The best thing a president could possibly do for nerds or just about anyone else is stop making more laws and start dismantling the stupid laws perpetrated by previous governments.
In "The Moon Is a Harsh Mistress" (a seminal book), Robert Heinlein proposed a two-house legislature: one makes laws, one repeals them. At my university, the Dean recently lamented, "We are better at creating new things than dismantling" old, outdated and unneeded things.
From the leader: "The crux of the issue is that the Fairplay DRM that is at the heart of the iTunes/iPod universe doesn't work with anything else, meaning that if you want access to the cast [vast?] iTunes library, you have to buy an iPod."
No iPod is necessary. Just get iTunes (free), buy songs, burn to CD. No more DRM. Except for the ostensible loss of quality (I say that because I cannot detect it), I cannot see how the DRM is an obstacle. Inconvenient, ill-conceived, irritating -- yes. Insurmountable -- no.
How about regular click an edge to move the entire window, and control-click-drag anywhere on an edge to resize? (or vice versa)
The #1 thing I want in OS X, even ahead of a Finder that refreshes itself 100% of the time (and that's one very sore need): thin borders on windows. Just a few pixels for a window border would resolve my most persistent and irritating UI error -- I try to manipulate a scroll bar on a window, miss my target, and wind up activating the window lying below, which often comes from a different app. Additionally, it's a lot easier to move windows about if there are more places on which to grab them.
Just 2 pixels or so of border around the window, like Mac OS had before OS X. Too much to ask?
"Under democracy one party always devotes its chief energies to trying to prove that the other party is unfit to rule - and both commonly succeed, and are right." -- H. L. Mencken
I have put my cell on the DNC list, on the very day it opened, and even though I was informed it wasn't necessary. That doesn't matter to someone willing and able to flout the law.
I admit to some trepidation regarding listing my phones on the DNC. In a sense, it could be like confirming to a spammer that a certain email address is legit. What if a dishonest telemarketer got a hold of this prime list of guaranteed-real phone numbers? "Boy-oh-boy, let's call all of these!".
The DNC list was working for me, too... but now I'm getting a slug of calls to my cell phone, playing recorded messages in a foreign language (Spanish). They don't block Caller ID, so I can see that the calls originate in various places in the U.S.. I have saved some of the numbers, but for what purpose? The fly-by-nights will never follow the rules. They're not supposed to ever call cell phones anyway, right?
But regarding the established telemarketers, as I said, it's worked wonders. Shortly before the DNC debued, there was a stretch where I was getting pestered by AT&T, including one week where I received at least 3 calls per day. It amused me that they were fighting so hard for my $5/month in long distance business. The DNC is saving them money.
Jef also opined on the source of aerodynamic lift, giving rise to airplane flight, with an explanation that runs counter to the traditional one based on the Bernoulli effect.
On this page, Jef discusses the Coanda effect, which is familiar to anyone who's been annoyed by water or juice running down the side of a pitcher (instead of getting into your glass). An interesting read, no matter your stake on the matter.
For certain purposes (including most of what I do), fortran is unmatched.
Agreed. I'm not a programmer, I'm a scientist who writes programs (very important difference), and since I deal with equations, there's no comparison w/r/t code readability and error avoidance between Fortran and C. For me, Fortran wins hands down. Three reasons why:
(1) equations look like equations, and arrays look like arrays. A block of my code might look like this, involving calculations moving over a 3D domain
tp(i,j,k) = tm(i,j,k)-u(i,j,k)*(t(i+1,j,k)-t(i-1,j,k))...... which in C -- unless there is a good and still clear way around it -- would look like
(2) Arrays naturally start at index 1, not 0. So for Fortran array TB(NZ), there are NZ elements, numbered 1...NZ, not 0...NZ-1. The most common error I've seen students make in C, by far, is trying to access array element NZ in array TB[NZ]. IMHO, those trained in mathematics, such as myself, find starting from index 1 much more natural, and expecting to be able to access array element NZ in an array dimensioned same also very natural.
(3) IME, Fortran compilers give much better diagnostics regarding errors than their C counterparts, and point with greater accuracy to the line number of the code where the error occurs. Maybe I've only had crappy C/C++ compilers, but far too often the compiler fails to realize where the problem really lies.
