Maybe you ought to try making your own lecture notes and slides available, and not teach from a book at all. As a college student, that's the kind of class I prefer!
There are several issues here. First, at a research university, a professor's primary focus is usually on research Devoting the time to develop material for an introductory course would take away from the pursuits upon which tenure, promotion, and overall status is based. Most faculty in this setting teach in a manner which does not consume too much time and lets them devote their primary energy to their research. Maybe this isn't the way it should be, but it is generally this way at most US research universities.
Second, writing a decent textbook is actually quite difficult. There are a number of gifted authors who can explain notions in a way that many students can get, giving useful explantions without outrageous/distracting amounts of detail. Those people should write texts. Other people should not. There are a huge number of terrible texts written by people who thought they could do a better job than the standard texts and do not realize how difficult it is to communicate effectively on the written page. Even very gifted teachers who excel in the classroom (which is much more interactive setting than a text, even in 300+ student lectures) often have a hard time writing a good, readable textbook.
Third, a decent text is actually a significant effort on the part of the publisher. These days, for example, people expect snazzy 3d figures and a level of clarity in figures and examples that is well beyond what a typical textbook author would be able to do themselves (it's really a different skillset.) Given how many errors there are in a typical text, it may be hard to believe but it also takes a fair amount work proofreading, working out exercises so the investment is significant. People tend to underestimate the work it takes to do thing right, which is perhaps why there are so many terrible textbooks out there.
It's worth mentioning that Carleson was on the faculty at UCLA, usually spending at least the winter quarter there (it doesn't take a genius to prefer Los Angeles in February to Uppsala in February.) I think all of the graduate students he advised were in Sweden though, which seems to be the case from the math genealogy site: http://www.genealogy.ams.org/html/id.phtml?id=1978 1 He did at least intermittently teach the first-year graduate analysis course at UCLA, and made those students suffer (and learn...)
Despite the limitations on the accuracy of the data, this is interesting and seems to describe well behavior I've seen. I have about 12 Apple laptops for student use (they develop code to run on our clusters and having their own laptops works great for that) and have had several for my own use since 2000 and have been very happy with the machines overall. AppleCare has worked very well- it's quick and effective. I used to use Sony Vaio subnotebooks, which were nice laptops and ran Linux nicely, but they were quite flimsy and Sony's repair process was ridiculous. They seemed to think it was reasonable to have the laptop sent to them for a month- being without a machine for a month is a huge hit, especially if it is your only/main machine. So for the Sonys I would wait until there were three or four problems before sending them in, since the process was a huge, slow hassle. For the Apples, they express mailed a box and you expressed it back at their expense, and they were never gone for long, and the process was very smooth, especially compared to Sony's nightmare repair process. I gave up on Sony and don't know if they have improved things, but a company that things that it's no big deal to keep your computer for a month and keep you in the dark about when its coming back isn't selling anything to me anymore. I did more repairs on the Apples than the Sonys, since I got a batch of iBooks which were vulnerable to the motherboard video weirdness, but the repairs were so smooth that it was no big deal so I wasn't waiting until there were a bunch of things wrong to send them back.
I found Apple very generous with the motherboard replacement on our iBooks, doing several for free even after the period and AppleCare expired. They did refuse to do the replacement until it actually failed, so there were a couple that we were just waiting to flake out, but overall that process was quick and painless, so all those high replacement motherboard failures in the 2001-2003 G3 iBook range really aren't as bad as they might look.
A few quibbles with the analysis on the table:
1) They don't normalize "dropping" for age of the computer, so that column is essentially meaningless. The fact that only 2% of the 17" PB G4 HRs have been dropped has a lot more to do with the fact that the machine has only been available since October 2005 than anything else, so highlighting that as "better than one standard deviation below the mean" is meaniningless.
2) There are different populations for the iBook and PB crowd, so it's a bit unfair to compare their reliability. That is, the iBooks are more likely to be used by students and the PB are more likely to be used by grownups. I think the iBooks are significantly sturdier machines (given how well they've held up to my students use) but given that on average they are subjected to less careful treatment (on average- I know some very careful students and some reckless faculty, but those are exceptions), their sturdiness doesn't show up as much in the table as a true apples v. apples (sorry) comparison would show, I suspect.
3) As in point #1, it appears that they didn't normalize for age of computer in any of the "component failure" columns, so the machines that are worse than 1 std. dev. are all older machines which have had more time to fail, so that should have been explained or corrected for.
