Because if they didn't, what profit motive would Microsoft have to support this conspiracy?
I'm not really sure where you're getting 'conspiracy' from here.
The point being is that we simply wouldn't have bought new machines (in the case of non x86-64 machines) or updated the OS (in the ones that were). As an aside XP64 was not (according to our MS rep) licensed for us (so still an OS upgrade) and lacked some hardware support.
It means "practically speaking, if you need more than 4GB of RAM get a 64 bit cpu/OS and you'll be much better off."
However only if that's the general use case. Otherwise the term 'practical' doesn't seem to apply.
Yes absolutely. But so what?
(I think I accidentally chopped some of my response - bad tag maybe!) Pretty much what I say above. That you're assuming that > 4GB in a single application is the general use case of having from 4 to 8 GB. For the low end I rather suspect that it's not.
but I don't think that it can't be objectively measured.
Didn't say it couldn't however I said it's difficult. By which I mean - difficult to do usefully.
The first and most key aspect of an intuitive interface is how readily it is grasped by the user. A chair, for example, is an incredibly intuitive interface. The form of an intuitive interface matches the function it exports. The degree to which a user can be introduced to an interface concept and understand it is something that can be measured.
That's one way of defining it. Another is that the UI of a chair is simply consistent to your pre-established actions for sitting. Someone who was trained to sit differently (say only zazen) might find a chair completely incomprehensible.
To me using the term 'intuitive' seems to appeal to some intrinsic value not some learned behavior.
The utilization of focused user trials can help discover the problem areas of an interface and that information can be used to improve intuitiveness.
Possibly, but how much research is there on the utility of 'focus user trials' in UI design;-). That is to say how much involved say a large SRS of field users.
The first is the degree to which the interface conforms to the user's preconceived concept of the function.
Yes, yes. D. A. Norman referred to this as the users 'mental model' of the device. I think you're oversimplifying considerably though. Sure most objects we encounter require some kind of manipulation but 'direct'? Again there seem to be pretty heavy assumptions behind that term. Sure I can watch my daughter interact with a ball and watch it move in response to her actions. However before she can walk she is operating simple machines where the mechanism is hidden. Norman put it better (and way less pretentiously than you) by simply stating that the user has *some kind of idea* how the machine works and the closer the operation mimics the users mental model the better it is designed...
...and that's just one term 'direct'. IMHO each of your other statements is replete with terms which contain similar "ten ton assumptions" for which there are easy common counter-examples. e.g. I've certainly witnessed scores of older people be unable to operate touchscreens because their mental model doesn't include having poor calibration or capacitive sensing. Is this the general case? Don't know - but it's enough to call your statement into question.
Herein lies the problem. I don't doubt that you can objectively test some things related to UI design - there may even be some value in doing so. However it seems pretty difficult to determine what exactly you are testing. Especially if you're going to leap to stamping any of these results 'intuitive'. Which is better off in the marketing literature.
I'd agree that there's a lot of contextual information when interpreting sentences and "there is a 4GB architectural limit" is no exception but what does "for nearly all practical purposes" mean then? I mean if we really are "in the era of 64-bit budget computers" doesn't that imply that people who are using more memory than 4GB are likely at the lower end (i.e. >4GB
To me anyway MS can license their OS any way they want (memory, CPUs, number of keys on your keyboard) however we probably would have held off on upgrading a number of machines if we simply could have upgraded their memory.
Not to whine but without LAN what are you doing about firewall traversal? Plenty of wireless, LAN and mobile networks block assloads of ports and perform overzealous packet inspection. I work for an educational institution and we do this to differing degrees. One place where we do it the most is on the wireless network. So since in this situation how easy is it going to be to set up a game using only battle.net.
Give me five examples of a workflow that is easier on the Iphone's UI, compared with all other phones on the market? No wait - give me one?