I never truly appreciated Fortran until I had to become familiar with C, because that's what students were being taught these days. So, when some silly geek scoffs, "You program in Fortran?" the proper answer is, "damn right I do".
Couldn't this article have been posted earlier, and not a few hours before the action.
Don't worry, the dupe will be posted next week. That will be 8592 hours before the action returns next year. Surely, that is sufficient advance notice.
My school days were spent learning interesting and new things, spending time at labs and building stuff for science fairs... Do not make science and math easier - make the[m] enjoyable.
You had a wonderful education. I wonder how many of us can say the same, with equal fervor. I fear the answer would be depressing. How to change this?
Somewhere along the line, the emphasis of education at any level -- but especially high school -- shifted from understanding to doing, and from quality to quantity. I suspect the rise of AP classes and standardized tests, like AP, tied to specifically to courses had a lot to do with this. The SAT isn't as bad, since at least part of the test purports to assess aptitude. But AP tests are geared to a particular curriculum, and if you don't cover all the subjects, your students are at a disadvantage. There is a lot to cover; more every year, or so it seems.
I teach at the university level, and (because I can) in my courses I find I actually cover fewer topics in a term than I did in years past, but select them more carefully and cover them more completely and interactively. I came to realize that I could neglect topics of lesser importance because the students, if they understood the basics, could learn those on their own. Making students more interactively involved in the learning process makes it more fun, and leads to deeper and more permanent understanding. Less lecture and more discussion.
But students are less likely to learn those ancillary topics on their own if the class breezed by the basics, so the fundamentals were not learned sufficiently well. We can be in such a hurry to complete the syllabus that we forget that sometimes concepts need time to mature, to marinate, to germinate.
Am I wrong in tapping AP as the villain? I graduated high school before the AP phenomenon. It seems to me to be the epitome of 'teaching to the test'.
Re:Wait... so you're telling me...
on
A New Ice Age?
·
· Score: 1
You stated, "As soon as I can see an accurate 5 day weather forcast [sic] I'll start paying more attention" to model predictions of future climate. I pointed out the fallacy of your argument which unfairly and inaccurately confuses weather with climate modeling, and offered to show you a quite accurate climate model reconstruction of climate variations of the last millennium. The idea being, since the model has skill at hindcasting, it may also be skillful in simulating the climate of the next century or so.
The offer is made. If you are comfortable with your present attitude and level of knowledge, then by all means don't take me up on it.
Re:rising temps cause iceage theory?
on
A New Ice Age?
·
· Score: 1
This is a nice page describing the THC and its import.
Re:rising temps cause iceage theory?
on
A New Ice Age?
·
· Score: 1
Not just melting ice, but also (and IMHO, more importantly) increased precipitation in the north Atlantic can bring more freshwater into the ocean, impeding the sinking of Gulf Stream waters. This is called the thermohaline circulation (THC).
This is a nice page describing the THC and its import.
Re:Wait... so you're telling me...
on
A New Ice Age?
·
· Score: 1
no, what I said was that the world was significantly warmer before the year 1100, then a large amount of fresh water got dumped into the atlantic which in the course of 70 years dropped the temp to about 50 to 55 degrees and it was like that for about 300-400 years. we are now recovering from that event known as the little ice age and this "catastrophic" global warming that is going on is not catastrophic at all, but part of the natural cycle of earth.
Where are you getting your numbers? The quoted text appears to suggest your "50-55 deg" is a global average temperature. That cannot be correct; that is a huge drop from the recent global average surface temperature of about 60F.
During the "medieval warmperiod", which concluded about 1100AD, the global average temperature was about 0.25 degrees C (about 0.5F) higher than the 1901-1960 average. This was eclipsed by 1940, so since then the global average temperature has been higher than has been recorded during the last millennium, reaching about 0.7C (1.3F) above the 1901-1960 average by the end of the 20th century.
Source: articles published by Jones et al., 1998 and 2003.
Re:Wait... so you're telling me...
on
A New Ice Age?
·
· Score: 5, Insightful
As soon as I can see an accurate 5 day weather forcast I'll start paying more attention.
No, no, a thousand times no.
Nothing personal; you're just repeating what you've been told, but you have been told wrong.
The fact that short-range weather forecasts for individual locales lose skill at roughly 10 days does not mean that accurate 50+ year climate simulations are not possible. Why? The short answer is weather != climate.