4) The power brick/dropped correlation is described but not explained well. Those little white bricks have failed fairly often and I think are a significant problem. I think they don't hold up well under rough treatment (frayed cord where it joins the brick and frayed cord at the computer end were the most common problems) and "rough treatment" correlates well with dropping the computer, presumably. Certainly I have seen many bricks fall to the floor as they are dragged by their cords and repeated dropping takes it toll. I got several extra bricks to keep in our campus lab since students would often want to borrow mine after they left theirs at home.
Those are quibbles- I think the data there is interesting. Even though it has shortcomings, you can only work with the data you have...
I think that the questions are skewed to make things appear worse than they are, presumably because the survey is done by AOL and the National Cyber Security Alliance, who presumably have an interest in scaring people into their products and services. Aside from the obvious Linux/Mac issues described by other posters, "properly configured" firewall is a pretty strong definition and I expect many quite adequate firewalls could be classified as "improperly configured" even though they were effective against the bulk of the current attacks. Similarly, only counting anti-virus software if it has been updated in the last week is going to skew things- there is a big difference between having no AV at all and having AV that is running but has definitions that are two weeks or a month old, and the metric chosen groups those two cases together.
Charles Schumer, the senior senator from NY, has been big on this in the last month (and has been big on affordable drug policies for years...) First he said it was inexusable that Roche was putting their profits ahead of widespread safety, and last week said that Roche was being unreasonable in refusing to take steps to make the drug more widely available (not stepping up production, not meeting with other potential producers). Schumer threatened Congressional action to ignore the patent if there was no action from Roche in 30 days. Now it sounds like Roche relented, at least a little in the US market, and has agreed to step up production and at least talk with other potential producers about licensing.
The story about the moon has been on the linked webpage since at least the last Slashdot story
No- that was a different object (2003EL61) which has a moon- it is new news that 2003UB313 has one, and the prospect of using the moon to accurately weigh 2003UB313 is huge.
Michael Brown, one of the scientists on the team that discovered the planet and now its moon, has an excellent website about 2003UB313 and has been keeping it current. I've been checking it out to see if there are any interesting developments about the team that apparently claimed the discovery of 2003UB313 without mentioning the fact that they at least visited the logs of the telescope Brown's team was using, if not outright deducing its existence from those logs. It's great to see this kind of rapid dissemination from the principals. By the way, he also has an extensive website about his newborn daughter's sleep patterns which is pretty impressive too...
Actually, no- the CLT says that samples drawn from any distribution, no matter how non-normal it is, will have their averages behave closer and closer to a normal distribution as the sample size increases. It may be that the sample size needs to be large to have something that is a good approximation of the normal, but no matter how "thick-tailed" or "U shaped" a distribtion is, its sample averages will converge to the normal distribution as the sample size increases.
In actual practice, it may be that the sample size needs to be quite large for the averages to look "normal", but there is no getting around the CLT. The assumptions that are made to use standard statistical methods are exactly that- assumptions. If the assumptions are flawed (the distribution of an actual parameter is normal, rather than the averages, for example) then the conclusions are of course dubious. Famous examples of assumptions coming back to bite analysts in a big way are generally of two types- assuming something is normal that is not, as you point out, and also assuming that events are independant that are not. The failure of the independance assumption is thought to underlie the spectactular LCTM collapse, the Asian currency collapse in 97 and other domino-type collapses.
The classic example of "crowd wisdom" is the jellybeans-in-the-jar
experiment, often used in introductory MBA classes to convince
people that open markets value securities (basically) fairly. The
experiment goes like this: the professor brings a jar of jellybeans
and asks everyone to guess how many there are in the jar. The
individual estimates may vary quite a lot, but the average of
the estimates in the class is usually close, in fact often
closer than the closest estimate of any of the students depending upon
the size of the class. That
is in a situation where the students have very little information
about the jar and perhaps no experience with such estimates.
If there were greater experience and/or they were allowed more
information then presumably the individual and average estimates
would be even closer. Basically, this can be described as
the "Central Limit Theorem" in action- that the standard deviation of
averages is smaller than the standard deviation of the individuals
by a factor of the square root of the sample size, as
illustrated in this applet
or in this Mathworld description.
The CLT actually says more- that as the sample size increases, the
distibution of averages approaches that of the normal ("bell curve") distribution,
so the distribution of avergaes is roughly normal, and then techniques
designed to analyze the normal distribution can be applied with greater
certainty.
There is a long tradition in the astronomy community of respecting the priority of data gathered. Telescope time is often allocated by peer-review committees, where the proposals must include enough specifics about the desired projects to convince the committee to allocate the time. An unethical committee member could see a promising project during the review process and then do their own study. This rarely/never happens for several reasons:
A researcher caught using data or a project proposal inappropriately would have a hard time getting telescope time ever again at a significant observatory or funding from an agency which uses peer-review for allocation, such as the US National Science Foundation.