I'm not really part of the "It's really natural" camp. In fact I don't really believe in intrinsically intuitive interfaces at all - just consistent ones. Anyway the term "workflow" is kind of broad but the most useful thing I find about the iPhone UI toolkit is the "pinch" gesture. For magnifying part of the screen. Compared to all the other web-enabled portable devices I've used previous (several WinCE PDA/Smartphones, Blackberrys all the way back to the original). I find the "pinch" to be very easy to focus on a particular portion of a document. The next one would be they way it auto-magnifies form fields. In the interest of balance - the things I could go either way on: "thumb flick" for scrolling seems equivalent to me to a jog dial. Auto-landscape isn't really any better than a button that does the same thing (sometimes it saves me time, others its annoying and there are clear compromises in the auto functionality which I suspect are the drivers behind it's modality). Things I hate: Copy/Paste is annoying, Applications don't always have an obvious way to ascend and descend their hierarchies. "Shake to shuffle" is cute but having it on by default is lame and I don't see how anyone can use this feature and travel.
it's fundamentally unanswerable, you just have to accept it as true.
I'm sure there's a big slice of dogma amongst those folks. That said, I'd almost make that statement about the vast majority of UI claims. Why? It's simply difficult to test. I mean ask yourself what unit is "intuitiveness" expressed in? Time to reach competency? What is competency? Can competency be defined in an objective way? (i.e. Say we were looking at "typing" - during my highschool education 40WPM was considered competency. This figure was chosen - according to my teacher - because that's what she understood to be the threshold businesses wanted. Which is fine but is there any similar "objective" for the iPhone - other than entering the SMS olympics;-)
No argument that the 360 was more poorly designed than the PS3. When I look at things like the MS Surface (and I do mean "look" we have one on site). I get the feeling that MS's internal opinion of their hardware development skills is rather incongruous with that of most other engineers.
Linux isn't written by lonely nerds hiding out in their parents' basements. It's written by people working for major companies...who just happen to also be nerds and like to be close to mom.
Uh sure...but not by much. ~10Cals/lb. So negligible that the most widely used BMR formulae don't even take it into account. It's one of a big list of facts about weight management where the spoken frequency is inversely proportional to it's utility.
That said most forced exercise people do (i.e. walking) doesn't scale well with weight.
So muscle or lack thereof doesn't appear to be the primary variable in why increased weight doesn't significantly limit weight gain. Rather that the extra exercise you get with a heavier body is insignificant compared to the caloric intake it took to get there.
Meh...so the way I'm going to interpret the defining silence here is that beguyId doesn't really know anything. He's like that other guy nine-times (who had hauntingly similar opinions - maybe ones a sockpuppet?). The kind of person who just drops off a study (not that I'm convinced they even READ the studies attached to their link).
The "calories are only one factor among many" idea is interesting from a social point of view. It gives the appearance (to me anyway) that the person is claiming that somehow caloric reduction is not the main thing people should be worrying about in terms of weight loss.
But it is...
Even after going through the only human study indirectly linked to by beguyId (and putting aside some of my caveats like having an n=12 and an R=0.5 and some ambiguity in terms of what kind of caloric restriction these people were actually on and how well controlled it was). It doesn't seem to say that at all. Even if we take all the claims there at face value (and apply them universally) it looks - at best - you could cut an extra 10% of body weight by restriction your fat intake when you are dieting (and only after you had already lost some weight).
Ok now back to what was actually being talked about. Someone who seems to claim that they should be obese (by their eating habits) but are not. They specifically mention that when they are allegedly overeating they are eating fatty foods but their weight doesn't move. This is exactly the opposite correlation that the study this moron linked to implies. If fat is not restricted then weight-loss is less efficient.
Next even if beguyid had a synapse or two firing so they could get the direction of the correlation correct. The disparity between the alleged weights is too big to account for even by the best case for the study. The person I was responding to was talking about closer to a 17% gap between consumption and weight.
Or you're insisting the research findings must be changed to fit your THEORY.
First of all, you need to differentiate between research and someone just posting "I counted calories". Even if the research completely explained and perfectly predicted this persons alleged results - which it doesn't appear to. Assuming all other sources of error are zero seems to be pretty bad reasoning. More the kind I would personally associate with those pandering "junk science".
While it is so nice and tidy to say it's as simple as calories, it just isn't true.
Actually I never said that. Which makes you kind of an idiot. Nor do you adequately explain what is meant by "as simple as".
As only one factor, among many, recent research has shown that the particular bacteria in a person's gut makes a large difference in how much food is actually absorbed.