The climate model is not concerned with predicting the temperature and skycover at London at 3PM on April 1, 2078. It cannot do so. It is interested in the broad -- global, regional -- statistics: means, variances, seasonal/annual/decadal precip totals and averages, etc.. It is possible to get those right even though forecasts at fixed points in space in time are wrong. We're looking at the forest here, not the trees.
If you take a short-term weather forecast model and perturb its starting conditions, even by a wee little bit, you will wind up with a very different result in short order -- in under a week. One simulation might be predicting sunny for a fixed point, the other cloudy. One cooler than normal, one warmer; one wet, one dry. Chaos theory, and all that.
But it's still the same climate. Please understand this. Yes, the skill in assessing "weather" fluctuations about the climate mean has disappeared, but the climate remained the same.
What climate models are trying to do is ascertain whether the climate itself is changing. Are climate models perfect, complete, 100% skillful? NO, of course not. Do they have a long way to go towards improvement? YES. Are they useless? Well, you be the judge.
I have a very nice figure showing how well a climate model was able to reproduce climate (NOT weather) variations -- specifically, global average temperature -- over the last millennium. Model predictions are superposed on climate data reconstructed from proxies. The model was run numerous times, with perturbed starting conditions, to yield an "ensemble", helping to assess the range of uncertainty.
I can't find this image on the web, and don't want to put it somewhere where it might be slashdotted, but if you really care enough, email me at rfovell at yahoo dot com and I will send it to you, along with an explanation of what you are looking at. It's an excellent reconstruction. So good you simply have to pay attention to what these models are saying about the future.
Do we expect most car drivers to change their own oil, perform maintinence, and change their timing belts? Most people don't. I know many people who have never even opened their hood.
Hey, here's a car analogy that finally makes sense to me.
"I know many people who have never even opened their hood." And that's true now more so than ever before. In previous generations, a lot of people at least opened the hood to check their own oil, and a fairly large fraction would swap out their own plugs, etc.. Even farther back, all that was quite necessary.
But cars were simpler machines then. Nowadays, the majority of us just swoosh off to Jiffy Lube for simple things or to the dealer. Look under the hood; to how many of us does anything even make sense? Despite the fact that engine compartments tend to be better organized and labeled than ever, it remains they are more complicated than ever. It's a bloody mess in there.
Same with computers. I'm a geek but I just don't want to root through the inner workings and hidden mechanisms anymore. These things used to be very easy to troubleshoot, to mess with, but that's not true anymore. With greater utility, more doo-dads, comes complexity.
The answer isn't really to make computers or cars simpler, but rather to make then easier to maintain and service. Honda rolls all the services, major and minor, into packages tied to time and distance... I follow that schedule and am relieved from having to think about the details. Is that the most efficient way of servicing my car from a cost standpoint? Well, yes, if you factor in the value of my time and of the simplicity this brings to an overly complicated life.
People may resist taking in their computers to a service station every X-thousand hours -- way too much hassle unplugging and such, for desktops -- so think about piano tuners. Regular visits from a tuner. My computer tells me when there are updates, and that's good enough for me (now, anyway), but there are those who don't want to deal with even that, and besides a periodic 'tuning' of sorts may strike them as sensible anyway. This exists now, of course, but it's small; maybe this will be a big business in the future.
In the same way ease of use is vital for unix to gain marketshare.
What people want is a tool that will be there when they want it, work on demand, not break a lot, and someone to turn to when there's trouble. Like a car. Something they don't have to think about, because Lord knows they have enough to deal with already. It is *that* part of 'ease of use' that computers so often fail on.
Gatorade's label says (or used to say) "Scientifically formulated to taste best when you need it most." Not provide the greatest benefit, but taste best.
Disclaimer: Yes, I am a (satisfied) iTunes Music Store customer. However, I buy most music on CD and rip it the traditional way still.
Same here, and it might be worth a little wondering why.
iTMS is great for sampling music from the convenience of my home. The samples are of a goodly length and, more often than not, taken from an interesting and representative portion of the track. A lot of care seems to go into their selections.
But when it comes to buying an album (I rarely purchase single tracks), I just can't seem to push that button, at least not very often. I don't think I care about the lossy codec; I rip my CDs using that anyway, and can't tell the difference. There's something about the original pressed CD, the original printed inserts, that still has value to me. The full album price (typically $9.99) isn't cheap enough compared to the kind of discounted prices I can get on "real" pressed CDs to make up for that perhaps silly desire to "own the original".