Dishonest people usually go into far more lucrative professions than optical astronomy.
In such a community, people are used to being somewhat secretive about what they are studying, but not to the point of compulsion. (Sometimes people enter intentionally erroneous information in targeting logs to keep secret what they are looking at, but this is rare and usually a sign of someone that is quite paranoid...) Brown was not paranoid and did not have good reason to be, given the standards of the astronomy community. He used the name generated by his software in the abstract, which happened to be keyed off of the date and turned out to be sufficient information for Ortiz to find the targeting data and then claim the discovery as his own. Since Ortiz did not credit Brown in any way and does not seem to have responded or tried to offer any explanation, it seems quite clear to me that he acted unethically and hopefully his institute will dismiss him- as it is, the director of Ortiz' institute has promised an investigation and will probably not want the institute to be known primarily as the home of Ortiz and this unethical behavior.
I saw a similar thing in the Geneva airport- next to some pay-to-connect WiFi networks, an open wireless network named "hidden." I don't know the story, but the idea that Mssrs. Payperconnect asked the owner of the open one to make his hidden, and so he named it "hidden"- that cracked me up...
I have seen in internet cafes "locked down" computers in cabinetry with only the USB ports accessible through cutouts. That is a nice convenience at such cafes that don't have Wifi and are nervous about letting me using an Ethernet cable since I can still move outgoing mail from my laptop to a keychain drive and then take two minutes to upload and download mail to catch up on offline. I hope that USB keychain exploits don't result in cafes closing up that last hole...
Indeed- remember also that 68k/PPC "fat binaries" weren't even necessary, really. They were a convenience during the 68k/PPC transition, since even pure 68k Mac code ran pretty much seamlessly and reasonably on PPC with the built-in emulators, and in fact continues to do that today, more than 15 years later. I have old Mac apps from the System 6 and before days that run fine on my modern TiBook (under Classic, of course.) Yes, emulation is not as fast as native, but somehow I think a 1.25 GHz G4 can cope with running stuff as well as my old 1993 33MHz 68k Powerbook Duo- it's not like there were 1.25 GHz native 68k processors around, anyway...
One thing that should be mentioned is that the amount is not as important as the level of the work, at least in math and science. If the homework is too easy, then it is tedious and repetitive and of little value. If it is too hard, the students have no idea where to start and make no progress and it is of little value. If it is at an appropriate level, the students are stuck for a little bit, perhaps, but then figure it out (maybe after looking at some other problems or other material, getting a gentle hint from someone or just having an "aha" moment) and it is useful. One problem is that in a given class, there will be a range of students and some problems that are the right level for some students will be too easy or too hard for others.
No one like being stuck on a problem, but to be honest that is the time that real learning occurs- trying to puzzle it out, and then hopefully succeeding. Even though students don't like being stuck and often just want to know how to do a problem, simply being told how to do problems generally ruins their value as a learning tool.
As a research mathematician, I spend almost every productive waking moment "stuck" on something and I am very happy with this arrangement! (There is also the sometimes tedious process of writing down stuff formally for publication once I've worked it out, but that part is less gratifying than doing the actual research and I'm always happy when one of my co-authors is willing to do that part...)
It's not clear how they sorted out the professions, but I think they were just using data from a more all-encompassing standard survey and probably didn't have as much choice as if they had done the survey themselves (in which case it would have been easier to get a larger sample, I would think.) They do mention as a specific example that doctors are considered neutral because they have both technical and emotional components to their success and training. My guess is that "flight attendant" and "librarian" weren't choices in the survey but who knows?
Actually- yes, your situation would contribute just as much to their empirical data as the same situation with biological children.
The survey data does not distinguish between adopted and biological children (those questions weren't asked) so it is possible that the effect is entirely one of choice in adopted cases, or at least that it is a contributing factor.
Conception has no memory, but the distribution of X and Y chromosomes in sperm has genetic links. That is, some men produce more X or more Y chromosomal sperm. So that is where these notions come from. Too bad Henry VIII 's wives didn't realize this- it would have saved a lot of trouble if they had just snuck off and met up with someone who already had produced a family of 8 boys...
Both of the linked articles are pretty flismy- the first claims that switching professions may increase the chance of having a child of a particular gender (confusing correlation with causation...) and the second one marvels at the notion that a sequence of children of the same gender is more likely than randomness would suggest (which is already well-established as there is some genetic predisposition towards male sperm having uneven fractions of X and Y chromosome shares).
