designed here, and all I know is I have seen plenty enough of iDevices riding on Tokyo's metro. Heck, it seems that Japanese (of all people!) crave U.S. designs -- they'd love to drive american cars, for example! Must be marketing or something, since U.S. cars are supposedly crap?
You may find interesting my anecdote then: in most "budget" stores around here (Walmart, Dollar General, Dollar Tree), we started running into plenty of seemingly "cheap and easy" products that were made in U.S.A. Stuff that "screams" that it should be made in a sweatshop somewhere since the margins are so thin, like simple molded or extruded plastic household goods, is often made in the U.S. -- I even bothered to verify that some of those manufacturers existed and were, in fact, registered as a proper kind of a business (manufacturer, not importer).
So tell us, is this true (as per one of the videos), that in China you'll get approved for 10:1 leverage (property value vs income), compared to "typical" U.S. 3:1? Is that "saner", then? Wasn't overleveraging one of the reasons the U.S. housing bubble burst?
As for the high speed rail: it seems like a yet another argument that China is spending for sake of increasing GDP and nothing else. For high speed rail to make any sense at Europe-to-China distances, it'd have to go about as fast as planes fly. I fail to see how it could be any cheaper to operate such a system on the ground than in the air. Even at Shinkansen speeds, it'd make no sense to travel, say, from Beijing to Berlin. That's 7500 km -- at 300 km/h (top Shinkansen speed, thus unrealistic), it'd take 25 hours. Now compare this to a direct flight taking about 9.5 hours on average. Even with connecting travel the factor doesn't improve.
Google lets you use gmail on your own domain. It used to be free and still is for grandfathered domains, I don't know how it works for new domains.
No, it's almost impossible. Not with everyone and their dog getting spambot malware on their systems. It'd be very, very unlikely that one's family and friends are all extremely tech-literate, to the extent that they get no malware at all, over a course of more than a decade. I find it pretty much an unbelievable scenario.
What is it with people who don't know or don't care how to set up their email clients/readers so that stuff gets automatically filtered? I've seen plenty of subscribers complain of excessive traffic on mailing list "X", and I always think: WTF? This stuff should go to its own folder automatically, and get organized by threads, too. I mean this functionality has been available for more than a decade... I use email as my own private archive of many mailing lists, it's very convenient. It is all locally indexed, too!
#1 is fine and dandy until someone in your family gets infected by malware, then it'll show up on dozens of email lists within days. Or until data gets siphoned out from your financial institution. BTDT on both counts, worse -- multiple times in both scenarios.
Well, thank you very much for contributing your bit, too ;)
So educate me. A U.S. President has nothing much to do with economics of U.S.A., or the world at large, for that matter. Their net worth is usually very low, and the influence they have in the matters of economical "ecosystem" is I'd say apocryphal at best. It's demonstrated over and over when they can't get shit done.
Their core competency is to get shit done, even when it means reengineering everything from scratch. Not much different than what was called for in JWST. Of course JWST called for stuff that was never done before, but hey, SpaceX's feat was pretty much never done before either: a completely integrated launch vehicle supplier.
Hmm, I don't know, I'm working with an electrical engineer "close to retirement" with plenty of previous big project / corporate experience, and the guy is a joy to work with and there's a shitload of stuff I'm learning from him. Of course it's just an anecdote. Even the "bureaucracy" aspects of his experience are useful. We seriously needed some insight into product lifecycle management, and the guy just came knowing all this, knowing what worked well and what was a pain in the butt, etc., all from front-line experience in a corporate environment. YMMV...
I have to agree. Let's see at how you could spend $200k per year (for 20 kids), in a low cost of living state like Ohio:
- a Ph.D. to teach the kids: $90k
- classroom rental in an office complex somewhere: $25k
- transportation contract with a small business: $25k
- classroom supplies: $40k
- roadtrips: $10k
- surplus/cushion: $10k
I'm assuming that healthcare can be taken care of by the parents. Heck, they have to have a pediatrician and yearly
checkups anyway, I don't think a "school nurse" is really necessary. Have the teacher go through a CPR and first
aid training and it should be fine.
Those are all very realistic numbers, and I can't but see an amazing experience for the kids that could go through
such a program, and for the teacher, too. Yes, you could easily get a group of very, very good kids after 12 years
of this.
I'm wondering how much would it cost if SpaceX's engineering and management could be used to run it. They'd probably pull it off for $750M or somesuch, and end up launching it on their own freaking rocket to boot. Heck, they probably could launch a technology testbed/prototype (with a smaller mirror) to check things out, too.