So, I've made far fewer iTMS purchases than I had anticipated I would.
I wonder if it makes sense for Apple to consider teaming up with Amazon (or similar) and provide a "buy pressed CD" option. You use iTMS to select the album and, if desired, have Amazon ship it to you (usually free of tax and shipping, at least for me). Apple gets a cut, Amazon gets a sale, I get to cut out the middle step of switching to my browser.
The only issue is whether there are a sufficient number of traditionalists that for whatever reason, still want the pressed CD and packaging...
And really, 99c for a song isn't even that great of a deal. That makes a 15 song cd = $15.
When you visit iTMS you will see that most albums having more than 10 tracks have prices capped at $9.99. Yes, there are counterexamples (Kitaro's "An Ancient Journey" is $18.81, the aggregate cost of the individual tracks), but they strike me as quite rare. I noticed a few albums are actually less than $9.99, and those have 9 or fewer tracks.
So, FYI, your blanket statement "That makes a 15 song CD = $15" is far more often incorrect than correct.
At my university, we also have a prof rating website (www.bruinwalk.com/professors/). The reviewers do indeed tend to be self-selected, tho on our boards, the top segment of the class also (thankfully) seems motivated to post. However, the biggest difference may be that the online reviews tend to be posted after the course grade is known. That can have unintended consequences.
As a professor myself, I have had mixed feelings about this board. However, some recent changes, and also this slashdot discussion, have encouraged me to look at our site more favorably. Those are:
* The site now shows distributions of scores, instead of just listing average stats. The frequency distribution can be most illuminating. * A poor review, coupled with a "s/he's too hard!!" complaint can have a most salutary effect on the composition of your next class:-) * Compared to what I've seen on other sites, the student comments on my school's site -- even and especially the negative ones -- appear to be quite mature. Sure, there are a few childish and transparent ones, but they're the exception (and, besides, likely to have the opposite of the intended effect).
Certainly, there is no universally perfect teacher or class. I think all anyone can ask is that posters be honest in their comments. I've read hundreds and hundreds of student comments (not only my own, but also in committee work), and I have to say that I am pretty impressed with the students' candor.
Anyway, does your scientist friend note how many people were on the planet before and after each of these plagues? The number of people on the planet in the 1800s is so much greater than in the Roman era that the 1800s era *should* have been much much much hotter than the Roman time and that 'little ice age' should have only reduced the global temp by a small number which would have still left it well above the Roman era.
You are presuming the effect of agriculture on climate scales with population. This is likely wrong. A single acre of farmland can support far, far more people now than during the Little Ice Age, to say nothing of the Roman era. The critical factors include acreage, along with specifics of the agricultural land use. Being compared here is land that is actively farmed with land that has fallen out of use. That doesn't just involve CO2 uptake, by the way; albedo and surface moisture fluxes are also relevant and those vary with land type and use.
The challenge for the proponent of this theory - beyond the patent obstacle of correlation/causation - is to find a convincing proxy for forestation. He then has to show that forestation changes have led the temperature variations. Right now, all the article notes is an apparent variation with population decreases (owing to plague) leading the onset of cooler spells. This may be only coincidence, as I have already intimated, but it is an intriguing one and I trust the issue won't be settled by the closed-minded.
Perhaps, dear prudence, if I get some time, I can explain "cause" and "effect" to you.
Instead, why don't you spend that time RTFA or, more likely, finding someone to explain it to you?
If I could be bothered explaining things to witless ACs, I would make sure you realized that the article's subject involves hypothesized climate changes owing to the spread of agriculture. That these changes represent deviations from the Milankovitch cycles and are possibly related to degrees of human-induced forest clearance and agricultural activities in general. That greater forestation means less atmospheric carbon dioxide, which is a greenhouse gas. That a test of this hypothesis is: does cooling result from periods of relatively greater forest coverage?
I would further note that the scientist quoted in the article believes that historically cooler periods, such as the Little Ice Age, followed periods of widespread plague, during which times human populations decreased and "a lot of previously farmed land turned back into forest." That is, he believes the cause is decreased agriculture and increased forestation and the effect is a period of cooling beyond that which could be attributed to orbital variations. I would conclude by noting more evidence is necessary to satisfy the scientifically literate, and other contributory factors were likely involved, but this seems to be a novel explanation for periods such as the Little ice Age.
