I remember bookmarking his Python pages online, and I thought to myself, "Awesome, this is, like, the future, man. I'm not going to download it all and keep a hard copy, I can just access it anytime. The future is, like, now, dude." Wholly my own fault, but I feel strangely... weird. My cloud-faith is... shaken. Maybe I should start printing out all my emails like it's 1993...
...but I'm not going to read the other 400 comments.
It would decrease my productivity by, like, 92%, but has anyone considered NO monitors? Seriously, 0 for IDE, 0 for logs, and zero for debugging. No feedback whatsoever! let's do it! The zero monitor movement starts here!
Dominious had the right idea - the vowels change the sound; additionally, consonants have weight in many tonal languages, so they actually perceive the sound differently. You think of them as the same phoneme, because in English, they are - even if they are allophonic. Taiwanese speakers have an additional distinction, so their phonological space is tighter in that area, and thus they are able to make finer phonetic distinctions - which is confusing, because you're (parent of parent) telling them they're all the same sound. So they're thinking, "this guy doesn't even know these sounds are different in his own language! Why listen to him? Ha ha ha."
Also, all this Wade-Giles/Yale gibberish, even pinyin, is useless. Try the IPA.
Those would all make for interesting stories, but technically speaking, we're talking about movies, and movies typically mean hollywood movies. I leave it as an exercise for the reader to determine the proof of whether hollywood movie == good story. They would all be very fine multi-hour documentaries, if they weren't dumbed down for public consumption. The problem isn't only directors - it's you, and the rest of us. People don't like what we're interested in. The parent you replied to meant Normal Humans.
Can you imagine a Kens Burnsy documentary on pre-moon NASA engineering? With famous actors reading lines of machine code and voltages and lots of graphs of rocket test results? Comparing graphs of all the different compounds in IGNITE! That would be like the PBS of my dreams. They could call it "Rocket Science," and if they had to reschedule the show, they could say, "this isn't Rocket Science, people."
The point is that however reasonable it might be, we now have at least one structured study (albeit on only one subject group), where we can say, look, modified speech is whatever% shorter than non-modified speech, the child's acquisition proceeded by this amount in this period of time, the complexity of the utterances was some degree less than normal speech based specifically on these verb/Det/prep phrases, these certain grammatical structures were characterized spontaneously by caregivers as "simpler," etc. If we get another cognitive scientist to do it, we'll have another dataset. A few more centuries and thousands of recorded case studies of phonotactics, timing, and intonation, and we'll be able to discover if 4 billion years of evolution really offers the optimal acquisition solution. Evolution is case dependent, there's lots of room for improvement to the human condition. It also wouldn't hurt to suss out the whole "module" structure of human grammar, perception and production - if only to understand ourselves better, if not to create really bitchin' NLP A.I.s
I've forgotten where I heard the quote - something like, if your Grandmother knew it, don't do a paper on it. That's simply not the case in cognitive science/language acquisition. Everything we know about language is mostly wrong, and built on popular misconceptions. We can all speak competently to communicate, but humans are not therefore language specialists - I know when a car isn't working, because things are exploding or not moving, but it doesn't mean I can lay out a detailed plan to build a great car.
Child preference is not "known" for baby-talk - the studies that indicate that can easily be explained by the high frequency differentials involved in motherese, in the same way that bright, primary colors stimulate their visual field. Stimulation that elicits laughter doesn't necessarily mean preference, and if it does, it still does not indicate higher differential utility than other stimulus, such as adult talk. What cognitive scientists are trying (or would like to be trying) to discover is the effect of these different patterns on first language acquisition and development. There is still a pitched battle raging on nature vs. nurture in the area of child language acquisition, and if there is, in fact, a difference, there is money to be made and careers to be launched. If evolution has fostered baby-talk to optimally stimulate language development, then great. If there is rather some other optimal method, if humans can acquire language more efficiently, then all bets are off. I like to imagine in the latter scenario a global arms race in infant socialization techniques.
It is SO HARD to get ethics committee approval for the proper experiments, of course.