I skimmed the papers. I admit I'm a little intrigued over your interpretation here. What is large? Large enough to cover this alleged 11500 calorie gap? Significantly overshadowing calories taken in?
It's not impossible. My daughter has coeliac disease, and i've been tested (blood test only) and come back negative but it's not impossible that I have something like that going on.
Impossible is not really the issue. If you are passing 11550 calories of food. That's anywhere from 1.4Kg to 2.8Kg of extra poop per day (and who knows how many cc's). You would be going to the bathroom so often people would think they are holding the Oscars there.
A lot of the time when I take my temp it is around 37.5-38
Congratulations you're normal. Point being if you are somehow not excreting this insane amount of caloric surplus. It has to be *doing* something, something that produces very little matter (otherwise you would gain weight) and it has to do so in a short period of time. The most efficient way it can do this is to emit heat.
Again, the smart money is on inaccurate measurement.
I don't think it's quite that simple for everyone. I've been through periods of eating really badly (high fat takeaway for lunch every day for weeks on end) and then really healthy, with identical exercise level (~none), and my weight never moved outside the 69-71kg range that i seem to have been stuck at for the last 5-10 years. I've added up the calories I intake vs the exercise I do (next to none) and by all calculations I should be a balloon. 69-71kg would be about right for my height if there was a bit more muscle on me.
If you are implying that you are of normal BMI (we'll say this is 25) but due to your diet you should be obese (BMI 90) Then you are almost assuredly mis-measuring.
Think about it, when you eat food the choices are either excrete it or metabolize it. Making some guesses as to your height (168 cm) you appear to be saying that according to your measurements you are eating enough calories to put you at the 85Kg mark. That's something like eating at a 115500 calorie surplus. Come on. Do you really think you are passing that much extra food? or if you're not passing it. How exactly is your body burning this without some other physiological symptom? (i.e. tachycardia, +2 degree c fever).
Interesting you should mention that. I just read some of Walter Mischel's work available on google scholars.(Advances in experimental social psychology, Volume 7) they did an experiment where they measured the mean time children would wait based but asking them to think (or "ideate") with one of three modalities. i) Fun - that is think of something fun to do, ii) The reward - think of the thing they are going to get and iii) No particular instruction. They did this test with both the reward visible and obscured.
The results were kind of interesting: With the rewards visible - thinking about nothing was the worst strategy (to improve waiting time) or par with thinking about the rewards with the reward not visible. Thinking about something fun was the best strategy in either case but when the rewards weren't visible - no instruction was almost as good.
So to a degree it seems that it can be taught however it doesn't answer if this 'learned' behavior has the same effect (improvements in test scores, etc..) and it is difficult to verify that the child was actually thinking what they were asked to think about.
I suppose it depends on what you mean by "lots of tiny chunks". Clearly doing a single "burst" transfer is better than lots of small ones but if you are still planning to process all these "chunks" of data at the same time then there's no reason why you couldn't just ship them all together and process them individually. Perhaps even from shared memory.
Unless of course we're taking about a bunch of chunks that are not going to be worked on simultaneously which goes back to my statement about the degree of parallelism that can be achieved being your chief worry.
Yes, this true for a lot of people who speak outside of their area of expertise. Some recognize it (Feynman comes to mind in "The Meaning of it All" where he IIRC says in effect that has he strays farther from talking about physics people should give less credit to what he's talking about) others don't.
I had an opportunity to hear Kurzweil speak and his reasoning (as given) was a little flawed. One he talked about how computing power increased over time in a predictable pattern doubling every 18-24 months (moore's law). I thought it was interesting that he didn't seem to discern between transistor capacity of chips (closer to what that law talks about) and computing performance. If computing performance was actually doubling each and every 18 months. Then next year I could replace my machine with one that runs an (algorithm) twice as fast for roughly the same price. However that generally doesn't happen and just by glancing at other benchmarks I suspect that increases in computing performance are largely domain specific. i.e. highly parallelizable tasks have recently enjoyed a better than average performance increase due to multicore architectures. Whereas non-parallizable tasks probably have crept along.