However, I shall not do you this service since I have no truck with cowardly gits. Sorry, boy-o.
Slashdot Mirror
The best thing a president could possibly do for nerds or just about anyone else is stop making more laws and start dismantling the stupid laws perpetrated by previous governments.
In "The Moon Is a Harsh Mistress" (a seminal book), Robert Heinlein proposed a two-house legislature: one makes laws, one repeals them. At my university, the Dean recently lamented, "We are better at creating new things than dismantling" old, outdated and unneeded things.
Let's do both.
No iPod is necessary. Just get iTunes (free), buy songs, burn to CD. No more DRM. Except for the ostensible loss of quality (I say that because I cannot detect it), I cannot see how the DRM is an obstacle. Inconvenient, ill-conceived, irritating -- yes. Insurmountable -- no.
How about regular click an edge to move the entire window, and control-click-drag anywhere on an edge to resize? (or vice versa)
The #1 thing I want in OS X, even ahead of a Finder that refreshes itself 100% of the time (and that's one very sore need): thin borders on windows. Just a few pixels for a window border would resolve my most persistent and irritating UI error -- I try to manipulate a scroll bar on a window, miss my target, and wind up activating the window lying below, which often comes from a different app. Additionally, it's a lot easier to move windows about if there are more places on which to grab them.
Just 2 pixels or so of border around the window, like Mac OS had before OS X. Too much to ask?
"Under democracy one party always devotes its chief energies to trying to prove that the other party is unfit to rule - and both commonly succeed, and are right."
-- H. L. Mencken
True then and now.
I admit to some trepidation regarding listing my phones on the DNC. In a sense, it could be like confirming to a spammer that a certain email address is legit. What if a dishonest telemarketer got a hold of this prime list of guaranteed-real phone numbers? "Boy-oh-boy, let's call all of these!".
But regarding the established telemarketers, as I said, it's worked wonders. Shortly before the DNC debued, there was a stretch where I was getting pestered by AT&T, including one week where I received at least 3 calls per day. It amused me that they were fighting so hard for my $5/month in long distance business. The DNC is saving them money.
Remember when Macs couldn't multitask? ;-)
Jef also opined on the source of aerodynamic lift, giving rise to airplane flight, with an explanation that runs counter to the traditional one based on the Bernoulli effect.
On this page, Jef discusses the Coanda effect, which is familiar to anyone who's been annoyed by water or juice running down the side of a pitcher (instead of getting into your glass). An interesting read, no matter your stake on the matter.
For certain purposes (including most of what I do), fortran is unmatched.
... ... which in C -- unless there is a good and still clear way around it -- would look like
Agreed. I'm not a programmer, I'm a scientist who writes programs (very important difference), and since I deal with equations, there's no comparison w/r/t code readability and error avoidance between Fortran and C. For me, Fortran wins hands down. Three reasons why:
(1) equations look like equations, and arrays look like arrays. A block of my code might look like this, involving calculations moving over a 3D domain
tp(i,j,k) = tm(i,j,k)-u(i,j,k)*(t(i+1,j,k)-t(i-1,j,k))
tp[i][j][k] = tm[i][j][k] - u[i][j][k]*(t[i+1][j][k]-t[i-1][j][k])...;
(2) Arrays naturally start at index 1, not 0. So for Fortran array TB(NZ), there are NZ elements, numbered 1...NZ, not 0...NZ-1. The most common error I've seen students make in C, by far, is trying to access array element NZ in array TB[NZ]. IMHO, those trained in mathematics, such as myself, find starting from index 1 much more natural, and expecting to be able to access array element NZ in an array dimensioned same also very natural.
(3) IME, Fortran compilers give much better diagnostics regarding errors than their C counterparts, and point with greater accuracy to the line number of the code where the error occurs. Maybe I've only had crappy C/C++ compilers, but far too often the compiler fails to realize where the problem really lies.
I never truly appreciated Fortran until I had to become familiar with C, because that's what students were being taught these days. So, when some silly geek scoffs, "You program in Fortran?" the proper answer is, "damn right I do".
Couldn't this article have been posted earlier, and not a few hours before the action.
Don't worry, the dupe will be posted next week. That will be 8592 hours before the action returns next year. Surely, that is sufficient advance notice.