The actual article (Journal of Theoretical Biology, 233, p589-599 "Engineers have more sons, nurses have more daughters: an evolutionary psychological extension of Baron-Cohen's extreme male brain theory of autism" by Satoshi Kanazawa and Griet Vandermassen and available through Elsevier's Science Direct) came out in December 2004 an is available online for those whose institutions subscribe, notes the following correlations:
This is based on survey data from US professions of around 1500 people. Only some of the professions are categorized as "systemizing" and "empathizing" so presumably the sample size is much smaller than that . The sample size isn't listed directly in the article but it appears to be about 20% of the 1500 with at least one parent so categorized profession, for around 300 people or so. Most professions are neutral in the "systematizing/empathizing" continuum, apparently.
Amoung those with "systemizing occupations" had regression coefficients of.35 with the number of sons and.14 with number of daughters, and those with "empathizing occupations" had coefficients of.27 with #sons and.40 with #daughters. (As a side note, it appears that "empathizing professions" have more reproduction overall, consistent with common myths about lonely geeky engineers...)
From the classification of professions:
Systemizing occupations
Executative, managerial, adminstrative such as financial managers, analysts, etc.
I've been reading CR since the 70s and have found that though there have been some changes over the years, the basic mission and role is more or less the same. I don't have the on-line version so I can't compare that but the magazine (having gone through many facelifts) does a reasonable job of comparing things. As another poster points out, sometimes their perspective is nit-picking and sometimes they don't do such a great job on products that I know more about than them (bikes, computers...) In general, the quality of manufacture, testing and design of products has risen with time so it may seem like they don't trash things as much as in the old days. Overall they do a reasonable job and have put a great deal of effort into maintaining their integrity (at great expense.)
If you get a chance to visit the Consumer's Union open house (each year in early October at their headquarters in Yonkers, NY reachable from NYC by transit easily)- make an effort to turn up. It's really great to meet everyone there and chat with them. These people love what they do and care a great deal about their methods and approach and are happy to talk about things. I find it very uplifting to go on such a pilgramage when I can. (It is a biased sample, since presumable the most enthusiastic and interested people are the ones who are willing to turn up and host an open house in their labs, but still it's great!)
ACM Transactions on Algorithms, impressive editorial board, formed from Journal of Algorithms by resignation and reconstitution
I know that other scientific disciplines have stronger histories of expensive journals and that their typesetting needs (color photos, etc.) may be greater than that of math and CS, so perhaps it is not so surprising to see math and CS being more of the pioneers here. There are plenty of good expensive journals now but the point is- why not have good inexpensive or free journals instead?
Lazy tenured professors are a problem everywhere and elicit no sympathy from me.
The approriate thing to do is put the same energy that used to be directed toward for-profit journals instead now to free journals or journals that are reasonably priced (for example, those run by professional societies not international publishing giants, as described eloquently by Knuth in his letter.)
By editing for, refereeing for, and contributing to free/modestly priced journals, those will benefit from the effective donation of prestige and we may move to a better model for the dissemination of scientific knowledge.
Many researchers have complained about the high price of academic research journals and some of us are doing something about it. The fundamental problem is that there are some prestigious, very expensive journals that libraries feel like they must subscribe to and authors feel compelled to submit there because they are prestigious. But things are changing at least in some disciplines. The cost of a journal is not so much for distribution- there are other costs, but those are largely actually borne by universities. A typical life story of a research article:
Brilliant researcher at Oxbridge University (who pays his salary) comes up with great idea, writes it up, submits it electronically by emailing it to an editor at the Snooty Journal,
The editor, a professor at Enormus State University (who pays his salary and has him teach a little less because of his prestigous editorship) thinks of an appropriate anonymous referee and sends off the article to be refereed. Snooty Journal may give ESU some money to cover part of the cost of a secretary, but does not pay his salary.
Professor at IviedHalls University (who pays his salary) receives the article to refereee, reads it, sends it back with comments after letting it molder on his desk/inbox for a bit.
Editor accepts or rejects the paper, possibly asking for modifications based upon the referee's recommendation, possibly some iteration at this step
Original author prepares the article in electronic format using LaTeX with Snooty Journal's style files and uploads it to their web site.
Snooty Journal staff typeset the paper, messing a few things up because they are not experts in the appropriate field, and send the "galley proofs" to the author to review.
Original author points out typos introduced in their typsetting process, sends back corrected galleys.
Snooty Jounal releases the article on their paid-subscription webpage and prints it as a dead-tree volume to send to libraries around the world that can afford it.
As you can see, the hard part of the labor (writing, reviewing, refereeing) is not done by anyone at the publisher-- various universities pay the salaries of those folks and they pay again for the journal in dead-tree form.
So you can see that there may be some objection to the arrangement. In the old days, the journal staff actually typset things and dead-trees were the only game in town, but most of the typesetting is done by the author.