My Mom's old boyfriend did a study on anecdotes and it turned out that anecdotes are generally accurate. That's just one study, but it's pretty easy to extrapolate to a larger sample.
...but it was so awful I had to comment. Awful, awful - like Shannara-awful. Utterly terrible fan-fiction/mary-sue non-david-foster-wallace-ishy david-foster-wallace footnoting in the main text, modern idioms, just ridiculous. The translation also has some pretty glaring errors - lots of present perfect tenses stuck inside of relative clauses consistently (you keep using that tense - it doesn't mean what you think it means). Not a terrible or unexpected translation for an L2 translator, but it adds to the cumulative effect. Sorry, but bad.
As a skeptic, I (as you suggest) dismiss your example as delusional, but your sample is too flawed and your factors too unmeasurable for me to consider it a "statistical fluke." I like that you admit an astrologer would accept a lot less data to make a conclusion, and I'm certain you arrived validly at "on-in-a-million" odds, but your data set is disastrously small. Any small error will grossly disrupt your error; i.e. one of the set lied about their birthday, one of your set was born in a differing time zone or continent that would effect date relative to absolute birth, the division of the birthday into hours/minutes (what do you consider a "28 day cycle?" Literally 24*28, or 60*24*28, or 24*28+-an hour - what's the variance of your target?), the vagaries of how we meet friends - all of this compounded by the inherent difficulty of measuring "friendship-ness," and what that factor would entail if it were real.
This is similar to the Birthday Problem - where 99% probability is reached with only 57 people. Except for your constraints, instead of 1/365 possible days, we have to merely be 28 days apart - again, I don't understand the limitations of your bio-rythm thing, if it's |28| or any multiple of 28, or whatever. But not a statistical fluke. Statistics is the study of flukes, and their flukiness, with more advanced stuff being for real outliers - your sample can be dealt with effectively by classical statistical methods. Your sample isn't very fluky.
What's really interesting is the logical flaws in your reasoning, and I don't mean that to disparage you, I'm only pointing out the logical error that undergirds the mistake you made. You found your significant correlation, but what if you were wrong? What about all the other possible hypotheses that you didn't consider? What if it wasn't this silly bio-rythm of 28 days thing, but rather some significant multiple of the time needed to grow a braincell in a young brain, or a multiple of a vibrating helium atom, or a zero of some large number sequence based on our future time-of-death?
spelling, not grammar, and there's certainly a punctuation problem there, but still, no. There is a difference. Grammar Nazis are the least well-educated of all of our nazis.
Sure, I can't. We're talking both public and international primary and high schools in Taiwan, Vietnam, and Korea. I'm only generalizing about those specific examples where I've had the actual experience, and of course it isn't representative, but while yes, the particular students I had did those things, the teachers and proctors as well found it perfectly acceptable. A lot of the time you can also go in to administer a placement test (or the TOEFL, but I heard it secondhand, although I was there for the GRE), for example, and the reality (as they've explained it to me after I'd take some student's test away) is that if you stop them from cheating on the test, there are simply other avenues for the kids to pursue - namely bribery between the parents and the weakest administrative link, and bribery is something that many of these governments are actively trying to curtail, so it's a question of which is the lesser evil, made simpler by the fact that, again, the cultural stance against cheating is, stereotypically, not as bad a thing as it is in the west. Your mileage may vary, and I hate generalizing about some mystical eastern collectivist-facey culture thing, but evidence usually bears it out, in this case.
Also, protip: "naive" is "evian" spelled backwards (without the diaeresis) and vice versa! Unless you're using "e" to signify the diaeresis, which is kinda smart actually.
It's unfortunate that someone decided to mod the parent as 'troll' - regardless of whatever native language this "scientist" speaks, "several" would have to be exceedingly large, and there would have to be a process in place to keep non-optimal martian children from having sex (which is ironically what this whole conversation is largely about). It is truly sad the utter lack of understanding of differential reproductive success and just general natural selection knowledge this individual exhibits, and this is compounded by him using his credentials as... some kind of scientist, and then reporting incorrect impressions of a separate field. Since he said it to a reporter, he presumably actually believes this account of evolution, which isn't particularly encouraging, but often even other biologists have a tenuous grasp on natural selection and only hear about it in a watered-down version in a couple undergrad classes. He is, as per parent, an "idiot," and an embarrassment.