My school days were spent learning interesting and new things, spending time at labs and building stuff for science fairs... Do not make science and math easier - make the[m] enjoyable.
You had a wonderful education. I wonder how many of us can say the same, with equal fervor. I fear the answer would be depressing. How to change this?
Somewhere along the line, the emphasis of education at any level -- but especially high school -- shifted from understanding to doing, and from quality to quantity. I suspect the rise of AP classes and standardized tests, like AP, tied to specifically to courses had a lot to do with this. The SAT isn't as bad, since at least part of the test purports to assess aptitude. But AP tests are geared to a particular curriculum, and if you don't cover all the subjects, your students are at a disadvantage. There is a lot to cover; more every year, or so it seems.
I teach at the university level, and (because I can) in my courses I find I actually cover fewer topics in a term than I did in years past, but select them more carefully and cover them more completely and interactively. I came to realize that I could neglect topics of lesser importance because the students, if they understood the basics, could learn those on their own. Making students more interactively involved in the learning process makes it more fun, and leads to deeper and more permanent understanding. Less lecture and more discussion.
But students are less likely to learn those ancillary topics on their own if the class breezed by the basics, so the fundamentals were not learned sufficiently well. We can be in such a hurry to complete the syllabus that we forget that sometimes concepts need time to mature, to marinate, to germinate.
Am I wrong in tapping AP as the villain? I graduated high school before the AP phenomenon. It seems to me to be the epitome of 'teaching to the test'.
You stated, "As soon as I can see an accurate 5 day weather forcast [sic] I'll start paying more attention" to model predictions of future climate. I pointed out the fallacy of your argument which unfairly and inaccurately confuses weather with climate modeling, and offered to show you a quite accurate climate model reconstruction of climate variations of the last millennium. The idea being, since the model has skill at hindcasting, it may also be skillful in simulating the climate of the next century or so.
The offer is made. If you are comfortable with your present attitude and level of knowledge, then by all means don't take me up on it.
This is a nice page describing the THC and its import.
This is it
Not just melting ice, but also (and IMHO, more importantly) increased precipitation in the north Atlantic can bring more freshwater into the ocean, impeding the sinking of Gulf Stream waters. This is called the thermohaline circulation (THC).
This is a nice page describing the THC and its import.
no, what I said was that the world was significantly warmer before the year 1100, then a large amount of fresh water got dumped into the atlantic which in the course of 70 years dropped the temp to about 50 to 55 degrees and it was like that for about 300-400 years. we are now recovering from that event known as the little ice age and this "catastrophic" global warming that is going on is not catastrophic at all, but part of the natural cycle of earth.
Where are you getting your numbers? The quoted text appears to suggest your "50-55 deg" is a global average temperature. That cannot be correct; that is a huge drop from the recent global average surface temperature of about 60F.
During the "medieval warmperiod", which concluded about 1100AD, the global average temperature was about 0.25 degrees C (about 0.5F) higher than the 1901-1960 average. This was eclipsed by 1940, so since then the global average temperature has been higher than has been recorded during the last millennium, reaching about 0.7C (1.3F) above the 1901-1960 average by the end of the 20th century.
Source: articles published by Jones et al., 1998 and 2003.
As soon as I can see an accurate 5 day weather forcast I'll start paying more attention.
No, no, a thousand times no.
Nothing personal; you're just repeating what you've been told, but you have been told wrong.
The fact that short-range weather forecasts for individual locales lose skill at roughly 10 days does not mean that accurate 50+ year climate simulations are not possible. Why? The short answer is weather != climate.
The climate model is not concerned with predicting the temperature and skycover at London at 3PM on April 1, 2078. It cannot do so. It is interested in the broad -- global, regional -- statistics: means, variances, seasonal/annual/decadal precip totals and averages, etc.. It is possible to get those right even though forecasts at fixed points in space in time are wrong. We're looking at the forest here, not the trees.
If you take a short-term weather forecast model and perturb its starting conditions, even by a wee little bit, you will wind up with a very different result in short order -- in under a week. One simulation might be predicting sunny for a fixed point, the other cloudy. One cooler than normal, one warmer; one wet, one dry. Chaos theory, and all that.
But it's still the same climate. Please understand this. Yes, the skill in assessing "weather" fluctuations about the climate mean has disappeared, but the climate remained the same.