The choice is hard for some people that really need to publish in the expensive journals to get tenure, recognition, grants, etc. But for people who already have tenure, some are resistant to the journal extortion. Some may have a policy like mine- I do not submit to expensive journals or agree to referee for expensive journals, now that I have the advantage of tenure.
There have been some successes of editorial boards that resigned wholesale, then started a free/inexpensive journal. Hopefully this becomes more common.
I've wondered about this also. Much of my moving around is by bicycle, which gives a good appreciation of the importance of coasting and travelling smoothly. In that case, it's important to be efficient and conserve since it's your own power that you don't want to squander. In a car, I also drive smoothly, anticipating changing lights, coasting when appropriate, not zooming off, and so on and I find that I usually get better mileage than others driving the same car, though not as dramatic I as I thought (usually 5-10%, depending upon how aggressive drivers they are and the car's setup and gearing, presumably.)
One big difference I notice with driving gently, though, is how long things last. My friends who zip around are always replacing clutches, brakes, tires, etc. and my mechanic is always marvelling at how long things last in our cars and my truck. It is a biased sample- as a mechanic, he sees people who are hard on their vehicles much more often than he sees me, so some of that is explained by the frequency of visit. I had to put a new clutch in my truck after 11 years and 150,000 of my miles and he was shocked that it was the original- and the truck is usually driven in the mountains full of camping/climbing gear or towing a small trailer with kayaks, boards and/or boats or full of someone's stuff that I'm helping them move, so it's not like it has an easy life. (The guy who had it for the 35,000 miles before me apparently drove like an idiot, so I usually blame him for things that break, but that is getting to be a stretch after all these years...)
Newer streamlined cars wouldn't see as dramatic effect as I do on the wind
That's true for wind resistance from weather-related wind, but actually, small fuel-efficient streamlined cars are quite sensitive to increased wind resistance, moreso than larger vehicles, at least when it comes to added wind resistance from bikes or luggage on the roof, for example. Even roof racks without anything on them can affect mileage noticably- I remember an old friend's 83 Honda Civic which got 38 MPG with empty roof racks installed and 44 MPG with them off the car or inside. That was far more than the 5-10% we had guessed before performing the experiment. When I see a car with Colorado plates in Maine with a large bulky souvenir lobster trap attached to their roof, I wonder if they realize that the lobster trap that cost $20 to buy may also add $50 to their fuel bill on the ride home (2000 miles,30MPG vs 22MPG, gas $2/gallon, rough calculation...)
Slashdot Mirror
There are several issues here. First, at a research university, a professor's primary focus is usually on research Devoting the time to develop material for an introductory course would take away from the pursuits upon which tenure, promotion, and overall status is based. Most faculty in this setting teach in a manner which does not consume too much time and lets them devote their primary energy to their research. Maybe this isn't the way it should be, but it is generally this way at most US research universities.
Second, writing a decent textbook is actually quite difficult. There are a number of gifted authors who can explain notions in a way that many students can get, giving useful explantions without outrageous/distracting amounts of detail. Those people should write texts. Other people should not. There are a huge number of terrible texts written by people who thought they could do a better job than the standard texts and do not realize how difficult it is to communicate effectively on the written page. Even very gifted teachers who excel in the classroom (which is much more interactive setting than a text, even in 300+ student lectures) often have a hard time writing a good, readable textbook.
Third, a decent text is actually a significant effort on the part of the publisher. These days, for example, people expect snazzy 3d figures and a level of clarity in figures and examples that is well beyond what a typical textbook author would be able to do themselves (it's really a different skillset.) Given how many errors there are in a typical text, it may be hard to believe but it also takes a fair amount work proofreading, working out exercises so the investment is significant. People tend to underestimate the work it takes to do thing right, which is perhaps why there are so many terrible textbooks out there.
It's worth mentioning that Carleson was on the faculty at UCLA, usually spending at least the winter quarter there (it doesn't take a genius to prefer Los Angeles in February to Uppsala in February.) I think all of the graduate students he advised were in Sweden though, which seems to be the case from the math genealogy site: http://www.genealogy.ams.org/html/id.phtml?id=1978 1 He did at least intermittently teach the first-year graduate analysis course at UCLA, and made those students suffer (and learn...)
