I have always thought that Richard Feynman's demonstration of how the space shuttle's o-ring material behaved in cold conditions was beautiful. Beautiful in it's simplicity.
From this obituary:
Frustrated by witnesses' vague answers and by slow bureaucratic procedures, he conducted an impromptu experiment that proved a key point in the investigation: He dunked a piece of the rocket booster's O-ring material into a cup of ice water and quickly showed that it lost all resiliency at low temperatures.
Beautiful, because it took this theory that was being argued over, and cast it into complete, inarguable certainty that was easily understood by everyone.
the absolute worst that can happen is that some time and money will be spent in finding that the results are not sound
Not so. The worst that can happen is that a lot of time and money will be spent in finding that the results are not sound. Meanwhile people less familiar with the scientific process will see the special on Dateline saying the energy revolution is here, and believe it.
I don't feel qualified to say much about the merits of this particular experiment, but I did just get finished reading Park's Voodo Science (and so, you know, I'm an expert). In it he discusses many ways that the label "science" had been attached to products and experiments that end up tricking people. One example he uses in his book is the fear people have had that power lines cause cancer. After around 20 years of research, no statistically relevant connection has been found. Hundreds of millions of dollars and huge amounts of time and energy spent by people, activists in an uproar, etc.
Park might have a bit of a tendency to be condescending or quick to jump to conclusions, but he seems to have a lot of experience and maybe he sounds jaded for a reason. I'm convinced his basic point is sound: the scientific community and it's accepted methods are effective and exist for a reason. Bad things can happen when the accepted procedures of peer review aren't adhered to.
I did a report on benchmarking in an embedded systems class I took last year. The great thing about EEMBC is that they try to eliminate some of the problems that are typically associated with benchmarking. They have members ranging from chipmakers like NEC, Intel, and Hitachi, to software makers like Redhat and WindRiver. They have five categories of tests, corresponding to different processor applications (like industrial, network, office apps). They also have varying levels to which chip makers can tweak the code to run on their hardware, from basic out-of-the box scores that reflect a basic compiler's output, down to hand tuning the assembly language. This is great because it shows you how much improvement you can get if you're really willing to dig into the algorithms' implementation. EEMBC should be able to put an end to misleading benchmark scores.
I think we could easily see the kind of thing you envision happening, especially with the power-over-ethernet capabilities, which is what we use for our phones where I work.
There is one complication I see, using my place of employment as an example: Each cubicle has 6 network jacks in it. Each one is connected to a different cable run, back to a patch room. The whole idea here was that you had 6 different network drops. The standard configuration for these is one jack for the corporate network, one for the lab network, one for voice connections, and I belive the other three are currently unused for future expansion. Although this new 3com switch does have the extra two jacks, which could be used for this purpose. So I guess the real question is whether you need multiple network drops to the final destination, or multiple connections to the same network.
rice_burners_suck thinks microsoft-sucks and finds a way to bring it up. This is pure opinion with no fact, just the standard anti-MS accusations. There's nothing new or useful here. This is neither insightful nor on-topic.
Slashdot Mirror
Oh dude, I hate that game!
Can YOU handle a Big Gulp Gin and Tonic?!
Looks like they should have spent some time building a web monkey to go along with it.
The fall-over capabilities of the NonStop systems are unmatched.
Frustrated by witnesses' vague answers and by slow bureaucratic procedures, he conducted an impromptu experiment that proved a key point in the investigation: He dunked a piece of the rocket booster's O-ring material into a cup of ice water and quickly showed that it lost all resiliency at low temperatures.
Beautiful, because it took this theory that was being argued over, and cast it into complete, inarguable certainty that was easily understood by everyone.
I'm picturing a lowered Atlas V with tinted windows, a giant spoiler on the back, and a huge POWERED BY RUSSIA sticker across the windshield.
Not so. The worst that can happen is that a lot of time and money will be spent in finding that the results are not sound. Meanwhile people less familiar with the scientific process will see the special on Dateline saying the energy revolution is here, and believe it.
I don't feel qualified to say much about the merits of this particular experiment, but I did just get finished reading Park's Voodo Science (and so, you know, I'm an expert). In it he discusses many ways that the label "science" had been attached to products and experiments that end up tricking people. One example he uses in his book is the fear people have had that power lines cause cancer. After around 20 years of research, no statistically relevant connection has been found. Hundreds of millions of dollars and huge amounts of time and energy spent by people, activists in an uproar, etc.
Park might have a bit of a tendency to be condescending or quick to jump to conclusions, but he seems to have a lot of experience and maybe he sounds jaded for a reason. I'm convinced his basic point is sound: the scientific community and it's accepted methods are effective and exist for a reason. Bad things can happen when the accepted procedures of peer review aren't adhered to.
I did a report on benchmarking in an embedded systems class I took last year. The great thing about EEMBC is that they try to eliminate some of the problems that are typically associated with benchmarking. They have members ranging from chipmakers like NEC, Intel, and Hitachi, to software makers like Redhat and WindRiver. They have five categories of tests, corresponding to different processor applications (like industrial, network, office apps). They also have varying levels to which chip makers can tweak the code to run on their hardware, from basic out-of-the box scores that reflect a basic compiler's output, down to hand tuning the assembly language. This is great because it shows you how much improvement you can get if you're really willing to dig into the algorithms' implementation. EEMBC should be able to put an end to misleading benchmark scores.
I think we could easily see the kind of thing you envision happening, especially with the power-over-ethernet capabilities, which is what we use for our phones where I work. There is one complication I see, using my place of employment as an example: Each cubicle has 6 network jacks in it. Each one is connected to a different cable run, back to a patch room. The whole idea here was that you had 6 different network drops. The standard configuration for these is one jack for the corporate network, one for the lab network, one for voice connections, and I belive the other three are currently unused for future expansion. Although this new 3com switch does have the extra two jacks, which could be used for this purpose. So I guess the real question is whether you need multiple network drops to the final destination, or multiple connections to the same network.
rice_burners_suck thinks microsoft-sucks and finds a way to bring it up. This is pure opinion with no fact, just the standard anti-MS accusations. There's nothing new or useful here. This is neither insightful nor on-topic.
So are you suggesting we'll see 4-dimensional games soon?
Which will from now on be known as Smarketing.
Can orbital microbrew be far behind? If one millilitre isn't a microbrew, I don't know what is.