What climate models are trying to do is ascertain whether the climate itself is changing. Are climate models perfect, complete, 100% skillful? NO, of course not. Do they have a long way to go towards improvement? YES. Are they useless? Well, you be the judge.
I have a very nice figure showing how well a climate model was able to reproduce climate (NOT weather) variations -- specifically, global average temperature -- over the last millennium. Model predictions are superposed on climate data reconstructed from proxies. The model was run numerous times, with perturbed starting conditions, to yield an "ensemble", helping to assess the range of uncertainty.
I can't find this image on the web, and don't want to put it somewhere where it might be slashdotted, but if you really care enough, email me at rfovell at yahoo dot com and I will send it to you, along with an explanation of what you are looking at. It's an excellent reconstruction. So good you simply have to pay attention to what these models are saying about the future.
Thanks for reading this far.
Do we expect most car drivers to change their own oil, perform maintinence, and change their timing belts? Most people don't. I know many people who have never even opened their hood.
Hey, here's a car analogy that finally makes sense to me.
"I know many people who have never even opened their hood." And that's true now more so than ever before. In previous generations, a lot of people at least opened the hood to check their own oil, and a fairly large fraction would swap out their own plugs, etc.. Even farther back, all that was quite necessary.
But cars were simpler machines then. Nowadays, the majority of us just swoosh off to Jiffy Lube for simple things or to the dealer. Look under the hood; to how many of us does anything even make sense? Despite the fact that engine compartments tend to be better organized and labeled than ever, it remains they are more complicated than ever. It's a bloody mess in there.
Same with computers. I'm a geek but I just don't want to root through the inner workings and hidden mechanisms anymore. These things used to be very easy to troubleshoot, to mess with, but that's not true anymore. With greater utility, more doo-dads, comes complexity.
The answer isn't really to make computers or cars simpler, but rather to make then easier to maintain and service. Honda rolls all the services, major and minor, into packages tied to time and distance... I follow that schedule and am relieved from having to think about the details. Is that the most efficient way of servicing my car from a cost standpoint? Well, yes, if you factor in the value of my time and of the simplicity this brings to an overly complicated life.
People may resist taking in their computers to a service station every X-thousand hours -- way too much hassle unplugging and such, for desktops -- so think about piano tuners. Regular visits from a tuner. My computer tells me when there are updates, and that's good enough for me (now, anyway), but there are those who don't want to deal with even that, and besides a periodic 'tuning' of sorts may strike them as sensible anyway. This exists now, of course, but it's small; maybe this will be a big business in the future.
In the same way ease of use is vital for unix to gain marketshare.
What people want is a tool that will be there when they want it, work on demand, not break a lot, and someone to turn to when there's trouble. Like a car. Something they don't have to think about, because Lord knows they have enough to deal with already. It is *that* part of 'ease of use' that computers so often fail on.
Gatorade's label says (or used to say) "Scientifically formulated to taste best when you need it most." Not provide the greatest benefit, but taste best.
Disclaimer: Yes, I am a (satisfied) iTunes Music Store customer. However, I buy most music on CD and rip it the traditional way still.
Same here, and it might be worth a little wondering why.
iTMS is great for sampling music from the convenience of my home. The samples are of a goodly length and, more often than not, taken from an interesting and representative portion of the track. A lot of care seems to go into their selections.
But when it comes to buying an album (I rarely purchase single tracks), I just can't seem to push that button, at least not very often. I don't think I care about the lossy codec; I rip my CDs using that anyway, and can't tell the difference. There's something about the original pressed CD, the original printed inserts, that still has value to me. The full album price (typically $9.99) isn't cheap enough compared to the kind of discounted prices I can get on "real" pressed CDs to make up for that perhaps silly desire to "own the original".
So, I've made far fewer iTMS purchases than I had anticipated I would.
I wonder if it makes sense for Apple to consider teaming up with Amazon (or similar) and provide a "buy pressed CD" option. You use iTMS to select the album and, if desired, have Amazon ship it to you (usually free of tax and shipping, at least for me). Apple gets a cut, Amazon gets a sale, I get to cut out the middle step of switching to my browser.
The only issue is whether there are a sufficient number of traditionalists that for whatever reason, still want the pressed CD and packaging...