It's hard to explain really, but it involves face and the assumption of entitlement. If you've gotten to the exam stage, it's just a formality to pass the test, and preventing you from doing so is assumed to be contrary to societal norms. I'm not the OP of course, but I too have spent time outside the west, and proctored a few exams. What the OP relates is accurate. We had some discussions about ethics in one class, and the students were utterly mystified by the western attitude towards cheating. The best students in the class asked how the top-graded students could abandon their lesser classmates like that. If you have the knowledge and you refuse to share it with everyone, you're seen as being very arrogant and greedy. These are students who will walk up to a student being questioned and hit them on both sides of the face, grab their chin, and say to the teacher "see - he's stupid. Don't ask him questions, we need to help him." This, by the way, is how students pass the TOEFL in asia to get into U.S. universities, despite any lack of technical skills. This is why the GRE is useless for foreigners (perfectly acceptable to text and look up things on wikipedia during administration). And, incidentally, it is why all the great math scores that come out of standardized testing in Asian tiger nations showing an achievement gap are utterly baseless and useless as a comparative measure. The scores are of the top students, and all the other scores are the lesser students copying answers from the top students. The teachers actively promote cheating in most public, non-IB schools, as well as at the university level. They brazenly cheat from the youngest ages because there is no corollary to our ethical prohibition.
Slashdot Mirror
I remember bookmarking his Python pages online, and I thought to myself, "Awesome, this is, like, the future, man. I'm not going to download it all and keep a hard copy, I can just access it anytime. The future is, like, now, dude." Wholly my own fault, but I feel strangely... weird. My cloud-faith is... shaken. Maybe I should start printing out all my emails like it's 1993...
...but I'm not going to read the other 400 comments.
It would decrease my productivity by, like, 92%, but has anyone considered NO monitors? Seriously, 0 for IDE, 0 for logs, and zero for debugging. No feedback whatsoever! let's do it! The zero monitor movement starts here!
Dominious had the right idea - the vowels change the sound; additionally, consonants have weight in many tonal languages, so they actually perceive the sound differently. You think of them as the same phoneme, because in English, they are - even if they are allophonic. Taiwanese speakers have an additional distinction, so their phonological space is tighter in that area, and thus they are able to make finer phonetic distinctions - which is confusing, because you're (parent of parent) telling them they're all the same sound. So they're thinking, "this guy doesn't even know these sounds are different in his own language! Why listen to him? Ha ha ha." Also, all this Wade-Giles/Yale gibberish, even pinyin, is useless. Try the IPA.
Those would all make for interesting stories, but technically speaking, we're talking about movies, and movies typically mean hollywood movies. I leave it as an exercise for the reader to determine the proof of whether hollywood movie == good story. They would all be very fine multi-hour documentaries, if they weren't dumbed down for public consumption. The problem isn't only directors - it's you, and the rest of us. People don't like what we're interested in. The parent you replied to meant Normal Humans.
Can you imagine a Kens Burnsy documentary on pre-moon NASA engineering? With famous actors reading lines of machine code and voltages and lots of graphs of rocket test results? Comparing graphs of all the different compounds in IGNITE! That would be like the PBS of my dreams. They could call it "Rocket Science," and if they had to reschedule the show, they could say, "this isn't Rocket Science, people."
Thank you for that, sir.