Perhaps this is just me but...probably what you had for lunch is pretty low on most peoples "care-dar". When I get together with my friends...know how often we talk about lunch...almost never. Know how may SMSs I've received about peoples lunch? or IMs or emails for that matter? Those figures hang pretty close to zero too. But Twitter? From my modest sampling of tweets it seems like it's pretty close to mandatory to shoutout about your ingestibles. I can think of some reasons for why this particular subject comes up but the real revelation for tweeters (or twits or whatever you call yourselves) should be that MOST OF YOU ARE REACHING PRETTY DAMN FAR TO COME UP WITH SOMETHING TO TALK ABOUT!!
It's difficult to actually figure out what you are talking about here..from what I see this article is about writing code to the AMD stream framework and have it target X86 (or AMD GPUs).
If your concern is shipping object code to a card to be processed may end up being so time consuming that it would not be worth it. Then I'd say that most examples of this kind of processing I've seen are doing some specific highly scalable task (e.g. MD5 hashing, portions of h264 decode). So clearly you have to do a cost/benefit like you would with any type of parallelization. That said, the cost of shipping code to the card is pretty small. So I would expect any reasonably repetitive task would afford some improvement. You're probably more worried about how well the code can be parallelized rather than the transfer cost.
Too bad. Evidence says that your pontification is probably incorrect.
After all, it's easy to point out the specific issues with arguments that are irrational, spurious and pedantic and demonstrate them to be such. The former two are easy to show contradiction and the latter it's easy to show a gap in relevance to the issue at hand. As long as your opponent is willing to talk (which I clearly am).
But instead of simply pointing out the specific places where my arguments fail. You seem to have a penchant for *pronouncing* things as wrong in some vague way (i.e. irrational, irrefutably researched, yadda yadda..) and then running away without actually making your case.
Any reason your running away like this would be compelling? Don't you think there's better evidence to believe that you are simply engaging in face-saving or dissonance preserving here?
Anyway whichever of your stall tactics you engage now. I think it's pretty clear that you've lost this round.
Slashdot Mirror
Because if they didn't, what profit motive would Microsoft have to support this conspiracy?
I'm not really sure where you're getting 'conspiracy' from here.
The point being is that we simply wouldn't have bought new machines (in the case of non x86-64 machines) or updated the OS (in the ones that were). As an aside XP64 was not (according to our MS rep) licensed for us (so still an OS upgrade) and lacked some hardware support.
Of the ones I was talking about? They all either supported > 4GB or supported 4GB but could only utilize 3GB.
It means "practically speaking, if you need more than 4GB of RAM get a 64 bit cpu/OS and you'll be much better off."
However only if that's the general use case. Otherwise the term 'practical' doesn't seem to apply.
Yes absolutely. But so what?
(I think I accidentally chopped some of my response - bad tag maybe!) Pretty much what I say above. That you're assuming that > 4GB in a single application is the general use case of having from 4 to 8 GB. For the low end I rather suspect that it's not.
but I don't think that it can't be objectively measured.
;-). That is to say how much involved say a large SRS of field users.
...and that's just one term 'direct'. IMHO each of your other statements is replete with terms which contain similar "ten ton assumptions" for which there are easy common counter-examples. e.g. I've certainly witnessed scores of older people be unable to operate touchscreens because their mental model doesn't include having poor calibration or capacitive sensing. Is this the general case? Don't know - but it's enough to call your statement into question.
Didn't say it couldn't however I said it's difficult. By which I mean - difficult to do usefully.
The first and most key aspect of an intuitive interface is how readily it is grasped by the user. A chair, for example, is an incredibly intuitive interface. The form of an intuitive interface matches the function it exports. The degree to which a user can be introduced to an interface concept and understand it is something that can be measured.
That's one way of defining it. Another is that the UI of a chair is simply consistent to your pre-established actions for sitting. Someone who was trained to sit differently (say only zazen) might find a chair completely incomprehensible.
To me using the term 'intuitive' seems to appeal to some intrinsic value not some learned behavior.
The utilization of focused user trials can help discover the problem areas of an interface and that information can be used to improve intuitiveness.
Possibly, but how much research is there on the utility of 'focus user trials' in UI design
The first is the degree to which the interface conforms to the user's preconceived concept of the function.