Despite the limitations on the accuracy of the data, this is interesting and seems to describe well behavior I've seen. I have about 12 Apple laptops for student use (they develop code to run on our clusters and having their own laptops works great for that) and have had several for my own use since 2000 and have been very happy with the machines overall. AppleCare has worked very well- it's quick and effective. I used to use Sony Vaio subnotebooks, which were nice laptops and ran Linux nicely, but they were quite flimsy and Sony's repair process was ridiculous. They seemed to think it was reasonable to have the laptop sent to them for a month- being without a machine for a month is a huge hit, especially if it is your only/main machine. So for the Sonys I would wait until there were three or four problems before sending them in, since the process was a huge, slow hassle. For the Apples, they express mailed a box and you expressed it back at their expense, and they were never gone for long, and the process was very smooth, especially compared to Sony's nightmare repair process. I gave up on Sony and don't know if they have improved things, but a company that things that it's no big deal to keep your computer for a month and keep you in the dark about when its coming back isn't selling anything to me anymore. I did more repairs on the Apples than the Sonys, since I got a batch of iBooks which were vulnerable to the motherboard video weirdness, but the repairs were so smooth that it was no big deal so I wasn't waiting until there were a bunch of things wrong to send them back.
I found Apple very generous with the motherboard replacement on our iBooks, doing several for free even after the period and AppleCare expired. They did refuse to do the replacement until it actually failed, so there were a couple that we were just waiting to flake out, but overall that process was quick and painless, so all those high replacement motherboard failures in the 2001-2003 G3 iBook range really aren't as bad as they might look.
A few quibbles with the analysis on the table:
1) They don't normalize "dropping" for age of the computer, so that column is essentially meaningless. The fact that only 2% of the 17" PB G4 HRs have been dropped has a lot more to do with the fact that the machine has only been available since October 2005 than anything else, so highlighting that as "better than one standard deviation below the mean" is meaniningless.
2) There are different populations for the iBook and PB crowd, so it's a bit unfair to compare their reliability. That is, the iBooks are more likely to be used by students and the PB are more likely to be used by grownups. I think the iBooks are significantly sturdier machines (given how well they've held up to my students use) but given that on average they are subjected to less careful treatment (on average- I know some very careful students and some reckless faculty, but those are exceptions), their sturdiness doesn't show up as much in the table as a true apples v. apples (sorry) comparison would show, I suspect.
3) As in point #1, it appears that they didn't normalize for age of computer in any of the "component failure" columns, so the machines that are worse than 1 std. dev. are all older machines which have had more time to fail, so that should have been explained or corrected for.
4) The power brick/dropped correlation is described but not explained well. Those little white bricks have failed fairly often and I think are a significant problem. I think they don't hold up well under rough treatment (frayed cord where it joins the brick and frayed cord at the computer end were the most common problems) and "rough treatment" correlates well with dropping the computer, presumably. Certainly I have seen many bricks fall to the floor as they are dragged by their cords and repeated dropping takes it toll. I got several extra bricks to keep in our campus lab since students would often want to borrow mine after they left theirs at home.
Those are quibbles- I think the data there is interesting. Even though it has shortcomings, you can only work with the data you have...
I think that the questions are skewed to make things appear worse than they are, presumably because the survey is done by AOL and the National Cyber Security Alliance, who presumably have an interest in scaring people into their products and services. Aside from the obvious Linux/Mac issues described by other posters, "properly configured" firewall is a pretty strong definition and I expect many quite adequate firewalls could be classified as "improperly configured" even though they were effective against the bulk of the current attacks. Similarly, only counting anti-virus software if it has been updated in the last week is going to skew things- there is a big difference between having no AV at all and having AV that is running but has definitions that are two weeks or a month old, and the metric chosen groups those two cases together.
Charles Schumer, the senior senator from NY, has been big on this in the last month (and has been big on affordable drug policies for years...) First he said it was inexusable that Roche was putting their profits ahead of widespread safety, and last week said that Roche was being unreasonable in refusing to take steps to make the drug more widely available (not stepping up production, not meeting with other potential producers). Schumer threatened Congressional action to ignore the patent if there was no action from Roche in 30 days. Now it sounds like Roche relented, at least a little in the US market, and has agreed to step up production and at least talk with other potential producers about licensing.
No- that was a different object (2003EL61) which has a moon- it is new news that 2003UB313 has one, and the prospect of using the moon to accurately weigh 2003UB313 is huge.
Michael Brown, one of the scientists on the team that discovered the planet and now its moon, has an excellent website about 2003UB313 and has been keeping it current. I've been checking it out to see if there are any interesting developments about the team that apparently claimed the discovery of 2003UB313 without mentioning the fact that they at least visited the logs of the telescope Brown's team was using, if not outright deducing its existence from those logs. It's great to see this kind of rapid dissemination from the principals. By the way, he also has an extensive website about his newborn daughter's sleep patterns which is pretty impressive too...