And really, 99c for a song isn't even that great of a deal. That makes a 15 song cd = $15.
When you visit iTMS you will see that most albums having more than 10 tracks have prices capped at $9.99. Yes, there are counterexamples (Kitaro's "An Ancient Journey" is $18.81, the aggregate cost of the individual tracks), but they strike me as quite rare. I noticed a few albums are actually less than $9.99, and those have 9 or fewer tracks.
So, FYI, your blanket statement "That makes a 15 song CD = $15" is far more often incorrect than correct.
Vorbis was named after a character in Terry Pratchett's "Small Gods", which is one very witty book. The best parody of religion I ever read.
But, Vorbis was the drop dead evil bad guy, so I wonder why it was chosen. Besides, doesn't it sound like the name of a Windows virus?
Good points.
:-)
At my university, we also have a prof rating website (www.bruinwalk.com/professors/). The reviewers do indeed tend to be self-selected, tho on our boards, the top segment of the class also (thankfully) seems motivated to post. However, the biggest difference may be that the online reviews tend to be posted after the course grade is known. That can have unintended consequences.
As a professor myself, I have had mixed feelings about this board. However, some recent changes, and also this slashdot discussion, have encouraged me to look at our site more favorably. Those are:
* The site now shows distributions of scores, instead of just listing average stats. The frequency distribution can be most illuminating.
* A poor review, coupled with a "s/he's too hard!!" complaint can have a most salutary effect on the composition of your next class
* Compared to what I've seen on other sites, the student comments on my school's site -- even and especially the negative ones -- appear to be quite mature. Sure, there are a few childish and transparent ones, but they're the exception (and, besides, likely to have the opposite of the intended effect).
Certainly, there is no universally perfect teacher or class. I think all anyone can ask is that posters be honest in their comments. I've read hundreds and hundreds of student comments (not only my own, but also in committee work), and I have to say that I am pretty impressed with the students' candor.
Classic mode in Mac OS X, that is, is a tile game called Ishido: The Way of Stones. Looks great, plays even better.
Free download, BTW. Wish there were a native OS X version!
Anyway, does your scientist friend note how many people were on the planet before and after each of these plagues? The number of people on the planet in the 1800s is so much greater than in the Roman era that the 1800s era *should* have been much much much hotter than the Roman time and that 'little ice age' should have only reduced the global temp by a small number which would have still left it well above the Roman era.
You are presuming the effect of agriculture on climate scales with population. This is likely wrong. A single acre of farmland can support far, far more people now than during the Little Ice Age, to say nothing of the Roman era. The critical factors include acreage, along with specifics of the agricultural land use. Being compared here is land that is actively farmed with land that has fallen out of use. That doesn't just involve CO2 uptake, by the way; albedo and surface moisture fluxes are also relevant and those vary with land type and use.
The challenge for the proponent of this theory - beyond the patent obstacle of correlation/causation - is to find a convincing proxy for forestation. He then has to show that forestation changes have led the temperature variations. Right now, all the article notes is an apparent variation with population decreases (owing to plague) leading the onset of cooler spells. This may be only coincidence, as I have already intimated, but it is an intriguing one and I trust the issue won't be settled by the closed-minded.
Perhaps, dear prudence, if I get some time, I can explain "cause" and "effect" to you.
Instead, why don't you spend that time RTFA or, more likely, finding someone to explain it to you?
If I could be bothered explaining things to witless ACs, I would make sure you realized that the article's subject involves hypothesized climate changes owing to the spread of agriculture. That these changes represent deviations from the Milankovitch cycles and are possibly related to degrees of human-induced forest clearance and agricultural activities in general. That greater forestation means less atmospheric carbon dioxide, which is a greenhouse gas. That a test of this hypothesis is: does cooling result from periods of relatively greater forest coverage?
I would further note that the scientist quoted in the article believes that historically cooler periods, such as the Little Ice Age, followed periods of widespread plague, during which times human populations decreased and "a lot of previously farmed land turned back into forest." That is, he believes the cause is decreased agriculture and increased forestation and the effect is a period of cooling beyond that which could be attributed to orbital variations. I would conclude by noting more evidence is necessary to satisfy the scientifically literate, and other contributory factors were likely involved, but this seems to be a novel explanation for periods such as the Little ice Age.
However, I shall not do you this service since I have no truck with cowardly gits. Sorry, boy-o.