The point is that however reasonable it might be, we now have at least one structured study (albeit on only one subject group), where we can say, look, modified speech is whatever% shorter than non-modified speech, the child's acquisition proceeded by this amount in this period of time, the complexity of the utterances was some degree less than normal speech based specifically on these verb/Det/prep phrases, these certain grammatical structures were characterized spontaneously by caregivers as "simpler," etc. If we get another cognitive scientist to do it, we'll have another dataset. A few more centuries and thousands of recorded case studies of phonotactics, timing, and intonation, and we'll be able to discover if 4 billion years of evolution really offers the optimal acquisition solution. Evolution is case dependent, there's lots of room for improvement to the human condition. It also wouldn't hurt to suss out the whole "module" structure of human grammar, perception and production - if only to understand ourselves better, if not to create really bitchin' NLP A.I.s
I've forgotten where I heard the quote - something like, if your Grandmother knew it, don't do a paper on it. That's simply not the case in cognitive science/language acquisition. Everything we know about language is mostly wrong, and built on popular misconceptions. We can all speak competently to communicate, but humans are not therefore language specialists - I know when a car isn't working, because things are exploding or not moving, but it doesn't mean I can lay out a detailed plan to build a great car.
Child preference is not "known" for baby-talk - the studies that indicate that can easily be explained by the high frequency differentials involved in motherese, in the same way that bright, primary colors stimulate their visual field. Stimulation that elicits laughter doesn't necessarily mean preference, and if it does, it still does not indicate higher differential utility than other stimulus, such as adult talk. What cognitive scientists are trying (or would like to be trying) to discover is the effect of these different patterns on first language acquisition and development. There is still a pitched battle raging on nature vs. nurture in the area of child language acquisition, and if there is, in fact, a difference, there is money to be made and careers to be launched. If evolution has fostered baby-talk to optimally stimulate language development, then great. If there is rather some other optimal method, if humans can acquire language more efficiently, then all bets are off. I like to imagine in the latter scenario a global arms race in infant socialization techniques.
It is SO HARD to get ethics committee approval for the proper experiments, of course.
You read it wrong. There's thousands of three-year-old horses up there, running around and around and around...
You're right. I would never peg that site as a real journal if I was just wandering by. It's very timecubey.
Hopefully with this we can literally, for all intensive purposes, put behind this us.
!siellA rammarG etah ew ,htraE rou nO
Make a time machine and set it to 6 months...
My Mom's old boyfriend did a study on anecdotes and it turned out that anecdotes are generally accurate. That's just one study, but it's pretty easy to extrapolate to a larger sample.
it's the classic "think globally, introduce non-native species locally."
...but it was so awful I had to comment. Awful, awful - like Shannara-awful. Utterly terrible fan-fiction/mary-sue non-david-foster-wallace-ishy david-foster-wallace footnoting in the main text, modern idioms, just ridiculous. The translation also has some pretty glaring errors - lots of present perfect tenses stuck inside of relative clauses consistently (you keep using that tense - it doesn't mean what you think it means). Not a terrible or unexpected translation for an L2 translator, but it adds to the cumulative effect. Sorry, but bad.
Crap. Does this mean we have to push the singularity back again?
I only read the first and last sentence of long posts, and yours makes no sense. How do you make hours for your dad?
As a skeptic, I (as you suggest) dismiss your example as delusional, but your sample is too flawed and your factors too unmeasurable for me to consider it a "statistical fluke." I like that you admit an astrologer would accept a lot less data to make a conclusion, and I'm certain you arrived validly at "on-in-a-million" odds, but your data set is disastrously small. Any small error will grossly disrupt your error; i.e. one of the set lied about their birthday, one of your set was born in a differing time zone or continent that would effect date relative to absolute birth, the division of the birthday into hours/minutes (what do you consider a "28 day cycle?" Literally 24*28, or 60*24*28, or 24*28+-an hour - what's the variance of your target?), the vagaries of how we meet friends - all of this compounded by the inherent difficulty of measuring "friendship-ness," and what that factor would entail if it were real.
This is similar to the Birthday Problem - where 99% probability is reached with only 57 people. Except for your constraints, instead of 1/365 possible days, we have to merely be 28 days apart - again, I don't understand the limitations of your bio-rythm thing, if it's |28| or any multiple of 28, or whatever. But not a statistical fluke. Statistics is the study of flukes, and their flukiness, with more advanced stuff being for real outliers - your sample can be dealt with effectively by classical statistical methods. Your sample isn't very fluky.