In actual practice, it may be that the sample size needs to be quite large for the averages to look "normal", but there is no getting around the CLT. The assumptions that are made to use standard statistical methods are exactly that- assumptions. If the assumptions are flawed (the distribution of an actual parameter is normal, rather than the averages, for example) then the conclusions are of course dubious. Famous examples of assumptions coming back to bite analysts in a big way are generally of two types- assuming something is normal that is not, as you point out, and also assuming that events are independant that are not. The failure of the independance assumption is thought to underlie the spectactular LCTM collapse, the Asian currency collapse in 97 and other domino-type collapses.
The classic example of "crowd wisdom" is the jellybeans-in-the-jar experiment, often used in introductory MBA classes to convince people that open markets value securities (basically) fairly. The experiment goes like this: the professor brings a jar of jellybeans and asks everyone to guess how many there are in the jar. The individual estimates may vary quite a lot, but the average of the estimates in the class is usually close, in fact often closer than the closest estimate of any of the students depending upon the size of the class. That is in a situation where the students have very little information about the jar and perhaps no experience with such estimates. If there were greater experience and/or they were allowed more information then presumably the individual and average estimates would be even closer. Basically, this can be described as the "Central Limit Theorem" in action- that the standard deviation of averages is smaller than the standard deviation of the individuals by a factor of the square root of the sample size, as illustrated in this applet or in this Mathworld description. The CLT actually says more- that as the sample size increases, the distibution of averages approaches that of the normal ("bell curve") distribution, so the distribution of avergaes is roughly normal, and then techniques designed to analyze the normal distribution can be applied with greater certainty.
In such a community, people are used to being somewhat secretive about what they are studying, but not to the point of compulsion. (Sometimes people enter intentionally erroneous information in targeting logs to keep secret what they are looking at, but this is rare and usually a sign of someone that is quite paranoid...) Brown was not paranoid and did not have good reason to be, given the standards of the astronomy community. He used the name generated by his software in the abstract, which happened to be keyed off of the date and turned out to be sufficient information for Ortiz to find the targeting data and then claim the discovery as his own. Since Ortiz did not credit Brown in any way and does not seem to have responded or tried to offer any explanation, it seems quite clear to me that he acted unethically and hopefully his institute will dismiss him- as it is, the director of Ortiz' institute has promised an investigation and will probably not want the institute to be known primarily as the home of Ortiz and this unethical behavior.
I saw a similar thing in the Geneva airport- next to some pay-to-connect WiFi networks, an open wireless network named "hidden." I don't know the story, but the idea that Mssrs. Payperconnect asked the owner of the open one to make his hidden, and so he named it "hidden"- that cracked me up...
I have seen in internet cafes "locked down" computers in cabinetry with only the USB ports accessible through cutouts. That is a nice convenience at such cafes that don't have Wifi and are nervous about letting me using an Ethernet cable since I can still move outgoing mail from my laptop to a keychain drive and then take two minutes to upload and download mail to catch up on offline. I hope that USB keychain exploits don't result in cafes closing up that last hole...
Indeed- remember also that 68k/PPC "fat binaries" weren't even necessary, really. They were a convenience during the 68k/PPC transition, since even pure 68k Mac code ran pretty much seamlessly and reasonably on PPC with the built-in emulators, and in fact continues to do that today, more than 15 years later. I have old Mac apps from the System 6 and before days that run fine on my modern TiBook (under Classic, of course.) Yes, emulation is not as fast as native, but somehow I think a 1.25 GHz G4 can cope with running stuff as well as my old 1993 33MHz 68k Powerbook Duo- it's not like there were 1.25 GHz native 68k processors around, anyway...
No one like being stuck on a problem, but to be honest that is the time that real learning occurs- trying to puzzle it out, and then hopefully succeeding. Even though students don't like being stuck and often just want to know how to do a problem, simply being told how to do problems generally ruins their value as a learning tool.
As a research mathematician, I spend almost every productive waking moment "stuck" on something and I am very happy with this arrangement! (There is also the sometimes tedious process of writing down stuff formally for publication once I've worked it out, but that part is less gratifying than doing the actual research and I'm always happy when one of my co-authors is willing to do that part...)
It's not clear how they sorted out the professions, but I think they were just using data from a more all-encompassing standard survey and probably didn't have as much choice as if they had done the survey themselves (in which case it would have been easier to get a larger sample, I would think.) They do mention as a specific example that doctors are considered neutral because they have both technical and emotional components to their success and training. My guess is that "flight attendant" and "librarian" weren't choices in the survey but who knows?
The survey data does not distinguish between adopted and biological children (those questions weren't asked) so it is possible that the effect is entirely one of choice in adopted cases, or at least that it is a contributing factor.