Yes, yes. D. A. Norman referred to this as the users 'mental model' of the device. I think you're oversimplifying considerably though. Sure most objects we encounter require some kind of manipulation but 'direct'? Again there seem to be pretty heavy assumptions behind that term. Sure I can watch my daughter interact with a ball and watch it move in response to her actions. However before she can walk she is operating simple machines where the mechanism is hidden. Norman put it better (and way less pretentiously than you) by simply stating that the user has *some kind of idea* how the machine works and the closer the operation mimics the users mental model the better it is designed...
Herein lies the problem. I don't doubt that you can objectively test some things related to UI design - there may even be some value in doing so. However it seems pretty difficult to determine what exactly you are testing. Especially if you're going to leap to stamping any of these results 'intuitive'. Which is better off in the marketing literature.
I'd agree that there's a lot of contextual information when interpreting sentences and "there is a 4GB architectural limit" is no exception but what does "for nearly all practical purposes" mean then? I mean if we really are "in the era of 64-bit budget computers" doesn't that imply that people who are using more memory than 4GB are likely at the lower end (i.e. >4GB
To me anyway MS can license their OS any way they want (memory, CPUs, number of keys on your keyboard) however we probably would have held off on upgrading a number of machines if we simply could have upgraded their memory.
...the usual scaling for actual battery life? 0.5 * (Figure provided by vendor) = (hours of useful work)?
Not to whine but without LAN what are you doing about firewall traversal? Plenty of wireless, LAN and mobile networks block assloads of ports and perform overzealous packet inspection. I work for an educational institution and we do this to differing degrees. One place where we do it the most is on the wireless network. So since in this situation how easy is it going to be to set up a game using only battle.net.
Give me five examples of a workflow that is easier on the Iphone's UI, compared with all other phones on the market? No wait - give me one?
;-)
I'm not really part of the "It's really natural" camp. In fact I don't really believe in intrinsically intuitive interfaces at all - just consistent ones. Anyway the term "workflow" is kind of broad but the most useful thing I find about the iPhone UI toolkit is the "pinch" gesture. For magnifying part of the screen. Compared to all the other web-enabled portable devices I've used previous (several WinCE PDA/Smartphones, Blackberrys all the way back to the original). I find the "pinch" to be very easy to focus on a particular portion of a document. The next one would be they way it auto-magnifies form fields. In the interest of balance - the things I could go either way on: "thumb flick" for scrolling seems equivalent to me to a jog dial. Auto-landscape isn't really any better than a button that does the same thing (sometimes it saves me time, others its annoying and there are clear compromises in the auto functionality which I suspect are the drivers behind it's modality). Things I hate: Copy/Paste is annoying, Applications don't always have an obvious way to ascend and descend their hierarchies. "Shake to shuffle" is cute but having it on by default is lame and I don't see how anyone can use this feature and travel.
it's fundamentally unanswerable, you just have to accept it as true.
I'm sure there's a big slice of dogma amongst those folks. That said, I'd almost make that statement about the vast majority of UI claims. Why? It's simply difficult to test. I mean ask yourself what unit is "intuitiveness" expressed in? Time to reach competency? What is competency? Can competency be defined in an objective way? (i.e. Say we were looking at "typing" - during my highschool education 40WPM was considered competency. This figure was chosen - according to my teacher - because that's what she understood to be the threshold businesses wanted. Which is fine but is there any similar "objective" for the iPhone - other than entering the SMS olympics
No argument that the 360 was more poorly designed than the PS3. When I look at things like the MS Surface (and I do mean "look" we have one on site). I get the feeling that MS's internal opinion of their hardware development skills is rather incongruous with that of most other engineers.
Focused significantly more on quality control and then simply shipped every second 360 casing with packing peanuts and achieved the same result?
Linux isn't written by lonely nerds hiding out in their parents' basements. It's written by people working for major companies...who just happen to also be nerds and like to be close to mom.
Uh sure...but not by much. ~10Cals/lb. So negligible that the most widely used BMR formulae don't even take it into account. It's one of a big list of facts about weight management where the spoken frequency is inversely proportional to it's utility.
That said most forced exercise people do (i.e. walking) doesn't scale well with weight.
So muscle or lack thereof doesn't appear to be the primary variable in why increased weight doesn't significantly limit weight gain. Rather that the extra exercise you get with a heavier body is insignificant compared to the caloric intake it took to get there.