What's really interesting is the logical flaws in your reasoning, and I don't mean that to disparage you, I'm only pointing out the logical error that undergirds the mistake you made. You found your significant correlation, but what if you were wrong? What about all the other possible hypotheses that you didn't consider? What if it wasn't this silly bio-rythm of 28 days thing, but rather some significant multiple of the time needed to grow a braincell in a young brain, or a multiple of a vibrating helium atom, or a zero of some large number sequence based on our future time-of-death?
spelling, not grammar, and there's certainly a punctuation problem there, but still, no. There is a difference. Grammar Nazis are the least well-educated of all of our nazis.
Sure, I can't. We're talking both public and international primary and high schools in Taiwan, Vietnam, and Korea. I'm only generalizing about those specific examples where I've had the actual experience, and of course it isn't representative, but while yes, the particular students I had did those things, the teachers and proctors as well found it perfectly acceptable. A lot of the time you can also go in to administer a placement test (or the TOEFL, but I heard it secondhand, although I was there for the GRE), for example, and the reality (as they've explained it to me after I'd take some student's test away) is that if you stop them from cheating on the test, there are simply other avenues for the kids to pursue - namely bribery between the parents and the weakest administrative link, and bribery is something that many of these governments are actively trying to curtail, so it's a question of which is the lesser evil, made simpler by the fact that, again, the cultural stance against cheating is, stereotypically, not as bad a thing as it is in the west. Your mileage may vary, and I hate generalizing about some mystical eastern collectivist-facey culture thing, but evidence usually bears it out, in this case.
Two words: frog DNA!
Also, protip: "naive" is "evian" spelled backwards (without the diaeresis) and vice versa! Unless you're using "e" to signify the diaeresis, which is kinda smart actually.
It's unfortunate that someone decided to mod the parent as 'troll' - regardless of whatever native language this "scientist" speaks, "several" would have to be exceedingly large, and there would have to be a process in place to keep non-optimal martian children from having sex (which is ironically what this whole conversation is largely about). It is truly sad the utter lack of understanding of differential reproductive success and just general natural selection knowledge this individual exhibits, and this is compounded by him using his credentials as... some kind of scientist, and then reporting incorrect impressions of a separate field. Since he said it to a reporter, he presumably actually believes this account of evolution, which isn't particularly encouraging, but often even other biologists have a tenuous grasp on natural selection and only hear about it in a watered-down version in a couple undergrad classes. He is, as per parent, an "idiot," and an embarrassment.
It's hard to explain really, but it involves face and the assumption of entitlement. If you've gotten to the exam stage, it's just a formality to pass the test, and preventing you from doing so is assumed to be contrary to societal norms. I'm not the OP of course, but I too have spent time outside the west, and proctored a few exams. What the OP relates is accurate. We had some discussions about ethics in one class, and the students were utterly mystified by the western attitude towards cheating. The best students in the class asked how the top-graded students could abandon their lesser classmates like that. If you have the knowledge and you refuse to share it with everyone, you're seen as being very arrogant and greedy. These are students who will walk up to a student being questioned and hit them on both sides of the face, grab their chin, and say to the teacher "see - he's stupid. Don't ask him questions, we need to help him." This, by the way, is how students pass the TOEFL in asia to get into U.S. universities, despite any lack of technical skills. This is why the GRE is useless for foreigners (perfectly acceptable to text and look up things on wikipedia during administration). And, incidentally, it is why all the great math scores that come out of standardized testing in Asian tiger nations showing an achievement gap are utterly baseless and useless as a comparative measure. The scores are of the top students, and all the other scores are the lesser students copying answers from the top students. The teachers actively promote cheating in most public, non-IB schools, as well as at the university level. They brazenly cheat from the youngest ages because there is no corollary to our ethical prohibition.
That's how statistics works! Sometimes, you get an outlier, because if you didn't, the mean would be far more meaningful. Awesome.
If you want to be doubly sure, you can soak your insulation in holy water and bless your construction workers with a priest. Why take the risk?