Conception has no memory, but the distribution of X and Y chromosomes in sperm has genetic links. That is, some men produce more X or more Y chromosomal sperm. So that is where these notions come from. Too bad Henry VIII 's wives didn't realize this- it would have saved a lot of trouble if they had just snuck off and met up with someone who already had produced a family of 8 boys...
The actual article (Journal of Theoretical Biology, 233, p589-599 "Engineers have more sons, nurses have more daughters: an evolutionary psychological extension of Baron-Cohen's extreme male brain theory of autism" by Satoshi Kanazawa and Griet Vandermassen and available through Elsevier's Science Direct) came out in December 2004 an is available online for those whose institutions subscribe, notes the following correlations:
This is based on survey data from US professions of around 1500 people. Only some of the professions are categorized as "systemizing" and "empathizing" so presumably the sample size is much smaller than that . The sample size isn't listed directly in the article but it appears to be about 20% of the 1500 with at least one parent so categorized profession, for around 300 people or so. Most professions are neutral in the "systematizing/empathizing" continuum, apparently.
Amoung those with "systemizing occupations" had regression coefficients of
From the classification of professions:
Systemizing occupations
Empathizing occupations
Presumably other professions are regarded as neutral in this spectrum.
If you get a chance to visit the Consumer's Union open house (each year in early October at their headquarters in Yonkers, NY reachable from NYC by transit easily)- make an effort to turn up. It's really great to meet everyone there and chat with them. These people love what they do and care a great deal about their methods and approach and are happy to talk about things. I find it very uplifting to go on such a pilgramage when I can. (It is a biased sample, since presumable the most enthusiastic and interested people are the ones who are willing to turn up and host an open house in their labs, but still it's great!)
I know that other scientific disciplines have stronger histories of expensive journals and that their typesetting needs (color photos, etc.) may be greater than that of math and CS, so perhaps it is not so surprising to see math and CS being more of the pioneers here.
There are plenty of good expensive journals now but the point is- why not have good inexpensive or free journals instead?
The approriate thing to do is put the same energy that used to be directed toward for-profit journals instead now to free journals or journals that are reasonably priced (for example, those run by professional societies not international publishing giants, as described eloquently by Knuth in his letter.)
By editing for, refereeing for, and contributing to free/modestly priced journals, those will benefit from the effective donation of prestige and we may move to a better model for the dissemination of scientific knowledge.
As you can see, the hard part of the labor (writing, reviewing, refereeing) is not done by anyone at the publisher-- various universities pay the salaries of those folks and they pay again for the journal in dead-tree form.
So you can see that there may be some objection to the arrangement. In the old days, the journal staff actually typset things and dead-trees were the only game in town, but most of the typesetting is done by the author.
The choice is hard for some people that really need to publish in the expensive journals to get tenure, recognition, grants, etc. But for people who already have tenure, some are resistant to the journal extortion. Some may have a policy like mine- I do not submit to expensive journals or agree to referee for expensive journals, now that I have the advantage of tenure.
There have been some successes of editorial boards that resigned wholesale, then started a free/inexpensive journal. Hopefully this becomes more common.
One big difference I notice with driving gently, though, is how long things last. My friends who zip around are always replacing clutches, brakes, tires, etc. and my mechanic is always marvelling at how long things last in our cars and my truck. It is a biased sample- as a mechanic, he sees people who are hard on their vehicles much more often than he sees me, so some of that is explained by the frequency of visit. I had to put a new clutch in my truck after 11 years and 150,000 of my miles and he was shocked that it was the original- and the truck is usually driven in the mountains full of camping/climbing gear or towing a small trailer with kayaks, boards and/or boats or full of someone's stuff that I'm helping them move, so it's not like it has an easy life. (The guy who had it for the 35,000 miles before me apparently drove like an idiot, so I usually blame him for things that break, but that is getting to be a stretch after all these years...)
That's true for wind resistance from weather-related wind, but actually, small fuel-efficient streamlined cars are quite sensitive to increased wind resistance, moreso than larger vehicles, at least when it comes to added wind resistance from bikes or luggage on the roof, for example. Even roof racks without anything on them can affect mileage noticably- I remember an old friend's 83 Honda Civic which got 38 MPG with empty roof racks installed and 44 MPG with them off the car or inside. That was far more than the 5-10% we had guessed before performing the experiment. When I see a car with Colorado plates in Maine with a large bulky souvenir lobster trap attached to their roof, I wonder if they realize that the lobster trap that cost $20 to buy may also add $50 to their fuel bill on the ride home (2000 miles,30MPG vs 22MPG, gas $2/gallon, rough calculation...)
Most of us used the "tape counter" three digit counter as a directory. 005: brickout 020: lemonade stand 045: eliza and so on...