Meh...so the way I'm going to interpret the defining silence here is that beguyId doesn't really know anything. He's like that other guy nine-times (who had hauntingly similar opinions - maybe ones a sockpuppet?). The kind of person who just drops off a study (not that I'm convinced they even READ the studies attached to their link).
The "calories are only one factor among many" idea is interesting from a social point of view. It gives the appearance (to me anyway) that the person is claiming that somehow caloric reduction is not the main thing people should be worrying about in terms of weight loss.
But it is...
Even after going through the only human study indirectly linked to by beguyId (and putting aside some of my caveats like having an n=12 and an R=0.5 and some ambiguity in terms of what kind of caloric restriction these people were actually on and how well controlled it was). It doesn't seem to say that at all. Even if we take all the claims there at face value (and apply them universally) it looks - at best - you could cut an extra 10% of body weight by restriction your fat intake when you are dieting (and only after you had already lost some weight).
Ok now back to what was actually being talked about. Someone who seems to claim that they should be obese (by their eating habits) but are not. They specifically mention that when they are allegedly overeating they are eating fatty foods but their weight doesn't move. This is exactly the opposite correlation that the study this moron linked to implies. If fat is not restricted then weight-loss is less efficient.
Next even if beguyid had a synapse or two firing so they could get the direction of the correlation correct. The disparity between the alleged weights is too big to account for even by the best case for the study. The person I was responding to was talking about closer to a 17% gap between consumption and weight.
Sorry 11550 should be 115500
Sorry 11500 should be 115500
Or you're insisting the research findings must be changed to fit your THEORY.
First of all, you need to differentiate between research and someone just posting "I counted calories". Even if the research completely explained and perfectly predicted this persons alleged results - which it doesn't appear to. Assuming all other sources of error are zero seems to be pretty bad reasoning. More the kind I would personally associate with those pandering "junk science".
While it is so nice and tidy to say it's as simple as calories, it just isn't true.
Actually I never said that. Which makes you kind of an idiot. Nor do you adequately explain what is meant by "as simple as".
As only one factor, among many, recent research has shown that the particular bacteria in a person's gut makes a large difference in how much food is actually absorbed.
I skimmed the papers. I admit I'm a little intrigued over your interpretation here. What is large? Large enough to cover this alleged 11500 calorie gap? Significantly overshadowing calories taken in?
It's not impossible. My daughter has coeliac disease, and i've been tested (blood test only) and come back negative but it's not impossible that I have something like that going on.
Impossible is not really the issue. If you are passing 11550 calories of food. That's anywhere from 1.4Kg to 2.8Kg of extra poop per day (and who knows how many cc's). You would be going to the bathroom so often people would think they are holding the Oscars there.
A lot of the time when I take my temp it is around 37.5-38
Congratulations you're normal. Point being if you are somehow not excreting this insane amount of caloric surplus. It has to be *doing* something, something that produces very little matter (otherwise you would gain weight) and it has to do so in a short period of time. The most efficient way it can do this is to emit heat.
Again, the smart money is on inaccurate measurement.
I don't think it's quite that simple for everyone. I've been through periods of eating really badly (high fat takeaway for lunch every day for weeks on end) and then really healthy, with identical exercise level (~none), and my weight never moved outside the 69-71kg range that i seem to have been stuck at for the last 5-10 years. I've added up the calories I intake vs the exercise I do (next to none) and by all calculations I should be a balloon. 69-71kg would be about right for my height if there was a bit more muscle on me.
If you are implying that you are of normal BMI (we'll say this is 25) but due to your diet you should be obese (BMI 90) Then you are almost assuredly mis-measuring.
Think about it, when you eat food the choices are either excrete it or metabolize it. Making some guesses as to your height (168 cm) you appear to be saying that according to your measurements you are eating enough calories to put you at the 85Kg mark. That's something like eating at a 115500 calorie surplus. Come on. Do you really think you are passing that much extra food? or if you're not passing it. How exactly is your body burning this without some other physiological symptom? (i.e. tachycardia, +2 degree c fever).
Interesting you should mention that. I just read some of Walter Mischel's work available on google scholars.(Advances in experimental social psychology, Volume 7) they did an experiment where they measured the mean time children would wait based but asking them to think (or "ideate") with one of three modalities. i) Fun - that is think of something fun to do, ii) The reward - think of the thing they are going to get and iii) No particular instruction. They did this test with both the reward visible and obscured.
The results were kind of interesting: With the rewards visible - thinking about nothing was the worst strategy (to improve waiting time) or par with thinking about the rewards with the reward not visible. Thinking about something fun was the best strategy in either case but when the rewards weren't visible - no instruction was almost as good.
So to a degree it seems that it can be taught however it doesn't answer if this 'learned' behavior has the same effect (improvements in test scores, etc..) and it is difficult to verify that the child was actually thinking what they were asked to think about.
I suppose it depends on what you mean by "lots of tiny chunks". Clearly doing a single "burst" transfer is better than lots of small ones but if you are still planning to process all these "chunks" of data at the same time then there's no reason why you couldn't just ship them all together and process them individually. Perhaps even from shared memory.
Unless of course we're taking about a bunch of chunks that are not going to be worked on simultaneously which goes back to my statement about the degree of parallelism that can be achieved being your chief worry.
Yes, this true for a lot of people who speak outside of their area of expertise. Some recognize it (Feynman comes to mind in "The Meaning of it All" where he IIRC says in effect that has he strays farther from talking about physics people should give less credit to what he's talking about) others don't.
I had an opportunity to hear Kurzweil speak and his reasoning (as given) was a little flawed. One he talked about how computing power increased over time in a predictable pattern doubling every 18-24 months (moore's law). I thought it was interesting that he didn't seem to discern between transistor capacity of chips (closer to what that law talks about) and computing performance. If computing performance was actually doubling each and every 18 months. Then next year I could replace my machine with one that runs an (algorithm) twice as fast for roughly the same price. However that generally doesn't happen and just by glancing at other benchmarks I suspect that increases in computing performance are largely domain specific. i.e. highly parallelizable tasks have recently enjoyed a better than average performance increase due to multicore architectures. Whereas non-parallizable tasks probably have crept along.
...lost! Oh the humanity.
Perhaps this is just me but...probably what you had for lunch is pretty low on most peoples "care-dar". When I get together with my friends...know how often we talk about lunch...almost never. Know how may SMSs I've received about peoples lunch? or IMs or emails for that matter? Those figures hang pretty close to zero too. But Twitter? From my modest sampling of tweets it seems like it's pretty close to mandatory to shoutout about your ingestibles. I can think of some reasons for why this particular subject comes up but the real revelation for tweeters (or twits or whatever you call yourselves) should be that MOST OF YOU ARE REACHING PRETTY DAMN FAR TO COME UP WITH SOMETHING TO TALK ABOUT!!
It's difficult to actually figure out what you are talking about here..from what I see this article is about writing code to the AMD stream framework and have it target X86 (or AMD GPUs).
If your concern is shipping object code to a card to be processed may end up being so time consuming that it would not be worth it. Then I'd say that most examples of this kind of processing I've seen are doing some specific highly scalable task (e.g. MD5 hashing, portions of h264 decode). So clearly you have to do a cost/benefit like you would with any type of parallelization. That said, the cost of shipping code to the card is pretty small. So I would expect any reasonably repetitive task would afford some improvement. You're probably more worried about how well the code can be parallelized rather than the transfer cost.
Too bad. Evidence says that your pontification is probably incorrect.
After all, it's easy to point out the specific issues with arguments that are irrational, spurious and pedantic and demonstrate them to be such. The former two are easy to show contradiction and the latter it's easy to show a gap in relevance to the issue at hand. As long as your opponent is willing to talk (which I clearly am).
But instead of simply pointing out the specific places where my arguments fail. You seem to have a penchant for *pronouncing* things as wrong in some vague way (i.e. irrational, irrefutably researched, yadda yadda..) and then running away without actually making your case.
Any reason your running away like this would be compelling? Don't you think there's better evidence to believe that you are simply engaging in face-saving or dissonance preserving here?
Anyway whichever of your stall tactics you engage now. I think it's pretty clear that you've lost this round.
I'm guessing that by your lack of response you figured out that your definition was silly.
If not, try removing "intent" from a strawman argument. Still seems like a strawman doesn't it?